et al. 2008 - Addi - University of the Basque Country

Document technical information

Format pdf
Size 3.5 MB
First found May 22, 2018

Document content analysis

Category Also themed
Language
English
Type
not defined
Concepts
no text concepts found

Persons

Organizations

Places

Transcript

Phylogeography and non-invasive landscape genetics of
the European pine marten (Martes martes L. 1758):
Insights into ancient and contemporary processes
shaping genetic variation
Front cover photo: Peter Cairns © www.northshots.com
Back cover photo: Laurie Campbell ©
ACTA DE GRADO DE DOCTOR
ACTA DE DEFENSA DE TESIS DOCTORAL
DOCTORANDO DON/ÑA. Aritz Ruiz González
TITULO DE LA TESIS: Phylogeography and non-invasive landscape genetics of
the European pine marten (Martes martes L. 1758): Insights into ancient and
contemporary processes shaping genetic variation
El Tribunal designado por la Subcomisión de Doctorado de la UPV/EHU para calificar
la Tesis Doctoral arriba indicada y reunido en el día de la fecha, una vez efectuada la
defensa por el doctorando y contestadas las objeciones y/o sugerencias que se le han
formulado, ha otorgado por___________________la calificación de:
unanimidad ó mayoría
En
a
de
EL/LA PRESIDENTE/A,
de
EL/LA SECRETARIO/A,
Fdo.:
Fdo.:
Dr/a: ____________________
Dr/a: ___________________
VOCAL 1º,
VOCAL 2º,
VOCAL 3º,
Fdo.:
Fdo.:
Fdo.:
Dr/a:
Dr/a:
Dr/a:
EL/LA DOCTORANDO/A,
Fdo.: Aritz Ruiz- González
AUTORIZACION DEL/LA DIRECTOR/A DE TESIS
PARA SU PRESENTACION
Dr/a. Benjamín Gómez Moliner con N.I.F. 14245903W
como Director/a de la Tesis Doctoral: Phylogeography and non-invasive landscape
genetics of the European pine marten (Martes martes L. 1758): Insights into
ancient and contemporary processes shaping genetic variation
realizada en el Departamento Zoología y Biología Celular Animal
por el Doctorando Don/ña. Aritz Ruiz-González
autorizo la presentación de la citada Tesis Doctoral, dado que reúne las condiciones
necesarias para su defensa.
En Vitoria a 14de Diciembre de 2010
EL/LA DIRECTOR/A DE LA TESIS
Fdo.:
AUTORIZACION DEL/LA DIRECTOR/A DE TESIS
PARA SU PRESENTACION
Dr/a. María José Madeira García con N.I.F. 72728664L
como Director/a de la Tesis Doctoral: Phylogeography and non-invasive landscape
genetics of the European pine marten (Martes martes L. 1758): Insights into
ancient and contemporary processes shaping genetic variation
realizada en el Departamento Zoología y Biología Celular Animal
por el Doctorando Don/ña. Aritz Ruiz-González
autorizo la presentación de la citada Tesis Doctoral, dado que reúne las condiciones
necesarias para su defensa.
En Vitoria a 14de Diciembre de 2010
EL/LA DIRECTOR/A DE LA TESIS
Fdo.:
Departamento de Zoología y Biología Celular Animal
Facultad de Farmacia
Phylogeography and non-invasive landscape genetics
of the European pine marten (Martes martes L. 1758):
Insights into ancient and contemporary processes
shaping genetic variation
Tesis Doctoral dirigida por:
Dr. Benjamín J. Gómez-Moliner
Dra. Mª José Madeira García
Memoria que para optar al grado de doctor presenta
ARITZ RUIZ GONZÁLEZ
Vitoria-Gasteiz
2011
© 2011, Aritz Ruiz González
“La utopía está en el horizonte. Camino dos pasos, ella se
aleja dos pasos y el horizonte se corre diez pasos más allá.
¿Entonces para que sirve la utopía? Para eso, sirve para
caminar.”
Eduardo Galeano
“We only preserve what we love, we only love what we
understand, we only understand what we study.”
Tibetan Saying
Son tantas las personas que me gustaría incluir en este apartado que quizá necesitaría un capítulo entero
para agradecerles a todas y cada una de ellas específicamente la ayuda prestada durante esta tesis doctoral.
Muchos y variados son los protagonistas secundarios de este trabajo, pero quiero recalcar que todos
aquellos que habéis aportado vuestro granito de arena, habéis permitido que esta tesis sea hoy una realidad
tangible.
En primer lugar, quisiera mostrar mi más sincero agradecimiento a mis directores de tesis Benjamín
Gómez-Moliner y María José Madeira. A ti Benja no solo por apoyarme desde un primer momento,
donde las ideas y la realidad asociadas a este trabajo se encontraban muy alejadas la una de la otra, sino por
tu pasión y empeño que han permitido que lleguemos a buen puerto. A ti Marijo, por prestar tu
conocimiento y compañerismo a lo largo de esta tesis que ha ido in crescendo a medida que nos
acercábamos a la recta final. Incluso con Irati en camino tu apoyo ha sido decisivo: Mila esker!!
Si hay alguien a quien esta tesis le debe mucho, es a Jonathan Rubines. Además de haber sido cómplice en
el laboratorio de mis primeras PCR-s, extracciones, y un largo etc. te has convertido en un compañero de
viaje imprescindible en todo aquello relacionado con el estudio de la fauna silvestre. Mientras nacía esta
tesis se forjaron las bases de la asociación NATURESFERA que a día de hoy sigue dando sus frutos. Y
desde luego a Cebrián, por descubrirme el apasionante mundo de las huellas, rastros y señales de los
carnívoros (así como otras muchas cosas) que me han permitido llevar a cabo un trabajo donde las técnicas
tradicionales de campo se aúnan con las más novedosas técnicas moleculares.
Junto con Joni, siempre han estado Arantza “ilegorri” y Maite “la visona”, quienes a pesar de estar ahora de
post-doc por tierras belgas, habéis seguido siempre ahí de una manera u otra. Gracias Arantza por haber
acompañado tantos cigarros que aliviaban el stress, así como por tu capacidad incansable para mostrar e
incitar una sonrisa cuando más se necesitaba. A ti Maite, por ser fuente continua de bibliografía y un
ejemplo revelador de que el trabajo bien hecho siempre recibe su merecida recompensa.
No puedo olvidarme de Oskar Berdión, quien jugó un papel fundamental en las primeras fases de este
trabajo y me brindó la oportunidad de conocer en primera persona a la verdadera protagonista de esta tesis
(la marta) mediante su trabajo de radioseguimento en la Sierra de Elgea-Urkilla. Oskar, no me olvido:
¡Aún nos debes una comida!
Agradecimientos AGRADECIMIENTOS
Y no, no me he olvidado de vosotras, las futuras doctoras María Vergara y Oihana Razkin. Sin duda
vuestra ayuda en los dos últimos años ha sido indispensable para que la vorágine de muestras en la que me
veía inmerso estén a día de hoy etiquetadas, ordenadas e identificadas a la perfección. Mila eker!! Y como
no, a Luisja. Una persona (o mejor dicho personaje) que siempre te hace la vida más alegre y siempre esta
disponible para echarte un mano. También hay un hueco especial en estos agradecimientos para Ainhoa
“calamita” y Vanessa “dalmatina”. Dos luchadoras y trabajadoras natas, que me han demostrado en todo
este tiempo que es posible crecerse ante la adversidad poniéndole “al mal tiempo buena cara”, donde la
ilusión por conseguir tus metas siempre supera las barreras que nos ponga la vida. A Mikel Gurrutxaga, por
ser un impulsor de nuevas ideas y un apoyo fundamental en la elaboración de esta tesis. Mikel, este trabajo
te debe mucho… También me gustaría agradecer al recién llegado Urtzi Goiti, por sus comentarios, ideas
y conversaciones que siempre inducen a darle una vuelta más de tuerca a cualquier cuestión. No quiero
olvidarme tampoco de quienes nos acompañaron un día (Aritz Zubielki, Ariane, Xabi, Esti…) y de quienes
acaban de llegar a nuestro laboratorio (Amaia, Álvaro, Adriana, Marina…)
Quisiera agradecer también al Director del Conservation Genetics Lab (ISPRA) Ettore Randi por
brindarme la oportunidad de realizar varias estancias y acogerme como a uno más en su laboratorio de
Ozzano (Italia). Grazie mille non solo a Ettore, il mio periodo di ricerca in Italia me ha permesso di parlare
(un po) una bella e nuova lingua e conoscere tanti personaggi incredibile: Romolo Caniglia “Dr. Otsoa”,
Andrea “Il pipistreli de la birra”, Francesca Davoli “la Mamma”, Rita Oliveira “la gata portuguesa”, Ilaria
Celentano, Andrea Grill, Nadia Muci, Marco Galaverni, Elena Fabri, Claudia Greco, Mario, Adriano,
Fabio…Grazie mille a tuti!!!
Grazie a Alessandro Balestrieri, Luigi Remonti e Claudio Prigioni (University of Pavia) per la fruttuose
collaborazioni in la ricerca della martora nella pianura del fiume Po. Spero che possiamo continuare a
lavorare insieme nella conoscenza di questa specie…
I would like to thank also to Michael Schwartz and Keith Aubry for providing me the great opportunity to
contribute as co–author to the book chapter entitled “Martes Conservation Genetics: Using molecular
genetics to assess within species movements, barriers and corridors” that will be included in “Biology and
Conservation of Marten, Sables, and Fisher: a new synthesis”. I would like to thank also the other coauthors of this book chapter, Cino Pertoldi and Ryuchi Masuda.
I’m indebted to the wildlife photographers Peter Cairns and Laurie Campbell for kindly providing me
several amazing photographs of the European pine marten to include in this thesis.
Quisiera agradecer de manera especial también a Isabel Barja y al resto de su equipo por el enorme trabajo
desempeñado en los Montes do Invernadeiro.
Como no, agradecer a Marian Martínez de Pancorbo y a todo su equipo del banco de ADN que además de
compañeros de fatigas habéis sido siempre algo más que eso… Xabi, Laura, Marijo, Carmen, Naiara,
Maite, Ameli, David, Leire, Alejandra, Jose, Jose Aznar…Pero de entre todos ellos quiero hacer una
mención especial a Sergio Cardoso, el “tío elegante de Donosti”, y al recién nombrado doctor Adrían
Odriozola ¡Que hubiera sido de mí sin la “otra” mitad del cerebro para tramitar papeleos!
Me gustaría recalcar que este trabajo no hubiera sido posible sin la ayuda desinteresada de las más de 90
personas e instituciones que han colaborado en la recolección de muestras. Gracias a ellos no solo he
conseguido fuente de ADN para este trabajo, sino buenas amistades a lo largo y ancho del área de
distribución de la marta en toda Europa y parte de Rusia. Sin duda un gran mérito de esta tesis se os debe
a vosotros: Patricia Lizarraga, Laura Lorza y al resto del equipo de Martioda (CRF Martioda-DFA);
Servicio de vigilancia y SAP de la sección de Parques Naturales de la Diputación Foral de Álava (Juan
Carlos Ortíz, Ricardo Ortíz, María Elena García, Kepa García, Sonia Benítez, Elisabeth Cabanillas, Iker
Ayala, Arantza Puente, Jesús Gómez, Iñaki Martínez, Lidia Lacha, Jokin Sáez de Camara); Felipe Canales
y Miguel Ángel Campos (CRN); Javier Pinedo, Mario Corral y Asier Martínez (DFA); Enrique Arberas;
Oskar Berdión; Nerea Ruiz de Azua; Txema Fernández y Javier Sesma (IKT); Javier López de Luzuriaga;
Haizea Aguirre; Pablo Pérez; Alvaro Osés; Peio Lozano; Igor Aginako, Julio Ruiz Guijarro, Ignacio
Martínez, Eneko Díaz, Juanma Pérez de Ana, David Vado y Ander Eguia (DFB); Iñigo Zuberogoitia;
Fermín Urra (VRFN); Marta Barral (NEIKER); Amaya García de Albéniz; Asun Gómez; Aitor Valdeón; a
todo el equipo de ACCA; al personal de los parques Nacionales de Ordesa y Monte Perdido (Elena
Villagrasa; Fernando Carmena; Ramón Castillo; Alberto Iglesias; J. Argandoria; J. Pacheco; P. García;
Jesús Martín; Francisco Moratalla; Carlota Olivan; Blanca Nevot; Javier Estradera; Isabel Nerín, José
Antonio Escalonia; Paqui García), Aigüestortes i Estany de Saint Mauricy (Jordi Canut, Lluis Garzón,
Angel Monsó, Pep Gilabert) y Picos de Europa (Amparo Mora, Miguel Diez de Diego, Ángel Tejedor,
Miguel Ángel Bermejo, Cristóbal Chopitea, Felix Rojo, Marcelino Fernández ); Sociedad Galega de
Historia Natural (Manuel Arzúa, Catuxa Varela); Juan Fernando Berezo; Rubén Portas; Isabel Barja y su
equipo (Univ. Autónoma de Madrid); Alberto Fernández Gil (Univ. Oviedo, CSIC); Christian Gortázar
(IREC); Alfonso Hartasanchez (FAPAS); Gerardo Dominguez; Francesc López-Giraldez (DNA and Tissue
Collection, Univ. Pompeu Fabra); Pere Aymerich (PN Alt Pirineu); Francisco Purroy (Univ. León); Rita
Oliveira (CIBIO); BTVS-ICN; Licia Colli (Istituto di Zootecnica, Facoltà di Agraria Università Cattolica
del S. Cuore); Ettore Randi (Conservation Genetics Laboratory, ISPRA); Paolo Casula (Ente Foreste della
Sardegna, Direzione Generale); Adriano de Faveri (ISPRA); Alessandro Ballestrieri (Dpt. di Biologia
Animale, Università di Pavia); Museo Zoologico "la Specola" Sezione del MSN-Universita degli studio di
Firenze; Ana Galov (Dpt. of Animal Physiology, Faculty of Science of Zagreb); Angus Davison (Institute
of Genetics, School of Biology, Queen's Medical Centre, University of Nottingham); Geraldine Veron
(Muséum National d'Histoire Naturelle, Département Systématique et Evolution); Jan Herr (Dpt. of
Biology and Environmental Science, University of Sussex); Barbara Herzig (Mammal Collection, The
Natural History Museum Vienna); Hein van Grouw (National Museum of Natural History, Naturalis.
Netherlands); Herman Ansorge (Staatliches Museum fur Naturkunde Gorlitz. Germany); Franz Müller;
Natalia Martinkova (Institute of Vertebrate Biology, Academy of Science of the Czech Republic);
Malgorzata Pilot (Museum and Institute of Zoology, Polish Academy of Sciences); Robert Myslajek
(Mammal Research Institute, Poland); Maddis Podra (Institute of Mathematics and Natural Sciences,
Tallinn University); Gabor Csorba (Hungarian Natural History Museum); Peter Mortensen (Swedish
Museum of Natural History); Øystein Wiig (Mammal Collection, Natural History Museum, University
of Oslo); Dimitry V. Skumatov; Alexei Abramov (Laboratory of Mammalogy, Zoological Institute,
Russian Academy of Sciences).
De entre todos los que compartieron, comparten y seguirán compartiendo mi vida hay quienes sin duda
han vivido esta periodo de una forma más intensa. Por ello, esta tesis la deberías firmar también vosotros:
mi familia al completo, pero sobre todo mis Aitas y mi amona, por la fuerza, el cariño y el ánimo que me
habéis ofrecido siempre…mila esker bihotz bihotzez. A Josu y a Luis…a vosotros no hace falta que os diga
porque esta tesis también tiene algo vuestro. A Azula, por compartir las penas y las alegrías de embarcarse
en la siempre compleja tarea de hacer una tesis…Pero si hay alguien a quien quiero dedicarle esta tesis con
un cariño especial por el apoyo incondicional que me ha brindado esa eres tú… Mariajo. Has sido mi
consejera, mi guía, y mi luz al otro lado del túnel cuando parecía que esto no terminaba nunca. Gracias
por seguirme en este largo y a veces tortuoso camino que con tu ayuda siempre fue más fácil…
La presente tesis ha sido realizada gracias a la financiación procedente del programa de especialización
tecnológica de postgraduados en Materia de Biodiversidad (Dpto. de Ordenación del Territorio y Medio
ambiente, Dirección de Biodiversidad, Gobierno Vasco. 2005-2006) y el programa de becas FPI (Ref.
BFI06.396) del Dpto. de Educación, Universidades e Investigación del Gobierno Vasco durante el periodo
comprendido entre 2006-2010. Este trabajo ha sido financiado parcialmente por la Universidad del País
Vasco y el Dpto. de Medio Ambiente, Planificación Territorial, Agricultura y Pesca del Gobierno Vasco a
través de la empresa IKT SA, en el marco de la convocatoria de proyectos Universidad-Empresa (Ref.
UE0702). Así mismo, este estudio ha sido parcialmente financiado por el Dpto. de Educación,
Universidades e Investigación del Gobierno Vasco mediante la ayuda al grupo de investigación
“Sistemática, Biogeografía y Dinámica de Poblaciones” (Ref. IT317-10; GIC10/76).
contents CONTENTS
ABSTRACT.……...…………………………...………………………………………...……………i
RESUMEN…...………………...……………………………………..…...………………………..iii
LABURPENA…...……………...……………………………………..…...………………………...v
LIST OF PAPERS……....…………………………….………………...………………...……..…vii
PREFACE…………….…………………………..…………………………………………….…...ix
AIMS OF THE THESIS………...…………………..…...…………………………………...…….xi
CHAPTER 1. INTRODUCTION…………..…..…………………………………...……………..1
PAPER I: Martes conservation genetics: Using molecular genetics to assess within species movements
barriers and corridors………………..……………………………………………………...3
Abstract
Introduction
The role of conservation genetics
Genetic diversity and inbreeding
The impact of habitat fragmentation and connectivity on genetic diversity
Genetic monitoring
Molecular markers
Phylogenetics and DNA sequence data
Neutral and non-neutral markers (microsatellites, SNPs)
Phylogenetic inference within species
Martes population genetic studies
Landscape genetics
Future directions
Conclusions
CHAPTER 2. PHYLOGEOGRAPHY…………………………………..…………………………43
PAPER II: New insights into the cryptic northern glacial refugia: Phylogeography of the forest dwelling
European pine marten (Martes martes) …………………………………..………………...45
Abstract
Introduction
Material and Methods
Results
Discussion
CHAPTER 3. SPECIES IDENTIFICATION…………………………………………………..…87
PAPER III: A non-invasive genetic method to identify the sympatric mustelids pine marten (Martes
martes) and stone marten (Martes foina): preliminary distribution survey on the northern
Iberian Peninsula…..……..………………………………………...……………………89
Abstract
Introduction
Material and Methods
Results
Discussion
CHAPTER 4. RELIABLE FAECAL DNA GENOTYPING OF SYMPATRIC MARTENS.…...107
PAPER IV: Reliable faecal DNA genotyping of sympatric marten species (Martes martes and Martes foina):
The impact of sample collector field experience on species and individual identification
success rates…….………………………………………...………………….…………...109
Abstract
Introduction
Material and Methods
Results
Discussion
CHAPTER 5. LANDSCAPE GENETICS………………………………………………………..139
PAPER V: Non-invasive landscape genetics of the European pine marten (Martes martes): assessing spatial
genetic structure and distribution in a heterogeneous landscape …………….……………141
Abstract
Introduction
Material and Methods
Results
Discussion
PAPER VI: Landscape genetics as a tool for the empirical assessment of a regional ecological network: The
European pine marten (Martes martes) as a target-species……………………....……....…179
Abstract
Introduction
Material and Methods
Results
Discussion
CONCLUDING REMARKS….…….……………………..……………...…………………...…205
APPENDICES.…….………….……………………………...…………………………..……….207
APPENDIX A (Paper IIIa)……………………………………………………...209
Rosellini S, Osorio E, Ruiz-González A, Piñeiro A, Barja I (2008) Monitoring the small-scale
distribution of sympatric European pine martens (Martes martes) and stone martens (Martes
foina): a multievidence approach using faecal DNA analysis and camera-traps. Wildlife Research,
35: 434–440
APPENDIX B (Paper IIIb)…………………………………………………………..…219
Balestrieri A, Remonti L, Ruiz-González A, Gómez-Moliner BJ, Vergara M, Prigioni C (2010)
Range expansion of the pine marten (Martes martes) in an agricultural landscape matrix
(NWItaly). Mammalian Biology, 75: 412–419
APPENDIX C (Paper IIIc)…………………………………………………………..…229
Balestrieri A, Remonti L, Ruiz-González A, Capelli E, Vergara M, Gómez-Moliner BJ,
Prigioni C. Food habits of genetically identified pine marten (Martes martes) expanding in
agricultural lowlands (NW Italy). Acta Theriologica. In press
abstract ABSTRACT
Understanding the physical and temporal factors that structure populations is essential to their
conservation and management. The main objective of this thesis is to identify ancient and contemporary
processes shaping genetic variation of the forest dwelling European pine marten (Martes martes L. 1758),
trough two distinct disciplines: phylogeography (i.e combination of phylogenetics and biogeography) and
landscape genetics (i.e combination of population genetics and landscape ecology). Firstly, this thesis
reviews the intraspecific phylogenetic, population genetic, and landscape genetic studies conducted on the
8 recognized Martes species populations; discusses commonalities found across the species in terms of
habitats deemed important for connectivity; and identifies knowledge gaps for understanding movement
and substructure. Secondly, we investigated the phylogeographic pattern of the European pine marten,
providing new insights into the cryptic northern glacial refugia and postglacial recolonization of Europe as
well as into genetic relationships between M. martes and M. zibellina. Thirdly, in order to understand how
landscape and environmental features influence population genetic structure, non-invasive genetic
sampling (i.e sampling genetic material without disturbing the target species) of faeces in combination
with landscape genetics methods have been used for the study of this rare and elusive species. For the
consecution of this objective we initially designed and applied a non-invasive species identification method
to asses the distribution of sympatric martens from fecal DNA (M. martes and M. foina) and thereafter, by
means of microsatellite analysis we investigated the spatial genetic structure of the pine marten inhabiting
a fragmented landscape. Bayesian spatial and non-spatial models allowed us to detect several populations
with genetic discontinuities being associated to human induced landscape transformations (i.e. urban
areas, road networks and forest fragmentation processes). Finally, we evaluated the influence of several
landscape features on pine marten gene flow within a regional ecological network (Basque country, Spain)
through individual-based least-cost distance GIS analysis. Our results confirmed the validity of the
designed corridors and the need to keep taking them into account in land use planning. Besides, we
identified areas of intensive agriculture and the major road networks as landscape features which limit pine
marten gene flow. To the best of our knowledge, this is the first time that a GIS modeled ecological
network has been evaluated through a landscape genetics approach.
Keywords: Phylogeography, landscape genetics; Non-invasive genetic sampling; Martes martes
‐ i ‐ ‐ ii ‐ Resumen RESUMEN
El estudio de los factores físicos y temporales que influyen en los procesos de estructuración poblacional
resulta esencial para una adecuada gestión y conservación de las especies. El principal objetivo de esta tesis
es identificar los procesos históricos y contemporáneos que han determinado la distribución actual de la
variación genética de una especie eminentemente forestal como la marta europea (Martes martes L. 1758) a
través de dos disciplinas: la filogeografía (i.e. la combinación de la filogenia y la biogeografía) y la genética
del paisaje (i.e. la combinación de la genética de poblaciones y la ecología del paisaje). En primer lugar,
esta tesis hace una revisión exhaustiva de los diferentes estudios llevados a cabo en las 8 especies del género
Martes en lo concerniente a las filogenias intra-específicas, la genética de poblaciones y la genética del
paisaje; discute los patrones generales detectados en las diferentes especies en relación a los hábitats
considerados de especial relevancia para la conectividad; e identifica las principales lagunas de
conocimiento en lo que respecta a la comprensión de los patrones dispersivos y la sub-estructuración
poblacional. En segundo lugar, investigamos el patrón filogeográfico de la marta europea aportando nuevas
evidencias sobre la existencia de refugios crípticos más allá de las Penínsulas mediterráneas; sobre los
procesos de recolonización post-glacial de Europa así como acerca de las relaciones genéticas entre M.
martes y M. zibellina. En tercer lugar, la aplicación de técnicas moleculares en muestras no–invasivas (i.e la
obtención de material genético sin necesidad de capturar o interferir con la especie objetivo) en
combinación con métodos de genética del paisaje nos permitió el estudio de una especie rara y elusiva con
el objetivo de comprender los factores ambientales y territoriales que influyen sobre su distribución y sobre
la estructuración de sus poblaciones. Para ello, se desarrolló inicialmente un método genético no-invasivo
que permitió determinar la distribución de las especies simpáticas del género Martes (M. martes y M. foina)
a partir del ADN extraído de muestras fecales y, posteriormente, investigar mediante el uso de marcadores
microsatélites la estructuración genética espacial de la marta en un ambiente fragmentado. Los modelos
Bayesianos espaciales y no-espaciales nos permitieron detectar diferentes poblaciones cuyas
discontinuidades genéticas aparecen asociadas a transformaciones paisajísticas de carácter antropogénico
(i.e. áreas urbanas, infraestructuras viarias, y procesos de fragmentación forestal). Por último, evaluamos la
influencia de diferentes variables paisajísticas sobre el flujo génico de la marta en el marco de una red
ecológica regional (País Vasco, España) mediante la utilización de distancias de mínimo-coste y análisis
GIS. Nuestros resultados confirmaron la validez de los corredores diseñados y por tanto, la necesidad de
seguir considerándolos como herramientas útiles en la planificación territorial. Además, identificamos que
‐ iii ‐ los elementos paisajísticos que limitan mayoritariamente el flujo génico de la marta son las áreas de
agricultura intensiva, las principales infraestructuras viarias así como las zonas urbanizadas. Este trabajo es
pionero en la evaluación a través de un estudio basado en genética del pasaje, de una red ecológica
modelizada mediante GIS.
Palabras clave: Filogeografía, genética del paisaje, técnicas moleculares no-invasivas, Martes martes
‐ iv ‐ Laburpena LABURPENA
Espezieen kontserbazio eta kudeaketa egokirako, populazioen egituran eragiten duten eragile fisiko eta
denborazkoak ikertzea ezinbestekoa da. Tesi honen helburu nagusia, basoko espeziea den lepahoriaren
(Martes martes L. 1758) gaur egungo aniztasun genetikoaren banaketan eragin duten prozesu historiko eta
garaikideak identifikatzea da bi diziplina ezberdinen bidez: filogeografia (i.e. filogenia eta biogeografiaren
konbinazioa) eta paisaiaren genetika (i.e. populazioen genetika eta paisaiaren ekologiaren konbinazioa).
Lehenik eta behin, tesi honek Martes generoko 8 espezieen filogenia intraespezifikoa, populazioen genetika
eta paisaiaren genetika arloetan gauzatu diren ikerketak berrikusi; espezie ezberdinetan konektibitaterako
garrantzia berezia duten habitatetan antzeman diren patroi orokorrak eztabaidatu eta sakabanatze patroi
eta populazioen barne egiturei dagokien ulermen hutsune nagusiak identifikatzen ditu. Bestalde, lepahori
europarraren patroi filogeografikoa ikertuz, mediterranear Penintsulakoez gain babesleku kriptikoen
existentziaren ebidentziak, Europako glaziazio ondorengo birkolonizazio prozesuak zein M. martes eta M.
zibellina arteko erlazio genetikoei buruzko ekarpenak egin ditugu. Teknika molekular ez inbaditzaileen
(i.e. material genetikoa eskuratu espeziea harrapatu edo oztopatu gabe) eta paisaiaren genetika metodoen
konbinazioaren erabilera, espezie arraro eta sahieskorren ikerketa egokiak burutzeko eraginkorra da.
Teknika hauek, populazioen egituraketan eragiten duten ingurune eta lurralde eragileak ulertzea
ahalbidetzen dute. Ondorioz, helburu hau lortzeko lehenik Martes generoko (M. martes eta M. foina)
espezie sinpatrikoen banaketa zehazten duen metodo genetiko ez inbaditzailea garatu zen. Ondoren,
mikrosatelite markatzaileen bidez, lepahoriak ingurune zatikatuan duen egitura genetiko espaziala ikertu
zen. Modelo Bayesiano espazial eta ez espazialek, populazio desberdinetan etenaldi genetikoa izaera
antropogenikoa duen paisaiaren eraldaketarekin (i.e. bide azpiegiturak eta basoko zatiketa prozesuak)
loturik dagoela antzematea ahalbidetu dute. Azkenik, eskualde mailako sare ekologiko batean (Euskal
Autonomia Erkidegoa, Espainia) eta distantzia kostu minimoak eta SIG azterketa erabiliz, lepahoriaren
fluxu genikoan aldagai paisajistiko ezberdinek duten eragina aztertu genuen. Gure emaitzek, diseinatutako
korridoreen balioa berretsi eta lurralde planifikazioetan tresna baliagarri moduan aintzakotzat hartzen
jarraitzea beharrezkoa dela baieztatzen dute. Horretaz gain, nekazaritza intentsiboa, bide azpiegitura
nagusiak eta gune urbanizatuak, lepahoriaren fluxu genikoa mugatzen duten elementu paisajistiko
nagusiak direla identifikatu genuen. Lan hau, paisaiaren genetikan oinarritutako ikerketa baten bidez SIG
bitartez modelizatutako sare ekologiko baten azterketan aitzindaria da.
Gako-Hitzak: Filogeografia, paisaiaren genetika, teknika molekular ez inbaditzaileak, Martes martes
‐ v ‐ ‐ vi ‐ This thesis is based on the following papers which will be referred to by their Roman numerals:
I
Schwartz M, Ruiz-González A, Pertoldi C, Masuda, R. Martes Conservation Genetics: Using
Molecular Genetics to Assess Within Species Movements, Barriers and Corridors. In: Biology
and Conservation of Marten, Sables, and Fisher: a new synthesis. Aubry et al. Eds. Cornell
University Press. Submitted Manuscript *
II
Ruiz-González A, Madeira MJ, Randi E, Abramov A, Gómez Moliner BJ. New insights into
the cryptic northern glacial refugia: Phylogeography of the forest dwelling European pine
marten (Martes martes). Manuscript
III
Ruiz-González A, Rubines J, Berdión O, Gómez-Moliner BJ (2008) A non-invasive genetic
method to identify the sympatric mustelids pine marten (Martes martes) and stone marten
(Martes foina): preliminary distribution survey on the northern Iberian Peninsula. European
Journal of Wildlife Research, 54 (2): 253-261
IIIa
Rosellini S, Osorio E, Ruiz-González A, Piñeiro A, Barja I (2008) Monitoring the small-scale
distribution of sympatric European pine martens (Martes martes) and stone martens (Martes
foina): a multievidence approach using faecal DNA analysis and camera-traps. Wildlife Research,
35: 434–440.*
IIIb
Balestrieri A, Remonti L, Ruiz-González A, Gómez-Moliner BJ, Vergara M, Prigioni C (2010)
Range expansion of the pine marten (Martes martes) in an agricultural landscape matrix
(NWItaly). Mammalian Biology, 75: 412–419*
IIIc
Balestrieri A, Remonti L, Ruiz-González A, Capelli E, Vergara M, Gómez-Moliner BJ,
Prigioni C. Food habits of genetically identified pine marten (Martes martes) expanding in
agricultural lowlands (NW Italy). Acta Theriologica. In press*
IV
Ruiz-González A, Madeira MJ, Randi E, Gómez-Moliner BJ. Reliable faecal DNA genotyping
of sympatric marten species (Martes martes and Martes foina): The impact of sample collector
field experience on species and individual identification success rates. Manuscript
V
Ruiz-González A, Madeira MJ, Randi E, Gómez-Moliner BJ. Non-invasive landscape genetics
of the European pine marten (Martes martes): assessing spatial genetic structure and distribution
in a heterogeneous landscape. Manuscript
VI
Ruiz-González A, Gurrutxaga M, Madeira MJ, Randi E, Gómez-Moliner BJ. Landscape
genetics as a tool for the empirical assessment of a regional ecological network: The European
pine marten (Martes martes) as a target-species. Manuscript
* The introduction of this thesis is based on paper I. Papers IIIa, IIIb and IIIc are derived
contributions from the application of the developed species identification method (Paper III)
and are included as Appendix.
‐ vii ‐ List of
Papers LIST OF PAPERS
‐ viii ‐ preface PREFACE
I subdivide this thesis into six main papers and three additional ones, which address different aspects of the
European pine marten (Martes martes) concerning the wide research topics of phylogeography, landscape
genetics and non-invasive genetics.
The general framework of this thesis is given by the introduction (Paper I), where the intraspecific
phylogenetic, population genetic, and landscape genetic studies conducted on Martes populations are
reviewed; discussed commonalities found across the species in terms of habitats deemed important for
connectivity; and identified knowledge gaps for understanding movement and substructure of the 8
species of the genus Martes.
In chapter 2 (Phylogeography), we investigate the phylogeographic patterns of the pine marten throughout
the current species’ European distribution. In this chapter we provide new insights into the cryptic
northern glacial refugia and postglacial recolonization of Europe as well as into genetic relationships
between M. martes and M. zibellina (Paper II). We found a complex phylogeographic history for M.
martes indicating a mixed pattern of recolonization of northern Europe from both Mediterranean and
non-Mediterranean refugia. Each of the inferred phylogroups showed a clear correlation to specific
biogeographic regions which could probably represent different ecotypes.
Later, we address different aspects of analytical advances with non-invasive genetics for species (Chapter 3)
and individual identification (Chapter 4) of sympatric martens (M. martes and M. foina). In paper III, we
describe a reliable non-invasive Polymerase Chain Reaction–Restriction Fragment Length Polymorphism
(PCR-RFLP) method for distinguishing between M. martes and M. foina based on DNA extracted from
faeces samples. The efficiency of this technique was evaluated through a preliminary field study across the
potential sympatric distribution areas of both marten species in the northern Iberian Peninsula. Moreover,
this method has been used to: i) monitor the small-scale distribution of sympatric martens in combination
with camera-traps (Paper IIIa); to asses the range expansion of the pine marten in an agricultural landscape
matrix (NW-Italy) (Paper IIIb); and to efficiently investigate pine marten food habits (Paper IIIc).
Thereafter, we describe a reliable multiplex panel for faecal DNA genotyping of sympatric marten species
(M. martes and M. foina) (Paper IV). The application of this panel comprising 15 microsatellite markers
‐ ix ‐ facilitated species distinction discarding the presence of putative hybrids between both species. Moreover,
we asses the impact of sample collectors field experience on species and individual identification success
rates.
These two non-invasive researches (Paper III and Paper IV) set the frame for the following papers, treating
the landscape genetic approaches of pine marten in order to understand how landscape and environmental
features influence population genetic structure (Chapter 5). In paper V, we analyze the distribution
patterns of sympatric martens (M. martes and M. foina) and the spatial genetic structure of European pine
marten on a heterogeneous landscape through non-invasive genetic sampling and Bayesian analyses.
Spatial and non-spatial Bayesian models allow us to detect several populations with genetic discontinuities
being associated to human induced landscape transformations (i.e. roads, forest fragmentation).
In the final paper (paper VI), we focus on individual-based analyses predicting the genetic differences
among all pairs of individuals based on the least cost distances between them as functions of multiple
landscape resistance hypotheses: i) the resistance map which was drawn up in the design of the regional
Ecological Network in the Basque Country (North Spain) and ii) different binary resistance maps which
covered a gradient from greater to lesser preference of the focal species in relation to forest environments. For each of these papers we provide an abstract and introduction, describe methods and present and
discuss the results. The concluding remarks chapter (Chapter 5) summarises the major findings of the
previous chapters and points out consequences of these findings for pine marten conservation and
management.
‐ x ‐ The main objective of this thesis is to apply different molecular tools to study the phylogeography, genetic
variability, distribution and spatial genetic structure of the European pine marten (Martes martes) in order
to identify ancient and contemporary processes shaping genetic variation. More specifically, the aims have
been:
1. Review the intraspecific phylogenetic, population genetic, and landscape genetic studies conducted
worldwide on the 8 recognized Martes species.
2. Use a phylogeographic approach in order to assess the impact of Quaternary glaciations on the current
genetic structure of the European pine marten and to identify location of refugia as well as post-glacial
recolonization routes after the Last Glacial Maximum.
3. Develop a reliable non-invasive Polymerase Chain Reaction–Restriction Fragment Length
Polymorphism (PCR-RFLP) method for distinguishing between M. martes and M. foina based on
mtDNA extracted from faecal samples.
3.1 Assess the field suitability of the PCR-RFLP technique, through a preliminary study
across the potential distribution area of both marten species in the northern Iberian
Peninsula.
3.2 Apply this protocol for improving our knowledge of different bio-ecological traits of
sympatric marten species.
4. Develop a reliable and accurate multiplex panel of 15 microsatellites for individual identification of M.
martes and M. foina from non-invasively collected faecal samples.
4.1 Verify possible hybridization events between both marten species in sympatric areas.
4.2 Assess the impact of sample collector field experience on species and individual
identification success rates.
5. Develop a non-invasive landscape genetic approach in order to understand how landscape and
environmental features influence population genetic structure.
‐ xi ‐ Aims of the
Thesis AIMS OF THE THESIS 5.1 Infer the spatial genetic structure of European pine marten using 15 highly variable
genetic markers and spatial and non- spatial Bayesian clustering analyses (GENELAND and
STRUCTURE).
5.2 Develop an individual-based non-invasive landscape genetic approach in order to
evaluate alternative resistance maps relating landscape structure to gene flow. The resistance
maps tested were: i) different binary resistance maps which covered a gradient from greater to
lesser preference of the focal species in relation to forest environments and ii) The resistance
map used to design the regional ecological network of the Basque Country (North Spain).
‐ xii ‐ CHAPTER
1
INTRODUCTION
PAPER I
Martes conservation genetics: Using molecular genetics to assess
within species movements barriers and corridors
-1-
1
CHAPTER 1
Introduction
-2-
Martes conservation genetics
Paper I
PAPER I
Martes conservation genetics: Using molecular genetics to assess
within species movements barriers and corridors
ABSTRACT
Understanding the physical and temporal factors that structure Martes populations is essential to the
conservation and management of the 8 recognized Martes species. Recently, advances in three distinct
subdisciplines in molecular ecology have provided insight into historical and contemporary environmental
factors that have created population substructure and influenced movement patterns of several of the
Martes species. Intraspecific phylogenetics has allowed us to understand the role of large scale historical
events such as the last glacial maxima and their associated refugia on Martes populations in Europe, Asia
and North America in at least five species (M. americana, M. martes, M. melapus, M. pennanti, M.
ziberllina). In addition, population genetics has examined how connected Martes populations are within
species across space, and in some cases how this level of connectivity has changed over recent time by
examining historical samples in multiple populations. These studies have been conducted on M.
americana, M. martes, and M. pennanti. More recently several landscape genetic analyses, including graph
theoretic and least cost path approaches, have been used to evaluate the correlation between landscape
features and genetic relatedness among individuals (within a species) in a landscape.
These new
approaches are showing promising results in empirically evaluating multiple habitat features at multiple
scales that foster or hinder connectivity. Different forms of this approach have been applied on M.
americana, M. martes, and M. pennanti in portions of their range. This paper reviews the intraspecific
phylogenetic, population genetic, and landscape genetic studies conducted on Martes populations;
discusses commonalities found across the species in terms of habitats deemed important for connectivity;
and identifies knowledge gaps for understanding movement and substructure of the 8 Martes species.
-3-
CHAPTER 1
Introduction
INTRODUCTION
Understanding the biotic and abiotic forces that influence the movement of wildlife has been a central
focus of wildlife management for nearly a century. This topic has come to the forefront of wildlife biology
in recent years as the perils of habitat fragmentation and climate change are becoming clearer and more
pronounced. In many areas lands that were once abundant and connected are now becoming small,
degraded or completely isolated. In fact, we can consider these habitat changes on a gradient from changes
that completely eliminate movement potentials of animals to those changes that marginally limit the
probability of a successful dispersal of an individual. For example, urbanization of a once forested area may
act as a complete barrier to movement, isolating populations, whereas various forest thinning treatments
may remove the necessary cover for mid-sized and small carnivores to disperse, thus exposing them to
predation risks that they did not historically face in unmanaged forests.
Some habitat changes are tied to natural cycles occurring on a temporal scale of centuries or millennia,
while others are either functions of short term natural changes or are anthropogenically linked and occur
on time scales of one or several generations. Understanding if the contemporary distribution and
substructure patterns of animals are a function of long-term influences, such as the last glacial maximum,
or more recent landscape uses is critical for managing and conserving wildlife, as this will allow us to infer
if putative barriers are historic and out of management control or contemporary and possibly human
induced. From these understandings, we will also be able to infer if our management actions, such as
corridor protection or habitat improvement will likely increase animal movement and gene flow across
their range.
Fortunately, recent advances in molecular ecology are now providing tools that can examine connectivity,
and its converse genetic substructure, at multiple time scales. Typically intraspecific phylogenetic methods
that examine sequence data (Avise et al. 1987) have been used for evaluating events that occur over longer
time scales, and population genetic methods that use highly variable nuclear DNA markers have been used
for evaluating contemporary patterns. One of the most exciting recent contributions is the use landscape
genetic (Manel et al. 2003) data, which combines landscape ecology approaches and population genetics
data, to disentangle the roles of historical and contemporary forces in structuring existing populations. For
example, in recent work on California mule deer (Odocoileus hemionus) Pease et al. (2009) used landscape
genetic analyses in conjunction with environmental niche modeling (ENM) to discern that clustering of
California mule deer are largely a function of current ecological factors (topography, temperature,
precipitation, seasonality, and plant community) and not from historical factors present since the last
-4-
Martes conservation genetics
Paper I
glacial maximum. Armed with this understanding managers can now try to change the environmental and
habitat factors that are limiting deer distributions and movement.
In this review, we first provide a basic primer on molecular ecology methods and tools. Subsequently, we
discuss recent finding on movement and substructure based on intra-specific phylogenetic information.
We do not cover newer interspecific phylogenetic information that provides understandings of the
evolutionary history among Martes species, as this is well covered in chapter X (Koepfli et al. 2008, Koepfli
et al. This Volume). Next, we review the published population genetic studies from three Martes species
(M. americana, M. martes, and M. pennanti; Table 1) in various parts of each species geographic
distribution. Lastly, we summarize the landscape genetics literature on 3 Martes species (M. americana, M.
martes, and M. pennanti). We also discuss the latest methods developed in this field, provide a critique of
where we think the major advances will occur (e.g. the use of more complex landscape hypotheses, and
more advanced molecular markers and models) which will lead to better understandings on how to
manage Martes species in light of climate change.
-5-
CHAPTER 1
Introduction
Intraspecific Phylogenetic
Population Genetic
Landscape
Genetic
Martes americana
Carr and Hicks 1997; Stone and
Broquet 2006, Kyle 2000, Kyle and Strobeck
Broquet
Cook 2002, Stone et al. 2002;
2003,
Wasserman 2008
Slauson et al. 2008, Dawson, 2010
Swanson 2006, 2007, Williams 2007
Davison 2001, Pertoldi 2008, Ruiz-
Kyle 2003, Mergey 2007, Pertoldi 2008
Mergey 2007
Williams et al. 2000, Drew 2003,
Kyle 2001, Carr 2007, Rella-Hapeman 2006,
Carr
Vinkey 2006, Schwartz 2007, Knaus
Schwartz
Garroway 2008,
et al. In Review.
Williams 1999, Williams 2000
Hosoda et al. 1999, Kurose et al.
Petrovskaya 2007
McGowan
1999,
Small
2003,
2006,
Martes flavigula
Martes foina
Martes gwatkinsi
Martes martes
González et al. 2009
Martes melampus
Hosoda et al. 1999, Kurose et al.
1999, Murakami et al. 2004, Sato et
al. 2009, Inoue et al. 2010
Martes pennanti
Martes zibellina
1999,
Murakami
et
al.
unpublished,
Wisely
2004,
2007,
2004,
Malyarchuk et al. 2010, Inoue et al.
2010
Table 1. A list of interaspecific phylogenetic, population genetic or landscape genetic studies conducted on Martes populations, many
reviewed in this mansucript. There has been no studies conducted to date on M. flavigula, M. foina, or M. gwatkinsi, and no detailed
population or landscape genetic studies conducted on M. melampus or M. zibellina. The table is color coded to denote areas where there
has been ample work (green), some work conducted in limited geographic areas (yellow), and no work conducted (red).
THE ROLE OF CONSERVATION GENETICS
Genetic Diversity and Inbreeding
During the last two decades, the role of genetics in conservation biology, and in ecology in general, has
been greatly emphasized (see for reviews: Frankham 1995, 2005, Allendorf and Luikart 2007, Pertoldi et
al. 2007). The conservation genetics field has historically been subject to fierce criticisms from population
ecologists arguing that genetic variation is not really relevant when a population faces extinction due to
demographic problems. Understanding the consequences of demographic stochasticity in populations
-6-
Martes conservation genetics
Paper I
requires detailed knowledge of local fluctuations in population size, extinction probability, colonization
potential as well as reproductive success, which can be gained from population dynamics analyses. The
difficulties associated with the collection of the demographic data necessary for such analyses make them
extremely time consuming. Environmental factors and their changes are mirrored in the genetic
composition of affected populations. Therefore, as DNA analyses are increasingly used to estimate and
infer the causes of spatio-temporal dynamics, conservation geneticists and ecologists are reconciling. Even
small alterations of environmental conditions can affect the genetic composition of the populations, both
via demographic and selective responses. In many cases, changes in the genetic composition of populations
can be more readily detected than the concomitant demographic changes (Tallmon et al. 2010).
Measuring the health of a population is an extremely difficult task, especially in elusive and/or nocturnal
species like most of the Martes species are, there is therefore a need for a unit of measurement, which can
help to quantify, not only in a qualitative way, the status of endangerment of a given population, species or
group of species. Help comes from conservation genetics, which is likely to play a key role in developing a
strategy for both the short and long-term preservation of genetic diversity. The assessment of genetic
diversity in endangered animal is now pervasive. This phenomenon is due to acquisitions of powerful
methods for DNA analyses, which are being increasingly used to infer the causes of the spatio-temporal
dynamics of the populations as well as to estimate the extent of genetic diversity within populations and
the organization of genetic diversity between populations (Allendorf and Luikart 2007).
Genetic diversity can be defined in several ways. First genetic diversity can be quantified by molecular
genetic methods and be expressed as proportion of polymorphic loci, proportion of heterozygous loci and
number of alleles in these loci and the allelic frequencies. An implicit assumption often made in
conservation genetics is a causal relationship between genetic variability and the evolutionary
precariousness of a species, both in the short-term and in the long-term (Ouborg et al. 2010a,b). Expected
heterozygosity (HE) can provide an indication of the immediate evolutionary potential of the population,
although this measure has no deterministic relationship to its future value. Given an initial pool of
unrelated founder genes, the potential changes and losses of genetic diversity can be assessed by the
increase in relatedness. The initial genetic variability of a population is reduced as relatedness between
individuals increases. The decline is proportional to the reduction in heterozygosity or increase of average
inbreeding coefficient (F) of the parents, caused by inbreeding and genetic drift. The degree of inbreeding
within an isolated population can be quantified by a genetic parameter called FIS which vary from -1 to 1
with 0 representing non-inbred population in Hardy Weinberg equilibrium (HWE). FIS equal to 1 is
caused when observed heterozygosity (HO) is smaller than HE which is the level of heterozygosity that
-7-
CHAPTER 1
Introduction
would be observed if the population is in HWE. Inbreeding depression (the reduction in a fitness trait due
to inbreeding) may arise due to mating between close relatives that increases homozygosity, and is most
commonly observed in species that normally outcross.
Contrary, increased genetic divergence between the parents may be an advantage increasing heterozygosity
of the offspring, which could cause a “heterosis” effect, or “hybrid vigor effect” but only until the genetic
distance between the parents reaches a limit. Beyond this limit the divergence and differences in coadaptation between the parents may reduce fitness in the offspring due to outbreeding depression
(Tallmon et al. 2005). Therefore, genetic divergence between the parents of an individual can be viewed as
a continuum with varying fitness consequences for the offspring. Inbreeding is of major interest in
conservation genetics, because its increase in small populations is believed to influence the probability of
extinction, both in the short-term because of inbreeding depression and the long-term because of loss of
variability. Reports on populations which have suffered severe bottlenecks, but nevertheless prosper
currently, challenge the assumption that inbreeding depression is a severe threat for the survival of
populations. Evidently, this ambiguity could also arise from the difficulties of detecting inbreeding
depression effects under natural conditions, from the fact that we are unable to monitor the fate of inbred
populations on evolutionary time-scales, or due to the fact that inbreeding depression is only expressed
under stressful environmental conditions.
An additional form of genetic variation is quantitative genetic variation, which is correlated to the adaptive
potential, or the capacity of a population to adapt in an evolutionary way altering the allelic frequency in
their genetic pool. Conservation geneticists try to estimate the quantitative genetic variation by means of
neutral markers but the correlation between neutral and quantitative variation is weak and therefore
divergence among populations at neutral loci is potentially uninformative, as it cannot exclude local
adaptations (Lynch 1996).
The impact of habitat fragmentation and connectivity on genetic diversity
Several factors are responsible for population decline; however the principal factors are the anthropogenic
ones, such as land development and urbanization that are the primary causes of extinction. These primary
anthropogenic factors have ramifying ecological and genetic effects, which contribute to extinction risk.
The role of ecological factors, such as metapopulation dynamics, are of fundamental importance for the
persistence of populations, bearing in mind that land development causes habitat fragmentation and
isolation of small populations, and intensification of metapopulation dynamics. Habitat fragmentation is a
major factor limiting the distribution of species in man-altered landscapes and it has always been
-8-
Martes conservation genetics
Paper I
considered to have deleterious consequences for the organisms and may be perceived differently by
different species, depending on their ability to overcome introduced barriers. For some species, roads have
been shown to act as a considerable barrier to dispersal, especially in Martes species where the main impact
of roads consists in increased disturbance or mortality which can reduce local populations’ size and may
lead to local extinction. For example, American marten tend to avoid roads (Robitaille and Aubry 2000)
and are sensitive to habitat fragmentation. Forsey and Baggs (2001) suggested that tracts of treeless land
may act as complete barriers to marten if greater than 5 km in length dispersal (Gibilisco 1994; See also
Thompson and Fryxell This Volume).
Many examples exist of species and populations in danger of going extinct or losing significant proportions
of their genetic variability due to a restriction of their habitat. Typical cases involve species which were
previously abundant and distributed over large geographical areas but are now found in only a few, small
isolated populations. Many members of the genus Martes and family Mustelidae have experienced such a
fate and specific studies conducted on several species will be discussed below.
Generally, small fragmented populations are genetically depauperated. This loss of genetic variability has
two potential consequences: (a) Low genetic variability can be a threat in the long-term for adapting and
evolving under changing environmental conditions and in disturbed habitats and (b) Small fragmented
and isolated populations can suffer from inbreeding depression due to increasing relatedness between
individuals (i.e., inbreeding). One of the most common practical conservation strategies to offset concerns
regarding inbreeding is to increase the level of connectivity, and thus gene-flow between populations as the
combination of small population size and isolation may lower the fitness of a population because of
reduced genetic diversity from drift and inbreeding. However, too high levels of gene flow can either
reduce or impede the capacity of adaptation to a stressor or they can alternatively introduce essential new
genes for future adaptation. In a population, the actual degree of adaptation is the effect of the dynamic
interaction between the selective pressure acting on the population and gene flow. Gene flow among
populations can be studied using an evolutionary time frame or as current gene flow.
Changes in gene flow can be estimated by comparing ‘historical‘ estimates based on genetic differentiation
(using FST or other related substructure measures which range from 0, or no genetic differentiation
between populations, to 1, complete genetic differentiation or no gene-flow) to current estimates based on
assignment tests which allows the detection of migrants.
Alternatively, evolutionary questions concerning the role of gene flow for long-term genetic diversity,
population differentiation, species identity, and speciation, emphasize the evolutionary time scale and is
-9-
CHAPTER 1
Introduction
usually estimated with phylogenetic analysis of sequence data. One of the main goals of conservation
genetics is the identification of evolutionary significant units (ESU) and the preservation of genetic
diversity, which should allow the evolutionary processes of natural selection and adaptation to continue in
the future. The identification of separate ESU currently requires significant divergence of allele frequencies
at nuclear loci and reciprocal monophyly of mtDNA. However this definition of ESU is only based on
neutral genetic markers, whereas a broader definition including non-neutral markers would be more
appropriate. New statistical methods, which seek to identify the number of ‘populations’ in a group of
samples and/or assign individuals to population of origin, are being widely applied (see for example Carr et
al. 2007a).
Genetic Monitoring
Recently, there have been many efforts that use molecular markers to monitor wild populations of fish,
wildlife and plants (Boulanger et al. 2004, Schwartz et al. 2007). These methods either use diagnostic
molecular markers to infer individuals, population and species and monitor changes in estimated
parameters such as abundance using traditional wildlife biology tools, or monitor changes in population
genetic metrics (McComb et al. 2010). These approaches may be highly useful for the study of Martes
given the secretive nature of these species. Deciding, the best strategy for monitoring will depend largely
on the number of markers available for the species of interest and the power associated with various
metrics. Tallmon et al. (2010) has recently shown that monitoring the population genetic metrics, such as
effective population size, may be as powerful, or potentially more powerful, than monitoring change in
abundance for detecting abundance and trends in simple populations.
One common population genetic metric to monitor is the effective population size (NE). NE has been
considered as the most important and critical surrogate parameter to describe the status of small
populations and is based on relatedness within and among individuals and populations. Populations with a
NE smaller than few hundred individuals are nearly independent of the strength of selection and are therefore
governed by mutation-drift regimes. With small NE genetic drift and inbreeding will affect both target traits
and neutral alleles, while with large NE selection will only affect target and linked genes. Hence, small NE
make the populations unable to adapt in an evolutionary way to an environmental change. Populations
with large NE have the potentiality to react to the selective pressures generated by environmental changes if
their genetic variability is high enough and if the speed of the environmental changes is not to high (if the
rate of adaptive evolution at least matches the rate of environmental change). Therefore, the NE of a
population can predict a population’s capacity to survive in changing environment more reliably than the
- 10 -
Martes conservation genetics
Paper I
census size and/or the amount of genetic variability, furthermore it determines the speed at which the
genetic variability is lost (Schwartz et al. 1999, Luikart et al. 2010). The notion of NE can therefore be
viewed as a bridging point between ecology and genetics with, the ecological characteristics, including
aspects of life history, social structure and population dynamics being put on the focus for determining NE
and the rate of loss of genetic variation.
To predict the long-term persistence of animal populations, accurate estimates of population size are also
necessary, thus abundance is also a commonly monitored metric. Census methods based on direct counts
may be inaccurate because individuals are difficult to detect. Furthermore, statistical tools to estimate
population size based on trapping methods often depend on unrealistic assumptions. New molecular
techniques for the analysis of faeces or hairs typed for diagnostic markers may be of help to count
individuals in a population by determining the number of unique multilocus genotypes in the population
(Luikart et al. 2010).
This possibility has created a relatively new discipline: non-invasive genetics which is a set of field,
laboratory and analytical techniques that allow studying the biology of natural populations, without
observing or capturing of individuals (Broquet et al. 2007, Long and Mackay et al. This Volume).
Microsatellites are the molecular marker of choice, used to identify species and to detect individual
genotypes from non-invasive samples (Broquet et al. 2007). However, low amplification rates and
genotyping errors due to non-invasive DNA degradation, risk to generate false genotypes, which do not
correspond to any extant individual (Bonin et al. 2004). To improve the genotyping success and
reliability, it is preferable to amplify DNA fragments as short as possible. The use of single nucleotide
polymorphisms (SNPs) requires only amplification of very short fragments and this makes them
particularly attractive for non-invasive genetic monitoring projects (Seddon et al. 2005). Using SNPs
requires potentially high-throughput to be compared with microsatellite markers, which makes them
particularly promising. The main obstacles in the use of SNPs remains the difficulty of their identification
in non-model organisms (Smith et al. 2004, Ryynänen et al. 2007) and their uncertain performance on
non-invasive genetic samples of lower quality.
- 11 -
CHAPTER 1
Introduction
MOLECULAR MARKERS
Phylogenetics and DNA sequence data
Most accounts of intraspecific phylogenetics trace the roots of mitochondrial DNA (mtDNA) because of
its properties such as maternal transmission, extensive intraspecific variation, general lack of interspecific
variation, and usual absence of genetic recombination. Intraspecific studies conducted to date on Martes
sp. thus involved analysis of mtDNA sequences mainly or exclusively (Table 2). mtDNA has a relatively
fast rate of nucleotide divergence, well suited to examining events occurred over the last few million years.
Two mitochondrial molecular markers have been used mainly on intraspecific phylogenetic studies of
Martes species: the control region (D-loop) and the cytochrome b (cyt b). The sequence length is usually
short (300-500 bp). Several studies used both markers to obtain more resolution while others are
complemented with RFLPs or with additional mtDNA markers such as the NADH dehydrogenase
subunit 2 gene, several tRNA (Thr & Pro) or with internal spacer regions of the nuclear ribosomal DNA
(rDNA) (see Table 2 for detailed information). Several studies focusing on the intraspecific phylogenetics
have not detected clear structuring (Hosoda et al. 1999, 2000; Kurose et al. 1999, Davison et al. 2001),
possibly due to insufficient information, because most of the studies used a small fragment of mtDNA
(Cyt b or D-loop) which did not provide enough resolution to resolve intraspecific patterns. Appropriate
efforts in the conservation of Martes, therefore, require that genetic diversity should be analyzed more
comprehensively, with different markers and/or with sufficient DNA sequence length, so that conservation
units can be defined in terms of geographical distribution of the obtained lineages. Each DNA sequence
has its own genealogy and they may evolve at different rates. Furthermore, the various methods of analysis
probe different aspects of the molecular and spatial history. Consequently, to reconstruct a species
phylogeographic history one would ideally like to use a range of sequences (including nuclear, cytoplasmic,
sex linked, autosomal, conserved, neutral, high and low mutation rate DNA fragments) and apply a suite
of pertinent data analyses (Sato et al. 2009; Balloux 2010). New techniques are becoming quickly available
to sequence the entire mitochondrial genome, which promise to provide a multitude of new data and
insights to Martes intraspecific relationships (Knaus et al. In Review).
Neutral and non-neutral markers (Microsatellites, SNPS)
Microsatellites, which are regions of the genome with high mutation rates, have been widely used in
conservation genetic investigations (Schwartz and Monfort 2008) and many software programs are
available for analysis of these data (Excoffier and Heckel 2006). Microsatellites, which are considered
- 12 -
Martes conservation genetics
Paper I
neutral genetic markers, have been very useful in estimating effective population size, abundance, gene
flow, hybridization, and measuring genetic diversity in natural populations. However, focusing on
quantitative traits or traits under selection, rather than neutral markers, may be a more direct avenue for a
better understanding of the adaptive potential of populations and the consequences of inbreeding and
outbreeding depression. Molecular markers like microsatellites cannot identify the likelihood of loss of
genetic variance in traits of ecological significance, as the correlation between neutral molecular diversity
and quantitative genetic variation is weak and becomes even weaker in expanding or declining
populations. Recently there has been a surge of new molecular genetic tools that will allow researchers to
examine markers under selection.
One of the most promising new molecular tools is the single nucleotide polymorphisms (SNPs). SNPs
represent the most widespread source of sequence variation within genomes (Brumfield et al. 2003) and
hold the potential to significantly expand the ability to survey both neutral variation as well as genes under
selection in natural populations as they have the very favorable characteristic of being detectable in both
non-expressed and expressed sequences. SNPs have only emerged recently as valuable genetic markers in
conservation genetics as they are biallelic markers. However, for some applications microsatellites will still
be the molecular marker of choice as SNPs are inherently less informative compared to the multiallelic
microsatellites, thus less informative when used for individual identification, parentage analysis and
population genetics.
- 13 -
CHAPTER 1
Introduction
Intraspecific
Species
Phylogenetic
Molecular Marker (lenght bp)
References
Data
Martes americana
Yes
Cyt b (441 bp)
Cyt b (1140 bp); ald C (241 bp)
Cyt b (441 bp) and complete Cyt b (1140 bp) in
combination with RFLPs;
1428 bp Cyt- b (1140 bp), tRNA-Pro (25 bp) and
D-loop (263 bp)
mtDNA and 14 Microsatelites
Carr and Hicks, 1997
Stone and Cook, 2002
Stone et al. 2002
Slauson et al. 2008
Dawson, 2010
Martes flavigula
No
-
-
Martes foina
No
-
-
Martes gwatkinsii
No
-
-
Martes martes
Yes
D-loop (321bp)
D-loop (350bp)
Cyt b, tRNA-Thr , tRNA-Pro, complete D-loop and
rDNA 12S (1608 bp)
Davison et al. 2001
Pertoldi et al. 2008
Ruiz-González et al. 2009
Martes melampus
Yes
Cyt b (402 bp)and RFLPs of rDNA
Cyt b (1140 bp);
Cyt b , tRNA-Thr, tRNA-Pro , D-loop (521-524
bp)
Cyt b, D-loop, and NADH-2 (2814 bp)
D-loop (535-537 bp)
Hosoda et al. 1999
Kurose et al. 1999
Murakami et al. 2004
Sato et al. 2009
Inoue et al. 2010
Martes pennanti
Yes
Isoenzymes
D-loop (301 bp);
D-loop (301 bp) & Cyt-b (428 bp)
Williams et al. 2000
Drew et al. 2003;
Vinkey et al. 2006; Schwartz,
2007.
Martes zibellina
Yes – parts of
range
Cyt b (402 bp)and RFLPs of rDNA
Cyt b (1140bp);
Cyt b, tRNA-Thr, tRNA-Pro , D-loop (521-524
bp)
Cyt b (702 bp)
D-loop (535-537 bp)
Hosoda et al. 1999
Kurose et al. 1999
Murakami et al. 2004
Malyarchuk et al. 2010
Inoue et al. 2010
Table 2. A list of interaspecific phylogenetic studies conducted on Martes species with the type of molecular marker
used.
- 14 -
Martes conservation genetics
Paper I
PHYLOGENETIC INFERENCE WITHIN SPECIES
With the development and more routine use of molecular methods, it is now possible to investigate
geographical variation using different molecular markers, and to deduce phylogenetic inference within
species. Intraspecific genetic variations applied within the framework of phylogeography provides further
insight into the history of range expansions and contractions of many Martes species (e.g. Stone et al.
2002; Davison et al. 2001; Ruiz-González et al. 2009). DNA sequences have been crucial in identifying
lines of descent both at the intra- and inter-population levels (e.g. Pertoldi et al. 2008b, Sato et al. 2009),
identifying the genetic legacy of the species translocations (Drew et al. 2003, Vinkey et al. 2006,
Schwartz 2007), elucidating taxonomic doubts or delineating conservation units (e.g. Stone et al. 2002,
Slauson et al. 2008, Sato et al. 2009), reconstructing colonization histories (Davison et al. 2001, RuizGonzalez et al. 2009), and even exploring temporal variation with the use of ancient DNA (e.g. Pertoldi et
al. 2001, Schwartz 2007, Pertoldi et al. 2008a). Molecular (and morphological) studies of the extant
species, in combination with palaeoecology, may provide opportunities to test hypotheses related to the
effect of fluctuations during Pleistocene ice ages on genetic diversity in extant populations. Thus,
molecular genetics can describe intraspecific geographical structures by identifying genetic lineages, and,
consequently, can reveal postglacial recolonization routes from the glacial maximum refugia, of which
locations are known (Davison et al. 2001; Stone et al. 2002; Ruiz-González et al. 2009).
Research on intraspecific genetics of Martes species are limited and biased towards some species (see Table
1 and 2). While some robust research on American martens (Martes americana) (Stone et al. 2002;
Slauson et al. 2008; Dawson 2010; Dawson and Cook This Volume), fishers (Martes pennanti) (Drew et
al. 2003, Vinkey et al. 2006, Schwartz et al. 2007) and the Japanesse marten (Marten melampus) (Hosoda
et al. 1999; Kurose et al. 1999; Murakami et al. 2004; Sato et al. 2009; Inoue et al. 2010) are available,
other species, such as the European pine marten (Martes martes) (Davison et al. 2001; Pertoldi et al. 2008;
Ruiz-González et al. 2009), and the sable (Martes zibellina) (Hosoda et al., 1999; Kurose et al.,1999;
Murakami et al. 2004; Inoue et al. 2010; Malyarchuk et al. 2010) are only partially or insufficiently
studied (Table 1 and 2). The absence of intraspecific phylogenetics studies for the Stone Marten (Martes
foina), the Nilgiri Marten (Martes gwatkinsii) and the Yellow-Throated Marten (Martes flavigula) is
noticeable.
During the last decade, DNA sequence-based studies on the phylogeny, evolutionary history and
taxonomy have begun to change our understanding about the genetic relationships between Martes species
(Hosoda et al. 1997; Hosoda et al. 2000; Stone et al. 2002, Sato et al. 2003, Koepfli et al. 2008). These
studies using both mitochondrial and nuclear DNA sequences highlight that some phylogenetic
- 15 -
CHAPTER 1
Introduction
relationships within the genera remain uncertain or unknown. Recently, a nearly complete generic-level
phylogeny of the Mustelidae using ~12,000 bp of mitochondrial and nuclear DNA data obtained from 22
gene segments (Koepfli et al. 2008) was even unable to resolve some relationships betwen Martes species
(See Koepfli This Volume). Consequently, in this framework of very closely related species, phylogenetic
inference within species is a challenging endeavour. Furthermore, as the species boundaries are sometimes
difficult to assess and can be questionable, data on intraspecific variation between closely related species is
sometimes difficult to elucidate (Davison et al. 2001; Sato et al 2009). Therefore, more slowly evolving
sequences are required in combination with mtDNA markers to investigate deeper phylogenetic history
that could also elucidate hybridization processes between closely related species commonly occurring
between many Martes species. On the other hand, research on phylogenetic inference needs to include
closely related species to evaluate both, intra- and inter –specific patterns. Below we go through our
intraspecific phylogenetic understanding for each of the Martes species:
The Pine Marten (Martes martes)
While information about the intraspecific relationships of the pine marten is very limited, there are a few
studies concerning its phylogeography (Davison et al. 2001, Pertoldi et al. 2008b; Ruiz-González et al.
2009). The phylogeography of M. martes has been investigated using a small fragment (321 bp) of the
control region (Davison et al. 2001). This work suggests that the present-day M. martes in central and
northern Europe is the result of colonization from one or several different refugia with subsequent
intermixing of expanding populations. However this work gives no clear clues for glacial refuges location
and the post-glacial recolonization of central Europe as it was based on a small fragment of DNA that gave
poor resolution to resolve the phylogeography of the species. Moreover, a lack of samples from the main
Mediterranean refuge areas (only 1 specimen from the Iberian Peninsula, 3 from Italy and 2 from
Northern Balkans) limited the support of the recolonization hypothesis. Davison et al. (2001) reported
evidence for historic introgression of M. martes with the sable (M. zibellina) in Fennoscandia, along with
mtDNA and morphological evidence for introgression with American martens (M. americana caurina) in
England later confirmed by microsatellite data (Kyle et al. 2003).
However, no samples of M. martes and M. zibellina from Russian populations were included to underline
the processes for the presence of a divergent lineage in Fennoscandia. More recently, Ruiz-Gonzalez et al.
(2009) investigated the unresolved questions posed in previous work (Davison et al. 2001), and reexamined the phylogeographyc pattern of the European pine marten throughout the current species’
distribution. This study was more comprehensive in terms of number of specimens and length of mtDNA
- 16 -
Martes conservation genetics
Paper I
sequences (1,600 bp). The sampling also covered a broader geographic range than the previous
phylogeographic study, sampling individuals and populations from Scandinavia in the North, Russia in
the East, and the Iberian Peninsula in the South-West. For clarifying the relationships between M. martes
and the sable M. zibellina in Fennoscandia and Russia, sable samples were also included. Ruiz-Gonzalez et
al. (2009) revealed the presence of 69 different haplotypes for M. martes and 11 for M. zibellina, which are
split into two major assemblages: European-Mediterranean and Fennoscandian-Russian clades. The first
clade, including all M. martes samples collected throughout its entire current European distribution, is
subdivided into two groups joining haplotypes distributed in central-northern Europe and in the
Mediterranean regions. Most interestingly, haplotypes in the Mediterranean clade apparently did not
contribute to the postglacial recolonization of most of the Palaearctic range of the species. It seems that
Central-Northern Europe was recolonized by a pine marten phylogroup that survived the last glaciations
in an undetermined Central European refuge as it has been previously proposed by paleontological data
(Sommer and Benecke, 2004). In addition to this complex recolonization of Europe, genetically
differentiated populations of pine marten, distributed in Fennoscandia and Russia, are introgressed with
mtDNA of M. zibellina. In conclusion, the study of Ruiz-Gonzalez et al. (2009) indicates a complex
phylogeographic history for M. martes, a species sufficiently adaptable to survive, facing historical climate
changes, both in southern and northern forest habitats.
In a more restrictive study in terms of spatial distribution, Pertoldi et al. (2008b) studied the genetic
differentiation and changes over time in genetic variability of the pine marten in three isolated geographic
regions from northern Europe: Jutland and Sealand (Denmark) and southern Scania (southernmost
Sweden), by sequencing the hypervariable domain of the mtDNA D-loop (350 bp). Both recent and
museum samples were analysed in order to evaluate any temporal loss of genetic variability. Pertoldi et al.
(2008) found eight different haplotypes: two haplotypes were shared by individuals from all three regions,
but unique haplotypes were also found in all localities. When comparing these data with previous
haplotype analysis (Davison
et al. 2001), they confirmed the presence of the two main distinct
haplogroups of central and Northern Europe along this region, with the samples from southern Scania
being well differentiated from central Sweden samples. The obtained genotypic data for Jutland and
Sealand suggest a recent independent evolutionary history for the Danish pine marten.
The American Marten (Martes americana)
The intraspecific phylogeny of the American marten has been debated during several decades, with clear
doubts about sub-specific and specific status. Traditionally, 8 subspecies of M. americana have been
- 17 -
CHAPTER 1
Introduction
described (Clark et al. 1987) and placed into two morphologically distinct groups, americana and caurina
(Merriam 1890; Clark et al. 1987). Although several studies (e.g. Merriam 1890; Anderson 1970; Hall
1981; Clark et al. 1987; Carr & Hicks 1997, Stone and Cook 2002, Stone et al. 2002, Small et al. 2003,
Dawson 2010) have corroborated the separation of M. americana into these two groups, the level of
distinctiveness between them has been largely debated.
Before 1953, these two groups of martens were recognized as distinctive species in North America: Martes
americana and Martes caurina (Merriam 1890). However, Wright (1953) noted that both species
intergrade in British Columbia and Montana, and he proposed that they should be combined into one
species. Since then, they have been synonymised under Martes americana, and the conclusion has been
reflected by the current taxonomy (Wilson & Reeder 1993).
However, the dichotomy (americana and caurina) in North American marten has continued to be
acknowledged. Preliminary molecular data corroborate the distinction of caurina and americana as two
monophyletic mitochondrial clades (Carr & Hicks 1997; Stone et al. 2002). Carr & Hicks (1997)
compared the divergence of the two groups with that of 4 Palearctic species of Martes. As the mtDNA
divergence levels found between the americana and caurina clades were similar to those existing among the
other four Palearctic species, Carr & Hicks (1997) concluded that M. americana and M. caurina should be
recognized as distinct species. However, subsequent studies gave subspecific status to caurina and
americana clades (Stone and Cook 2002; Stone et al. 2002).
These two lineages are largely allopatric. The americana clade is widespread from interior Alaska south to
Montana and eastward to Newfoundland and New England (i.e. northwestern, north-central and
northeastern North America). By contrast, the caurina clade occurs in western North America, extending
from Admiralty Island in southeastern Alaska south to Oregon and Wyoming (Wright 1953; Hall 1981;
Carr and Hicks 1997; Stone et al. 2002). Within the americana clade, little or no geographical structure
was present among populations, while within caurina, several haplotypes were confined to single
populations (Stone et al. 2002). Interbreeding between these two lineages where they come into contact
(i.e. Montana and one island in southeastern Alaska) have been demonstrated by DNA studies, (Stone and
Cook 2002; Stone et al. 2002). According to Stone et al (2002), these lineages appear to have diverged due
to isolation in distinct southern glacial refugia (caurina populations isolated in the west and americana
populations isolated in the east of the United States, respectively). They hypothesized that the individuals
belonging to the caurina clade represent an early Holocene colonization northward along the coast as
coastal ice receded at the end of the last glaciation, whereas americana populations represent a later
- 18 -
Martes conservation genetics
Paper I
colonization from continental source populations that expanded through river corridors traversing the
coastal mountains.
Later Small et al. (2003) explored the distinctive histories of caurina and americana populations using 14
nuclear microsatellite markers. Microsatellite studies corroborated the population structure patterns
obtained by the DNA sequences of the mitochondrial cytochrome b gene fragment (Carr & Hicks 1997;
Stone et al. 2002), the nuclear aldolase C gene fragment (Stone & Cook 2002) as well as earlier
morphological works (Merriam 1890; Anderson 1970). Small et al. (2003) suggest that these lineages may
well represent distinctive species and that further investigations of ecological, behavioural or physiological
characteristics should be conducted to elucidate this question.
More recently Dawson (2010) and Dawson and Cook (This Volume) reviewed the previous molecular
studies and developed a more detailed view about the genetic differentiation across the whole range of
North American marten lineages. In this work this author addressed the question of how many species of
marten are extant in North America. Mitochondrial DNA studies identified 2 monophyletic groups
within North American Martes that corresponded to the 2 morpho-species described. Investigations using
nuclear loci were also consistent with species-level differences in North American Martes. Based on a series
of studies, they concluded that M. americana and M. caurina are valid species paralleling the original
descriptions.
The subspecific status of several American marten populations from Oregon and California has been also
investigated by Slauson et al. (2008). These authors investigated the subspecific identity of a rediscovered
population of American martens within the range of a presumed extinct subspecies (Martes americana
humboldtensis). They compared mtDNA (1,428 bp) sequence diversity of contemporary specimens within
the described range of M. a. humboldtensis, close to ranges of M. a. caurina and M. a. sierrae, and a
museum specimen of M. a. humboldtensis. The historical Humboldt marten museum sample shared one
haplotype with martens from the rediscovered population, coming from coastal regions of Oregon and
California. This result suggests that the rediscovered population represents descendants of a relictual
population that previously existed in coastal California. They also concluded that subspecific boundary
between M. a. humboldtensis and M. a. caurina may not be valid, because this haplotype was shared with
coastal Oregon and coastal California current populations and no known contemporary or historical
biogeographic barriers prevent north–south movement. Thus, marten populations currently located in
coastal forests of California and Oregon should be managed collectively to preserve the connectivity that
these results suggest. Moreover, M. a. sierrae differed substantially from both M. a. humboldtensis and M.
- 19 -
CHAPTER 1
Introduction
a. caurina, suggesting marten populations were not one large, genetically homogeneous population along
the Pacific states and that the divergence may have occurred in separate glacial refugia.
The Fisher (Martes pennanti)
Intraspecific genetic research on the fishers (Martes pennanti) has been conducted to examine the genetic
consequences of past and future translocations and examine the validity of morphologically based subspecies designations. Translocations to re-establish extirpated populations or to maintain declining ones
have historically been carried out without genetic information on source or target populations or adequate
consideration of the potential effects of mixing genetic stocks. The first research on this topic was limited
to translocations within the eastern range of the fisher, and showed little genetic subdivision among
populations (probably because of the low variability of the molecular marker use; Williams et al. 2000).
Subsequently, Drew et al. (2003) considered the conservation status of Martes pennanti and evaluated the
potential genetic consequences of past and future translocations by examining population variation of Dloop sequences. They sampled populations throughout the fisher range in North America including five
populations unaffected by translocations and two western populations that had received long-distance
translocations. Populations in Oregon, Montana, and Idaho received several translocations and, as a result,
these three populations showed greater similarities to source populations than to adjacent ones. Additional
sequences obtained from museum specimens collected prior to any translocation suggested historical gene
flow among populations in British Columbia, Washington, Oregon, and California. This study concluded
that anthropogenic impacts in that region have greatly reduced and isolated extant populations in Oregon
and California. Therefore, British Columbia would be the most appropriate source population for future
translocations to recover those of Washington and some localities in Oregon and California. This result
was confirmed using the same molecular markers by Warheit (2004) as reported in Lewis and Hayes
(2004). Recent work by Knaus et al. (In Review) has re-examined some of the Drew et al. (2003) results
using complete mitochondrial genomes (~16,000 bp). The most striking result was that D-loop region
sequences incorrectly identified full genome population structure. For instance, Drew et al. (2003) showed
that both northern and southern California shared a common haplotype, suggesting gene flow, yet the full
genome revealed that these geographic areas each had unique haplotypes, concordant with microsatellite
data (Wisely et al. 2004) and consistent with long term isolation.
Until recently it was assumed that Martes pennanti specimens living in the Rocky Mountains all were
descendants from reintroduced stocks. However, a recent study reported that mtDNA (428 bp of Cyt-b
and 301 bp of D-loop) haplotypes found only in west-central Montana fishers were likely derived from a
- 20 -
Martes conservation genetics
Paper I
relict population that escaped harvests conducted in the early 20th century (Vinkey et al. 2006). Schwartz
(2007), using the same molecular markers as the study of Vinkey et al.(2006) compared fishers in westcentral Montana with samples from north-central Idaho and found no differences between these groups.
One museum specimen, collected in 1896 in north-central Idaho before any known translocation, had the
same haplotype as the ‘‘native Montana haplotype’’ discovered in the recent study of Vinkey et al. (2006).
Thus, fishers in north-central Idaho and west central Montana are the only confirmed native fishers in the
Rocky Mountains, and one of a few populations in the West that have maintained native genetic
characters. Fishers from Idaho and Montana are not all descendants of translocated individuals, but are
also the descendants of fishers that persisted despite early 20th century trapping. Recent data by Knaus et
al. (In Review) confirm these results.
The Japanese Marten (Martes melampus)
The Japanese marten is an endemic species to Japan, where it is distributed on the main Japanese islands,
except Hokkaido Island: Honshu, Shikoku and Kyushu Islands (Masuda et al. 2009). By contrast, the
sable (Martes zibellina) is present only in Hokkaido Island within Japan (Murakami et al. 2009). However,
M. melampus was artificially introduced to Hokkaido from Honshu, and is currently expanding in
southern Hokkaido, whereas the native M. zibellina is distributed in central and eastern Hokkaido. The
contact zone between the two species is in central Hokkaido (Masuda et al. 2009; Murakami et al. 2009).
The Japanese marten has a complicated taxonomic history. Moreover, the presence of the closely related
species, Martes zibellina (Hosoda et al., 1997, 2000; Sato et al., 2003, 2006; Koepfli et al., 2008) makes
correct intraspecific assignment more difficult. Several studies have focused on the genetic relationships
within and between the Japanese marten, and M. zibellina brachyura (Temminck, 1844) (Hosoda et al.
1999; Kurose et al. 1999). Hosoda et al. (1999) used the restriction fragment length polymorphism
(RFLP) of the nuclear ribosomal DNA (rDNA) spacer and the mitochondria1 cytochrome b (402bp) gene
fragment sequences, and Kurose et al. (1999) sequenced the entire Cyt-b (1140 bp), to reveal the extent
of intra- and inter-specific variation between the two species. Both studies showed high genetic differences
between these two species. However, the clustering of haplotypes in phylogenetic trees did not correspond
with the geographically expected relationships between populations of the different Japanese islands. Only
the Tsunima island populations (Martes melampus tsuensis) showed some genetic differentiation from the
other populations of M. melampus. The results suggested that mtDNA introgression between local
populations of M. melampus might have resulted from the incomplete geographic isolation within each
- 21 -
CHAPTER 1
Introduction
island, and/or that M. melampus might have recently expanded to the Japanese islands during a short
period.
Based on 521-524 bp fragments of mtDNA including D-loop, Murakami et al. (2004) further
investigated these questions and indicated that the two Martes species in Hokkaido were closely related, in
contrast to previous studies (Hosoda et al. 1999; Kurose et al. 1999) which indicated that M.melampus
and M. zibellina were grouped in two different groups. However, only Hosoda et al. (1999) included one
individual from Hokkaido were both species coexist. Murakami et al. (2004) revealed two distinct clusters,
each containing both haplotypes from M. zibellina and M. melampus providing three possible
explanations: past hybridization between both species might have occurred; the two species might have
similar heteroplasmy of mtDNA; or these haplotypes might have come from the nuclear genome.
By contrast, Inoue et al. (2010) examined 535-537 bp of mtDNA D-loop for Japanese martens from
native populations of Honshu and Kyushu and from introduced populations of Hokkaido, together with
sables of Hokkaido and Russia, and showed high genetic differentiation between the two species. In
addition, Inoue et al. (2010) reported that there were neither individuals showing melampus-type mtDNA
haplotype together with zibellina morphology, nor those showing reciprocal characters in specimens from
Hokkaido. These results indicated that no hybridization occurred between these two species on this island.
The genetic diversity of the introduced populations of the Japanese martens in Hokkaido was lower than
that found in the native populations of Honshu and Kyushu. This could be due to founder effects.
One of the most comprehensive studies regarding phylogenetic inference within M. melampus populations
was recently published by Sato et al. (2009). The first studies focused on the molecular phylogeography of
the Japanese marten did not detect different genetic units (Hosoda et al. 1999, 2000; Kurose et al. 1999),
possibly due to insufficient information from the molecular markers used. Furthermore, the populations
studied were poorly sampled. Previous works showed that genetic diversity of the Japanese marten should
be analyzed more comprehensively, so that conservation units could be defined in terms of geographical
population. Sato et al. (2009) conducted a more reliable research in terms of molecular markers (mtDNA
and nDNA), sequence length and sampling, to overcome this lack of information. In this study, they
performed molecular phylogenetic analyses of 49 specimens of Japanese martens collected from several
areas in Japan and focused on three mtDNA loci (Cyt-b, D-loop, and the NADH subunit 2 gene) and one
nuclear gene (the growth hormone receptor gene, including the polymorphic intron regions). Evaluating
the phylogeny and the genetic variation estimated by mtDNA and nDNA sequences, Sato et al. (2009)
identified nine intraspecific groups. The grouping was not correlated with winter coat color, but was
consistent with geography of the Japanese islands. In particular, they obtained the monophyly of the
- 22 -
Martes conservation genetics
Paper I
Tsushima martens, M. m. tsuensis, supporting the view that the Tsushima marten’s long history of
isolation on small islands is responsible for its genetic distinctiveness and uniformity, validating the
Tsushima population as an evolutionarily significant unit.
The Sable (Martes zibellina)
The Sable shows substantial interpopulation variation of morphological characters and a multiplicity of
local forms which make the elaboration of unified intraspecific taxonomy complicated (Monakhov, 1976;
Pavlinov and Rossolimo, 1976). Moreover, the massive human introductions and re-introductions
throughout most of its geographic range in Russia in the 20th century complicate this goal (Monakhov,
2001).
Several studies have been focused on the Hokkaido population, to investigate genetic relationship within
and between the sable, which is classified as a separate subspecies M. zibellina brachyura (Temminck,
1844), and the closely related Japanese marten (Martes melampus), which was introduced to this island (see
above). However, the results are of limiting significance for the intraspecific relationship within Martes
zibellina after taking into account the wide range of the species over most of Asia.
In the Russian populations, previous analysis of the population genetic polymorphism in sables pointed to
the existence of intraspecific heterogeneity (Balmysheva and Solovenchuk, 1999a;b; Petrovskaya et al.
2007). Analysis of restriction polymorphism of the mtDNA Cyt-b in populations of sables from Siberia
and Far East showed the prevalence of three different haplotypes, which probably represented three
monophyletic lineages (Balmysheva and Solovenchuk, 1999a;b; Petrovskaya et al. 2007). To confirm these
findings, fine-scale analysis of mtDNA variation at the level of nucleotide sequences was recently
conducted by Malyarchuk et al. (2010). This study was focused on the analysis of phylogenetic
relationships of the mtDNA Cyt-b sequences in 17 sables from Magadan oblast, Kamchatka, and
Khabarovsk Krai, and also included previously published data of Martes zibellina brachyura from
Hokkaido, Japan.
These authors identified two supported phylogenetic groups of sable mtDNA. The first group was
predominantly represented by sables from different regions of Northeast Asia, including Kamchatka,
Khabarovsk Krai, and the Magadan oblast together with samples from Hokkaido Island (Japan). The
other group was composed by haplotypes of sables from Magadan oblast and Khabarovsk Krai, without
haplotypes representing populations of Kamchatka or Hokkaido. Interestingly, a pine marten specimen
from Sweden was clustered within this group, supporting previously reported evidence for historic
- 23 -
CHAPTER 1
Introduction
introgression of pine martens with sable (M. zibellina) in Fennoscandia (Davison et al. 2001, RuizGonzález et al. 2009). The high sequence divergence values obtained between both haplogroups imply
that the ancestral gene pool of sables was once split into two parts, probably, as a result of glaciation. Later
on, during the last deglaciation period, the two gene lineages were reunified in a new contact zone
(Malyarchuk et al. 2010).
In the populations from Central Kamchatka, all sables examined belonged to the first clade, as well as all
examined Martes zibellina brachyuran individuals from Hokkaido (n = 10). Hosoda et al. (1999) also
identified one unique haplotype on Hokkaido which is closely related to one of the two identified
haplotypes in Russia. They hypothesized that divergence between Hokkaido and the Russian Far East
populations has been a recent process. However the limited number of specimens analyzed limits the
support of this hypothesis. Hokkaido is thought to be a refugium for Martes zibellina during the last
glacial age (Kurose et al. 1999). However, Kurose et al. (1999) hypothesised that populations of the sable
could have expanded in Hokkaido recently. Otherwise, the repeated processes of expansion and reduction
of their habitats through the glacial and interglacial ages would have impeded the fixation of haplotypes to
local populations. Further studies are required to examine genetic variation within each local population
by using more specimens from comprehensive areas including the continent and using more polymorphic
DNA markers such as microsatellites. On the other hand, the observed spatial heterogeneity of the sable
populations of Magadan oblast with individuals from both clades, could be explained by the introduction
of Kamchatka and Khabarovsk sables, starting in the 1950s (Petrovskaya et al. 2007; Malyarchuk et al.
2010).
It seems likely that morphological differences between sables belonging to each haplogroup still have
certain contributions to the contemporary phenotypic diversity of sable populations. This suggestion is
supported by morphologic specific differences of Kamchatka sables, which are treated as a separate
subspecies M. z. kamtschadalica (Pavlinov and Rossolimo, 1976, Anderson, 1970), and characterized by
the presence of mtDNA haplotypes of the first group (Malyarchuk et al. 2010). It should be noted that
haplotypes from both groups are in contact with sable inhabiting Khabarovsk Krai and the Magadan
oblast and historically belonging to another subspecies, M. z. jakutensis.
Overall, a more reliable study in terms of spatial distribution and number of individuals is necessary to
better understand the phylogenetic relationships within Martes zibellina populations over the wide range
of the species.
- 24 -
Martes conservation genetics
Paper I
Martes POPULATION GENETIC STUDIES
The phylogenetic studies represented above consistently show complex evolutionary history, markedly
influenced by large scale events like glacial refugia and subsequent reunitions. These intraspecific
phylogenetic studies set the stage for examining more fine scale temporal and spatial relationships among
populations. To date there have been population genetic studies on three of the Martes species: the
American marten, the pine marten and the fisher, many of them examining the impact of habitat
fragmentation on the movement of populations within a species.
The pine marten is distributed throughout Europe and has been subject to long term decline in numbers
in most regions (Mitchell-Jones et al. 1999). The pine marten is a habitat specialist confined to mature
deciduous and coniferous forests (Domingo- Roura, 2002), has a limited dispersal ability compared with
other mustelids (Kyle et al. 2000), and a slow reproduction rate, rendering it particularly vulnerable to
habitat changes (Bright, 2000, Webster 2001).
The pine marten populations has been shown to have a higher level of genetic structure (with an overall
FST value of 0.18, range: 0.016–0.330) and lower genetic variation (HE range excluding the insular
populations: 53.8%-63.8%) than their North American sibling species, the American marten (Martes
americana), sampled throughout Canada (HE Yukon: 69%; Kyle et al. 2003). Even if it is difficult to
exclude more ancient processes such as the influence of glaciations as a cause of the differences observed
between these two species, it is suggested that the greater level of persecution and habitat fragmentation
experienced by the pine marten could be the reason (Kyle et al. 2003).
Despite the fact that the continental pine marten populations shows a relatively homogenous levels of
genetic variation, with no significant differences among them (Kyle et al. 2003), the level of genetic
differentiations between the populations appear to be correlated with the geographic distance between the
populations. This is true even if the pairwise FST did not correlate so strongly with distance (r = 0.31, p =
0.11), as other measures of genetic differentiations such as Nei’s genetic distance (DS) and the genotype
likelihood ratio (DLR) have shown both highly significant correlations (DS: r = 0.55, p= 0.007, DLR: r =
0.91, p = 0.00006; Kyle et al. 2003).
These results may be related to the relative differences in the level and duration of anthropogenic
disturbances in Europe and northern North America. These influences may have resulted in smaller, more
isolated populations in Europe where the effects of genetic drift would lead to more genetic structure.
Differences in NE can also be the causes of the observed pattern as smaller NE is translated in a quicker loss
- 25 -
CHAPTER 1
Introduction
of genetic variability. For these reasons, it is important to monitor the levels gene flow and of genetic
variation between populations to help identify populations where conservation actions may be appropriate.
However, a cautionary approach must be undertaken when trying to interpret differences in the level of
genetic structure or variability between species as these differences can also be attributed to other factors.
Kyle et al. (2003) discussed the possibility that the structure observed in their study could also reflect a
greater degree of philopatry in the pine marten compared with the American marten. Additionally they
discussed the possibility that more ancient processes still influence the gene frequencies, such as postglacial founder effects and historical introgression from sable (Martes zibellina) in Fennoscandia.
Sometimes, studies conducted on insular populations are more easy to interpret as often there is no
“confounding” gene-flow occurring and the populations has typically been isolated for a well known
period of time. This allows us to make more accurate predictions about the fate of genetic variability and
substructuring. Among island populations of the two siblings species, the Scottish pine marten population
revealed a similar level of structure and variation to the Newfoundland martens (Martes americana atrata)
(HE Scotland: 42.3%, HE Martes Americana atrata: 44.6%), however Ireland (HE Ireland: 34%), was more
differentiated with less genetic variation (Kyle et al. 2003).
Abundance estimates of the Newfoundland martens (which seems to be genetically similar to mainland
populations despite their biogeographic separation of nearly 10,000 years; Carr and Hicks 1997;
McGowan et al. 1999) decreased from 630–875 animals in 1986 to only 300 animals in 1995 (Snyder
1986, Forsey et al. 1995). Such a drastic and rapid population decline has raised concern that inbreeding
depression could affect the average fitness of this population (Forsey et al. 1995). The level of genetic
variability of the English pine marten (HE England: 66.1%), could not be used for comparisons as Kyle et
al. (2003) provided further evidence for the possibility that a hybridization event between Martes
americana and Martes martes has occurred (see phylogenetic section above).
Martes americana caurina is a distinct clade of American marten inhabiting the far western North America
(Stone et al. 2002), when compared with Martes americana americana. In fact the genetic differences
found between Martes americana caurina and Martes americana americana are at a level observed among
other Martes species, which have lead some to consider this group two distinct species – Martes caurina
and Martes americana (Carr and Hicks 1997, Stone et al. 2002, Small et al. 2003; Dawson 2010 and
Dawson and Cook This Volume) (see phylogenetic section above). Small et al. (2003) have shown that
northern insular populations of M. caurina have higher genetic differences among populations and lower
within population genetic diversity compared to northern populations of M. americana likely caused by
- 26 -
Martes conservation genetics
Paper I
the longer periods of isolation in coastal forests that were fragmented during the early Holocene period
(Small et al. 2003). The lack of differences among M. americana populations has been attributed to either
continued gene flow or a more recent expansion throughout the Pacific Northwest (Small et al. 2003)
The comparisons of genetic parameters between species can be quite misleading both because of the
different genetic markers utilized but also because of the species different histories of postglacial
recolonization and eventual introgression. Hence, a better approach, in order to understand the reasons for
the observed genetic pattern seems to be to compare the degree of genetic variability and genetic structure
within the same species but comparing populations living in different habitats. An example of such an
approach, (using a generous sample size of individuals and markers), has been conducted by Kyle and
Strobeck (2003) again on the American marten (1,262 individuals, genotyped at 11 microsatellite loci)
which compared the level of genetic variability and of genetic differentiations between populations living
in the unfragmented habitat of northern Canada with the genetic variability and structure of the
populations living in the more southern Canadian regions which have a more fragmented habitat. As
expected, and in agreement with previous studies, little genetic structure was observed in northern regions,
where few barriers to marten dispersal are thought to exist. However, contrary to their expectations, no
strong breaks in gene-flow were observed between any of the 35 sampled regions with the exception of the
insular Newfoundland population. The lack of genetic structure observed may suggest very large NE of the
populations and that, at a larger scale, marten dispersal is not as limited by some landscape features as was
previously thought. In Canadian populations of martens (with the exception of the insular Newfoundland
population) the lack of genetic structure may be explained by large NE and relatively continuous habitat,
conditions that are not present for the European species.
Hence, despite the fact that life history traits of European pine martens more closely resemble those of the
American marten, the demographic trends of this species seem to quite similar to the pattern observed for
fisher (Martes pennanti) which often live in similar geographic areas as the American marten.
Kyle et al. (2001) which investigated the fisher populations sampled from across the Canadian provinces,
revealed relatively high levels of genetic structuring (FST = 0.14; range: 0.028-0.261) compared to the FST
found for the American marten (FST = 0.020) and a relatively high genetic variability (HE = 62%) over
short geographic distances. The level of structure in fishers could be a reflection of philopatry and the large
demographic changes that affected most populations of this species in the early 1900’s. Fishers were
extirpated from much of their range as a result of anthropogenic influences (fur harvests and logging).
- 27 -
CHAPTER 1
Introduction
Therefore, mainland populations of European martens may be similarly structured to fishers where smaller
NE has potentially led to more genetic drift between populations. However the fisher showed relatively
high levels of genetic variability, despite the recent history of populations decline. The level of genetic
variability found for fisher (Kyle et al. 2001, Wisely et al. 2004) and successively by Carr et al. (2007a,b)
(HE range: 0.599–0.679) has been suggested to be due to the partitioning of genetic variation into
multiple refugia during the period of population fragmentation, where the presence of multiple refugia has
probably maintained relatively higher levels of genetic variation than predicted for a single source.
Subsequently, contact among expanding reproductive fronts may have counteracted loss of genetic
variation within any one cluster by increasing the gene-flow among clusters (Thompson 2000).
However, an examination of the distribution of genetic structure of fisher at different distances showed
higher rates of gene flow than predicted under a strict isolation by distance model at small distances (40
km) within clusters and at larger distances up to 100 km among clusters. Such a pattern has been
associated with expanding reproductive fronts and translocations events (of individuals from known
sources: Berg 1982) as many of the patches where the fisher disappeared during the population decline
have now been recolonized. It should also be mentioned that Wisely et al. (2004) found much lower
values of HE for fishers in the fragmented populations along the Pacific coast, ranging from(HE range:
0.16-0.42) associated with relatively high estimates of FIS values, suggesting inbreeding. Also this pattern
of reduced heterozygosity appears to follow a north-south gradient, with fisher populations in the southern
part of the Sierra Nevada population showing the lowest levels of genetic variation as compared to sister
populations to the North (Wisely et al. 2004).
As previously mentioned the comparisons of genetic variability and genetic differentiations (among species
or among populations within the same species) can be quite problematic when using different genetic
markers in different species, but can be quite problematic also when comparing genetic differentiations
among populations collected at different geographical scales. One way to overcome the latter problem is to
perform a Mantel test, which consists in plotting the pairwise geographic and genetic distances between
populations, and such an approach can also be used to illustrate the difference between the different
species. This has been illustrated by Kyle and Strobeck (2003), who plotted a linear regression of the DS
and geographic distances for mainland populations of pine marten, fisher and American marten (Figure
1).
- 28 -
Martes conservation genetics
Paper I
M. Pennanti (West Coast, United States)
M. Martes (Europe)
Genetic
Distance
M. Pennanti (Canada)
G. gulo
M. Americana (Canada)
L. canadensis
0
1000
2000
3000
4000
5000
Km
Figure 1. A schematic showing the approximate relationship between genetic distance and Euclidean distance for
several Martes species. Dotted lines show the approximate relationship for other mid-sized carnivores and is presented
as a reference.
The highest level of structure per unit distance (which is directly correlated with the degree of genetic substructuring) was found for the pine marten: 0.140/1,000 km, followed by fisher 0.092/1,000 km and
American marten 0.057/1,000 km (Figure 1). The Mantel test is a rough test and nowadays there are
more sophisticated techniques which have been developed by the emerging field of landscape genetics
which allow the detection of subtle sub-structure and of barriers.
In addition to genetic substructure analyses, population genetic analyses have also provided insight into
hybridization among Martes species. As previously mentioned microsatellite data has confirmed mtDNA
data suggesting hybridization between M. americana (introduced) and M. martes (native) in England
(Davison et al. 2001; Kyle et al. 2003), although there still remains several native M. martes populations in
Great Britain that show no sign of genetic introgression. Hybridization has also been identified using
microsatellite data between M. caurina and M. americana in two regions, the Kuiu Island in southeastern
Alaska and in Montana, USA (Small et al. 2003, see above). To date no studies have examined the extent
of the hybrid zone or the fitness of these hybrids relative to the parental populations.
- 29 -
CHAPTER 1
Introduction
Population genetic data has also been used to evaluate the success of reintroductions of Martes. Williams et
al. (2000) show that older reintroductions of fisher show significant allele frequency differences from their
source populations. While some of this difference may be due to initial sampling error, as typical
reintroductions use relatively few individuals, the fact that recent introductions do not show significant
allele frequency differences form source populations suggests that this is due to genetic drift. Drift can
occur rapidly in small populations, especially with species that exhibit polygynous mating. Swanson et al.
(2006) and Swanson and Kyle (2007) examined reintroduced populations of marten in Michigan, USA
and found that the reintroduced populations had high levels of genetic variation. This may be due to the
use of multiple source locations or the temporal separation of the reintroductions, which occurred over a
24 year period (Williams and Scribner 2007, Swanson and Kyle 2007).
Landscape Genetics
Population genetic data has been used to effectively delineate substructure, identify isolated populations,
and define units of conservation. However, by definition these data rely on group or population statistics.
This can be difficult when species appear to be continuously distributed across a landscape and groups are
not readily apparent. In addition, while some elements of population genetics, such as measures of
between population genetic distance, are inherently spatial, they do not specifically take the landscape into
account. The field of landscape genetics is an extension of population genetics that uses either individual
or population data, explicit spatial information, and associated covariates (i.e., elevation, forest type,
distance to roads, etc) to make inference to environmental variables that influence species movement.
Landscape genetic approaches are relatively new, but since the term was coined in 2003 (Manel et al.
2003) there have been over 400 published papers that reference or use these methods. The most common
landscape genetics approach used is to compare ecological distances among either individuals or
populations to a matrix of genetic distances (or the inverse, genetic relatedness). These ecological distances
can be distance among populations or individuals measured in stream distances, distance through forest
cover, distances through riparian zones, distances across non-human habitation, distances across savanna
or steppe, or any other environmental variable deemed important to the organism’s life history, survival
and ability to disperse. This approach becomes more complicated when the landscape is a mosaic of
habitat patches, and there is not a continuous path within the ecological covariate of choice forcing
populations or individuals to move through non-optimal habitats to interact. Here the standard landscape
genetics approach has been to impose cost values on habitats of different quality and type, and conduct
least cost-path modeling to derive a matrix of least cost paths among individuals or populations. Given
- 30 -
Martes conservation genetics
Paper I
that a specific cost per habitat type is rarely known, often multiple models with different cost penalties are
created. These multiple models are then evaluated by comparing the many matrices of least cost paths to
the matrices of genetic distances. In more complex models, these resistance values can be an aggregation of
costs imposed by multiple variables or can be evaluated using multiple matrix regression modeling, where
each covariates influence on genetic relatedness can be evaluated (Balkenhol et al. 2009).
In addition to least cost path modeling, there have been several graph theoretic based approaches that have
been developed for landscape genetic analyses. These approaches allow identification and prioritization of
important locations and populations for maintaining connectivity. The most widely used graph theoretic
approach is one based in electrical circuit theory and is incorporated into the program CIRCUTSCAPE
(McRae and Beier 2007, McRae and Shah 2009). This model simultaneously considers all possible paths
connecting individuals or populations based on resistance distances. This approach is similar to least-cost
path modeling, but can provide different results as it can simultaneously evaluate contributions for
multiple dispersal pathways which can identify areas where connectivity is most tenuous (i.e., pinchpoints, McRae and Shaw 2009).
The field of landscape genetics is relatively new and has not been used with many Martes species yet,
however, there are some important exceptions. Broquet et al. (2006a) tested whether American marten in
the boreal forest of Ontario, Canada showed isolation by distance (i.e., genetic distance was positively
correlated with geographic distance) and found no significant relationship. However, the samples were
collected in 11 different habitat patches that could be categorized as logged (regenerating tree stands of
different ages and types) and unlogged old-growth forest (>80 years old). Examining patterns of isolation
by distance in unlogged landscapes revealed a significant pattern of isolation by distance (Mantel test
P=0.01), whereas isolation by distance was not significant in the logged replicates (Mantel test P=0.42).
This suggests that marten dispersal is changed in suboptimal, logged habitats compared to intact
environments and is consistent with demographic studies on marten movement (Broquet et al. 2006a).
The authors subsequently used least-cost path modeling to confirm these results. Interesting in this
ensuing work they also showed that this result was partially dependent on the resolution of the maps, with
intermediate grid cell sizes showing the strongest relationship (Broquet et al. 2006b). The authors explain
this result by suggesting that the largest grid cell sizes (coarsest resolution) may miss important landscape
features that marten are sensitive to, yet the smallest grid cell sizes (finest resolution) would require more
complex parameterization of habitat features (by changing the resistance of particular landscape features
such as rivers, or by adjusting resistance to the width, shape and orientation to other landscape elements)
to produce more accurate model results (Broquet et al. 2006b).
- 31 -
CHAPTER 1
Introduction
Similar least cost path modelling has also been conducted on American marten in Idaho, U.S.A. and
European pine marten in Ardennes, La Bresse, and L’Isere, France (Wasserman 2008, Mergey 2007) both
showing the importance forest structure for dispersal across large landscapes. In addition, Wasserman
(2008) showed the importance of elevation, which was a proxy for snowpack, with marten avoiding lower
elevations and dispersing in mid to high elevation forests characterized by moist cool sites with subalpine
fir (Abies lasiocarpa) and Engelmann spruce (Picea engelmannii).
Several new landscape genetics approaches have been used to evaluate a recolonizing fisher population in
southern Ontario, Canada (Carr et al. 2007a, b, Garroway 2009). Initial research tested the idea that
Algonquin Provincial Park was the source population for this colonization by examining microsatellite
profiles of 35 sites (groups of samples or “populations”) surrounding the Park (Carr et al. 2007a). The
authors found that these 35 sites could be clustered into 5 discrete genetic groups first showing multiple
origins. These origins were not Algonquin Park as initially predicted, but rather remnant populations
within Ontario, Quebec, and New York (U.S.A.). Carr et al. (2007a) also showed that these populations
were rapidly homogenizing among expansion fronts. Subsequent research used assignment tests to infer
the proportion of immigrants into each of the 5 genetic clusters and relate the proportion of immigrants to
habitat variables (Carr et al. 2007b). Carr et al. (2007b) showed a positive relationship between snow
depth and the proportion of immigrants, and a negative relationship between the proportion of coniferous
forest in the landscape and the proportion of immigrants. The best regression model was one that included
both snow depth and proportion of coniferous forest, suggesting that the most suitable landscapes for
fishers had low snow and more coniferous forests.
Lastly, this same dataset was used in a graph theoretical framework to examine network structure for
evaluating habitat quality, gene flow, and population substructure (Garroway et al. 2009). The graph
theoretical framework is a new approach in the landscape genetics arena that can be used to evaluate
complex systems of connectivity that lead to system-level properties not readily discerned by examining
among population relationships. This analytical approach has been adopted in the fields of social network
analysis, neurobiology, and transportation efficiency network analysis (Costa et al. 2007). Basically every
complex network, in this case a network of connectivity among fisher populations (or groups of samples),
has very specific topological features that typify its connectedness and how it responds to perturbation
(Costa et al. 2007). Garroway et al. (2009) showed that the fisher network in Ontario displayed high levels
of clustering, and short mean path lengths connecting pairs of nodes (populations). Using the graph
theoretic approach also allowed exploration of the effect that removal of nodes had on system connectivity
and robustness. Garroway’s removal analysis suggested that harvest (removal of nodes) is unlikely to affect
- 32 -
Martes conservation genetics
Paper I
genetic connectivity given current conditions. In addition, Garroway et al. (2009) was able to show a
negative relationship between measures of node connectivity and both the proportion of immigrants into a
node and snow depth, confirming the previous results of Carr et al. (2007b).
Overall, these landscape genetic approaches are allowing us to test more detailed hypotheses regarding
gene flow through different environments. While these approaches are still nascent, they already have
confirmed some of our ecological ideas about what hinders and enhances connectivity among the Martes
species. In addition, once we have gained understandings of gene flow from these landscape genetic
models we can create putative corridor maps for target species (Cushman et al. 2009, Schwartz et al.
2009).
FUTURE DIRECTIONS
It is an exciting time to be working with molecular genetic data. The field of genomics is exploding which
is providing unprecedented power to ask detailed questions regarding relationships among populations
within a species, and the processes that created these patterns. Furthermore, we can use genomic data to
enhance our understanding of the ecological requirements of the Martes species. Additionally, the
development of new theoretical models and the use of computer simulations will significantly contribute
to the conservation biology of Martes through, for example, the integration of genetic or genomic data
into metapopulation frameworks and the developments of predictive models which incorporate both
environmental and genetic data sets (Bouchy et al. 2005, Nomura 2005). Using a Bayesian approach, the
integration of genetic and non-genetic data is also possible in order to go beyond the simple estimation of
parameters and tests of hypotheses about the factors that control demographic and genetic changes. In
particular, the development of Bayesian models aimed to infer historical population dynamics and
population parameters are extremely promising (Pertoldi and Topping 2004; Bach et al. 2006). We
portend the further development of genetic models that integrate both spatial variability (i.e.,
heterogeneous landscapes) and temporal variability (i.e., metapopulation dynamics), to examine how these
variations influence the genetic structure of populations and thereby our interpretations of genetic
structure.
The combination of individual-based models and genetics is just emerging now, but it will soon be feasible
to evaluate the impact of environmental changes on genetic composition of populations. Models should be
developed to address the unresolved questions in conservation genetics. The future promise of the
development of these theoretical models and the use of computer simulations is to support conservation
genetics investigations through: (a) modelling alternative scenarios for the dynamics of genetic diversity
- 33 -
CHAPTER 1
Introduction
within and among populations exposed to different environmental regimes and evaluation of short and
long-term risks; (b) linking the genotype with phenotype, for example, modeling how a given trait (lifehistory or morphological trait) would develop in a given scenario. If the information obtained can be
combined with empirical and molecular data, the models will be a powerful tool for understanding realworld dynamics.
In addition, we can now use landscape genetic approaches, including graph theoretic approaches to
evaluate how various environmental elements influence the flow of genes through landscapes. We can then
use these understandings to predict corridors under contemporary conditions and under future climate
change scenarios. For example, how will fisher populations interact with one another given a reduction in
snowpack in some regions and a change in forest types? Or how will existing connectivity change for pine
marten in Europe given anthropogenic development and habitat restoration projects?
As previously mentioned in this chapter, one new and very promising tool are SNPs that may avoid some
of the problems attached to microsatellites. SNPs hold the potential to significantly expand our ability to
survey both neutral variation as well as genes under selection in natural populations (Beja-Pereira et al.
2009). Furthermore, fast and inexpensive methods are continuously developed to screen hundreds or
thousands of SNPs per sample in populations (Chen and Sullivan 2003, Ellegren 2008, Wang et al. 2009).
SNP genotypes based on single nucleotide changes, are universally comparable and do not require
standardization across detection platforms. In contrast, it is difficult to compare microsatellite data sets
produced by different laboratories, do to inconsistencies in allele size calling (Vignal et al. 2003). This
enhanced ability to collaborate between laboratories in different countries or continents will aide in our
ability to understand Martes population dynamics.
Additionally, there will be in the near future the possibility to create for every Martes species of interest, a
subset of very informative SNPs (48 or 96 SNPs) that could be applied relatively cheap. The cost of SNP
microchip beads application is gradually decreasing, making its application available to a wider range of
users. The planned cost of a panel of 50 loci using VeraCode SNPs, SNPlex or Fluidigm EP1 system
would e.g. cost less than $10, which is 1/20 the amount of the application of 12 microsatellites with 4
PCR and genotyping repetitions. Given the developments in modeling, the new power provided by
cheaper and more molecular markers, we anticipate rapid advances in our understandings of the various
Martes species by the time of the next Martes synthesis.
- 34 -
Martes conservation genetics
Paper I
CONCLUSIONS
There are few generalities that can be made across the studies of Martes, as each species has evolved a
unique niche. However, we can see a few general patterns. For instance, it is clear that the complex glacial
histories of Europe and North America left refugia populations that are only recently coming back into
contact. Detailed phylogenetic studies have provided an understanding of how Martes have responded to
previous climate changes, and these studies are absolutely essential for distinguishing short-term
anthropogenic changes versus longer term climatic changes that have structured Martes populations.
In the field of landscape genetics, which examines the finer-scale movements of Martes we are confirming
field data that shows the importance of intact old forests with ample structure for several species.
Furthermore, for species such as fisher we can use both population genetic and landscape genetic data to
further elucidate understandings regarding their avoidance of areas with heavy snow. It will be critical to
predict how these areas will change given general circulation models, which often predict the
disappearance of snow pack.
Overall, we still have an incomplete taxonomic and evolutionary framework for a significant portion of the
Martes complex. The current knowledge about intraspecific genetics of genus Martes is limited and
devoted to only a few species (Table 1). However, the advance in DNA technology is producing a wealth
of data for intraspecific phylogenetic studies of Martes and there are concomitant developments in
analytical methods to deduce demographic history and evolutionary relationships and to test their
significance. In this case, whole-genome analyses will provide unprecedented phylogenetic resolution and
the power to distinguish even extremely closely related groups. For genera that have emerged relatively
recently, such as the Martes genus (Koepfli et al. 2008), whole genomes will provide fine-scale
differentiation. With next-generation sequencing technologies making sequencing cheaper and faster,
whole-genome intraspecific phylogenies will soon become a reality for a growing number of marten
species.
New scientific studies concerning several species and several geographic regions not studied, is needed to
improve our understanding about intraspecific genetics of Martes worldwide (Table 1). We hope that
knowledge gaps identified in this review will be addressed by research organizations in near future so that,
with the development of new techniques in this field, a more precise and more reliable data of all Martes
species are expected to be fulfilled during the next decade.
- 35 -
CHAPTER 1
Introduction
ACKNOWLEDGEMENTS
A. Ruiz-González holds a Ph.D. fellowship awarded by the Dept. of Education Universities and Research of the
Basque Government (Ref. BFI09.396) and is member of the Research group “Systematic, Biogeography and
Population Dynamics” of the University of the Basque Country supported by the Basque Goverment (Ref. IT31710;GIC10/76). Michael Schwartz was partially funded to conduct this research by a Presidential Early Career Award
for Science and Engineers and through funding from the Rocky Mountain Research Station’s Wildlife and
Terrestrial Ecosystems Program.
REFERENCES
Allendorf, F.W. and G. Luikart. 2007. Conservation and the Genetics of Populations. Blackwell Publishing, Malden
MA, USA.
Anderson, E. 1970.Quaternary Evolution of the Genus Martes (Carnivora, Mustelidae). Acta Zoologica Fennica 130:
1–132.
Avise, J.C., J. Arnold, R.M. Ball, E. Bermingham, T. Lamb, J.E. Neigel, C.A. Reeb, and N.C. Saunders. 1987.
Intraspecific phylogeography: the mitochondrial DNA bridge between pouplation genetics and systematics.
Annual Review of Ecology and Systematics 18: 489-522.
Bach, L.A. Thomsen, R., Pertoldi C. and V. Loeschcke. 2006. Evolution of density-dependent dispersal in an
individual-based adaptative metapopulation model including explicit kin competition and demographic
stochasticity. Ecological modelling 192: 658-666.
Balkenhol, N. L.P. Waits, and R.J. Dezznai. 2009. Statistical approaches in landscape genetics: an evaluation of
methods for linking landscape and genetic data. Ecography 32: 818-830.
Balloux, F. 2010. The worm in the fruit of the mitochondrial DNA tree. Heredity 104: 419-420.
Balmysheva, N.P. and L.L Solovenchuk. 1999a. Association between Mutations of Mitochondrial DNA Genes for
Citochrome b and NADH Dehydrogenase 5/6 in Sable Martes zibellina L. Russian Journal of. Genetics
35:1447–1451.
Balmysheva, N.P. and L.L. Solovenchuk. 1999b. Genetic Variation of the Mitochondrial DNA Gene Encoding
Cytochrome b in the Magadan Population of Sable Martes zibellina L. Russian Journal of. Genetics
35:1077–1081.
Beja-Pereira, A., R. Oliveira, P.C. Alves, M.K. Schwartz, and G. Luikart. (2009) Advancing ecological
understandings through technological transformations in noninvasive genetics. Molecular Ecology
Resources 9: 1279-1301.
Berg, W.E. 1982. Reintroduction of fisher, pine mar- ten and river otter. Pp. 159-173 in Midwest fur- bearer
management (G. C. Sanderson, ed.). Proceedings of Symposia 43rd Midwest Fish and Wildlife Conference,
Wichita, Kansas.
Bonin, A., Bellemain, E., Eidesen, P.B., Pompanon, F., Brochmann, C. and P. Taberlet. 2004. How to track and
assess genotyping errors in population genetics studies. Molecular Ecology 13: 3261-3173.
Bouchy, P., Theodorou, K. and D. Couvet. 2005. Metapopulation viability: influence of migration. Conservation
Genetics 6: 75-78.
Boulanger , J. S. Himmer, and C. Swan. 2004. Monitoring a grizzly bear population trend and demography using
DNA mark-recapture methods in the Owikeno Lake area of British Columbia. Canadian Journal of
Zoology 82:1267-1277
Bright, P.W. 2000. Lessons from lean beats: conservation biology of the mustelids. Mammal Review 30: 217–226.
Broquet, T., C.A. Johnson, E. Petit, I. Thompson, F. Burel, and J.M. Fryxell. 2006a. Dispersal and genetic structure
in the American marten, Martes americana. Molecular Ecology 15 : 1689-1697.
Broquet, T., N. Ray, E. Petit, J.M. Fryxell, and F. Burel. 2006b. Genetic isolation by distance and landscape
connectivity in the American marten (Martes Americana). Landscape Ecology 21:877-889.
- 36 -
Martes conservation genetics
Paper I
Broquet, T., Ménard, N. and E. Petit. 2007. Noninvasive population genetics: a review of samples source, diet,
fragment length and microsatellite motif effects on amplification success and genotyping error rates.
Conservation Genetics 8:249-260.
Brumfield, R.T., Beerli, P., Nickerson, D.A. and S.V. Edwards. 2003. The utility of single nucleotide
polymorphisms in inferences of population history. Trends in Ecology and Evolution 18:249-256.
Carr, D., Bowman, J., Kyle C.J., Tully S.M., Koen, E.L., Robitaille, J.-F. and P.J. Wilson. 2007a. Rapid
homogenization of multiple sources: genetic structure of a recolonizing population of fishers. Journal of
Wildlife Management 71:1214–1219.
Carr, D., J. Bowman, and P.J. Wilson. 2007b. Density-dependent dispersal suggests a genetic measure of habitat
suitability. Oikos 116: 629-635.
Carr, M., and S.A. Hicks. 1997. Are there two species of pine marten in North America? Genetic and evolutionary
relationships within Martes. In Martes: Taxonomy, Ecology, Techniques, and Management: Proceedings of
the Second International Martes Symposium, Edmonton, Canada, August 1995. Edited by G. Proulx,
H.N. Bryant, P.M. Woodard. Provincial Museum of Alberta, Edmonton. pp. 15–28.
Chen, X. And P.F. Sullivan 2003. Single nucleotide polymorphism genotyping: biochemistry, protocol, cost and
throughput. The Pharmacogenomics Journal 3:77-96.
Clark, T.W., Anderson, E., Douglas, C., M. Strickland. 1987. Martes americana. Mammalian Species, 289:1-8.
Costa, L.da F., F.A. Rodrigues, G. Travieso, and P.R. Villas Boas. 2007. Characterization of complex networks: a
survey of measurements. Advances in Physics 56, 167-242.
Cushman, S.A., K.S. McKelvey, M.K. Schwartz. 2009. Evaluating habitat connectivity and mapping of corridors
between Yellowstone National Park and the Canadian Border with landscape genetics and least cost path
analysis. Conservation Biology 23: 368-376.
Davison, A., Birks, J.D., Brookes, R.C., Messenger, J.E. and H.I. Griffiths. 2001. Mitochondrial phylogeography
and population history of pine martens Martes martes compared with polecats Mustela putorius. Molecular
Ecology 10: 2479–2488.
Dawson, N. 2010. Tracking historical diversification and contemporary structure in high latitude mesocarnivores.
Ph.D. dissertation, The University of New Mexico, New Mexico, USA.
Domingo-Roura 2002
Dawson, N. and J. Cook. This Volume. Behind the genes: diversification of North American Marten. Biology and
Conservation of Marten, Sables, and Fisher: a new synthesis. Aubry et al. Editors. Cornell University Press.
Domingo-Roura, X. 2002. Genetic distinction of marten species by fixation of a microsatellite region. Journal of
Mammology 83: 907–912.
Drew, R..E., Hallet, J.G., Aubry, K.B., Cullings, K.W., M.Koepfs, S., and W.J. Zielinski. 2003. Conservation
genetics of the fisher (Martes pennanti) based on mitochondrial DNA sequencing. Molecular Ecology
12:51–62.
Ellegren, H. 2008. Sequencing goes 454 and takes large-scale genomics into the wild. Molecular Ecology 17: 16291635.
Excoffier, L. and G. Heckel. 2006. Computer programs for population genetics data analysis: a survival guide.
Nature Reviews Genetics 7: 745-758
Forsey, E.S. and E.M. Baggs. 2001. Winter activity of mammals in riparian zones and adjacent forests prior to and
following clearcutting at Copper Lake, Newfoundland, Canada. Forest Ecological Management 145: 163–
171.
Forsey, O., Bissonette, J., Brazil. J., Curnew, K., Lemon, L., Mayo, L., Thompson, I., Bateman, L., and L.
O’Driscoll. 1995. National recovery plan for the Newfoundland marten. Recovery of Nationally
Endangered Wildlife Committee, Ottawa. Rep. No. 14.
Frankham, R. 1995. Conservation Genetics. Annual Review in Genetics 29: 305-327.
Frankham, R. 2005. Genetics and extinction (review article). Biological Conservation 126: 131-140.
Garroway, C.J., J. Bowman, D. Carr, and P.J. Wilson. 2008. Applications of graph theory to landscape genetics.
Evolutionary Applications 1:620–630.
Gibilisco, C.J. 1994. Distributional dynamics of American martens and fishers in North America. In Martens, sables,
and fishers: biology and conservation. Edited by S.W. Buskirk, A. Harestad, and M. Raphael. Cornell
University Press, Ithaca, N.Y. pp. 59–71.
Hall, E.R.1981. The Mammals of North America. Second Edition. John Wiley and Sons, New York, USA.
- 37 -
CHAPTER 1
Introduction
Hosoda, T., Suzuki, H., Harada, M., Tsuchiya, K:, Han, S.H., Zhang, Y.P., Kryukov , A.P., L.K. Lin. 2000.
Evolutionary trends of the mitochondrial lineage differentiation in species of genera Martes and Mustela.
Genes & Genetic Systems 75:259-267.
Hosoda, T., Suzuki, H., Iwasa M.A., Hayashida, M., Watanabe, S., Tatara, M., K. Tsushiya. 1999. Genetic
relationships within and between the Japanese marten Martes melampus and the sable M..zibellina, based on
variation of mitochondrial DNA and nuclear ribosomal DNA. Mammal Study 24:25–33
Hosoda, T., Suzuki, H., Tsuchiya, K., Lan, H., Shi, L. and A.P. Kryukov. 1997. Phylogenetic relationships within
Martes based on nuclear ribosomal DNA and mitochondrial DNA. In (G. Proulx, H. N. Bryant, and P. M.
Woodard., eds.) Martes: Taxonomy, Ecology, Techniques, and Management. Pp. 3-14. Provincial Museum
of Alberta, Edmonton.
Inoue, T., Murakami, T., Abramov, A,V., and Masuda, R. 2010. Mitochondrial DNA control region variations in
the sable Martes zibellina of Hokkaido Island and the Eurasian Continent, compared with the Japanese
marten M. melampus. Mammal Study 35: In press.
Koepfli This Volume. Patterns and processes of diversification in the Guloninae within the Mustelidae as inferred
from phylogenomic data. In Biology and Conservation of Marten, Sables, and Fisher: a new synthesis.
Aubry et al. Editors. Cornell University Press.
Knaus, B.J., Cronn, R., A. Liston, K. Pilgrim, and M.K. Schwartz. In Review. Genomic data illuminates lineages of
conservation concern. BMC Evolution.
Koepfli, K.P., Deere, K.A., Slater, G.J. Begg, K., Grassman, L., Lucherini, M., Veron, G. and R.K. Wayne. 2008.
Multigene Phylogeny of the Mustelidae: Resolving Relationships, Tempo and Biogeographic History of a
Mammalian Adaptive Radiation. BMC Biol. 6: 1–22
Kurose N., Masuda, R., Siriaroonrat, B., M.C. Yoshida .1999. Intraspecific variation of mitochondrial cytochrome b
gene sequences of the Japanese marten Martes melampus and the sable Martes zibellina (Mustelidae,
Carnivora, Mammalia) in Japan. Zoological Sciecne 16: 693–700
Kyle, C.J. and C. Strobeck. 2003. Genetic homogeneity of Canadian mainland marten populations underscores the
distinctiveness of Newfoundland pine martens (Martes americana atrata). Canadian Journal of Zoology
81:57–66.
Kyle, C.J., Davis, C.S. and C. Strobeck. 2000. Microsatellite analysis of North American pine marten (Martes
americana) populations from the Yukon and Northwest Territories. Canadian Journal of Zoology 78:
1150–1157.
Kyle, C.J., Davison, A. and C. Strobeck. 2003. Genetic structure of European pine martens (Martes martes) and
evidence for introgression with M. americana in England. Conservation Genetics 4: 179-188.
Kyle, C.J., Robitaille, J.F. and C. Strobeck. 2001. Genetic variation and structure of fisher (Martes pennanti)
populations across North America. Molecular Ecology, 10: 2341-2347.
Lewis, J.C., and G.E. Hayes. 2004. Feasibility assessment for reintroducing fishers to Washington. Washington
Department Fish and Wildlife, Olympia, WA 70pp.
Long , R. and P. Mackay. In Press. Noninvasive methods for surveyingand monitoring martens, fisher and sables. In
Biology and Conservation of Marten, Sables, and Fisher: a new synthesis. Aubry et al. Editors. Cornell
University Press.
Luikart, G., N. Ryman, D.A. Tallmon, M.K. Schwartz, and F.W. Allendorf. 2010. Estimation of census and
effective popualtion sizes: the increasing usefulness of DNA-based approaches. Conservation Genetics:11
355-373.
Lynch, M. 1996. A quantitative-genetic perspective on conservation issues. In: Conservation Genetics. Case Histories
from Nature, (eds. Avise, J. C. and Hamrick, J. L.) pp. 471-501. Chapman & Hall, New York.
Malyarchuk, B.A., Petrovskaya, A.V., M.V. Derenko. 2010. Intraspecific structure of sable Martes zibellina L.
Inferred from nucleotide variation of the mitochondrial DNA cytochrome b gene. Russian Journal of
Genetics 46:1, 64-68
Manel S., M.K. Schwartz, G. Luikart, and P. Taberlet (2003) Landscape genetics: combining landscape ecology and
population genetics. Trends in Ecology and Evolution, 18, 189–197.
Masuda, R. 2009. Martes melampus (Wagner, 1840). In “The Wild Mammals of Japan (Ohdachi, S.D., Ishibashi,
Y., Iwasa, M.A., Saitoh, T., eds.)”, pp. 250-251, Shoukadoh, Kyoto.
McComb, B. B. Zuckerberg, D. Vesely, and C. Jordan. 2010. Monitoring animal populations and their habitats: a
practitioner’s guide. CRC Press, Boca Raton, FL, USA.
- 38 -
Martes conservation genetics
Paper I
McGowan, C., Howes, L.A., and W.S. Davidson. 1999. Genetic analysis of an endangered pine marten (Martes
americana) population from Newfoundland using randomly amplified polymorphic DNA markers.
Canadian Journal of Zoology 77: 661–666.
McRae, B. H., and P. Beier. 2007. Circuit theory predicts gene flow in plant and animal populations. Proceedings of
the National Academy of Sciences of the United States of America 104:19885–19890.
McRae, B.H., and Shah, V.B. 2009. Circuitscape User Guide. ONLINE. The University of California,
Mergey, M. 2007. Réponses des populations de martres d’Europe (Martes martes) à la fragmentation de l'habitat :
mécanismes comportementaux et consequences. Disseration University of Reims Campagne-Ardenne,
France. P. 211.
Merriam, C.H. 1890. Description of a new marten (Mustela caurina) from the north-west coast region of the United
States. North American Fauna 4:27-29.
Mitchell-Jones, A.J., Amori, G., Bogdanowicz, W., Kriimagetufek, B., Reijnders, P.J.H., Spitzenberger, F., Stubbe,
M., Thissen, J.B.M., Vohralik, V. and J. Zima. 1999. The atlas of European mammals. London: Academic
Press.
Monakhov VG. 2001. Phenetic analysis of aboriginal and introduced populations of sable (Martes zibellina) in
Russia. Russian Journal of Genetics 37:1074–1081.
Monakhov, G.I. 1976. Geographic variability and taxonomic structure of Sable in the USSR. Pages: 54-86 In: Proc.
Res. Inst. for Game Farming and Animal Breeding, Kirov, Russia.
Mullins, J. 2010. Estimating the size and structure of pine marten populations using non-invasive genetic sampling.
Department of Chemical and Life Sciences, Waterford Institute of Technology.
Murakami, T. 2009. Martes zibellina (Linnaeus, 1758). In “The Wild Mammals of Japan (Ohdachi, S.D., Ishibashi,
Y., Iwasa, M.A., Saitoh, T., eds.)”, pp. 252-253, Shoukadoh, Kyoto, Japan.
Murakami, T., Asano, M., and N. Ohtaishi. 2004. Mitochondrial DNA variation in the Japanese marten Martes
melampus and Japanese sable, Martes zibellina. Japanese Journal of Veterinary Research 51:135-142
Nomura, T. 2005. Methods for minimizing the loss of genetic diversity in conserved populations with overlapping
generations Conservation Genetics 6: 655-663.
Ouborg, N.J., F. Angeloni, and P. Vergeer. 2010a. An essay on the necessity and feasibility of conservation
genomics. Conservation Genetics 11:643-653.
Ouborg J., C. Pertoldi, V. Loeschcke, R. Bijlsma, P.W. Hedrick. 2010b Conservation genetics in transition to
conservation genomics, Trends in Genetics 26: 177-187.
Palsboll, P.J., M. Berube, and F.W. Allendorf. 2007. Identification of management units using population genetic
data. Trends in Ecology and Evolution. 22:11-16.
Pavlinov, I.Y.and O.L.Rossolimo. 1979. Geographic Variability and Intraspecies Systematics of Sable (Martes
zibellina L.) in the USSR. Pages 241-256 in Proc. Zool. Museum Moscow State University, Moscow,
Russia.
Pease, K. M., A. H. Freedman, J. P. Pollinger, J. E. McCormack, W. Buermann, J. Rodzen, J. Banks et al. 2009.
Landscape genetics of California mule deer (Odocoileus hemionus): the roles of ecological and historical
factors in generating differentiation. Molecular Ecology 18:1848–1862.
Pertoldi, C. and C. Topping. 2004. Impact assessment predicted by means of genetic agent-based modelling Critical
Reviews in Toxicology 34: 487-498.
Pertoldi, C., Bijlsma, R. and V. Loeschcke. 2007. Conservation genetics in a globally changing environment: present
problems, paradoxes and future challenges. Biodiversity and Conservation, 16: 4147-4163.
Pertoldi, C., Hansen, M.M., Loeschcke, V., Madsen, A.B., Jacobsen, L. and H. Baagoe. 2001. Genetic consequences
of population decline in European Otter Lutra lutra: An assessment of microsatellite DNA variation in
Danish otters from 1883 to 1993. Proceedings B. Royal Society 268: 1775-1781.
Pertoldi, C., Madsen, A.B., Barker, S.F., Jørgensen, H., Randi, E., Muñoz, J., Baagoe, H. and V. Loeschcke. 2008a.
Spatio-temporal population genetic survey of the Danish pine marten (Martes martes). Biological Journal of
the Linnean Society, 93: 457-464.
Pertoldi, C., Munoz, J., Madsen, A.B., Barker, J.S.F., Andersen, D.H.H., Baagøe, J., Birch, M. and V. Loeschcke.
2008b. Genetic variability in the mitochondrial DNA of the Danish Pine marten. Journal of Zoology 276:
168-175
- 39 -
CHAPTER 1
Introduction
Petrovskaya, A.V. 2007. Genetic Structure of the Sable Martes zibellina L. Populations from Magadan Oblast as
inferred from Mitochondrial DNA variation. Russian Journal of Genetics 43:424–429.
Robitaille, J.F. and K. Aubry. 2000. Occurrence and activity of American martens Martes americana in relation to
roads and other routes. Acta Theriologica 45: 137–143.
Ruiz-González, A., Madeira, M.J., Randi, E., Abramov, A., and B.J. Gómez Moliner. 2009. Phylogeography of the
European pine marten (Martes martes). In 5th International Martes Symposium Biology and Conservation
of Martens, sables, and Fishers: a New Synthesis. Seattle, Washington, USA:
Ryynänen, H.J., Tonteri, A., Vasemägi, A. And C.R. Primmer. 2007. A comparison of biallelic markers and
microsatellites for the estimation of population and conservation genetic parameters in Atlantic Salmon
(Salmo salar). Journal of Heredity 7: 692-704.
Sato J.J., Hosoda, T., Wolsan, M., Tsuchiya, K., Yamamoto, Y. and H. Suzuki. 2003. Phylogenetic relationships and
divergence times among mustelids (Mammalia: Carnivora) based on nucleotide sequences of the nuclear
interphotoreceptor retinoid binding protein and mitochondrial cytochrome b genes. Zoological Sciences
20:243–264
Sato, J.J., Yasuda, S.P. and T. Hosoda. 2009. Genetic Diversity of the Japanese Marten (Martes melampus) and Its
Implications for the Conservation Unit. Zoological Science 26:457–466
Schwartz MK and Monfort SL. 2008. DNA and Endocrine Sampling. In: Long RA, MacKay P, Ray JC, Zielinski
WJ (eds) Noninvasive survey methods for North American carnivores. Island Press, Washington D.C.
Schwartz MK, Luikart G, and Waples RS. 2007. Genetic Monitoring as a promising tool for conservation and
management. Trends in Ecology and Evolution 22: 25-33.
Schwartz MK, Tallmon DA, and Luikart G. 1999. Using genetics to estimate the size of wild populations: many
methods, much potential, uncertain utility. Animal Conservation 2: 320-322
Schwartz, M.K. 2007.Ancient DNA confirms native Rocky Mountain fisher (Martes pennanti) avoided early 20th
Century extinction. Journal of Mammalogy 88:921–925
Schwartz, M.K., J.P. Copeland, N.J. Anderson, J.R. Squires, R.M. Inman, K.S. McKelvey, K.L. Pilgrim, L.P. Waits,
and S.A. Cushman. (2009) Wolverine gene flow across a narrow climatic niche. Ecology 90:3222-3232.
Seddon, J.M., Parker, H.G., Ostrander, E.A. and H. Ellegren. 2005. SNPs in ecological and conservation studies: a
test in the Scandinavian wolf population. Molecular Ecology 14:503–511.
Slauson, K.M., Zielinski, W. J., and K.D. Stone. 2008.Characterizing the molecular variation among American
marten (Martes americana) subspecies from Oregon and California. Conservation Genetics 10:1337-1341
Small, M.P., Stone, K.D. AND J.A. Cook. 2003. American marten (Martes americana) in the Pacific Northwest:
population differentiation across a landscape fragmented in time and space. Molecular Ecology 12:89–103.
Smith, S., Aitken, N., Schwarz, C. and A. Morin 2004. Characterization of 15 single nucleotide markers for
chimpanzees (Pan troglodytes). Molecular Ecology Notes, 4: 348-351.
Snyder, J. 1986. Updated status report on the marten (Newfoundland population) Martes americana atrata.
Committee on the Status of Endangered Wildlife in Canada, Ottawa, Ont.
Sommer, R. and N. Benecke, (2004) Late- and post-glacial history of the mustelidae in Europe. Mammal Review
34:249–284.
Stone, K.D. and J.A.Cook,. 2002. Molecular Evolution of Holarctic Martens (Genus Martes, Mammalia: Carnivora:
Mustelidae). Molecular Phylogenetic and Evolution. 24:169–179.
Stone, K.D., Flynn, R.W., and J.A. Cook. 2002. Post-glacial colonization of northwestern North America by the
forest-associated American marten (Martes americana, Mammalia: Carnivora: Mustelidae). Molecular
Ecology 11: 2049–2063.
Swanson, B.J., L.R. Peters, and C.J. Kyle. 2006. Demographic and genetic evaluation of an American marten
reintroduction. Journal of Mammalogy 87:272–280.
Swanson, B.J. and C.J. Kyle. 2007. Relative influence of temporal and geographic separation of source populations in
a successful marten reintroduction. Journal of Mammalogy 88: 13461348.
Tallmon, D.A., G. Luikart, R.S. Waples. (2005) The alluring simplicity and complex reality of genetic rescue. Trends
in Ecology & Evolution 16: 330–342.
Tallmon, D.A., D. Gregovich, R. Waples, C.S. Baker, J. Jackson, B. Taylor, F. Archer, F.W. Allendorf, and M.K.
Schwartz. 2010. When are genetic methods useful for estimating contemporary abundance and detecting
population trends? Molecular Ecology Resources. doi: 10.1111/j.1755-0998.2010.02831.x
- 40 -
Martes conservation genetics
Paper I
Thompson, I.D. 2000. Forest vertebrates of Ontario: patterns of distribution. Pages 54–73 in A. H. Perera, D. L.
Euler, and I. D. Thompson, editors. Ecology of a managed terrestrial landscape: patterns and processes of
forest landscapes in Ontario. UBC Press, Vancouver, British Columbia, Canada.
Thompson, I., and J. Fryxell. This Volume. review of marten habitat requirements in North America. Biology and
Conservation of Marten, Sables, and Fisher: a new synthesis. Aubry et al. Editors. Cornell University Press.
Vignal, A., Milan, D., SanCristobal, M. And A. Eggen. 2002. A review on SNP and other types of molecular
markers and their use in animal genetics. Genetics Selection Evolution 34:275–305.
Vinkey, R. 2006. When reintroduction efforts are augmentations:the genetic legacy of fisher (Martes pennanti) in
Montana. Journal of Mammalogy 87:265–271.
Wang, j, Lin.M., Crenshaw, A., Hutchinson, A., Hicks, B., Yeager, M., Berndt, S., Huang, W-Y., Hayes, R.B.,
Chanock, S.J., Jones, R.C. and R. Ramakrishnan. 2009. High-throughput single nucleotide polymorphism
genotyping using nanofluidic Dynamic Arrays. BMC Genomics 10: 561-574.
Waples, R.S., and O. Gaggiotti (2006) What is a population? An empirical evaluation of some genetic methods for
identifying the number of gene pools and their degree of connectivity. Molecular Ecology 15: 1419-1439.
Warheit, K.I. 2004. Fisher (Martes pennant) control region sequences from Alberta, and re-analysis of sequences
from other regions in North America: recommendations for the reintroduction of fishers into Washington.
Draft Report. Washington Department of Fish and Wildlife, Olympia, WA, USA.
Wasserman, T.N. 2008. Habitat relationships and gene flow of Martes americana in northern Idaho. Thesis, Western
Washington University. P.128.
Webster, J.A. 2001. A review of the historical evidence of the habitat of the pine marten in Cumbria. Mammal
Review 31: 17–31.
Williams, R.N., Rhodes O.E. Jr, Serfass T.L. 2000. Assessment of genetic variance among source and reintroduced
fisher populations. Journal of Mammalogy 81:895-907.
Williams, B.W. and K.T. Scribner. 2007. Demographic and genetic evaluation of an American marten
reintroduction: a comment. Journal of Mammalogy 88: 1342-1345.
Wilson, D.E. and D.M. Reeder. 1993. Mammal Species of the World. A Taxonomic and Geographic Reference.
Second edition. Smithsonian Institution Press, Washington, DC.
Wisely, S.M., Maldonado, J.M. and M.C. Fleischer. 2004. A technique for sampling ancient DNA that minimizes
damage to museum specimens. Conservation Genetics 5: 105-107.
Wright, P.L. 1953.Intergradation between Martes americana and Martes caurina in Western Montana. Journal of
Mammalogy 34: 74-86.
- 41 -
CHAPTER 1
Introduction
- 42 -
CHAPTER
1
Phylogeography
PAPER II
New insights into the cryptic northern glacial refugia:
Phylogeography of the forest dwelling European pine marten
(Martes martes)
- 43 -
2
CHAPTER 2
Phylogeography
- 44 -
Phylogeography of the European pine marten
Paper II
PAPER II
New insights into the cryptic northern glacial refugia:
Phylogeography of the forest dwelling European pine marten
(Martes martes)
ABSTRACT
The role of southern European peninsulas as glacial refugia for temperate species has been widely
established, but the role of cryptic northern refugia is being only recently addressed. Here we describe the
phylogeographic pattern of the forest dwelling European pine marten (Martes martes) in order to assess the
impact of Quaternary glaciations on the genetic structure and to identify location of refugia as well as
post-glacial recolonization routes. We used mtDNA sequences 1600bp long generated from 287 samples,
which were collected from 21 countries throughout the pine marten distribution. Aiming at clarifying the
relationships between M. martes and the sable M. zibellina in Fennoscandia and Russia, 10 sable samples
were also included. Our results reveal the presence of 69 different haplotypes for M. martes and 10 for M.
zibellina, which are split into three major assemblages: Mediterrenean, central-northern European and
Fennoscandian-Russian clades, each of them related to specific biogeographic regions which could
probably represent different ecotypes. The Mediterranean phylogroup apparently did not contribute to the
postglacial recolonization of most of the Palaearctic range of the species, suggesting continuous gene flow
across southern Europe during the late Pleistocene. It seems that most of Europe was rather colonized by
the central-northern European phylogroup probably surviving the last glaciations in northern cryptic
refugia, as it has been previously proposed by paleontological data. A highly divergent phylogroup has
been discovered in Fennoscandia-Rusia, wich comprises specimens from both M. martes and M. zibellina
morphospecies. All divergences estimates fall within the Pleistocene suggesting that glacial-interglacial
cycles played an important initiating phylogeographic differentiation as well as sculpting this pre-existing
phylogeographic variety into today’s sister species M. martes and M. zibellina. Overall, our study indicates
a complex phylogeographic history for M. martes indicating a mixed pattern of recolonization of northern
Europe from both Mediterranean and non-Mediterranean refugia.
Keywords: Phylogeography, cryptic refugia, Martes martes, glacial refugia, mitochondrial DNA,
Quaternary glaciations
- 45 -
CHAPTER 2
Phylogeography
INTRODUCTION
The dramatic climate changes during the Quaternary have had strong impacts on the range dynamics of
species within the Palaearctic region (Webb & Bartlein 1992; Hewitt 2000; Hewitt 2004). It has long
been recognized that the species living in temperate forest habitats, were forced to shift their distribution
range, so that they would have survived mainly in the glacial maxima in Mediterranean refuge areas
(Taberlet et al. 1998; Hewitt 2001; Randi 2007). Interglacial and postglacial recolonizations of central and
northern Europe could therefore have arisen from these Mediterranean glacial refugia (Taberlet et al.
1998; Hewitt 2001). Although different species may have had different refugia and colonisation routes
according to their ecological and biogegraphical traits (Bhagwat & Willis 2008; Stewart et al. 2010), this
pattern of glacial survival in the South, followed by post-glacial recolonization of northern regions, seems
to be a general pattern among a variety of temperate taxa (Taberlet et al. 1998; Hewitt 2001). More
recently, however, in addition to traditional southern refugia for temperate species, cryptic refugia have
been proposed in the North during glacials (i.e glacial refugia for temperate taxa situated at higher
latitudes than the expected areas of suitable habitat to the South; Stewart & Lister 2001; Stewart et al.
2010). Indeed, the cryptic northern refugium hypothesis (Stewart & Lister 2001) has received significant
support since its publication (Willis & van Andel 2004; Bhagwat & Willis 2008; Provan & Bennett 2008;
Zoltán 2010), with phylogeographic studies finding evidence for northern refugia in various temperate
organisms (Jaarola & Searle 2002; Deffontaine et al. 2005; Kotlik et al. 2006, Saarma et al. 2007; Teacher
et al. 2009). Further, evidence in support of this hypothesis has come from the study of mammal fossil
records (Sommer & Benecke 2004; Sommer & Nadachowski 2006), fossil pollen data and macrofossil
remains (Willis & van Andel 2004) and species distribution modelling (Svenning et al. 2008; Fløjgaard et
al. 2009) which suggested relatively widespread distributions for some mammals and tree species to the
north of the traditional southern refugia during the Last Glacial Maximum (LGM).
Recently, there have been attempts to reconstruct the glacial refugia of mustelids on the basis of fossil
evidence that in addition to the traditional Mediterranean refugia suggest the existence of a cryptic glacial
refuge in the Carpathians for the forest-dwelling European pine marten (Sommer & Benecke, 2004,
Sommmer & NadachowskI 2006). Moreover, recent studies using both molecular and fossil data for the
bank vole (Myodes glareorus) evidenced the existence of central European refugia for this species
(Deffontaine et al. 2005; Kotlik et al 2006). Taking into account that the bank vole is one of the main
prey species of the pine marten (Zalewski et al. 2004), we could expect a similar phylogeographic pattern
for the pine marten. Moreover, Bhagwat & Wills (2008) suggest that species that have persisted in
northern refugia have shared biogreographical traits that matched with those found in the European pine
- 46 -
Phylogeography of the European pine marten
Paper II
marten: a present-day northern distribution, small body size and cold tolerance. Taking into account all
these evidences suggesting for the existence of northern cryptic refugia, the time is ripe to ask whether the
pine marten have persisted only in southerly refugia or have had populations farther north in Europe
during glacials.
The aim of this study is to detail the phylogeographic pattern of the European pine marten (Martes
martes), a carnivore species closely associated with forest habitats (Proulx et al. 2004; Zalewski &
Jędrzejewski 2006). The pine marten (Martes martes) occurs throughout much of Europe and northern
and central Asia, from northern Portugal to western Siberia (Proulx et al. 2004). The pine marten is
generally associated with forest habitats, mainly mature coniferous and mixed forests (Proulx et al. 2004).
The phylogeography of the pine marten is poorly known. Davison et al. (2001) have suggested that the M.
martes populations currently distributed in central and northern Europe originated from several different
refugia with subsequent admixture. However, this study gave no clear clues for the post-glacial
recolonization of central Europe, because it was based on a small fragment of mtDNA (325 base pairs, bp)
that were not enough to resolve the species phylogeography. Moreover they did not analyse enough
samples for the main Mediterranean refuge areas (Balkans, Iberian Peninsula and Italy) and from the
eastern Russian populations. Here, we present a more comprehensive study in terms of number of
specimens (n=287) and length of mtDNA sequence (1600bp). This study enlarges the number of
specimens and the range of distribution of previous phylogeographic works, getting to Scandinavia in the
North, Russia in the East, and the Iberian Peninsula in the South-West. Sable (Martes zibellina) samples
were also included to better understand the relationships between M. martes and the closely related M.
zibellina in Fennoscandia and Russia. Indeed, recent studies (Koepfli et al. 2008), support that the
subgenus Martes diversified during the Plio-Pleistocene and recognized M. martes and M. zibellina as sister
species within this subgenus. Thus, additional analyses, including population genetic-level sampling,
between these related species, will be needed to confidently resolve relationships among these recently
evolved species (Koepfli et al. 2008). Thus, we aim to 1) identify the main phylogeographic patterns in
this species; 2) reconstruct the post-glacial colonization routes of central-Europe and 3) obtain first
information on genetic structure of eastern European Martes populations with special emphasis on the
genetic relationship between M. martes and M. zibellina
- 47 -
CHAPTER 2
Phylogeography
MATERIAL AND METHODS
Samples and Laboratory procedures
Tissue and hair specimens were collected from 287 pine martens from throughout 21 countries that
correspond to the main areas of its distribution range (Table 1 and Fig.1). We also added 10 sable samples
from Russia. These specimens were obtained from collaborators and museums collections and were used to
isolate DNA from M.martes and M.zibellina (see Supplementary material 1). DNA was isolated from
tissues and hairs using the Qiagen DNeasy® Tissue DNA extraction kit according to the manufacturer’s
instructions.
The mitochondrial DNA region selected in this study includes the final part of the cytochrome b gene,
tRNAPro, tRNAThr, the control region (D-loop) and the initial part of rRNA12S used on a previously
work about genetic variability of Mustela putorius in Europe (Pertoldi et al. 2006). This fragment of c.a.
1600
bp
length
was
amplified
AGAACACCCATTCATCATTATCG-
3´),
using
and
the
the
forward
reverse
primer
primer
LutbF
LLU12SH91
(5´(5’-
CTAGAGGGATGTAAAGCA CCG- 3´) (Pertoldi et al. 2006). The standard PCR amplifications were
conducted in 15 μL reactions containing 1 μL diluted template DNA, 3.2 pmol of each primer, 1.75μM
dNTP, 1.33 μM MgCl2, 1.56 μL buffer STR 10 and 0.6U Taq DNA polymerase using the following
cycling conditions: an initial denaturing step at 94 ºC for 5min; 35 cycles of denaturing at 94 ºC for 50 s,
annealing at 58.5 ºC for 45 s and extending at 72 ºC for 90 s; and a final extending step of 72 ºC for 10
min.
PCR products were purified using EXO-SAP IT (USB, Cleveland, OH, USA) and sequenced using the
BigDye Terminator Kit V1.1 (Applied Biosystems, Foster City, CA, USA) in an ABI PRISM Model 3130
Genetic Analyzer (Applied Biosystems). Eelectropherograms were visually inspected using SeqScape 2.5
(Applied Biosystems) and nucleotide sequences were aligned using the default parameters of CLUSTAL X
version 2.0 (Larkin et al. 2007) and manually checked in the BIOEDIT sequence editing program v. 5.0.9
(Hall 1999). The minisatellite repetition of the control region [(TACGCACACG)-n] was removed from
the phylogenetic analysis to reduce ambiguous sites with the selected outgroups.
Phylogenetic analyses
The data set used for phylogenetic analysis includes the haplotype sequences of the selected mtDNA
region that were obtained from 287 pine martens and 10 sables. Three related species were selected as
outgroups: Mustela putorius (the polecat; AY962040), Martes flavigula (the yellow throated marten;
- 48 -
Phylogeography of the European pine marten
Paper II
FJ719367) and Gulo gulo (wolverine NC_009685). Number of polymorphic sites, transitions and
transversions, and haplotype (h) and nucleotide (π) diversities were obtained with ARLEQUIN version 3.0
(Excoffier et al. 2005).
Phylogenetic reconstructions were performed by a distance method using the neighbour-joining algorithm
(NJ) (Saitou & Nei 1987) and using maximum-parsimony criterion (MP) (Fitch 1971) algorithm
implemented in PAUP version 4.0b10 (Swofford 2002). For distance analyses, the HKY85 model with rate
heterogeneity and invariable sites (α= 0.732, I= 0.8693) was selected as the best-fit model of nucleotide
substitution for the molecular data set by the Akaike information criteria approach using MODELTEST 3.6
(Posada & Crandall 1998). Therefore, we used this model and parameters for inferring distance matrices.
MP analysis was conducted with the heuristic search algorithm, tree-bisection–reconnection (TBR)
swapping and a maximum number of trees constrained to 1000. Phylogenetic trees were rooted with
homologous region for the selected outgroups (Mustela putorius, Martes flavigula and Gulo gulo). The
robustness of the trees was assessed by bootstrap resampling (BS) (10000 random replications for NJ
analysis and 5000 for MP; Felsenstein 1985).
We also performed a Bayesian phylogeny estimation using MRBAYES 3.0b4 (Huelsenbeck & Ronquist
2001). Metropolis-coupled Markov chain Monte Carlo sampling was performed using four chains run for
2,000,000 iterations and using the most suitable model determined by MODELTEST. Bayesian posterior
probabilities(BPP) were picked from the 50% majority rule consensus of trees sampled every 20
generations, after removing trees obtained before the chains reached an apparent plateau (‘burn in’
determined by empirical checking of likelihood values). The whole procedure was repeated three times
starting from different random trees and the tree topologies obtained were the same.
Phylogeographic analyses
The NETWORK software version 4.1.0.6 (Bandelt et al. 1999; http.www. fluxus-engineering.com) was
used to construct a median joining (MJ) network. MJ is a powerful method that reconstructs phylogenies
based on intraspecific genetic differentiation (Posada & Crandall, 1998; Bandelt et al. 1999). The data
matrix included the nucleotide sequences of the mtDNA region from all the martens and sables
sequenced.
The genetic structure of populations was examined using an analysis of molecular variance (AMOVA)
performed in the ARLEQUIN version 3.0 (Excoffier et al. 2005). AMOVA was conducted at three
hierarchical levels of populations subdivision: among geographical groups; (see Table 1 and Fig. 1), among
- 49 -
CHAPTER 2
Phylogeography
populations within regional groups and within populations (See Table 1 for population designation). The
significance of these parameters were estimated by 10 000 permutations of the distance matrix.
Demographic analysis
Demographic histories of different phylogroups were inferred by a pairwise mismatch distribution analysis
between individuals (Rogers & Harpending 1992) computed under a population growth-decline model in
DNASP version 5.0 (Librado & Rozas 2009). Multimodal distributions were consistent with demographic
stability, while sudden expansion would generate a unimodal pattern (Slatkin & Hudson 1991).
Hypotheses of demographic expansion were tested using Fu and Li’s F (Fu & Li 1993) and Tajima’s D
statistics (Tajima 1989). Significances for FS statistics were obtained by means of coalescent simulations of
a panmictic population of constant size conditioning on the number of segregating sites. For each case,
1000 simulations were run in DNASP and the number of trees with values of interest statistic equal or
more extreme than the observed value was recorded.
Estimation of divergence times
We estimated divergence times of splits using the Bayesian relaxed phylogenetic approach implemented in
BEAST v1.4.6 (Drummond & Rambaut 2007). Analyses were performed using the HKY model of
nucleotide substitution (Hasegawa et al 1985). Rate variation among sites was modelled using a gamma
distribution with four rate categories. The uncorrelated lognormal relaxed molecular clock model was used
to estimate substitution rates for all nodes in the tree, with uniform priors on the mean (0, 100) and
standard deviation (0, 10) of this clock model. We employed the expansion growth as the coalescent prior
and, with the ingroup assumed to be monophyletic with respect to the outgroup.
Molecular dating was derived using as calibration point the age of the fossil record of the extinct M. vetus,
which has been considered ancestral to both M. martes and M. zibellina (Anderson 1994). This date ranges
from almost the beginning of the Pleistocene to about 400 kya (kilo years ago) (Wolsan 1993). A
lognormal distribution suitable for modelling fossil data (Ho 2007) was used as prior with parameter
values of 300 kya as the minimum age (lower bound parameter), 400 kya as the mean and the standard
deviation of the distribution was chosen to 1. Moreover, we set the mean of the normal distribution of the
root height prior to 4.8 Mya (million years ago), assuming this time as the time of divergence between M.
flavigula and the subgenus Martes (Koepfli et al. 2008) with a standard deviation of 1.0 Mya. Three
independent MCMC runs of 20,000,000 steps were performed. Samples from the three chains, which
yielded similar results, were combined to estimate the posterior distribution of the substitution model and
- 50 -
Phylogeography of the European pine marten
Paper II
tree model parameters, as well as node ages. Analyses of these parameters in TRACER 1.4 (Rambaut &
Drummond 2007) suggested that the number of MCMC steps was more than adequate, with effective
sample sizes of all parameters often exceeding 100 and Tracer plots showing strong equilibrium after
discarding burn-in.
To compare the genetic findings with subfossil records, spatial and temporal information on the
distribution of the pine marten during the Pleistocene was obtained from Sommer & Bennecke (2004)
and Sommer & Nadachowski (2006).
RESULTS
Pattern of sequence variation
We identified a total of 69 haplotypes among the 287 pine marten specimens analyzed. The 10 analyzed
Martes zibellina samples showed 10 different haplotypes. We included two genebank sequences that
correspond to a new sable haplotype (Mz11). In the alignment comprising both species (1566 bp), there
were 95 variable sites of which 59 were parsimony informative. The average transitions/transversion ratio
was = 18.4. When considering M. martes haplotypes only (excluding M. martes haplotypes belonging to
Fennoscandia-Rusian phylogroups; see below), there were 47 variable sites of which 25 were parsimony
informative.
Phylogenetic and phylogeographic analyses
The geographic distribution and frequency of the 69 and 11 mtDNA haplotypes found in the 287
European pine martens and the 10 sables is shown in Table 1. The haplotype distribution clearly
differentiated sequences from three main geographic regions: (i) Southern Europe (i.e Mediterranean
peninsulas), (ii) Central-Northern Europe, and (iii) Fennoscandia-Russia. The first group (Haplotypes:
Mm1 - Mm27) included unique haplotypes discovered only in the three main Mediterranean peninsulas
(Mm1 – Mm8, Mm10 - 17, Mm19), some shared haplotypes between southern Europe and CentralNorthern Europe (Mm9 and Mm18) and some closely related haplotypes that have been discovered only
in Central-Northern Europe (Mm21 – Mm27) and Ireland (Mm20). The second group (Mm28-Mm55)
included haplotypes mainly distributed in Central-Northern Europe, some of them shared across a wide
geographic range of this region (Mm29 and Mm31), Fennoscandia (Mm31, Mm45-Mm46), Russia
(Mm49-Mm55) and Scotland (Mm28). The northern parts of Fennoscandia and Russia include unique
haplotypes not found in any other regions where both M. martes and M. zibellina haplotypes were mixed
- 51 -
CHAPTER 2
Phylogeography
together (Mm51-59 and Mz1-8). Twenty-nine out of the 69 haplotypes detected in M. martes specimens
were restricted to a single individual.
The NJ reconstruction of phylogenetic relationships between haplotypes is shown in Fig. 2, the MP, and
Bayesian trees showing identical topologies. The pine martens split into two major groups: a
Fennoscandian-Russian (FNR) clade (BS: ≈96%, BPP: 1.0 including haplotypes: Mm56-Mm65 & Mz1Mz11) and a large clade grouping all the other pine martens haplotypes (Mm1-Mm55) (BS: 100%, BPP:
1.00). This last group, including pine martens from all the current European distribution of the species, is
separated into two different phylogroups: the Mediterranean (MED) and Central-Northern European
(CNE) phylogroups (Fig 1).
Fig. 1 Geographic distribution of the pine marten (Martes martes) (n=287) and sable (Martes zibeliina) (n=10)
samples represented as dots and stars, respectively. The correspondence of each sample with the discovered
phylogroups is also shown. Mediterranean (MED), Central-Northern European (CNE), Fennoscandian-Russian 1
(FNR1) and Fennoscandian-Russian 2 (FNR2) are represented in white, light grey, black grey and black respectively.
The Mediterranean phylogroup (BS: ≈63%, BPP:0.86) is made up predominantly of animals from the
main three Mediterranean peninsulas (corresponding to 101 out of 123 samples in this group i.e. 82.11%;
Spain: Mm1 - Mm9, n = 59; Portugal: Mm9, n = 4; Italy-Sardinia: Mm9 - Mm19 n = 34; Croatia: Mm19
n = 4) and some individuals (n = 12) from central Europe (France n = 2; Austria n = 1; Luxembourg n = 1
- 52 -
Phylogeography of the European pine marten
Paper II
Germany n = 4; Netherlands n = 1; Latvia n = 1; Czech Republic n = 1 Southern Sweden n = 4) and
Ireland (n = 6) (see Fig 1 and Table1). The Central-Northern group (BS: ≈67%, BPP: 1.00) is widely
distributed throughout Europe with the exceptions of the Mediterranean region and the northern area of
Fennoscandia (see Figs 1 and 2 and Table 1). The Fennoscandian-Russian phylogroup includes pine
martens from northern Sweden (Mm67 and Mm69) Norway (Mm68), Finland (Mm58, Mm59, Mm69)
and Russia (Mm56, Mm57 Mm60-66, Mm69) which are grouped together with Martes zibellina
haplotypes (Mz1-Mz11). This phylogroup is subdivided into two mayor haplogroups: the first one
(FNR1) (BS: ≈80%, BPP: 1.00) is composed of Martes zibellina specimens from Russia (Alexandrovskoe;
South Transbaikalia; Kamtchatka) and Martes martes from Russia (Chelyabinck Province, Leningrad
province), North Sweden and Norway; the second haplogroup (FNR2) (BS: ≈98%, BPP: 1.00) is
composed of Martes zibellina specimens from Russia (Alexandrovskoe; North Transbaikalia) and Martes
martes from Finland and Russia (Penza Province, Kirov region, Tver province) (Figs 1 and 2 and Table 1).
- 53 -
CHAPTER 2
Phylogeography
Table. 1 Geographic distribution and frequency of the 69 mtDNA haplotypes found in the 287 M.martes (Mm) and
11 haplotypes found in 10 M.zibellina (Mz) samples and two EMBL/Genbank sequences. The correspondence of
haplotypes with the discovered phylogroups is also shown. Mediterranean, Central-Northern European,
Fennoscandian-Russian 1 and Fennoscandian-Russian 2 are shown in white, light grey, black grey and black
respectively. EMBL/GenBank database with Accession numnbers are indicated for each haplotype.
Southern Europe
GenBank Acc
HM025990
HM025991
HM025992
HM025993
HM025994
HM025995
HM025996
HM025997
HM025998
HM025999
HM026000
HM026001
HM026002
HM026003
HM026004
HM026005
HM026006
HM026007
HM026008
HM026009
HM026010
HM026011
HM026012
HM026013
HM026014
HM026015
HM026016
HM026017
HM026018
HM026019
HM026020
HM026021
HM026022
HM026023
HM026024
HM026025
HM026026
HM026027
HM026028
HM026029
HM026030
HM026031
HM026032
HM026033
HM026034
HM026035
HM026036
HM026037
HM026038
HM026039
HM026040
HM026041
HM026042
HM026043
HM026044
HM026045
HM026046
HM026047
HM026048
HM026049
HM026050
HM026051
HM026052
HM026053
HM026054
HM026055
HM026056
HM026057
HM026058
HM026059
HM026060
HM026061
HM026062
HM026063
HM026064
HM026065
HM026066
HM026067
HM026068
FJ429093
Haplotype
Mm1
Mm2
Mm3
Mm4
Mm5
Mm6
Mm7
Mm8
Mm9
Mm10
Mm11
Mm12
Mm13
Mm14
Mm15
Mm16
Mm17
Mm18
Mm19
Mm20
Mm21
Mm22
Mm23
Mm24
Mm25
Mm26
Mm27
Mm28
Mm29
Mm30
Mm31
Mm32
Mm33
Mm34
Mm35
Mm36
Mm37
Mm38
Mm39
Mm40
Mm41
Mm42
Mm43
Mm44
Mm45
Mm46
Mm47
Mm48
Mm49
Mm50
Mm51
Mm52
Mm53
Mm54
Mm55
Mm56
Mm57
Mm58
Mm59
Mm60
Mm61
Mm62
Mm63
Mm64
Mm65
Mm66
Mm67
Mm68
Mm69
Mz1
Mz2
Mz3
Mz4
Mz5
Mz6
Mz7
Mz8
Mz9
Mz10
Mz11
n
Central‐Northern Europe
SP_NE SP_NW POR IT_CS IT_IS
19
2
1
1
2
3
1
1
29
4
1
6
2
1
2
IT_N CRO IR
SC FR LUX AU NTH DE CR PL
5
4
5
Fennoscandia
LT EST HU
RO
SW
FIN
NW
Russia
RUS
2 1
2
2
4
4
1
6
1
4
1
1
1
1
1
1
1
12
1
1
4
3
6
3
2
5
5
1
2
2
1
1
1
1
3
3
7
2
1
8
1
5
1
1
2
1
1
11
2
2
2
1
8
13
2
7
1
1
2
1
2
4
1
1
23
36
4
12
14
SP_NE SP_NW POR IT_CS IT_IS
Southern Europe
12
4
6
IT_N CRO IR
12 7 6
3
10 21 1 16
SC FR LUX AU NTH DE CR PL
1 15 5
LT EST HU
Central‐Northern Europe
1
RO
2
2
23
SW
7
FIN
1
1
1
1
1
1
2
3
3
NW
Fennoscandia
1
1
1
1
1
1
1
1
1
1
1
2
57
RUS
Russia
299
TOTAL
Geographical groups are indicated: Southern Europe, Central-Northern Europe, Fennoscandia and Russia. Abbreviations of
populations: Sp_NE, Spain northeast; Sp_NW, Spain northwest; Por, Portugal; It_S, south Italy; ; It_IS, Italy, Island of Sardina; ;
It_N, North Italy; Cr, Croatia; IR, Ireland; SC, Scotland; FR, France; LU, Luxembourg; NTH, The Netherlands; DE, Germany;
PL, Poland; LT, Lithuania; EST, Estonia; HU, Hungary; CR, Czeck Republic; AU, Austria; RO, Romania; FIN, Finland; SW,
Sweden; NW, Norway; RUS; Russia.
- 54 -
Phylogeography of the European pine marten
Paper II
Fig. 2 Neighbour-joining tree of the 69 Martes martes and 11 Martes zibellina mtDNA haplotypes. Bootstrap values
(%) obtained by the NJ and MP as well as Bayesian posterior probabilities are shown. See Table 1 for the haplotype
designations and distribution.
- 55 -
CHAPTER 2
Phylogeography
From the geographic distribution of each phylogroup (Fig. 1) some suture zones are evidenced between
them. A suture zone between MED and CNE phylogroups located in the Pyrenees and in the Alps; and
two suture zones between CNE and FNR phylogroups located in central Sweden and in the Ural
Mountains, respectively.
The median-joining network gave complementary information and confirmed the existence of these four
marten phylogroups (Fig. 3). The distinction among phylogroups was strongly supported respectively by 6
and > 30 mutations which separated the Mediterranean and the Central-Northern European, and the
Fennoscandian-Russian phylogroups. In contrast, sequence divergence within each of theses three
haplogroups was low. Central-Northern phylogroup is organised around a dominant haplotype (Mm35).
This group shows star-like topology, suggesting exponential growth of populations from a small numbers
of individuals. Also the Mediterranean phylogroup, is organised around a dominant haplotype (Mm9)
with star-like topology. The Fennoscandian-Russian haplogroups is also subdivided into two different
subgrroups (FNR1&FNR2), separated by 15 mutations.
Fig. 3 Median-joining network of the pine marten (Martes martes) and sable (Martes zibellina) mtDNA haplotypes.
Numbers of mutations (greater than one) between haplotypes are indicated on branches. See Table 1 for the
haplotype designations.
- 56 -
Phylogeography of the European pine marten
Paper II
In spite of the presence of two well differentiated subgroups within Fennoscandia-Russia clade, we
consider it as one unique phylogroup for remaining analysis taking into account the low number of
samples and the presence of admixture of M. martes and M. zibellina haplotypes within both subgroups.
The mismatch distribution (Fig. 4) and Tajima’s D and Fu and Li’s F statistics (Table 2) also suggested
varied demographic histories for the marten phylogroups. The negative and statistically significant values
of Fu’s statistic (Table 2) and the bellshaped mismatch distributions are indicative of population
expansions in the past within the MED and the CNE phylogroups (Fig. 4). The FNR (FNR1 & FNR2)
phylogroup showed a multimodal mismatch distribution (Fig. 4) that could indicate the admixture of two
expanding populations as also suggested by the positive result of Tajima test. However, non significant
result of Fu’s and Tajima test could indicate also a long-term stability and the reduced sample numbers
could also be responsible for not detecting significance; thus, a larger number of samples is needed to
clarify the demographic history of this group.
Fig. 4 Mismatch distribution analysis for the three major phylogroups.
Mediterranean
Central-Northern European
Fennoscandian-Russian
- 57 -
CHAPTER 2
Phylogeography
Table 2 Genetic variability observed within the main genetic phylogroups and Tajima’s D and Fu and Li’s F
statistics test.
n
Total
MED
CNE
FNR
299
123
139
37
Number of Nucleotide diversity Haplotype diversity
haplotypes π (±SD percentage)
(h ±SD)
79
26
28
14
0.647±0.048
0.120±0.010
0.234±0.012
0.707±0.054
0.955±0.006
0.837±0.027
0.936±0.006
0.930±0.030
Tajima’sD
Fu and Li’s F
-2.1672 (P<0.05) -3.340 (P<0.01)
-1.68615 (P>0.05) -2.42733 (P<0.05)
0.03059 (P>0.05) -0.09011 (P>0.05)
Population structure and genetic diversities statistics
AMOVA results were also consistent with the regional subdivision of samples in three main groups as
suggested by the median-joining network and the phylogenetic trees. Most probable phylogeographic
structures were those with maximum and statistically significant percentages of variation explained by
differences among groups (Table 3). The AMOVA showed that the majority of the total mtDNA
variation (%73.11) was distributed among geographical groups whereas a low percentage (%12.83) was
observed among populations within the groups. Moreover, the φ statistic suggests a low level of gene flow
between populations (ϕCT = 0.859, P < 0.001).
Table 3 Analyses of molecular variance based on mtDNA data from the main geographical groups.
Source of variation
Variance
Percentage P value
components of variation
ϕ statistics
< 0.001 ϕCT = 0.859
Among groups
6.841
73.11
Among populations/ groups
1.201
12.83
< 0.001 ϕSC = 0.477
Within populations
1.315
14.06
< 0.001 ϕST = 0.731
Intra- and inter-group genetic distances were very low: between 0.001-0.008 and 0.006-0.021, respectively
(Table 4). The genetic distance between CNE and MED phylogroups was 0.006. The genetic distance
between the FNR and CNE and MED phylogroups were slightly higher: 0.021 and 0.020, respectively
(Table 4).
- 58 -
Phylogeography of the European pine marten
Paper II
The highest value of nucleotide diversity was found in the FNR phylogroup, and the lowest in the MED
phylogroup (Table 2). On the other hand, haplotype diversity was similar in CN and FNR phylogroups,
being higher than in the MED phylogroup (Table 2).
Table 4 Matrix of distances between phylogeographic groups inferred from the data (below diagonal) assuming a
HKY85 model with rate heterogeneity and invariable sites (α = 0.732, I= 0.8693). Values within regions and are
shown in bold.
MED
CNE
FNR
MED
0.001
0.060
0.020
CNE
FNR
0.001
0.021
0.008
MED: Mediterrenean phylogroup; CNE: Central-Northern European
phylogroup; FNR: Fennoscandian-Rusian phylogroup (FNR1&FNR2).
Divergence times
Assuming the fossil record of M. vetus as calibration for the divergence point between M. martes-M.
zibellina complex, different periods of diversification can be recognized for Martes populations, falling all
of them within the Pleistocene period: (i) 0.29 Mya (95% HPD: 0.12-0.48) for the TMRCA of the
Fenoscandia-Russian phylogroup which began to differentiate before the two European phylogroups, (ii)
0.15 Mya (95% HPD: 0.06-0.26) and 0.13 (95% HPD: 0.04-0.22) for the timing of FNR1 and FNR2
groups respectively, (iii) 0.16 Mya (95% HPD: 0.064-0.28) for the separation time between the two
major European phylogroups (i.e. MED and CNE), (iv) 0.092 Mya (95%HPD:0.03-0.15) and 0.081
Mya (95% HPD: 0.03-0.14) as the divergence times for the CNE and MED phylogropus respectively
DISCUSSION
The role of Pleistocene glaciations on phylogeographic patterns and species diversification
There are several previous studies focused on phylogeny, evolutionary history and taxonomy of Martes
genera species which confirmed that true martens (subgenus Martes) are a monophyletic group (Hosoda et
al. 2000; Stone & Cook 2002; Sato et al. 2003; Marmi et al. 2004; Sato et al. 2006; Koepfli et al. 2008).
Within this subgenus, Martes foina has generally been recognized as basal to a clade containing the
- 59 -
CHAPTER 2
Phylogeography
remaining present-day species (M. americana, M melampus, M. zibellina and M. martes) (Anderson 1970;
Carr & Hicks 1997; Hosoda et al. 2000; Sato et al. 2003; Sato et al. 2006; Koepfli et al. 2008). However,
previous studies were not able to confidently resolve the phylogenetic relationships of the remaining four
members of the subgenus Martes (Koepfli et al. 2008). Indeed, this four species have been described as a
superspecies complex of closely related, yet largely allopatrically distributed taxa (Hagmeier, 1961;
Anderson, 1970). In the most recent and complete study published up to now, Koepfli et al. (2008), using
22 gene segments (~12,000 base pairs), were also uneven to fully resolve the phylogenetic relationships
within this group and found different phylogenetic placements of M. americana and M. melampus relative
to the clade which comprises M. martes and M. zibellina as sister species. Koepfli et al. (2008) suggested a
relatively recent speciation event within this group and calculated that diversification of this forest
dwelling species, coincides with expansion of this type of habitat across the Holarctic region during the
Plio-Pleistocene. This finding is consistent with fossil records that indicate that extinct ancestral species to
true martens primarily evolved in forested habitats (Anderson 1970; Anderson 1994; Koepfli et al. 2008).
The early Miocene Martes laevidens has been proposed as the oldest known member of the subgenus
Martes (Anderson 1994). However, the basicranial anatomy indicates that this species is not congeneric
with current living martens (Sato et al. 2003). Thus, the earliest known true marten appears to be M.
wenzensis (Anderson, 1970, 1994) from the early Villafranchian (4Mya). M. wezensis may have been
ancestral to M. vetus (Anderson 1994). The fossil record of the extinct M. vetus, which has been considered
ancestral to both M. martes and M. zibellina (Anderson 1994), ranges from almost the beginning of the
Pleistocene to about 400 kya (Wolsan 1993). Thus, using the most recent fossil record of M. vetus as
calibration of the divergence point of the M. martes-M. zibellina complex, our divergence analysis
estimates that the separation time between the two major European phylogroups (i.e MED and CNE) of
pine marten took place during the middle-late Pleistocene c.a 160.000 BP [ 0.16Mya (95% HPD: 0.0640.28)], probably during the late Riss or early Würm glaciations. However, the two different lineages of the
FNR phylogroup, which comprises specimens of both morphospecies (Martes martes and Martes zibellina),
begin to differentiate before the European phylogroups, c.a 290.000 BP, but also associated to the glacial
periods of the middle-late Pleistocene [0.29 Mya (95% HPD: 0.12-0.48)]. All divergences estimates fall
within the Pleistocene, which is typical of mammalian intraspecific phylogroups (Avise 2000). The earliest
record of M .martes is from the Riss-Wurm interglacial (ca. 120kya) in central Europe (Anderson 1994).
However, the earliest known records of Martes zibellina is much more posterior, from late Pleistocene in
age (Anderson 1994). However, the absence of fossil record data from Asia, limits the information
regarding the latter species and probably more ancient records of sable could exist in this region. Overall,
- 60 -
Phylogeography of the European pine marten
Paper II
our results suggest that Pleistocene conditions played an important initiating phylogeographic
differentiation as well as sculpting pre-existing phylogeographic variety into today’s consider sister species’
M. martes and M. zibellina. A combination of evidences from the fossil record and the divergence times
ontained in the present study indicates that, the populations of the common ancestor to both species (i.e
M. vetus) became reproductively and genetically isolated and started to diverge into two different lineages
giving first to the origin of the FNR phylogroup, where all the specimens morphologically identified as M.
zibellina are included, intermixed with some phenotypically M. martes indivuals, and latter to the pine
marten European phylogroups. We therefore propose that isolation of the FNR phylogroup in a refugium
presumably located in Eastern Asia (see below), may have played an important role in the origin of the
current M. zibellina morphospecies which occurs from East of Ural across all the Siberian coniferous taiga
forests.
The mtDNA groups inferred in this study shows a strict phylogeographic pattern throughout the species
range. Altogether three major mtDNA phylogenetic lineages have been discovered with a high correlation
between the currently recognized biogeographic regions in Europe (EEA 2008) and the distribution of
each phylogroup. Although some overlap occurs between specimens of different groups, overall a spatial
segregation pattern is evidenced. The MED phylogroup is closely associated with the Mediterrenean
peninsulas but mainly linked to the Atlantic and Alpine areas where the temperate mixed forests are
predominat. The CNE phylogroup, which covers most of the pine marten distribution, is clearly
distributed across the Continental biogeographic region. Finally, the FNR phylogroup is strongly related
to the Boreal region. These results suggest that each of the inferred groups could be adapted to specific
environmental conditions; and could represent different ecotypes. Indeed, there is a clear pattern of
latitudinal variation in pine marten body size, decreasing from south to north (Reig 1992) but also in diet
composition, food niche breadth, and prey size for martens in Europe (Zalewski 2004). However, in all
the pine marten distribution range the bank vole is the main prey species suggesting a close relationship
between both forest species (see below). Probably geographic variation in habitat and diet plays an
important role in shaping pine marten evolutionary adaptations, life-history strategies and ecological roles.
Additionally, similar phylogeographical patterns have been described in other forest species (Deffontaine et
al. 2005) suggesting that similar processes of coevolution can occur in species related to this specific
habitat.
- 61 -
CHAPTER 2
Phylogeography
Gene flow across southern European glacial refugia during the LGM
The role of southern European peninsulas (Iberia, Italy and the Balkans) as glacial refuge for temperate
species has been widely established (Taberlet et al. 1998; Hewit 2000; Weiss & Ferrand 2006). Sommer &
Nadachowski (2006) provide information about pine marten fossil remains from different archaeological
sites from the Younger Palaeolithic, which are precisely dated to the Last Glacial Maximum (23,000–
16,000 BP ). They confirmed the existence of the pine marten during the LGM in the well-known
refugium of the Mediterranean peninsulas. In spite of they found records only in the Italian and the
Balkan Peninsulas, the few records of pine marten reported, probably reflect only single records of its
former distributions in potential refuge areas, rather than a complete range of distribution (Sommer &
Nadachowski 2006). Indeed, in a previous work focused on the late and post-glacial colonization of
mustelids, Sommer & Bennecke (2004) suggested that pine marten could also be present into the Iberian
Peninsula during the LGM taking into account that subfossil records from the Late Glacial until the
Middle Holocene were continuously present in the Iberian Peninsula. According to fossil data (Sommer &
Bennecke 2004; Sommer & Nadachowski 2006), and the fact that our phylogeographic analyses reveal a
mtDNA phylogroup joining all the pine martens populations from these three regions (Figs 1 and 2),
there is strong evidence for the widely recognized refuge of the Mediterranean peninsulas for the pine
marten. Moreover, these results suggest a large Mediterranean population during the late Pleistocene,
where gene flow between populations was possible. This pattern of continuous geneflow across southern
Europe has been also reported in Brown bear populations (Valdiosera et al. 2007).
For many temperate species southern refugial areas currently exhibit high genetic diversity (Hewit 2004).
However, in this study the Mediterranean phylogoup shows the lowest genetic variability of the three
identified phylogroups. The low nucleotide and haplotype diversities characterizing this phylogroup could
be associated with population fragmentation followed by severe population bottlenecks during the
Quaternary glaciations.
Central-Northern phylogroup: New insights into the cryptic northern glacial refugia
The low proportion of pine martens from the Mediterranean lineage (18 out of 123 i.e 14.6%) identified
in central-northern Europe strongly suggests that this lineage has not been the source of major postglacial
recolonizations of this region. Moreover, no haplotypes from CNE phylogroup was found in any of the
know southern refuges of Europe (except 2 samples near the contact zone in the Italian Alps, that may
come from a recent recolonization from CNE lineage populations in the Alps; Balestrieri et al. 2010),
suggesting that central-northern Europe was rather recolonized by a pine marten phylogroup surviving the
- 62 -
Phylogeography of the European pine marten
Paper II
last glaciations in a central European glacial refugia as it has been previously proposed by paleontological
data (Sommer & Benecke, 2004; Sommer & Nadachowski 2006). Indeed, the CNE phylogroup presently
covers most of the pine marten distribution range in the Paleartic region (Fig. 1) and was subjected to a
recent population expansion (Figs 3 and 4). Additionally, this phylogroup is characterized by a high
haplotype and nucleotide diversities in comparison to the MED phylogroup, suggesting that this
populations have been affected by less severe population bottlenecks. Moreover, fossil records of pine
marten were found during the LGM in the East of the Carpathians, in Moldova (Markova et al. 1995;
Sommer & Benecke, 2004) and in Deszczowa and Mamutova Cavesin in southern Poland (Sommer &
Nadachowski 2006). Consequently, we can reasonably assume that this region acted as glacial refugium
for the CNE pine marten phylogroup.
The existence of a non Medtirrenean refugium would also be consistent with past and present ecological
traits of pine martens. The current range distribution of the pine marten includes coniferous forests and
cold environments. As a cold tolerant species, the pine martens would have been able to survive at
northern latitudes even at former boundary between woodland and tundra. Indeed, the northernmost
subfossil record from the Late-Glacial was situated in Denmark assigned to the younger Dryas (14 kya).
Moreover, Bhagwat & Wills (2008) suggest that the persistence of species in northerly glacial refugia is
closely related to some biological and biogeographical traits, such as small body size, a present-day
northerly distribution and cold-tolerance, which in fact strongly matches with those traits found on the
pine marten.
There is mounting evidence that some European temperate species did not respond to the last glaciation
by simply shifting their distributions to the Mediterranean region but also survived at higher latitudes
previously considered inhospitable (Willis & van Andel 2004; Bhagwat & Willis 2008; Provan & Bennett
2008; Stewart et al. 2010). Evidences in support to this model came from phylogeographic studies (Jaarola
& Searle 2002; Deffontaine et al. 2005; Kotlik et al. 2006, Saarma et al. 2007; Teacher et al. 2009)
mammal fossil records (Sommer & Benecke 2004; Sommer & Nadachowski 2006), fossil pollen data and
macrofossil remains of tree species (Willis & van Andel 2004) as well as from species distribution
modeling (Svenning et al. 2008; Fløjgaard et al. 2009).
The present genetic study, as well as fossil remains (Sommer & Benecke, 2004; Sommer & Nadachowski
2006), indicates that a temperate forest species such as the pine marten was also able to survive in central
European regions during glacial periods, providing new insights into the existence of cryptic northern
glacial refugia.
- 63 -
CHAPTER 2
Phylogeography
The Fennoscandian-Russian group
In our analysis, we identified the existence of a third continental phylogroup that joined pine martens
form Norway, Northern Sweden, Finland and several regions of Russia together with Martes zibellina
specimens collected from a wide geographic area East of Urals (from Urals to Katnchatka). Moreover this
phylogroup is subdivided into two diferent subgroups sugessting two different mtDNA lineages present in
marten populations of Fennoscandia-Rusia (Fig 2). This group is characterized by very different
demographic history, genetic diversity and genetic divergence as compared to the other European
phylogroups (Figs 2, 3; Tables 2 and 5). However, we have not found correspondence between these two
subgroups, neither with their geographic distribution, nor with the two morphospecies considered. Thus,
in this phylogroup (i.e FNR) the morphological species concept does not correspond to the phylogenetic
species concept (De Queiroz et al. 2007). Our results suggest that, Pleistocene conditions played an
important initiating phylogeographic differentiation as well as sculpting this pre-existing phylogeographic
variety into today’s sister species’ pine marten and sable. It seems likely that this ancient phylogeographic
varieties (i.e FNR1&FNR2) could be therefore evolved given to the origin of the sable east of Urals with a
secondary contact between pine marten and sable giving rise to a process of genetic hybridization of both
morphospecies in Fennoscandia and Russia region. Previous studies have found that M. martes and M.
zibellina formed a monophyletic group with M. martes paraphyletic with respect to M. zibellina (Stone &
Cook 2002, Marmi et al. 2004). Although incomplete lineage sorting can result in paraphyletic
relationships (Davison et al. 1999), these two species are not reproductively isolated and successful
hybridization between them is possible (Grakov 1994; Rozhnov et al. 2010). Indeed, Davison et al. (2001)
detected some individuals on Fennoscandia with mtDNA of M. zibellina and explained the existence of
those specimens as the past introgression of the sable mtDNA into pine martens. More recently, Rozhnov
et al. (2010) found a high degree of reciprocal mtDNA introgression in sympatric populations of these
species in the sympatric zone in the northern Urals. Hybridization or mtDNA introgression between other
species of the subgenus Martes has been previously documented (M. martes x M. americana, Kyle et al.
2003; M. zibellina x M. melampus, Murakami et al. 2004) confirming that interspecific hybridization is a
common event within this subgenus. The four species of the subgenus Martes have been considered as a
superspecies by some authors, suggesting close phylogenetic relationships (Hagmeier1961; Anderson,
1970) and defined as a monophyletic group of allopatric species that are morphologically too different to
be included in a single species. Although some species begin to separate during the Pleistocene, in most
cases it took more than two million years to reach full speciation (Avise et al. 1999). This consideration is
concordant with the results obtained for the FNR phylogroup. Probably, the origin of Post-Glacial
- 64 -
Phylogeography of the European pine marten
Paper II
recolonization during the Holocene lies in one or more Asian glacial refugia (Sommer & Bennecke 2004)
that could be located in southern or eastern Urals or even in the Caucasus. However, this phylogroup
should be further investigated by the use of nuclear markers and a more extensive sampling from these
regions. Moreover, a comparison with the phylogeographic patterns of the sable (currently underway) will
probably provide valuable information concerning the evolutionary history of this different marten
phylogroup and whether refugial isolation could have lead to speciation between these species.
Postglacial re-colonization of pine marten populations
The impact of Quaternary glaciations and the identification of the main post-glacial colonisation routes
from glacial refugia have been widely studied in different mammal species across Eurasia (Taberlet et al.
1998; Hewitt 2004: Stewart et al. 2010).
In this study, the evidence of population structuring, into three different phylogroups, found within
European pine marten populations is the clear sign of a postglacial recolonization from different refuge
areas with posterior intermixing. The high levels of haplotype diversity and low levels of nucleotide
diversity found (Table 2) may suggest rapid demographic expansions from small effective population sizes,
multiple refuges and secondary contact of populations from different refuges. The geographic distribution
of some shared haplotypes and phylogeographic groups (Table 2, Fig. 1) agrees with this last possibility.
The study by Davison et al. (2001), found a similar population structure in Europe with the existence of
the same three groups inferred in this study. However the low support values obtained for each of the
detected groups, due to the small mtDNA fragment used (325bp) and the limited sampling from
Mediterranean and from the eastern Russian populations, did not allow providing clear clues for the postglacial recolonization of central Europe.
Our data suggest that during the LGM, the two lineages retreated in separate refugia (southern European
peninsulas for the MED lineage and the Carpathians or more eastern refugium for the CNE lineage).
Populations of MED lineage probably went through a bottleneck during the last glaciation, and after the
LGM, the population expansion from Mediterranean peninsulas was likely associated with a haplotype
diversification, as suggested by the star-like phylogeny (Fig. 3) and the low nucleotide and high haplotype
diversities (Table 2). The MED lineage expanded to the north up to southern Sweden. However, the low
proportion of pine martens from the MED lineage discovered in Central-Northern Europe strongly
suggests that this lineage has not been the source of major postglacial recolonizations of this region.
- 65 -
CHAPTER 2
Phylogeography
The persistence of CNE pine marten populations in Central European refugia must have significantly
reduced the times by which recolonizing animals reached the northern parts of Europe after the LGM
(Sommer & Nadachoswki, 2006). Indeed, while the MED phylogroup is more restricted to the southern
European areas, the CNE phylogroup arrives up to northen Urals. The rapid recolonization of Central
Europe by CNE populations which survived in central European refugia, could be the reason for the
presence of pine marten in regions like Denmark and Czech Republic during Younger Dryas and the
Magdalennian: (c.a 17.00-9.000 BP), respectively. Probably, the CNE phylogroup was the first to colonise
central Europe limiting the expansion processes of the MED phylogroup. Consequently, the low presence
of MED lineage on central Europe could be interpreted as a minor recolonization process from the
Mediterrenean penisulas to this area.
According to our results, it is noticeable that the current island populations of Britain and Ireland are
represented by the two main European phylogroups: CNE and MED, respectively. However, the recent
discovery of an haplotype representative of the Iberian peninsula martens in museum specimens from
Wales, indicates that both phylogroups were present in Britain (C. O’Reilly, unpublished data; Mullins et
al. 2009).
There are two main hypotheses explaining the post- glacial re-establishment of the Brithish islands martens
according to this data, which are not mutually exclusive: a natural post glacial re-colonization of the MED
group from the Iberian Peninsula, tracking the coast line for both Ireland and Britain; and a natural
colonization of CNE phylogroup from continental Europe to Britain. However, we cannot discard
anthropic origin of these populations, which is also congruent with the early trade routes that have been
established between south-west Europe and Ireland since Mesolithic, as has been previously proposed
(Searle 2008).
The presence of the three major phylogroups in Fennoscandia suggests that the pine marten recolonized
this area from the northeast, by the FNR phylogroup, and from the south, by the MED and CNE
phylogroups. Similar north–south phylogeographical patterns with a suture zone in central Fennoscandia
have been described for several other mammals strongly corroborating this recolonization model (e.g.
Taberlet et al. 1995; Jaarola et al. 1999, Brunhoff et al. 2003).
At present, it is impossible to know exactly neither how hybridization occurred between martens in
Fennoscandia-Rusia, nor whether is the exact location of the FNR populations refugia. However, as M.
zibellina is an eastern species that is better cold adapted than M. martes, it may have been the first to
colonize the northeast of Fennoscandia from and undetermined Eastern refugia in Asia (Sommer &
- 66 -
Phylogeography of the European pine marten
Paper II
Bennecke 2004). As the climate ameliorated, M. martes could have replaced M. zibellina, with
mitochondrial introgression as the dwindling M. zibellina population mated with M. martes (Grakov
1994; Rozhnov et al. 2010). An alternative scenario is suggested by Davison et al. (2001). M. zibellina is
generally limited by the Ural mountains to the west, but during the “little ice age” (c.a.1550-1850 BP)
there is evidence that the sable penetrated deep into Europe providing a suitable scenario for mating with
the pine marten. As the temperature increased again, the range of M. zibellina was restricted once again to
the Urals. Actually, while the FNR phylogroups arrives up to central Sweden, the CNE phylogroups is
restricted Eastward up to the Ural Mountains.
Prey-predator relationships: a linked phylogeographic pattern between the bank vole and the pine marten
Coevolved relationships may lead to high congruence of distributional history that could be strong
between predators and their potential prey species (Abrams 2000). The bank vole is one of the main prey
species of the pine marten across its entire distribution area (Zalewski et al. 2004). Consequently, a linked
phylogeogrphic pattern could be expected for prey-predators which inhabit the same forest habitats.
Indeed, recent phylogeophic studies conducted in the bank vole (Myodes glareolus) have found a close
pattern to that found in the pine marten (Deffontaine et al. 2005; Kotlik et al. 2006). Interestingly, in
southern Europe the bank vole has three different Mediterranean lineages belonging to each of the
Mediterranean peninsulas (Deffontaine et al. 2005), while we found a unique Mediterranean lineage for
the pine marten that suggest continuous gene flow across southern Europ for the latter. These traits are
congruent with the restricted dispersal capabilities of the bank vole in comparison to a highly mobile midsized carnivore such as the pine marten. Moreover, Kotlik et al. (2006) provided the clearest
phylogeographic evidence of a northern glacial refugium for temperate species in central Europe. Thus, the
same location of a glacial refugium for the main predator of the bank vole gives strong support for central
European refugia in temperate forest mammals. Regarding the Fennoscandia and Russia region (west of
Urals) an Ural phylogroup was identified also for the bank voles which were closely related to the redbacked vole Myodes rutilus, the species found east of Urals. These data is congruent with the FNR
phylogroup where pine marten and sable haplotypes are admixed together. Thus, similar patterns between
closely related species (M. martes and M. zibellina; M. glareolus and M. rutylus) in Fennoscandia-Russia
region could therefore highlight the importance of the Ural Mountains as the main barrier for gene flow
between lineages, given rise to species diversification processes. Indeed, the Urals has been evidenced as a
important suture zone (Hewitt 1996), and probably had an important role on species diversification in the
- 67 -
CHAPTER 2
Phylogeography
Paleartic region taking into account that this mountain barrier suppose the distribution limit between a
number of related taxa.
CONCLUSIONS
The role of glacial refugia in intraspecific evolution has been widely addressed for many different taxa
through phylogeographic analysis (Avise 2009). However, the respond of each species due to climatic
changes of the Quaternary depends largely on their adaptations and environmental tolerances (Stewart et
al. 2010). The mtDNA groups inferred in this study shows a strict phylogeographic pattern throughout
the species range with the presence of three major phylogroups MED, CNE and FNR each of them
related to specific biogeographic regions: Mediterranean-Atlantic, Continental and Boreal, respectively,
which could represent different ecotypes. Overall, our study indicates a complex phylogeographic history
for M. martes indicating a mixed pattern of recolonization of northern Europe from both Mediterranean
and non-Mediterranean refugia. The presence of CNE lineage, widespread across northern Europe, which
did not correspond to the lineages present in any of the three peninsular refugia, suggests that the source of
this lineages lie elsewhere, possibly in a northern cryptic refugia located in the Carpathians, as it has been
previously proposed by paleontological data. These results provide new insights into the evidence that in
addition to traditional southern refugia for temperate species, cryptic refugia existed in Northern Europe
during glacials. However, this does not exclude the importance of Mediterranean peninsulas as a relevant
source of diversity for pine martens. Moreover, a highly divergent phylogroup has been discovered in
Fennoscandia-Rusia, wich comprises specimens of both M. martes and M. zibellina morphospecies, which
appears characterized by very different demographic history as compared to the other European
phylogroups. Our incipient results about this phylogroup suggest that Pleistocene conditions played an
important initiating phylogeographic differentiation as well as sculpting this pre-existing phylogeographic
variety into today’s sister species’ pine marten and sable. However, attention should be paid in the future
to this region and a deep and needed study will probably give useful information concerning Martes
complex evolution in this area.
Finally, the linked phylogeographic patterns found between the pine marten and the bank vole, provided
clear evidence about the Quaternary effects on the evolution of forest dwelling species and suggest that
coevolved prey-predator relationships lead to a stronger congruence of their phylogeographic history.
- 68 -
Phylogeography of the European pine marten
Paper II
ACKNOWLEDGEMENTS
The authors wish to thank to all the researchers and institutions listed in Table S1 for providing samples used in this
study. This research has been partially funded by the Basque Government through the Research group on
“Systematics, Biogeography and Population Dynamics” (Ref. IT317-10; GIC10/76) and by the Conservation
Genetics Laboratory (ISPRA). Ruiz-González holds a Ph.D. fellowship awarded by the Dept. of Education
Universities and Research of the Basque Government (Ref. BFI09.396).
REFERENCES
Abrams PA (2000) The evolution of predator-prey interactions: Theory and evidence. Annual Review of Ecology and
Systematics, 31, 79-105.
Anderson E. (1970) Quaternary evolution of the genus Martes (Carnivora, Mustelidae). Acta Zooogica Fennica, 130,
1–132.
Anderson E. (1994) Evolution, prehistoric distribution, and systematics of Martes. In: Buskirk SW, Harestad AS,
Raphael MG, Powell RA (Eds.), Martens, Sables, and Fishers: Biology and Conservation. Cornell University
Press, Ithaca, NY, pp. 13–25.
Avise JC (2000) Phylogeography: The History and Formation of Species. Harvard University Press, Cambridge, MA
Avise JC (2009) Phylogeography: retrospect and prospect. Journal of Biogeography, 36, 3-15.
Balestrieri A, Remonti L, Ruiz-Gonzalez A, Gomez-Moliner BJ, Vergara M, Prigioni C (2010) Range expansion of
the pine marten (Martes martes) in an agricultural landscape matrix (NW Italy). Mammalian Biology, 75,
412-419.
Bandelt HJ, Forster P, Rohl A (1999) Median-joining networks for inferring intraspecific phylogenies. Molecular
Biology and Evolution, 16, 37-48.
Bhagwat SA, Willis KJ (2008) Species persistence in northerly glacial refugia of Europe: a matter of chance or
biogeographical traits? Journal of Biogeography, 35, 464-482.
Brunhoff C, Galbreath KE, Fedorov VB, Cook JA, Jaarola M (2003) Holarctic phylogeography of the root vole
(Microtus oeconomus): implications for late Quaternary biogeography of high latitudes. Molecular Ecology,
12, 957-968. Cox, C. B. & Moore, P. D. 2005: Biogeography: an ecological and evolutionary approach.
Blackwell Publishing, Oxford.
Carr SM, Hicks SA (1997) Are there two species of marten in North America? Genetic and evolutionary
relationships with in Martes. In: Proulx G, Bryant HN, Woodard PM. Martes: Taxonomy, Ecology,
Techniques, and Management (eds) pp. 15-28. Provincial Museum of Alberta, Edmonton, Alberta,
Canada.
Davison A, Birks JDS, Brookes RC, Messenger JE, Griffiths HI (2001) Mitochondrial phylogeography and
population history of pine martens Martes martes compared with polecats Mustela putorius. Molecular
Ecology, 10, 2479-2488.
De Queiroz K (2007) Species concepts and species delimitation. Systematic Biology, 56, 879-886.
Deffontaine V, Libois R, Kotlik P, Sommer R, Nieberding C, Paradis E, Searle JB, Michaux J (2005) Beyond the
Mediterranean peninsulas: evidence of central European glacial refugia for a temperate forest mammal
species, the bank vole (Clethrionomys glareolus). Molecular Ecology, 14, 1727-1739.
Drummond AJ, Rambaut A (2007) BEAST: Bayesian evolutionary analysis by sampling trees. Bmc Evolutionary
Biology, 7.
Excoffier L, Laval G, Schneider S (2005) Arlequin (version 3.0): An integrated software package for population
genetics data analysis. Evolutionary Bioinformatics, 1, 47-50.
Felsenstein J (1985) Confidence-limits on phylogenies with a molecular clock. Systematic Zoology, 34, 152-161.
Fitch WM (1971) Toward defining course of evolution - minimum change for a specific tree topology. Systematic
Zoology, 20, 406-&.
Flojgaard C, Normand S, Skov F, Svenning JC (2009) Ice age distributions of European small mammals: insights
from species distribution modelling. Journal of Biogeography, 36, 1152-1163.
Hagmeier, E.M., 1961. Variation and relationships in North American marten. Can. Field-Nat. 75, 122–137
- 69 -
CHAPTER 2
Phylogeography
Hall TA (1999) BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows
95/98/NT. Nucleic Acids Symp. Ser. 41, 95–98
Hasegawa M, Kishino H, Yano TA (1985) Dating of the human ape splitting by a molecular clock of
mitochondrial-dna. Journal of Molecular Evolution, 22, 160-174.
Hewitt G (2000) The genetic legacy of the Quaternary ice ages. Nature, 405, 907-913.
Hewitt GM (2001) Speciation, hybrid zones and phylogeography - or seeing genes in space and time. Molecular
Ecology, 10, 537-549.
Hewitt GM (2004) Genetic consequences of climatic oscillations in the Quaternary. Philosophical Transactions of the
Royal Society of London Series B-Biological Sciences, 359, 183-195.
Ho SYW (2007) Calibrating molecular estimates of substitution rates and divergence times in birds. Journal of Avian
Biology, 38, 409-414.
Hosoda T, Suzuki H, Harada M, Tsuchiya K, Han SH, Zhang YP, Kryukov AP, Lin LK (2000) Evolutionary
trends of the mitochondrial lineage differentiation in species of genera Martes and Mustela. Genes &
Genetic Systems, 75, 259-267.
Huelsenbeck JP, Ronquist F (2001) MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics, 17, 754755.
Jaarola M, Searle JB (2002) Phylogeography of field voles (Microtus agrestis) in Eurasia inferred from mitochondrial
DNA sequences. Molecular Ecology, 11, 2613-2621.
Jaarola M, Tegelstrom H, Fredga K (1999) Colonization history in Fennoscandian rodents. Biological Journal of the
Linnean Society, 68, 113-127.
Koepfli KP, Deere KA, Slater GJ, Begg C, Begg K, Grassman L, Lucherini M, Veron G, Wayne RK (2008)
Multigene phylogeny of the Mustelidae: Resolving relationships, tempo and biogeographic history of a
mammalian adaptive radiation. Bmc Biology, 6.
Kotlik P, Deffontaine V, Mascheretti S, Zima J, Michaux JR, Searle JB (2006) A northern glacial refugium for bank
voles (Clethrionomys glareolus). Proceedings of the National Academy of Sciences of the United States of
America, 103, 14860-14864.
Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, McWilliam H, Valentin F, Wallace IM, Wilm
A, Lopez R, Thompson JD, Gibson TJ, Higgins DG (2007) Clustal W and clustal X version 2.0.
Bioinformatics, 23, 2947-2948.
Librado P, Rozas J (2009) DnaSP v5: a software for comprehensive analysis of DNA polymorphism data.
Bioinformatics, 25, 1451-1452.
Markova AK, Smirnov NG, Kozharinov AV, Kazantseva NE, Simakova, N.E.K. & Kitaev, L.M. (1995) Late
pleistocene distribution and diversity of mammals in Northern Eurasia. Paleontologia i Evolucio, 29, 5–
143.
Marmi J, Lopez-Giraldez JF, Domingo-Roura X (2004) Phylogeny, evolutionary history and taxonomy of the
Mustelidae based on sequences of the cytochrome b gene and a complex repetitive flanking region.
Zoologica Scripta, 33, 481-499.
Mullins J, Statham MJ, Roche T, Turner PD, O'Reilly C (2010) Remotely plucked hair genotyping: a reliable and
non-invasive method for censusing pine marten (Martes martes, L. 1758) populations. European Journal of
Wildlife Research, 56, 443-453.
Pertoldi C, Breyne P, Cabria MT, Halfmaerten D, Jansman HAH, Van Den Berge K, Madsen AB, Loeschcke V
(2006) Genetic structure of the European polecat (Mustela putorius) and its implication for conservation
strategies. Journal of Zoology, 270, 102-115.
Posada D, Crandall KA (1998) MODELTEST: testing the model of DNA substitution. Bioinformatics, 14, 817818.
Proulx G Aubry KB, Birks J, Buskirk SW, Fortin C, Frost HC, Krohn WB, Mayo L, Monakhov V, Payer D, Saeki
M, Santos-Reis M, Weir R, Zielinski WJ (2004) World distribution and status of the genus Martes in
2000. In: Harrison DJ, Fuller AK, Proulx G (Eds) Martens and fishers (Martes) in human- altered
environments: an international perspective. New York: Springer-Verlag, pp 77–98
Provan J, Bennett KD (2008) Phylogeographic insights into cryptic glacial refugia. Trends in Ecology & Evolution,
23, 564-571.
Rambaut A, Drummond AJ (2007) Tracer v1.4, Available from http://beast.bio.ed.ac.uk/Tracer
- 70 -
Phylogeography of the European pine marten
Paper II
Reig S (1992) Geographic-variation in pine marten (Martes-martes) and beech marten (m-foina) in europe. Journal
of Mammalogy, 73, 744-769.
Rogers AR, Harpending H (1992) Population-growth makes waves in the distribution of pairwise geneticdifferences. Molecular Biology and Evolution, 9, 552-569.
Rozhnov VV, Meschersky IG, Pishchulina SL, Simakin LV (2010) Genetic analysis of sable (Martes zibellina) and
pine marten (M. martes) populations in sympatric part of distribution area in the northern Urals. Russian
Journal of Genetics, 46, 488-492.
Saarma U, Ho SYW, Pybus OG, Kaljuste M, Tumanov IL, Kojola I, Vorobiev AA, Markov NI, Saveljev AP,
Valdmann H, Lyapunova EA, Abramov AV, Mannil P, Korsten M, Vulla E, Pazetnov SV, Pazetnov VS,
Putchkovskiy SV, Rokov AM (2007) Mitogenetic structure of brown bears (Ursus arctos L.) in
northeastern Europe and a new time frame for the formation of European brown bear lineages. Molecular
Ecology, 16, 401-413.
Saitou N, Nei M (1987) The neighbor-joining method - a new method for reconstructing phylogenetic trees.
Molecular Biology and Evolution, 4, 406-425.
Sato JJ, Hosoda T, Wolsan M, Tsuchiya K, Yamamo M, Suzuki H (2003) Phylogenetic relationships and
divergence times among mustelids (Mammalia : Carnivora) based on nucleotide sequences of the nuclear
interphotoreceptor retinoid binding protein and mitochondrial cytochrome b genes. Zoological Science, 20,
243-264.
Sato JJ, Wolsan M, Suzuki H, Hosoda T, Yamaguchi Y, Hiyama K, Kobayashi M, Minami S (2006) Evidence from
nuclear DNA sequences sheds light on the phylogenetic relationships of pinnipedia: Single origin with
affinity to musteloidea. Zoological Science, 23, 125-146.
Searle JB (2008) The colonization of Ireland by mammals. In: . Davenport JL, Sleeman DP, Woodman PC (Eds.)
Mind the Gap: Postglacial colonization of Irelandd, Ir Nat J (Special supplement), pp. 109-116.
Slatkin M, Hudson RR (1991) Pairwise comparisons of mitochondrial-dna sequences in stable and exponentially
growing populations. Genetics, 129, 555-562.
Sommer R, Benecke N (2004) Late- and Post-Glacial history of the Mustelidae in Europe. Mammal Review, 34,
249-284.
Sommer RS, Nadachowski A (2006) Glacial refugia of mammals in Europe: evidence from fossil records. Mammal
Review, 36, 251-265.
Stewart JR, Lister AM (2001) Cryptic northern refugia and the origins of the modern biota. Trends in Ecology &
Evolution, 16, 608-613.
Stewart JR, Lister AM, Barnes I, Dalen L (2010) Refugia revisited: individualistic responses of species in space and
time. Proceedings of the Royal Society B-Biological Sciences, 277, 661-671.
Stone KD, Cook JA (2002) Molecular evolution of Holarctic martens (genus Martes, Mammalia : Carnivora :
Mustelidae). Molecular Phylogenetics and Evolution, 24, 169-179.
Svenning JC, Normand S, Kageyama M (2008) Glacial refugia of temperate trees in Europe: insights from species
distribution modelling. Journal of Ecology, 96, 1117-1127.
Swofford, D.L., 2002. PAUP*: Phylogenetic analysis using parsimony (* and other methods), version 4.0b10.
Sinauer, Sunderland, Massachusetts
Taberlet P, Fumagalli L, Wust-Saucy AG, Cosson JF (1998) Comparative phylogeography and postglacial
colonization routes in Europe. Molecular Ecology, 7, 453-464.
Tajima F (1989) Statistical-method for testing the neutral mutation hypothesis by dna polymorphism. Genetics,
123, 585-595.
Valdiosera CE, Garcia N, Anderung C, Dalen L, Cregut-Bonnoure E, Kahlke RD, Stiller M, Brandstrom M,
Thomas MG, Arsuaga JL, Gotherstrom A, Barnes I (2007) Staying out in the cold: glacial refugia and
mitochondrial DNA phylogeography in ancient European brown bears. Molecular Ecology, 16, 5140-5148.
Webb T, Bartlein PJ (1992) Global changes during the last 3 million years - climatic controls and biotic responses.
Annual Review of Ecology and Systematics, 23, 141-173.
Weiss, S. & Ferrand, N. (eds) (2006) The phylogeography of southern European refugia. Springer, Dordrecht.
Willis KJ, van Andel TH (2004) Trees or no trees? The environments of central and eastern Europe during the Last
Glaciation. Quaternary Science Reviews, 23, 2369-2387.
- 71 -
CHAPTER 2
Phylogeography
Wojcik JM, Kawalko A, Markova S, Searle JB, Kotlik P (2010) Phylogeographic signatures of northward post-glacial
colonization from high-latitude refugia: a case study of bank voles using museum specimens. Journal of
Zoology, 281, 249-262.
Wolsan M (1993) Phylogeny and classification of early european mustelida (mammalia, carnivora). Acta
Theriologica, 38, 345-384.
Zalewski A (2004) Geographical and seasonal variation in food habits and prey size of European pine martens. In:
Harrison DJ, Fuller AK, Proulx G (Eds) Martens and fishers (Martes) in human- altered environments: an
international perspective. New York: Springer-Verlag, pp 21–76
Zalewski A, Jedrzejewski W (2006) Spatial organisation and dynamics of the pine marten Martes martes population
in Bialowieza Forest (E Poland) compared with other European woodlands. Ecography, 29, 31-43.
Zoltán V (2010) Extra-Mediterranean Refugia, Post-Glacial Vegetation History and Area Dynamics in Eastern
Central Europe. In: Relict Species: Phylogeography and Conservation Biology J.C. Habel and T. Assmann
(eds.), Springer-Verlag Berlin Heidelberg, pp: 57- 87
- 72 -
Phylogeography of the European pine marten
- 73 -
Paper II
Phylogeography
Eueopean pine marten
(Martes martes)
Species
Spain
Country
SP_NE
SP_NE
SP_NE
14T
15T
8BU
4
5
6
431/06
10-TF
016/07
67KA
14-EA
160/07
253/09
1-ESK
13T
071/07
128-OR
12
13
14
15
16
17
18
19
21
22
20
037/06
MA1103
11
199/05
9
10
SL
8
SP_NE
SP_NE
SP_NE
SP_NE
SP_NE
SP_NE
SP_NE
SP_NE
SP_NE
SP_NE
SP_NE
SP_NE
SP_NE
SP_NE
SP_NE
SP_NE
SP_NE
11T
3
7
SP_NE
9T
14AP
SP_NE
8T
Huesca;Torla; Ordesa
- 74 -
Alava; Arzubiaga; N-1 PK357
Navarra;Egués NA-150 Km 3’3
Alava; Escota
Alava; Amezaga; N-622 pk23;5
Alava; Aguiñiga
Alava; Amurrio
Bizkaia; Mañaria
Alava; Llodio
Alava; Zuya; Altube
Álava; Legutiano
Álava
Álava;Ozaeta
Álava; Ozaeta
Guipuzkoa;Salinas de Leniz
Burgos;Aostri de Losa-Tunel Peña Angulo
Lleida;La Guingueta d Áneu;Cruce de
estaron
Navarra; Esparza NA-178 Km 40
Navarra; Enderitz; N-121 Km 12’2
Navarra;Cruce Inbuluzqueta N-135 Km 16’8
Navarra; Enderitz; N-121 Km 12’5
Navarra; Etulain-Anue N-121 Km 18
Country_Code Sample Location
2
Sample code
1
ID
42,64928800
42,87793161
42,82427081
42,83752108
42,96125248
43,12529689
43,00646502
43,12155152
43,12529689
42,97636399
42,98052195
42,90974855
42,91867282
42,91867282
42,98705089
42,53022900
42,93879694
42,85387300
42,89318802
42,92489700
42,89066785
42,93763387
UTMX
-0,05779499
-2,61897701
-1,55257416
-2,94442004
-2,84315285
-2,98821596
-2,94023940
-2,65138170
-2,98821596
-2,86927798
-2,64509459
-2,89371216
-2,50151183
-2,50151183
-2,57253164
1,19150193
-3,10801844
-1,10017200
-1,60622056
-1,54141837
-1,60616621
-1,60867352
UTMY
Fernando Carmena. Parque Nacional de Ordesa y
Fermin Urra.VRFN. Gobierno de Navarra
Patricia Lizarraga/ Laura Elorza. Centro de Recuperación de
Fauna de Mártioda. DFA
Iker Ayala
Enrique Arberas Mendibil
Patricia Lizarraga/ Laura Elorza. Centro de Recuperación de
Fauna de Mártioda. DFA
Patricia Lizarraga/ Laura Elorza. Centro de Recuperación de
Fauna de Mártioda. DFA
Julio Ruiz Guijarro
Txema Fernández. IAN
Patricia Lizarraga/ Laura Elorza. Centro de Recuperación de
Fauna de Mártioda. DFA
Museo de Ciencias Naturales de Álava
Patricia Lizarraga/ Laura Elorza. Centro de Recuperación de
Fauna de Mártioda. DFA
Pere Aymerich
Patricia Lizarraga/ Laura Elorza. Centro de Recuperación de
Fauna de Mártioda. DFA
Patricia Lizarraga/ Laura Elorza. Centro de Recuperación de
Fauna de Mártioda. DFA
Patricia Lizarraga/ Laura Elorza. Centro de Recuperación de
Fauna de Mártioda. DFA
Aitor Valdeón
Fermin Urra.VRFN. Gobierno de Navarra
Fermin Urra.VRFN. Gobierno de Navarra
Fermin Urra.VRFN. Gobierno de Navarra
Fermin Urra.VRFN. Gobierno de Navarra
Fermin Urra.VRFN. Gobierno de Navarra
Institution and/or Collector
Mm3
Mm2
Mm2
Mm1
Mm1
Mm1
Mm1
Mm1
Mm1
Mm1
Mm1
Mm1
Mm1
Mm1
Mm1
Mm1
Mm1
Mm1
Mm1
Mm1
Mm1
Mm1
Haplotype
Appendix S1 List of origins and collectors of pine marten samples used in this study. The sample code of museums and of samples from our collection is included with the
correspondent haplotype found. See Table 1 for remaining correspondences.
CHAPTER 2
SP_NW
SP_NW
SP_NW
SP_NW
SP_NW
18-SG
23-SG
26-SG
33-SG
32-SG
1JNA
33
34
35
36
37
16JNA
JNA01
408/04
34KA
44-SG
53-SG
M405/08
CPV-01
79EA(II)
Mma12RO
46
47
48
49
50
51
52
53
54
15JNA
13JNA
11JNA
10JNA
6JNA
45
44
43
42
41
40
39
38
3JNA
SP_NW
17-SG
31
SP_NW
SP_NW
SP_NW
SP_NW
SP_NW
SP_NW
SP_NW
SP_NE
SP_NW
SP_NW
SP_NW
SP_NW
SP_NW
SP_NW
SP_NW
SP_NW
SP_NW
SP_NW
SP_NW
SP_NW
32
7-JB
28
SP_NW
SP_NW
6-JB
27
SP_NW
SP_NW
8-RP
1-JB
26
Mma19RO
19 -SG
M 768/08
25
SP_NW
30
M154
24
SP_NE
29
481/07
23
Phylogeography of the European pine marten
Asturias; Cortrella
Cantabria; Praves
- 75 -
Leon; Posada de Valdeón; Collada Solano
Asturias; El pando
Lugo; Xermade; A Fraguela
Lugo; Angueiro
Bizkaia; Karrantza
Álava; Villlamaderne
Asturias; Teverga;Taja
Asturias
Asturias;Cangas del Narcea; La linde
Asturias
Asturias;Santo Adriano; Tuñon
Asturias
Asturias; Cangas del Narcea; Portiella
Asturias; San Antolín de Beón
Asturias; Narcea; Calabazos
Lugo; Fazai (Carballido)
A coruña; Cedeira; Esteiro
A coruña; Cedeira; Meizoso
Lugo; A fonsagrada
A coruña; Mera; Ortigueira
A Coruña;Cerdido; Rebordaos
Lugo;A fonsagrada; As rodas
Galicia
Lugo; Bóveda
Lugo; O corgo; N-546 km6;4
Lugo; O corgo; N-546 km6;6
Lugo; O corgo; N-546 km7
Asturias; Piloña
Asturias
Alava; Legutiano; A-2620 KM19
43,39460400
43,26901068
43,12387191
43,37807400
43,35138328
43,01632300
43,26901068
43,26901068
43,29174898
43,24782365
43,06308705
43,24782365
43,29174898
43,24782365
43,39460400
43,03735976
43,39460400
43,04032479
43,63527589
43,70355400
43,12299165
43,67571353
43,62383500
43,12977133
42,27444466
42,60446666
42,93691954
42,93428649
42,92882818
43,32024600
43,29761456
42,99377749
-6,07794899
-3,28426301
-4,99164176
-5,31081300
-7,80821744
-7,48073297
-3,28426301
-3,28426301
-5,98186293
-5,11365973
-6,50702975
-5,11365973
-5,98186293
-5,11365973
-6,07794899
-6,86972424
-6,07794899
-7,45345505
-8,04385823
-7,95118899
-7,07350302
-7,84246369
-7,99496099
-7,08367270
-7,12388891
-7,45921666
-7,48630786
-7,48351762
-7,48015053
-5,35208800
-5,28447506
-2,63755586
Paper II
Rita Oliveira. CIBIO. (Universidad de Oviedo)
Igor Aguinaco Amigo
Parque Nacional de Picos de Europa
Manuel Arzúa Piñeiro. Sociedad Galega de Historia Natural
Lida Mamán/Christian Gortázar. Instituto de Investigación
en Recursos Cinegéticos
Manuel Arzúa Piñeiro. Sociedad Galega de Historia Natural
Igor Aguinaco Amigo
Alberto Fdz. Gil. EBD-CSIC. Universidad de Oviedo
Patricia Lizarraga/ Laura Elorza. Centro de Recuperación de
Fauna de Mártioda. DFA
Alberto Fdz. Gil. EBD-CSIC. Universidad de Oviedo
Alberto Fdz. Gil. EBD-CSIC. Universidad de Oviedo
Alberto Fdz. Gil. EBD-CSIC. Universidad de Oviedo
Alberto Fdz. Gil. EBD-CSIC. Universidad de Oviedo
Alberto Fdz. Gil. EBD-CSIC. Universidad de Oviedo
Alberto Fdz. Gil. EBD-CSIC. Universidad de Oviedo
Alberto Fdz. Gil. EBD-CSIC. Universidad de Oviedo
Alberto Fdz. Gil. EBD-CSIC. Universidad de Oviedo
Manuel Arzúa Piñeiro. Sociedad Galega de Historia Natural
Catuxa Varela Sociedad Galega de Historia Natural
Manuel Arzúa Piñeiro. Sociedad Galega de Historia Natural
Manuel Arzúa Piñeiro. Sociedad Galega de Historia Natural
Manuel Arzúa Piñeiro. Sociedad Galega de Historia Natural
Manuel Arzúa Piñeiro. Sociedad Galega de Historia Natural
Manuel Arzúa Piñeiro. Sociedad Galega de Historia Natural
Rita Oliveira. CIBIO. (Vicente)
Rubén Portas Pérez
Juan Fernando Berezo
Juan Fernando Berezo
Juan Fernando Berezo
Patricia Lizarraga/ Laura Elorza. Centro de Recuperación de
Fauna de Mártioda. DFA
Montserrat Bosch. Institut de Recerca i Tecnologia
Agroalimentàries
Lida Mamán/Christian Gortázar. Instituto de Investigación
en Recursos Cinegéticos
Monteperdido.
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Mm8
Mm7
MM6
Mm6
Mm6
Mm5
Mm5
Mm4
CHAPTER 2
Italy
Portugal
Mma12
M1IT14
M1IT15
6-SAR
4-SAR
M5IT48
M5IT57
M5IT53
M5IT54
M5IT56
M5IT58
73
74
75
76
77
78
79
80
81
82
Mma18
69
72
Mma 28
68
M1IT6
Mma 27
67
Mma19
M1ITL2
66
71
M1IT9
65
70
M1IT3
08-0084
63
64
07-0108
62
Mma20RO
59
05-0014
Mma16RO
58
61
Mma15RO
57
04-0045
Mma14RO
56
60
Mma13RO
55
IT_IS
IT_IS
IT_IS
IT_IS
IT_IS
IT_IS
IT_IS
IT_IS
IT_CS
IT_CS
IT_CS
IT_CS
IT_CS
IT_CS
IT_CS
IT_CS
IT_CS
IT_CS
IT_CS
POR
POR
POR
POR
SP_NW
SP_NW
SP_NW
SP_NW
SP_NW
Sardinia; Nuoro;Loceri
Sardinia; Nuoro;Barisardo
Sardinia; Nuoro;Lanusei
- 76 -
Sardinia; Cagliari;Fluminimaggiore
Sardinia; Nuoro;Urzulei
Sardinia; Cagliari
Sardinia; North
Sardinia; North
Umbria
Roma; RN Macia dil Barco
Roma
Viterbo; Lagp di vico
Viterbo
Viterbo
Viterbo
Viterbo; Lago di vico
Viterbo
National Parc Peneda Geres; Terras de Bouro
National Parc Peneda Geres; Terras de Bouro
National Parc Peneda Geres; Terras de Bouro
National Parc Peneda Geres
Galicia
Galicia; Muñiz
Asturias; Calabazos
Asturias;Somiedo
Asturias;Somiedo
39,86016701
39,85494709
39,87920199
39,44478297
40,09453700
39,43216161
40,98009438
40,98009438
42,52842098
42,04875250
42,04875250
42,34100751
42,39108129
42,37349444
42,36014230
42,34100751
42,39333954
41,71407188
41,71407188
41,74559200
41,88073377
42,41527777
43,13032850
43,39460400
43,12180876
43,11872900
9,58501799
9,64836872
9,54136699
8,49820097
9,50861100
9,22946860
9,23969070
9,23969070
12,69643916
12,67613778
12,67613778
12,15811383
12,11295292
12,08214117
12,07199528
12,15811383
12,09319886
-8,29960650
-8,29960650
-8,16972899
-8,23275132
-7,22222222
-6,89816883
-6,07794899
-6,26139072
-6,28008600
Phylogeography
Ettore Randi. Conservation Genetics Laboratory. ISPRA
Licia Colli. Istituto di Zootecnica - Facoltà di Agraria
Università Cattolica del S. Cuore
Licia Colli. Istituto di Zootecnica - Facoltà di Agraria
Università Cattolica del S. Cuore
Paolo Casula. Ente Foreste della Sardegna Direzione
Generale
Paolo Casula. Ente Foreste della Sardegna Direzione
Generale
Licia Colli. Istituto di Zootecnica - Facoltà di Agraria
Università Cattolica del S. Cuore
Licia Colli. Istituto di Zootecnica - Facoltà di Agraria
Università Cattolica del S. Cuore
Licia Colli. Istituto di Zootecnica - Facoltà di Agraria
Università Cattolica del S. Cuore
Licia Colli. Istituto di Zootecnica - Facoltà di Agraria
Università Cattolica del S. Cuore
Licia Colli. Istituto di Zootecnica - Facoltà di Agraria
Università Cattolica del S. Cuore
Licia Colli. Istituto di Zootecnica - Facoltà di Agraria
Università Cattolica del S. Cuore
Ettore Randi. Conservation Genetics Laboratory. ISPRA
Ettore Randi. Conservation Genetics Laboratory. ISPRA
Licia Colli. Istituto di Zootecnica - Facoltà di Agraria
Università Cattolica del S. Cuore
Ettore Randi. Conservation Genetics Laboratory. ISPRA
Ettore Randi. Conservation Genetics Laboratory. ISPRA
BTVS-ICN
Licia Colli. Istituto di Zootecnica - Facoltà di Agraria
Università Cattolica del S. Cuore
Licia Colli. Istituto di Zootecnica - Facoltà di Agraria
Università Cattolica del S. Cuore
Licia Colli. Istituto di Zootecnica - Facoltà di Agraria
Università Cattolica del S. Cuore
BTVS-ICN
BTVS-ICN
BTVS-ICN
Rita Oliveira. CIBIO. (Vicente)
Rita Oliveira. CIBIO. (Pablo Siena)
Rita Oliveira. CIBIO. (Universidad de Oviedo)
Rita Oliveira. CIBIO. (Universidad de Oviedo)
Rita Oliveira. CIBIO. (Universidad de Oviedo)
Mm15
Mm15
Mm15
Mm15
Mm14
Mm14
Mm14
Mm14
Mm13
Mm13
Mm12
Mm11
Mm11
Mm10
Mm10
Mm10
Mm10
Mm10
Mm10
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Ireland
Croatia
MM2-AB
95
2-DA
4-CR
105
107
3-CR
104
1-DA
2-CR
103
106
1-CR
102
M5IT59
M1ITL5
94
101
MM1-AB
93
M5IT55
M1ITL7
92
100
Mma 11
91
M1ITL1
Mma 10
90
99
Mma 9
89
2-SAR
Mma 8
88
98
M1ITL6
87
13451
M1ITL4
86
2009-0027
M1IT8
85
97
1999-1002
84
96
M5IT60
83
IR
IR
CRO
CRO
CRO
CRO
IT_IS
IT_IS
IT_CS
IT_IS
IT_IS
IT_IS
IT_N
IT_N
IT_N
IT_N
IT_N
IT_N
IT_N
IT_N
IT_N
IT_N
IT_CS
IT_CS
IT_IS
Phylogeography of the European pine marten
Near Zagreb
Near Zagreb
Near Zagreb
Near Zagreb
- 77 -
Sardinia; Cagliari;gonnosfanadiga
Sardinia; Nuoro;Villagrande/Fonni
Pisa
Sardinia; North
Colle d'Orano-Marciana; Isola de Elba
Toscana; Livorno; Isola de Elba
Riserva della Garzaia di Valenza
Cuneo; Centallo
Riserva della Garzaia di Valenza
Alessandria
Weissmatteu
Weissmatteu
Weissmatteu
Weissmatteu
Torino; Piobesi Torinesi
Cuneo
Varese; Zenna;Lago di vico
Varese; Zenna;Lago di vico
Sardinia; Cagliari;Guspini
53,41291000
53,41291000
45,72817047
45,72817047
45,72817047
45,72817047
39,48447939
40,10529427
43,41975358
40,98009438
42,78769899
42,83555283
45,19260555
44,45001818
45,19260555
44,97905929
45,74671245
45,74671245
45,74671245
45,74671245
44,94046082
44,37075659
46,10173488
46,10173488
39,54604904
-8,24388999
-8,24388999
15,92560545
15,92560545
15,92560545
15,92560545
8,66104444
9,25232293
10,66944733
9,23969070
10,11273900
10,40609787
8,71495019
7,61965020
8,71495019
8,43353631
7,82974144
7,82974144
7,82974144
7,82974144
7,61058238
7,31307139
8,75477319
8,75477319
8,61804930
Paper II
Adriano de Faveri. ISPRA (F. Gannini)
Paolo Casula. Ente Foreste della Sardegna Direzione
Generale
Licia Colli. Istituto di Zootecnica - Facoltà di Agraria
Università Cattolica del S. Cuore
Licia Colli. Istituto di Zootecnica - Facoltà di Agraria
Università Cattolica del S. Cuore
Licia Colli. Istituto di Zootecnica - Facoltà di Agraria
Università Cattolica del S. Cuore
Ana Galov.Department of Animal Physiology. Faculty of
Science. Zagreb (Konjevié)
Ana Galov.Department of Animal Physiology. Faculty of
Science. Zagreb (Konjevié)
Ana Galov.Department of Animal Physiology. Faculty of
Science. Zagreb (Huber. Konjevié)
Ana Galov. Department of Animal Physiology. Faculty of
Science. Zagreb (Konjevié)
Angus Davison.Institute of Genetics, School of Biology,
Queen's Medical Centre, University of Nottingham (J.
Higgins)
Angus Davison.Institute of Genetics, School of Biology,
Queen's Medical Centre, University of Nottingham (J.
Ettore Randi. Conservation Genetics Laboratory. ISPRA
Licia Colli. Istituto di Zootecnica - Facoltà di Agraria
Università Cattolica del S. Cuore
Alessandro Ballestrieri. Dipartimento di Biologia Animale,
Università di Pavia.
Licia Colli. Istituto di Zootecnica - Facoltà di Agraria
Università Cattolica del S. Cuore
Alessandro Ballestrieri. Dipartimento di Biologia Animale,
Università di Pavia.
Museo Zoologico "la Specola" Sezione del MSN-Universita
degli studio di Firenze
Ettore Randi. Conservation Genetics Laboratory. ISPRA
Ettore Randi. Conservation Genetics Laboratory. ISPRA
Ettore Randi. Conservation Genetics Laboratory. ISPRA
Adriano de Faveri. ISPRA (E. Lenzo)
Licia Colli. Istituto di Zootecnica - Facoltà di Agraria
Università Cattolica del S. Cuore
Licia Colli. Istituto di Zootecnica - Facoltà di Agraria
Università Cattolica del S. Cuore
Licia Colli. Istituto di Zootecnica - Facoltà di Agraria
Università Cattolica del S. Cuore
Licia Colli. Istituto di Zootecnica - Facoltà di Agraria
Università Cattolica del S. Cuore
Mm20
Mm20
Mm19
Mm19
Mm19
Mm19
Mm9
Mm9
Mm9
Mm9
Mm9
Mm9
Mm48
Mm48
Mm47
Mm47
Mm18
Mm18
Mm18
Mm18
Mm17
Mm17
Mm16
Mm16
Mm15
CHAPTER 2
France
Scotland
TC-174
TC-271
1051
6759
127
128
129
Mma 25
123
126
Mma 24
122
TC-166
Mma 23
121
125
Mma 22
120
1999-316
Mma 21
119
124
21-DA
11-DA
116
Mma 20
10-DA
115
118
9-DA
114
117
8-DA
6-DA
111
113
5-DA
110
7-DA
4-DA
109
112
3-DA
108
FR
FR
FR
FR
FR
FR
SC
SC
SC
SC
SC
SC
SC
SC
SC
SC
SC
SC
IR
IR
IR
IR
- 78 -
Forêt domaniale de Chinon
47,16604588
48,61890900
46,00977400
France (carrefour N 17 D 205; Nantiat;
Haute Vienne)
Boissise-la-Bertrand; Seine et Marne
47,30777631
44,41987426
48,64291551
57,42135742
57,00774112
57,00774112
57,00774112
57,11662923
57,11662923
57,00774112
56,41041002
56,41041002
56,41041002
56,41041002
56,41041002
53,41291000
53,41291000
52,85897400
Cher; Neuvy sur Barangeon
Landes; road Biscarrosse to lac of Navarosse
Rambouillet (National Road 306)
Inverness
Speyside
Northern Highlands
Northern Highlands
North Western Seabord
North Western Seabord
A9 road opposite Crubenmore Junction
Road near Oban
Field Barcaldine; Oban 22
Benderloch Road; Oban 20
Benderloch Road; Oban
Clare
53,41291000
0,23811010
2,97564000
1,17575300
2,45025979
-1,17303537
1,82584801
-4,25429108
-4,16618402
-4,16618402
-4,16618402
-5,12257116
-5,12257116
-4,16618402
-5,46972300
-5,46972300
-5,46972300
-5,46972300
-5,46972300
-8,24388999
-8,24388999
-9,11004899
-8,24388999
Phylogeography
Geraldine Veron. Muséum National d'Histoire Naturelle.
Ettore Randi. Conservation Genetics Laboratory. ISPRA
Geraldine Veron. Muséum National d'Histoire Naturelle.
Département Systématique et Evolution ( R. MichelKerneur)
Geraldine Veron. Muséum National d'Histoire Naturelle.
Département Systématique et Evolution (Y. Grugier)
Geraldine Veron. Muséum National d'Histoire Naturelle.
Département Systématique et Evolution (R. Dohogne et Y.
Grugier)
Geraldine Veron. Muséum National d'Histoire Naturelle.
Département Systématique et Evolution (R. Dohogne et Y.
Grugier)
Geraldine Veron. Muséum National d'Histoire Naturelle.
Département Systématique et Evolution (R. Cornette)
Ettore Randi. Conservation Genetics Laboratory. ISPRA
Ettore Randi. Conservation Genetics Laboratory. ISPRA
Ettore Randi. Conservation Genetics Laboratory. ISPRA
Ettore Randi. Conservation Genetics Laboratory. ISPRA
Ettore Randi. Conservation Genetics Laboratory. ISPRA
Angus Davison.Institute of Genetics, School of Biology,
Queen's Medical Centre, University of Nottingham
(Congella McGuire)
Angus Davison.Institute of Genetics, School of Biology,
Queen's Medical Centre, University of Nottingham (Paddy
Sleeman)
Angus Davison.Institute of Genetics, School of Biology,
Queen's Medical Centre, University of Nottingham (Pat
Smiddy)
Angus Davison.Institute of Genetics, School of Biology,
Queen's Medical Centre, University of Nottingham (Pat
Smiddy)
Angus Davison.Institute of Genetics, School of Biology,
Queen's Medical Centre, University of Nottingham
Angus Davison.Institute of Genetics, School of Biology,
Queen's Medical Centre, University of Nottingham
Angus Davison.Institute of Genetics, School of Biology,
Queen's Medical Centre, University of Nottingham
Angus Davison.Institute of Genetics, School of Biology,
Queen's Medical Centre, University of Nottingham
Angus Davison.Institute of Genetics, School of Biology,
Queen's Medical Centre, University of Nottingham
Angus Davison.Institute of Genetics, School of Biology,
Queen's Medical Centre, University of Nottingham
Higgins)
Mm31
Mm31
Mm31
Mm22
Mm21
Mm29
Mm28
Mm28
Mm28
Mm28
Mm28
Mm28
Mm28
Mm28
Mm28
Mm28
Mm28
Mm28
Mm20
Mm20
Mm20
Mm20
Germany
Netherlands
Austria
Luxembourg
M4008
ALE219
3-FM
151
152
06/004
06/002
04/175
04/029
04/006
00/369
99/044
99/035
94/043
150
149
148
147
146
145
144
143
142
141
140
2006/398
93/046
6-LUX
136
139
5-LUX
135
2004/343
4-LUX
134
138
3-LUX
133
2003/261
2-LUX
132
137
1-LUX
M242
131
130
DE
DE
DE
NTH
NTH
NTH
NTH
NTH
NTH
NTH
NTH
NTH
NTH
AU
AU
AU
LUX
LUX
LUX
LUX
LUX
LUX
FR
Phylogeography of the European pine marten
- 79 -
Niederrode ;Krifulda; Hessen
Hohenzell/Kreis Main-Kinzig;Spessart
Niesky 2km O; Strabe Ritchung Horka
Motor way A1; near Apeldoorn; NL
Motor way N786; between Laag Soeren and
Eerbeek
Heelsum; NL
Eerbeek
Veluwe
Wekerom
Motor way A28; near ‘t Harde
Garderen
Motor way N-304; near Apeldoorn; NL
Hulshorst
Wiltz
Bous
Canach
Steinsel
Bridel
Near Dijon
50,52138888
50,32333611
51,28768652
52,17307025
51,98049300
52,09244944
52,09795507
52,14657115
52,10190800
52,41337751
52,18600411
52,21845546
52,36370443
48,25000000
48,43333333
48,50000000
49,94910426
49,55431689
49,59883146
49,59883146
49,68762102
49,65564185
47,51140500
9,60416666
9,53581388
14,85677038
5,97146268
5,75599400
6,08905671
6,06849646
5,88543480
5,72655039
5,90082814
5,91938830
5,72483899
5,73225469
13,80000000
14,46666666
14,26666666
5,91456533
6,35190424
6,30241122
6,30241122
6,12912355
6,05601978
5,28013900
Paper II
Franz Müller
Franz Müller
Département Systématique et Evolution (Office National
de la Chasse)
Montserrat Bosch. Institut de Recerca i Tecnologia
Agroalimentàries
Jan Herr.Department of Biology and Environmental
Science.University of Sussex.UK
Jan Herr.Department of Biology and Environmental
Science.University of Sussex.UK
Jan Herr.Department of Biology and Environmental
Science.University of Sussex.UK
Jan Herr.Department of Biology and Environmental
Science.University of Sussex.UK
Jan Herr.Department of Biology and Environmental
Science.University of Sussex.UK
Jan Herr.Department of Biology and Environmental
Science.University of Sussex.UK
Barbara Herzig. Mammal Collection. The Natural History
Museum Vienna (Braunschmid Otto)
Barbara Herzig. Mammal Collection. The Natural History
Museum Vienna (Plass Jurgen)
Barbara Herzig. Mammal Collection. The Natural History
Museum Vienna
Hein van Grouw. National Museum of Natural History,
Naturalis. Lieden, Netherlands
Hein van Grouw. National Museum of Natural History,
Naturalis. Lieden, Netherlands
Hein van Grouw. National Museum of Natural History,
Naturalis. Lieden, Netherlands
Hein van Grouw. National Museum of Natural History,
Naturalis. Lieden, Netherlands
Hein van Grouw. National Museum of Natural History,
Naturalis. Lieden, Netherlands
Hein van Grouw. National Museum of Natural History,
Naturalis. Lieden, Netherlands
Hein van Grouw. National Museum of Natural History,
Naturalis. Lieden, Netherlands
Hein van Grouw. National Museum of Natural History,
Naturalis. Lieden, Netherlands
Hein van Grouw. National Museum of Natural History,
Naturalis. Lieden, Netherlands
Hein van Grouw. National Museum of Natural History,
Naturalis. Lieden, Netherlands
Herman Ansorge. Staatliches Museum fur Naturkunde
Gorlitz. Germany
Mm29
Mm27
Mm18
Mm30
Mm29
Mm29
Mm30
Mm29
Mm29
Mm30
Mm29
Mm29
Mm18
Mm25
Mm32
Mm32
Mm31
Mm31
Mm29
Mm42
Mm26
Mm31
Mm31
CHAPTER 2
5-POL
23-POL
81-POL
2-POL
18-POL
175
176
177
178
1-FM
170
174
M3959
169
84-POL
2-FM
168
173
ALE207
167
21-POL
M4007
166
172
M4223
165
Poland
ALE218
164
NM303
M4222
163
171
ALE212
162
Czech
Republic
ALE209
158
161
M5422
157
M5623
M3997
156
ALE208
ALE206
155
160
ALE205
154
159
4-FM
ALE204
153
PL
PL
PL
PL
PL
PL
PL
CR
DE
DE
DE
DE
DE
DE
DE
DE
DE
DE
DE
DE
DE
DE
DE
DE
DE
DE
Steudenec
- 80 -
Obrelsbach/ Rhön-Grabfeldkreis; Bayern
Cablau Zerna
Altenfeld/kuifulda; Hesse
Kreis Marburg-Biedenkopf; Wetter
KaupPa; Strabe am DelitsZcher Teich
Niesky Aral Tankstelle B115
Limeshain-Hainchen; Wetteran-Kreis
Rhön-Grabfeldkreis; Hausen/Lange Rhön
Klosterwald B99 zwischen Schelegel und
Marienthal
Hunsnick; Morbad
Zell-Barl/Mosel
Ostritz Hirschfelde B99 Klosterwald
Maukendorf; Abzweig B96 Spohla; Richtung
Hoyerswerda
Lieske südlich B156
Rhön-Grabfeldkreis; Lange Rhön
Rhön-Grabfeldkreis; Lange Rhön
Rhön-Grabfeldkreis; Lange Rhön
Wickers;Krifulda; Hessen
51,57623102
53,58333333
53,75000000
54,08333333
52,33333333
51,28333333
53,83333333
49,21666666
50,48333333
51,28149297
50,45809722
50,44059444
51,25220174
51,27904931
50,27044166
50,98366191
50,50277777
49,81071388
50,03336388
51,40770937
50,96617874
51,32097643
50,47472222
50,50861111
50,54166666
50,54282222
23,55001938
22,75000000
21,91691666
23,09969444
20,75000000
19,66666666
23,25007777
16,06666666
10,03333333
14,23630794
9,87573888
8,76850000
14,48862585
14,81818783
8,99528611
14,89154142
10,06666666
7,11984722
7,14998333
14,29062200
14,88957468
14,53358656
10,02500000
10,10555555
10,06388888
9,98613333
Phylogeography
Franz Müller
Natalia Martinkova.Institute of Vertebrate Biology,
Academy of Science of the Czech Republic.
Malgorzata Pilot. Museum and Institute of Zoology, Polish
Academy of Sciences.
Malgorzata Pilot. Museum and Institute of Zoology, Polish
Academy of Sciences.
Malgorzata Pilot. Museum and Institute of Zoology, Polish
Academy of Sciences.
Malgorzata Pilot. Museum and Institute of Zoology, Polish
Academy of Sciences.
Malgorzata Pilot. Museum and Institute of Zoology, Polish
Academy of Sciences.
Malgorzata Pilot. Museum and Institute of Zoology, Polish
Academy of Sciences.
Malgorzata Pilot. Museum and Institute of Zoology, Polish
Academy of Sciences.
Franz Müller
Herman Ansorge. Staatliches Museum fur Naturkunde
Gorlitz. Germany
Herman Ansorge. Staatliches Museum fur Naturkunde
Gorlitz. Germany
Herman Ansorge. Staatliches Museum fur Naturkunde
Gorlitz. Germany
Herman Ansorge. Staatliches Museum fur Naturkunde
Gorlitz. Germany
Herman Ansorge. Staatliches Museum fur Naturkunde
Gorlitz. Germany
Herman Ansorge. Staatliches Museum fur Naturkunde
Gorlitz. Germany
Herman Ansorge. Staatliches Museum fur Naturkunde
Gorlitz. Germany
Franz Müller
Franz Müller
Franz Müller
Herman Ansorge. Staatliches Museum fur Naturkunde
Gorlitz. Germany
Herman Ansorge. Staatliches Museum fur Naturkunde
Gorlitz. Germany
Herman Ansorge. Staatliches Museum fur Naturkunde
Gorlitz. Germany
Franz Müller
Franz Müller
Franz Müller
Mm34
Mm34
Mm33
Mm33
Mm33
Mm31
Mm24
Mm9
Mm9
Mm9
Mm41
Mm37
Mm34
Mm34
Mm33
Mm33
Mm32
Mm31
Mm31
Mm31
Mm31
Mm31
Mm29
Mm29
Mm29
Mm29
8-EST
11-EST(1)
11-EST(2)
13-EST(1)
16-EST
17-EST
2-EST
3-EST
4-EST
5-EST
12-EST(1)
1-EST
191
192
193
194
195
196
197
198
199
200
201
202
6-POL
187
7-EST
41-POL
186
190
30-POL
185
6-EST
27-POL
184
189
74-POL
183
Estonia
65-POL
182
1-LT
48-POL
181
188
44-POL
180
Latvia
26-POL
179
EST
EST
EST
EST
EST
EST
EST
EST
EST
EST
EST
EST
EST
EST
LT
PL
PL
PL
PL
PL
PL
PL
PL
PL
Phylogeography of the European pine marten
Selja
Hiuma Island; Leluselja
Hiuma Island
Hiuma Island
Hiuma Island
Hiuma Island
Hiuma Island; Käina
Hiuma Island; Määvli
Hiuma Island; Suuremoisa
Hiuma Island; Körgessaare
Hiuma Island; Körgessaare
Hiuma Island
Hiuma Island
Hiuma Island
- 81 -
59,28944842
58,81344300
58,88090343
58,88090343
58,88090343
58,88090343
58,82722800
58,94729400
58,87107853
58,98205801
58,98205801
58,88090343
58,88090343
58,88090343
55,75500000
53,95000000
49,75000000
52,33333333
53,83333333
52,08333333
51,16666666
52,08333333
51,78333333
53,58346388
24,53996106
22,56266999
22,69212648
22,69212648
22,69212648
22,69212648
22,77641199
22,77162200
22,94396588
22,47103186
22,47103186
22,69212648
22,69212648
22,69212648
26,23000000
23,08335277
22,25000000
20,50000000
21,50000000
21,16666666
23,20000000
21,16666666
20,08333333
21,00000000
Paper II
Asun Gómez. TRAGSEGA
Robert Myslajek
Maddis Podra.Institute of Mathematics and Natural
Sciences, Tallinn University
Maddis Podra.Institute of Mathematics and Natural
Sciences, Tallinn University
Maddis Podra.Institute of Mathematics and Natural
Sciences, Tallinn University
Maddis Podra.Institute of Mathematics and Natural
Sciences, Tallinn University
Maddis Podra.Institute of Mathematics and Natural
Sciences, Tallinn University
Maddis Podra.Institute of Mathematics and Natural
Sciences, Tallinn University
Maddis Podra.Institute of Mathematics and Natural
Sciences, Tallinn University
Maddis Podra.Institute of Mathematics and Natural
Sciences, Tallinn University
Maddis Podra.Institute of Mathematics and Natural
Sciences, Tallinn University
Maddis Podra.Institute of Mathematics and Natural
Sciences, Tallinn University
Maddis Podra.Institute of Mathematics and Natural
Sciences, Tallinn University
Maddis Podra.Institute of Mathematics and Natural
Sciences, Tallinn University
Maddis Podra.Institute of Mathematics and Natural
Sciences, Tallinn University
Malgorzata Pilot. Museum and Institute of Zoology, Polish
Academy of Sciences.
Malgorzata Pilot. Museum and Institute of Zoology, Polish
Academy of Sciences.
Malgorzata Pilot. Museum and Institute of Zoology, Polish
Academy of Sciences.
Malgorzata Pilot. Museum and Institute of Zoology, Polish
Academy of Sciences.
Malgorzata Pilot. Museum and Institute of Zoology, Polish
Academy of Sciences.
Malgorzata Pilot. Museum and Institute of Zoology, Polish
Academy of Sciences.
Malgorzata Pilot. Museum and Institute of Zoology, Polish
Academy of Sciences.
Malgorzata Pilot. Museum and Institute of Zoology, Polish
Academy of Sciences.
Malgorzata Pilot. Museum and Institute of Zoology, Polish
Academy of Sciences.
Mm50
Mm40
Mm40
Mm40
Mm40
Mm40
Mm33
Mm33
Mm33
Mm33
Mm33
Mm33
Mm33
Mm33
Mm23
Mm44
Mm36
Mm35
Mm35
Mm34
Mm34
Mm34
Mm34
Mm34
CHAPTER 2
NRM20035159
NRM20055085
NRM20075094
NRM985120
NRM985574
NRM985582
NRM985586
NRM985639
NRM985187
NRM985198
NRM985146
NRM985179
NRM985601
NRM20035073
NRM845006
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
M275
NRM995178
217
235
NRM985172
216
M273
NRM915124
215
234
NRM20035099
214
M277
NRM985128
213
233
NRM20075092
212
Finland
NRM20055214
211
1-RU
86.7.1.
208
NRM20035102
97.31.3
207
210
4090
206
209
20241
205
Sweden
87.71.1
204
10-EST
Romania
Hungary
203
FIN
FIN
FIN
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
RO
HU
HU
HU
HU
HU
EST
Kihnio¨
Kihnio¨
Kihnio¨
Luleå; Norrbotten
- 82 -
Luovaure; Snesudden; Lule lappmark
Björsjö; Dalarna
Blidsberg; Västergötland
Blidsberg; Västergötland
Fredriksberg; Dalarna
Fredriksberg; Dalarna
Närkes Kil; Närke
Bofors target rage; Värmland
Närkes Kil; Närke
Bofors target rage; Värmland
Road E18; Närke
Vellinge; Bernstorp;Skåne
Vissefjärda; Hulan; Småland
Ösmo - Muskö; Södermanland
Stockholm; Norra Djurgården; Uppland
Filipstad; Värmland
Strängnäs; Södermanland
Falsterbo; Falsterbonäset;Skåne
Skoghall;Västmanland
Vellinge; Bernstorp;Skåne
Vellinge; Foteviken; Bernstorp;Skåne
Falsterbo; Falsterbonäset;Skåne
South Romania
Zala; Keszthely
Györ-Sopron
Csongrád;Makó
Borsod-Abaúj-Zemplén
Baranya;Pécs
Hiuma Island; Laasi
62,20369302
62,20369302
62,20369302
65,58415700
66,14525598
60,04679012
57,92933532
57,92933532
60,12893525
60,12893525
59,36883208
59,35504330
59,37543690
59,35504330
59,24024556
55,45109543
56,55138699
58,98933335
59,35795744
59,67238588
59,29713428
55,39719335
59,33585674
55,44633740
55,44861761
55,39108868
43,99382476
46,76803410
47,61008268
46,16918629
48,28404276
46,09237287
58,89824909
23,17845301
23,17845301
23,17845301
22,15474900
20,36294469
15,34237103
13,49247522
13,49247522
14,39594607
14,39594607
15,07351632
14,54297605
15,12697889
14,54297605
14,61522526
12,97441338
15,67207999
18,09881774
18,06153989
14,03150943
17,09430090
12,82884889
13,47392284
12,97735054
12,97323506
12,84022203
24,45744393
17,21430563
16,80659278
20,28222455
20,64434073
18,22513205
22,31818988
Phylogeography
Peter Mortensen. Swedish Museum of Natural History
Montserrat Bosch. Institut de Recerca i Tecnologia
Agroalimentàries
Montserrat Bosch. Institut de Recerca i Tecnologia
Agroalimentàries
Montserrat Bosch. Institut de Recerca i Tecnologia
Agroalimentàries
Peter Mortensen. Swedish Museum of Natural History
Peter Mortensen. Swedish Museum of Natural History
Peter Mortensen. Swedish Museum of Natural History
Peter Mortensen. Swedish Museum of Natural History
Peter Mortensen. Swedish Museum of Natural History
Peter Mortensen. Swedish Museum of Natural History
Peter Mortensen. Swedish Museum of Natural History
Peter Mortensen. Swedish Museum of Natural History
Peter Mortensen. Swedish Museum of Natural History
Peter Mortensen. Swedish Museum of Natural History
Peter Mortensen. Swedish Museum of Natural History
Peter Mortensen. Swedish Museum of Natural History
Peter Mortensen. Swedish Museum of Natural History
Peter Mortensen. Swedish Museum of Natural History
Peter Mortensen. Swedish Museum of Natural History
Peter Mortensen. Swedish Museum of Natural History
Peter Mortensen. Swedish Museum of Natural History
Peter Mortensen. Swedish Museum of Natural History
Peter Mortensen. Swedish Museum of Natural History
Peter Mortensen. Swedish Museum of Natural History
Peter Mortensen. Swedish Museum of Natural History
Peter Mortensen. Swedish Museum of Natural History
Gabor Csorba. Hungarian Natural History Museum
Dumitru Murariu. "Grigore Antipa" National Museum of
Natural History
Gabor Csorba. Hungarian Natural History Museum
Gabor Csorba. Hungarian Natural History Museum
Gabor Csorba. Hungarian Natural History Museum
Gabor Csorba. Hungarian Natural History Museum
Maddis Podra.Institute of Mathematics and Natural
Sciences, Tallinn University
Mm58
Mm58
Mm58
Mm69
Mm69
Mm67
Mm46
Mm46
Mm45
Mm45
Mm45
Mm45
Mm45
Mm45
Mm45
Mm45
Mm45
Mm45
Mm45
Mm31
Mm31
Mm31
Mm18
Mm18
Mm18
Mm18
Mm39
Mm43
Mm29
Mm43
Mm38
Mm44
Mm50
Russia
Norway
T3-DS
T2-DS
3-DS
1-DS
25-DS
26-DS
10-DS
11-RUS
12-RUS
T4-DS
T5-DS
T6-DS
T7-DS
7-DS
24-DS
27-DS
28-DS
13-DS
16-DS
17-DS
31-DS
32DS
14-DS
9-RUS
T1-DS
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
NW-11849
242
243
NW-11425
241
239
NW-11365
Mma18RO
238
240
M279
Mma17RO
237
M278
236
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
NW
NW
NW
FIN
FIN
FIN
FIN
Phylogeography of the European pine marten
Kirov; Falyonskiy
Novgorod; Valdai Upland
Komi;Vuktyl
Kirov; Kilmez
Komi
Komi;Vuktyl
Komi;Vuktyl
Komi; Vuktyl
Kirov;Rogovoe
Kirov;Omutninsk
Kirov;Omutninsk
Kirov; Kilmez
Kirov;Falyonskiy
Kirov;Falyonskiy
Kirov;Falyonskiy
Kirov; Falyonskiy
Leningrad Province
Leningrad ; Vyritsa
Kirov;Rogovoe
Kirov;Omutninsk
Kirov;Omutninsk
Kirov;Kotelnich
Kirov;Kirov
Kirov; Falyonskiy
Kirov; Falyonskiy
- 83 -
Hedmark county;Fiskebekklia
Hedmark county; Hummelneset
Hedmark county; Fuggdal
Central Finland
Central Finland
Padasjoki
Padasjoki
58,28645716
58,17450399
64,08524410
56,93952999
63,45077326
63,79284541
63,74323549
63,86481284
58,55322641
58,47725498
58,60975100
57,10802810
58,28645716
58,28645716
58,28645716
58,28645716
59,83960937
59,40242953
58,55322641
58,50550889
58,68192200
57,90956709
58,64113102
58,28645716
58,28645716
60,66666666
60,66666666
61,63244887
64,61541603
64,35136365
61,35012100
61,35012100
51,72576388
33,25446861
57,78018880
51,07117099
58,13947871
57,85138037
57,22442054
57,49370460
50,72889005
52,53821763
52,15365141
50,35695684
51,72576388
51,72576388
51,72576388
51,72576388
32,60710947
30,34069745
50,72889005
52,20951833
52,45993112
48,42841075
49,73890856
51,72576388
51,72576388
10,00000000
10,00000000
8,12179595
27,41689010
26,38804286
25,27454015
25,27454015
Paper II
Dimitry V. Skumatov
Dimitry V. Skumatov
Alexei Abramov. Laboratory of Mammalogy, Zoological
Institute, Russian Academy of Sciences
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Rita Oliveira. CIBIO. (University of Jyväskylä)
Øystein Wiig. Mammal Collection. Natural History
Museum. University of Oslo
Øystein Wiig. Mammal Collection. Natural History
Museum. University of Oslo
Øystein Wiig. Mammal Collection. Natural History
Museum. University of Oslo
Rita Oliveira. CIBIO. (University of Jyväskylä)
Montserrat Bosch. Institut de Recerca i Tecnologia
Agroalimentàries
Montserrat Bosch. Institut de Recerca i Tecnologia
Agroalimentàries
Mm53
Mm52
Mm52
Mm51
Mm51
Mm51
Mm51
Mm51
Mm51
Mm51
Mm51
Mm51
Mm51
Mm51
Mm51
Mm51
Mm50
Mm50
Mm50
Mm50
Mm50
Mm50
Mm50
Mm50
Mm49
Mm68
Mm68
Mm68
Mm69
Mm59
Mm58
Mm69
Sable
(Martes zibellina)
CHAPTER 2
Russia
33DS
7-RUS
20-RUS
16-RUS
17-RUS
4-DS
14-RUS
Rsep9
Rsep7
T8-DS
Rsep8
Rsep11
19-RUS
Rsep6
10-RUS
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
1-MZ
20-DS
21-DS
293
294
291
292
4-MZ
19-DS
290
2-MZ
15-RUS
271
289
15-DS
270
18-DS
9-DS
269
288
2-DS
22-DS
268
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
RUS
Tomsk; Alexandrovskoe
Tomsk; Alexandrovskoe
- 84 -
Transbaikalia; Buryatia North; Barguzin Area
Tomsk;Alexandrovskoe
Transbaikalia; Buryatia South; Dzhida Area
Kamtchatka
Tomsk;Alexandrovskoe
Leningrad; Boksitogorsk
Tyumen region
Chelyabinsk; Krasnoarmeisk
Tyumen region
Tyumen region
Kirov;Falyonskiy
Near Kirov city
Near Kirov city
Penza; Sosnovka Village
Kirov; Rogovoe
Tver ; Bezhetsk
Tver ; Bezhetsk
Chelyabinsk; Krasnoarmeyskiy
Leningrad ; Podporozhye
Kirov; Kilmez
Ural; Perm ; Kizel
Komi;Vuktyl
Kirov;Slobodskoy
Kirov;Kotelnich
Kirov; Kirov
59,17328866
59,17328866
54,22023171
59,17328866
54,22023171
54,61531892
59,17328866
59,46937200
55,45643150
55,61168551
55,45643150
55,45643150
58,28645716
58,60344169
58,60344169
53,27628074
58,55322641
57,78443061
57,78443061
55,31316016
60,91789328
56,93952999
59,00919378
63,97211679
58,55322641
58,11629476
58,87975907
Dimitry V. Skumatov
Alexei Abramov. Laboratory of Mammalogy, Zoological
Institute, Russian Academy of Sciences
Dimitry V. Skumatov
Alexei Abramov. Laboratory of Mammalogy, Zoological
Institute, Russian Academy of Sciences
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Alexei Abramov. Laboratory of Mammalogy, Zoological
Institute, Russian Academy of Sciences
Dimitry V. Skumatov
Alexei Abramov. Laboratory of Mammalogy, Zoological
Institute, Russian Academy of Sciences
Alexei Abramov. Laboratory of Mammalogy, Zoological
Institute, Russian Academy of Sciences
Alexei Abramov. Laboratory of Mammalogy, Zoological
Institute, Russian Academy of Sciences
Alexei Abramov. Laboratory of Mammalogy, Zoological
Institute, Russian Academy of Sciences
Dimitry V. Skumatov
Alexei Abramov. Laboratory of Mammalogy, Zoological
Institute, Russian Academy of Sciences
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
79,08158564
79,08158564
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Alexei Abramov. Laboratory of Mammalogy, Zoological
111,47659877 Institute, Russian Academy of Sciences
79,08158564
Dimitry V. Skumatov
Alexei Abramov. Laboratory of Mammalogy, Zoological
158,39028767 Institute, Russian Academy of Sciences
Alexei Abramov. Laboratory of Mammalogy, Zoological
111,47659877 Institute, Russian Academy of Sciences
79,08158564
33,86022200
69,43747698
62,14651528
69,43747698
69,43747698
51,72576388
49,53794759
49,53794759
45,27853761
50,72889005
36,72009810
36,72009810
61,99983177
34,18951020
51,07117099
57,67843556
57,94858758
50,72889005
48,04265553
49,30391681
Phylogeography
Mz7
Mz6
Mz5
Mz4
Mz3
Mz2
Mz1
Mm69
Mm66
Mm66
Mm65
Mm64
Mm63
Mm62
Mm61
Mm60
Mm57
Mm56
Mm56
Mm55
Mm54
Mm53
Mm53
Mm53
Mm53
Mm53
Mm53
Unknow
NC011579
299
Rsep3
297
FJ429093
Rsep2
296
298
Rsep1
295
RUS
RUS
RUS
Phylogeography of the European pine marten
Tomsk; Alexandrovskoe
Tomsk; Alexandrovskoe
Tomsk; Alexandrovskoe
- 85 -
59,10532493
59,10532493
59,10532493
79,15567035
79,15567035
79,15567035
Paper II
Dimitry V. Skumatov
Dimitry V. Skumatov
Dimitry V. Skumatov
Mz11
Mz11
Mz10
Mz9
Mz8
CHAPTER 2
- 86 -
Phylogeography
CHAPTER
1
Species Identification
PAPER III*
A non-invasive genetic method to identify the sympatric
mustelids pine marten (Martes martes) and stone marten (Martes foina):
preliminary distribution survey on the northern Iberian
Peninsula
*The applicability and the usefulness of this method for improving our knowledge of different bio-ecological
traits of sympatric marten species has been demonstrated in other three different researches which are included as
Appendix A (Paper IIIa), B (Paper IIIb) and C (Paper IIIc).
- 87 -
3
CHAPTER 3
Species Identification
- 88 -
Non-invasive identification Martens
Paper III
PAPER III
A non-invasive genetic method to identify the sympatric
mustelids pine marten (Martes martes) and stone marten (Martes foina):
preliminary distribution survey on the northern Iberian
Peninsula
ABSTRACT
The closely related mustelids European pine marten (Martes martes) and stone marten (Martes foina)
sympatrically inhabit a large area of Europe. However, given our limited knowledge of their bioecological
relationships, their extremely elusive behaviour and the fact that their faeces cannot be distinguished on
the basis of morphology alone, it is very difficult to monitor their populations. In this study, we describe a
reliable non-invasive polymerase chain reaction (PCR)–restriction fragment length polymorphism (PCRRFLP) method for distinguishing between M. martes and M. foina based on the analysis of
deoxyribonucleic acid extracted from faeces samples. The method was specifically designed to avoid
possible interference from potential prey mammals and other sympatric carnivores. The procedure consists
of PCR amplifying a mitochondrial D-loop region followed by digesting the resulting 276-bp-long
amplicons with the restriction enzymes HaeIII and RsaI. To assess the efficiency of this technique, we
conducted a preliminary field study across the potential sympatric distribution areas of both marten
species in the northern Iberian Peninsula. Out of 359 faeces samples collected, we identified 80 as
specimens from the stone marten and 235 from the pine marten. Unequivocal species identification was
thus possible in 88% of the faeces samples collected. These findings reveal the combined use of noninvasive genetic sampling and GIS technology to be a reliable and cost-effective procedure for improving
our knowledge of the spatial distributions of sympatric marten species. This protocol could also be used to
identify and improve information gaps, to develop effective research and management programmes and in
population and landscape genetics studies of marten species.
Keywords Non-invasive genetic sampling, Genetic species identification, Faecal DNA
- 89 -
CHAPTER 3
Species Identification
INTRODUCTION
Accurate species identification is a key step in conservation biology and the basis for any study concerned
with wildlife management and conservation (Frankham et al. 2002). However, obtaining this kind of
information for elusive or cryptic carnivore species can be logistically difficult, particularly if relying
entirely on field signs such as hair, faeces or tracks (e.g. see Piggott and Taylor 2003 for a recent
discussion). These types of signs have been traditionally subjected to a variety of morphological analyses to
establish their species of origin (Kohn and Wayne 1997). Nevertheless, there are situations in which
samples deposited by carnivores of similar body size living in sympatry with the target species preclude any
reliable identification at the species level on the basis of morphology alone (Davison et al. 2002; Birks et
al. 2004). Thus, the misidentification of species from scats is a common event, as has been indicated for
several sympatric carnivore species (Davison et al. 2002; Dalén et al. 2004; Riddle et al. 2003; Pilot et al.
2006).
The European pine marten (Martes martes) and stone marten (Martes foina), similar in terms of their
morphology, feeding and behaviour, live sympatrically across a large area of Europe. M. martes is a
threatened or scarce species in many countries where forest habitat loss and fragmentation are major
threats for conservation. In contrast, the number of stone martens has increased in many countries of their
current distribution range (Proulx et al. 2004).
Traditionally, the presence of martens has been determined through a variety of techniques such as
roadkill or hunting information, live trap and sighting surveys (Messenger and Birks 2000), the use of
track plates (Zielinski and Kucera 1995), camera traps (Zielinski and Kucera 1995), hair-snagging devices
(Messenger and Birks 2000; Lynch et al. 2006) or scat-based surveys (Birks et al. 2004). Each of these
methods has its relative merits or biases depending on the size of the area surveyed, the sampling effort and
the cost and efficiency of the method (Birks et al. 2004). The monitoring of martens inferred from records
of their scats seems to be the most effective for widespread surveys, as the animals are minimally disturbed.
However, a necessary step in this type of survey is the reliable identification of the species from which the
faeces was derived (Riddle et al. 2003; Dalén et al. 2004; Gómez-Moliner et al. 2004). Unfortunately,
these two marten species cannot be reliably identified solely on the basis of faeces morphology (Marchesi
et al. 1989; Pilot et al. 2006) and may be easily mistaken for species such as the red fox (Vulpes vulpes;
Davison et al. 2002), European polecat (Mustela putorius), commom genet (Genneta genneta) and their
related mustelid species stoats (Mustela erminea), American minks (Mustela vison; Birks et al. 2004) and
European minks (Mustela lutreola).
- 90 -
Non-invasive identification Martens
Paper III
Several genetic species identification methods have been recently developed for studies on carnivores based
on noninvasive sampling. These methods include the amplification and direct sequencing of
mitochondrial deoxyribonucleic acid (mtDNA; Höss et al. 1992; Murakami 2002; Davison et al. 2002),
the use of species-specific mtDNA primers (Dalén et al. 2004), microsatellite allele frequencies (Randi and
Lucchini 2002), species-specific allele lengths of a given microsatellite (Domingo-Roura 2002; Kalz et al.
2006; Pilot et al. 2006) or polymerase chain reaction (PCR)–restriction fragment length polymorphism
(RFLP) methods (Hansen and Jacobsen 1999; Riddle et al. 2003; Gómez-Moliner et al. 2004; Vercillo et
al. 2004; Colli et al. 2005). However, little attention has been paid to the reliability and utility of these
laboratory techniques for non-invasive field studies or conservation programmes (Schwartz et al. 1998;
Taberlet et al. 1999; Piggott and Taylor 2003; Kalz et al. 2006). Moreover, most of these reports only
describe and present the particular molecular technique. There is therefore a need for studies assessing the
field use of DNA techniques in species surveys to confirm their validity for widespread use in monitoring
or conservation programmes (e.g. Mowat and Paetkau 2002; Pilot et al. 2006).
Recent molecular methods described for the genetic identification of M. martes and M. foina have been
based on microsatellite analysis (Domingo-Roura 2002; Pilot et al. 2006) or PCR-RFLP of the Cyt-b gene
(Vercillo et al. 2004; Colli et al. 2005). Pilot et al. (2006) proposed a method based on the simultaneous
use of the microsatellites Mel 10 (Domingo-Roura 2002) and Ma 18 that is suitable for non-invasive
samples and takes into account the two main sources of error in non-invasively collected material (allelic
dropout and false alleles; Taberlet et al. 1996; Broquet and Petit 2004). However, the lower amplification
rate and need for a higher number of replicates of nuclear DNA (nDNA) compared to mtDNA is a
significant drawback of the broad-scale application of this method to large numbers of faecal samples.
In contrast, the PCR-RFLP methods developed by Vercillo et al. (2004) and Colli et al. (2005) have been
mainly used on tissue samples, their feasibility for use on faeces has not yet been analyzed and their utility
has not been tested in a real field study. We considered these techniques, but both require the use of
primers known to work on a wide variety of vertebrates (Kocher et al. 1989). Given that the DNA of nontarget species such as prey or other carnivores is often amplified during the identification procedure
(personal observation), the use of these methods on faeces samples can compromise the quality of the data
provided (Piggott and Taylor 2003).
Thus, the aim of this study was to develop a reliable and unambiguous non-invasive PCR-RFLP method
for distinguishing between the scats of M. martes and M. foina through the analysis of mtDNA. We
intended to avoid the interference from potential mammalian prey species or other carnivore species
showing similar scat morphology because this is a common source of error when using field samples. Our
- 91 -
CHAPTER 3
Species Identification
analysis focused on the mtDNA control region because of its rapid evolution rate, generating different
haplotypes even among closely related species and populations (Tamura 2000), and because it relies upon
the high copy number of mtDNA. Finally, to assess the field suitability of the technique, we conducted a
preliminary study across the potential distribution area of both marten species in the northern Iberian
Peninsula. Our findings revealed the presence and range of the two marten species and suggest the
efficiency of the technique for use on large sample numbers.
MATERIALS AND METHODS
Design of PCR-RFLP assay
Sampling and DNA extraction
Fresh tissue and hair specimens from roadkill animals were used to isolate DNA from the target marten
species, M. martes and M. foina. Marten specimens were collected mainly from several regions of the
northern Iberian Peninsula but were also obtained from other European countries (France, Luxembourg,
Estonia, Finland, Belarus and Russia). DNA was also isolated from several prey species and other
sympatric carnivore species showing a similar scat morphology that could easily be misidentified as
martens. The species, sample numbers and sample types are provided in Supplement 1. DNA was isolated
from tissues and hairs using the Qiagen DNeasy ® Tissue DNA extraction kit according to the
manufacturer’s instructions.
Primer design
To design primers for the Martes species, we used the CLUSTAL X programme (Thompson et al. 1997)
to check and align the mitochondrial D-loop sequences available from GenBank for the target marten
species, other carnivore species, prey species and humans (Homo sapiens). Supplement 2 provides the
GenBank accession numbers for the control region nucleotide sequences for all the species included in the
alignment. Inter-specific D-loop sequence differences were targeted as potential marten-specific primer
sites. We designed a forward primer Mm_L1 (5 - CCCAAAGCTGACATTCTAAC-3 adapted for the
Martes genus from the L1607 primer of Davison et al. 1999) and a reverse primer designed specifically for
this study, Mm_H1 (5 -ATGGGCCCGGAGCGAGAAGAGGTACAC-3 ). The primers were designed
to amplify a short fragment of 276 bp to maximize the probability of amplifying degraded DNA as
expected in faecal samples.
- 92 -
Non-invasive identification Martens
Paper III
To verify the specificity of the primers designed for the two target species, we searched for primer binding
sites using Amplify 3.1 software on all the sequences listed in Supplement 2. Their specificity was also
evaluated using the primers on DNA extracted from fresh tissue and hair samples for all the species listed
in Supplement 1.
PCR amplification of the D-loop region
A 2-μL volume of the DNA extraction mixture was added to 23 μL of the PCR mixture containing 0.5 μL
of forward primer Mm_L1 and 0.5 μL of reverse primer Mm_H1 (20 pmol/μL), 2.5 μL 10× reaction
buffer, 0.75 μL MgCl2
(50 mM), 1.4 μL deoxynucleotide triphosphates (2.5 mM), 1 μL of bovine
serum albumin (10 mg/mL), 16.15 μL of sterile water and 0.2 μL Taq polymerase (5 U/μL). After
incubation for 5 min at 94°C, the samples were subjected to 35 amplification cycles in a BIO-RAD
iCycler (Version 3.021, BIO-RAD Laboratories) consisting of denaturation for 1 min at 94°C, annealing
for 1 min at 63.5°C and a final extension stage of 1 min at 72°C. Annealing temperature was selected after
temperature gradient PCR on tissuederived DNA. Four microlitres of the final amplified product was
analysed by electrophoresis on 2% agarose gel. Negative controls were used to check for contamination.
DNA sequencing
PCR products were purified using the QIAquick® Tissue PCR Purification kit according to the
manufacture’s instructions and sequenced using the dRhodamine Terminator Cycle Sequencing Ready
Reaction Kit (Applied Biosystems) in an ABI PRISM Model 310 Genetic Analyzer (Applied Biosystems).
The new nucleotide sequences reported in this paper were deposited in GenBank under accession numbers
EF200700 (M. martes), EF200701 and EF200702 (M. foina).
Restriction enzyme analysis
The CLUSTAL X programme (Thompson et al. 1997) was used to check and to align the mitochondrial
D-loop sequences and their different haplotypes for the two target marten species and for the carnivore
species whose DNA sequences were also amplified by the primer pairs developed (M. putorius, M. lutreola,
M. erminea and M. vison; Supplement 2). The sable (Martes zibellina) was also included, as its distribution
range overlaps that of M. martes around the Ural Mountains in central Russia, where hybridisation
between the two species is common (Grakov 1994; Helldin 1998; Davison et al. 2001).
The red fox and common genet were not included in the RFLP test because their DNA was not amplified
by the primers developed.
- 93 -
CHAPTER 3
Species Identification
Polymorphic sites were determined using In silico software (Bikandi et al. 2004). The restriction enzymes
RsaI and HaeIII were identified as those generating different RFLP patterns for each species (Fig. 1). The
robustness of the assay was tested using the selected enzymes to digest PCR products obtained from all the
species listed in Supplement 1.
DNA digestions by endonucleases were run in 14-μL volumes, consisting of 1.4 μL of the appropriate 10×
reaction buffer (supplied by manufacturer with the respective enzyme), 0.2–0.4 μL of restriction enzyme
solution (10 U/μL), 5 μL of the PCR product and the remaining volume of pure water. Incubations were
performed for 6 h at 37°C, followed by a 3% agarose gel electrophoresis of 10 μL of the digestion
products. Ethidium bromide-stained bands were visualized and photographed using an UVIdoc LCD
system (UVITEC).
Fig. 1 Aligned sequences of the amplified mitochondrial D-loop fragment in Martes martes and Martes foina. The
primer sequences are shown in bold. The restriction sites for HaeIII (recognition sequence GGCC) and RsaI
(recognition sequence GTAC) are underlined and in bold. Restriction patterns (Table 1) are indicated in brackets for
each species.
Use of the PCR-RFLP assay on faecal samples: preliminary field study of the genus Martes in the
northern Iberian Peninsula
The faeces identification technique was tested in a study performed across the current sympatric
distribution range of the two Martes species in the northern Iberian Peninsula (Proulx et al. 2004). From
September 2002 to September 2006, we collected 359 faeces samples initially identified as Martes sp.
based on scat morphology. Sampling was conducted along linear features, such as forest trails and paths,
for which pine martens show a preference. Universal Transversal Mercator (UTM) coordinates were
recorded for all the samples collected using a global positioning system (Garmin eTtrex). The faecal
samples were stored in autoclaved tubes containing ethanol 96% and frozen at −20°C until processed.
- 94 -
Non-invasive identification Martens
Paper III
The faecal DNA extraction procedure was based on the protocol described by Gómez-Moliner et al.
(2004). Extraction blanks were included to check for crosscontamination and were processed as samples in
the subsequent amplifications. DNA extractions were carried out in a specialized laboratory to avoid
contaminating DNA.
The PCR-RFLP assay described above for tissue/hair samples was used with the necessary modifications
on faecal samples. These modifications comprised of increasing DNA extraction solution from 2 to 4 μL
and inclusion of positive controls obtained from tissue/hair samples of known M. martes and M. foina
specimens in all processes. The PCR protocol was also slightly modified to a touchdown PCR procedure.
This involved eight extra cycles after the first denaturation step at 94°C for 2 min of 1 min at 95°C, 40 s at
72°C (reducing the temperature 0.5°C per cycle) and 45 s at 72°C to enhance the enrichment of the
specific product over any non-specific one. Finally, the UTM coordinates corresponding to each sample
were projected onto a GIS (Arcview 9.0. ESRI) along with the species identification data provided by the
PCR-RFLP assay.
RESULTS
PCR-RFLP assay
Primer design and PCR amplification
The accuracy of the newly designed D-loop primers Mm_L1 and Mm_H1 was tested on sequences for the
two target species (Supplement 2.1) using the Amplify 3.1 programme and on sequences obtained from
the tissue/hair specimens from the 50 M. martes and 32 M. foina roadkill individuals collected from
different regions of the martens’ current European distribution area (Supplement 1.1). In all virtual
analyses (Amplify v.3.1), the primer pair Mm_L1/ Mm_H1 consistently generated the expected amplicons
of 276 bp length, which it subsequently also did in reality in all tissue/hair DNA samples tested (Fig. 1).
In contrast, when Amplify 3.1 was used on the other carnivore species sequences (Supplement 2.2), no
virtual amplification products were obtained except for M. putorius, M. lutreola, M. erminea and M. vison
(Table 1). Moreover, of the potential prey species tested (Supplement 2.3), only Sorex araneus rendered an
amplification product of 900 bp using Amplify 3.1. These theoretical results were checked by running
assays on DNA extracted from tissue and hair specimens of all the carnivore and prey species (Supplement
1). No amplification products were obtained for any of the species, except the four Mustela species and
Sorex coronatus. This crosscheck confirmed the high specificity of the markers developed for Martes and
- 95 -
CHAPTER 3
Species Identification
Mustela species, whose scats are similar in size and shape. The primers failed to amplify DNA
corresponding to the potential marten prey species examined with the exception of the Sorex spp. that
rendered a 900-bp amplicon. Notwithstanding, this 900-bp amplicon was not produced when working
with degraded DNA, such that we could rule out any interference effects of prey species when the PCR
assay was used on faeces samples. No human DNA amplicons were generated by the PCR assay
(Supplements 1.4, 2.4)
Table 1 Amplicon lengths in base pairs and diagnostic restriction enzyme patterns generated using HaeIII and RsaI
Fragment sizes that could be visually identified on agarose gel for each enzyme are shown in italics. Fragments italicized and set in
bold could not always be identified on agarose gel. a The red fox and common genet were not included because their DNA was
not amplified by the PCR method.
Restriction fragment length polymorphism
Through a detailed comparison of the restriction maps of the sequences obtained here and those from
GenBank (Supplement 2), the position of cleavage sites for the restriction endonucleases HaeIII and RsaI
were found to be suitable for differentiating M. martes from M. foina (Fig. 1), from the other carnivore
species whose DNA was amplified by the selected primers (M. putorius, M. lutreola, M. erminea and M.
vison) and from M. zibellina (Table 1). M. martes displayed a 220- and a 51-bp fragment (restriction
pattern A) after digestion with HaeIII (the remaining 5-bp-long band was not visible on agarose gel).
Further, M.martes rendered two restriction patterns after mitochondrial D-loop digestion with RsaI
because of a base substitution (T–C) in the cleavage site of the enzyme (Fig. 1). Digestion pattern A
- 96 -
Non-invasive identification Martens
Paper III
included a thick electrophoretic band corresponding to 97- and 94-bp-long fragments and a second 62bp-long band. Digestion pattern B was a thick electrophoretic band corresponding to the 97- and 94- bplong fragments and a second band of 42 bp length. M. foina showed a 271-bp-long fragment (restriction
pattern B; the remaining 5-bp-long band was not visible on agarose gel) after digestion with HaeIII. RsaI
digestion produced five fragments of 94, 82, 62, 23 and 15 bp length appearing as three electrophoresis
bands (restriction pattern C). All the RFLP patterns are shown in Fig. 2.
Thus, HaeIII enzyme digestion served to differentiate the two species, but to avoid false identifications and
obtain species-specific restriction patterns, the simultaneous use of RsaI was required.
Unambiguous interpretation of the results could be achieved visually without the need of computer
analysis. Using this technique, two haplotypes can be discriminated for M. martes (AA, AB) and one for
M. foina (BC; Fig. 2). Restriction digestion patterns that could not be ascribed to any of the species
investigated were not observed. Moreover, other European mustelids producing morphologically similar
faeces showed different RFLP haplotypes (Table 1). In 115 out of 119 samples analysed (Supplement 1;
M. martes, n=52; M. foina, n=32; M. putorius, n=10; M. lutreola, n=10; M. erminea, n=5; M. vison,
n=10), our RFLP analysis yielded results that matched the theoretical RFLP patterns obtained using In
silico software (Bikandi et al. 2004). One of the exceptions corresponded to a specimen erroneously
identified as stone marten (M. foina), which showed pine marten (M. martes) restriction patterns. On the
other hand, one specimen labelled as pine marten from Finland was identified as M. zibellina by our
PCRRFLP method.
Fig. 2 Photograph of a 3% agarose gel showing ethidiumbromide stained bands and diagnostic restriction enzyme
patterns (AB; AA; BC) generated using HaeIII and RsaI for Martes martes (Mm) and Martes foina (Mf )
- 97 -
CHAPTER 3
Species Identification
Use of PCR-RFLP on faecal samples: a preliminary field study of the Martes genus in the northern
Iberian Peninsula
Out of 359 faecal samples collected from the entire sympatric distribution range of both marten species in
the northern Iberian Peninsula (Proulx et al. 2004), 316 were classified as one of the target species by our
PCR-RFLP method. Thus, unequivocal species identification was possible in 88% of the samples. This
represents a high species identification rate for the technique used on faecal samples. In the remaining
12%, the DNA extracted was not amplified by the primers used.
The geographical locations for the 316 correctly identified faecal samples in the Iberian Peninsula are
shown in Fig. 3a. We effectively identified 80 faecal samples as stone marten and 235 as pine marten. One
sample was identified as the European polecat, indicating visual confusion of the scat in the field survey.
While in some forested sampling areas, the prevalence of the pine marten was clear (sampling areas 1, 2, 3
and 6), in other forested areas, the ranges of both marten species fully overlapped (sampling areas 4 and 5).
The overlapping distribution ranges of the martens in sampling area 4 are depicted in Fig. 3b. This area
has been selected because of the high number of samples of both Marten species (N=190) analysed.
Fig. 3 a GIS determined geographical locations and number of samples ascribed to the target species after applying
the PCR-RFLP method to faecal samples collected in the northern Iberian Peninsula. Sampling areas are designated
by successive numbers (1–6). b Detail of the overlapping distribution of Martes martes and Martes foina within
sampling area 4. Forest and urban areas appear shaded in grey and black, respectively. Black stars and circles
represent the geographical locations of faecal samples identified as Martes martes and Martes foina by the proposed
PCR-RFLP method, respectively
- 98 -
Non-invasive identification Martens
Paper III
DISCUSSION
PCR-RFLP method
The use of the technique developed here on all published partial D-loop sequence haplotypes and tissue
samples of the target martens and other carnivores from several regions of Europe included in our study
revealed the method to be highly robust. Our assay was also highly reliable, as positive results were
obtained only for Martes and Mustela species. Other sympatric carnivore species, in particular the red fox
and common genet, whose scats can be easily mistaken for those of martens, rendered no amplicons. A
wide screen for potential mammal preys of the martens of interest gave no amplicons except for Sorex spp.
However, Sorex spp. amplicons were very different in size (900 bp) compared to those yielded by the
mustelids, thus precluding misidentification. Moreover, a large 900-bp fragment is unlikely to provide an
intact template for amplification (Kohn and Wayne 1997; Broquet et al. 2006). Thus, interference
because of the presence in faeces samples of genetic material from species consumed by the martens can be
in principle ruled out.
When we tested the method in faecal samples collected in the field survey, a high species identification rate
was obtained (88%). The remaining 12% of samples did not amplify. The possibility that some samples
had been incorrectly identified in the field as marten scats would mean an even higher correct
identification rate. Particularly, confusing fox as marten scats is a common error, as indicated by Davison
et al. (2002).
One of the advantages of our protocol is that it relies on the high copy number of mtDNA for improved
species resolution in low-quantity, low-quality samples. Further, the use of multiple restriction sites
reduces the risk of misidentification. In addition, a short DNA fragment (276 bp) is more likely to provide
an intact template for amplification (Kohn and Wayne 1997; Broquet et al. 2006) when working with
degraded DNA and can be effective even in old and/or rain-washed samples.
Some of the restriction patterns yielded by the method were found to be shared by M. vison, M. lutreola
and M. putorius. Thus, previously described specific methods (Gómez-Moliner et al. 2004) should be used
to distinguish between these species. Moreover, the discrimination of all the mustelid species treated in the
present paper might already become possible using RsaI digestion alone with polyacrylamide gels, which
would allow visualizing DNA fragments differing in a few bases length. However, the application of
agarose gels facilitates the use of this method even in basic field laboratories.
- 99 -
CHAPTER 3
Species Identification
Comparison with other methods
The PCR-RFLP method described in this paper was especially designed for non-invasively collected
samples and specifically for faecal material. With this purpose in mind, we were able to reliably distinguish
other carnivore species whose scats could be confused as marten scats. Besides, interference because of prey
DNA was avoided through the use of mustelid-specific primers. Similar PCR-RFLP methods based on
Cyt-b (Vercillo et al. 2004; Colli et al. 2005) have been mainly developed for tissue samples, and their
feasibility or efficiency for use on faecal DNA has not been tested. Notwithstanding, given (1) that the
primers used by both sets of authors are generic for vertebrates (Kocher et al. 1989; personal observation),
(2) that both protocols give rise to similar diagnostic fragments using faecal DNA from other carnivores
(Pilot et al. 2006; personal observation) and (3) that potential prey DNA (e.g. from Cletryonomys glareolus,
Microtus agrestis, Mus musculus) will be amplified and interfere with the restriction patterns for the martens
(personal observation), these methods are unlikely to be useful for identifying martens in scat surveys. The
approach recently developed by Pilot et al. (2006) for the genetic identification of both marten species by
microsatellite analyses has the main advantage that the costs of species identification would be entirely
absorbed in the pursuit of intraspecific genetic analysis by genotyping of faeces. Nevertheless, the success
rate of the PCR-RFLP method proposed based on the mitochondrial D-loop region is higher than that
based on microsatellites: 88 vs 53.4%. This can be explained by the higher copy numbers of mtDNA
compared to nDNA, conferring a greater success rate to the method for screening large numbers of
samples across broad geographical areas. The amplification of nDNA requires two or even three replicates
to obtain a valid result when working with degraded DNA (Taberlet et al. 1996; Miller et al. 2002;
Broquet and Petit 2004). This would mean an increased time and cost expenditure for the nDNA analysis
compared to our PCR-RFLP technique.
Preliminary field study of the genus Martes in the northern Iberian Peninsula
The development of non-invasive DNA techniques for species identification has recently offered
researchers the possibility to design reliable programmes for monitoring elusive species as an alternative to
the use of traditional sources of distribution data.
In the case of the sympatric marten species examined here, there is a clear need to monitor their presence
and distribution across their distribution ranges (Proulx et al. 2004). These two species show inverse
demographic trends, a different population status and probably compete for the same resources. Thus,
their management needs are likely to differ. Moreover, given our limited knowledge of their bioecological
- 100 -
Non-invasive identification Martens
Paper III
relationships, their extremely elusive behaviour and the fact that their faeces are morphologically
indistinguishable, it is not easy to monitor populations of these mustelids.
Most scat surveys on the pine marten conducted in its sympatric distribution area with the stone marten
have assumed that stone martens rarely penetrate forests (e.g. Delibes 1983; Russell and Storch 2004).
This assumption, however, is clearly an oversimplification of the complex sympatric relationship between
the two species and can lead to false interpretations of their distribution patterns (Pilot et al. 2006).
The results obtained in the present preliminary field study indicate that sympatric relationships differ
depending on the habitat type, forest structure, as well as on landscape effects of forest fragmentation and
human disturbance of the surrounding habitat. Despite the fact that, as a general rule, the pine marten has
a clear dominance over the stone marten in late-successional forests (Fig 3a sampling areas 1, 2, 3 and 6)
and displaces the stone marten to surrounding areas, both species are known to co-exist in many forested
areas (Fig 3a sampling areas 4 and 5). Our results suggest that the increasing effects of forest fragmentation
and current inverse demographic trend of the two species have led to a high complexity of their
bioecological relationships. These factors have determined that the co-existence of both species has become
common on forest patches within mosaic habitats (Fig. 3b).
The results provided by the combined use of non invasive genetic analysis of faecal samples and GIS
technology tested in this preliminary field study suggest that our protocol would serve to assess the effects
of human activities on distribution patterns, identify and resolve information gaps and design effective
research and management programmes.
The information obtained in our field study requires confirmation through an extended sampling effort.
However, it is a first step towards improving our knowledge of distribution patterns of sympatric marten
species, at both broad and small scales in the northern Iberian Peninsula.
In conclusion, our results indicate that the PCR-RFLP method proposed for use on non-invasive samples
(faeces or remotely plucked hair) is a reliable, efficient, time-saving and cost-effective procedure for
improving our knowledge of the spatial distributions of sympatric marten species. This method has
applications in studies focusing on the population genetics, phylogeographic variation and landscape
genetics of marten species.
- 101 -
CHAPTER 3
Species Identification
ACKNOWLEDGEMENTS
This study was funded by the Biodiversity Section, Dept. of Territorial Planning and Environment of the Basque
Government. A. Ruiz-González holds a Ph.D. fellowship awarded by the Dept. of Education Universities and
Research (Basque Government). The authors wish to thank the following persons and institutions for supplying
tissue and faecal samples: the technical staff of the National Parks of Ordesa and Monte Perdido (E. Villagrasa), Picos
de Europa (A. Mora) and Aigüestortes i Estany de Saint Mauricy (J. Canut); J.Herr (Department of Biology and
Environmental Science, University of Sussex.), Dr. A. Abramov (Zoological Institute, Russian Academy of Sciences),
A. Fernandez (Doñana Biological Station-University of Oviedo), Dr. I. Barja and his research group (UAM), F.
López-Giraldez (DNA and Tissue Collection, Pompeu Fabra University), P. Aymerich (PN Alt Pirineu), J.M.
Fernandez (IAN), G. Belamendia (CEA–MCN Álava), P. Lizarraga and L. Lorza (CRF Martioda-DFA), J. Pinedo
(DFA), I. Amigo (DFV), H. Aguirre, P. Pérez and G. Dominguez. We are also indebted to Mikel Gurrutxaga (Dept.
of Natural Environment and Geographical Information System, IKT) for his help with the GIS treatment of data
and preparing maps.
REFERENCES
Adams JR, Kelly BT, Waits P (2003) Using faecal DNA sampling and GIS to monitor hybridization between Red
Wolves (Canis rufus) and Coyotes (Canis latrans) Mol Ecol 10:2175-2186
Bikandi J, San Millán R, Rementeria A, Garaizar J (2004) In silico analysis of complete bacterial genomes: PCR,
AFLP-PCR, and endonuclease restriction Bioinformatics 20:798-799
Birks JDS, Messenger JE, Braithwaite TC, Davison A, Brookes RC, Strachan C (2004) Are scat surveys a reliable
method for assessing distribution and population status of pine martens? In: Harrison DJ, Fuller AK,
Proulx G (eds) Martens and fishers (Martes) in human-altered environments: an international perspective.
Springer-Verlag, New York: pp 235–252
Broquet T, Ménard N, Petit E (2006) Noninvasive population genetics: a review of sample source, diet, fragment
length and microsatellite motif effects on amplification success and genotyping error rates. Conserv Genet
8:249-260
Broquet T, Petit E (2004) Quantifying genotyping errors in noninvasive population genetics. Mol Ecol 13:3601–
3608
Colli L, Cannas R, Deiana AM, Gandolfi G, Tagliavini J (2005) Identification of mustelids (Carnivora: Mustelidae)
by mitochondrial DNA markers. Mamm Biol 6:384-389
Dalén L, Götherström A, Angerbjörn A (2004) Identifying species from pieces of faeces. Conserv Genet 5:1–3
Davison A, Birks JDS, Brookes RC, Braithwaite TC, Messenger JE (2002) On the origin of faeces: morphological
versus molecular methods for surveying rare carnivores from their scats. J Zool (Lond) 257:141–143
Davison A, Birks JDS, Brookes RC, Messenger JE, Griffiths HI (2001) Mitochondrial phylogeography and
population history of pine martens Martes martes compared with polecats Mustela putorius. Mol Ecol
10:2341-2347
Davison A, Birks JDS, Griffiths HI, Kitchener AC, Bigginns D, Butlin RK (1999) Hibridization and the
phylogenetic relationships between polecats and domestic ferrets in Britain. Biol Conserv 87:155-161
Delibes M (1983) Interspecific competition and the habitat of the stone marten Martes foina (Erxleben, 1777) in
Europe. Acta Zool Fenn 174:229–231
Domingo-Roura X (2002) Genetic distinction of marten species by fixation of a microsatellite region. J Mammal
83:907–912
- 102 -
Non-invasive identification Martens
Paper III
Ernest HB, Penedo MCT, May BP, Syvanen M, Boyce WM (2000) Molecular tracking of mountain lions in the
Yosemite Valley region in California: genetic analysis using microsatellites and faecal DNA. Mol Ecol
9:433–441
Farrel LE, Romant J, Sunquist ME (2000) Dietary separation of sympatric carnivores identified by molecular analysis
of scats. Mol Ecol 9:1583–1590
Frankham R, Ballou JD, Briscoe DA (2002) Introduction to Conservation Genetics. Cambridge University Press
Cambridge, UK
Gómez-Moliner BJ, Cabria MT, Rubines J, Garin I, Madeira MJ, Elajalde A, Aihartza J, Fournier P, Palazón S
(2004) PCR-RFLP identification of mustelid species: European mink (Mustela lutereola), American mink
(Mustela vison) and polecat (Mustela putorius) by analysis of excremental DNA. J Zool (Lond) 262:311–316
Grakov NN (1994) Kidus-a hybrid of the sable and the pine marten. Lutreola, 1:1-4
Hansen MM, Jacobsen L (1999) Identification of mustelid species: otter (Lutra lutra), American mink (Mustela
vison) and polecat (Mustela putorius), by analysis of DNA from faecal samples. J Zool (Lond) 247:177–181
Helldin JO (1998) Pine marten (Martes martes) population limitation. Dissertation, Swedish University, Agricultural
Sciences, Uppsala, Sweeden
Höss M, Kohn M, Pääbo S, Knauer F, Scgroder W (1992) Excremental analysis by PCR Nature 359:199
Kocher T D, Thomas W K, Meyer A, Edwards SV, Pääbo S, Villablanca FX, Wilson A C (1989) Dynamics of
mitochondrial DNA evolution in animals: amplification and sequencing with conserved primers. Proc Natl
Acad Sci USA 86:6196–6200
Kohn MH, Wayne RK (1997) Facts from feces revisited. TREE 12:223–227
Kurose N, Masuda R, Tatara M (2005) Fecal DNA analysis for identifying species and sex of sympatric carnivores: A
noninvasive method for conservation on the Tsushima Islands, Japan. J Hered 96:688-697
Lynch ÁB, Brown MJF, Rochford JM (2006) Fur snagging as a method of evaluating the presence and abundance of
a small carnivore, the pine marten (Martes martes). J Zool (Lond) 270:330–339
Marchesi P, Lachat N, Lienhard R, Debieve Ph, Mermod C (1989) Comparaison des régimes alimentaires de la
fuine (Martes foina Erxl) et de la martre (Martes martes L) dans une région du Jura suisse. Rev Suisse Zool
96 127–146
Messenger JE, Birks JDS (2000) Monitoring the very rare pine marten populations in England and Wales. In:
Griffiths HI (ed) Mustelids in a modern world: Management and conservation aspects of small carnivore:
human interactions. Backhuys Publishers, Leiden: pp. 153-162
Mills LS, Pilgrim KL, Schwartz MK, McKelvey KS (2000) Identifying lynx and other North American felids based
on mtDNA analysis. Conserv Genet 1:285–288
Mowat G, Paetkau D (2002) Estimating marten Martes americana population size using hair capture and genetic
tagging. Wildl Biol 8:201–209
Murakami T (2002) Species identification of mustelids by comparing partial sequences on mitochondrial DNA from
fecal samples. J Vet Med Sci 64:321–323
Palomares F, Godoy JA, Piriz A, O’Brien J, Johnson WE (2002) Faecal genetic analysis to determine the presence
and distribution of elusive carnivores: design and feasibility for the Iberian lynx. Mol Ecol 11:2171–2182
Paxinos E, Mcintosh C, Ralls K, Fleischer R (1997) A non-invasive method for distinguishing among canid species:
amplification and enzyme restriction of DNA. Mol Ecol 6:483-486
Piggott MP, Taylor AC (2003) Remote collection of animal DNA and its applications in conservation management
and understanding the population biology of rare and cryptic species. Wildl Res 30:1-13
Pilot M, Gralak B, Goszczyński J & Posłuszny M (2006) A method of genetic identification of pine marten (Martes
martes) and stone marten (Martes foina) and its application to faecal samples. J Zool (Lond) 271:140-147
Proulx G, Aubry KB, Birks J, Buskirk SW, Fortin C, Frost HC, Krohn WB, Mayo L, Monakhov V, Payer D, Saeki
M, Santos-Reis M, Weir R, Zielinski WJ (2004) World distribution and status of the genus Martes in 2000
In: Harrison DJ, Fuller AK, Proulx G (eds) Martens and fishers (Martes) in human-altered environments:
an international perspective. Springer-Verlag, New York: pp 21–76
Randi E, Lucchini V (2002) Detecting rare introgression of domestic dog genes into wild wolf (Canis lupus)
populations by Bayesian admixture analyses of microsatellite variation. Conserv Genet 3:31–45
Reed JZ, Tollit DJ, Thompson PM, Amos W (1997) Molecular scatology: the use of molecular genetic analysis to
assign species, sex and individual identity to seal faeces. Mol Ecol 6:225–234
- 103 -
CHAPTER 3
Species Identification
Riddle AE, Pilgrim KL, Mills LS, McKelvey KS, Ruggiero LF (2003) Identification of mustelids using mitochondrial
DNA and non-invasive sampling. Conserv Genet 4:241–243
Russell AJM, Storch I (2004) Summer food of sympatric red fox and pine marten in the German Alps. Eur J Wildl
Res 50:53–58
Schwartz MK, Tallmon DA, Luikart G (1998) Review of DNA based census and effective population size estimators.
Anim Conserv 1:293–299
Sugimoto T, Nagata JV, Aramilev V, Belozor A, Higashi S, McCullough DR (2006) Species and sex identification
from faecal samples of sympatric carnivores Amur leopard and Siberian tiger, in the Russian Far East.
Conserv Genet 7:799-802
Taberlet P, Griffin S, Goossens B, Questiau S, Manceau V, Escaravage N, Waits LP, Bouvet J (1996) Reliable
genotyping of samples with very low DNA quantities using PCR. Nucleic Acids Res 24:3189–3194
Taberlet P, Waits LP, Luikart G (1999) Noninvasive genetic sampling: look before you leap. TREE 14:323–327
Tamura K (2000) On the estimation of the rate of nucleotide substitution for the control region of human
mitochondrial DNA. Gene 259:189-197
Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG (1997) The ClustalX windows interface: flexible
strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res 24:4876-4882
Vercillo F, Lucentini L, Mucci N, Ragni B, Randi E, Panara F (2004) A simple and rapid PCR-RFLP method to
distinguishing Martes martes and Martes foina. Conserv Genet 5:869–871
Zielinski WJ, Kucera TE (1995) American marten, fisher, lynx, and wolverine: survey methods for their detection.
General Technical Report PSW-157 US Department of Agriculture Forest Service, Pacific Southwest
Research Station, Berkeley, California, USA
- 104 -
Non-invasive identification Martens
Paper III
SUPPLEMENTARY MATERIAL
S1 Hair/Tissue samples analyzed a
1. Marten target species
Martes martes (European pine marten; n:50, h:22-t:28); Martes foina (stone marten; n:32, h:11-t:21).
2. Carnivore species showing a similar scat morphology that could easily be misidentified as martens
Vulpes vulpes (red fox; n:10, h:2-t:8) Mustela putorius (European polecat; n:10, h:1-t:9); Genneta genneta (commom
genneta; n:10 h:8-t:2); Mustela erminea (stoat; n:5, h:5); Mustela vison (American mink; n:10, h:1- t:9) Mustela
lutreola (European mink; n:10, h:8-t:2).
3. Potential mammalian prey species
Apodemus flavicolis (yellow-necked field mouse; n:3, t:3); Apodemus sylvaticus (European woodmouse; n:4, t:4);
Crocidura russula (white-toothed shrew; n:4; t:4); Mus musculus (house mouse; n:4, t:4); Microtus agrestis (short-tailed
field vole; n:4, t:4); Microtus arvalis (common vole; n:4, t:4); Sorex coronatus (crowned shrew; n:5, t:5); Suncus
etruscus (white-toothed pygmy shrew; n:3, t:3); Neomys fodiens (Eurasian water shrew; n:5, t:5); Clethryonomis
glareolus (bank vole, n:4, t:2-h:2); Glis glis (fat dormouse; n:3, h:3); Sciurus vulgaris (Eurasian red squirrel; n:4, h:1t:3); Elyomis quercinus (garden dormouse; n:2, h:2); Talpa europaea (European mole; n:4, t:4); Erinaceus europaeus
(western European hedgehog; n:5, t:5); Cervus elaphus (red deer; n:5, t:5), Capreolus capreolus (western roe deer; n:5,
t:5 ); Sus scrofa (wild boar; n:5, t:5).
4. Human
Homo sapiens (Human; n:4, h:4)
a
n: number of samples analysed; t: tissue sample; h: hair sample
S2 GenBank accession numbers of D-loop sequences analyzed
Newly obtained sequences of Martes martes and Martes foina were analysed together with sequences from the
following mammalian species for the PCR-RFLP design. The partial sequences that did not fully incorporate the
fragment analysed were included with complete sequences to check for possible changes in the D-loop sequences.
1. Marten Species
Martes martes (European pine marten; AF336949-AF336964, AF336968, AF336969 AJ585357); Martes foina (stone
marten; AF336973, EF200701, EF200702); Martes zibellina (Sable; AF336965-AF336967, AF336970, EF200700).
2. Carnivore species showing similar scat morphology a
Vulpes vulpes (red fox; AJ585358, AF487753, AF487752, AF487736-AF487746, AF338789, AF338802-AF338790
D83639, AF098155, AM181037); Mustela putorius (European polecat; AY962022-AY962045, AF207718,
AF207726, AF068560, AF068566-AF068568, AF068570); Mustela erminea (stoat; AB061215, AB049782AB049788, AB049777-AB049780; AJ585353-AJ585356, AB006729-AB006733); Mustela vison (American mink;
AJ585352-AJ585350, AB052720); Mustela lutreola (European mink; AF207720-AF207725, AJ548803-AJ548814).
3. Potential mammalian prey species
Sciurus vulgaris (Eurasian red squirrel; AJ238588, AB249881, NC_002369, AF111001-AF111026); Apodemus
sylvaticus (European woodmouse; AY588255, AY588252, AY588264, AY588253, AJ410624, AJ410626AJ410628,); Apodemus flavicolis (yellow-necked field mouse; AY588253, AY588264, AJ410624-AJ410631); Mus
musculus (house mouse; NC_005089, AB042809, AB049357, AB042523, AB042524); Mus spretus (western wild
- 105 -
CHAPTER 3
Species Identification
mouse; AF287305, U47539); Rattus norvegicus (Norway rat; X57080, X57081, DQ897633-DQ897638,
DQ673907-DQ673917, AJ428514, NC_001665, AC_000022, X14848, AY172581, X52757, X04733, X04734);
Rattus rattus (black rat; AB211039, DQ009781-DQ009794, X04735); Microtus agrestis (short-tailed field vole;
F267270, AY948542) Microtus rosameridionalis (southern vole; DQ323955, NC_008064, DQ015676); Microtus
liechtensteini (AF267260-AF267266, AF267278-AF267283); Microtus subterraneus (European pine vole; AF267246,
AF267247, AF267250, AF267271- AF267275); Microtus oeconomus (root vole; AF267268, AF267269); Microtus
arvalis (common vole; AJ009883, DQ121418, AY708394-AY708459, DQ181947-DQ182012, AF267285)
Microtus nivalis (snow vole; AF267284); Clethrionomys glareolus (bank vole; AJ236833, Y07543, AY185801AY185810, AY133323-AY133326, AY133303-AY133315, AF165533) Erinaceus europaeus (western European
hedgehog; NC002080, X88898, AF379703-AF379749); Talpa europaea (European mole; Y19192); Galemys
pyrenaicus (Iberian desman; DQ305040); Crocidura russula (white-toothed shrew; NC006893, AY918341AY918368, AY769264 AY769263, AF343009-AF343017, X90952); Sorex araneus (European shrew; X90951,
AY781981-AY781060, AH014450-AH0144849); Oryctolagus cuniculus (European rabbit; AJ001588, AJ535817AJ535822, AJ535791-AJ535792, AJ535785, AJ535784, AJ563720-AJ563721, AF534080-AJ563710, AJ293843AJ293832, AJ293837, AJ293839, Z83356-Z83363, Z83351, Z83347, Z83345, Z83367, Z83353, Z83355,
Z83349, Z83340-Z83344, AF157459, U62924-U62927, AJ535793-AJ535816, AJ535786-AJ535790, AJ535786AJ535788, AJ563722, AJ563711-AJ563719, AJ293838-AJ293840, AJ293841-AJ293842, AJ293836-AJ293831,
Z83364-Z83366, Z83352, Z83354, Z83358, Z83350, Z83346, AF003189-AF003195); Lepus castroviejoi (Broom
Hare; DQ883207); Lepus europaeus (European hare; DQ645450, DQ883196-DQ8832016, AY876080 -AY876118,
AY466782-AY466853, AY300032-AY300036, AY1546661-AY154666, AY103494-AY103531, AY163356AY163376, AF1574363-AF157454, DQ645432 -DQ645449, AJ421471, Y15315); Cervus elaphus (red deer;
AM419026, DQ386106-DQ386110, AM279271, DQ452075-DQ452087, NC007704, AB012385, AF016972AF016973); Capreolus capreolus (roe deer; AM419028, DQ114784, DQ384640-DQ384708, DQ114745DQ114783, AM279273, AY625732-AY625892, Z70318, AJ287365-AJ287383); Sus scrofa (wild boar; D17739,
D16483, AJ854456).
4. Human
Homo sapiens (Human; AF254896 and D38112).
a
As there were no available D-loop sequences for Gennetta gennetta in Genbank, the PCR procedure was only tested
on tissue/hair samples.
- 106 -
CHAPTER
1
RELIABLE FAECAL DNA GENOTYPING OF
SYMPATRIC MARTENS
PAPER IV
Reliable faecal DNA genotyping of sympatric marten species
(Martes martes and Martes foina): The impact of sample collector field
experience on species and individual identification success
rates
- 107 -
4
CHAPTER
4
- 108 -
Paper IV
RELIABLE FAECAL DNA GENOTYPING OF SYMPATRIC MARTENS
PAPER IV
Reliable faecal DNA genotyping of sympatric marten species
(Martes martes and Martes foina): The impact of sample collector field
experience on species and individual identification success
rates
ABSTRACT
Non-invasive wildlife research using DNA from faeces has become increasingly popular. In this study, we
assessed the reliability and success of using genetic techniques to determine species and individual
identification of sympatric martens (Martes martes and M. foina) by genotyping non-invasively collected
faecal samples. First, we developed a novel and accurate multiplex panel of 15 microsatellites loci, selected
by cross species amplification of 41 loci. The application of this panel facilitated species distinction,
discarding the presence of putative hybrids. Further, in order to evaluate the impact of sample collector
experience on DNA quality, we conducted a genetic survey all over the sympatric range of both marten
species in the Iberian Peninsula by three different kinds of sample collectors (wildlife biologist; trained
volunteers and technical staff from natural parks). In order to achieve this goal, we evaluate the a) success
of PCR-RFLP identification of Martes sp. faecal samples and b) genotyping success and error rates of
faecal M. martes samples. Our results show that the difference in the level of expertise between sample
collectors significantly influence the success rate of microsatellite genotyping of pine marten faecal samples
(mean 45% n=317) but not the species identification success rate (mean 84%, n=634). Based on our
results, we recommend conducting sampling by an experienced biologist to maximize non-invasive
sampling and DNA quality so as to ensure accurate genotyping success. Application of our methods to
field collected scats can be used in a cost-effective way so as to investigate distribution, patterns of genetic
diversity and structure as well as to estimate population abundance for sympatric martens.
Keywords Non-invasive genetic sampling, Microsatellites, Genotyping success, Faecal DNA, Genetic
species identification, Martes
- 109 -
CHAPTER
4
INTRODUCTION
The European pine marten (Martes martes) and stone marten (Martes foina), species, which are similar in
morphology and feeding habits, are closely related mustelids that live to a large extent sympatrically over
an extensive area of Europe (Proulx et al. 2004). The North Iberian Peninsula constitutes the present
south-western distribution limit of the pine marten and consequently the southernmost sympatric area
with the stone marten (López-Martín, 2007). M. martes is either threatened or scarce in many countries
where forest habitat loss and fragmentation are major threats (Bright 1999). By contrast, the number of
stone martens has increased in many countries where they are currently distributed (Proulx et al. 2004)
and they have even invaded urban areas (Herr et al. 2009). However, given our limited knowledge of their
ecological niches in sympatric areas, their extremely elusive behaviour, and the fact that their faeces are
morphologically indistinguishable, there is considerable difficulty in studying and monitoring their
populations (Ruiz-González et al. 2008). As these inverse demographic trends and probable competition
between the two species may lead to the reduction of the pine marten’s range (Goszczyński et al. 2007)
and the geographic expansion of the stone marten, it is important to monitor the presence and numbers of
these species in different parts of Europe (Proulx et al. 2004).
Moreover, molecular phylogenetic studies confirmed that these two species are closely related, although
they are not the most related within the subgenus Martes (Stone & Cook 2002; Sato et al. 2003; Marmi et
al. 2004; Koepfli et al. 2008). Hybridization between Martes species has been previously documented
(Davison et al. 2001; Kyle et al. 2003; Grakov et al. 1994). Nevertheless, sympatric, M. martes and M.
foina do not interbreed, or at least this has not been previously reported (Davison et al. 2001).Thus,
additional analyses, including population genetic-level sampling, will be very useful in confidently
resolving relationships among these recently evolved species (Koepfli et al. 2008) and verifying possible
hybridization events.
Molecular methods incorporating non-invasive sampling via the collection of scats or hairs have become
common for population monitoring of carnivores, providing valuable information about species and
individual identification (Taberlet & Luikart 1999; Piggott & Taylor 2003; Waits & Paetkau 2005;
Schwartz & Monfort 2008; Beja-Pereira et al. 2009).
The standard method for monitoring marten populations are scat transect surveys (Birks et al. 2004).
Consequently, faeces should provide a plentiful non-invasive source material that is easy to collect for
genetic surveys (Beja-Pereira et al. 2009; Ruiz-González et al. 2008). However, the applicability of noninvasive techniques to a new species must be explored before embarking on large-scale studies (Taberlet &
- 110 -
Paper IV
RELIABLE FAECAL DNA GENOTYPING OF SYMPATRIC MARTENS
Luikart 1999; Waits & Paetkau, 2005; Valiere et al., 2007). The drawbacks of non-invasively collected
faecal samples of mustelids are (a) that species determination of faecal samples may be misidentified even
by experienced collectors (Davison et al. 2002; Birks et al. 2004) and (b) extracts from field-collected
faeces usually yield little target DNA and may contain polymerase chain reaction (PCR) inhibitors,
potentially leading to false allele amplification (FA) and/or allelic drop-out (ADO) on microsatellite
genotyping procedures (Waits & Paetkau, 2005). Consequently, working with non-invasively obtained
faecal samples of sympatric martens requires a previous determination of species identity so that we can
thereafter apply a valuable and informative microsatellite panel useful on faecal samples. In this context,
previous studies have recommended that any study requiring a non-invasive genetic method should be
preceded by a pilot study in which the probability of identity, as well as genotyping errors, should be
assessed (Taberlet & Luikart 1999; Valière et al.2007).
Recent progress in molecular techniques has supplied several non-invasive genetic methods for the species
identification of martens and related mustelids (e.g. Gómez-Moliner et al. 2004; Livia et al. 2007;
O´Reilly et al. 2008; Ruiz-González et al. 2008) which can be applied to field monitoring. However, there
are currently no microsatellites developed specifically for M. martes and M.foina for individual
identification, which could provide valuable information for population censuses and to address questions
regarding the population genetics of free ranging marten populations (Beja-Pereira 2009). Nevertheless,
cross-amplification of loci identified in related species should be used. Previous studies focused on pine
marten have given information about the genetic variability and structure of several European populations
(Kyle et al. 2003; Mergey 2007; Pertoldi et al. 2008; Mullins et al. 2010). However, information
regarding a valuable microsatellite panel for Iberian Peninsula populations of pine martens or stone
martens is unavailable. Moreover, theses studies were mainly focused on tissue (Kyle et al. 2003, Mergey
2008; Pertoldi et al. 2008) or remotely plucked hair (Mullins et al. 2010) so no information about the
effectiveness of microsatellite markers on faecal marten samples is currently available.
In addition, to effectively and efficiently apply faecal DNA analysis in large-scale studies, it is important to
identify the variables that impact PCR amplification success (Waits & Paetkau 2005; Murphy et al. 2007).
Although laboratory procedures can be optimised to reduce the impact of genotyping errors and increase
genotyping success [e.g. faecal preservation method (Wasser et al. 1997; Murphy et al. 2000; Murphy et
al. 2002; Piggott & Taylor 2003) DNA extraction method (Flagstad et al. 1999; Goossens et al. 2000;
Frantz et al. 2003; Wehausen et al. 2004), and amplification method (Goossens et al. 2000; Bellemain &
Taberlet 2004; Piggott et al. 2004)], other extrinsic field factors [e.g. age of the faecal sample (Lucchini et
al. 2002; Piggot 2004, Santini et al. 2007; Brinkman et al. 2010), weather conditions (Farrell et al. 2000;
- 111 -
CHAPTER
4
Lucchini et al. 2002; Piggot 2004; Brinkman et al. 2010), diet (Murphy 2003; Murphy et al. 2007)
and/or season (Maudet et al. 2004; Hajkova et al. 2006)], are usually difficult to control in non-invasive
field surveys (Piggot & Taylor, 2003). Also, the considerable remaining variability among studies implies
that other unidentified parameters are acting (Broquet et al. 2007). In many different non-invasive genetic
studies the sample collection procedure was conducted by different staff with very different field
experience [e.g. game wardens and hunters (Bellemain et al. 2005); experienced field volunteers (Jacob et
al. 2010), specialised wildlife biologists (Rossellini et al. 2008) rangers and technical staff from protected
areas and/or a combination of different personnel (Fabri et al. 2007)]. The difference in the level of
expertise between sample collectors potentially impacts accuracy in the species identification of faecal
samples (Davison et al. 2002; Zuercher et al. 2003; Prugh & Ritland 2005) and results in population size
estimation surveys of rare or elusive species (e.g. Jacob et al. 2010). This bias may be limited if volunteers
have the opportunity to train and acquire more experience in species identification and fieldwork (Prugh
& Ritland 2005; Jacob et al. 2010). In spite of the clear relevance of this field parameter for reliable and
cost effective non-invasive genetic studies, to our knowledge the effect of sample collector experience on
species identification via mitochondrial DNA and on genotyping success and errors rates has not
previously been evaluated.
Thus, the aims of this study were: a) First, to develop a reliable and accurate mutiplex panel of
microsatellites for individual identification of M. martes and test its applicability to M. foina in both tissue
and scat samples. b) Second, to apply this panel for the detection of possible hybridization events between
both species in sympatric areas and c) finally, to assess the impact of the field experience of sample
collectors on the success of PCR-RFLP identification of Martes sp. faecal samples and d) on genotyping
success and the error rates of faecal pine martens samples by conducting a genetic survey across the
sympatric area of both marten species in the northern Iberian Peninsula.
MATERIAL AND METHODS
Sample collection and DNA extraction
Tissue samples
Fresh tissue specimens from roadkilled animals were used to isolate DNA from the target marten species,
M. martes and M. foina. Marten specimens were collected across a wide geographic area in the northern
Iberian Peninsula which covers the sympatric distribution range of both species (López-Martín 2007).
- 112 -
Paper IV
RELIABLE FAECAL DNA GENOTYPING OF SYMPATRIC MARTENS
DNA was isolated from tissues and hairs using the Qiagen DNeasy Tissue DNA extraction kit (Qiagen,
Hombrechtikon, Switzerland) according to the manufacturer’s instructions.
Faecal samples
The faecal sampling was conducted between 2006 and 2009 all over the sympatric range of both marten
species in the northern Iberian Peninsula (López-Martín 2007).
To assess the impact of sample collector experience on the results of microsatellite genotyping and species
identification rates, samples were collected opportunistically by three different kinds of staff: a) Wildlife
biologists with recognized experience in non-invasive surveys of carnivores (WB), b) Volunteers who
received a training course on how to identify Martes sp. scat samples (TV) and c) rangers and technical
staff from different National Parks (TS). The wide study area and long-term programme did not permit
the standardization of sampling efforts across space and time. However, the proportion of the samples
obtained for each kind of sample collector was similar in each season.
Before sampling, we mailed a brief description of the project, the sampling protocol and material for
sample collection (individually labelled 2 ml autoclaved tubes containing ethanol 96%) to each of the
collaborators. European pine martens, stone martens and other carnivores use forest roads and frequently
defecate on them as a way of visual-scent marking (Barja 2005). Thus, sampling was conducted
systematically along linear features, such as forest trails and paths. Morphological characteristics, such as
size and shape, were used to presumably distinguish faeces of the genus Martes from those of mediumsized carnivores such as the red fox and the wildcat (Ruiz-González et al. 2008).
For each faecal sample, sampling date, sample collector and Universal Transversal Mercator (UTM)
coordinates using a global positioning system (Garmin eTtrex) were recorded. Collaborators were warned
to collect only fresh samples (<5 days). Fresh faeces were characterised by the presence of a mucus layer
and lack of any sign of dehydration (Rosselinni et al. 2009). Moreover, to assess the validity of the fresh
designation in each faecal sample a pre-selection procedure was applied by evaluating a photograph of each
faecal sample, discarding samples typified as fresh that presumably correspond to this category.
A portion of each “marten like” faecal sample was picked up by with stick and preserved in 2 ml
autoclaved tubes containing 96% ethanol and preserved at room temperature until processed (GómezMoliner et al. 2004; Ruiz González et al. 2008). A different stick and tube were used for each sample.
In each extraction step, special care was taken to avoid cross-contamination, which is an important
problem especially when handling DNA from non-invasive samples (Taberlet et al. 1999; Pompanon et al.
- 113 -
CHAPTER
4
2005). Thus, extractions were performed in a dedicated laboratory on a sterile bench, using filter tips
throughout. To monitor potential contamination, we included one negative extraction control per
extraction process. DNA was then extracted using the DNA Stool MiniKit (Qiagen, Hombrechtikon,
Switzerland) following the manufacturer’s protocol for DNA extraction from stool samples. The DNA was
eluted from the silica membrane twice using 100 μL buffer AE (Qiagen).
Species identification
Martes species identification of each faecal sample was accomplished by a polymerase chain reaction–
restriction fragment length polymorphism (PCR-RFLP) method, providing for an effective genetic
identification of sympatric marten species (Ruiz-González et al. 2008). A small fragment (276 bp) of
mitochondrial DNA (mtDNA), D-loop region, was amplified with specific primers developed for Martes
and Mustela by PCR. The primers were designed to amplify small fragments to maximise the probability
of amplification of degraded DNA. Following DNA amplification, PCR products were digested with the
restriction enzymes HaeIII and RsaI. The combined use of both enzymes produced a species-specific
banding pattern allowing the scats of M. martes and M. foina to be discriminated (see Ruiz-González et al.
2008 for further details).
Microsatellite analysis
Development of a microsatellite multiplex protocol for M.martes and M.foina useful for faecal DNA genotyping
Forty one microsatellite loci identified in the genomic DNA of other mustelid species were tested for
cross-species amplification and polymorphism in M. martes and M. foina. The 41 microsatellite primer sets
were originally developed for M. americana (Ma1, Ma2, Ma3, Ma4, Ma5, Ma8, Ma9, Ma10, Ma11,
Ma15, Ma18, Ma19; Davis & Strobeck 1998 ), Gulo gulo (Gg7, Gg14, Davis & Strobeck 1998; Gg454,
Walker et al. 2001) Meles meles (Mel1, Mel6, Bijlsma et al. 2000; Mel10, Domingo-Roura et al. 2003),
Mustela lutreola (MLUT27, MLUT04, Cabria et al. 2007), Neovison vison (Mvis020, Mvis022, Mvis072,
Mvis075, Fleming et al. 1999; Mvi39, Mvi57, O’Connell et al. 1996), M. erminea (Mer 22, Mer 41 and
Mer 95; Fleming et al. 1999); Lutra lutra (Lut604 and Lut615, Dallas & Piertney 1998) and for Martes
pennanti (MP0059, MP0084, MP0100, MP0144, MP0175, MP0188, MP0200, MP0234, MP0247,
MP0288, Jordan et al. 2007). These loci were selected for their high polymorphism and heterozygosity as
show in previous studies of pine martens (Kyle et al. 2003; Pertoldi et al. 2008) or related species (Jordan
et al. 2007).
- 114 -
Paper IV
RELIABLE FAECAL DNA GENOTYPING OF SYMPATRIC MARTENS
PCR conditions were firstly optimised for each primer in singleplex and the annealing temperature was
selected after temperature gradient PCR on tissue-derived DNA. Microsatellite amplifications were
performed in a total volume of 15 μL with 2 μL DNA, 0.2 μL of each primers (20 pmol/μL), 1.7 μL 10×
reaction buffer, 0.64 μL MgCl2 (50 mM), 0.5 μL deoxynucleotide triphosphates (2.5 mM), 0.1 μL of
bovine serum albumin (10 mg/mL), 11.56 μL of sterile water and 0.1 μL Taq polymerase (5 U/μL). The
following PCR conditions were used for all amplifications: After incubation for 5 min at 94 °C, the
samples were subjected to 40 amplification cycles in a BIO-RAD iCycler (Version 3.021, BIO-RAD
Laboratories) consisting of denaturation for 1 min at 94 °C, annealing for 1 min at 54-59 °C and a final
extension stage of 1 min at 72 °C.
Loci with a low rate of amplification, monomorphic in both species, and with a high quantity of
unspecific amplification were discarded. This first screening was evaluated with 15 tissues, and 10 faecal
samples of each species from the Iberian Peninsula (data not shown).
From the 41 screened loci we selected a suite of 15 markers to design a Multiplex protocol with special
emphasis on demonstrated variability in pine martens, amplification strength with small quantities of
DNA (faecal DNA) and powerful enough to allow individual identification. Forward primers were
fluorescently labelled with different dyes (6-FAM, NED, PET, VIC) to enable multiplex electrophoresis of
microsatellite products. In order to avoid noise from variable adenylation during the PCR, the ‘pigtail’
sequence GTTTCTT was added to the 5 -end of each reverse primer (Brownstein et al. 1996). PCR
multiplex amplifications were carried out with QIAGEN Multiplex PCR kits using the manufacturer’s
protocol in a total volume of 10 μL with 2 μL of DNA and 2 pmol of each fluorescence labelled forward
and unlabelled reverse primers. We applied a hot-start thermocycling protocol. The initial polymerase
activation (HotStart PCR) was done at 95 °C for 15 min, followed by 42 cycles (35 for tissue samples) of
denaturation at 94 °C for 30s, primer annealing at 57 °C for 90s, and sequence extension at 72 °C for 60s,
and a final extension step at 60 °C for 30 min.
In addition to the negative controls for extraction, negative PCR controls were included as proposed by
Pompanon et al. (2005). Additionally, we amplified a reference sample as a positive control and to test
that the electrophoretic mobility of the fragments was consistent across runs (Davison & Chiba 2003).
In non-invasive genetic sampling, genotyping errors occur due to increased rates of null alleles, allelic
dropout (ADO) and false alleles (FA) (Taberlet et al. 1999; Pompanon et al. 2005). Therefore, we
followed a modified multiple-tube approach from Taberlet et al. (1996), amplifying each DNA extract in
four replicates and in separate rooms dedicated to low DNA-content samples.
- 115 -
CHAPTER
4
First, to test the applicability of the multiplex protocol designed on non-invasive genetic surveys, 21 PCRRFLP identified faecal samples of each species were genotyped four times per locus to estimate error rates
and genotyping success. These samples were randomly chosen from each kind of sample collector (7 from
WB, 7 from TV and 7 from TS). Moreover, 10 of the 35 tissue samples of each species were amplified
four times per locus to estimate error rates for tissue samples. As errors were almost absent from tissue
samples (see Results), the remaining 25 tissue samples were amplified only once per locus.
Later, in order to assess genotyping success and error rates on field studies for each of the different kinds of
samples collector, all the faecal samples identified by the PCR-RFLP method as pine marten were
genotyped at 15 loci. DNA quality was initially screened by PCR-amplifying each DNA sample four times
at four loci (Multiplex 1: MP0188; MP0059; Gg-7; Ma1). Only samples showing >50% positive PCRs
were further amplified four times for the remaining 10 loci. Samples with ambiguous results after four
amplifications per locus or with <50% successful amplifications across loci were removed from further
analyses as they were not considered reliable genotypes.
Multiplex PCR products were run on an ABI (Foster City, CA) 3130XL automated sequencer (Applied
Biosystems), with the internal size standard GS500 LIZ™ (Applied Biosystems). Fragment analyses were
conducted using ABI software GENEMAPPER 4.0.
Data analysis
Probability of identity, genotype checking, and individual identification
To test the discrimination power of the set of 15 microsatellites, we computed the probability of pairs of
individuals bearing an identical multilocus genotype (PID) with GIMLET v 1.3.4 (Valière 2002) using a
data set of genotypes obtained from tissue DNA of 35 pine and stone martens, respectively. PID
calculations were performed with both the unbiased equation for small sample size and the equation for
siblings. The more conservative PID for full-sibs (PID-Sib) was estimated as an upper limit to the probability
that pairs of individuals would share the same genotype. As advocated in Waits et al. (2001), the observed
probability in the population ranged from unbiased PID to PID-Sib (Waits et al. 2001).
Consensus genotypes from four replicates were reconstructed using Gimlet v 1.3.4 (Valière 2002),
accepting heterozygotes if the two alleles were seen at least in two replicates and homozygotes if a single
allele was seen at least in three replicates. Gimlet was also used to estimate PCR success and errors: ADO
(Taberlet et al. 1996), and FA (Gagneux et al. 1997).
- 116 -
Paper IV
RELIABLE FAECAL DNA GENOTYPING OF SYMPATRIC MARTENS
We used GENETIX (Belkhir et al. 2004) to estimate observed (HO) and expected heterozygosities (HE),
the number of total alleles (A), shared alleles (SA) and private alleles (PA) for each locus and for each of
the tested species in this study (Mm= Martes martes; Mf= Martes foina).
Deviation from Hardy-Weinberg equilibrium (HWE) was tested using the exact test implemented in
GenePop version 4.0 (Raymond & Rousset 1995; Rousset 2008). Statistical significance was evaluated by
running a Markov Chain Monte Carlo (MCMC) consisting of 10,000 batches of 10,000 iterations each,
with the first 10,000 iterations discarded before sampling (Guo & Thompson 1992). Significance levels
were adjusted with sequential Bonferroni correction in order to correct for the effect of multiple tests (Rice
1989).
MICRO-CHECKER software (Van Oosterhout et al. 2004) was used to check for potential scoring errors
and the presence of null alleles.
Bayesian admixture analysis of sympatric martens
STRUCTURE version 2.2 (Pritchard et al. 2000; Falush et al. 2003; Falush et al. 2007) was used to
determine the level of admixture, calculated as the proportion of the genome of an individual that is
originated from each of the two parental species. Simulations were run using a burn-in period of 105
sweeps followed by 106 MCMC iterations. Independent runs of K (i.e. number of clusters or gene pools
assumed) were performed from one to four clusters and repeated ten times to check for consistency in the
results. Admixture ancestry and correlated allele frequency models were used without prior knowledge of
genetic information. Individual assignment probabilities refer to the proportion of the individual’s genome
which originated from European pine marten (qi) or stone marten (qj). Moreover, we calculated pair-wise
FST values (Weir & Cockerham 1984) and tested for pairwise genetic differences among clusters
identified by STRUCTURE using a randomization procedure implemented in FSTAT v 2.9.3.2 (Goudet
1995; Goudet 2001). Statistical significance at the 0.05 level was evaluated after the Bonferonni correction
for multiple comparisons (Rice 1989)
Effect of sample collector experience on species identification and genotyping success
In order to evaluate the impact of the field experience of sample collectors on faecal DNA sample quality
for genetic analysis, we evaluate five different parameters for each of the three different kind of sample
collector: i) Martes sp. identification rate (proportion of samples identified by the PCR-RFLP method as
M. martes or M. foina: Msp.), ii) microsatellite amplification rate (PCR replicates that yielded at least one
scoreable allele: PCR+), iii) genotyping error rates (FA and ADO), iv) the proportion of samples that did
not pass the microsatellite quality screening (SCR-), and v) the proportion of samples yielding a reliable
- 117 -
CHAPTER
4
genotype (samples correctly genotyped at 15 loci: G+). The first parameter was calculated from all the
analyzed samples and the other four from all the faecal samples identified as pine marten.
To compare the proportions obtained for each parameter and to determine whether they can be
considered as equal for the different sample collectors or if at least two proportions are significantly
different we used a Chi-square test based on 10,000 Monte Carlo simulations. The Marascuilo procedure
(Marascuilo & Slaughter, 1981) was also used if the Monte Carlo simulation rejected H0, as the
Marascuilo procedure compares all pairs of proportions, which enabled the pairwise differences to be
identified. Statistical analysis was performed using XLSTAT (Addinsoft™, New York, USA). Results were
considered statistically significant at p =0.05 level.
RESULTS
Reliability of a microsatellite multiplex panel for individual identification of M. martes and M. foina:
Genetic variability, probability of identity and genotyping errors
Details of the multiplex panel of 15 microsatellites loci, selected by cross species amplification of 41 loci,
are shown in Table 1. Allele frequencies at all loci were estimated based on tissue samples from 35
different individuals of each species. All but three of the analyzed microsatellite loci (MP0188, Mel-10 and
Ma-19 were monomorphic in M. foina) were polymorphic in both mustelid species, with an average
number of alleles per polymorphic locus of 4.13 and 3.31 (ranging from 2 to 6) in M. martes and M.
foina, respectively (Table 1). The presence of private alleles is considered one of the most significant
indicators of population or species distinction (Beaumont et al. 2001; Oliveira et al. 2008). Only 13 out of
the 96 unique alleles found in both species (13.54%) were shared alleles, whereas 33 (34.38%) and 50
(52.08%) were private alleles for M. martes and M. foina, respectively. Interestingly, alleles found at 7
microsatellite loci were species-specific with slight differences in allele size (Table 1).
Comparison of HE and HO revealed slight differences between pine martens and stone martens (Table 1).
Observed heterozygosity was lower than expected heterozygosity in both species. Mean expected
heterozygosity over all polymorphic loci was 0.586 and 0.469 and mean observed heterozygosity was
0.552 and 0.462 for M. martes and Martes foina, respectively (Table 1).
Significant departure from HW equilibrium after Bonferroni correction was observed at two loci in pine
martens (MP0059 and Lut-435), whereas only one locus deviated from HW equilibrium proportions in
stone martens (Ma-1). The results of MICROCHECKER rejected the hypothesis that putative null alleles
were causing bias in genetic diversity and population differentiation estimates.
- 118 -
Paper IV
RELIABLE FAECAL DNA GENOTYPING OF SYMPATRIC MARTENS
High genetic differentiation was detected between the two mustelid species based on 15 microsatellite data
(FST=0.485; P<0.001).
Table 1. Properties of the 15 microsatellite multiplexed loci used in this study and summary of the genetic
variability assessed per locus and species. The table includes: number of alleles (NA), shared alleles (SA),
private alleles (PA), observed (HO) and expected (HE) heterozygosities, rates of positive PCR (PCR+),
dropout (ADO) and false allele (FA) for each locus and for each of the species tested in this study (Mm=
Martes martes; Mf= Martes foina).
MULTIPLEX
Locus
Gg-7
MULT_1
MULT_3
NED
MP0059*
VIC
Lut-453
164-
166-
172
174
204-
203-
210
213
140-
139-
148
147
FAM
121
6-
108-
102-
FAM
112
108
265-
264-
275
272
Mel-10
NED
Lut-435*
VIC
Ma-19
PET
Lut-615
Mf
113-
PET
Mvi072
Size range
Mm
6-
Mel-1
Mvi-57
MULT_4
PET
Ma-1*
MP0188
MULT_2
DYE
158166
105
127
129-
135-
143
139
206210
211
6-
100-
100-
FAM
114
108
260-
264-
276
270
6-
111-
113-
FAM
115
119
167-
169-
179
177
150-
156-
160
164
198-
186-
204
194
NED
Ma-2
NED
Mer41
VIC
Mlut-27
PET
NA
SA
Mf
Mm
3
5
6
PA
HE
HO
PCR+
ADO
FA
Mf
Mm
Mf
Mm
Mf
Mm
Mf
Mm
Mf
Mm
Mf
Mm
2
1
3
0.342
0.713
0.371
0.638
0.13
0.96
0.333
0.143
0
0
4
0
6
4
0.675
0.520
0.629
0.444
0.94
0.91
0.345
0.167
0
0
4
4
0
4
4
0.707
0.615
0.628
0.472
0.96
0.98
0.150
0.092
0
0.072
1
3
0
1
3
-
0.483
-
0.418
0.81
0.97
0
0.139
0
0
3
3
1
2
2
0.292
0.513
0.228
0.5000
1
0.96
0.400
0.093
0
0.033
3
5
0
3
5
0.378
0.643
0.371
0.667
0.98
0.87
0.281
0.174
0
0.009
1
5
0
1
5
-
0.536
-
0.472
1
0.88
0
0.375
0
0.014
3
3
1
2
2
0.493
0.508
0.543
0.361
0.94
0.86
0.333
0.114
0.167
0.027
1
3
0
1
3
-
0.578
-
0.556
0.96
0.85
0
0.192
0
0
4
5
2
2
3
0.484
0.416
0.543
0.444
1
0.94
0.300
0.167
0
0.014
4
6
2
2
4
0.477
0.726
0.457
0.806
0.94
0.77
0.190
0.183
0
0.023
4
2
1
3
1
0.469
0.490
0.400
0.528
0.88
0.89
0.389
0.269
0
0.014
4
5
2
2
3
0.629
0.672
0.628
0.583
0.85
0.93
0.167
0.274
0
0.008
3
5
2
1
4
0.277
0.757
0.286
0.805
0.94
0.87
0.361
0.316
0
0
2
4
0
2
4
0.408
0.616
0.457
0.583
0.94
0.91
0.325
0.333
0
0
13
33
50
0.469
0.586
0.462
0.552
0.88
0.90
0.238
0.202
0.011
0.014
Total
46
62
Mean
3.31
4.133
Means for the number of alleles and Heterozigosities are computed over polymorph loci. Loci marked with an asterisk
deviated from Hardy–Weinberg proportions at the p =0.05 level (Ma-1 for M. foina; MP0059 and Lut-435 for M.
martes).
- 119 -
CHAPTER
4
Probability of identity was calculated on the basis of allele frequencies estimated from the 35 tissue
samples of each species. The overall PID using all loci for pine martens and stone martens was 5.42x10-11
and 5.42x10-7 and the overall PID-sibs was 4.47x10-5 and 1.95x10-3, respectively. When locus Gg-7 was
excluded, which was the locus that was difficult to type for stone marten faeces samples, the overall PID
became 1.21x10-6 and the overall PID-sibs was 2.79x10-3. Thus PID-sibs was again lower than the 0.01
threshold necessary to prevent the shadow effect, i.e. the presence of two or more individuals with the
same multilocus genotype (Mills et al. 2000) (Fig. 1).
Thus, the 15 loci used in this study were necessary, yet sufficient, to distinguish with 99% certainty
between sibling pine or stone martens (Mills et al. 2000) (Fig. 1). Moreover, the PID-sib showed that the
proportion of individuals with identical profiles dropped to zero if the most informative loci were used (3
in M. martes and 8 in M. foina) (Fig. 1).
Figure 1. Probability of identity (PID) and probability of identity for full-sibling (Pid-sibs) values with the
addition of loci in decreasing order of heterozygosity for M. martes (Mm) and M. foina (Mf). The 1% cutoff line represents the point where enough loci are typed to distinguish between individuals with 99%
certainty.
Genotyping errors from tissue were estimated from 10 samples of each marten species. No ambiguous
genotypes were detected. The proportion of PCR+s varied between 90 and 100% among loci and from 86
- 120 -
Paper IV
RELIABLE FAECAL DNA GENOTYPING OF SYMPATRIC MARTENS
to 100% among samples for both species. Only 1 ADO was detected, at locus Gg-7 in pine martens and 3
in the same locus for the stone marten.
Genotyping success and error rates for faeces were preliminarily estimated from 42 samples (21 of each
species). 13 faecal samples of M. martes (61.9%) and 12 (57.14%) of M. foina were completely and
successfully genotyped. The other samples showed less than 50% amplification success or high genotyping
error rates after 4 replications at 15 loci, so they were discarded as we were not able to resolve consensus
genotypes from repeated attempts. Gg-7 was not valuable for faecal sample genotyping of stone martens
because of the low amplification success rate (0.13).
The average proportion of positive PCRs for full multilocus microsatellite genotypes of faecal samples was
88% for the stone marten and 90% for the pine marten (Table 1) and varied among loci from 77% and
100% in both species. The mean ADO rate for M. foina and M. martes was 0.238 and 0.202, respectively.
The ADO rates were not homogenous in each of the multiplexed reactions and each of the species (see
Table 1). False allele rates averaged 0.011 for stone martens and 0.014 for pine martens (Table 1). Error
types other than ADO or false allele FA, were also detected but they appeared at very low rates (< 0.001)
and we considered them negligible.
Bayesian admixture analysis for species assignment
Analyses using STRUCTURE with both the admixture ancestry and correlated allele frequency models
supported the existence of two distinct genetic clusters (k=2) based on the log probability of the data (ln
[Pr(X/K)]) given the model. Moreover, the modal value of ΔK (Evanno et al. 2005) was also shown at K=2
(data not shown). Cluster I grouped all the samples identified as M.martes by both phenotype and PCRRFLP method, with an estimated average proportion of membership (QI) higher than 0.99 (Figure 3).
Cluster II included all the individuals identified as stone martens by phenotype and PCR-RFLP with QII >
0.99 (Figure 2).
- 121 -
CHAPTER
4
Figure 2. Cluster identification by STRUCTURE without prior genetic information. Each individual is
represented as a vertical line partitioned into two segments whose length is proportional to the individual’s
estimated membership coefficients. The analyses were performed assuming two (K=2) distinct genetic
clusters.
Effect of sample collector experience on Martes sp. identification and pine marten genotyping success of
faecal DNA
Sample collection and Martes species identification
Out of 849 faecal samples collected from the entire sympatric distribution range of both marten species in
the northern Iberian Peninsula (López-Martín 2007), 215 (25.32%) were finally discarded as they were
not assigned as fresh samples according to the pre-selection procedure (sample collector indications and
photograph verification). The percentage of discarded samples was greater on samples collected by TS
(29.38%) than by TV (28.97%) and WB (20.00%), respectively (Chi-square test, P=0.009) (Table 2).
The Marascuilo procedure indicated that the difference between the proportions of discarded samples was
statistically significant between WB and TS and between WB and TV. After sample pre-selection
procedure, we obtained a total number of 634 fresh faecal DNA samples suitable for genetic analysis
(Table 2). 532 were identified as one of the target species by our PCR-RFLP method. Thus, unequivocal
species identification was possible in 83.91% of the samples. In the remaining 12%, the DNA extracted
was not amplified by the primers used. The species identification success rate was similar for all three
sample collectors with slightly higher values for WB (85.56%) and TV (84.58%) in comparison with TS
(79.87%) (Figure 3, Table 2). However, differences were not significant (P=0.284).
Genotyping success and error rates of pine marten faecal samples
From the total scat samples identified as pine marten (n=317) we evaluated the effect of sample collector
experience on microsatellite amplification rate (PCR+), genotyping error rates (FA and ADO), the
- 122 -
Paper IV
RELIABLE FAECAL DNA GENOTYPING OF SYMPATRIC MARTENS
proportion of samples that failed to pass quality screening (SCR-) and samples yielding a reliable genotype
(G+). The first quality-screening test was not passed by 121 samples (38.17%), which were immediately
discarded. The proportion of the samples that did not pass quality-screening was higher, and statistically
significant (P=0.003), in both TV (49.23%) and TS (46.83%) in comparison to WB (30.06%) (Table 2).
The Marascuilo procedure identifies WB collected samples as most differentiated from both TV and TS
samples (Table 2). The remaining 196 samples (61.83%) were amplified for the other 11 loci. After
multiple-tube genotyping, 52 samples (16.4% of the total pine marten samples analysed) were then
discarded because they showed <50% PCR success, or because of high failure rates. Full multilocus
microsatellite genotypes were obtained for the remaining 144 samples. (73.47% from the samples that
passed the screening and 45.43% from the total M. martes samples analyzed). After a regrouping
procedure, we identified 114 individual genotypes (Table 2). The highest genotyping success rate
(52.60%) was obtained for samples collected by WB, followed by TV (40.00%) and TS (34.18%). Chisquare test showed significant difference between the proportion of samples correctly genotyped across
sample collectors (p=0.018), and the Marascuilo procedure showed that the only significant difference was
that obtained between WB and TS collected samples. The overall PCR success rate (calculated only for
correctly genotyped samples) was 95% with not significant differences (P=0.118) and similar values across
sample collectors. The observed average error rates across loci were: ADO =0.189 and FA =0.021. The
results obtained for the three sample collectors were very similar and not statistically significant (ADO,
P=0.071; FA, P=0.527) (Figure 3 Table 2).
- 123 -
CHAPTER
4
Table 2. Number of samples collected (SC), discarded (SD) and analyzed (SA); number (in brackets) and
percentage of samples non-amplified by mtDNA (AMP-), identified as M. martes or M. foina (Msp),
identified as M. foina (Mf), identified as M. martes (Mm); identified as M. martes that failed to pass
screening (SCR-), not correctly genotyped (G-) and correctly genotyped (G+) and rates of positive PCR
(PCR+), dropout (ADO) and false allele (FA) for each of the three different kinds of sample collectors:
wildlife biologist (WB); trained volunteers (TRV) and technical staff (TS).
WB
355
71
284
14.44
(41)
85.56
(243)
24.65
(70)
60.92
(173)
30.06
(52)
17.34
(30)
52.60
(91)
0.96
0.142
0.014
69
SC
SD*
SA
AMPMsp
Mf
Mm
SCR-*
GG+*
PCR+
ADO
FA
IDIND
Sample Collector
TRV
283
82
201
15.42
(31)
84.58
(170)
52.24
(105)
32.34
(65)
49.23
(32)
10.76
(7)
40.00
(26)
0.94
0.137
0.02
18
TS
211
62
149
20.13
(30)
79.87
(119)
26.85
(40)
53.02
(79)
46.83
(37)
18.99
(15)
34.18
(27)
0.95
0.159
0.015
27
Total
849
215
634
16.09
(102)
83.91
(532)
33.91
(215)
50.00
(317)
38.17
(121)
16.4
(52)
45.43
(144)
0.95
0.189
0.021
114
*P-value < 0.05 for Chi-square test statistic.
For positive Chi-square test, the Marascuilo procedure was used to determine which pairs of proportions, corresponding to
each of the sample collectors, were statistically significant. For the three parameters in which the Chi-square test was
positive the statistically significant pairs were:
SD (WB-TV; WB-TS); SCR- (WB-TS; WB-TS); G+ (WB-TS).
SD
SCR-
G+
Contrast
Value
|p(WB) - p(TV)|
|p(WB) - p(TS)|
|p(TV) - p(TS)|
|p(WB) - p(TV)|
|p(WB) - p(TS)|
|p(TV) - p(TS)|
|p(WB) - p(TV)|
|p(WB) - p(TS)|
|p(TV) - p(TS)|
0.090
0.094
0.004
0.192
0.168
0.024
0.126
0.184
0.058
- 124 -
Critical
value
0.084
0.093
0.101
0.174
0.162
0.205
0.175
0.160
0.198
Significant
Yes
Yes
No
Yes
Yes
No
No
Yes
No
Paper IV
RELIABLE FAECAL DNA GENOTYPING OF SYMPATRIC MARTENS
Figure 3. Comparison of percentage (%) of discarded samples by pre-selection procedure (SD), Martes
species identification success (Msp), genotyping success (G+), positive PCR (PCR+), dropout (ADO) and
false allele (FA) for each of the three different kinds of sample collectors: wildlife biologist (WB); trained
volunteers (TRV) and technical staff (TS).
*
*
*
*P-value < 0.05 for Chi-square test statistic.
For positive Chi-square test, the Marascuilo procedure was used to determine which pairs of proportions, corresponding to
each of the sample collectors, were statistically significant. For the three parameters in which the Chi-square test was positive
the statistically significant pairs were:
SD (WB-TV; WB-TS); SCR- (WB-TS; WB-TS); G+ (WB-TS).
From the total analyzed samples (n=634) 16.08% could not be assigned to species level, 33% were
identified as M. foina, 8% failed to pass quality-screening and 19.08 % were not finally reliably genotyped.
Thus, approximately 23% of all the analysed samples were correctly genotyped (45.43% if we only
consider samples identified as M. martes by PCR-RFLP). The global and partial results for each sample
collector are summarized in Figure 3. The partial results obtained for each sample collector are slightly
different and mainly biased by the differences in the percentage of samples identified as M. foina.
- 125 -
CHAPTER
4
Figure 4. Percentage (%) of samples from the total analyzed samples that were: non-amplified by mtDNA
(AMP-), identified as M. foina (Mf), identified as M. martes that failed to pass quality-screening (SCR-),
identified as M. martes not reliably genotyped (G-) and identified as M. martes correctly genotyped (G+)
for each of the three different kinds of sample collectors (wildlife biologist=WB; trained volunteers=TRV
and technical staff=TS) and for the whole dataset.
DISCUSSION
Research on non-invasive genetics of mustelids are limited and biased towards species identification
methods (e.g. Gómez-Moliner et al. 2004; Livia et al. 2007; Ruiz-González et al. 2008). Microsatellite
genotyping of non-invasive samples has also been carried out on a limited number of species and used
mainly for individual identification and population size estimation (e.g. European otter, Hajkova et al.
2009; wolverine, Hedmark & Ellegren 2007; badger, Wilson et al. 2003; Frantz et al. 2006).
To our knowledge, this study is the first to investigate and evaluate a microsatellite panel for individual
identification of both sympatric marten species (M. martes and M. foina) while also being valuable for
faecal DNA genotyping. Moreover, this paper is the first attempt to elucidate genetic relationships
between pine and stone martens by microsatellites. Additionally, this research is the first to compare the
effect of sample collectors experience on Martes species identification and microsatellite genotyping success
and error rates in non-invasively collected faecal samples.
- 126 -
Paper IV
RELIABLE FAECAL DNA GENOTYPING OF SYMPATRIC MARTENS
Reliability of the microsatellite panel for individual identification and variability assessment of
sympatric martens
Reliable amplification of microsatellite markers is a prerequisite for estimating the size and structure of
populations within the landscape (Taberlet et al. 1999). In this paper, we standardized individual genetic
identification of sympatric martens from faecal DNA with protocols offering some advantages over past
efforts (Mullins et al. 2010). Firstly, species identification of the field-collected samples ensured that only
target species samples were used for further analysis. This initial step towards species identification is
crucial in landscapes where both species overlap extensively (Ruiz-González et al. 2008). In addition, the
screening of a sufficiently large number of 41 microsatellite loci for the initial assessment of variability
improved our ability to ascertain the combination of loci, resulting in higher precision in probability of
identity. This enabled us to get an optimal combination of a few very variable loci. This study identified,
by cross species amplification of 41 loci, a valuable panel of 15 microsatellite loci to consistently genotype
faecal samples of sympatric pine and stone martens.
The most conservative estimate of PID, i.e. PID-Sib statistics, suggested that DNA profiles consisting of the
15 loci used in our study would be sufficient to distinguish between individuals of each of the study
species, including siblings, with 99% certainty (Figure 1). The PID-Sib for the 15 loci (14loci in stone
marten, after excluding the most difficult to type Gg-7) used is 2.79x10-3 for stone martens and 5.42x10-7
for pine martens and can be used to distinguish even closely related individuals. Therefore they can also be
used in studies requiring individual identification for population size estimation based on non-invasive
scat sampling.
The genotyping success (G+) and error rates obtained from the 21 faecal samples from each species for the
microsatellite panel optimization procedure (G+: Mm=61.9%, Mf=57.14%; ADO: Mm=20.2%,
Mf=23.8%; FA: Mm=1.4% and Mf=1.1%) were similar to what was obtained for the whole pine marten
survey (GS: 45.43%; ADO: 1.89% FA: Mm=2.1%). Thus, the partial results of the pilot study reflected
the magnitude of the success and error rates that could be obtained on more extensive research.
The success rate of pine marten microsatellite genotyping (presented here as percentage of faecal samples
that provided consensus genotypes for all loci needed for reliable individual identification) in our study
was 45.43%. This result is around average in comparison with many other studies undertaken on other
carnivore species, which reported successful genotyping of faecal samples in the range of 14–79% [e.g. The
badger (Meles meles): 62%, Frantz et al. 2003. The wolverine (Gulo gulo): 54% Hedmark & Ellegren
2007; 70%, Flagstad et al. 2004. Otter (Lutra lutra): 14% Lanszki et al. 2008; 20% Dallas et al. 2003;
- 127 -
CHAPTER
4
21% Ferrando et al. 2008; 24% Kalz et al. 2006; 41% Prigioni et al. 2006; 44% Arrendal et al. 2007; 5563% Hajkova et al. 2009; 65% Hung et al. 2004; 73%, Janssens et al. 2008. Coyote (Canis latrans): 28%
Adams et al. 2007. Wolf (Canis lupus): 59%, Lucchini et al. 2002; 79 %, Creel et al. 2003; 52.8%,
Marucco et al. 2009, 50-61%, Stenglein et al. 2010]. However, many of these studies used less than 10
microsatellites so the ease with which a complete genotype can be obtained is greater if we compare that
complete genotyping with the 15 microsatellites used in this study.
Another important issue in non-invasive genetic studies is the detection and elimination of genotyping
errors, as they can dramatically affect survey results, especially when estimating abundance via genetic
sampling (Waits & Leberg 2000; Creel et al. 2003; Pompanon et al. 2005). The rate of genotyping errors,
therefore, should always be carefully computed and clearly reported (e.g. Broquet & Petit 2004). This
information can be considered as proof of data quality and thus a means of measuring the credibility of the
results (Bonin et al. 2004). The frequency of allelic dropout and false alleles in our study (ADO =17.8%,
FA =2.9%) was similar to that obtained in other non-invasive genetic studies of mustelids (e.g.
ADO=15.8%, FA=2%, Ferrando et al. 2008; ADO=27%, FA=8% Frantz et al. 2003).
There are a few studies focused on the genetic variability of pine martens (Kyle et al. 2003; Mergey 2008;
Pertoldi et al. 2008; Mullins et al. 2010), while no previous studies are available for M. foina. Kyle et al.
(2003) documented the genetic structure of several European pine marten populations; Mergey (2008)
studied the genetic structure of French populations inhabiting a fragmented landscape; while Pertoldi et
al. (2008) assessed the genetic variation of Danish pine martens across both space and time. More recently,
Mullins et al. (2010) developed a reliable and non-invasive method for censusing Irish pine marten
populations from remotely plucked hair. Eleven of the 27 different microsatellite loci used in these
previous studies have also been included in the present work.
Studying pine marten populations across Europe, Kyle et al. (2003) found microsatellite HE values in the
range of 0.56–0.64 for continental and 0.34-0.66 for insular populations (8 microsatellite loci). Mergey
obtained similar values for French populations (11 loci, HE=0.2). Pertoldi et al. (2008) found He values in
the range of 0.67-0.79 in Danish populations(11 loci), while Mullins et al. (2010) found slightly lower
values of HE for the reduced and isolated Irish populations (0.35, 17 loci). In this study using 15 variable
loci we detected HE values of 0.586. So, taking into account the high number of variable loci used in this
study we can estimate that in comparison to continental European populations, the Iberian population of
pine martens has lower levels of genetic variability. However, caution is called for when trying to interpret
differences in the level of genetic variability between different studies as these differences can also be
attributed to the different number of microsatellites used.
- 128 -
Paper IV
RELIABLE FAECAL DNA GENOTYPING OF SYMPATRIC MARTENS
The lower heterozigosity levels found in M. foina, in comparison to M. martes (Mf=0.469; Mm=0.586),
could be explained by the lower polymorphism of microsatellite markers in stone martens (Mm: A=4.13;
Mf: A=3.31). The microsatellite multiplex panel was primarily developed to be variable for pine martens as
landscape genetic research on this species is ongoing (Ruiz-Gonzalez et al. in prep.). Thus, further
optimisation of other polymorphic markers for M. foina may also be required for population/individual
level genetic analysis in this species.
Reliable species identification: no evidence of genetic hybridization
The genetic analyses performed in this study were based on different molecular markers (mtDNA and
nDNA), and gave consistent results that provided detailed information regarding reliable species
identification of sympatric marten species. The microsatellite data confirmed the validity of the species
identification results obtained by the PCR-RFLP method in mtDNA. The presence of species-specific
microsatellite alleles and loci, together with Bayesian admixture analysis, facilitated pure-bred species
distinction, discarding the presence of putative hybrids.
All the samples identified at species level by the PCR-RFLP method for the development of the
microsatellite panel (Mm=48; Mf=47), were identified as pure European pine marten or stone marten
based on microsatellite data analyses. The microsatellite panel generated in this study, together with
Bayesian analysis, provided an efficient method for corroborating the pure ancestry of sympatric martens.
These preliminary results, with the absence of phenotypically intermediate individuals described in the
literature, indicated that there is no hybridization between these two sympatric marten species in the
Iberian Peninsula.
Scats vs. hairs for microsatellite genotyping
Faecal or hair sampling methods have relative merits or biases depending on the size of the area surveyed,
the sampling effort and the cost and efficiency of the method (Beja-Pereira 2009; Schwartz & Monfort
2008). In terms of estimating population size, intensive surveys in reduced areas may be more cost
effective by genotyping remotely plucked hair as a result of requiring fewer amplifications to obtain
reliable genotypes (Mullins et al. 2010). However, more sophisticated field work is needed to obtain hair
samples in comparison to the easy collection of scats (Beja-Pereira 2009). Moreover, in areas with
sympatric martens (i.e. most of Europe) scat-based surveys could be more cost effective as a distribution
assessment and target DNA of each species can be obtained in one step. Thus, scats seem to be the most
effective for widespread surveys of sympatric martens where a large number of samples can be obtained by
- 129 -
CHAPTER
4
different collectors with a straightforward scat sampling protocol. Unfortunately, despite relatively easy
sampling in the field, subsequent genetic analysis of pine marten faeces, using nuclear microsatellite loci, is
challenging.
Effect of sample collector experience on Martes sp. identification and pine marten genotyping success of
faecal DNA
Faecal samples could potentially result in highly variable DNA quality and quantity, resulting in different
genotyping success and error rates (Beja-Pereira et al. 2009). The difference in the level of expertise among
sample collectors can potentially impact the results of microsatellite genotyping as has been documented in
this study.
The proportion of samples discarded from the total samples collected showed significant differences
among samples collected by different personnel, demonstrating that the sampling conducted by wildlife
biologists had the highest proportion of valuable fresh samples for genetic analysis. For example, a
sampling conducted by WB increased the overall available fresh sample for consecutive genetic analyses by
9% and 10% in comparison to TV and TS, respectively.
An overall high species identification rate was obtained (mean 84%), confirming previous results of the
application of the PCR-RFLP method on scat samples for monitoring the presence and distribution of
sympatric marten species in the northern Iberian Peninsula (Ruiz-González et al. 2008; Rosellini et al.
2008) and Italy (Balestrieri et al. 2009). These results are above the mean in comparison to that found in
Broquet et al. (2007), with an average amplification success rate of 73% of mtDNA reported in the noninvasive papers reviewed. Martes sp. identification rate was not statistically significant among different
sample collectors (WB=85.56%; TV=84.58%; TS=79.87%). Thus, after a sample pre-selection procedure,
sample collector experience had no significant effect on developing a reliable distribution survey of
sympatric martens, as all three groups of samples collectors achieved similar identification rates. Thus, the
work of non-specialized personnel may still be valuable in genetic studies focused on distribution
assessment as mtDNA methods can effectively work with highly variable DNA quality and quantity.
On the other hand, we demonstrate that sample collector experience can have a significant effect on
nuclear DNA quality and on the ability to obtain full genotypes from pine marten scats. Surprisingly, this
effect has been largely ignored in the faecal DNA literature.
Five different measures were used to quantify nDNA quality of PCR-RFLP identified pine marten scat
samples, but only two of these measures were significantly different among sample collectors.The
- 130 -
Paper IV
RELIABLE FAECAL DNA GENOTYPING OF SYMPATRIC MARTENS
proportion of the samples that did not pass the quality-screening was lower, and statistically significant for
wildlife biologist collected samples (30.06%) in comparison to trained volunteers (49.23%) and technical
staff (46.83). Thus, samples collected by WB increased the overall available samples for further genotyping
by 19.17% and 16.77% in comparison to TV and TS, respectively.
Genotyping success rates showed significant differences among samples collected by different personnel
demonstrating that the degree of experience in collecting fresh quality samples increases the probability of
obtaining a correct genotype. For example, a sampling conducted by WB increased the overall per sample
finalized genotyping success by 12.6% and 18.3% in comparison to TV and TS, respectively. By contrast,
PCR amplification success and error rates, because FA and ADO were not significantly different among
sample collectors. However, these two measures are less important indicators of differences in DNA
quality as PCR amplification products alone do not necessarily reflect true genotypes (Broquet & Petit
2004). Thus, in large-scale non-invasive genetic sampling projects, the field experience of sample collectors
can significantly increase the number of completed genotypes and maximize output per unit effort.
Recommendations for faecal genetic surveys of martens in sympatric areas
As already suggested by previous non-invasive genetic studies conducted on carnivore species (Piggot
2004; Santini et al. 2007; Murphy et al. 2007), success rates are greatly improved by the use of very fresh
samples. Thus, the effectiveness of sample collectors in meeting this objective is a crucial parameter. While
sample quality is a limiting factor for microsatellite genotyping, it is not so for species identification rate by
PCR-RFLP of mtDNA, after conducting a freshness sample pre-selection procedure.
The Martes sp. identification rate, based on a PCR-RFLP on mtDNA, was not influenced by sample
collection staff. Thus, the development of a reliable distribution survey of sympatric martens could be
done by different volunteers, independently of their previous experience, as the mtDNA methods seem
not to be affected by the quality of fresh samples. This can be explained by the higher copy numbers of
mtDNA compared to nDNA, conferring a greater success rate even with non-fresh samples that are not
suitable for genotyping analysis. However, the success variability of faecal DNA genotyping reported in
the present study for different sample collectors indicates that it is highly recommended that researchers
carry out a pilot study prior to implementing a full study, where the experience of collaborators in field
non-invasive surveys should be taken into account.
Factors that may influence success include intrinsic ones such as the individual, season and diet, over
which the researcher has minimal control (Piggot & Taylor 2004). However, extrinsic sources of variation
- 131 -
CHAPTER
4
in success such as sample collectors experience may be minimised by the collection of samples by
experienced wildlife biologists.
Finally, the results of the present work give some guidelines which should be taken into account before
starting faecal genotyping studies focused on pine martens living in sympatric areas with stone marten. In
our study, 45.43% of the samples identified as pine marten were fully genotyped by the 15 microsatellites.
However, when considering all the samples analysed, only 22.7 % were correctly genotyped (16.08%
could not be assigned to species level, 33% were identified as M. foina, 8% did not pass the microsatellite
screening, and the remaining 19.08 % were not fully genotyped). These results reveal the extent of the
effort needed to develop non-invasive individual identification studies in species of the genus Martes with
extensive overlapping areas. These observations could be extrapolated to non-invasive studies of other
sympatric carnivores.
The difference in the level of expertise among sample collectors potentially impacts the results of
microsatellite genotyping, while there are no significant differences in species identification rate. Noninvasive genetic studies may benefit from the help of different sample collectors with differences in the
level of expertise, as exemplified in Bellemain et al. (2005), Jacob et al. 2010 or in our present study.
However, in order to maximize non-invasive sampling and DNA quality which will insure accurate
genotyping success, we recommend, whenever possible, that sampling be conducted by experienced
biologists.
Application of our methods to field collected scats can be used in a cost-effective way to investigate species
biology (including distribution, patterns of genetic diversity, relatedness and population connectivity) as
well as to estimate population abundance for sympatric martens in the wild.
- 132 -
Paper IV
RELIABLE FAECAL DNA GENOTYPING OF SYMPATRIC MARTENS
ACKNOWLEDGEMENTS
A. Ruiz-González holds a Ph.D. fellowship awarded by the Dept. of Education, Universities and Research of the
Basque Government (Ref. BFI09.396). This study has been partially funded by the Basque Government through the
Research group on “Systematics, Biogeography and Population Dynamics” (Ref. IT317-10; GIC10/76) and by the
University of the Basque Country (UPV-EHU) and the Department of Environment, Territorial Planning,
Agriculture and Fisheries (Basque Country Government) through IKT S.A under University-Enterprise research
program (Ref. UE07/02). The authors wish to thank the following persons and institutions for supplying tissue and
faecal samples: Dr. F. Urra (VRFN-Navarre Government); Dr. I. Barja and his research group (UAM); Alfonso
Hartasanchez (FAPAS); the technical staff of the National Parks of Ordesa and Monte Perdido (E. Villagrasa), Picos
de Europa (A. Mora) and Aigüestortes i Estany de Saint Mauricy (J. Canut), and the Natural Parks of Alava (DFA);
Museo de Ciencias Naturales de Álava; F. López-Giraldez (DNA and Tissue Collection, Pompeu Fabra University),
P. Lizarraga and L. Lorza (CRF Martioda-DFA); J.M. Fernandez (IAN); N. Ruiz de Azua; G. Belamendia (CEA); J.
Pinedo (DFA); H. Aguirre, P. Pérez ; G. Dominguez.; F. Canales & M.A. Campos (CRN); E. Arberas; Oskar
Berdión, M. Corral; J. López de Luzuriaga; I. Aginako; H. Aguirre; P. Pérez; J. Ruiz Guijarro; I. Martínez; E. Díaz;
J.M. Pérez de Ana; D. Vado; A. Eguia; A. García de Albéniz; I. Zuberogoitia; Dr. M. Barral (NEIKER).
REFERENCES
Adams JR, Waits LP (2007) An efficient method for screening faecal DNA genotypes and detecting new individuals
and hybrids in the red wolf (Canis rufus) experimental population area. Conservation Genetics, 8, 123-131.
Arrendal J, Vila C, Bjorklund M (2007) Reliability of noninvasive genetic census of otters compared to field
censuses. Conservation Genetics, 8, 1097-1107.
Barja I (2005) Winter distribution of European pine marten (Martes martes) scats in a protected area of Galicia,
Spain. Mammalia, 69, 435-438.
Beaumont M, Barratt EM, Gottelli D, Kitchener AC, Daniels MJ, Pritchard JK, Bruford MW (2001) Genetic
diversity and introgression in the Scottish wildcat. Molecular Ecology, 10, 319-336.
Beja-Pereira A, Oliveira R, Alves PC, Schwartz MK, Luikart G (2009) Advancing ecological understandings through
technological transformations in noninvasive genetics. Molecular Ecology Resources, 9, 1279-1301.
Belkhir K, Borsa P, Chikhi L, Raufaste N, Bonhomme F (2000) Genetix 4.02, Logiciel sous windows pour la génétique
des populations. Laboratoire Génome, Populations, Interactions, Université de Montpellier II, Montpellier,
France.
Bellemain E, Swenson JE, Tallmon O, Brunberg S, Taberlet P (2005) Estimating population size of elusive animals
with DNA from hunter-collected feces: Four methods for brown bears. Conservation Biology, 19, 150-161.
Bellemain E, Taberlet P (2004) Improved noninvasive genotyping method: application to brown bear (Ursus arctos)
faeces. Molecular Ecology Notes, 4, 519-522.
Bijlsma R, Van de Vliet M, Pertoldi C, Van Apeldoorn RC, Van de Zande L (2000) Microsatellite primers from the
Eurasian badger, Meles meles. Molecular Ecology, 9, 2216-2217.
Birks JDS, Messenger JE, Braithwaite TC, Davison A, Brookes RC, Strachan C (2004) Are scat surveys a reliable
method for assessing distribution and population status of pine martens? In: Harrison, D.J., Fuller, A.K. &
Proulx, G (eds) Martens and fishers (Martes) in human-altered environments: an inter- national
perspective.New York: Springer-Verlag, pp 235–252.
Bonin A, Bellemain E, Eidesen PB, Pompanon F, Brochmann C, Taberlet P (2004) How to track and assess
genotyping errors in population genetics studies. Molecular Ecology, 13, 3261-3273.
Brinkman TJ, Schwartz MK, Person DK, Pilgrim KL, Hundertmark KJ (2010) Effects of time and rainfall on PCR
success using DNA extracted from deer fecal pellets. Conservation Genetics, 11, 1547-1552.
- 133 -
CHAPTER
4
Broquet T, Menard N, Petit E (2007) Noninvasive population genetics: a review of sample source, diet, fragment
length and microsatellite motif effects on amplification success and genotyping error rates. Conservation
Genetics, 8, 249-260.
Broquet T, Petit E (2004) Quantifying genotyping errors in noninvasive population genetics. Molecular Ecology, 13,
3601-3608.
Brownstein MJ, Carpten JD, Smith JR (1996) Modulation of non- templated nucleotide addition by Taq DNA
polymerase: primer modifications that facilitate genotyping. Biotechniques 20, 1004–1010
Cabria MT, Gonzalez EG, Gomez-Moliner BJ, Zardoya R (2007) Microsatellite markers for the endangered
European mink (Mustela lutreola) and closely related mustelids. Molecular Ecology Notes, 7, 1185-1188.
Creel S, Spong G, Sands JL, Rotella J, Zeigle J, Joe L, Murphy KM, Smith D (2003) Population size estimation in
Yellowstone wolves with error-prone noninvasive microsatellite genotypes. Molecular Ecology, 12, 20032009.
Dallas JF, Coxon KE, Sykes T, Chanin PRF, Marshall F, Carss DN, Bacon PJ, Piertney SB, Racey PA (2003)
Similar estimates of population genetic composition and sex ratio derived from carcasses and faeces of
Eurasian otter Lutra lutra. Molecular Ecology, 12, 275-282.
Dallas JF, Piertney SB (1998) Microsatellite primers for the Eurasian otter. Molecular Ecology, 7, 1248-1251.
Davis CS, Strobeck C (1998) Isolation, variability, and cross-species amplification of polymorphic microsatellite loci
in the family Mustelidae. Molecular Ecology, 7, 1776-1778.
Davison A, Birks JDS, Brookes RC, Braithwaite TC, Messenger JE (2002) On the origin of faeces: morphological
versus molecular methods for surveying rare carnivores from their scats. Journal of Zoology, 257, 141-143.
Davison A, Birks JDS, Brookes RC, Messenger JE, Griffiths HI (2001) Mitochondrial phylogeography and
population history of pine martens Martes martes compared with polecats Mustela putorius. Molecular
Ecology, 10, 2479-2488.
Davison A, Chiba S (2003) Laboratory temperature variation is a previously unrecognized source of genotyping error
during capillary electrophoresis. Molecular Ecology Notes, 3, 321-323.
Domingo-Roura X, Macdonald DW, Roy MS, Marmi J, Terradas J, Woodroffe R, Burke T, Wayne RK (2003)
Confirmation of low genetic diversity and multiple breeding females in a social group of Eurasian badgers
from microsatellite and field data. Molecular Ecology, 12, 533-539.
Falush D, Stephens M, Pritchard JK (2003) Inference of population structure using multilocus genotype data:
Linked loci and correlated allele frequencies. Genetics, 164, 1567-1587.
Falush D, Stephens M, Pritchard JK (2007) Inference of population structure using multilocus genotype data:
dominant markers and null alleles. Molecular Ecology Notes, 7, 574-578.
Ferrando A, Lecis R, Domingo-Roura X, Ponsa M (2008) Genetic diversity and individual identification of
reintroduced otters (Lutra lutra) in north-eastern Spain by DNA genotyping of spraints. Conservation
Genetics, 9, 129-139.
Flagstad O, Hedmark E, Landa A, Broseth H, Persson J, Andersen R, Segerstrom P, Ellegren H (2004) Colonization
history and noninvasive monitoring of a reestablished wolverine population. Conservation Biology, 18, 676688.
Flagstad O, Roed K, Stacy JE, Jakobsen KS (1999) Reliable noninvasive genotyping based on excremental PCR of
nuclear DNA purified with a magnetic bead protocol. Molecular Ecology, 8, 879-883.
Fleming MA, Ostrander EA, Cook JA (1999) Microsatellite markers for American mink (Mustela vison) and ermine
(Mustela erminea). Molecular Ecology, 8, 1352-1354.
Frantz AC, Pope LC, Carpenter PJ, Roper TJ, Wilson GJ, Delahay RJ, Burke T (2003) Reliable microsatellite
genotyping of the Eurasian badger (Meles meles) using faecal DNA. Molecular Ecology, 12, 1649-1661.
Gagneux P, Boesch C, Woodruff DS (1997) Microsatellite scoring errors associated with noninvasive genotyping
based on nuclear DNA amplified from shed hair. Molecular Ecology, 6, 861-868.
Gomez-Moliner BJ, Cabria MT, Rubines J, Garin I, Madeira MJ, Elejalde A, Aihartza J, Fournier P, Palazon S
(2004) PCR-RFLP identification of mustelid species: European mink (Mustela lutreola), American mink
(M.vison) and polecat (M.putorius) by analysis of excremental DNA. Journal of Zoology, 262, 311-316.
Goossens B, Chikhi L, Utami SS, Ruiter Jd et al (2000) A multi- samples, multi-extracts approach for microsatellite
analysis of faecal samples in an arboreal ape. Conservation Genetics 1:157–162
- 134 -
Paper IV
RELIABLE FAECAL DNA GENOTYPING OF SYMPATRIC MARTENS
Goszczynski J, Posluszny M, Pilot M, Gralak B (2007) Patterns of winter locomotion and foraging in two sympatric
marten species: Martes martes and Martes foina. Canadian Journal of Zoology-Revue Canadienne De Zoologie,
85, 239-249.
Goudet J (1995) Fstat (version 1.2): a computer program to calculate F-statistics. Journal of Heredity, 86, 485–486.
Goudet J (2001) FSTAT, A program to estimate and test gene diversities and fixation indices. Version 2.9.3, updated
from Goudet 1995. Available from http://www.unil.ch/izea/softwares/fstat.html
Grakov NN (1994) Kidus – a hybrid of the sable and the pine marten. Lutreola, 1, 1–4.
Guo SW, Thompson EA (1992) Performing the exact test of hardy-weinberg proportion for multiple alleles.
Biometrics, 48, 361-372.
Hajkova P, Zemanova B, Bryja J, Hajek B, Roche K, Tkadlec E, Zima J (2006) Factors affecting success of PCR
amplification of microsatellite loci from otter faeces. Molecular Ecology Notes, 6, 559-562.
Hajkova P, Zemanova B, Roche K, Hajek B (2009) An evaluation of field and noninvasive genetic methods for
estimating Eurasian otter population size. Conservation Genetics, 10, 1667-1681.
Hedmark E, Ellegren H (2007) DNA-based monitoring of two newly founded Scandinavian wolverine populations.
Conservation Genetics, 8, 843-852.
Herr J, Schley L, Roper TJ (2009) Socio-spatial organization of urban stone martens. Journal of Zoology, 277, 54-62.
Hung CM, Li SH, Lee LL (2004) Faecal DNA typing to determine the abundance and spatial organisation of otters
(Lutra lutra) along two stream systems in Kinmen. Animal Conservation, 7, 301-311.
Jacob G, Debrunner R, Gugerli F, Schmid B, Bollmann K (2010) Field surveys of capercaillie (Tetrao urogallus) in
the Swiss Alps underestimated local abundance of the species as revealed by genetic analyses of non-invasive
samples. Conservation Genetics, 11, 33-44.
Janssens X, Fontaine MC, Michaux JR, Libois R, de Kermabon J, Defourny P, Baret PV (2008) Genetic pattern of
the recent recovery of European otters in southern France. Ecography, 31, 176-186.
Jordan MJ, Higley M, Matthews SM, Rhodes OE, Schwartz MK, Barrett RH, Palsboll PJ (2007) Development of
22 new microsatellite loci for fishers (Martes pennanti) with variability results from across their range.
Molecular Ecology Notes, 7, 797-801.
Kalz B, Jewgenow K, Fickel J (2006) Structure of an otter (Lutra lutra) population in Germany - results of DNA and
hormone analyses from faecal samples. Mammalian Biology, 71, 321-335.
Koepfli KP, Deere KA, Slater GJ, Begg C, Begg K, Grassman L, Lucherini M, Veron G, Wayne RK (2008)
Multigene phylogeny of the Mustelidae: Resolving relationships, tempo and biogeographic history of a
mammalian adaptive radiation. Bmc Biology, 6.
Kyle CJ, Davison A, Strobeck C (2003) Genetic structure of European pine martens (Martes martes), and evidence
for introgression with M-americana in England. Conservation Genetics, 4, 179-188.
Lanszki J, Hidas A, Szentes K, Revay T, Lehoczky I, Weiss S (2008) Relative spraint density and genetic structure of
otter (Lutra lutra) along the Drava River in Hungary. Mammalian Biology, 73, 40-47.
Livia L, Francesca V, Antonella P, Fausto P, Bernardino R (2007) A PCR-RFLP method on faecal samples to
distinguish Martes martes, Martes foina, Mustela putorius and Vulpes vulpes. Conservation Genetics, 8, 757759.
López-Martín JM (2007) Martes martes (Linnaeus, 1758). In: Palomo LJ, Gisbert J, Blanco JC (Eds.) Atlas y Libro Rojo
de los Mamíferos Terrestres de España. Dirección General de Biodiversidad-SECEM-SECEMU, Madrid, pp
302–304
Lucchini V, Fabbri E, Marucco F, Ricci S, Boitani L, Randi E (2002) Noninvasive molecular tracking of colonizing
wolf (Canis lupus) packs in the western Italian Alps. Molecular Ecology, 11, 857-868.
Marmi J, Lopez-Giraldez JF, Domingo-Roura X (2004) Phylogeny, evolutionary history and taxonomy of the
Mustelidae based on sequences of the cytochrome b gene and a complex repetitive flanking region.
Zoologica Scripta, 33, 481-499.
Marucco F, Pletscher DH, Boitani L, Schwartz MK, Pilgrim KL, Lebreton JD (2009) Wolf survival and population
trend using non-invasive capture-recapture techniques in the Western Alps. Journal of Applied Ecology, 46,
1003-1010.
Maudet C, Luikart G, Dubray D, Von Hardenberg A, Taberlet P (2004) Low genotyping error rates in wild
ungulate faeces sampled in winter. Molecular Ecology Notes, 4, 772-775.
Mergey M (2007). Réponses des populations de martres d’Europe (Martes martes) à la fragmentation de l'habitat :
mécanismes comportementaux et consequences. Dissertation University of Reims Campagne-Ardenne, France.
- 135 -
CHAPTER
4
Mills LS, Citta JJ, Lair KP, Schwartz MK, Tallmon DA (2000) Estimating animal abundance using noninvasive
DNA sampling: Promise and pitfalls. Ecological Applications, 10, 283-294.
Mullins J, Statham MJ, Roche T, Turner PD, O'Reilly C (2010) Remotely plucked hair genotyping: a reliable and
non-invasive method for censusing pine marten (Martes martes, L. 1758) populations. European Journal of
Wildlife Research, 56, 443-453.
Murphy MA, Kendall KC, Robinson A, Waits LP (2007) The impact of time and field conditions on brown bear
(Ursus arctos) faecal DNA amplification. Conservation Genetics, 8, 1219-1224.
Murphy MA, Waits LP, Kendall C (2000) Quantitative evaluation of fecal drying methods for brown bear DNA
analysis. Wildlife Society Bulletin, 28, 951-957.
Murphy MA, Waits LP, Kendall KC (2003) The influence of diet on faecal DNA amplification and sex
identification in brown bears (Ursus arctos). Molecular Ecology, 12, 2261-2265.
Murphy MA, Waits LP, Kendall KC, Wasser SK, Higbee JA, Bogden R (2002) An evaluation of long-term
preservation methods for brown bear (Ursus arctos) faecal DNA samples. Conservation Genetics, 3, 435-440.
O´Connell M, Wright JM, Farid A (1996) Development of PCR primers for nine polymorphic American mink
Mustela vison microsatellite loci. Molecular Ecology, 5, 311-312.
O’Reilly C, Statham MJ, Mullins J, Turner PD, O’Mahony D (2008) Efficient species identification of pine marten
(Martes martes)and red fox (Vulpes vulpes) scats using a 5′ nuclease real-time PCR assay. Conservation
Genetics, 9, 735–738.
Oliveira R, Godinho R, Randi E, Ferrand N, Alves PC (2008) Molecular analysis of hybridisation between wild and
domestic cats (Felis silvestris) in Portugal: implications for conservation. Conservation Genetics, 9, 1-11.
Pertoldi C, Barker SF, Madsen AB, Jorgensen H, Randi E, Munoz J, Baagoe HJ, Loeschcke V (2008) Spatiotemporal population genetics of the Danish pine marten (Martes martes). Biological Journal of the Linnean
Society, 93, 457-464.
Piggott MP (2004) Effect of sample age and season of collection on the reliability of microsatellite genotyping of
faecal DNA. Wildlife Research, 31, 485-493.
Piggott MP, Bellemain E, Taberlet P, Taylor AC (2004) A multiplex pre-amplification method that significantly
improves microsatellite amplification and error rates for faecal DNA in limiting conditions. Conservation
Genetics, 5, 417-420.
Piggott MP, Taylor AC (2003) Remote collection of animal DNA and its applications in conservation management
and understanding the population biology of rare and cryptic species. Wildlife Research, 30, 1-13.
Pompanon F, Bonin A, Bellemain E, Taberlet P (2005) Genotyping errors: Causes, consequences and solutions.
Nature Reviews Genetics, 6, 847-859.
Prigioni C, Remonti L, Balestrieri A, Sgrosso S, Priore G, Mucci N, Randi E (2006) Estimation of european otter
(Lutra lutra) population size by fecal DNA typing in southern Italy. Journal of Mammalogy, 87, 855-858.
Pritchard JK, Stephens M, Donnelly P (2000) Inference of population structure using multilocus genotype data.
Genetics, 155, 945-959.
Proulx G, Aubry KB, Birks J, Buskirk SW, Fortin C, Frost HC, Krohn WB, Mayo L, Monakhov V, Payer D, Saeki
M, Santos-Reis M, Weir R, Zielinski WJ (2004) World distribution and status of the genus Martes in
2000. In: Harrison DJ, Fuller AK, Proulx G (Eds) Martens and fishers (Martes) in human- altered
environments: an international perspective. New York: Springer-Verlag, pp 21–76
Prugh LR, Ritland CE (2005) Molecular testing of observer identification of carnivore feces in the field. Wildlife
Society Bulletin, 33, 189-194.
Raymond M, Rousset F (1995) Genepop (Version-1.2) - Population-Genetics Software for Exact Tests and
Ecumenicism. Journal of Heredity, 86, 248-249.
Rice WR (1989) Analyzing tables of statistical tests. Evolution, 43, 223–225
Rosellini S, Osorio E, Ruiz-Gonzalez A, Isabel AP, Barja I (2008) Monitoring the small-scale distribution of
sympatric European pine martens (Martes martes) and stone martens (Martes foina): a multievidence
approach using faecal DNA analysis and camera-traps. Wildlife Research, 35, 434-440.
Rousset F (2008) GENEPOP ' 007: a complete re-implementation of the GENEPOP software for Windows and
Linux. Molecular Ecology Resources, 8, 103-106.
Ruiz-Gonzalez A, Rubines J, Berdion O, Gomez-Moliner BJ (2008) A non-invasive genetic method to identify the
sympatric mustelids pine marten (Martes martes) and stone marten (Martes foina): preliminary distribution
survey on the northern Iberian Peninsula. European Journal of Wildlife Research, 54, 253-261.
- 136 -
Paper IV
RELIABLE FAECAL DNA GENOTYPING OF SYMPATRIC MARTENS
Santini A, Lucchini V, Fabbri E, Randi E (2007) Ageing and environmental factors affect PCR success in wolf (Canis
lupus) excremental DNA samples. Molecular Ecology Notes, 7, 955-961.
Sato JJ, Hosoda T, Wolsan M, Tsuchiya K, Yamamo M, Suzuki H (2003) Phylogenetic relationships and divergence
times among mustelids (Mammalia : Carnivora) based on nucleotide sequences of the nuclear
interphotoreceptor retinoid binding protein and mitochondrial cytochrome b genes. Zoological Science, 20,
243-264.
Stenglein JL, De Barba M, Ausband DE, Waits LP (2010) Impacts of sampling location within a faeces on DNA
quality in two carnivore species. Molecular Ecology Resources, 10, 109-114.
Stone KD, Cook JA (2002) Molecular evolution of Holarctic martens (genus Martes, Mammalia : Carnivora :
Mustelidae). Molecular Phylogenetics and Evolution, 24, 169-179.
Schwartz MK, Monfort SL (2008) Genetic and endocrine tools for carnivore surveys. In: Long RA, Mackay P,
Zielinski WJ, Ray JC (eds) Noninvasive survey methods for carnivores. Island Press, Washington DC, pp 238–
262
Taberlet P, Griffin S, Goossens B, Questiau S, Manceau V, Escaravage N, Waits LP, Bouvet J (1996) Reliable
genotyping of samples with very low DNA quantities using PCR. Nucleic Acids Research, 24, 3189-3194.
Taberlet P, Luikart G (1999) Non-invasive genetic sampling and individual identification. Biological Journal of the
Linnean Society, 68, 41-55.
Taberlet P, Waits LP, Luikart G (1999) Noninvasive genetic sampling: look before you leap. Trends in Ecology &
Evolution, 14, 323-327.
Valiere N (2002) GIMLET: a computer program for analysing genetic individual identification data. Molecular
Ecology Notes, 2, 377-379.
Valiere N, Bonenfant C, Toigo C, Luikart G, Gaillard JM, Klein F (2007) Importance of a pilot study for noninvasive genetic sampling: genotyping errors and population size estimation in red deer. Conservation
Genetics, 8, 69-78.
Van Oosterhout C, Hutchinson WF, Wills DPM, Shipley P (2004) MICRO-CHECKER: software for identifying
and correcting genotyping errors in microsatellite data. Molecular Ecology Notes, 4, 535-538.
Waits JL, Leberg PL (2000) Biases associated with population estimation using molecular tagging. Animal
Conservation, 3, 191-199.
Waits LP, Paetkau D (2005) Noninvasive genetic sampling tools for wildlife biologists: A review of applications and
recommendations for accurate data collection. Journal of Wildlife Management, 69, 1419-1433.
Walker CW, Vila C, Landa A, Linden M, Ellegren H (2001) Genetic variation and population structure in
Scandinavian wolverine (Gulo gulo) populations. Molecular Ecology, 10, 53-+.
Wasser SK, Houston CS, Koehler GM, Cadd GG, Fain SR (1997) Techniques for application of faecal DNA
methods to field studies of Ursids. Molecular Ecology, 6, 1091-1097.
Wehausen JD, Ramey RR, Epps CW (2004) Experiments in DNA extraction and PCR amplification from bighorn
sheep feces: the importance of DNA extraction method. Journal of Heredity, 95, 503-509.
Wilson GJ, Frantz AC, Pope LC, Roper TJ, Burke TA, Cheeseman CL, Delahay RJ (2003) Estimation of badger
abundance using faecal DNA typing. Journal of Applied Ecology, 40, 658-666.
Zuercher GL, Gipson PS, Stewart GC (2003) Identification of carnivore feces by local peoples and molecular
analyses. Wildlife Society Bulletin, 31, 961-970.
- 137 -
CHAPTER
4
- 138 -
CHAPTER
1
Landscape genetics
PAPER V
Non-invasive landscape genetics of the European pine marten
(Martes martes): assessing spatial genetic structure and
distribution in a heterogeneous landscape
PAPER VI
Landscape genetics as a tool for the empirical assessment of a
regional ecological network: The European pine marten
(Martes martes) as a target-species
- 139 -
5
CHAPTER
V
Landscape Genetics
- 140 -
Paper V
Spatial genetic structure
PAPER V
Non-invasive landscape genetics of the European pine marten
(Martes martes): assessing spatial genetic structure and
distribution in a heterogeneous landscape
ABSTRACT
Non-invasive genetic sampling in combination with landscape genetics provides a valuable framework for
the study of rare and elusive species in order to understand how landscape features influence population
genetic structure. In this study, we used non-invasive methods to assess the distribution of sympatric
martens (Martes martes and M. foina) and the spatial genetic structure of the forest dwelling European pine
marten (M. martes) inhabiting a fragmented landscape, in the northern Iberian Peninsula. Out of 798
faeces samples analysed by a PCR-RFLP method, we identified 323 stone martens and 347 pine martens,
which allowed us to determine species distribution. We used a number of different analytical approaches
for identifying spatial genetic structure and to analyse patterns of gene flow for the 140 individual pine
marten genotypes identified by multilocus microsatellite analyses of 15 loci. Results from spatial and nonspatial Bayesian clustering methods were mainly concordant and detected the presence of different genetic
clusters distributed across the study area with non-overlapping distribution. The data reported in this
study suggested that habitat loss and fragmentation could be one of the main causes of the spatial genetic
structure found in a typical forest dwelling species. Other factors, such as, direct persecution, probable
interspecific competition with the stone marten and the fact that the study area is situated at the limit of
the species distribution, could be acting synergically to shape the current spatial genetic structure.
Keywords: Landscape genetics, Non-invasive genetic sampling, Martes martes, Bayesian analysis, Faecal
DNA
- 141 -
CHAPTER
V
Landscape Genetics
INTRODUCTION
Landscape genetics has emerged as a synthetic discipline which integrates population genetics, landscape
ecology and spatial statistics (Manel et al. 2003; Storfer et al. 2007; Holderegger & Wagner 2008).
Landscape genetics explicitly quantifies the effects of landscape composition, configuration, and matrix
quality on spatial patterns in neutral and adaptive genetic variation and underlying microevolutionary
processes (Manel et al. 2003; Storfer et al. 2007; Holderegger and Wagner 2008; Balkenhol et al. 2009).
All species are to some extent influenced by spatially heterogeneous landscapes (Segelbacher et al. 2010).
Spatial heterogeneity affects, among other things, dispersal and, consequently, gene flow (Holderegger et
al. 2006). An important issue is whether changes in landscape features create barriers to gene flow and as
such lead to population structure. Multiple factors that act at various spatial scales affect spatial
connectivity and rates of gene flow across a landscape (Anderson et al. 2010). These factors can include
biotic (e.g. forest structure, landscape uses, and presence of prey, predators or sympatric competitors),
abiotic (e.g. rivers, topography, environmental conditions) and/or anthropogenic features (e.g. roads,
urban areas, crops). Thus, emergent patterns of gene flow are the result of the interactions between
structural landscape connectivity and how organisms respond to landscape structure (Manel et al. 2003).
In this context, assessing genetic population structure is crucial for defining appropriate conservation and
management units (Frankham et al. 2002; Palsbøll et al. 2007), and for maintaining gene flow and genetic
diversity (Moritz 2002).
Until recently, genetic structuring was studied by defining populations on an a priori basis, and gene flow
was measured mainly by FST or similar parameters (Pearse & Crandall, 2004). However, landscape
genetics has shifted towards individual-based sampling and analysis, especially when organisms are
continuously distributed (Manel et al. 2007; Clark et al. 2008; Frantz et al. 2009; Cushman & Landguth
2010a,b; Segelbacher et al. 2010; Storfer et al. 2010). Recent landscape genetic approaches largely focus
on describing and mapping populations (e.g. Guillot et al. 2005; Dionne et al. 2008) and on identifying
factors that influence rates and patterns of gene flow within and between populations (e.g. McRae et al.
2005; Coulon et al. 2006; Cushman et al. 2006; McRae and Beier 2007; McCairns & Bernatchez 2008;
Schwartz et al. 2009), and deducing from these which features restrict or promote movements of
individuals. Bayesian clustering algorithms (Pritchard et al. 2000; Guillot et al. 2005; Francois & Durand
20010) have become popular tools for characterizing population genetic structure based on individual
genotypes. These methods probabilistically assign individuals to groups based on their multi-locus
genotypes by minimizing Hardy–Weinberg and linkage disequilibria, without presuming pre-defined
populations (Pearse & Crandall 2004). A recent extension of these methods addresses the spatial nature of
- 142 -
Paper V
Spatial genetic structure
the problem of locating genetic discontinuities by including the geographical coordinates of individuals in
their prior distributions (Francois & Durand 2010). These models offer a powerful tool to answer
questions in ecology, conservation and wildlife management, as genetic discontinuities within populations
can be correlated with landscape features (Manel et al. 2003). These methods lead to more meaningful
population delineations, and they are particularly useful for describing population structure in highly
mobile or continuously distributed species, such as large and midsized carnivores (e.g. Cegelski et al. 2003;
McRae et al. 2005; Pilot et al. 2006; Riley et al. 2006; Millions & Swanson 2007; Gula et al. 2009;
Zalewski et al. 2009; Mucci et al. 2010; Williams & Scribner 2010). However, sufficient sample collection
for these purposes is a difficult task, especially in rare and elusive species, in which sampling is a limiting
factor (Schwartz & Monfort 2008). In this context, molecular methods incorporating non-invasive
sampling via the collection of scats or hairs have become common for population monitoring of
carnivores, providing valuable DNA sources for genetic surveys (Taberlet & Luikart 1999; Piggott &
Taylor 2003; Waits & Paetkau 2005; Schwartz & Monfort 2008; Beja-Pereira et al. 2009).
The pine marten (Martes martes L. 1758) is a typical woodland-dwelling mustelid that occurs throughout
most of Europe and northern and central Asia, extending from northern Portugal to western Siberia
(Proulx et al. 2004). The pine marten is generally associated with forest habitats, mainly mature coniferous
and mixed forests (Delibes 1983, Buskirk 1992, Proulx et al. 2004). Nonetheless, they have also been
recently reported in fragmented landscapes where woods consist of isolated, small fragments within an
agricultural landscape matrix (Pereboom et al. 2008; Balestrieri et al. 2010). Deforestation and forest
fragmentation have been reported to affect the distribution and density of pine martens (Brainerd et al.
1994, Kurki et al. 1998), which are believed to need a minimum woodland area to survive (Zalewski and
Jędrzejewski 2006) and tend to avoid treeless areas (Storch et al. 1990, Brainerd and Rolstad 2002,
Pereboom et al. 2008, Ruiz-González et al. 2008). Given their strong associations with structural
complexity in forests, the species is particularly sensitive to human influences on their habitats, including
habitat loss, and landscape-scale effects of habitat fragmentation (Brainerd 1990; Bright 1993; Pereeboom
et al. 2008). Consequently, the pine marten is a well suited species for carrying out studies focused on the
effects of forest fragmentation on genetic structure and gene flow. However, little is known about the
genetic population structure (Kyle et al. 2003) and the influence of habitat fragmentation on pine marten
gene-flow (Mergey 2007). Previous studies focused on the pine marten have given information about the
genetic variability and structure of several European populations (Kyle et al. 2003; Mergey 2007; Pertoldi
et al. 2008; Mullins et al. 2010). However, information about the genetic variability and population
structure of the Iberian populations found at the southern edge of the species distribution is negligible.
- 143 -
CHAPTER
V
Landscape Genetics
Previous studies focused on forest dwelling species (e.g. Roe deer, Coulon et al. 2004, Coulon et al. 2006;
American marten, Broquet et al. 2006; Wasserman et al. 2010) supported the hypothesis that the
landscape has a significant influence on the structuring of the populations under study. Indeed, recent
works based on a least cost distance approach, indicated that several landscape features act as moderators of
gene flow because of a high resistance to pine marten movements (Mergey 2008; Ruiz-Gonzalez et al. in
prep.). Thus, this study could provide a new insight and help to validate previous results obtained in forest
associated species.
Terrestrial organisms respond to complex landscape structure at their own unique set of spatial and
temporal scales (Anderson et al. 20010), based on their inherent dispersal abilities and sensitivity to
environmental change (D’Eon et al. 2002). Landscape spatial heterogeneity not only influences, directly or
indirectly, landscape connectivity (Spear et al. 2010), but also the probability of site occupancy
(Goszczyński et al. 2007; Ruiz-González et al. 2008). Moreover, the spatial organization of a species in a
landscape is influenced, at least in part, by the presence of sympatric competitors (Linnell & Strand 2000).
In this context, the interference interactions between sympatric carnivores can potentially influence
population genetic structure as they may compete for the same resources and consequently synergically
increase the effects of other biotic or abiotic factors. Thus, the probable interspecific competition between
the pine marten and the more widespread sympatric stone marten (Martes foina) may influence pine
marten distribution range (Delibes 1983; Goszczyński et al. 2007) and hence affect population structure.
Consequently, prior to conducting any landscape genetic study focused on the pine marten via noninvasive sampling of scats in sympatric areas with the stone marten, a distribution assessment through the
application of a species identification genetic method is an essential previous step (Ruiz-González et al.
2008). Although some studies on the distribution of sympatric martens have been conducted in the
Iberian Peninsula (Rosellini et al. 2008; Ruiz-González et al. 2008), there is a remarkable absence of
information on the distribution of the European pine marten and on the bio-ecological relationships
between these species (Proulx et al. 2004). The north Iberian Peninsula, houses the south-western
distribution limits of a number of temperate forest species (e.g. Glis glis, Myodes glareoulus) including the
focal pine marten (Palomo et al. 2007). Thus, this area has remarkable conservation priorities for the pine
marten and for biodiversity conservation in general, as distributional limits are thought to be particularly
important as long-term stores of genetic diversity and hot spots for speciation (Hampe & Petit 2005).
Moreover, the study area plays an important connector role constituting a natural link between the
Cantabrian Mountains and the Pyrenees and is considered of strategic importance for the conservation of
ecological connectivity in south-western Europe (Worboys et al. 2010).
- 144 -
Paper V
Spatial genetic structure
Thus, the main aims of our study were to i) first, determine the spatial distribution of pine marten and
stone marten and infer sympatric relationships between both species through non-invasive genetic
sampling of scats. As pine marten faeces cannot be distinguished from those of the sympatric stone marten
(Martes foina), which is widespread in the study area (López-Martín, 2008), and can also be easily
confused with those of other carnivores (Davison et al. 2002; Harrington et al. 2010), molecular
techniques were applied for the identification of faecal samples (Ruiz-González et al. 2008). Our second
objective was to ii) characterize the spatial genetic structure of the European pine marten using 15 highly
variable genetic markers on previously PCR-RFLP indentified faecal DNA. Therefore, we inferred
population structure in the pine marten by comparing two genetic assignment approaches that use a
Bayesian statistical framework to delineate genotype clusters without the need to infer gene flow with a
priori subjective groupings. Complementarily, spatial autocorrelation analyses were used to infer spatial
genetic structure. Barriers to gene flow among populations were tested combining landscape derived data
with GIS. This study will allow us to detect which spatial and landscape features influence pine marten
gen-flow and should be considered as potential barriers for the migration movements of individuals, with
particular respect to the differentiation of management units.
As habitat specialist species are usually considered highly vulnerable to habitat fragmentation (Bright
1993; Devictor et al. 2008) we predicted that, as a forest dwelling species, pine martens would be better
represented in forested areas and would present spatial genetic structure due to the fragmented nature of
the landscape under study.
MATERIAL AND METHODS
Study area
The Basque Country and Navarre are located in the northern Iberian Peninsula, bordering the Cantabrian
Sea and sit between the mountain ranges of the Pyrenees and the Cantabrian Mountains (Figure 1). They
belong to the Atlantic and Mediterranean biogeographic regions, with the exception of the Northeastern
area of Navarre, which belongs to the Alpine region.
The Basque Country comprises an area of 7,235 km2 and has a high population density of 298 inhabitants
per square kilometer (INE, 2008) and a dense road network. The area is still covered by remnant zones of
natural and semi-natural vegetation composed of deciduous oak and beech forests. However, extensive
industrial development and increasing urbanization together with an increased density in communication
infrastructures have given rise to a heterogeneous landscape characterized by a clear pattern of forest loss
- 145 -
CHAPTER
V
Landscape Genetics
and fragmentation. Natural forest covers approximately 28% of its area, forestry plantations 29%, nonwooded mountains 24%, cultivated land 14%, and urban land and infrastructures 5.7%. Contrastingly,
Navarre comprises an area of 10,390 km2 and has a low population density of 60 inhabitants per square
kilometer (INE, 2008). The area is covered by a non-fragmented natural forest system, concentrated in the
North, while cultivated land and urban areas are located in the South. Natural forest covers nearly 36% of
its area, forestry plantations 6%, non-wooded mountains 10%, cultivated land 46%, and urban land and
infrastructures 1.4%.
Non-invasive sample collection and DNA extraction
As the pine marten is an elusive, scarce and stress sensitive species (Barja et al. 2007), we obtained DNA
from non-invasive sampling of faeces. This DNA source is regularly used for non-invasive genetic studies
of carnivores (Schwartz & Monfort 2008; Beja-Pereira, 2009) and was used in previous pine marten
studies (Ruiz-González et al. 2008; Rosellini et al. 2009; Balestrieri et al. 2010). We conducted a multistage sampling scheme, in which samples from a pilot study were used to assess the appropriateness of the
sampling with respect to the research questions (Balkenhol 2009). Thus, two scat-based surveys were
conducted between 2004 and 2009 all over the sympatric range of both marten species in the study area
(López-Martín 2007). The first one, conducted in the 2004-2005 period was used to initially estimate the
distribution range of the two species of the genus Martes in the study area. The second one, conducted
between 2006 and 2009 was used to refine species distribution assessment and to obtain a higher number
of fresh samples of M. martes for microsatellite genotyping after a PCR-RFLP species identification
process. Fresh faeces were characterised by their strong smell, the presence of a mucus layer and lack of any
sign of dehydration. European pine martens, stone martens and other carnivores use forest roads and
frequently defecate on them as a means of visual-scent marking (Barja 2005). Thus, sampling was
conducted systematically along linear features, such as forest trails and paths. Morphological
characteristics, such as size and shape, were used with a view to distinguishing faeces of the genus Martes
from those of other medium-sized carnivores such as the red fox and the wildcat (Rosellini et al. 2008;
Ruiz-González et al. 2008). The territory of the European pine marten has been calculated to be nearly 1.3
km2 for males and 1.0 km2 for females in some forested areas of the study area (O. Berdión, pers. comm.).
The mean territory size of the stone marten is similar to that of the European pine marten, being about
0.96 km2 (López-Martín 2003). Thus, in order to minimize multiple sampling of the same individual and
to maximize the number of sampled individuals, samples were collected if separated by a spatial distance of
≥ 1km from the next nearest sample. Transects were uniformly distributed throughout the study area in an
- 146 -
Paper V
Spatial genetic structure
effort to survey all the territory, with special emphasis on forested areas. Universal Transversal Mercator
(UTM) coordinates were recorded for all the samples collected using a global positioning system (Garmin
eTtrex). The faecal samples were stored in autoclaved tubes containing ethanol 96% and frozen at −20°C
until processed (Gómez-Moliner et al. 2004; Ruiz González et al. 2008).
Sampling was carried out thanks to the collaboration of more than 30 volunteers and students, personnel
of the regional Natural parks and a specialized field biologist. The widespread study area and long-term
programme did not permit the standardization of sampling in space and time. Collaborators were trained
to collect only fresh samples (<5days) (Ruiz-González et al. 2008).
At each extraction step, special care was taken to avoid cross-contamination, which is an important
problem especially when handling DNA from non-invasive samples (Taberlet et al. 1999; Pompanon et al.
2005). Thus, extractions were performed in a dedicated laboratory on a sterile bench, using filter tips
throughout. To monitor potential contamination, we included one negative extraction control per
extraction process. DNA was then extracted using the DNA Stool MiniKit (Qiagen, Hombrechtikon,
Switzerland) following the manufacturer’s protocol for DNA extraction from stool samples. The DNA was
eluted from the silica membrane using twice 100 μL buffer AE (Qiagen).
Additionally, fresh tissue specimens from road-kill pine martens were included in the study, when
possible. DNA was isolated from tissues using the Qiagen DNeasy Tissue DNA (Qiagen, Hombrechtikon,
Switzerland) extraction kit according to the manufacturer’s instructions.
Species identification and spatial distribution
The specific identification of faecal samples was accomplished by a polymerase chain reaction – restriction
fragment length polymorphism (PCR-RFLP) method, providing for an effective genetic identification of
sympatric marten species (Ruiz-González et al. 2008). A small fragment (276 bp) of mitochondrial DNA
(mtDNA), D-loop region, was amplified with specific primers developed for Martes and Mustela by PCR.
The primers were designed to amplify small fragments to maximise the probability of amplification of
degraded DNA. Following DNA amplification, PCR products were digested with the restriction enzymes
HaeIII and RsaI. The combined use of both enzymes produced a species-specific banding pattern allowing
the scats of M. martes and M. foina to be discriminated (see Ruiz-González et al. 2008 for further details).
To assess the spatial distribution of both species in the study area the UTM coordinates corresponding to
each sample were projected onto a GIS (Arcview 9.0. ESRI) along with the species identification data
provided by the PCR-RFLP method. Finally, we characterized the location of each species with regards to
- 147 -
CHAPTER
V
Landscape Genetics
landscape uses. Landscape use data was obtained in vector format from the most recent Spanish forestry
map (Spanish Ministry of the environment, 2006) while the road network map was obtained from the
Spanish National Geographic Institute (2008).
DNA amplification and microsatellite genotyping
The extracted DNA was PCR-amplified and genotyped using 15 microsatellite primers (Ma1, Ma2, Ma19
Gg7, Davis and Strobeck 1998; Mel1, Bijlsma et al. 2000; Mel10, Domingo-Roura et al. 2003; MLUT27,
Cabria et al. 2007; Mvis072, Mer 41 Fleming et al. 1999; Mvi57 O’Connell et al. 1996; Lut 615 Dallas
& Piertney 1998; MP0059, MP0188, Jordan et al. 2007). These loci were selected by cross species
amplification of 41 loci and based on demonstrated variability in Iberian pine marten populations,
amplification strength with small quantities of DNA (faecal DNA), and powerful enough to allow
individual identification according to Ruiz-González et al. (in prep.).
Forward primers were fluorescently labelled with different dyes (6-FAM, NED, PET, VIC) to enable
multiplex electrophoresis of microsatellite products (Ruiz-González et al. in prep.). In order to avoid noise
from variable adenylation during the PCR, the ‘pigtail’ sequence GTTTCTT was added to the 5 -end of
each reverse primer (Brownstein et al. 1996). PCR multiplex amplifications were carried out with
QIAGEN Multiplex PCR kits using the manufacturer’s protocol in a total volume of 10 μL with 2 μL of
DNA and 2 pmol of each fluorescence labeled forward and unlabelled reverse primers. We applied a hotstart thermocycling protocol. The initial polymerase activation (HotStart PCR) was done at 95°C for 15
min, followed by 42 cycles (35 for tissue samples) of denaturation at 94°C for 30s, primer annealing at
57°C for 90s, and sequence extension at 72°C for 60s, and a final extension step at 60°C for 30 min.
In addition to the negative controls for extraction, negative PCR controls were included as proposed by
Pompanon et al. (2005). Additionally, we amplified a reference sample as a positive control and to test
that the electrophoretic mobility of the fragments was consistent across runs (Davison & Chiba 2003).
In non-invasive genetic sampling, genotyping errors occur due to increased rates of null alleles, allelic
dropout (ADO) and false alleles (FA) (Taberlet et al. 1996; Pompanon et al. 2005). Therefore, we
followed a modified multiple-tube approach from Taberlet et al. (1996), amplifying each DNA extract in
four replicates and in separate rooms dedicated to low DNA-content samples.
All the faecal samples identified by the PCR-RFLP method as pine marten were genotyped at 15 loci.
DNA quality was initially screened by PCR-amplifying each DNA sample four times at four loci
(Multiplex 1: MP0188; MP0059; Gg-7; Ma-1). Only samples showing >50% positive PCRs were further
- 148 -
Paper V
Spatial genetic structure
amplified four times at the remaining 11 loci. Samples with ambiguous results after four amplifications per
locus or with <50% successful amplifications across loci were removed from further analysis as they were
not considered reliable genotypes.
Multiplex PCR products were run on an ABI (Foster City, CA) 3130XL automated sequencer (Applied
Biosystems), with the internal size standard GS500 LIZ™ (Applied Biosystems). Fragment analyses were
conducted using the software ABI software GENEMAPPER 4.0.
Probablity of identity, genotyping checking and individual identification
To test the discrimination power of the set of 15 microsatellites, we computed the probability of pairs of
individuals bearing an identical multilocus genotype [i.e. probability of identity (PID)] with the software
GIMLET v 1.3.4 (Valière 2002). PID calculations were performed with both the unbiased equation for
small sample size and the equation for siblings. The more conservative PID for full-sibs (PID-Sib) was
estimated as an upper limit to the probability that pairs of individuals would share the same genotype.
Consensus genotypes from four replicates were reconstructed using GIMLET v 1.3.4 (Valière 2002),
accepting heterozygotes if the two alleles were seen at least in two replicates and homozygotes if a single
allele was seen at least in three replicates. Gimlet was also used to estimate PCR success and errors (ADO
and FA).
Data analysis
We used different analytical approaches for identifying spatial genetic structure and for analysing patterns
of genetic diversity and gene flow across the entire study area. As suggested in previous studies (e.g. Pearse
& Crandall, 2004, Frantz et al. 2009; Francois & Durand, 2010) we applied two Bayesian model-based
clustering algorithms, one spatial and one non-spatial to infer population structure (i.e. number of
clusters, K) and to assign individuals (probabilistically) to populations (or clusters) without a priori
knowledge of population units and limits: GENELAND v 3.2.2 (Guillot et al.2005a,b; Guillot 2008;
Guillot et al. 2008) and STRUCTURE v 2.1 (Pritchard et al. 2000; Falush et al. 2003; Falush et al.
2007). Both of these approaches assume that populations are panmictic units with distinct allele
frequencies. We checked ex-post that the inferred groups were significantly differentiated by FST values
and at Hardy-Weinberg equilibrium (HWE). Moreover, to test whether individual dispersal was restricted
in space, we conducted spatial autocorrelation analysis. Finally, past bottleneck and recent migration
among populations (within the last few generations) was assessed using BOTTLENECK (Cornuet &
Luikart 1996; Piry et al. 1999) and a Bayesian model implemented in MIGRATE software version 3.0.3
- 149 -
CHAPTER
V
Landscape Genetics
(Beerli & Felsenstein 2001; Beerli 2006), respectively. Barriers to gene flow among populations were
tested combining landscape derived data with GIS. Details for all analyses are given below.
Genetic clustering analyses
GENELAND procedure
First we analysed population genetic structure using GENELAND version 3.2.2 (Guillot et al. 2005a, b
Guillot 2008; Guillot et al. 2008) an extension of program R 2.11.1 (Ihaka & Gentleman 1996).
GENELAND implements a Bayesian clustering algorithm similar to STRUCTURE and uses an MCMC
re-sampling method to estimate unknown parameters including the number of genotype clusters.
GENELAND integrates the spatial coordinates of individuals together with the genetic information and
so provides an improved definition of the spatial genetic units when compared with non-spatial clustering
methods. This model assumes that populations are spatially organized as a set of non-overlapping polygons
with no gaps (Guillot et al. 2005a, b). Thus, areas of genetic discontinuity were also detected as
geographical areas of global low posterior probability of population membership. Unlike STRUCTURE,
GENELAND treats the number of genotype clusters as an additional parameter (Guillot et al. 2005a).
Due to substantial algorithm improvement implemented in the recent versions of GENELAND software
(from version 3.0.0 onwards), we used the correlated frequency model that allowed us to detect subtle
structures in the presence of low genetic differentiation that would probably remain undetected using an
uncorrelated frequencies model (Guillot 2008).
The number of clusters was determined by running the MCMC iterations 50 times, allowing K to vary,
with the following parameters: 500,000 MCMC iterations, maximum rate of the Poisson process fixed to
150, uncertainty attached to the spatial coordinates fixed at 1 km, minimum K = 1, maximum K = 10,
maximum number of nuclei in the Poisson–Voronoi tessellation fixed to 450, bur-in of 100,000 in the
post-processing, as well as the option to filter null alleles.
We included uncertainty (1 km) in the spatial coordinates for each individual to account for any
measurement error, movement of individuals, and the potential for observed locations to reflect the true
locations inaccurately (Guillot et al. 2005a). The uncertainty associated with the spatial coordinates was
set to 1 km, based on a 1 km2 estimate of home range size in the study area.
Each run provided (in a single step) an estimate of K and a map of the estimated populations. We
calculated the mean logarithm of posterior probability for each of the 50 runs. The runs were then sorted
according to their mean posterior density and only the best 3 runs were considered in the analysis. These 3
- 150 -
Paper V
Spatial genetic structure
runs were then post-processed (with a burn-in of 100,000 iterations) in order to obtain posterior
probabilities of population membership for each individual and each pixel of the spatial domain. We
finally checked visually for the consistency of results across these 3 runs.
STRUCTURE procedure
STRUCTURE version 2.2 (Pritchard et al. 2000; Falush et al. 2003; Falush et al. 2007) was also used to
investigate the genetic structure of the pine marten. STRUCTURE is currently the most widely adopted
method of determining population structure for groups of individuals with unknown population affinities,
and thus provides a standard with which to compare the results produced by GENELAND. We used an
admixture model, which allows for multiple genetic sources of individuals, with correlated gene
frequencies (Falush et al. 2003). Simulations were run using a burn-in period of 105 sweeps followed by
106 MCMC iterations. Independent runs of K (i.e. number of clusters or gene pools assumed) were
performed from one to eight clusters and repeated twenty times to check for consistency in the results.
To determine the optimal number of clusters, a plot of the estimated log probabilities for each K value
(Pr(X|K)) was examined. If the plot showed a clear peak at one K value, this value with the highest
probability was chosen. If the plot increased gradually, and no clear peak exists, the smallest K value where
the plot reaches a plateau was chosen as the most parsimonious number of populations (Pritchard et al.
2000). Additionally, we computed the standardized second order rate K of change of Ln P(X|K) (Evanno
et al. 2005), to identify the K value that produces the greatest gain in probability of the model. Once the
true K was selected, the fractional membership of each individual in each cluster (q), averaging q over the
20 runs, was plotted on a map of the study region to assess geographical congruence of the clusters and
contrast the results obtained from the spatially explicit GENELAND results.
Moreover, to discuss the relationship between genetic discontinuities and landscape features, we plotted
the results of the modal population for each pixel of the study area according to GENELAND and the
mean membership coefficients of each individual for each cluster according to STRUCTURE.
Population genetic analysis of the inferred clusters
We summarized genetic variation through the number of alleles per locus (A) and expected (HE) and
observed (HO) heterozygosities using GENETIX v 4.05.2 (Belkhir et al. 2004) for each of the inferred
clusters (based on spatially explicit GENELAND results) and for the full dataset. Estimates of pairwise
linkage disequilibria for each pair of loci in each population and deviation from HWE was tested using the
exact test implemented in GenePop version 4.0 (Raymond & Rousset 1995 Rousset, 2008). Statistical
- 151 -
CHAPTER
V
Landscape Genetics
significance was evaluated by running a Markov Chain Monte Carlo (MCMC) consisting of 10,000
batches of 10,000 iterations each, with the first 10,000 iterations discarded before sampling (Guo &
Thompson 1992). Significance levels were adjusted with sequential Bonferroni correction in order to
correct for the effect of multiple tests (Rice 1989).
We calculated pair-wise FST values (Weir & Cockerham 1984) and tested for pairwise genetic differences
among clusters identified by Bayesian models using a randomization procedure implemented in FSTAT v
2.9.3.2 (Goudet 1995; Goudet 2001). Statistical significance at the 0.05 level was evaluated after the
Bonferonni correction for multiple comparisons (Rice 1989). Moreover, analysis of molecular variance
(AMOVA) was also conducted to test the significance of the inferred population structure with the
software ARLEQUIN v 3.1 (Excoffier et al. 2005) at three levels: among populations (considering Navarre
and the Basque country) among individuals within populations and within individuals. MICROCHECKER software (Van Oosterhout et al. 2004) was used to check for potential scoring errors, the
presence of null alleles and linkage disequilibrium.
Spatial autocorrelation analysis
We assessed fine-scale spatial genetic structure (SGS) using spatial autocorrelation analyses. Spatial
autocorrelation analyses, i.e. the analyses of genetic relatedness between pairs of individuals as a function
of geographical distance, were conducted using SPAGeDI 1.2 (Hardy & Vekemans 2002). These were
performed on the 15 polymorphic loci with kinship coefficients (Fij) (Loiselle et al. 1995). As suggested by
Vekemans and Hardy (2004), Fij was chosen as a pair-wise estimator of genetic relatedness, as it is a
relatively unbiased estimator with low sampling variance. The slope (b) of this linear regression does not
depend on an arbitrary choice of distance classes and, therefore, provides a good estimator of the degree of
SGS at this scale (Hardy & Vekemans 2002).
Because there is no consensus regarding how to generate distance classes, we used the recommendations of
Hardy and Vekermans (2002). In all cases, more than 50% of all individuals were represented at least once
in each interval, and the coefficients of variation of the number of times each individual was represented
was less than one (Hardy & Vekemans 2002). Thus, to illustrate the pattern of spatial autocorrelation, the
number of spatial distance classes was set to 22, leading to a minimum of 148 pair-wise comparisons per
distance class. A jackknife procedure over loci was used to estimate standard errors for each distance class
and 10,000 randomizations of individual spatial locations were performed to test for the overall spatial
structure (Hardy & Vekemans 2002).
- 152 -
Paper V
Spatial genetic structure
Bottleneck detection
Statistical methods have been developed to infer the demographic history of a population from a single
genetic sample. The program BOTTLENECK version 1.2.02 (Cornuet & Luikart 1996; Piry et al. 1999)
was applied to test for recent reductions in effective population size. During population bottlenecks, rare
alleles are lost due to drift at a rate faster than loss of heterozygosity. This disparity is used to detect past
bottlenecks. The analyses were performed under the three microsatellite mutational models available:
infinite allele model (IAM), stepwise (SMM) and two-phase model of mutation (TPM) (with 95%
stepwise mutation).
Migration rates among clusters
Evidence of recent migration events across clusters was assessed using the MIGRATE software version
3.0.3 (Beerli & Felsenstein 2001; Beerli 2006;). This software estimates the number of migrants (4Nm)
exchanged between populations per generation using an expansion of the coalescent theory in a Bayesian
approach. Following the recommendations of Beerli (2004), we did an initial run on our data set using
FST to find the start parameters, and we used the output of the initial run as the start parameters of our
second run. Because there were only minor differences between the outputs from the first and the second
runs, we presented only the output from the second run.
RESULTS
Non-invasive sample collection and species identification
Out of 977 faecal samples collected from the entire study area, 179 were discarded because they were not
fresh or because they presumably belong to the same individual (samples separated by < 1km). 670 out of
798 analyzed samples were classified as one of the target species by our PCR-RFLP method. Thus,
unequivocal species identification was possible in 84% of the samples. In the remaining 16%, the DNA
extracted was not amplified by the primers used. We effectively identified 323 faecal samples as stone
marten and 347 as pine marten.
Additionally, we obtained 74 tissue samples of pine marten and 109 of stone marten from road-kill
animals. Thus we obtained a total number of 421 samples of pine marten and 432 of stone marten. The
geographical locations for the 670 correctly identified faecal samples and 183 tissue samples of both
species in the study area are shown in Figure 1. Geographic information system (GIS) technology revealed
that areas of high quality pine marten habitat (natural forest) were interspersed in varying configurations
- 153 -
CHAPTER
V
Landscape Genetics
with non-forested areas and human- made non-suitable habitat (roads; urban areas, reservoirs) (Fig. 1). In
this heterogeneous landscape the stone marten is widely distributed across the whole study area while the
pine marten is restricted to the main forested areas.
Fig. 1. Study area, located in the regions of the Basque country and Navarre, Northern Spain. Grey and white circles
represent the geographical locations of samples identified as Martes martes and Martes foina by the PCR-RFLP
method, respectively (Faecal samples: n=670; Tissue samples: n=183). Natural forest appears shaded in white, urban
areas, roads and man-made reservoirs in black and other land-uses in grey.
Individual identification and genotype checking
Out of 347 faecal samples identified as pine marten 120 were not included in the microsatellite
genotyping procedure. These samples corresponded to the sampling period between 2004-2005 and were
used for a first distribution assessment of sympatric martens in the study area. Thus, 301 pine marten
samples (227 faecal samples and 74 tissue samples) were used for microsatellite genotyping. The first
quality-screening test was not passed by 103 non-invasive samples (45.4%), which were immediately
discarded. The remaining 124 samples (54.6%) were amplified at the other 11 loci. After multiple-tube
genotyping 41 samples from this sub-set (33.06%) were also discarded because they showed < 50% PCR
success, or because of high failure rates. Full multilocus microsatellite genotypes were obtained for the
remaining 83 samples (66.93% from the samples that passed the screening and 36.56 from the total
samples analyzed). The observed average error rates across loci were: ADO = 0.178 and FA = 0.029. The
average proportion of positive PCRs (calculated only for positive samples) was 91% and varied among loci
- 154 -
Paper V
Spatial genetic structure
from 78 to 98% and among samples from 68 to 100%. PID analysis showed that the set of 15 loci would
produce an identical genotype with a probability of 9.83x10-12, and with a probability of 2.09x10-5 for a
full-sib. Thus, only 2.09 pine martens in 100,000 siblings are expected to share by chance an identical
genotype, suggesting no “shadow effect” (i.e. all the genotypes identify distinct individuals; Mills et al.
2000), and that matching genotypes are recaptures of the same individual.
After a regrouping procedure we identified 66 individual genotypes from faecal samples. The 74 tissue
samples were correctly genotyped at 15 loci and provided new individuals. In total we identified 157
genotypes that correspond with 140 different individuals. The number of times each individual was
detected in the survey varied from 1 to 3 with a total number of 17 re-samplings.
Clustering analyses
Bayesian clustering with spatial information (GENELAND)
The three GENELAND runs giving the highest average posterior probability suggested the presence of
four geographically coherent genetic clusters in the study area. The modal population for each pixel of the
study area inferred by GENELAND indicated that there is a clear spatial pattern with a West-East
subdivision between the four inferred pine marten populations (Fig 2). The second and third best runs
gave rise to similar clusters (results not shown). Individuals from Navarre (almost continuously forested
area) formed a unique genetic cluster (N) located in the Eastern part of the study area. Individuals from
the Basque country (highly fragmented forested area) formed three different clusters: Eastern Basque (EB)
Central Basque (CB) and Western Basque (WB) clusters (Fig 2).
Bayesian clustering without spatial information (STRUCTURE)
Structure provided consistent results over 20 replicated runs tested for each K. The STRUCTURE
analysis showed that the likelihood of the data increased quickly up to K = 3 and then reached an
asymptote (Figure 3a). Calculation of ΔK value (Evanno et al. 2005) from the STRUCTURE output
produced a distinct apex value (300.75) when K = 3 (Fig. 3a), implying the likely presence of three
genetically distinct groups. The assignment of individuals to populations for K = 3 is presented in Fig. 3b
and their spatial location is provided in Fig 2.
- 155 -
CHAPTER
V
Landscape Genetics
Fig 2. Modal population for each pixel of the study area according to GENELAND (K=4) and spatial distribution of
the cluster membership coefficients according to STRUCTURE (K=3) for each individual. The four genetic clusters
are represented by different colours: Navarre (N) cluster-green, Eastern Basque (EB) cluster-red, Central Basque (CB)
cluster-blue, Western Basque (WB) cluster-yellow. Structure results: Each individual is represented by pie charts at
the location where they were sampled (n = 140). The colours indicate the average membership coefficients for each
individual to each of the three clusters uncovered by STRUCTURE (green, red and blue for cluster N, EB and CB,
respectively).
The modal assignments by STRUCTURE roughly corresponded to those identified with spatial
information in the GENELAND analyses (Fig 3b and Fid 2), with the exception of the WB cluster, which
was not identified in STRUCTURE analysis (Fig 3b and Fig 2). However this group consisted of only a
few (n=12), highly admixed individuals located in the western part of the study area (Fig 3b and Fig 2).
This pattern was clearly evident when we interpolated on the map of the study area the cluster
membership coefficients obtained with STRUCTURE and the modal population for each pixel of the
study area obtained by GENELAND (Fig. 2). Although GENELAND suggested the presence of an extra
genetic cluster, there was a clear convergence between the assignments of the individuals to each cluster
provided by both Bayesian methods. STRUCTURE inferred basically the same clusters (i.e. N, EB and
CB) with the exception of the WB cluster, which was only detected by GENELAND. 85.71% of the
individuals (120/140) were assigned probabilistically to the same genetic cluster i.e. N, EB and CB by
both methods (taking into account the assignment of the individuals by STRUCTURE to the population
for which the estimated membership was the highest). In the case of the 12 individuals (8.57%) identified
- 156 -
Paper V
Spatial genetic structure
as a different cluster by GENELAND (WB), these corresponded to individuals assigned by STRUCTURE
to the EB or CB cluster and/or clearly admixed individuals (q<0.7) and all of them were located in the
western part of the study area (Fig 3b and Fig 2). The 8 individuals (5.71%) for which each method
provided different population assignments were located between contact zones of different clusters (Fig 3b
and Fig 2).
Fig. 3 Structure analysis results. a) Estimation of the number of pine marten clusters (K) from 20 independent runs
for K = 1–8. Dotted line is the mean Ln probability of the data (Ln P(K)) (Pritchard et al. 2000) and the solid line is
the second-order rate of change (ΔK)(Evanno et al. 2005), inferring that K=3. (b) Distribution of the three genetic
clusters generated by STRUCTURE. The vertical lines are broken into coloured segments showing the proportion of
each individual assigned to each of the inferred K. The three genetic clusters are presented by different coloured
columns and are displayed according to East-West spatial location of individuals: Navarre (N) cluster-green, Eastern
Basque (EB) cluster-red, Central Basque (CB) cluster-blue. Letters at the bottom of the figure correspond to
Geneland inferred cluster composition for comparison. The Western Basque (WB) Geneland inferred cluster
corresponds to 12 admixed individuals.
Overall, both Bayesian methods indicate that the entire M. martes population is separated into at least
three subpopulations with a strong spatial pattern (Fig 3b and Fig 2) and with sharp borders between
populations. Thus, there is a clear spatial genetic pattern confirmed by spatially explicit (GENELAND)
and non-explicit (STRUCTURE) Bayesian methods. Comparing landscape data with population
boundaries (Fig 2), it is clear that gene-flow barriers revealed by both Bayesian methods coincided with
- 157 -
CHAPTER
V
Landscape Genetics
spatial distribution of anthropogenic habitat gaps (boundary between CB and EB cluster) and the presence
of the sympatric stone marten (boundary between N and EB cluster) (Fig 1 and Fig 2).
Genetic diversity and HWE
The overall pine marten dataset (n=140) is not at HWE (p=0.0003) (Table 1). These results were
consistent with the existence of population structure, which is to be expected at this broad geographic
level. On the other hand, we observed no deviation from HWE in each of the inferred populations (Table
1). These results confirmed the validity of the clusters inferred by Bayesian means. The clusters range in
sample size from 12 to 47, with an average size of 35 individuals.
Table 1 Genetic diversity indices of pine marten samples genotyped at 15 microsatellite loci.
Cluster
n
A
Ho
He
HWE P value(SE)
Overall FIS
EB
47
4.200
0.5292
0.5621
0.0891 (0.0056)
0.0692
WB
12
3.8677
0.5878
0.5699
0.3991 (0.0192)
0.01298
CB
38
3.7333
0.5191
0.5338
0.3976 (0.0098)
0.04092
N
43
5.1333
0.5608
0.5845
0.0559 (0.0047)
0.05245
Total
140
5.7333
0.5412
0.6191
0.0003 * (0.0002)
0.05130
Genetic clusters identified by Geneland: EB, Eastern Basque; WB, Western Basque;
CB, Central Basque; N, Navarre. n, number of individuals; A, mean number of alleles
per locus, Ho, the mean observed heterozygosity, He, expected heterozygosity, HWE,
Hardy-Weinberg equilibrium P value and standard error and overall Fis. Significant
values are marked with asterisk (p<0,001)
Linkage disequilibrium was not apparent for any pair of loci within any of the subpopulations after
performing Bonferroni corrections. The results of MICROCHECKER indicated that null alleles were
apparently present at two loci: MP0188 and Ma19. However, the estimated frequency of the null allele at
these loci occurred at a relatively low frequency (0.054-0.060). As such, it is valid to utilise the data
generated using these markers to assess the levels of genetic diversity within, and the structure between,
populations.
For all 140 pine martens, the average observed (HO) an expected (HE) heterozygosity with values at 0.542
and 0.6191, respectively (Table 1). All 15 loci were variable with the total numbers of alleles ranging
- 158 -
Paper V
Spatial genetic structure
between 3 and 12 per locus. The mean number of alleles per locus in the 4 subpopulations ranged from
3.73 (WB) to 5.13 (N). Similar levels of genetic diversity were observed within each of the 4
subpopulations (Table 1). Observed heterozygosities ranged from 0.511 in the CB subpopulation to
0.5878 in the WB subpopulation. FIS values were positive but not significantly different from zero in each
of the inferred clusters.
Genetic differentiation, migration rates and population bottleneck
Genetic differentiation among clusters was also revealed by FST values and migration rates. All intercluster pair-wise comparisons summarized by mean FST were significant (P < 0.05, Table 2). FST values
ranged from 0.056 to 0.170 (Table 2) and suggested low migration rates. The N cluster was the most
differentiated subpopulation with the highest FST value.
Table 2 F-statistic (FST) tests for pairwise population differentiation (GENELAND groups) based on microsatellite
loci frequencies.
Cluster
EB
WB
CB
N
EB
-
0.05684*
0.07981*
0.11379*
WB
-
-
0.07480*
0.07753*
CB
-
-
0.17262*
*P < 0.05.
This result is in agreement with AMOVA analysis showing a high percentage of variation (11.07) when
considering Navarre and the Basque country as the main groups in comparison to the 4.27% of variation
among individuals within inferred clusters. Considering all the individuals the percentage of variations was
84.66 supporting the evidence of population substructure. Migration rate analysis also indicated that the
migration rates between each genetic cluster were low, with bi-directional migration rates ranging between
2.5 and 3.6 individuals per generation (Table 3). However, our results indicated higher emigration from
the EB cluster to adjacent areas (i.e. CB and N clusters) than from adjacent areas to the EB cluster.
BOTTLENECK analysis found significant support (P < 0.05) for historical reductions in effective
population size in each of the inferred clusters under all three microsatellite mutational models (i.e IAM,
SMM, TPM).
- 159 -
CHAPTER
V
Landscape Genetics
Table 3 Bi-directional estimates of gene flow (Nm) between pine marten clusters as calculated with
MIGRATE.
Migration from
Migration to
EB
WB
CB
N
EB
-
2.51806
2.5973
2.86636
WB
2.84265
-
3.62889
2.85025
CB
3.09339
2.52047
-
2.65878
N
3.17622
2.51639
2.6798
-
Spatial autocorrelation
Spatial autocorrelation analysis also suggested local genetic structure within the study area. The negative
regression slope (b=-0.0437 SE=0.00955) between kinship coefficient and logarithmic distance between
individuals was significant (P < 0.001). There was significant deviation from the population mean kinship
estimate in the closest and most distant distance classes (Fig. 3). Positive values of kinship coefficient were
found at short distances, meaning that neighboring individuals had a higher genetic relatedness than
random pairs of individuals, whereas negative values of kinship occurred at larger distances, indicating
isolation by distance within the whole study area. The intercept of the correlogram with the x-axis was
approximately 70 km (Fig. 3), suggesting that within this distance individual pine martens are more
related than on average in the population.
Figure 4. Average kinship coefficients, Fij, between pairs of pine martens individuals plotted against the
Ln geographical distance. Dashed lines represent 95% confidence intervals for Fij under the null
hypothesis that genotypes are randomly distributed. Significant deviation of all distance classes from the
population mean (P<0.001) was observed with the exception of distance classes of 50 and 70km.
- 160 -
Paper V
Spatial genetic structure
DISCUSSION
In this study we used non-invasive genetic sampling of faeces to assess the distribution of sympatric
martens and to elucidate spatial genetic structure of the forest dwelling pine marten. First, as pine martens
are elusive and stress-sensitive (Barja 2007), and inhabit sympatrically with the stone marten we applied a
species identification method to asses spatial distribution of sympatric martens and to isolate the pine
marten DNA samples. This DNA source is regularly used for non-invasive genetic studies of carnivores
(Schwartz & Monfort 2008; Beja-Pereira et al. 2009). However, to our knowledge this is the first time
that faecal DNA has been used to asses spatial genetic structure of the European pine marten. This allowed
us to gather a higher number of samples for this study without any physical contact with individuals.
The application of a sufficiently variable panel of microsatellites for individual identification and
population characterization through Bayesian analyses have allowed us to detect the presence of at least 3
genetic clusters distributed across the study area with non-overlapping distribution, but also of physical
barriers to gene flow.
Spatial distribution of sympatric marten species
Our results indicated that the PCR-RFLP method (Ruiz-González et al. 2008) proposed for use on noninvasive faecal samples is a reliable, efficient, time-saving and cost-effective procedure for improving our
knowledge of the spatial distributions of sympatric marten species, and, subsequently, to isolate pine
marten samples for further microsatellite analyses. An overall high species identification rate was obtained
(670 out of 798 analyzed samples; mean 84%). Species identification success was rather high with respect
to other similar approaches (e.g.: 58%, Livia et al. 2007; 53.4%, Pilot et al. 2007) and similar to those
obtained, by the same method, in the northern Iberian Peninsula (Rosellini et al. 2008; Ruiz-González et
al. 2008) and Italy (Balestrieri et al. 2009).
The results obtained in the present study improved our previous knowledge about the spatial distribution
patterns of sympatric marten species (Proulx et al. 2004). Both species showed a notorious spatial
segregation in the study region, but they also co-exist locally in some forested areas. As a typical stenotopic
species (Storch et al. 1990; Brainerd & Rolstad 2002, Zalewski & Jedrzejewski 2006), the results of our
study confirmed that the pine marten is a species strongly associated with forest habitats, wherein its
distribution is restricted to forested areas. This distribution pattern leaves this species particularly
vulnerable to forest fragmentation (Proulx et al. 2004). Indeed, it was almost absent in the northern
Basque country area, where the natural forests are scarce and human occupancy and anthropic
- 161 -
CHAPTER
V
Landscape Genetics
derangements of natural habitats are more intense. It was also absent from wide intensive crop fields in the
Ebro Valley. Moreover, as a typical cold adapted Eurosiberian species, the pine marten was not present in
the southern part of the study area, characterized by being more influenced by Mediterranean climate,
although wide and well preserved natural deciduous forests are still present. In the absence of the pine
marten, the stone marten was widely distributed in this area. This is in agreement with previous studies
that indicated that in areas of the Iberian Peninsula without pine martens, stone martens are known to
occupy the higher quality forest habitat (Virgós et al. 2000; Virgós & Garcia 2002;). These results
suggested that climatic conditions are playing an important role in pine marten distribution, having a
considerable influence on the bioecological relationships with the stone marten as the study area is located
at the south-western limit of a typical Eurosiberian species (Proulx et al. 2004). The spatial distribution
limit of pine martens outlined in this study is a step towards understanding the factors affecting the
occurrence of the pine marten at the southern limit of its European distribution. Moreover, the more exact
knowledge of pine marten distribution derived from this study has an important conservation value for the
revision of the protective status of the species currently listed as “Rare” on the Basque Catalogue of
Threatened species (Decree 167/1996 and Order of July 8, 1997). By contrast, the stone marten seems to
be very adaptable and it was found not only in forests but also in a wide variety of different habitats
including open areas, as well as a number of places markedly transformed and intensively exploited by
humans. The latter species is generally regarded as generalistic in habitat with synanthropic behaviours in
Europe (Reig 1992; Broekhuizen & Müskens, 2000; Lanszki 2003;) and has even invaded urban areas
(Herr et al. 2009) but it can also establish home ranges in extensive agricultural areas near to rural villages
and also in woodlands (e.g. Delibes 1983; Genovesi et al. 1997; Virgós & Casanovas 1998; Broekhuizen
& Müskens, 2000; Virgós et al. 2000; Virgos & Garcia 2002; Santos & Santos-Reis 2010) as has been
documented in this study. The stone marten has a clear dominance over the pine marten in the southern
study area (i.e. Mediterranean area) and is the only Martes species found towards the south in the Iberian
Peninsula, suggesting that its more thermophile condition (López-Martín 2007) provides a comparative
advantage facing current climate change.
Wherever the two species occur sympatrically, the stone marten is often associated with rural and
suburban areas (Delibes 1983; Reig 1992), while the pine marten occupies mainly forested areas. In one of
the earliest studies about interspecific competition between martens, Delibes (1983) suggested that in
altered environments the stone marten is displaced to more urbanised areas, but would tend to become less
synanthropic wherever the density of the pine marten decreased. This compression of the spatial niche of
the stone marten has been explained as a consequence of interspecific competition, favouring the slightly
- 162 -
Paper V
Spatial genetic structure
bigger pine marten (Delibes 1983). In the study area, the rarity of the pine marten could be associated
with events of direct persecution during recent decades and the increased effects of forest loss and
fragmentation (López-Martín, 2007). This situation provides a better competitive framework to the stone
marten, which shows more behavioural plasticity than the pine marten, proving capable of adaptation to
different habitats and climatic conditions, and being less vulnerable to anthropogenic changes and human
presence (Herr et al. 2009). Consequently, the colonization of new areas by the pine marten will be
significantly reduced in areas where the stone marten is present, thus reducing gene flow between
populations.
Recent studies have shown contrasting results regarding the relationships between the two species on a
local scale. The pine marten was the only Martes in a mountainous and well conserved forest area in northwestern Spain (Rosellini et al. 2008), while the pine marten is expanding its range into an agricultural
landscape matrix in NW Italy which was previously considered a prerogative of the more synanthropic
stone marten. In a forest-field mosaic of central Poland the two martens exploited the same micro-habitats
(Pilot et al. 2007; Goszczyński et al. 2007), but they substantially differed in their ways in area searching
and foraging behaviour (Posluszny et al. 2007) as well as in their attitude towards open areas and human
settlements (Goszczyński et al. 2007).
Our results agreed with previous studies highlighting that sympatric relationships differ depending on
several factors. The ecological displacement (e.g. Rosellini et al. 2008) or co-existence (Pilot et al. 2007;
Goszczynsky et al. 2007) that occurs when both species overlap in the same area probably depends on a
combination of several factors, including the relative abundance of each predator species within the local
carnivore guild, the availability of trophic resources, climate conditions (cold adapted Martes martes vs
more thermophile Martes foina) and landscape effects of forest fragmentation and human derangement of
the surrounding habitats. Both species have inverse attitudes towards these factors (stenotopic Martes
martes vs more synanthrophic Martes foina). Competition for these resources and the adaptability to new
environmental conditions can play a fundamental role in the interactions between these species (Schoener
1974; Linnell & Strand 2000).
In summary, the results provided by the combined use of non-invasive genetic analysis of faecal samples
and GIS technology suggest that our protocol would serve to assess the landscape effects on distribution
patterns, identify and resolve information gaps and design effective research and management
programmes. Moreover, an ongoing work on habitat suitability modelling will allow us to assess the
interspecific relationships of martens in relation to environmental and landscape variables (Ruiz-González
et al. in prep.).
- 163 -
CHAPTER
V
Landscape Genetics
Faecal DNA genotype reliability
Research on non-invasive genetics of mustelids are limited and biased towards species identification
methods (e.g. Gómez-Moliner et al. 2004; Livia et al. 2007; Ruiz-González et al. 2008). In this study,
species identification of the field-collected scats ensured that only target pine marten samples were used for
further microsatellite analysis. Initial steps for species identification are thus crucial in landscapes where
both species overlap extensively (Ruiz-González et al. 2008) allowing reliable spatial distribution
assessment. Microsatellite genotyping of faecal samples has also been carried out on a small number of
mustelid species and used mainly for individual identification and population size estimation (e.g.
European otter, Arrendal et al. 2007; Hajkova et al. 2009; wolverine, Hedmark & Ellegren 2007; badger,
Wilson et al. 2003; Frantz et al. 2006). To our knowledge, this study is the first to investigate and evaluate
the spatial genetic structure of the European pine marten from the focus of faecal DNA.
Reliable amplification of microsatellite markers is a prerequisite for estimating the size and structure of
populations within the landscape (Taberlet et al. 1999; Valière et al. 2007). The most conservative
estimate of PID (i.e. PID-Sib statistics) suggested that DNA profiles consisting of the 15 loci used in our
study would be sufficient to distinguish between individuals of the study species, including siblings, with
99% certainty (PID-Sib=2,09x10-5). The success rate of pine marten microsatellite genotyping (presented
here as a percentage of faecal samples that provided consensus genotypes, for all loci needed for reliable
individual identification) in our study was 36.56%. This result is approximately around average in
comparison with many other non-invasive studies undertaken on other carnivore species (e.g Frantz et al.
2003; Hedmark & Ellegren 2007; Hajkova et al. 2009; Marucco et al. 2009; Lanszki et al. 2010;
Stenglein et al. 2010). Similarly, our error rates (ADO = 17.8%, FA = 2.9%) were comparable to those
obtained in other non-invasive genetic studies of mustelids (e.g Frantz et al. 2003; Hajkova et al. 2006;
Ferrando et al. 2008).
Genetic variability of the European pine marten
Previous studies focused on the pine marten have given information about the genetic variability and
structure of several European populations (Kyle et al. 2003; Mergey 2007; Pertoldi et al. 2008; Mullins et
al. 2010). So far, however, information about Iberian populations was unavailable. Moreover, these studies
were mainly focused on tissue (Kyle et al. 2003, Mergey 2007; Pertoldi et al. 2008) or remotely plucked
hair (Mullins et al. 2010) so this study marks an improvement in genetic surveys of pine marten through
faecal sampling.
- 164 -
Paper V
Spatial genetic structure
Studying pine marten populations across Europe, Kyle et al. (2003) found microsatellite HE values in the
range of 0.56–0.64 for continental and 0.34-0.66 for insular populations (8 microsatellite loci). In the
same study, the pine marten populations have been shown to have an overall FST value of 0.18 (range:
0.016–0.330) which comprise comparisons between geographically distant populations (e.g. FST value
between Italy and Germany 0.044). Mergey (2007) obtained similar values for different French
populations (11 loci, HE=0.518) and values of FST ranging between 0.047-0.052 for populations
separated by more than 130 km. Pertoldi et al. (2008) found HE and FST values in the range of 0.67-0.79
and 0.044-0.097, respectively for Danish populations (11 loci). Mullins et al. (2010) found slightly lower
values of HE for the reduced and isolated Irish populations (0.35, 17 loci). In this study, using 15 variable
loci, we detected HE values of 0.586. So, taking into account the higher number of variable loci used in
this study we can estimate that, compared with continental European populations, Iberian populations of
pine marten have lower levels of genetic variability. Interestingly, the FST values reported in this study,
which range from 0.056 to 0.113, were surprisingly high taking into account the regional scale of the
study area and the contiguous nature of the inferred clusters. However, caution is called for when trying to
interpret differences in genetic variability and FST values between different studies as these differences can
also be attributed to the different number of microsatellites used.
Spatial genetic structure and putative barriers to geneflow
Here we utilised individual-based Bayesian approaches for inferring the location of potential barriers to
gene flow and identifying population structure (Manel et al. 2003; Guillot et al. 2005a).
Our data did not support the hypothesis that pine martens in the study area exist as a single, panmictic
population. When we considered all the pine martens as a whole, we detected a significant deviation from
random mating, indicating that M. martes were not distributed in a random fashion and were instead
associated in genetically defined subpopulations. Subpopulation structuring of pine martens at regional
level is confirmed when looking for genetic partitions using Bayesian methods. The results generated
independently by STRUCTURE and GENELAND suggested the presence of at least three genetic units.
Levels of genetic differentiation among the clusters represented by pair-wise FST, were greater than similar
measures estimated among other pine marten populations sampled across much wider geographic areas
(Kyle et al. 2003; Mergey 2007; Pertoldi et al. 2008). Even though the number of clusters varied between
approaches (GENELAND, K=4 and STRCTURE k=3), the location of, and membership within, each of
the inferred clusters was largely concordant between methods. Both spatial and aspatial Bayesian clustering
methods inferred mainly the same three genetic clusters (i.e. N, EB and CB) and have a clear convergence
- 165 -
CHAPTER
V
Landscape Genetics
between the assignments of the individuals to each cluster (120 out of 140 individuals analysed were
assigned probabilistically to the same genetic cluster). Accordingly, it seems that pine martens did not
form a single reproductive unit, with a clear spatial subdivision within the study area.
A certain amount of non-convergence between different Bayesian clustering methods thus appears to be
relatively frequent (Frantz et al. 2009; Francois & Durand 2010). In recent applications of spatially
explicit and implicit models together to the same dataset, when more than one model is used, the studies
often report consensual results (Fontaine et al. 2007; Latch et al. 2008; Liu et al. 2009), but there are
interesting exceptions (Coulon et al. 2006; Rowe & Beebee 2007; Ball et al. 2010; Francois & Durand
2010). The detection of four clusters by GENELAND versus the three clusters identified by
STRUCTURE can be interpreted as an example of increased power in spatially structured populations as
has been previously documented (e.g. Coulon et al. 2006). However, recently Francois & Durand (2010)
showed that models without admixture (i.e. GENELAND) are not sufficiently robust to the inclusion of
admixed individuals in the sample, thus leading to an incorrect assessment of population genetic structure
in many generic cases. Consequently, taking into account the high proportion of admixed individuals
encompassed within the WB cluster, the validity of the fourth inferred GENELAND cluster should be
interpreted with caution. Moreover, this group is composed of a reduced number of individuals (n=12).
Therefore, the validity of the results regarding this cluster should be further evaluated with an increased
sample size.
As noted by the authors of the Bayesian algorithms used here (Pritchard et al. 2000; Guillot et al. 2005b)
and recently shown by simulation (Frantz et al. 2009; Schwartz & McKelvey, 2009), deviations from
random mating not caused by barriers to gene flow (i.e. spatial autocorrelation and isolation by distance)
and the sampling scheme may have impacts on the detection and interpretation of genetic structure. In
this study, spatial autocorrelation analysis strongly hinted an isolation by distance pattern. However, the
conformation of the inferred subpopulations to HWE expectations, and given the fact that the inferred
clusters showed significant differentiations as indicated by estimates of pairwise FST values, suggests that
these subpopulations have been appropriately defined (Guillot & Santos 2009). The low migration rates
detected between identified clusters also gave strong support to the genetic structure detected. Besides,
agreement between the results of similar models and algorithms (i.e. GENELAND and STRUCTURE)
give a high degree of confidence to the proposed clustering solution (Guillot et al. 2010).
Finally, our results clearly demonstrated the existence of a neat spatial pattern, with no overlapping
distribution between subpopulations and with sharp discontinuities between contiguous areas that are
related to different environmental features, while this was in no way included in the algorithm itself.
- 166 -
Paper V
Spatial genetic structure
The N, EB, and CB clusters were delineated with similar boundaries in STRUCTURE to those identified
with spatial information in the GENELAND analysis, both confirming that the main genetic
discontinuities found are spatially coherent. Even though the analysis with STRUCTURE did not use any
prior geographical information, it inferred the same genetic discontinuities between groups, with the only
exception of the boundary with the WB cluster, which was only corroborated by GENELAND. The
location of their boundaries suggested the existence of some putative barriers to geneflow in the study area,
but as has been previously documented for other forested species, the combination of several landscape
features with low permeability can lead to population differentiation (Coulon et al 2006; Broquet et al.
2006).
Different and non-exclusive hypotheses can explain the significant genetic structure observed in pine
marten populations at the regional scale. Firstly, over-harvesting processes (Helldin 2000) and direct
persecution due to their valuable fur greatly depleted pine marten populations in recent decades, also in
North Spain (López-Martín, 2007). The three primary genetic consequences of hunting are the alteration
of population structure, loss of genetic variation, and evolution resulting from selection (Harris et al.
2002; Allendorf et al. 2008; Coltman 2008, Allendorf & Hard 2009). Hunting activities, which stopped
only a few decades ago in North Spain, and its consequences could be responsible for the recent decline of
pine marten populations in the study area and thus, an important factor in the origin of the recurrent
population bottlenecks detected in all the inferred clusters. Thus, hunting is probably one of the factors
involved in the origin of the present genetic structure.
Secondly, habitat fragmentation may have important consequences for population genetic structure, which
may be even greater in habitat specialists that show a high dependence on particular resources and strong
sensitivity to habitat changes and loss (Coulon et al. 2006; Broquet et al. 2006; Devictor et al. 2008).
Habitat fragmentation may be due to the presence of (non-linear) areas of non-suitable landscape elements
that act as ‘moderators of gene flow’ and lead to the differentiation of genetic units (e.g. Coulon et al.
2006), as also to the existence of narrow linear elements (e.g. roads, rivers) that divide otherwise
continuous patches of habitat, or both. Indeed, depending on the abilities of organisms to move through
the matrix separating habitat patches, gene flow directions and quantities can be deeply modified, which
can in turn affect the genetic structure of the species (e.g. Coulon et al. 2006; Broquet et al. 2006;
Wasserman et al. 2010), even in the case of highly mobile carnivores (e.g. Latch et al. 2008, Zalewski et al.
2009). Certainly, deforestation and forest fragmentation have been reported to affect the distribution and
density of pine martens (Brainerd et al. 1995; Kurki et al. 1998), which are believed to need a minimum
woodland area to survive (Zalewski & Jędrzejewski 2006) and tend to avoid treeless areas (Storch et al.
- 167 -
CHAPTER
V
Landscape Genetics
1990; Brainerd and Rolstad 2002, Pereboom et al. 2008). As a result of habitat specialization this species
is particularly sensitive to changes in its original habitat, including habitat loss, and landscape-scale effects
of habitat fragmentation (Brainerd 1990; Bright 1993; Proulx et al. 2004; Pereeboom et al. 2008). This
specific habitat requirement is probably the cause of very limited gene flow between contiguous
subpopulations and the strong genetic structure observed between geographically close localities in the
highly fragmented forests in Basque country area, as demonstrated by the Bayesian clustering of groups of
individuals. On the other hand Navarre, formed by a largely continuous forested area, belongs to a single
panmicitic population. These results are in agreement with previous studies suggesting that, for typical
forest dwelling species, one of the main factors structuring present-day genetic diversity could be
attributed to forest fragmentation processes (e.g. Coulon et al. 2006; Broquet et al. 2006; Olivieri et al.
2008; Segelbacher et al. 2008; Craul et al. 2009; Liu et al. 2009; Pavlacky et al. 2009; Bruggeman et al.
2010).
When comparing the gene-flow discontinuities revealed by Bayesian means and the spatial pattern of all
landscape features, some evidences of geographical barriers could be identified. The genetic discontinuity
found between CB and EB clearly overlapped with the geographic location of an area characterized by a
high density of roads, the presence of large man-made reservoirs and the high presence of human
urbanization. Indeed, this area has been previously identified as one of the critical connectivity areas in the
framework of the Basque connectivity network (Gurrutxaga et al. 2010a, b). The reduced migration across
this boundary may be rooted in species behaviour (i.e. aversion to move across treeless areas; Storch et al.
1990, Brainerd & Rolstad 2002, Pereboom et al. 2008) and direct road mortality (Balkenhol & Waits
2009). The high differentiation detected between CB and EB subpopulations suggested that this area
strongly affected gene flow between these spatially contiguous areas. The genetic discontinuity located
between the EB and N clusters corresponded to an area where the pine marten distribution shows a clear
gap in spite of the presence of suitable forest habitat. Indeed, the distribution gap matched with the
presence of the more synathropic stone marten. There is no previous genetic evidence about the effect of
interspecific competition on dispersal and genetic structure on mustelids. However our results suggested
that interspecific competition between sympatric marten species could be playing an important role in
structuring pine marten populations. Spatial distribution data showed that each marten species does not
utilize the landscape uniformly. Thus, contrasting distribution patterns, differences in habitat utilization
and ecological traits in both stone and pine marten could also be playing an important role in structuring
pine marten populations. Consequently, the colonization of new areas by the pine marten will be
significantly reduced in areas where the stone marten is well established, thus becoming a barrier to gene
- 168 -
Paper V
Spatial genetic structure
flow between pine marten populations. These results provide an interesting insight into the effects of
interspecific competition on dispersal and genetic structure in sympatric carnivores.
Altogether, our results have shown that pine martens inhabiting a fragmented landscape in the Basque
country and Navarre may not be considered to belong to a single panmicitic population, comprising at
least three different genetic units with strong genetic differentiation despite a rather small distribution
area. The data reported in this study suggested that habitat loss and fragmentation could be the main
factors responsible for the spatial genetic structure found in a typical forest dwelling species. Several
landscape features (i.e. unforested areas, urbanized areas, roads and man-made reservoirs) acted as
moderators of gene flow because of their high resistance to pine marten migrations, and hence their
cumulative effect led to the differentiation of the inferred genetic units. Moreover, past over-harvesting
processes, together with the probable interspecific competition with the stone marten added to the fact
that the study area is placed at the southern limit of pine marten distribution, could be acting synergically
with forest fragmentation as factors shaping the current spatial genetic structure.
Implications for conservation and connectivity restoration
Understanding the spatial genetic structure of populations can provide insight into the ecological or
evolutionary processes of the species, and enable wise conservation decisions (Segelbacher et al. 2010).
Anthropogenic habitat degradation and fragmentation not only leads to distribution range contractions
and population extinctions, but may also have significant genetic, and thus evolutionary, consequences for
populations (Fischer & Lindenmayer 2007; Nabe-Nielsen et al. 2010). For the conservation of species
inhabiting fragmented habitats, securing dispersal between local habitat patches appears to be the major
challenge (Taylor et al. 1993; Cushman et al. 2009). Pine martens have experienced a large range
contraction in the last few decades in the study area, and, we found some landscape features that already
constrain connectivity among existing habitat patches. Further conservation measures thus need to
consider population processes operating at the landscape scale ensuring population connectivity (Crooks
2006). Improving habitat quality and connectivity in the study area will not only secure the long-term
persistence of the pine marten but also favour connectivity for other forest associated species (Gurrutxaga
et al. 2010b). The future coexistence and relative abundances of the two martens in forest habitats will
thus depend on the mode of forest management as well as on the existence of effective migratory corridors
connecting neighbouring forest patches (Goszczynski et al. 2007; Gurrutxaga et al. 2010b).
- 169 -
CHAPTER
V
Landscape Genetics
ACKNOWLEDGEMENTS
Ruiz-González holds a Ph.D. fellowship awarded by the Dept. of Education Universities and Research of the Basque
Government (Ref. BFI09.396). This study has been partially funded by the Basque Government through the
Research group on “Systematics, Biogeography and Population Dynamics” (Ref. IT317-10; GIC10/76) and by the
University of the Basque Country (UPV-EHU) and the Department of Environment, Territorial Planning,
Agriculture and Fisheries (Basque Government) through IKT S.A under the University-Enterprise research program
(Ref. UE07/02). The authors wish to thank the following persons and institutions for supplying tissue and faecal
samples: Dr. F. Urra (VRFN-Navarre Government); The technical staff of the Natural Parks of Alava (DFA); Museo
de Ciencias Naturales de Álava; P. Lizarraga and L. Lorza (CRF Martioda-DFA); J.M. Fernandez (IAN); N. Ruiz de
Azua; G. Belamendia (CEA); J. Pinedo (DFA); H. Aguirre, P. Pérez and G. Dominguez.; F. Canales; M.A. Campos
(CRN); E. Arberas; Oskar Berdión, M. Corral; J. López de Luzuriaga; I. Aginako; H. Aguirre; P. Pérez; J. Ruiz
Guijarro; I. Martínez; E. Díaz; J.M. Pérez de Ana; D. Vado; A. Eguia; A. García de Albéniz; I. Zuberogoitia;; M.
Barral (NEIKER). We are also indebted to M. Gurrutxaga (Dept. of Geography, UPV-EHU) for his help with the
GIS treatment of data and preparing maps.
REFERENCES
Allendorf FW, England PR, Luikart G, Ritchie PA, Ryman N (2008) Genetic effects of harvest on wild animal
populations. Trends in Ecology & Evolution, 23, 327-337.
Allendorf FW, Hard JJ (2009) Human-induced evolution caused by unnatural selection through harvest of wild
animals. Proceedings of the National Academy of Sciences of the United States of America, 106, 9987-9994.
Anderson CD, Epperson BK, Fortin MJ, Holderegger R, James PMA, Rosenberg MS, Scribner KT, Spear S (2010)
Considering spatial and temporal scale in landscape-genetic studies of gene flow. Molecular Ecology, 19, 35653575.
Arrendal J, Vila C, Bjorklund M (2007) Reliability of noninvasive genetic census of otters compared to field
censuses. Conservation Genetics, 8, 1097-1107.
Balestrieri A, Remonti L, Ruiz-Gonzalez A, Gomez-Moliner BJ, Vergara M, Prigioni C (2010) Range expansion of
the pine marten (Martes martes) in an agricultural landscape matrix (NW Italy). Mammalian Biology, 75, 412419.
Balkenhol N (2009) Evaluating and improving analytical approaches in landscape genetics through simulations and
wildlife case studies. PhD Thesis, University of Idaho.
Balkenhol N, Waits LP (2009) Molecular road ecology: exploring the potential of genetics for investigating
transportation impacts on wildlife. Molecular Ecology, 18, 4151-4164.
Balkenhol N, Waits LP, Dezzani RJ (2009) Statistical approaches in landscape genetics: an evaluation of methods for
linking landscape and genetic data. Ecography, 32, 818-830.
Ball MC, Finnegan L, Manseau M, Wilson P (2010) Integrating multiple analytical approaches to spatially delineate
and characterize genetic population structure: an application to boreal caribou (Rangifer tarandus caribou) in
central Canada. Conservation Genetics, 11, 2131-2143.
Barja I (2005) Winter distribution of European pine marten (Martes martes) scats in a protected area of Galicia,
Spain. Mammalia, 69, 435-438.
Barja I, Silvan G, Rosellini S, Pineiro A, Gonzalez-Gil A, Camacho L, Illera JC (2007) Stress physiological responses
to tourist pressure in a wild population of European pine marten. Journal of Steroid Biochemistry and Molecular
Biology, 104, 136-142.
Beerli P (2004) Effect of unsampled populations on the estimation of population sizes and migration rates between
sampled populations. Molecular Ecology, 13, 827-836.
- 170 -
Paper V
Spatial genetic structure
Beerli P (2006) Comparison of Bayesian and maximum-likelihood inference of population genetic parameters.
Bioinformatics, 22, 341-345.
Beerli P, Felsenstein J (2001) Maximum likelihood estimation of a migration matrix and effective population sizes in
n subpopulations by using a coalescent approach. Proceedings of the National Academy of Sciences of the United
States of America, 98, 4563-4568.
Beja-Pereira A, Oliveira R, Alves PC, Schwartz MK, Luikart G (2009) Advancing ecological understandings through
technological transformations in noninvasive genetics. Molecular Ecology Resources, 9, 1279-1301.
Belkhir K, Borsa P, Chikhi L, Raufaste N, Bonhomme F (2004) Genetix 4.02, Logiciel sous windows pour la génétique
des populations. Laboratoire Génome, Populations, Interactions, Université de Montpellier II, Montpellier,
France.(Raymond & Rousset 1995)
Bijlsma R, Van de Vliet M, Pertoldi C, Van Apeldoorn RC, Van de Zande L (2000) Microsatellite primers from the
Eurasian badger, Meles meles. Molecular Ecology, 9, 2216-2217.
Brainerd SM, Helldin JO, Lindstrom ER, Rolstad E, Rolstad J, Storch I (1995) Pine marten (Martes martes)
selection of resting and denning sites in scandinavian managed forests. Annales Zoologici Fennici, 32, 151-157.
Brainerd SM, Rolstad J (2002) Habitat selection by Eurasian pine martens Martes martes in managed forests of
southern boreal Scandinavia. Wildlife Biology, 8, 289-297.
Bright PW (1993) Habitat fragmentation - problems and predictions for british mammals. Mammal Review, 23,
101-111.
Broekhuizen S, Müskens GJDM. (2000) Utilization of rural and suburban habitat by pine marten Martes martes and
beech marten M. foina: species-related potential and restrictions for adaptation. Lutra, 43, 223-227.
Broquet T, Ray N, Petit E, Fryxell JM, Burel F (2006) Genetic isolation by distance and landscape connectivity in
the American marten (Martes americana). Landscape Ecology, 21, 877-889.
Brownstein MJ, Carpten JD, Smith JR (1996) Modulation of non- templated nucleotide addition by Taq DNA
polymerase: primer modifications that facilitate genotyping. Biotechniques 20, 1004–1010
Bruggeman DJ, Wiegand T, Fernandez N (2010) The relative effects of habitat loss and fragmentation on
population genetic variation in the red-cockaded woodpecker (Picoides borealis). Molecular Ecology, 19, 36793691.
Cabria MT, Gonzalez EG, Gomez-Moliner BJ, Zardoya R (2007) Microsatellite markers for the endangered
European mink (Mustela lutreola) and closely related mustelids. Molecular Ecology Notes, 7, 1185-1188.
Carr D, Bowman J, Kyle CJ, Tully SM, Koen EL, Robitaille JF, Wilson PJ (2007) Rapid homogenization of
multiple sources: Genetic structure of a recolonizing population of fishers. Journal of Wildlife Management,
71, 1853-1861.
Cegelski CC, Waits LP, Anderson NJ (2003) Assessing population structure and gene flow in Montana wolverines
(Gulo gulo) using assignment-based approaches. Molecular Ecology, 12, 2907-2918.
Clark RW, Brown WS, Stechert R, Zamudio KR (2008) Integrating individual behaviour and landscape genetics:
the population tructure of timber rattlesnake hibernacula. Molecular Ecology, 17, 719-730.
Coltman DW (2008) Molecular ecological approaches to studying the evolutionary impact of selective harvesting in
wildlife. Molecular Ecology, 17, 221-235.
Cornuet JM, Luikart G (1996) Description and power analysis of two tests for detecting recent population
bottlenecks from allele frequency data. Genetics, 144, 2001-2014.
Coulon A, Cosson JF, Angibault JM, Cargnelutti B, Galan M, Morellet N, Petit E, Aulagnier S, Hewison AJM
(2004) Landscape connectivity influences gene flow in a roe deer population inhabiting a fragmented
landscape: an individual-based approach. Molecular Ecology, 13, 2841-2850.
Coulon A, Guillot G, Cosson JF, Angibault JMA, Aulagnier S, Cargnelutti B, Galan M, Hewison AJM (2006)
Genetic structure is influenced by landscape features: empirical evidence from a roe deer population.
Molecular Ecology, 15, 1669-1679.
Craul M, Chikhi L, Sousa V, Olivieri GL, Rabesandratana A, Zimmermann E, Radespiel U (2009) Influence of
forest fragmentation on an endangered large-bodied lemur in northwestern Madagascar. Biological
Conservation, 142, 2862-2871.
Crooks KR, Sanjayan MA (2006) Connectivity Conservation: Maintaining Connections for Nature. Cambridge
University Press, Cambridge, UK.
Cushman SA, Landguth EL (2010a) Scale dependent inference in landscape genetics. Landscape Ecology, 25, 967979.
- 171 -
CHAPTER
V
Landscape Genetics
Cushman SA, Landguth EL (2010b) Spurious correlations and inference in landscape genetics. Molecular Ecology, 19,
3592-3602.
Cushman SA, McKelvey KS, Hayden J, Schwartz MK (2006) Gene flow in complex landscapes: Testing multiple
hypotheses with causal modeling. American Naturalist, 168, 486-499.
Cushman SA, McKelvey KS, Schwartz MK (2009) Use of Empirically Derived Source-Destination Models to Map
Regional Conservation Corridors. Conservation Biology, 23, 368-376.
D'Eon R, Glenn SM, Parfitt I, Fortin MJ (2002) Landscape connectivity as a function of scale and organism vagility
in a real forested landscape. Conservation Ecology, 6.
Dallas JF, Piertney SB (1998) Microsatellite primers for the Eurasian otter. Molecular Ecology, 7, 1248-1251.
Davis CS, Strobeck C (1998) Isolation, variability, and cross-species amplification of polymorphic microsatellite loci
in the family Mustelidae. Molecular Ecology, 7, 1776-1778.
Davison A, Chiba S (2003) Laboratory temperature variation is a previously unrecognized source of genotyping error
during capillary electrophoresis. Molecular Ecology Notes, 3, 321-323.
Devictor V, Julliard R, Jiguet F (2008) Distribution of specialist and generalist species along spatial gradients of
habitat disturbance and fragmentation. Oikos, 117, 507-514.
Dionne M, Caron F, Dodson JJ, Bernatchez L (2008) Landscape genetics and hierarchical genetic structure in
Atlantic salmon: the interaction of gene flow and local adaptation. Molecular Ecology, 17, 2382-2396.
Domingo-Roura X, Macdonald DW, Roy MS, Marmi J, Terradas J, Woodroffe R, Burke T, Wayne RK (2003)
Confirmation of low genetic diversity and multiple breeding females in a social group of Eurasian badgers
from microsatellite and field data. Molecular Ecology, 12, 533-539.
Evanno G, Regnaut S, Goudet J (2005) Detecting the number of clusters of individuals using the software
STRUCTURE: a simulation study. Molecular Ecology, 14, 2611-2620.
Excoffier L, Laval G, Schneider S (2005) Arlequin (version 3.0): An integrated software package for population
genetics data analysis. Evolutionary Bioinformatics, 1, 47-50.
Falush D, Stephens M, Pritchard JK (2003) Inference of population structure using multilocus genotype data:
Linked loci and correlated allele frequencies. Genetics, 164, 1567-1587.
Falush D, Stephens M, Pritchard JK (2007) Inference of population structure using multilocus genotype data:
dominant markers and null alleles. Molecular Ecology Notes, 7, 574-578.
Ferrando A, Lecis R, Domingo-Roura X, Ponsa M (2008) Genetic diversity and individual identification of
reintroduced otters (Lutra lutra) in north-eastern Spain by DNA genotyping of spraints. Conservation
Genetics, 9, 129-139.
Fischer J, Lindenmayer DB (2007) Landscape modification and habitat fragmentation: a synthesis. Global Ecology
and Biogeography, 16, 265-280.
Fleming MA, Ostrander EA, Cook JA (1999) Microsatellite markers for American mink (Mustela vison) and ermine
(Mustela erminea). Molecular Ecology, 8, 1352-1354.
Fontaine MC, Baird SJE, Piry S, Ray N, Tolley KA, Duke S, Birkun A, Ferreira M, Jauniaux T, Llavona A, Ozturk
B, Ozturk AA, Ridoux V, Rogan E, Sequeira M, Siebert U, Vikingsson GA, Bouquegneau JM, Michaux JR
(2007) Rise of oceanographic barriers in continuous populations of a cetacean: the genetic structure of
harbour porpoises in Old World waters. Bmc Biology, 5.
Francois O, Durand E (2010) Spatially explicit Bayesian clustering models in population genetics. Molecular Ecology
Resources, 10, 773-784.
Frantz AC, Cellina S, Krier A, Schley L, Burke T (2009) Using spatial Bayesian methods to determine the genetic
structure of a continuously distributed population: clusters or isolation by distance? Journal of Applied Ecology,
46, 493-505.
Frantz AC, Fack F, Muller CP, Roper TJ (2006) Faecal DNA typing as a tool for investigating territorial behaviour
of badgers (Meles meles). European Journal of Wildlife Research, 52, 138-141.
Frantz AC, Pope LC, Carpenter PJ, Roper TJ, Wilson GJ, Delahay RJ, Burke T (2003) Reliable microsatellite
genotyping of the Eurasian badger (Meles meles) using faecal DNA. Molecular Ecology, 12, 1649-1661.
Genovesi, P., Sinibaldi, I. and Boitani, L. (1997) Spacing patterns and territoriality of the stone marten. Canadian
Journal of Zoology, 75, 1966-1971
Gómez-Moliner BJ, Cabria MT, Rubines J, Garin I, Madeira MJ, Elejalde A, Aihartza J, Fournier P, Palazon S
(2004) PCR-RFLP identification of mustelid species: European mink (Mustela lutreola), American mink (M.
vison) and polecat (M. putorius) by analysis of excremental DNA. Journal of Zoology, 262, 311-316.
- 172 -
Paper V
Spatial genetic structure
Goszczynski J, Posluszny M, Pilot M, Gralak B (2007) Patterns of winter locomotion and foraging in two sympatric
marten species: Martes martes and Martes foina. Canadian Journal of Zoology-Revue Canadienne De Zoologie,
85, 239-249.
Goudet J (1995) fstat (version 1.2): a computer program to calculate F-statistics. Journal of Heredity, 86, 485–486.
Goudet J (2001) FSTAT, A program to estimate and test gene diversities and fixation indices. Version 2.9.3, updated
from Goudet 1995. Available from http://www.unil.ch/izea/softwares/fstat.html.
Guillot G (2008) Inference of structure in subdivided populations at low levels of genetic differentiation-the
correlated allele frequencies model revisited. Bioinformatics, 24, 2222-2228.
Guillot G, Estoup A, Mortier F, Cosson JF (2005a) A spatial statistical model for landscape genetics. Genetics, 170,
1261-1280.
Guillot G, Leblois R, Coulon A, Frantz AC (2009) Statistical methods in spatial genetics. Molecular Ecology, 18,
4734-4756.
Guillot G, Mortier F, Estoup A (2005b) GENELAND: a computer package for landscape genetics. Molecular Ecology
Notes, 5, 712-715.
Guillot G, Santos F, Estoup A (2008) Analysing georeferenced population genetics data with Geneland: a new
algorithm to deal with null alleles and a friendly graphical user interface. Bioinformatics, 24, 1406-1407.
Gula R, Hausknecht R, Kuehn R (2009) Evidence of wolf dispersal in anthropogenic habitats of the Polish
Carpathian Mountains. Biodiversity and Conservation, 18, 2173-2184.
Guo SW, Thompson EA (1992) Performing the exact test of hardy-weinberg proportion for multiple alleles.
Biometrics, 48, 361-372.
Gurrutxaga M, Lozano PJ, Del Barrio G (2010a) Assessing Highway Permeability for the Restoration of Landscape
Connectivity between Protected Areas in the Basque Country, Northern Spain. Landscape Research, 35, 529550.
Gurrutxaga M, Lozano PJ, del Barrio G (2010b) GIS-based approach for incorporating the connectivity of ecological
networks into regional planning. Journal for Nature Conservation, 18, 318-326.
Hajkova P, Pertoldi C, Zemanova B, Roche K, Hajek B, Bryja J, Zima J (2007) Genetic structure and evidence for
recent population decline in Eurasian otter populations in the Czech and Slovak Republics: implications for
conservation. Journal of Zoology, 272, 1-9.
Hajkova P, Zemanova B, Bryja J, Hajek B, Roche K, Tkadlec E, Zima J (2006) Factors affecting success of PCR
amplification of microsatellite loci from otter faeces. Molecular Ecology Notes, 6, 559-562.
Hajkova P, Zemanova B, Roche K, Hajek B (2009) An evaluation of field and noninvasive genetic methods for
estimating Eurasian otter population size. Conservation Genetics, 10, 1667-1681.
Hampe A, Petit RJ (2005) Conserving biodiversity under climate change: the rear edge matters. Ecology Letters, 8,
461-467.
Hardy OJ, Vekemans X (2002) SPAGEDi: a versatile computer program to analyse spatial genetic structure at the
individual or population levels. Molecular Ecology Notes, 2, 618-620.
Harris RB, Wall WA, Allendorf FW (2002) Genetic consequences of hunting: what do we know and what should we
do? Wildlife Society Bulletin, 30, 634-643.
Hedmark E, Ellegren H (2007) DNA-based monitoring of two newly founded Scandinavian wolverine populations.
Conservation Genetics, 8, 843-852.
Helldin JO (2000) Population trends and harvest management of pine marten Martes martes in Scandinavia.
Wildlife Biology, 6, 111-120.
Herr J, Schley L, Roper TJ (2009) Socio-spatial organization of urban stone martens. Journal of Zoology, 277, 54-62.
Herrman M (1994) Habitat use and spatial organization by the stone marten. In: Buskirk SW, Harestad AS, Raphael
MG, Powell RA (eds) Martens, sables and fishers. Biology and Conservation. Cornell University Press, Ithaca, pp
122–136
Holderegger R, Kamm U, Gugerli F (2006) Adaptive vs. neutral genetic diversity: implications for landscape
genetics. Landscape Ecology, 21, 797-807.
Holderegger R, Wagner HH (2008) Landscape genetics. Bioscience, 58, 199-207.
Ihaka R, Gentleman R (1996) R. A language for data analysis and graphics. Journal of Computational and Graphical
Statistics 5, 299–314.
Instituto Geográfico Nacional (2008) Base Cartográfica Numérica BCN200. Ministerio de Fomento, Madrid.
- 173 -
CHAPTER
V
Landscape Genetics
Jordan MJ, Higley M, Matthews SM, Rhodes OE, Schwartz MK, Barrett RH, Palsboll PJ (2007) Development of
22 new microsatellite loci for fishers (Martes pennanti) with variability results from across their range.
Molecular Ecology Notes, 7, 797-801.
Kurki S, Nikula A, Helle P, Linden H (1998) Abundances of red fox and pine marten in relation to the composition
of boreal forest landscapes. Journal of Animal Ecology, 67, 874-886.
Kyle CJ, Davison A, Strobeck C (2003) Genetic structure of European pine martens (Martes martes), and evidence
for introgression with M. americana in England. Conservation Genetics, 4, 179-188.
Lanszki J (2003) Feeding habits of stone martens in a Hungarian village and its surroundings. Folia Zoologica, 52,
367-377.
Lanszki J, Hidas A, Szentes K, Revay T, Lehoczky I, Jeney Z, Weiss S (2010) Genetic structure of otter (Lutra lutra)
populations from two fishpond systems in Hungary. Mammalian Biology, 75, 447-450.
Latch EK, Scognamillo DG, Fike JA, Chamberlain MJ, Rhodes OE (2008) Deciphering ecological barriers to North
American river otter (Lontra canadensis) gene flow in the Louisiana landscape. Journal of Heredity, 99, 265274.
Linnell JDC, Strand O (2000) Interference interactions, co-existence and conservation of mammalian carnivores.
Diversity and Distributions, 6, 169-176.
Liu ZJ, Ren BP, Wu RD, Zhao L, Hao YL, Wang BS, Wei FW, Long YC, Li M (2009) The effect of landscape
features on population genetic structure in Yunnan snub-nosed monkeys (Rhinopithecus bieti) implies an
anthropogenic genetic discontinuity. Molecular Ecology, 18, 3831-3846.
Livia L, Francesca V, Antonella P, Fausto P, Bernardino R (2007) A PCR-RFLP method on faecal samples to
distinguish Martes martes, Martes foina, Mustela putorius and Vulpes vulpes. Conservation Genetics, 8, 757-759.
Loiselle BA, Sork VL, Nason J, Graham C (1995) Spatial genetic-structure of a tropical understory shrub, psychotria
officinalis (Rubiaceae). American Journal of Botany, 82, 1420-1425.
López-Martín JM (2003). Aspectos de la ecología de la marta (Martes martes L. 1758) y la garduña (M. foina
Erx.1777) en los ambientes mediterráneos: interacciones con la gineta (Genetta genetta L. 1758). Ph.D.
Thesis, Universidad de Barcelona, Barcelona.
López-Martín JM (2007) Martes martes (Linnaeus, 1758). Pp. 302 – 304. In: Atlas y Libro Rojo de los Mamíferos
Terrestres de España. Palomo LJ, Gisbert J, Blanco JC (Eds.). Dirección General de Biodiversidad-SECEMSECEMU, Madrid.
Manel S, Berthoud F, Bellemain E, Gaudeul M, Luikart G, Swenson JE, Waits LP, Taberlet P, Intrabiodiv C (2007)
A new individual-based spatial approach for identifying genetic discontinuities in natural populations.
Molecular Ecology, 16, 2031-2043.
Manel S, Schwartz MK, Luikart G, Taberlet P (2003) Landscape genetics: combining landscape ecology and
population genetics. Trends in Ecology & Evolution, 18, 189-197.
Marucco F, Pletscher DH, Boitani L, Schwartz MK, Pilgrim KL, Lebreton JD (2009) Wolf survival and population
trend using non-invasive capture-recapture techniques in the Western Alps. Journal of Applied Ecology, 46,
1003-1010.
McCairns RJS, Bernatchez L (2008) Landscape genetic analyses reveal cryptic population structure and putative
selection gradients in a large-scale estuarine environment. Molecular Ecology, 17, 3901-3916.
McRae BH, Beier P, Dewald LE, Huynh LY, Keim P (2005) Habitat barriers limit gene flow and illuminate
historical events in a wide-ranging carnivore, the American puma. Molecular Ecology, 14, 1965-1977.
Mergey M (2007). Réponses des populations de martres d’Europe (Martes martes) à la fragmentation de l'habitat :
mécanismes comportementaux et consequences. Dissertation University of Reims Campagne-Ardenne,
France.
Millions DG, Swanson BJ (2007) Impact of natural and artificial barriers to dispersal on the population structure of
bobcats. Journal of Wildlife Management, 71, 96-102.
Ministerio de Medio Ambiente (1997-2006) Mapa forestal de España 1:50.000. Ministerio de Medio Ambiente,
Madrid.
Moritz C (2002) Strategies to protect biological diversity and the evolutionary processes that sustain it. Systematic
Biology, 51, 238-254.
Mucci N, Arrendal J, Ansorge H, Bailey M, Bodner M, Delibes M, Ferrando A, Fournier P, Fournier C, Godoy JA,
Hajkova P, Hauer S, Heggberget T, Heidecke D, Kirjavainen H, Krueger HH, Kvaloy K, Lafontaine L,
Lanszki J, Lemarchand C, Liukko UM, Loeschcke V, Ludwig G, Madsen AB, Mercier L, Ozolins J, Paunovic
- 174 -
Paper V
Spatial genetic structure
M, Pertoldi C, Piriz A, Prigioni C, Santos-Reis M, Luis TS, Stjernberg T, Schmid H, Suchentrunk F,
Teubner J, Tornberg R, Zinke O, Randi E (2010) Genetic diversity and landscape genetic structure of otter
(Lutra lutra) populations in Europe. Conservation Genetics, 11, 583-599.
Mullins J, Statham MJ, Roche T, Turner PD, O'Reilly C (2010) Remotely plucked hair genotyping: a reliable and
non-invasive method for censusing pine marten (Martes martes, L. 1758) populations. European Journal of
Wildlife Research, 56, 443-453.
Nabe-Nielsen J, Sibly RM, Forchhammer MC, Forbes VE, Topping CJ (2010) The Effects of Landscape
Modifications on the Long-Term Persistence of Animal Populations. Plos One, 5.
O´Connell M, Wright JM, Farid A (1996) Development of PCR primers for nine polymorphic American mink
Mustela vison microsatellite loci. Molecular Ecology, 5, 311-312.
Olivieri GL, Sousa V, Chikhi L, Radespiel U (2008) From genetic diversity and structure to conservation: Genetic
signature of recent population declines in three mouse lemur species (Microcebus spp.). Biological Conservation,
141, 1257-1271.
Palomo LJ, Gisbert J, Blanco JC (eds) (2007). Atlas y libro rojo de los mamíferos terrestres de España. Dirección
General de Biodiversidad-SECEM-SECEMU, Madrid.
Palsboll PJ, Berube M, Allendorf FW (2007) Identification of management units using population genetic data.
Trends in Ecology & Evolution, 22, 11-16.
Pavlacky DC, Goldizen AW, Prentis PJ, Nicholls JA, Lowe AJ (2009) A landscape genetics approach for quantifying
the relative influence of historic and contemporary habitat heterogeneity on the genetic connectivity of a
rainforest bird. Molecular Ecology, 18, 2945-2960.
Pearse DE, Crandall KA (2004) Beyond F-ST: Analysis of population genetic data for conservation. Conservation
Genetics, 5, 585-602.
Pereboom V, Mergey M, Villerette N, Helder R, Gerard JF, Lode T (2008) Movement patterns, habitat selection,
and corridor use of a typical woodland-dweller species, the European pine marten (Martes martes), in
fragmented landscape. Canadian Journal of Zoology-Revue Canadienne De Zoologie, 86, 983-991.
Pertoldi C, Barker SF, Madsen AB, Jorgensen H, Randi E, Munoz J, Baagoe HJ, Loeschcke V (2008) Spatiotemporal population genetics of the Danish pine marten (Martes martes). Biological Journal of the Linnean
Society, 93, 457-464.
Piggott MP, Taylor AC (2003) Remote collection of animal DNA and its applications in conservation management
and understanding the population biology of rare and cryptic species. Wildlife Research, 30, 1-13.
Pilot M, Jedrzejewski W, Branicki W, Sidorovich VE, Jedrzejewska B, Stachura K, Funk SM (2006) Ecological
factors influence population genetic structure of European grey wolves. Molecular Ecology, 15, 4533-4553.
Piry S, Luikart G, Cornuet JM (1999) BOTTLENECK: A computer program for detecting recent reductions in the
effective population size using allele frequency data. Journal of Heredity, 90, 502-503.
Pompanon F, Bonin A, Bellemain E, Taberlet P (2005) Genotyping errors: Causes, consequences and solutions.
Nature Reviews Genetics, 6, 847-859.
Pritchard JK, Stephens M, Donnelly P (2000) Inference of population structure using multilocus genotype data.
Genetics, 155, 945-959.
Proulx G, Aubry KB, Birks J, Buskirk SW, Fortin C, Frost HC, Krohn WB, Mayo L, Monakhov V, Payer D, Saeki
M, Santos-Reis M, Weir R, Zielinski WJ (2004) World distribution and status of the genus Martes in 2000.
In: Harrison DJ, Fuller AK, Proulx G (Eds) Martens and fishers (Martes) in human- altered environments: an
international perspective. New York: Springer-Verlag, pp 21–76
Quemere E, Crouau-Roy B, Rabarivola C, Louis EE, Chikhi L (2010) Landscape genetics of an endangered lemur
(Propithecus tattersalli) within its entire fragmented range. Molecular Ecology, 19, 1606-1621.
Raymond M, Rousset F (1995) GENEPOP (version-1.2) - population-genetics software for exact tests and
ecumenicism. Journal of Heredity, 86, 248-249.
Reig S (1992) Geographic-variation in pine marten (Martes-martes) and beech marten (M. foina) in europe. Journal
of Mammalogy, 73, 744-769.
Riley SPD, Pollinger JP, Sauvajot RM, York EC, Bromley C, Fuller TK, Wayne RK (2006) A southern California
freeway is a physical and social barrier to gene flow in carnivores. Molecular Ecology, 15, 1733-1741.
Rice WR (1989) Analyzing tables of statistical tests. Evolution, 43, 223–225.
- 175 -
CHAPTER
V
Landscape Genetics
Rosellini S, Osorio E, Ruiz-Gonzalez A, Isabel AP, Barja I (2008) Monitoring the small-scale distribution of
sympatric European pine martens (Martes martes) and stone martens (Martes foina): a multievidence approach
using faecal DNA analysis and camera-traps. Wildlife Research, 35, 434-440.
Rousset F (2008) GENEPOP ' 007: a complete re-implementation of the GENEPOP software for Windows and
Linux. Molecular Ecology Resources, 8, 103-106.
Rowe G, Beebee TJC (2007) Defining population boundaries: use of three Bayesian approaches with microsatellite
data from British natterjack toads (Bufo calamita). Molecular Ecology, 16, 785-796.
Ruiz-Gonzalez A, Rubines J, Berdion O, Gomez-Moliner BJ (2008) A non-invasive genetic method to identify the
sympatric mustelids pine marten (Martes martes) and stone marten (Martes foina): preliminary distribution
survey on the northern Iberian Peninsula. European Journal of Wildlife Research, 54, 253-261.
Santos MJ, Santos-Reis M (2010) Stone marten (Martes foina) habitat in a Mediterranean ecosystem: effects of scale,
sex, and interspecific interactions. European Journal of Wildlife Research, 56, 275-286.
Schoener, T. W. (1974). Resource partitioning in ecological communities. Science 185, 27–39.
Schwartz MK, Copeland JP, Anderson NJ, Squires JR, Inman RM, McKelvey KS, Pilgrim KL, Waits LP, Cushman
SA (2009) Wolverine gene flow across a narrow climatic niche. Ecology, 90, 3222-3232.
Schwartz MK, McKelvey KS (2009) Why sampling scheme matters: the effect of sampling scheme on landscape
genetic results. Conservation Genetics, 10, 441-452.
Schwartz MK, Monfort SL (2008) Genetic and endocrine tools for carnivore surveys. In: Long RA, Mackay P,
Zielinski WJ, Ray JC (eds) Noninvasive survey methods for carnivores. Island Press, Washington DC, pp 238–
262
Segelbacher G, Cushman SA, Epperson BK, Fortin MJ, Francois O, Hardy OJ, Holderegger R, Taberlet P, Waits
LP, Manel S (2010) Applications of landscape genetics in conservation biology: concepts and challenges.
Conservation Genetics, 11, 375-385.
Segelbacher G, Manel S, Tomiuk J (2008) Temporal and spatial analyses disclose consequences of habitat
fragmentation on the genetic diversity in capercaillie (Tetrao urogallus). Molecular Ecology, 17, 2356-2367.
Spear SF, Balkenhol N, Fortin MJ, McRae BH, Scribner K (2010) Use of resistance surfaces for landscape genetic
studies: considerations for parameterization and analysis. Molecular Ecology, 19, 3576-3591.
Stenglein JL, Waits LP, Ausband DE, Zager P, Mack CM (2010) Efficient, Noninvasive Genetic Sampling for
Monitoring Reintroduced Wolves. Journal of Wildlife Management, 74, 1050-1058.
Storch I, Lindstrom E, Dejounge J (1990) Diet and habitat selection of the pine marten in relation to competition
with the red fox. Acta Theriologica, 35, 311-320.
Storfer A, Murphy MA, Evans JS, Goldberg CS, Robinson S, Spear SF, Dezzani R, Delmelle E, Vierling L, Waits LP
(2007) Putting the 'landscape' in landscape genetics. Heredity, 98, 128-142.
Taberlet P, Griffin S, Goossens B, Questiau S, Manceau V, Escaravage N, Waits LP, Bouvet J (1996) Reliable
genotyping of samples with very low DNA quantities using PCR. Nucleic Acids Research, 24, 3189-3194.
Taberlet P, Luikart G (1999) Non-invasive genetic sampling and individual identification. Biological Journal of the
Linnean Society, 68, 41-55.
Taberlet P, Waits LP, Luikart G (1999) Noninvasive genetic sampling: look before you leap. Trends in Ecology &
Evolution, 14, 323-327.
Taylor PD, Fahrig L, Henein K, Merriam G (1993) Connectivity is a vital element of landscape structure. Oikos, 68,
571-573.
Valière N (2002) GIMLET: a computer program for analysing genetic individual identification data. Molecular
Ecology Notes, 2, 377-379.
Van Oosterhout C, Hutchinson WF, Wills DPM, Shipley P (2004) MICRO-CHECKER: software for identifying
and correcting genotyping errors in microsatellite data. Molecular Ecology Notes, 4, 535-538.
Vekemans X, Hardy OJ (2004) New insights from fine-scale spatial genetic structure analyses in plant populations.
Molecular Ecology, 13, 921-935.
Virgos E, Casanovas JG (1998) Distribution patterns of the Stone marten (Martes foina Erxleben, 1777) in
Mediterranean mountains of central Spain. Zeitschrift Fur Saugetierkunde-International Journal of Mammalian
Biology, 63, 193-199.
Virgos E, Garcia FJ (2002) Patch occupancy by stone martens Martes foina in fragmented landscapes of central
Spain: the role of fragment size, isolation and habitat structure. Acta Oecologica-International Journal of
Ecology, 23, 231-237.
- 176 -
Paper V
Spatial genetic structure
Virgos E, Recio MR, Cortes Y (2000) Stone marten (Martes foina Erxleben, 1777) use of different landscape types in
the mountains of central Spain. Zeitschrift Fur Saugetierkunde-International Journal of Mammalian Biology, 65,
375-379.
Waits LP, Paetkau D (2005) Noninvasive genetic sampling tools for wildlife biologists: A review of applications and
recommendations for accurate data collection. Journal of Wildlife Management, 69, 1419-1433.
Weir BS, Cockerham CC (1984) Estimating F-statistics for the analysis of population structure. Evolution, 38,
1358–1370
Williams BW, Scribner KT (2010) Effects of multiple founder populations on spatial genetic structure of
reintroduced American martens. Molecular Ecology, 19, 227-240.
Wilson GJ, Frantz AC, Pope LC, Roper TJ, Burke TA, Cheeseman CL, Delahay RJ (2003) Estimation of badger
abundance using faecal DNA typing. Journal of Applied Ecology, 40, 658-666.
Worboys GL, Francis W, Lockwood M. (eds) (2010) Connectivity conservation management: a global guide. Earthscan,
London.
Zalewski A, Jedrzejewski W (2006) Spatial organisation and dynamics of the pine marten Martes martes population
in Bialowieza Forest (E Poland) compared with other European woodlands. Ecography, 29, 31-43.
Zalewski A, Piertney SB, Zalewska H, Lambin X (2009) Landscape barriers reduce gene flow in an invasive
carnivore: geographical and local genetic structure of American mink in Scotland. Molecular Ecology, 18,
1601-1615.
Zhan AB, Hu JJ, Hu XL, Zhou ZC, Hui M, Wang S, Peng W, Wang ML, Bao ZM (2009) Fine-Scale Population
Genetic Structure of Zhikong Scallop (Chlamys farreri): Do Local Marine Currents Drive Geographical
Differentiation? Marine Biotechnology, 11, 223-235.
- 177 -
CHAPTER
V
Landscape Genetics
- 178 -
CHAPTER V
Landscape Genetics
PAPER VI
Landscape genetics as a tool for the empirical assessment of a
regional ecological network: The European pine marten
(Martes martes) as a target-species
ABSTRACT
Coherent ecological networks (ENs) composed of core areas linked by ecological corridors are being
developed worldwide with a view on the maintenance of landscape connectivity and biodiversity
conservation. Although these ENs are developed in accordance with the precautionary principle and
ecological theory, they have been subject to criticisms based on the lack of empirical evidence regarding
their efficiency in achieving their goals. Landscape genetics in combination with non-invasive genetic
sampling provides a valuable framework for the study of rare and elusive species in order to understand
how landscape features influence gene flow, by correlating inter-individual pairwise genetic distances with
least-cost modelling. Here we analysed the accuracy of the parameters used in a resistance map which was
drawn up in the design of the regional EN in the Basque Country (North Spain) with regards to the
geneflow of a target species: the European pine marten. Moreover, different binary resistance maps which
covered a gradient from greater to lesser preference of the focal species in relation to forest environments
have been evaluated to determine the land uses which restrict gene flow of the pine marten. Our results
confirmed the utility of the corridors designed and the need to keep taking them into account in land use
planning. Also, we identified areas of intensive agriculture and the major road networks as landscape
features which limit pine marten gene flow. To the best of our knowledge, this is the first time that a GIS
modelled EN has been evaluated through a landscape genetics approach.
Key words: Landscape genetics, non-invasive genetic sampling, Martes martes, landscape connectivity,
least-cost modelling, land use planning
- 179 -
Least Cost distance Analysis
Paper VI
INTRODUCTION
Long-term biodiversity conservation requires the preservation of ecological and evolutionary processes,
such as gene flow, dispersal movements and change shifts in biota (Opdam and Washer 2004). The ability
of individuals to move across changing landscapes is crucial for maintaining regional populations (Fahrig
2007; Cushman et al. 2009). The preservation of these processes requires in turn that landscape
connectivity be preserved, especially when we take into account the synergetic effects of habitat
fragmentation and climate change (Opdam & Washer 2004). Connectivity is defined as the degree to
which landscape facilitates or impedes movement of organisms among resource patches (Taylor et al.
1993). Connectivity is species-specific and describes the response of individuals to landscape features and
the patterns of dispersal and gene flow that result from these individual responses (Brooks 2003).
Landscape connectivity depends to a large extent on the spatial configuration of the habitat and on land
use (Opdam & Washer 2004).
Ecological networks have been promoted in the last few decades as coherent systems composed of core
areas linked by ecological corridors capable of facilitating the dispersal, migration and gene flow of wild
species in landscapes and regions (Bennett & Wit 2001; Jongman & Pungetti 2004). They are configured
and managed with the objective of maintaining ecological functions and conserving biodiversity (Bennett
& Witt 2001). Although the development of ecological networks is based on the precautionary principle
and on ecological theory (Jongman & Pungetti 2004), the absence of empirical evidence regarding their
effectiveness and the difficulty in obtaining this evidence has been a focus of criticism regarding the extent
to which they have in fact ensured landscape connectivity and increased biodiversity conservation (Boitani
et al. 2007).
In the design of ecological networks there is a need to predict regional ecological corridors and to quantify
the degree of expected landscape connectivity between specific areas (Segelbacher et al. 2010). ‘Least-cost
modelling’ is one commonly employed approach for designing ecological corridors (Adriaensen et al.
2003), in which resistance values are assigned to distinct habitat or land use types and the least-cost paths
(LCP) between specific locations are calculated using a geographical information system (GIS). How
landscape influences effective distances between locations is calculated as the accumulated cost through the
least cost paths (Adriaensen et al. 2003; Ray 2005). For most organisms, setting the resistance values is a
difficult process in which expert judgement and data available in the literature play an important role
(Adriaensen et al. 2003; Beier et al. 2008; Spear et al. 2010).
- 180 -
CHAPTER V
Landscape Genetics
Reliable mapping of corridors must be based on a correct representation of the local resistance relative to
movement of the organism in focus (Beier et al. 2008; Spear et al. 2010). Landscape genetics, a research
area that integrates landscape ecology, population genetics and spatial statistics, provides a valuable
framework for testing the influence of landscape structure and composition on dispersal and gene flow
(Manel et al. 2003; Balkenhol et al. 2009). It facilitates detection of the ways in which the resistance of
given landscape patches affects dispersal movements and, in consequence, gene flow (Holderegger &
Wagner 2006; Segelbacher et al. 2010). Thus, one of the principal applications of landscape genetics in
landscape planning and conservation biology is precisely to empirically test resistance maps or friction
surfaces (Epps et al. 2007; Shirk et al. 2010). This facilitates the optimal design of ecological corridors
(Epps et al. 2007; Beier et al. 2008; Cushman et al. 2009), the detection of barriers to gene flow and the
identification of the landscape features which favour dispersal (Coulon et al. 2004; Broquet et al. 2006;
Wang et al. 2008; Schwartz et al. 2009; Shirk et al. 2010).
Gene flow is a function of the degree to which a landscape resists dispersal, which, in turn, is highly
dependent on both the species and the landscape in question (Hastings & Harrison 1994; Spear et al.
2010). Thus, gene flow estimated through highly polymorphic genetic markers such as microsatellites
helps to validate landscape connectivity modelled by LCP (Coulon et al. 2004; Broquet et al. 2006; Epps
et al. 2007). The most common methodology adopted for landscape genetics studies aiming to relate
landscape patterns with gene flow processes is to correlate inter-individual genetic distances with effective
distances as derived from GIS LCP models and Euclidean distances with methods such as simple (Mantel
1967; Coulon et al. 2004, Broquet et al. 2006; Wang et al. 2008; Zhu et al. 2010) and partial Mantel tests
(Smouse et al. 1986; Cushman et al. 2006; Schwartz et al 2009; Wasserman et al. 2010). Thus, testing the
departure of genetic distances from isolation by distance pattern (IBD) (Wright 1943), through the
inclusion of effective distances, provides evidence of the specific effect of landscape features on genetic
structure.
Landscape genetics has shifted towards individual-based sampling and analysis, especially when organisms
are continuously distributed (Segelbacher et al. 2010). However, sufficient sample collection for this
purpose is a difficult task, especially in rare and elusive species in which sampling is a limiting factor (BejaPereira et al. 2009). In this context, molecular methods incorporating non-invasive sampling via the
collection of scats or hairs have become common for population monitoring of carnivores, providing a
valuable DNA source for genetic surveys (Taberlet et al. 1999; Piggott & Taylor 2003; Waits & Paetkau
2005). Noninvasive genetic sampling allows us to address studies of wildlife species without the need to
- 181 -
Least Cost distance Analysis
Paper VI
capture or even observe them (Waits & Paetkau 2005). Thus this study has the added advantage of
permitting us to infer landscape connectivity without needing to intervene directly in the species in focus.
In 2005 a regional ecological network was drawn up in the Basque Country (North Spain). This was
achieved by delimiting the ecological corridors linking protected areas (Gurrutxaga et al. 2010a). A
functional group of forest mammal species was selected so as to obtain parameters for a generic resistance
map which would serve as a basis for the design of the ecological corridors by least-cost modelling
(Adriaensen et al. 2003). These mammals were considered suitable target-species due to their sensitivity to
the most important recent fragmentation and homogenization dynamics in the regional landscape, such as
road construction, urbanization and agrarian intensification (Jongman 2002). The resistance of the
landscape matrix was parameterized through bibliographical review and expert opinion (Gurrutxaga et al.
2010a). The design of the resistance map was based on the assignation of different resistance levels to
different land uses; this was especially so given that the ecological network was drawn up specifically for
use in the framework of land use planning. The regional government of the Basque country incorporated
that coherent ecological network as a reference for the environmental assessment of plans, programmes
and projects in 2005 (Gurrutxaga et al. 2010a). In addition to its intrinsic internal relevance, the Basque
country has been chosen for its crucial role in the regulation of relevant biotic flows. This is because of its
strategic location between two important biodiversity reservoirs in south-western Europe, the mountain
chains of the Pyrenees and the Cantabrian Range (Jongman et al. 2006; Worboys et al. 2010).
Consequently, the preservation and restoration of connectivity in this transitional area between mountain
ranges requires adequate land use planning (Workboys et al. 2010).
In the set of functional forest mammals which was taken into account in the design of the coherent
ecological network in the Basque Country the European pine marten (Martes martes) is the most
woodland-dwelling species (Proulx et al. 2004). The pine marten is generally associated with forest
habitats, mainly mature forests (Zalewski & Jędrzejewski 2006; Proulx et al. 2004). Nonetheless, they have
also been recently reported in fragmented landscapes where woods consist of isolated, small fragments
within an agricultural landscape matrix (Pereboom et al. 2008; Balestrieri et al. 2010). Deforestation and
forest fragmentation have been reported to affect the distribution and density of pine martens (Brainerd et
al. 1995; Kurki et al. 1998), which are believed to need a minimum woodland area to survive (Zalewski &
Jędrzejewski 2006) and tend to avoid treeless areas (Storch et al. 1990, Brainerd & Rolstad 2002,
Pereboom et al. 2008, Ruiz-González et al. 2008). Given their strong associations with structural
complexity in forests, the species is particularly sensitive to human influences on their habitats, including
habitat loss, and landscape-scale effects of habitat fragmentation (Bright 2000, Pereboom et al. 2008).
- 182 -
CHAPTER V
Landscape Genetics
Consequently, the pine marten is a species which is well suited to studies focused on the effects of forest
fragmentation on genetic structure and gene flow.
In a recent study (Ruiz-Gonzalez et al. in prep) Bayesian analysis methods suggested that the pine marten
in the study area was divided into at least two populations where several human-induced landscape
transformations (i.e. roads, forest fragmentation) could well be mainly responsible for the observed genetic
discontinuities. However, individual-based analyses relating landscape structure to genetic distance across
heterogeneous landscapes permit a more accurate assessment of multiple alternative hypotheses relating
landscape pattern to gene flow, and thus indentifying the landscape variables responsible for the inferred
genetic structure.
The main objective of this research is to develop an individual-based non-invasive landscape genetic
approach in order to evaluate alternative resistance maps relating landscape structure to gene flow
Specifically, we aim to determine firstly i) whether or not the resistance map with which the regional
ecological network was originally designed in the Basque Country was correctly parametrized regarding the
European pine marten gene flow. This is done with a view to obtaining empirical validation of the EN
from the ecology of a target species which is sensitive to relevant landscape dynamics; and additionally test
ii) different binary resistance maps which covered a gradient from greater to lesser preference of the focal
species in relation to forest environments in order to identify which land uses favour or impede genetic
interchange of the European pine marten in the study area.
METHODS
Study area and spatial data
The region of the Basque Country is located to the north of the Iberian Peninsula (Fig. 1) and belongs to
the Atlantic and Mediterranean biogeographical regions. It comprises an area of 7235 km2 and has a
population density of 298 inhabitants per square kilometre. There are three NUTS-3 provinces: Alava;
Biscay; and Guipuzcoa. Approximately, forests cover 28% of the area, forestry plantations 29%, nonwooded mountains 24%, cultivated land 14%, and urban land and infrastructures 5.7%.
- 183 -
Least Cost distance Analysis
Paper VI
Figure 1. Distribution map of land uses in the study area.
Land use information was obtained in vector format from the most recent forest map of Spain (Spanish
Ministry for the Environment, 2006) and from national road network maps (Spanish National
Geographic Institute, 2008).
Non-invasive genetic sampling and species identification
We used non-invasive scat sampling to collect genetic samples from the Martes sp. (Martes martes and
Martes foina) in the study area between 2005 and 2009. We conducted a multi-stage sampling scheme, in
which samples from a pilot study were used to assess the appropriateness of the sampling with respect to
the research questions. Thus, two scat-based surveys were conducted between 2004 and 2009 all over the
sympatric range of both marten species in the study area. The first one, conducted in the 2004-2005
period was used to initially estimate the distribution range of the two sympatric species of the genus Martes
in the study area and isolate genetic samples of the focal species (M. marrtes). The second, conducted
between 2006 and 2009 was used to refine species distribution assessment and to obtain a higher number
of M. martes fresh samples for microsatellite genotyping after a PCR-RFLP species identification process.
Additionally, fresh tissue specimens from road-kill pine martens were included in the study, when
- 184 -
CHAPTER V
Landscape Genetics
possible. Universal Transversal Mercator (UTM) coordinates were recorded for all the samples collected
using a global positioning system (Garmin eTtrex). The faecal samples were stored in autoclaved tubes
containing ethanol 96% and frozen at −20°C until processed (Ruiz González et al. 2008). DNA was
isolated from tissues and scat using the Qiagen DNeasy Tissue DNA (Qiagen, Hombrechtikon,
Switzerland) and DNA Stool MiniKit (Qiagen, Hombrechtikon, Switzerland) according to the
manufacturer’s instructions, respectively. As pine marten faeces cannot be distinguished from those of the
sympatric stone marten (Martes foina), which is widespread in the study area, and can also be easily
confused with those of other carnivores (Davison et al. 2002), a molecular technique was applied for the
identification of faecal samples. Species identification was accomplished by a polymerase chain reaction –
restriction fragment length polymorphism (PCR-RFLP) method, providing for an effective genetic
identification of sympatric marten species according to Ruiz-González et al. (2008).
Microsatellite analyses and individual identification
Identification of individual pine martens used nuclear DNA following methods in Ruiz-González et al. (in
prep.) All the faecal samples identified by the PCR-RFLP method as pine marten were genotyped at 15
variable microsatellite loci using a multiplex protocol according to Ruiz-González et al. (in prep.) (Table
S1). DNA quality was initially screened by PCR-amplifying each DNA sample four times at four loci
(Multiplex 1: MP0188; MP0059; Gg-7; Ma-1). Only samples showing > 50% positive PCRs were further
amplified four times at the remaining 11 loci. Samples with ambiguous results after four amplifications per
locus or with <50% successful amplifications across loci were removed from further analysis as they were
not considered reliable genotypes. Multiplex PCR products were run on an ABI (Foster City, CA)
3130XL automated sequencer (Applied Biosystems), with the internal size standard GS500 LIZ™ (Applied
Biosystems). Fragment analyses were conducted using the ABI software Genemapper 4.0.
GIMLET software v 1.3.4 (Valière 2002) was used to calculate the probabilities of identity (PID and PIDsibs) so as to quantify the efficacy in discriminating the fifteen loci in combination. Consensus genotypes
from four replicates were reconstructed using GIMLET, accepting heterozygotes if the two alleles were
seen at least in two replicates and homozygotes if a single allele was seen at least in three replicates.
GIMLET was also used to estimate genotyping errors: allelic dropout (ADO) and false alleles (FA)
(Taberlet et al. 1996; Pompanon et al. 2005)
Genetic diversity and pairwise individual genetic distances
- 185 -
Least Cost distance Analysis
Paper VI
We summarized genetic variation through the number of alleles per locus (A) and expected (HE) and
observed (HO) heterozygosities using GENETIX v 4.05.2 software (Belkhir et al. 2004). Estimates of
pairwise linkage disequilibria for each pair of loci and deviation from HWE were tested using the exact test
implemented in GenePop version 4.0 (Raymond & Rousset 1995; Rousset, 2008) in each previously
defined subpopulation in the study area, according to Ruiz-González et al (in prep), and for the whole
dataset. Statistical significance was evaluated by running a Markov Chain Monte Carlo (MCMC)
consisting of 10,000 batches of 10,000 iterations each, with the first 10,000 iterations discarded before
sampling (Guo & Thompson 1992). Significance levels were adjusted with sequential Bonferroni
correction in order to correct for the effect of multiple tests (Rice 1989). MICRO-CHECKER software
(Van Oosterhout et al. 2004) was used to check for potential scoring errors, the presence of null alleles and
linkage disequilibrium.
The ar inter-individual genetic distance (Rousset 2000) were computed using the program SPAGeDI
(Hardy & Vekemans 2002) since this parameter of relatedness has been successfully applied to infer the
effect of landscape on genetic structure in other vertebrates (Coulon et al. 2004; Broquet et al. 2006).
Resistance maps and effective distances calculation
Ecological network resistance map
Different resistance maps were used with a raster cell resolution set to 50 m. In the first place a resistance
map was used which was analogous to that used in the design of the regional ecological network of the
Basque Country (EN resistance map). The friction surface outlined in 2005 (Gurrutxaga et al. 2010a), was
updated with the available vectorial information regarding land uses in the study area (Spanish Ministry
for the Environment 2006; Spanish National Geographic Institute 2008). Raster breaks in linear barrier
elements were avoided by the reinforcement of the size of national roads and highways (Adriaensen et al.
2003). Sections of highways which run through viaducts or tunnels were assigned the resistance value
corresponding to the land use of the surrounding area. Equally, a friction map was used which was a
variant of that used previously, with a view to testing the effect of noticeably decreasing the resistance
attributed to national roads, highways, urban areas, reservoirs and quarries (ENnb map) (Table 1).
Binary resistance maps
Different binary resistance maps were used for the purpose of determining the land uses which were
favourable and unfavourable to the dispersal movements of the martens. Broquet et al.’s (2006)
methodology was adopted for the assignment of friction values. In this way, preferential land uses for
- 186 -
CHAPTER V
Landscape Genetics
dispersal were assigned a value of 1, while non-favourable to dispersal habitat sites were assigned a value of
50. As pine martens are believed to need a minimum woodland area to survive (Zalewski & Jędrzejewski
2006) and tend to avoid treeless areas (Brainerd and Rolstad 2002; Pereboom et al. 2008; Ruiz-González
et al. 2008), the different binary resistance maps (A to G) covered a gradient from greater to lesser
preference of the focal species in relation to forest environments, ranging from strictly forest land up to
and including open spaces (Table 1, Fig. 2). Different land uses were combined to create a single predictor
variable (each resistance map). Additionally, in some resistance maps we included information regarding
the river network as a means of dispersal (Table 1). Raster breaks in fluvial linear elements were avoided by
the reinforcement of the size of rivers (Adriaensen et al. 2003). Friction surfaces were also used which
included as barrier features the national roads (resistance 200), highways, urban areas, reservoirs and
quarries (resistance 1000).
Table 1. Friction values corresponding to the resistance maps taken evaluated: EN) a resistance map analogous to
that used in the design of the ecological network of the Basque country; ENnb) a variant of the latter which
suppresses the barrier effect of national roads, highways, urban areas, water reservoirs and quarries; A to G) binary
resistance maps, on a gradient from greater to lesser preference of the focal species in relation to forest environment.
Maps with letter “r” include the river network with resistance 1. Maps with letter “b” include the barrier effect of
national roads, highways, urban areas, reservoirs and quarries.
Land uses
Forests
Forestry
plantations
Scrubland
Agroforestry
mosaics
Pastures
and
meadows
Rocks
Crops
Wetlands
National
roads
Highways,
urban areas,
reservoirs
and quarries
Rivers
A
1
B
1
Br
1
C
1
D
1
Dr
1
Db
1
E
1
Er
1
Eb
1
F
1
G
1
EN
1
ENnb
1
50
1
1
1
1
1
1
1
1
1
1
1
10
10
50
50
50
1
1
1
1
1
1
1
1
1
5
5
50
50
50
50
1
1
1
1
1
1
1
1
20
20
50
50
50
50
50
50
50
1
1
1
1
1
30
30
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
1
50
50
1
1
50
40
60
100
40
60
100
50
50
50
50
50
50
200
50
50
200
50
50
200
50
50
50
50
50
50
50
1000
50
50
1000
50
50
1000
50
-
-
1
-
-
1
-
-
1
-
-
-
-
-
- 187 -
Least Cost distance Analysis
Paper VI
Figure 2. Resistance maps A, B, C, D, E, Eb, F y G. Green-coloured cells represent areas with resistance
value 1, orange-coloured cells those with resistance value 50.
The effective and Euclidean distances between each pair of individuals were calculated with
PATHMATRIX 1.1. (Ray 2005). Pair-wise effective distances between individuals were calculated as the
accumulated cost through the least cost paths (LCP) throughout each friction surface (Adriaensen et al
2003; Ray 2005) (Fig. 3).
- 188 -
CHAPTER V
Landscape Genetics
Figure 3. Least cost paths (LCP) obtained between the 101 pine marten individuals in accordance with
the EN resistance map, analogous to that used in the design of the corridors in the ecological network of
the Basque Country.
Relationship between genetic and geographical distances: effective vs Euclidean
The pairwise genetic distances matrix (ar) was correlated with different matrices of geographical distances
encompassing a total number of 5151 pairwise comparisons i) Euclidean distance, to determine whether
the patterns of differentiation follow an isolation by distance pattern (null hypothesis) and ii) the effective
distances calculated for each of the resistance maps, to infer landscape structure effects on gene flow. We
also computed Partial Mantel tests (Smouse et al. 1986) between genetic distance and effective distance,
factoring out the effects of Euclidean distance. The correlation between distance matrixes was calculated
by means of the Mantel test (Mantel 1967) and partial Mantel tests (Smouse et al. 1986) as implemented
in Xlstat software (Addinsoft, New York, USA) using genetic distances and the log of the effective and
Euclidean distances, as recommended by Rousset (1997) for tests performed on species using a two
dimensional landscape. P-values were obtained using a permutation procedure (10,000 permutations). We
identified the model most supported as the one with the highest significant correlation value.
- 189 -
Least Cost distance Analysis
Paper VI
RESULTS
Non-invasive sample collection and species identification
Out of 733 faecal samples collected from the entire study area, 141 were discarded because they were not
fresh samples or because they presumably belong to the same individual (samples separated by <1km). 494
out of 592 analyzed samples were classified as one of the target species (M. martes and M. foina) by our
PCR-RFLP method. Thus, unequivocal species identification was possible in 83.45% of the samples. We
effectively identified 232 faecal samples as stone marten and 262 as pine marten. Additionally, we
obtained 57 tissue samples from road-killed pine martens.
Out of 262 faecal samples identified as pine marten, 108 were not included to the microsatellite
genotyping procedure. These samples correspond to the sampling period from 2004-2005, which was used
for a first distribution assessment of sympatric martens in the study area. Thus, 213 pine marten samples
(154 faecal samples and 59 tissue samples) were used for microsatellite genotyping.
Individual identification, genotype checking and genetic diversity,
The first quality-screening test, based on 4 replicates of four loci, was not passed by 73 non-invasive
samples (47.40%), which were immediately discarded. The remaining 81 samples (52.59%) were
amplified at the other 11 loci. After multiple-tubes genotyping 27 samples from this sub-set (17.53% from
the total analyzed samples) were then discarded because they showed < 50% PCR success, or because of
high failure rates. Full multilocus microsatellite genotypes were obtained for the remaining 54 samples.
(66.67% from the samples that passed the screening and 35.06 from the total samples analyzed). The
observed average error rates across loci were: ADO = 0.218 and FA = 0.017. PID analysis showed that the
set of 15 loci would produce an identical genotype with a probability of 1.69x10-10, and with a probability
of 4.45x10-5 for a full-sib. Thus, only 4.45 pine martens in 100,000 siblings are expected to share by
chance an identical genotype, suggesting no “shadow effect” (i.e. all the genotypes identify distinct
individuals; Mills et al. 2000), and that matching genotypes are recaptures of the same individual.
After a regrouping procedure we identified 42 individual genotypes from faecal samples. The 59 tissue
samples were correctly genotyped at 15 loci and provided new individuals. In total we identified 113
genotypes that correspond with 101 different individuals. The number of times each individual was
detected in the survey varied from 1 to 3, with a total number of 12 re-samplings.
The overall pine marten dataset (n=101) is not at HWE (p<0.001). These results were consistent with the
existence of population structure, as has previously been demonstrated by Ruiz-González et al. (in prep).
- 190 -
CHAPTER V
Landscape Genetics
In contrast, we observed no deviation from HWE in each of the previously inferred populations according
to Ruiz-González et al (in prep). Linkage disequilibrium was not apparent for any pair of loci within any
of the subpopulations after performing Bonferroni corrections. The results of MICROCHECKER
indicated that null alleles were apparently present at two loci: Mp0188 and Ma19. However, the estimated
frequency of the null allele at these loci occurred at a relatively low frequency (0.041-0.065).
For all 101 pine martens, the average observed (HO) and expected (HE) heterozygosity values were 0.53
and 0.58, respectively (Table S1). All 15 loci were variable with total numbers of alleles ranging between 3
and 8 per locus.
Correlations between genetic and geographical distances
Ecological network resistance map
A significant positive correlation was obtained between the genetic distances and Euclidean distances
(r=0.214; p<0.0001), bearing clear witness to the existence of a pattern of isolation by distance (IBD) (Fig.
4). The correlation between genetic and effective distance calculated on the basis of the EN map was
noticeably greater (r=0.256; p<0.0001) (Fig. 4). The degree of correlation when using the ENnb map was
less than that which obtained with EN, though still greater than that obtained using Euclidean distance
(r=0.236; p<0.0001) (Fig 4.). Both EN and ENnb models appeared better supported than the null model
of IBD as this latter retained a significant positive relationship with ar-based genetic distance even after
factoring out the effects of Euclidean distance (Table 2). However, the original model used in the design
of the ecological network (EN), which included a higher barrier effect for national roads, highways, urban
areas, reservoirs and quarries was better supported than the alternative model (ENnb).
- 191 -
Least Cost distance Analysis
Paper VI
Figure 4. Pearson correlation coefficients (Mantel r) between genetic distance and effective distance in the
different resistance maps examined. The correlation coefficient (Mantel r) between genetic and Euclidean
distance (isolation by distance, IBD) was 0.214. In all cases the correlation was significant (p<0.001).
Table 2. Partial Mantel’s test correlation with genetic distance (ar) for each resistance map after factoring
out the effects of the null model of isolation by distance (IBD).
Correlation
ar x EN
ar x ENnb
ar x A
ar x B
ar x Br
ar x C
ar x D
ar x Dr
ar x Db
ar x E
ar x Er
ar x Eb
ar x F
ar x G
Partial r
0.145
0.103
0.072
0.079
0.021
0.089
0.098
0.033
0.130
0.136
0.101
0.174
0.136
0.103
- 192 -
P value
<0.0001
<0.0001
<0.0001
<0.0001
0.04
<0.0001
<0.0001
0.021
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
CHAPTER V
Landscape Genetics
Binary resistance maps
When using binary resistance maps, the correlation between genetic and effective distance gradually
increased on including, in addition to natural forest (Map A), forestry plantations (B), scrublands (C),
agroforestry mosaics (D), and pastures and meadows (E) as environments favouring dispersal (Fig. 4). This
correlation did not vary on including rocky areas as dispersal environments (F), while it decreased on
including cultivated land (G). The correlation reached its maximum value on including in resistance map
E the barrier effect of highways, urban areas, reservoirs and quarries (friction map Eb) (Fig. 4). The
correlation decreased on including in the resistance maps the total river network as an environment
favouring dispersal (Br, Dr, Er). In all models, the relationship between genetic distance and effective
distance was always significant when Euclidean distance was factored out of the relationship (Table 2).
The correlation values after factoring out the effect of Euclidean distance showed the same pattern as the
correlation values obtained by means of a simple Mantel test. The best two supported models are Eb and
EN according to simple and partial Mantel r values.
DISCUSSION
Empirical evaluation of an ecological network through landscape genetics
Maps of corridors are commonly used by land management groups, but unfortunately are more often the
product of expert opinion rather than empirical data (Boitani et al. 2007). To our knowledge, this is the
first time that a GIS modelled ecological network (Gurrutxaga et al. 2010a) was evaluated through a
landscape genetics approach. Moreover, the use of non invasive genetic sampling has the added advantage
of permitting us to infer landscape connectivity without needing to intervene directly in the species in
focus.
In this study, even though contemporary gene flow follows a pattern of isolation by distance, the
resistances of the landscape matrix which were used in the design of the ecological network of the Basque
Country (EN friction map) (Gurrutxaga et al., 2010a) provide a much better explanation for the genetic
pattern observed. Oftentimes, geographical distance and landscape resistance will simultaneously influence
gene flow (Trizio et al. 2005). Individuals that are geographically distant are often also separated by highcost distances, making the distinction between purely spatial and landscape effects particularly challenging
in practice (Cushman et al. 2006). However, in this study while Euclidean distance was significantly
correlated with genetic distance the correlation was better using the effective distances. Moreover, partial
- 193 -
Least Cost distance Analysis
Paper VI
Mantel tests suggested that IBD it was not the main factor that strongly influenced pine marten gene flow
as the landscape structure had a significant independent effect on genetic distance when we controlled for
Euclidean distance.
The comparison with the correlation coefficients obtained in other similar studies based on the genetic
distance between individuals (e.g. Coulon et al. 2004; Broquet et al. 2006; Wang et al. 2008; Schwartz et
al. 2009), shows that the improvement obtained with respect to the pattern of isolation by distance is
relatively high. Even though some authors have used the length of LCPs as an indicator of landscape
connectivity between pairs of individuals (Broquet et al. 2006; Coulon et al. 2004), this study has used the
accumulated cost of LCPs since this constitutes a more exact indicator of the effect of landscape structure
on dispersal ability and gene flow. This can be attributed to the fact that individuals who move along
different LCPs of equal length are affected by the differing efforts, and thus they must exert due to the
differing resistance they meet with according to the land uses of the trajectories they move across (Stevens
et al. 2006). A number of studies have used binary surfaces to investigate landscape influences on genetic
connectivity (Broquet et al. 2006; Coulon et al. 2004). In this study, the resistance map used (EN)
parameterizes with different resistance values varying along a fairly wide range. Complex resistance maps
like this are more realistic than simple binary models which divide the habitat into favourable and nonfavourable, a tendency which has predominated in previous studies (Coulon et al. 2004; Broquet et al.
2006; Schwartz et al. 2009). These models assume that the surrounding matrix is uniformly inhospitable,
which really constitutes an important simplification of the reality (Holderegger & Wagner 2008). Models
reflecting a mosaic, such as EN resistance map, on the other hand provide with a heterogeneous model
which more closely reflect the differing patch-like nature of landscapes.
Influence of land uses on gene flow: insights into pine marten ecology
The most consistent marten-habitat relation appears to be a general association with forest habitats, and
avoidance of open, non-forested habitats (Proulx et al. 2004; Pereeboon et al. 2009). Thus, the marten’s
unwillingness to cross open habitats may restrict the species’ ability to disperse and colonise new forested
areas (Caryl 2009). Many previous landscape genetic studies have evaluated a single landscape resistance
hypothesis relative to a null model of isolation by distance (e.g. Coulon et al. 2004; Broquet et al. 2006;
Schwartz et al. 2009). To avoid the probable risk of inferential error due to spurious correlations with
alternative but untested hypotheses (Cushman & Landguth 2010), we evaluated several landscape models
independently. Here, different binary resistance maps, with different hypotheses concerning the friction to
pine marten dispersal arising from land uses, were used to infer landscape effects on gene flow. In order to
- 194 -
CHAPTER V
Landscape Genetics
obviate the limitations associate with binary resistance maps, which reflect an oversimplification of the
landscape matrix, in this study we combined different land uses so as to create a single predictor variable
(each resistance map from A-G), including step by step land uses ranging from strict forest to open spaces.
This methodology has allowed us to determine which land uses favour or impede dispersal and, in
consequence, the gene flow of a species. The correlation results obtained with the binary models suggest
that it is not alone forest masses which serve as favourable environments for dispersal. Scrubland,
agroforest mosaic, and grassland habitats also favour dispersal, since the correlation increases as, step by
step, these environments are included as predictor variables of pine marten geneflow. These results are in
consonance with recent studies, based on radio tracking, carried out on European pine martens, which
provide new data which differ substantially from traditional descriptions in the scientific literature (Caryl,
2008; Pereboom et al. 2008). These studies show that martens are not exclusively confined to extensive
forest patches but that they also use other patches including scrubland and agroforest mosaics (Mergey
2007; Caryl 2008; Pereboom et al. 2008; Balestrieri et al. 2010). Indeed, the inclusion of scrub habitat in
marten home ranges is likely to be related to its role in the connectivity of forest habitats (Caryl 2008;
Pereboom et al. 2008). In the same way, the improvement in correlation on including pastures and
meadows seems to indicate that the species does not always renounce to cross these open spaces when there
is forest habitat in the immediate vicinity as has been previously suggested by radiotracking data
(Pereboom et al. 2008). This is precisely the case in the area under study, where pastures and meadows are
inserted in areas in the immediate vicinity of forest. The inclusion of homogeneous croplands reduces the
correlation between genetic distance and effective distance, suggesting that zones with intensive agriculture
impede species dispersal. This could be due to the scarcity of natural vegetation in these zones and the
distance separating them from forested areas. When we include the total river network as a medium
favourable to dispersal, the correlation diminishes. This can be attributed to the excess in connectivity
between different favourable habitats without taking into account the ecological quality of the medium
itself and its various crossings.
The inclusion of the barrier effect of major roads and urban areas leads to a substantial improvement in
understanding the genetic pattern which was detected, since the correlation with models Eb and EN was
greater than that obtained with models E and ENnb, respectively. A previous study based on Bayesian
analysis detected these landscape features as one of the principal factors in explaining the genetic
discontinuity between the two inferred marten populations in the study area (Ruiz-González et al. in
prep). Both LCP and Bayesian analyses confirm the importance of these variables as barriers or moderators
of gene flow. Moreover, these results reaffirm the need to guarantee the permeability of infrastructures by
- 195 -
Least Cost distance Analysis
Paper VI
means of adequate wildlife crossings (Gurrutxaga et al. 2010b). Also adequate urban planning is required,
so as to avoid urban sprawl and the conversion of the remaining natural or seminatural areas especially
along valley bottoms.
Since our ability to detect the effects of landscape structure on genetic differentiation depends on both the
landscape features used and the relative costs of each feature, different resistance values could provide
different results (Spear et al. 2010). In this study, due to computational constraints we evaluated binary
models with only 1 and 50 friction values. Consequently, a more detailed insight into the effect of a range
of different resistance values would be desirable. However, similar landscape genetics studies have used
friction values similar to those used in the present study in order to infer landscape effect on gene flow
(Coulon et al. 2004; Broquet et al. 2006; Wang et al. 2008; Schwartz et al. 2009). Moreover, in order to
reduce these limitations we have used both binary and complex resistance maps so as to parameterize the
resistances inherent to each land use in different ways. Taking into account partial Mantel r values, the
best supported models were EN, Eb, and E. In spite of they differ in the friction assignment procedure
and in the relative values assigned to each land use, yet each model provide similar results and highlights
similar landscape effects on gene flow, even after factoring out the effects of Euclidean distance. The
agreement between binary and complex models, in spite of the difference in the friction values assigned to
each land use, provides support for the results obtained. Additionally, the concordance between landscape
genetics model inferences and species ecology provided by radio tracking (Pereboom et al. 2008), reinforce
the validity of the conclusions arrived at.
Ruiz-González et al. (2008) found that pine marten occurrence in the study area is highly dependent on
the presence of forest and consequently sensitive to forest fragmentation as has been previously suggested
in other studies (Brainerd et al. 1995, Kurki et al. 1998). Nevertheless, the presence of forest habitats is
not the only factor which explains the gene flow, indicating that the habitat selection and gene flow of
pine martens may be driven by different factors (Spear et al. 2009). This may be because gene flow is
driven by mating and dispersal events and habitat selection reflects the behaviour of individual organisms
to maximize fitness within home ranges.
Similar least cost path modeling has also been conducted on American marten in Ontario (Canada)
(Broquet et al. 2006) and on European pine marten in Ardennes, La Bresse, and L’Isere (France) (Mergey
2007) both showing the importance of forest structure for dispersal across large landscapes. In addition,
more recently Wasserman et al. (2010) has extended the least cost path approach by incorporating multiscale modeling and improved model selection approach for the study of American pine marten gene flow
- 196 -
CHAPTER V
Landscape Genetics
in Idaho (USA). In this study, showed that gene flow in the Northern Idaho (USA) American marten
population is driven by a gradient of landscape resistance that is a function of elevation, which was a proxy
for snowpack, with marten avoiding lower elevations and dispersing in mid to high elevation montane
forests.
Even though human-induced changes to the landscape have been quite recent in the study area, the
genetic structure nonetheless reflects current landscape structure. Landguth et al. (2010) suggested that
individual-based landscape genetic approaches are accurate enough to detect the effects of existing
landscape features on genetic structure and connectivity, as the lag time to barrier detection is short (1-15
generations), at least for animals with high dispersal abilities. According to the relationship between home
range and dispersal distance outlined in Bowman et al. (2002) median and maximal dispersal distance for
the pine marten are approximately 8km and 46km for males and 7km and 40km for females, if we
consider the mean size of the home range to be 1,3km2 and 1km2 respectively for pine marten males and
females in the study area (O. Berdión unpublished data). Thus, taking into account the relatively high
dispersal abilities of the pine marten, we assume that the strong correlation found in this study between
contemporary landscape patterns and pine marten gene flow reflects current landscape effects more than
past events. However, we cannot discard the influence of other historical events and landscapes which
could have long-term effects that confound inferences about the impacts of current landscape features on
gene flow (Landguth et al. 2010). Consequently, controlling for the influence of the historic landscape
configuration on genetic structure may be necessary if we are to disentangle historical and contemporary
processes affecting genetic variation (Zellmer & Knowles 2009)
CONCLUSION
Individual-based analyses correlating the genetic distances between individuals with the cost distances
between them obtained from multiple landscape resistance hypotheses is useful approach to understand
the effect of landscape matrix on gene flow. Overall, this study suggests that a standard isolation-bydistance model is not sufficient to explain the observed genetic pattern, and including landscape variables
through a resistance map significantly improves the prediction power of this standard model. Moreover,
the results obtained suggest that the population connectivity of pine martens in the study area may be
vulnerable to habitat loss and fragmentation processes, as the optimal resistance map identified suggest
that pine marten gene flow is predominately driven by the presence of natural and semi-natural land uses
(i.e forests, forestry plantations, scrublands, agroforestry mosaics, pastures and meadows) and the
avoidance of intensively human altered land uses (i.e. crops, reservoirs, roads, highways and urban areas).
- 197 -
Least Cost distance Analysis
Paper VI
Our results confirm also that landscape genetic studies are useful tools for the empirical assessment of a
regional ecological network. The parameterization found in the resistance map which was used to design
the regional corridors in the Basque Country (north Spain) was adequate to explain pine marten gene
flow. Given the importance of this species as bio-indicator of relevant landscape dynamics, the results
obtained underline the importance of continuing to incorporate regional corridors as a reference for land
use planning and management in the study area in order to preserve landscape connectivity.
- 198 -
CHAPTER V
Landscape Genetics
ACKNOWLEDGEMENTS
This study has been partially funded by the Basque Government through the Research group on “Systematics,
Biogeography and Population Dynamics” (Ref. IT317-10; GIC10/76) and by the University of the Basque Country
(UPV-EHU) and the Department of Environment, Territorial Planning, Agriculture and Fisheries (Basque
Government) through IKT S.A under the University-Enterprise research program (Ref. UE07/02). Ruiz-González
holds a Ph.D. fellowship awarded by the Dept. of Education Universities and Research of the Basque Government
(Ref. BFI06.396). The authors wish to thank the following persons and institutions for supplying tissue and faecal
samples: F. Urra (VRFN-Navarre Government); The technical staff of the Natural Parks of Alava (DFA); Museo de
Ciencias Naturales de Álava; P. Lizarraga and L. Lorza (CRF Martioda-DFA); J.M. Fernandez (IAN&IKT); N. Ruiz
de Azua; G. Belamendia (CEA); J. Pinedo (DFA); H. Aguirre; P. Pérez; G. Dominguez; F. Canales and M.A.
Campos (CRN); E. Arberas; Oskar Berdión; M. Corral; J. López de Luzuriaga; I. Aginako; J. Ruiz Guijarro; I.
Martínez; E. Díaz; J.M. Pérez de Ana; D. Vado; A. Eguia; A. García de Albéniz; I. Zuberogoitia; M. Barral
(NEIKER).
REFERENCES
Adriaensen F, Chardon JP, De Blust G, Swinnen E, Villalba S, Gulinck H, Matthysen E (2003) The application of
'least-cost' modelling as a functional landscape model. Landscape and Urban Planning, 64, 233-247.
Balestrieri A, Remonti L, Ruiz-Gonzalez A, Gomez-Moliner BJ, Vergara M, Prigioni C (2010) Range expansion of
the pine marten (Martes martes) in an agricultural landscape matrix (NW Italy). Mammalian Biology, 75,
412-419.
Balkenhol N, Gugerli F, Cushman SA, Waits LP, Coulon A, Arntzen JW, Holderegger R, Wagner HH (2009)
Identifying future research needs in landscape genetics: where to from here? Landscape Ecology, 24, 455463.
Beier P, Majka DR, Spencer WD (2008) Forks in the road: Choices in procedures for designing wildland linkages.
Conservation Biology, 22, 836-851.
Beja-Pereira A, Oliveira R, Alves PC, Schwartz MK, Luikart G (2009) Advancing ecological understandings through
technological transformations in noninvasive genetics. Molecular Ecology Resources, 9, 1279-1301.
Belkhir K, Borsa P, Chikhi L, Raufaste N, Bonhomme F (2004) Genetix 4.02, Logiciel sous windows pour la génétique
des populations. Laboratoire Génome, Populations, Interactions, Université de Montpellier II, Montpellier,
France.(Raymond & Rousset 1995)
Bennett G, Wit P (2001) The development and application of ecological networks: a review of proposals, plans and
programmes. Amsterdam: AIDEnvironment.
Boitani L, Falcucci A, Maiorano L, Rondinini C (2007) Ecological networks as conceptual frameworks or
operational tools in conservation. Conservation Biology, 21, 1414-1422.
Bowman J, Jaeger JAG, Fahrig L (2002) Dispersal distance of mammals is proportional to home range size. Ecology,
83, 2049-2055.
Brainerd SM, Helldin JO, Lindstrom ER, Rolstad E, Rolstad J, Storch I (1995) Pine marten (Martes martes)
selection of resting and denning sites in scandinavian managed forests. Annales Zoologici Fennici, 32, 151157.
Brainerd SM, Rolstad J (2002) Habitat selection by Eurasian pine martens Martes martes in managed forests of
southern boreal Scandinavia. Wildlife Biology, 8, 289-297.
Bright PW (2000) Lessons from lean beasts: conservation biology of the mustelids. Mammal Review, 30, 217-226.
Brooks CP (2003) A scalar analysis of landscape connectivity. Oikos, 102, 433-439.
- 199 -
Least Cost distance Analysis
Paper VI
Broquet T, Ray N, Petit E, Fryxell JM, Burel F (2006) Genetic isolation by distance and landscape connectivity in
the American marten (Martes americana). Landscape Ecology, 21, 877-889.
Caryl FM (2008) Pine marten diet and habitat use within a managed coniferous forest. PhD University of Stirling.
Scotland.
Coulon A, Cosson JF, Angibault JM, Cargnelutti B, Galan M, Morellet N, Petit E, Aulagnier S, Hewison AJM
(2004) Landscape connectivity influences gene flow in a roe deer population inhabiting a fragmented
landscape: an individual-based approach. Molecular Ecology, 13, 2841-2850.
Cushman SA, Landguth EL (2010) Spurious correlations and inference in landscape genetics. Molecular Ecology, 19,
3592-3602.
Cushman SA, McKelvey KS, Hayden J, Schwartz MK (2006) Gene flow in complex landscapes: Testing multiple
hypotheses with causal modeling. American Naturalist, 168, 486-499.
Cushman SA, McKelvey KS, Schwartz MK (2009) Use of Empirically Derived Source-Destination Models to Map
Regional Conservation Corridors. Conservation Biology, 23, 368-376.
Epps CW, Wehausen JD, Bleich VC, Torres SG, Brashares JS (2007) Optimizing dispersal and corridor models
using landscape genetics. Journal of Applied Ecology, 44, 714-724.
Fahrig L (2007) Non-optimal animal movement in human-altered landscapes. Functional Ecology, 21, 1003-1015.
Guo SW, Thompson EA (1992) Performing the exact test of hardy-weinberg proportion for multiple alleles.
Biometrics, 48, 361-372.
Gurrutxaga M, Lozano PJ, Del Barrio G (2010a) Assessing Highway Permeability for the Restoration of Landscape
Connectivity between Protected Areas in the Basque Country, Northern Spain. Landscape Research, 35,
529-550.
Gurrutxaga M, Lozano PJ, del Barrio G (2010b) GIS-based approach for incorporating the connectivity of ecological
networks into regional planning. Journal for Nature Conservation, 18, 318-326.
Hardy OJ, Vekemans X (2002) SPAGEDi: a versatile computer program to analyse spatial genetic structure at the
individual or population levels. Molecular Ecology Notes, 2, 618-620.
Holderegger R, Wagner HH (2006) A brief guide to landscape genetics. Landscape Ecology, 21, 793-796.
Holderegger R, Wagner HH (2008) Landscape genetics. Bioscience, 58, 199-207.
Jongman RHG (2002) Homogenisation and fragmentation of the European landscape: ecological consequences and
solutions. Landscape and Urban Planning, 58, 211-221.
Jongman RHG, Bouwma, IM, Van Doorn A (2006). Indicative map of the Pan–European ecological network in
Western Europe. Wageningen: Alterra.
Jongman RHG, Pungetti G (Eds) (2004) Ecological networks and greenways: concept, design, implementation.
Cambridge: Cambridge University Press.
Kurki S, Nikula A, Helle P, Linden H (1998) Abundances of red fox and pine marten in relation to the composition
of boreal forest landscapes. Journal of Animal Ecology, 67, 874-886.
Landguth EL, Cushman SA, Schwartz MK, McKelvey KS, Murphy M, Luikart G (2010) Quantifying the lag time
to detect barriers in landscape genetics. Molecular Ecology, 19, 4179-4191.
Mantel N (1967) Detection of disease clustering and a generalized regression approach. Cancer Research, 27, 209-&.
Mergey M (2007). Réponses des populations de martres d’Europe (Martes martes) à la fragmentation de l'habitat :
mécanismes comportementaux et consequences. Dissertation University of Reims Campagne-Ardenne, France.
Mills LS, Citta JJ, Lair KP, Schwartz MK, Tallmon DA (2000) Estimating animal abundance using noninvasive
DNA sampling: Promise and pitfalls. Ecological Applications, 10, 283-294.
Opdam P, Wascher D (2004) Climate change meets habitat fragmentation: linking landscape and biogeographical
scale levels in research and conservation. Biological Conservation, 117, 285-297.
Pereboom V, Mergey M, Villerette N, Helder R, Gerard JF, Lode T (2008) Movement patterns, habitat selection,
and corridor use of a typical woodland-dweller species, the European pine marten (Martes martes), in
fragmented landscape. Canadian Journal of Zoology-Revue Canadienne De Zoologie, 86, 983-991.
Piggott MP, Taylor AC (2003) Remote collection of animal DNA and its applications in conservation management
and understanding the population biology of rare and cryptic species. Wildlife Research, 30, 1-13.
Pompanon F, Bonin A, Bellemain E, Taberlet P (2005) Genotyping errors: Causes, consequences and solutions.
Nature Reviews Genetics, 6, 847-859.
Proulx G Aubry KB, Birks J, Buskirk SW, Fortin C, Frost HC, Krohn WB, Mayo L, Monakhov V, Payer D, Saeki
M, Santos-Reis M, Weir R, Zielinski WJ (2004) World distribution and status of the genus Martes in 2000.
- 200 -
CHAPTER V
Landscape Genetics
In: Harrison DJ, Fuller AK, Proulx G (Eds) Martens and fishers (Martes) in human- altered environments: an
international perspective. New York: Springer-Verlag, pp 77–98
Ray N (2005) PATHMATRIX: a geographical information system tool to compute effective distances among
samples. Molecular Ecology Notes, 5, 177-180.
Raymond M, Rousset F (1995) GENEPOP (version-1.2) - population-genetics software for exact tests and
ecumenicism. Journal of Heredity, 86, 248-249.
Rousset F (1997) Genetic differentiation and estimation of gene flow from F-statistics under isolation by distance.
Genetics, 145, 1219-1228.
Rousset F (2000) Genetic differentiation between individuals. Journal of Evolutionary Biology, 13, 58-62.
Rousset F (2008) GENEPOP ' 007: a complete re-implementation of the GENEPOP software for Windows and
Linux. Molecular Ecology Resources, 8, 103-106.
Ruiz-Gonzalez A, Rubines J, Berdion O, Gomez-Moliner BJ (2008) A non-invasive genetic method to identify the
sympatric mustelids pine marten (Martes martes) and stone marten (Martes foina): preliminary
distribution survey on the northern Iberian Peninsula. European Journal of Wildlife Research, 54, 253261.
Schwartz MK, Copeland JP, Anderson NJ, Squires JR, Inman RM, McKelvey KS, Pilgrim KL, Waits LP, Cushman
SA (2009) Wolverine gene flow across a narrow climatic niche. Ecology, 90, 3222-3232.
Segelbacher G, Cushman SA, Epperson BK, Fortin MJ, Francois O, Hardy OJ, Holderegger R, Taberlet P, Waits
LP, Manel S (2010) Applications of landscape genetics in conservation biology: concepts and challenges.
Conservation Genetics, 11, 375-385.
Shirk AJ, Wallin DO, Cushman SA, Rice CG, Warheit KI (2010) Inferring landscape effects on gene flow: a new
model selection framework. Molecular Ecology, 19, 3603-3619.
Smouse PE, Long JC, Sokal RR (1986) Multiple-regression and correlation extensions of the mantel test of matrix
correspondence. Systematic Zoology, 35, 627-632.
Spanish Ministry for the Environment. (2006) Mapa forestal de España 1:50.000. Ministerio de Medio Ambiente,
Madrid.
Spanish National Geographic Institut (2008) Base cartográfica numérica BCN200. Ministerio de Fomento, Madrid.
Spear SF, Balkenhol N, Fortin MJ, McRae BH, Scribner K (2010) Use of resistance surfaces for landscape genetic
studies: considerations for parameterization and analysis. Molecular Ecology, 19, 3576-3591.
Stevens VM, Verkenne C, Vandewoestijne S, Wesselingh RA, Baguette M (2006) Gene flow and functional
connectivity in the natterjack toad. Molecular Ecology, 15, 2333-2344.
Storch I, Lindstrom E, Dejounge J (1990) Diet and habitat selection of the pine marten in relation to competition
with the red fox. Acta Theriologica, 35, 311-320.
Storfer A, Murphy MA, Spear SF, Holderegger R, Waits LP (2010) Landscape genetics: where are we now?
Molecular Ecology, 19, 3496-3514.
Taberlet P, Griffin S, Goossens B, Questiau S, Manceau V, Escaravage N, Waits LP, Bouvet J (1996) Reliable
genotyping of samples with very low DNA quantities using PCR. Nucleic Acids Research, 24, 3189-3194.
Taberlet P, Waits LP, Luikart G (1999) Noninvasive genetic sampling: look before you leap. Trends in Ecology &
Evolution, 14, 323-327.
Taylor PD, Fahrig L, Henein K, Merriam G (1993) Connectivity is a vital element of landscape structure. Oikos, 68,
571-573.
Trizio I, Crestanello B, Galbusera P, Wauters LA, Tosi G, Matthysen E, Hauffe HC (2005) Geographical distance
and physical barriers shape the genetic structure of Eurasian red squirrels (Sciurus vulgaris) in the Italian
Alps. Molecular Ecology, 14, 469-481.
Valiere N (2002) GIMLET: a computer program for analysing genetic individual identification data. Molecular
Ecology Notes, 2, 377-379.
Van Oosterhout C, Hutchinson WF, Wills DPM, Shipley P (2004) MICRO-CHECKER: software for identifying
and correcting genotyping errors in microsatellite data. Molecular Ecology Notes, 4, 535-538.
Vignieri SN (2005) Streams over mountains: influence of riparian connectivity on gene flow in the Pacific jumping
mouse (Zapus trinotatus). Molecular Ecology, 14, 1925-1937.
Waits LP, Paetkau D (2005) Noninvasive genetic sampling tools for wildlife biologists: A review of applications and
recommendations for accurate data collection. Journal of Wildlife Management, 69, 1419-1433.
- 201 -
Least Cost distance Analysis
Paper VI
Wang YH, Yang KC, Bridgman CL, Lin LK (2008) Habitat suitability modelling to correlate gene flow with
landscape connectivity. Landscape Ecology, 23, 989-1000.
Wasserman TN, Cushman SA, Schwartz MK, Wallin DO (2010) Spatial scaling and multi-model inference in
landscape genetics: Martes americana in northern Idaho. Landscape Ecology, 25, 1601-1612.
Worboys GL, Francis W, Lockwood M. (eds) (2010) Connectivity conservation management: a global guide. Earthscan,
London
Zalewski A, Jedrzejewski W (2006) Spatial organisation and dynamics of the pine marten Martes martes population
in Bialowieza Forest (E Poland) compared with other European woodlands. Ecography, 29, 31-43.
Zellmer AJ, Knowles LL (2009) Disentangling the effects of historic vs. contemporary landscape structure on
population genetic divergence. Molecular Ecology, 18, 3593-3602.
- 202 -
CHAPTER V
Landscape Genetics
Table S1. Properties of the 15 microsatellite loci multiplexed used in this study and summary of the genetic
variability. The table includes: number of alleles (NA) and observed (HO) and expected (HE) heterozygosities for each
locus and for whole data set.
MULTIPLEX
Locus
NA
HE
HO
MULT_1
Gg-7
5
0.66
0.59
Ma-1
4
0.56
0.45
MP0059
5
0.65
0.49
MP0188
3
0.46
0.35
Lut-453
3
0.50
0.57
Mel-1
5
0.65
0.69
Mel-10
5
0.66
0.57
Lut-435
7
0.42
0.28
Ma-19
4
0.54
0.55
Mvi-57
6
0.37
0.36
Mvi072
8
0.71
0.62
Lut-615
2
0.47
0.53
Ma-2
5
0.68
0.62
Mer41
6
0.77
0.68
Mlut-27
4
0.66
0.62
4.13
0.58
0.53
MULT_2
MULT_3
MULT_4
Mean
- 203 -
Least Cost distance Analysis
Paper VI
- 204 -
From the studies performed in this PhD thesis the following conclusions can be drawn:
1. There is an incomplete taxonomic and evolutionary framework for a significant portion of the Martes
complex. The current knowledge about intraspecific genetics, population genetics and landscape genetics
of genus Martes is limited and devoted to only a few species (i.e Martes martes; Martes americana; Martes
pennanti).
2. The mtDNA groups inferred for Martes martes in this study shows a strict phylogeographic pattern
throughout the species range with the presence of three major phylogroups, each of them related to
specific biogeographic regions which could probably represent different ecotypes. Overall, our study
indicates a complex phylogeographic history for M. martes indicating a mixed pattern of recolonization of
central and northern Europe from both Mediterranean and non-Mediterranean refugia, providing new
insights into the cryptic northern glacial refugia. Pleistocene climatic conditions have played an important
role in initiating phylogeographic differentiation as well as sculpting this pre-existing phylogeographic
variety into today’s sister species M. martes and M. zibellina.
3. The PCR-RFLP method developed to identify samples of M. martes and M. foina is a reliable, efficient,
time-saving and cost-effective procedure for improving our knowledge of the spatial distributions and bioecological traits of these two sympatric species. The application of this method to non-invasively collected
samples allowed us to: i) infer sympatric distribution areas of both marten species in the northern Iberian
Peninsula, ii) monitor the small-scale distribution of sympatric martens in combination with camera-traps
(NW Spain), iii) asses the range expansion of the pine marten in an agricultural landscape matrix (NW
Italy) and iv) reliably investigate pine marten food habits.
4. The novel multiplex panel of 15 microsatellites loci designed is reliable and effective for faecal DNA
genotyping of sympatric marten species (M. martes and M. foina), in order to i) identify individuals, ii)
infer population structure, and iii) determine genetic variability. The application of this panel facilitated
species distinction discarding the presence of putative hybrids between both marten species. Our results
showed that the differences in the level of expertise between sample collectors significantly influence the
- 205 -
Concluding
remarks CONCLUDING REMARKS
success rate of microsatellite genotyping of pine marten faecal samples but not the species identification
success rate.
5. Non-invasive genetic sampling of faeces in combination with landscape genetics represent a valuable
framework for the study of rare and elusive species in order to understand how landscape and
environmental features influence population genetic structure.
5.1 Bayesian clustering methods (Geneland and Stucture), supported by HWE analysis, Fst
values, and the low migration rates between groups showed that the pine marten have a clear
spatial genetic structure with the existence of different genetic clusters showing nonoverlapping distribution. Our data suggested that habitat loss and fragmentation could be
one of the main causes of the spatial genetic structure found in a typical forest dwelling
species. Other factors, such as, direct persecution, probable interspecific competition with the
stone marten and the circumstance that the study area is situated at the limit of the pine
marten distribution, could be acting synergically to shape its current spatial genetic structure. 5.2 Individual-based least cost distance analysis suggested that the population connectivity of
pine martens in the study area may be vulnerable to habitat loss and fragmentation processes,
as the optimal resistance map identified put forward that pine marten gene flow is
predominately driven by the presence of natural and semi-natural land uses (i.e forests,
forestry plantations, scrublands, agroforestry mosaics, pastures and meadows) and the
avoidance of intensively human altered landscapes (i.e. crops, reservoirs, roads, highways and
urban areas).
6. Our results confirm that landscape genetic studies are useful tools for the empirical assessment of
regional ecological networks. The parameterization found in the resistance map which was used to design
the regional corridors in the Basque Country (north Spain) was adequate to explain pine marten gene
flow. Given the importance of this species as bio-indicator of relevant landscape dynamics, the results
obtained underline the importance of continuing to incorporate regional corridors as a reference tool for
land use planning and management in order to preserve landscape connectivity.
- 206 -
Appendices
1
Appendix A
(PaperIIIa)
Rosellini S, Osorio E, Ruiz-González A, Piñeiro A, Barja I
(2008)
Monitoring the small-scale distribution of sympatric
European pine martens (Martes martes) and stone martens
(Martes foina): a multievidence approach using faecal DNA
analysis and camera-traps.
Wildlife Research, 35: 434–440
Appendix B
(PaperIIIb)
Balestrieri A, Remonti L, Ruiz-González A, Gómez-Moliner
BJ, Vergara M, Prigioni C (2010)
Range expansion of the pine marten (Martes martes) in an
agricultural landscape matrix (NWItaly).
Mammalian Biology, 75: 412–419
Appendix c
(PaperIIIc)
Balestrieri A, Remonti L, Ruiz-González A, Capelli E,
Vergara M, Gómez-Moliner BJ, Prigioni C.
Food habits of genetically identified pine marten (Martes
martes) expanding in agricultural lowlands (NW Italy).
Acta Theriologica. In press
- 207 -
Paper IIIa
Appendix A
Appendix A
PAPER IIIa
Rosellini S, Osorio E, Ruiz-González A, Piñeiro A, Barja I (2008)
Monitoring the small-scale distribution of sympatric European
pine martens (Martes martes) and stone martens (Martes foina): a
multievidence approach using faecal DNA analysis and cameratraps
Wildlife Research, 35: 434–440
- 209 -
Appendix A
Paper IIIa
CSIRO PUBLISHING
Wildlife Research, 2008, 35, 434–440
www.publish.csiro.au/journals/wr
Monitoring the small-scale distribution of sympatric
European pine martens (Martes martes) and stone martens
(Martes foina): a multievidence approach using faecal
DNA analysis and camera-traps
Stefano Rosellini A,E, Enrique OsorioB, Aritz Ruiz-González C, Ana Piñeiro A
and Isabel BarjaD
A
Departamento de Biología, Unidad Zoología, Facultad de Ciencias, Universidad Autónoma
de Madrid, Campus de Cantoblanco, 28049 Madrid, Spain.
B
Departamento de Zoología, Facultad de Biología, Universidad de León, Campus de Vegazana,
24271 León, Spain.
C
Departamento de Zoología y Biología Celular Animal, Facultad de Farmacia, Universidad del
País Vasco, C/Paseo de la Universidad 7, 01006 Vitoria, Spain.
D
Departamento de Biología y Ciencias del Medio Ambiente, Facultad de Ciencias Experimentales,
Universidad SEK, C/Cardenal Zúñiga 12, 40003 Segovia, Spain.
E
Corresponding author. Email: [email protected]
Abstract. The European pine marten (Martes martes) and stone marten (Martes foina) are two closely related mustelids
that live sympatrically over a large area of Europe. In the northern Iberian Peninsula, the distribution ranges of both species
overlap extensively. The objectives of this study were (1) to verify whether, on a small scale, both species also live
sympatrically and (2) to compare camera traps and scat DNA as methods for detecting marten species. The study was
conducted in a protected area (province of Ourense, north-west Spain), which covers 6700 ha. To test the sympatry
hypothesis, 90 fresh faecal samples, identified as faeces of genus Martes on the basis of their morphology, were collected
from June 2004 to August 2006. The specific identification of faecal samples was conducted using polymerase chain
reaction–restriction fragment length polymorphism (PCR-RFLP) techniques. In addition, 20 camera-traps (916 cameratrap-nights) were in operation during the study period. Of the faecal samples collected, 88.8% were attributed to the
European pine marten, while the remaining 11.2% were not amplified by PCR and thus could not be assigned. The
European pine marten was identified in 57.9% of the photos of carnivores and the stone marten was not detected in any.
The faecal DNA analysis and camera-trap results supported previous conclusions about habitat preferences and the
distribution of the two species obtained using other methods. The two non-invasive methods that were used in this study
were shown to be reliable techniques that can be employed simultaneously, because each method has advantages and
disadvantages that are influenced by the size of the area inventoried, sampling effort, and cost and efficiency of the method.
The data gathered using these methods provided important information on the understanding of trophic and competitive
interactions between the species.
Introduction
The European pine marten (Martes martes Linnaeus, 1758) and
stone marten (Martes foina Erxleben, 1777), species similar in
both morphology and feeding habits, are closely related
mustelids that live broadly sympatrically over an extensive area
of Europe. Limited knowledge of their ecological niches in sympatric areas, their extremely elusive behaviour and the high
degree of difficulty in distinguishing their faeces on the basis of
morphology alone, contributes to the difficulty of studying and
monitoring their populations.
The European pine marten occupies most of Europe, from
Mediterranean biotopes to Fennoscandia taiga, and western
Siberia and Iran (Proulx et al. 2004). The stone marten is present
throughout continental Europe, but it is absent from Great
Britain and Ireland as well as most of the Mediterranean islands
© CSIRO 2008
except Crete (Proulx et al. 2004). On the Iberian Peninsula, the
European pine marten has a Eurosiberian distribution and is also
present on the islands of Majorca and Minorca (Mitchell-Jones
et al. 1999; Palomo and Gisbert 2002); the stone marten is
present throughout the Iberian Peninsula (Fig. 1) (Palomo and
Gisbert 2002).
The European pine marten is associated primarily with
Eurosiberian deciduous and coniferous forests (Delibes 1983;
Mitchell-Jones et al. 1999) and prefers mature forests and areas
with permanent watercourses (Barja 2005a). In contrast, the
stone marten can survive in a variety of habitats across its distribution range, including fragmented woodlands, villages and
their periphery, and areas near farms (Waechter 1975; Delibes
1983; Sacchi and Meriggi 1995; Blanco 1998; Mitchell-Jones
10.1071/WR07030
1035-3712/08/050434
Small-scale distribution of stone and pine marten
Wildlife Research
et al. 1999; Virgós and Garcia 2002; Lanszki 2003). Libois and
Waechter (1991) also observed that the stone marten tends to
principally occupy deforested and altered areas when the
European pine marten is present. In addition, its predilection for
rocky habitats allows it to use buildings and other man-made
structures (Delibes 1983). However, this species appears to be
highly flexible as it is also found in typical forest habitats in
Russia and is therefore not exclusively dependent on the presence of rocky terrain (Waechter 1975). The stone marten is also
more thermophilic than the European pine marten, allowing it to
occupy urban environments (Waechter 1975). The ecological
requirements of both species can sometimes overlap; in these
circumstances, the European pine marten stays in forested landscapes and the stone marten seems to be displaced to more
urbanised areas (Delibes 1983).
Accurate species identification is a key step in conservation
biology and a necessary component of wildlife management and
conservation studies. However, determining presence and
absence of elusive or cryptic species can be logistically difficult,
particularly when relying solely on field signs such as faeces,
hair or tracks (Piggott and Taylor 2003). Traditionally, this kind
of information is subjected to a variety of morphological
analyses for definitive species identification (Kohn and Wayne
1997). Nevertheless, there are situations in which such samples,
deposited by sympatrically occurring carnivores of similar body
size, cannot be assigned at the species level on the basis of morphology alone (Farrell et al. 2000; Davison et al. 2002; Birks
et al. 2004; Kurose et al. 2005). Misidentification of species
from faeces is probably common, and has been indicated for different sympatric carnivore species (Paxinos et al. 1997; Ernest
et al. 2000; Farrell et al. 2000; Davison et al. 2002; Palomares
et al. 2002; Dalén et al. 2004; Sugimoto et al. 2006; Pilot et al.
2007). Traditionally, the presence of martens has been determined from data obtained from road kills or hunting information, live-trapping surveys, sighting surveys (Messenger and
Birks 2000), track plates and camera-traps (Zielinski and
Kucera 1995), fur-snagging devices (Messenger and Birks
2000; Lynch et al. 2006) and scat-based surveys, which have
several limitations for detecting and identifying different marten
435
species (Birks et al. 2004). Each method has advantages and
disadvantages that are influenced by the size of the area inventoried, sampling effort, and cost and efficiency of the method
(Birks et al. 2004). Methods of non-invasive genetic sampling
of animal populations and camera-trapping are becoming more
common for identification of mammalian species (Karanth
1995; Zielinski and Kucera 1995; Paxinos et al. 1997;
Murakami 2002; Palomares et al. 2002; Azlan and Sharma
2003; Gómez-Moliner et al. 2004; Swann et al. 2004; Kurose
et al. 2005; Trolle and Kéry 2005). Because the faeces of the
European pine and stone marten cannot be reliably separated on
the basis of morphology alone (Marchesi et al. 1989; Pilot et al.
2007), the application of methods that will positively distinguish the two could be an invaluable tool for conducting noninvasive surveys to determine the presence and distribution,
trophic and habitat requirements, niche overlap and competitive
interactions of the mustelids in sympatric areas.
Although several studies on the distribution of mammals
have been conducted on the Iberian Peninsula (Blanco et al.
1992; Rodriguez and Delibes 1992; Virgós 2001; Palomo and
Gisbert 2002), there is a remarkable absence of information on
the distribution of the European pine marten and the bioecological relationship between it and the stone marten.
Consequently, the goal of this work was to apply a multievidence approach incorporating two different survey techniques (faecal DNA analysis and camera-traps) to monitor the
small-scale distribution of the two marten species in the
northern Iberian Peninsula. This study compared the effectiveness, logistics and monetary efficacy of both techniques for this
type of research.
Materials and methods
Study area
The study was performed in the Montes do Invernadeiro Natural
Park, a mountainous area located in the north-west Iberian
Peninsula (Universal Transverse Mercator (UTM) coordinates:
29T 064633–643 and 467462–472) (Fig. 1). The study area,
which covers 6700 ha, varies in altitude from 880 to 1707 m.
The climate is continental, with hot summers and cold winters
N
Iberian Peninsula
Montes do
Invernadeiro
National Park
4 km
100 km
Stone marten
European pine marten
Village
Fig. 1. Broad-scale distribution of the stone
marten and the European pine marten over the
Iberian Peninsula (modified from Mitchell-Jones
et al. 1999; and Palomo and Gisbert 2002) and
location of the study area.
436
Wildlife Research
S. Rosellini et al.
(Barja 2001). The flora is diverse, including Mediterranean
plant communities and Atlantic relic forests (Castroviejo 1977;
Barja 2001). The plant community is scrubland, dominated by
heather (Erica australis), prickled broom (Pterospartum tridentatum) and sandling (Halimium lasianthum). The original
forests remain in valleys and along watercourses and are formed
principally by associations of oak (Quercus robur), birch
(Betula celtiberica), holly (Ilex aquifolium) and rowanberry
(Sorbus aucuparia). Large tracts of extended forestlands are
formed by Scot pine (Pinus sylvestris), which has repopulated
deforested areas (Barja 2001). The park is occupied by carnivores such as the Iberian wolf (Canis lupus), badger (Meles
meles), European polecat (Mustela putorius), stoat (Mustela
erminea), European common weasel (Mustela nivalis), genet
(Genetta genetta), wildcat (Felis silvestris), red fox (Vulpes
vulpes) and otter (Lutra lutra). The study area appears to have
suitable resources to support both the European pine marten and
the stone marten. However, the presence of both species in the
study area has not previously been examined. There are no
human populations in the study area and the nearest village is
4 km from the park.
Genetic analysis of faecal samples
The specific identification of faecal samples was accomplished
by the polymerase chain reaction–restriction fragment length
polymorphism (PCR-RFLP). The faecal DNA extraction procedure was based on the protocol described in Gómez-Moliner
et al. (2004). Extraction blanks were included to monitor for
contamination and were processed as separate samples in the
subsequent amplifications. A small fragment (276 bp) of mitochondrial DNA (mtDNA), D-loop region, was amplified with
specific primers developed for Martes and Mustela by PCR.
The primers were designed to amplify small fragments to maximise the probability of amplification of degraded DNA.
Following DNA amplification, PCR products were digested
with the restriction enzymes HaeIII and RsaI. The combined
use of both enzymes produced a species-specific banding
pattern allowing the scats of M. martes and M. foina to be discriminated. This method was recently developed and provides a
reliable and effective molecular technique for unequivocal
genetic identification of sympatric marten species through the
use of species-specific differential haplotypes (Ruiz-González
et al. 2008).
Collection of faecal samples
The European pine marten, the stone marten and other carnivores use forest roads and frequently defaecate on them as a way
of visual-scent marking (Robinson and Delibes 1988; Barja
et al. 2004; Barja 2005b; Barja et al. 2005). Initially, two scatbased surveys were conducted to estimate the distribution range
of the genus Martes in the study area. The study area comprised
67 cells (UTM) of 1 km2 each. In the initial survey, two 400-m
transects in each cell were examined and 26 yielded Martes
scats. Later, in order to collect fresh Martes faecal samples,
transects 300 m long were surveyed on foot along forest roads
monthly from June 2004 to August 2006 in these 26 cells alone.
These surveys were conducted only in the 26 cells where signs
of the presence of Martes species (scats, footprints, sightings)
had previously been detected. We found fresh faeces in 17 of the
26 cells and old faeces in the remaining nine cells. Fresh faeces
were characterised by their strong smell, presence of a mucus
layer and lack of any sign of dehydration. Morphological characteristics, such as size and shape, were used to distinguish
faeces of the genus Martes from those of medium-sized carnivores such as red fox and wildcat. The territory of the European
pine marten is 1.3 km2 for males and 1.0 km2 for females
(O. Berdión, Department of Zoology and Animal Cell Biology,
Universidad País Vasco, Spain, pers. comm.). The mean territory size of the stone marten is similar to that of the European
pine marten, being about 0.96 km2 (López-Martín 2003). Thus,
in order to increase the likelihood of obtaining faecal samples
from separate individuals, the transects were placed 700 m
apart. The transects were uniformly distributed throughout the
study area and 218.7 km were surveyed. When fresh faeces of
the genus Martes were detected, the date, time and UTM cell
were recorded. A sample was collected from the fresh scat using
a gloved hand. A new glove was used for each sample to avoid
cross-contamination. All collected faecal samples were stored in
hermetic and sterilised tubes, preserved with ethanol (96%) and
maintained at –20°C until assayed (Piggott and Taylor 2003;
Gómez-Moliner et al. 2004).
Camera trapping
Every three months, from June 2004 to October 2006, 10–20
automatic line-triggered cameras (models: Moultrie, Canon
Prima AF 9s, Stealth Cam) were set in place and baited with
chicken placed 70 cm from the ground and 2 m from the
cameras. The cameras were equipped with different activation
systems. Sixteen used passive and active infrared activation and
four employed a plate system triggered by the pressure of the
animal’s weight. The cameras were placed in those cells where
Martes faeces had been detected during the surveys and in
optimal areas for both marten species even if signs of the
animals were not detected. The optimal areas were selected on
the basis of the known habitat preferences of both marten
species (Delibes 1983; Sacchi and Meriggi 1995; Virgós and
Garcia 2002; Barja 2005a). The cameras remained in the field
for 3–12 days (916 camera-trap-nights). The cameras were
placed in a total of 22 UTM cells (32.8% of cells had cameras
in them). Each camera was revisited every third day.
The European pine marten and the stone marten are similar
morphologically. Both species have a spot on the neck called a
‘bib’, which is yellow in the European pine marten and white in
the stone marten (Cabrera 1914). Therefore, bib colour was used
to distinguish between the two species. Spot patterns, size and
symmetry of the bibs differ between individuals of the same
species (Cabrera 1914). In the present study, the differences in
the bib morphology and distance between cameras allowed us to
identify individuals. To increase the probability of photographing the bibs, the cameras were baited with chicken placed 70 cm
from the ground and 2 m from the cameras. The morphology of
the bibs was analysed in the photographs obtained and when the
spot pattern, size and symmetry coincided, the photographs
were assigned to a single individual. Also, when the distance
between cameras was greater than 1.3 km, we considered that
the photographs obtained were of different individuals of the
same species. This assumption was supported by the morphological differences of the bibs in the photographs.
Small-scale distribution of stone and pine marten
Wildlife Research
Results
The faecal DNA analysis allowed us to detect the European pine
marten in 17 UTM cells and the camera-trapping in 9 cells (in
6 cells the species was detected by both methods). Therefore, the
presence of the European pine marten was confirmed in 20 cells
(29.9% of cells) (Fig. 2). However, neither method detected the
stone marten. Of 90 analysed faecal samples from throughout
the study area, after applying the PCR-RFLP method, we successfully identified 80 as being from the European pine marten
and none from the stone marten. Consequently, 88.8% of the
collected samples could be assigned to a species; in the remaining 11.2% the DNA extracted was not amplified by the primers
used. The mean number of fresh faecal samples analysed per
cell was 5.3; the mean number of scats detected was 1.97 per
month per kilometre surveyed. In 65.4% of the study cells (17 of
26), fresh scats were collected. The faecal DNA analysis confirmed the presence of the European pine marten in 100% of the
cells in which fresh faeces were collected.
The camera-traps captured 126 photographs of four different
terrestrial carnivores (Fig. 3). The European pine marten was
the most photographed species (57.9%), followed by red fox
(33.3%), wildcat (5.6%) and Iberian wolf (3.2%). This sampling
method confirmed the presence of the European pine marten in
437
40.9% of camera-equipped cells, red fox in 36.4% of the cells,
wildcat in 27.3% and Iberian wolf in only 18.2%. The intercamera distance (>1.3 km) allowed for the identification of nine
different individuals of European pine marten. Also, seven different individuals were identified according to the morphology
of the bibs. The cells where the presence of the European pine
marten was confirmed (using faecal DNA analysis and cameratrapping) included a high percentage of tree cover (coniferous
and deciduous forests), permanent watercourses and a high percentage of rocky substrate and outcroppings. In contrast, the
cells in which pine martens were not detected were dominated
by scrubland.
Discussion
In this study, both research methods indicated that, while the
European pine marten was present in the study area, the stone
marten was not. The PCR-RFLP technique allowed us to
achieve a high rate of species identification (88.8%). These
genetic analyses open the possibility for studying population
genetics of the species, recognising that nuclear marker genotyping from scats is more difficult than mtDNA amplification.
The remaining 11.2% of samples did not amplify. The possibility that some scat samples had been incorrectly identified in the
UTM
4673
72
71
70
69
68
Fig. 2. Placement of the camera-traps, location
of the faeces of the genus Martes and distribution
of the European pine marten obtained from two
different methods (faecal DNA analysis and
camera-traps).
67
66
65
64
63
1 km
62
33
34
35
36
37
38
39
40
Limits of Montes do Invernadeiro National Park
Camera-trap
Faeces of genus Martes
Presence of European pine marten
41
42
43
644
UTM
438
Wildlife Research
S. Rosellini et al.
field as belonging to the marten would mean an even higher
amplification success. Misidentifying fox scat as marten scat is
a particularly common error (Davison et al. 2002). However, the
primers developed for Martes and Mustela by PCR did not
amplify in fox scats.
Camera-traps detected martens in only 40.9% of the cells
(9 of 22), whereas faecal DNA detected them in 100% of the
cells in which fresh faeces were collected (17 of 17). However,
both methods have advantages and disadvantages that are influenced by the size of the area inventoried, sampling effort,
detectability and cost and efficiency of the method. Although a
large number of faecal samples can be collected, the faecal DNA
is highly degraded (Taberlet et al. 1999) and the faeces collected
can be from other carnivore species for which the primers used
do not amplify, implying important economic costs. Also, the
collection of fresh faeces requires a high field effort. However,
this method is advantageous for both large- and small-scale
investigations. An advantage of camera-trapping is the possibility of obtaining a broad range of significant information regarding the carnivore’s community. The absence of specific
attractants does not allow the species of interest to be selectively
photographed. Also, the greater effort required for instrument
placement, physical monitoring of the cameras and revisits to
the study area are a disadvantage for large-scale research efforts.
In the current study, camera-traps detected the European pine
marten in cells in which the faecal DNA analysis did not detect
its presence. Also, the use of camera-traps allowed us to visually
detect the European pine marten (and identify individuals) and
other carnivores (red fox, wildcat and Iberian wolf), which are
difficult to study due to their elusive and nocturnal habits.
Camera-trapping allows the collection of valuable information
80
Photographs
Minimum number of individuals
70
60
Frequency
50
40
30
20
10
0
Pine marten
Red fox
Wildcat
Iberian wolf
Fig. 3. Photographs of carnivores obtained using camera-traps, and an
estimate of the minimum number of individuals present.
about multiple species, including herbivores and carnivores,
within any given community (Karanth 1995; Peterson and
Thomas 1998; Azlan and Sharma 2003; Swann et al. 2004;
Trolle and Kéry 2005). Nevertheless, the simultaneous use of
both methods in a small-scale marten survey can enhance the
accuracy of data and result in a more ecologically relevant study.
Small mammals are relatively abundant in the study area
(64.4 animals ha–1) (I. Barja, S. Rosellini and A. Piñeiro,
unpubl. data) and are the primary prey of the European pine
marten all year round (Rosellini et al. 2007). In some areas of
the northern Iberian Peninsula small mammals have also been
found to constitute the largest proportion of the stone marten
diet (Delibes 1978). Nevertheless, despite the apparent presence
of adequate food resources for the survival of both species
within the study area, only the European pine marten is present.
The ecological displacement that occurs when both species
overlap in the same area is likely related to their relative densities and to the availability of trophic resources, even though prey
items appear to be abundant. Competition for these resources
can play a fundamental role in the interactions between individuals and species (Schoener 1983; Begon et al. 1986; Keddy
1989). Interspecific and intraspecific competition can cause a
decrease in individual fitness (Begon et al. 1986). Since reduced
fitness has significant negative consequences from an evolutionary standpoint, some kind of differentiation in the utilisation
of trophic resources and habitat must occur in order for both
species to coexist (Schoener 1974). Such differentiation can
allow both species and individuals to avoid depressed fitness,
that is, a decrease in fecundity and/or survival (Schoener 1974;
Barrientos and Virgós 2006). In sympatric carnivores, there is
ample evidence that ecological competition determines size and
stability of populations (Palomares et al. 1996; Barrientos and
Virgós 2006). Similar carnivores that coexist in the same area,
therefore, normally differ in their trophic strategies (Schoener
1974; Clevenger 1994; Barrientos and Virgós 2006). In
mustelids, the sympatry among species has been explained by
differences in body size, which reduces interspecific competition (Erlinge 1986). While the high degree of trophic overlap
between the two martens studied here has been reported by
some authors (Marchesi et al. 1989; Clevenger 1994), the
mechanisms of coexistence have been proposed only by
Clevenger (1994).
The results of this work suggest a functional ecological
mechanism for competition between the two studied species. On
the Iberian Peninsula, the distribution ranges of the European
pine marten and the stone marten overlap within the study area,
providing a situation for ecological competition, and species
segregation, on both a large and small scale. The exclusive
detection of the European pine marten within the study area is
likely associated with its preference for uniform and well conserved forests (Barja 2005a), resulting in the displacement of
the stone marten to disturbed, deforested and urban areas
(Delibes 1983; Libois and Waechter 1991).
On a smaller scale, habitat fragmentation can allow for sympatric distribution of very similar species (Pilot et al. 2007). The
observed spatial segregation of the two mustelids is probably
due to both physical size, the European pine marten being the
larger of the two (Delibes 1983; Barea and Ballesteros 1999;
Barja 2005a), and the more general and flexible habits of the
Small-scale distribution of stone and pine marten
stone marten (Waechter 1975; Delibes 1983; Sacchi and
Meriggi 1995; Mitchell-Jones et al. 1999; Virgós and Garcia
2002; Lanszki 2003). Interspecific competition is, therefore,
avoided and the stone marten, due to its more generalist habits,
shows a higher level of adaptability to human-altered environments, which may have implications for long-term viability of
the species (Clevenger 1994).
The genetic analysis of four faecal samples collected in the
periphery of the closest village to the study area, and the location of several dead mustelids, indicated that the European pine
marten was also present there. This information and the other
results obtained during this study seem to indicate that the
scarce human-caused environmental alterations might influence
in the absence of stone marten. In the study area and surrounding areas the density of humans is very low and the forest habitats are well conserved, with a progressive abandonment of
agricultural activities, including crops and animal husbandry.
We suggest, therefore, that further studies are needed to analyse
the sympatry hypothesis on a small scale, considering different
degrees of urbanisation and environmental disturbance. The
possible competition and habitat segregation between the two
marten species in the forests of the northern Iberian Peninsula is
a basis for future studies regarding the animal community of
forest habitats. These investigations provide fundamental information relative to habitat management directed principally at
the conservation of the European pine marten since it apparently depends a great deal, if not exclusively, on unimpacted
forest resources.
Acknowledgements
We thank the Nature Conservation Service of Ourense (Regional
Government of Galicia) for providing the permits required to conduct this
study in the Montes do Invernadeiro Natural Park. Thanks are due to Dr
Benjamín Gómez-Moliner for supervising the molecular work. We are
grateful to gamekeepers Tomás Pérez, Ricardo Prieto and Paco Barja for
their help with fieldwork, and to Toni Gago for assistance during the collection of field data. A. Ruiz-González was supported by a Ph.D. grant from
the Department of Education, Universities and Research (Basque
Government) and S. Rosellini was supported by a postgraduate scholarship
from the Universidad Autónoma de Madrid. This study has been supported
by the Project USEK, n° 7-2004-05, Universidad SEK de Segovia.
References
Azlan, J. M., and Sharma, D. S. K. (2003). Camera trapping the Indochinese
tiger, Panthera tigris corbetti, in a secondary forest in peninsula
Malaysia. The Raffles Bulletin of Zoology 51, 421–427.
Barea, J. M., and Ballesteros, E. (Eds.) (1999). ‘Carnívoros Ibéricos, Serie
de Estudios y Proyectos de Biología N°2.’ (Colegio Oficial de Biólogos
de Andalucía: Granada.)
Barja, I. (2001). La señalización en el lobo ibérico (Canis lupus signatus).
Comparación con dos especies de Hienas (Crocuta crocuta y Hyaena
hyaena). Ph.D. Thesis, Universidad Autónoma de Madrid, Madrid.
Barja, I. (2005a). Winter distribution of European pine marten Martes
martes scats in a protected area of Galicia, Spain. Mammalia 69,
435–438. doi:10.1515/mamm.2005.037
Barja, I. (2005b). Patrones de marcaje con heces por la marta europea
(Martes martes) en el noroeste de España: importancia para su estudio.
Galemys 17, 123–134.
Barja, I., de Miguel, F. J., and Bárcena, F. (2004). The importance of crossroads in faecal marking behaviour of the wolves (Canis lupus).
Naturwissenschaften 91, 489–492. doi:10.1007/s00114-004-0557-1
Wildlife Research
439
Barja, I., de Miguel, F. J., and Bárcena, F. (2005). Faecal marking behaviour
of Iberian wolf in different zones of their territory. Folia Zoologica 54,
21–29.
Barrientos, R., and Virgós, E. (2006). Reduction of potential food interference in two sympatric carnivores by sequential use of shared resources.
Acta Oecologica 30, 107–116. doi:10.1016/j.actao.2006.02.006
Begon, M., Harper, J. L., and Townsend, C. R. (Eds) (1986). ‘Ecology.
Individuals, Populations and Communities.’ (Blackwell Scientific
Publications: Oxford.)
Birks, J. D. S., Messenger, J. E., Braithwaite, T. C., Davison, A.,
Brookes, R. C., and Strachan, C. (2004). Are scat surveys a reliable
method for assessing distribution and population status of pine martens?
In ‘Martens and Fishers (Martes) in Human-altered Environments: an
International Perspective’. (Eds D. J. Harrison, A. K. Fuller and
G. Proulx.) pp. 235–252. (Springer-Verlag: New York.)
Blanco, J. C. (Ed.) (1998). ‘Mamíferos de España.’ (Geoplaneta: Madrid.)
Blanco, J. C., Reig, S., and Cuesta, L. (1992). Distribution, status and conservation problems of the wolf Canis lupus in Spain. Biological
Conservation 60, 73–80. doi:10.1016/0006-3207(92)91157-N
Cabrera, A. (Ed.) (1914). ‘Fauna Ibérica. Mamíferos.’ (Museo Nacional de
Ciencias Naturales: Madrid.)
Castroviejo, S. (Ed.) (1977). ‘Estudio sobre la Vegetación de la Sierra del
Invernadeiro (Ourense).’ (ICONA: Madrid.)
Clevenger, A. P. (1994). Feeding ecology of Eurasian pine martens and stone
martens in Europe. In ‘Martens, Sables and Fishers’. (Eds S. W. Duskirk,
A. S. Arestad, M. G. Raphael and R. A. Powell.) pp. 326–340. (Cornell
University Press: Ithaca.)
Dalén, L., Götherström, A., and Angerbjörn, A. (2004). Identifying species
from pieces of faeces. Conservation Genetics 5, 109–111. doi:10.1023/
B:COGE.0000014060.54070.45
Davison, A., Birks, J. D. S., Brookes, R. C., Braithwaite, T. C., and
Messenger, J. E. (2002). On the origin of faeces: morphological versus
molecular methods for surveying rare carnivores from their scats.
Journal of Zoology 257, 141–143. doi:10.1017/S0952836902000730
Delibes, M. (1978). Feeding habits of the stone marten, Martes foina
(Erxleben, 1777), in northern Burgos, Spain. Zeitschrift für
Saugetierkunde 43, 282–288.
Delibes, M. (1983). Interspecific competition and the habitat of the stone
marten Martes foina (Erxleben 1777) in Europe. Acta Zoologica
Fennica 174, 229–231.
Erlinge, S. (1986). Specialists and generalists among the mustelids. Lutra
29, 5–11.
Ernest, H. B., Penedo, M. C. T., May, B. P., Syvanen, M., and Boyce, W. M.
(2000). Molecular tracking of mountain lions in the Yosemite Valley
region in California: genetic analysis using microsatellites and faecal
DNA. Molecular Ecology 9, 433–441. doi:10.1046/j.1365-294x.2000.
00890.x
Farrell, L. E., Romant, J., and Sunquist, M. E. (2000). Dietary separation of
sympatric carnivores identified by molecular analysis of scats.
Molecular Ecology 9, 1583–1590. doi:10.1046/j.1365-294x.2000.
01037.x
Gómez-Moliner, B. J., Cabria, M. T., Rubines, J., Garin, I., Madeira, M. J.,
Elejalde, A., Aihartza, J., Fournier, P., and Palazón, S. (2004).
PCR-RFLP identification of mustelid species: European mink (Mustela
lutreola), American mink (M. vison) and polecat (Mustela putorius) by
analysis of excremental DNA. Journal of Zoology 262, 311–316.
doi:10.1017/S0952836903004667
Karanth, K. U. (1995). Estimating tiger Panthera tigris populations from
camera-trap data using capture–recapture models. Biological
Conservation 71, 333–338. doi:10.1016/0006-3207(94)00057-W
Keddy, P. A. (Ed.) (1989). ‘Competition.’ (Chapman and Hall: London.)
Kohn, M. H., and Wayne, R. K. (1997). Facts from feces revisited. Trends in
Ecology & Evolution 12, 223–227. doi:10.1016/S0169-5347(97)01050-1
Kurose, N., Masuda, R., and Tatara, M. (2005). Fecal DNA analysis for identifying species and sex of sympatric carnivores: a noninvasive method
440
Wildlife Research
S. Rosellini et al.
for conservation on the Tsushima Islands, Japan. The Journal of
Heredity 96, 688–697. doi:10.1093/jhered/esi124
Lanszki, J. (2003). Feeding habits of stone martens in a Hungarian village
and its surroundings. Folia Zoologica 52, 367–377.
Libois, R., and Waechter, A. (1991). La fouine (Martes foina). In
‘Encyclopédie des Carnivores de France, Vol. 10’. (Eds M. Artois, and
P. Delattre.) pp. 1–53. (Societé Française pour l’Étude et Protection des
Mammifères, Bohallard: Puceul.)
López-Martín, J. M. (2003). Aspectos de la ecología de la marta (Martes
martes L. 1758) y la garduña (M. foina Erx.1777) en los ambientes
mediterráneos: interacciones con la gineta (Genetta genetta L. 1758).
Ph.D. Thesis, Universidad de Barcelona, Barcelona.
Lynch, Á. B., Brown, M. J. F., and Rochford, J. M. (2006). Fur snagging as
a method of evaluating the presence and abundance of a small carnivore,
the pine marten (Martes martes). Journal of Zoology 270, 330–339.
doi:10.1111/j.1469-7998.2006.00143.x
Marchesi, P., Lachat, N., Leinhard, R., Debieve, P., and Mermod, C. (1989).
Comparaison des régimes alimentaires de la fouine (Martes foina Erxl.)
et de la martre (Martes martes L.) dans une région du Jura suisse. Revue
Suisse de Zoologie 96, 281–296.
Messenger, J. E., and Birks, J. D. S. (2000). Monitoring the very rare: pine
marten populations in England and Wales. In ‘Mustelids in a Modern
World. Management and Conservation Aspects of Small Carnivore:
Human Interactions’. (Ed. H. I. Griffiths.) pp. 153–162. (Backhuys
Publishers: Leiden.)
Mitchell-Jones, A. J., Amori, G., Bogdanowicz, W., Krystufek, B.,
Reijnders, P. J. H., Spitzemberg, F., Stubbe, M., Thissen, J. B. M.,
Vohralik, V., and Zima, J. (1999). ‘The Atlas of European Mammals.’
(Academic Press: London.)
Murakami, T. (2002). Species identification of mustelids by comparing
partial sequences on mitochondrial DNA from fecal samples. The
Journal of Veterinary Medical Science 64, 321–323. doi:10.1292/
jvms.64.321
Palomares, F., Ferreras, P., Fedriani, J. M., and Delibes, M. (1996). Spatial
relationships between Iberian lynx and other carnivores in an area of
south-west Spain. Journal of Applied Ecology 33, 5–13. doi:10.2307/
2405010
Palomares, F., Godoy, J. A., Piriz, A., O’Brien, J., and Johnson, W. E. (2002).
Faecal genetic analysis to determine the presence and distribution of
elusive carnivores: design and feasibility for the Iberian lynx. Molecular
Ecology 11, 2171–2182. doi:10.1046/j.1365-294X.2002.01608.x
Palomo, L. J., and Gisbert, J. (Eds.) (2002). ‘Atlas de los Mamíferos
Terrestres de España’. (Dirección General de Conservación de la
Naturaleza-SECEM-SECEMU: Madrid.)
Paxinos, E., Mcintosh, C., Ralls, K., and Fleischer, R. (1997). A noninvasive method for distinguishing among canid species: amplification
and enzyme restriction of DNA. Molecular Ecology 6, 483–486.
doi:10.1046/j.1365-294X.1997.00206.x
Peterson, L. M., and Thomas, J. A. (1998). Performance of Trailmaster
infrared sensors in monitoring captive coyotes. Wildlife Society Bulletin
26, 592–596.
Piggott, M. P., and Taylor, A. C. (2003). Remote collection of animal DNA
and its applications in conservation management and understanding the
population biology of rare and cryptic species. Wildlife Research 30,
1–13. doi:10.1071/WR02077
Pilot, M., Gralak, B., Goszczyński, J., and Posluszny, M. (2007). A method
of genetic identification of pine marten (Martes martes) and stone
marten (Martes foina) and its application to faecal samples. Journal of
Zoology 271(2), 140–147. doi:10.1111/j.1469-7998.2006. 00179.x
Proulx, G., Aubry, K. B., Birks, J., Buskirk, S. W., Fortin, C., et al. (2004).
World distribution and status of the genus Martes in 2000. In ‘Martens
and Fishers (Martes) in Human Altered Environments: an International
Perspective’. (Eds D. J. Harrison, A. K. Fuller, and G. Proulx.)
pp. 21–76. (Springer-Verlag: New York.)
Robinson, I. H., and Delibes, M. (1988). The distribution of faeces by the
Spanish lynx (Felis pardina). Journal of Zoology 216, 577–582.
Rodriguez, A., and Delibes, M. (1992). Current range and status of the
Iberian lynx (Felis pardina Temminck, 1824) in Spain. Biological
Conservation 61, 189–196. doi:10.1016/0006-3207(92)91115-9
Rosellini, S., Barja, I., and Piñeiro, A. (2007). Distribución y hábitos
alimenticios de la marta (Martes martes) en el Parque Natural Os
Montes do Invernadeiro (Galicia, NO de España). Galemys 19, 99–114.
Ruiz-González, A., Rubines, J., Berdión, O., and Gómez-Moliner, B. J.
(2008). A non-invasive genetic method to identify the sympatric
mustelids pine marten (Martes martes) and stone marten (Martes foina):
preliminary distribution survey on the N Iberian Peninsula. European
Journal of Wildlife Research 54, 253–261. doi:10.1007/s10344-0070138-7
Sacchi, O., and Meriggi, A. (1995). Habitat requirements of the stone
marten (Martes foina) on the Tyrrhenian slopes of the northern
Apennines. Hystrix 7, 99–104.
Schoener, T. W. (1974). Resource partitioning in ecological communities.
Science 185, 27–39. doi:10.1126/science.185.4145.27
Schoener, T. W. (1983). Field experiments on interspecific competition.
American Naturalist 122, 240–285. doi:10.1086/284133
Sugimoto, T., Nagata, J., Aramilev, V. V., Belozor, A., Higashi, S., and
McCullough, D. R. (2006). Species and sex identification from faecal
samples of sympatric carnivores, Amur leopard and Siberian tiger, in the
Russian Far East. Conservation Genetics 7, 799–802. doi:10.1007/
s10592-005-9071-z
Swann, D. E., Hass, C. C., Dalton, D. C., and Wolf, S. A. (2004). Infraredtriggered cameras for detecting wildlife: an evaluation and review.
Wildlife Society Bulletin 32, 357–365. doi:10.2193/0091-7648(2004)32
[357:ICFDWA]2.0.CO;2
Taberlet, P., Waits, L. P., and Luikart, G. (1999). Noninvasive genetic sampling: look before you leap. Trends in Ecology & Evolution 14,
323–327. doi:10.1016/S0169-5347(99)01637-7
Trolle, M., and Kéry, M. (2005). Camera-trap study of ocelot and other
secretive mammals in the northern Pantanal. Mammalia 69, 409–416.
doi:10.1515/mamm.2005.032
Virgós, E. (2001). Distribución y estatus del turón (Mustela putorius) en
España: un análisis basado en encuestas. Galemys 13, 39–61.
Virgós, E., and Garcia, F. J. (2002). Patch occupancy by stone martens
Martes foina in fragmented landscapes of central Spain: the role of fragment size, isolation and habitat structure. Acta Oecologica 23, 231–237.
doi:10.1016/S1146-609X(02)01142-6
Waechter, A. (1975). Ecologie de la fouine en Alsace. Terre et Vie 29,
399–457.
Zielinski, W. J., and Kucera, T. E. (Eds.) (1995). ‘American Marten, Fisher,
Lynx, and Wolverine: Survey Methods for their Detection.’ (US
Department of Agriculture Forest Service, Pacific Southwest Research
Station: Berkeley.)
Manuscript received 7 March 2007, accepted 13 May 2008
http://www.publish.csiro.au/journals/wr
Paper IIIb
Appendix B
Appendix B
PAPER IIIb
Balestrieri A, Remonti L, Ruiz-González A, Gómez-Moliner
BJ, Vergara M, Prigioni C (2010)
Range expansion of the pine marten (Martes martes) in an
agricultural landscape matrix (NW Italy).
Mammalian Biology, 75: 412–419
- 219 -
Appendix b
Paper IIIb
www.elsevier.de/mambio
ORIGINAL INVESTIGATION
Range expansion of the pine marten (Martes martes) in an agricultural
landscape matrix (NW Italy)
Alessandro Balestrieria, Luigi Remontia, Aritz Ruiz-Gonzálezb,
Benjamı́n J. Gómez-Molinerb, Maria Vergarab, Claudio Prigionia,
a
Department of Animal Biology, University of Pavia, Piazza Botta 9, 27100 Pavia, Italy
Department of Zoology and Animal Cell Biology, Zoology Laboratory, Facultad de Farmacia,
Universidad del Paı´s Vasco (UPV-EHU), C/ Paseo de la Universidad 7, 01006 Vitoria-Gasteiz, Spain
b
Received 27 March 2009; accepted 24 May 2009
Abstract
Habitat segregation is considered to favour the coexistence of sympatric pine martens Martes martes and stone
martens M. foina, the latter being displaced to agricultural and urbanised areas. Subsequent to the report of pine
martens in cultivated areas of the western River Po plain (NW Italy), we reviewed all available information on the
presence of this species in plain areas of northern Italy and, for two study areas, applied a non-invasive PCR-RFLP
method for the identification of Martes species from faecal mtDNA.
A total of 24 pine marten records were collected, grouped in the western part of the River Po plain. The number of
records showed an exponential increase from 1988 to 2007, the percentage of woods in a 10 km2 wide circular plot
surrounding the location of records being inversely correlated to its distance from the 300 m a.s.l. contour line. In the
two study areas, 36 out of 119 ‘‘marten-like’’ faeces were assigned to the pine marten, whilst none belonged to the stone
marten. In the best monitored area, the pine marten was present almost constantly.
Our results suggest that the pine marten is expanding its range to include cultivated areas which were previously
considered a prerogative of the more synanthropic stone marten.
r 2009 Deutsche Gesellschaft für Säugetierkunde. Published by Elsevier GmbH. All rights reserved.
Keywords: Non-invasive genetic sampling; Road-kills; Distribution; Interspecific competition
Introduction
The range of mammals contracts and expands over
time, according to changes in climate (Walther et al.
2002; Parmesan and Yohe 2003) and habitat (Frey 1992;
Benedict et al. 2000; Albert et al. 2004; Geluso 2004;
Geluso et al. 2005) often caused by human influence
(Kerr and Currie 1995; Brashares et al. 2001). In
Corresponding author. Tel.: +39 0382 986304;
fax: +39 0382 986290.
E-mail address: [email protected] (C. Prigioni).
particular, despite their high mobility, the expansion
of mammalian carnivores has been shown to be
influenced by both food availability (e.g., on mustelids:
King 1989; Jedrzejewski et al. 1995) and habitat
fragmentation (Carroll et al. 2001; Crooks 2002).
In the last twenty years, as a consequence of a combination of factors – including legal protection, pollution
control, habitat restoration and increasing prey availability -, the demographic trend of several mammalian
predators has reversed, resulting in either the recolonisation of areas where they had disappeared during
the 20th century (e.g., for Italy, the otter Lutra lutra,
1616-5047/$ - see front matter r 2009 Deutsche Gesellschaft für Säugetierkunde. Published by Elsevier GmbH. All rights reserved.
doi:10.1016/j.mambio.2009.05.003
Mamm. biol. 75 (2010) 412–419
A. Balestrieri et al. / Mamm. biol. 75 (2010) 412–419
Prigioni et al. 2007, and the wolf Canis lupus, Boitani
2003) or the occupancy of novel environments (e.g.
urban red foxes Vulpes vulpes in Switzerland; Gloor
2002).
The effects of range expansion on interspecific
relationships can potentially vary from coexistence to
the geographic replacement of native species by invaders, the latter having been reported in consequence
of the introduction of alien species (IUCN 2000).
While the top-down effects that carnivores have on
prey populations have been the subject of much research
(e.g., Skogland 1991; Palomares et al. 1995; Smedshaug
et al. 1999), those that predator species have on each
other have been far less studied (Linnell and Strand
2000).
Competitive interactions among predators include
intraguild predation (Palomares and Caro 1999) and
may induce weaker species to seek habitats avoided by
their competitors (‘‘refuges’’) (Durant 1998; Prigioni
et al. 2008), sharply limiting their population density
(Lindstrom et al. 1995). Competition is more likely to
occur between interacting related species of similar size
(see Powell and Zielinski 1983, about Martes and
Mustela species).
Some endangered carnivore species are effectively
limited by predator species that have recently expanded
their range owing to protection policies (Creel and Creel
1996; Gorman et al. 1998; Kelly et al. 1998).
The pine marten Martes martes and the closely related
stone marten Martes foina are quite similar in terms of
body size, morphology and feeding habits (Goszczynski
1976; Marchesi et al. 1989). Their distribution range
overlaps across a large part of continental Europe
(Proulx et al. 2004), the two species showing different
habitat preferences. Habitat segregation has been
regarded as a mechanism favouring their coexistence
(Waechter 1975; Delibes 1983).
The pine marten occurs primarily in well structured
deciduous and coniferous forests (Delibes 1983; Buskirk
1992), but in southern Europe it has shown great
ecological plasticity, having been reported in coppices,
the Mediterranean maquis and cultivated land with
woodland fragments (De Marinis and Massetti 1993;
Pittiglio 1996). Nonetheless, pine martens strongly avoid
open habitats (Storch et al. 1990; Brainerd and Rolstad
2002) and are believed to need a minimum woodland
area to survive, equal to about 2 km2 for the temperate
zone (Zalewski and Jedrzejewski 2006). The stone
marten can live in a variety of habitats, from boreal
woods (Novikov 1962) to agricultural areas (Sacchi and
Meriggi 1995; Lanszki 2003) and also occurs in villages
and towns (Broekhuizen 1983; Prigioni and Sommariva
1997). Wherever the two species occur sympatrically, the
stone marten is often associated with rural and
suburban areas (Jensen and Jensen 1970; Waechter
1975), whilst the pine marten occupies forested areas.
413
This compression of the spatial niche of the stone
marten has been explained as a consequence of
interspecific competition, favouring the slightly bigger
pine marten (Delibes 1983). Nonetheless, the two species
have also been reported to be syntopic (Kruger 1990;
Genovesi 1993; Pilot et al. 2007; Ruiz-González et al.
2008).
Both martens occur in the Italian peninsula (Genovesi
and De Marinis 2003). In northern Italy, the stone
marten occurs from Alpine meadows to cultivated
plains, whilst the pine marten is associated with
deciduous and coniferous forests between 1000 and
2000 m a.s.l. (Martinoli 2001; Bon et al. 1995). However,
road-killed pine martens have been recently reported
also for the western plain of the River Po (Sindaco 2006;
Savoldelli and Sindaco 2008), an intensively cultivated
area, deeply altered by production activities.
Road kills are unpredictable events which can help to
draw a large-scale picture of species distribution, but are
ineffective when trying to examine their actual range at a
local scale. In contrast, the systematic survey of field
signs can yield reliable results on the distribution,
abundance and habitat requirements of carnivores
(Sadlier et al. 2004).
The monitoring of marten populations is hindered by
our inability to distinguish the faeces of the two species,
making indirect survey methods unreliable. Moreover
marten faeces can be confused with those of other
carnivores, such as, for northern Italy, the red fox and
the polecat Mustela putorius (Davison et al. 2002).
Nonetheless, recent progress in molecular techniques
has supplied several non-invasive genetic methods for
the identification of martens (review in Ruiz-González
et al. 2008), which can be applied to field monitoring.
Thus, with the aim of assessing the spread of the pine
marten in plain areas of northern Italy, we used two
different approaches: i) we reviewed available road kill
records of the pine marten in plain areas of northern
Italy (Piedmont, Lombardy, Veneto and Friuli-Venezia
Giulia regions); ii) to confirm the presence of the species
in two sampling areas of the middle western plain of the
River Po we applied a non-invasive genetic method
based on the analysis of DNA extracted from faecal
samples.
Material and methods
Available road kill records for northern Italy were
considered below the 300 m a.s.l. contour line, which
broadly marks the upper limit of the plain (T.C.I. 1957).
For Piedmont, we consulted the Data Banks of
Piedmont Region, whilst for Lombardy and Veneto
regions we referred to the data collected for the regional
Atlases of Mammals from, respectively, 1980 to 2001
414
A. Balestrieri et al. / Mamm. biol. 75 (2010) 412–419
(Prigioni et al. 2001) and 1970 to 1995 (Bon et al. 1995).
The thirty-year monitoring of road kills provided useful
information about marten distribution in the eastern part
of the study area (L. Lapini, unpublished data). Papers on
Italian martens and the proceedings of several national
congresses, published between 1985 and 2008, were also
consulted. Moreover unpublished, original records, reported by both field naturalists and game-keepers, were
collected. All road killed martens were examined by
expert zoologists, minimizing the risk of misidentification.
In three cases, genetic analyses were performed on tissue
samples to confirm the identifications.
All records were digitalised in a Geographic Information System (ArcView 3.1, ESRI, California, USA) and
grouped in four five-year long periods (1988–1992,
1993–1997, 1998–2002 and 2003–2007). A regression
analysis was used to find the best equation describing
the trend of pine marten records between 1988 and 2007.
To highlight the progressive penetration of the pine
marten into the River Po plain, the smallest distance
between each record and the 300 m a.s.l. contour line
was measured. For each five-year period, the mean and
maximum distances from this contour line were then
calculated.
Considering the minimum woodland area of 2 km2
needed by pine martens according to Zalewski and
Jedrzejewski (2006), we assessed the relative percentage
of woods in the areas of marten presence by overlaying
10 km2 wide circular plots, centred on each record
location, to a land-cover digitalised map. Plot area was
chosen with reference to the mean home range size of
pine martens in neighbouring Switzerland (6607220 ha;
Marchesi 1989) and Germany (765-1500 ha; Kruger
1990). The percentage of woods was related to the
smallest distance from the 300 m contour line by
Pearson’s correlation test (r).
Surveys for ‘‘fresh’’ marten scats were conducted in
2007/08, in two different plain areas – the Natural
Reserve ‘‘Garzaia di Valenza’’ (SE Piedmont,
02/2007–04/2008, no 1 in Fig. 1) and the Natural
Reserve ‘‘S. Massimo’’ (SW Lombardy, 08/2008–
10/2008, no 2 in Fig. 1).
The first Reserve covers 12.3 km2 on the left bank of
the River Po. The whole territory is flat, extensively
covered by cultivated fields and poplar (Populus sp.)
plantations. Woods consist of willows (Salix cinerea,
S. alba), oak (Quercus robur), poplars (Populus alba and
various hybrids) and alder (Alnus glutinosa), bordering
an abandoned river meander and three naturalized
artificial lakes. The black locust (Robinia pseudoacacia)
is widespread along roads and man-made embankments.
Mean altitude is 90 m a.s.l. Here, two pine martens were
killed by cars in February 2003 and March 2004,
respectively.
The second area covers 4.9 km2 on the right bank of
the River Ticino, which flows about 5 km away. Alder
(A. glutinosa) woods and mixed hygrophilous assemblages, where alders are joined by willows (S. alba) and
poplars (P. alba and P. nigra) are merged into an
agricultural matrix (mainly rice and maize). Mean
altitude is 70 m a.s.l. Here, pine marten presence was
reported for the first time in October 2005.
For both study areas the climate is sub-continental
temperate, with an average temperature of 12.4 1C and
an annual average precipitation of 1000–1040 mm.
Sampling was conducted along linear features, such as
wood/field margins, paths and country roads so as to
cover both open and forested habitats. In 2007, surveys
were carried out once a week, whilst in 2008 their
frequency was doubled so as to reduce the risk of faecal
DNA degradation.
Faecal samples were initially assigned to the genus
Martes if less than 10 mm large and to the red fox if
larger than 15 mm. Samples with intermediate width
were attributed on the basis of their overall appearance.
A portion (about 30%) of each ‘‘marten-like’’ faecal
sample was picked up with sticks and preserved in 96%
ethanol and by freezing until DNA extraction, the rest
was retained for dietary analysis. The specific identification of faecal samples was accomplished by a recently
developed polymerase chain reaction – restriction
fragment length polymorphism (PCR-RFLP) method,
providing for an effective genetic identification of
sympatric marten species (Ruiz-González et al. 2008).
The faecal mtDNA extraction procedure was based
on the protocol described by Gómez-Moliner et al.
(2004). Two specifically designed primers were used,
which generate 276 bp long amplicons. These primers
amplify the DNA from the two Martes species and from
four Mustela species. The simultaneous use of the
restriction enzymes RsaI and HaeIII differentiated
M. martes from M. foina and both of them from the
other carnivore species whose mtDNA is amplified by
the selected primers (see Ruiz-González et al. 2008 for
further details).
For the study area Garzaia di Valenza, which was
continuously monitored through 14 months, the percentage of positive surveys for the pine marten (%P) was
expressed as the per cent ratio between the number of
months with at least one positive survey and the total
number of surveyed months.
Results
A total of 24 road kill records was collected, most
(58.3%) referring to the last five years (Table 1). The
mean altitude of records was 183.7776.5 (S.D.) m a.s.l.
(min ¼ 70 m a.s.l.). Two out of six individuals for which
the sex had been recorded were females. The number of
pine marten records showed an exponential increase
A. Balestrieri et al. / Mamm. biol. 75 (2010) 412–419
415
Fig. 1. Road mortality and genetic reports of pine martens in the western River Po plain, relative to the 300 m a.s.l. contour line
(1: ‘‘Garzaia di Valenza’’ and 2: ‘‘S. Massimo’’ are the two areas of genetic survey).
Table 1. Number of road mortality reports of the pine
marten and their mean and maximum distance (km) from the
300 m a.s.l. contour line.
Years
N
Mean D
Max D
1988–1992
1993–1997
1998–2002
2003–2007
2
3
5
14
10.7
10.4
10.9
14.3
20.5
28.5
22.6
41.1
(Fig. 2), the distance of recorded localities from the
300 m a.s.l. contour line being higher in the last five-year
period (Table 1).
The pine marten is actually reported from plain,
mainly cultivated areas of two regions (Piedmont and
Lombardy) of north-western Italy (Fig. 1). Records are
grouped in the western part of the River Po plain, whilst
in its eastern part only one dubious record, dating back
to 1981, was found (province of Udine, about 300 m
a.s.l.).
The percentage of woods in the area surrounding pine
marten records was inversely correlated (r ¼ -0.57,
p ¼ 0.0033, N ¼ 24) to its distance from the 300 m
a.s.l. contour line (Fig. 3). For 48.5% of records the
percentage of woods inside the 10 km2 circular plots was
less than 20%; in the central plain localities, woods were
almost totally absent.
Almost all records could be easily assigned to a river
valley, 75% of records occurring on four main rivers
(Ticino, Sesia, Po and Agogna).
Fig. 2. Exponential increase of the number of pine marten
records below 300 m a.s.l., from 1988 to 2007.
A total of 36 out 119 (30.3%) ‘‘marten-like’’ faeces
was successfully identified by the PCR-RFLP method.
All faeces were assigned to the pine marten, respectively
19/99 (19.2%) for the Garzaia di Valenza (11.3%
in 2007 and 52.6% in 2008) and 17/20 (85%) for
S. Massimo, whilst none belonged to the stone marten.
The remaining samples did not amplify.
For these two areas, the percentage of woods inside
the 10 km2 circular plots was, respectively, 2.4% and
16.2%.
416
A. Balestrieri et al. / Mamm. biol. 75 (2010) 412–419
Fig. 3. Percentage of woods in the 10 km2 circular plots
centred on pine marten records, plotted against the distance of
the records from the 300 m a.s.l. contour line (r ¼ 0.57,
p ¼ 0.0033, N ¼ 24).
In the Garzaia di Valenza, for which marten
monitoring was almost constant for the whole study
period, %P was 78.6 (N ¼ 14).
Discussion
Although the lack of effective sampling may limit the
accuracy of our method for some areas of the River Po
plain (e.g. eastern Lombardy), the expansion of the pine
marten seems to be a recent phenomenon occurring only
in the western part of the River Po plain.
The penetration of the pine marten in arable lands
probably follows the main watercourses, which offer
patches of semi-natural woodland. On the basis of
record distribution, the colonisation follows a northsouth direction, the rivers Sesia, Agogna and Ticino
representing the main dispersal routes (see also Canova
and Rossi 2008, about raccoons Procyon lotor from
Switzerland).
In the Republic of Ireland, the increasing range of the
pine marten has been imputed to reduced persecution
and increased rates of afforestation, favouring habitat
connectivity (O’ Mahony et al. 2006). In northern Italy,
marten expansion in the plain may depend on demographic pressure, inducing dispersers to come down
from the Italian Pre-Alps, together with increasing food
availability in agricultural areas, as a consequence of the
use, starting from the 1980s, of more sustainable farming techniques (Stout 1986; Robson 1997). A positive
demographic trend favours immigration and range
expansion (e.g., about Mustelids: Blandford 1987;
Strachan and Jefferies 1996) and high population
pressure has been imputed to trigger the colonization
of Swiss cities by foxes (Gloor 2002). Unfortunately,
information on numbers and trend of pine marten
populations in Italy is inadequate to support any
hypothesis (Genovesi and De Marinis 2003).
In the middle part of the River Po plain, which seems
to have been reached by martens only in the last fiveyear period, woods are generally absent or consist of
relatively isolated, small fragments within an agricultural landscape matrix. Our results contrast with those
obtained in both Scottish (Balharry 1993a; Halliwell
1997) and Polish (Zalewski and Jedrzejewski 2006) study
areas, where a minimum wooded area was reported to
be needed by adult pine martens for survival and
reproduction.
Considering the sharp aversion of martens for open
areas, the colonization of such an unsuitable environment implies that pine martens are flexible enough
to adapt rapidly to novel environmental conditions
(Pigliucci 2001; West-Eberhard 2003). Behavioural
flexibility enhances colonization success, enabling the
persistence of initial populations before adaptive evolution can occur (Sax et al. 2005). Plasticity in both
territorial (Balharry 1993b) and feeding behaviour
(Marchesi 1989; Jedrzejewski et al. 1993) probably
allows the pine marten to face a variety of environmental conditions, as suggested by its distribution in
Mediterranean Italy. Accordingly, in the British Isles,
where the pine marten is the only autochthonous
Martes, martens have been able to survive in spite of
extensive deforestation and currently exploit also open
and rocky landscapes (Birks et al. 2004).
Nonetheless, agricultural areas might be suboptimal
habitats for pine martens, acting as dispersal sinks
(Kawecki 1995; Kirkpatrick and Barton 1997); in these
terms, marten presence (rather than colonisation) in the
plain would be occasional, depending on the immigration of non-resident floating male martens from
surrounding areas, for lack of vacant wooded territories.
In our study areas, genetic sampling does not seem to
support this hypothesis. For the Garzaia di Valenza,
the presence of the pine marten was stable during the
15-month study period; considering the first records for
this area, the species has probably been living there at
least from 2003. For S. Massimo, three years have
passed between the first record and the genetic
confirmation of pine marten presence.
Moreover, the few road-killed individuals for which
the sex was recorded included both males and females,
pregnant females with male offspring representing the
minimal viable propagule for the creation of new
populations (Fitch et al. 1952).
Variation in scat abundance has been used to
distinguish between resident and temporary populations
of pine martens (Balharry et al. 1996), but no empirical
A. Balestrieri et al. / Mamm. biol. 75 (2010) 412–419
evidence supports the assumption that low scat densities
correspond to non-breeding martens, marking activity
being less intense in low density populations (Birks et al.
2004). As suggested by the increasing rate of success of
DNA amplification in 2008, the small number of
identified faeces probably depended largely on the
overlong interval between consecutive surveys carried
out in 2007 and, secondarily, the need for a ‘‘training’’
period to reduce the rate of faeces misidentification
(Davison et al. 2002).
Discussing his competition hypothesis, Delibes (1983)
suggested that in altered environments the stone marten
is displaced to more urbanised areas, but would tend to
become less synanthropic wherever the density of the
pine marten decreased. The expansion of pine martens
into an agricultural landscape offers the opportunity of
analysing the effects of the opposite process on the
distribution of the stone marten.
Recent studies have shown contrasting results about
the relationship between the two species at a local scale.
The pine marten was the only Martes in a mountainous
area of north-western Spain (Rosellini et al. 2008),
whilst in a forest-field mosaic of central Poland the two
martens exploited the same micro-habitats (Posluszny
et al. 2007).
The possibility of the two species coexisting probably
depends on a combination of several factors, including
the relative abundance of each predator species within
the local carnivore guild and food availability. Our
results for the best monitored study area (Garzaia di
Valenza) suggest that the pine marten could be
dominant over the stone marten, displacing the latter
from the residual wooded areas of the River Po plain.
The PCR-RFLP analysis of faecal DNA was shown
to be a powerful method for distinguishing the species of
the genus Martes and is suitable, together with traditional survey methods, for assessing the spatial distribution of these elusive mustelids. Particularly, it represents
a useful tool for investigating the ecological relationship
between sympatric martens at a small scale. An ongoing
work on microsatellite genotyping with a set of 14 loci
useful for providing individual-specific genotypes of
pine marten from faecal DNA, will allow to assess also
the number of individuals occurring within the study
areas.
Further field work is needed to confirm the stable
presence of the pine marten in other areas of the western
River Po plain and to test the interspecific relationships
of martens across a wider range of environmental
conditions.
Acknowledgements
The authors would like to thank the River Po Park
(Vercelli and Alessandria provinces) for financial
417
support. They are particularly grateful to Giovanni
Boano and Roberto Sindaco, who shared with them the
data banks of Piedmont region. Laura Gola, Paolo
Debernardi and Elena Patriarca provided further
unpublished records of the pine marten in north-western
Italy, whilst Luca Lapini supplied helpful information
about road-kill monitoring in the North-East. Sara
Genovese helped with field work for her degree thesis at
the University of Pavia. Christopher F. Mason kindly
improved the English language. A. Ruiz-González holds
a Ph.D. fellowship awarded by the Dept. of Education
Universities and Research (Basque Government).
References
Albert, S., Ramotnik, C.A., Gregory Schmitt, C., 2004.
Collared peccary range expansion in Northwestern New
Mexico. Southwest. Nat. 49 (4), 524–528.
Balharry, D., 1993a. Factors affecting the distribution and
population density of pine marten (Martes martes L.) in
Scotland. Ph.D. Thesis, University of Aberdeen.
Balharry, D., 1993b. Social organization in martens: an
inflexible system? Symp. Zool. Soc. London 65, 321–345.
Balharry, E.A., McGowan, G.M., Kruuk, H., Halliwel, E.,
1996. Distribution of pine martens in Scotland as determined by field survey and questionnaire. SNH Survey and
Monitoring Report No. 48. Scottish Natural Heritage,
Edinburgh, UK.
Benedict, R.A., Genoways, H.H., Freeman, P.W., 2000.
Shifting distributional patterns of mammals in Nebraska.
Trans. Nebr. Acad. Sci. 26, 55–84.
Birks, J.D.S., Messenger, J.E., Braithwaite, A.C., Davison, A.,
Brookes, R.C., Strachan, C., 2004. Are scat surveys a
reliable method for assessing distribution and population
status of pine martens? In: Harrison, D.J., Fuller, A.K.,
Proulx, G. (Eds.), Martens and Fishers (Martes) in HumanAltered Environments: An International Perspective.
Springer, New York, pp. 235–252.
Blandford, P.R.S., 1987. Biology of the polecat Mustela
putorius: a literature review. Mamm. Rev. 17, 155–198.
Bon, M., Paolucci, P., Mezzavilla, F., Battisti, R., Vernier, E.,
1995. Atlante dei Mammiferi del Veneto. Lav. Soc. Ven.
Sci. Nat. 21 (Suppl.).
Boitani, L., 2003. Wolf conservation and recovery. In: Mech,
L.D., Boitani, L. (Eds.), Wolves. Behaviour, Ecology,
and Conservation. The University of Chicago Press,
pp. 317–340.
Brainerd, S.M., Rolstad, J., 2002. Habitat selection by
Eurasian pine martens Martes martes in managed forests
of southern boreal Scandinavia. Wildl. Biol. 8, 289–297.
Brashares, J.S., Arcese, P., Sam, M.K., 2001. Human
demography and reserve size predict wildlife extinction
in West Africa. Proc. R. Soc. London B Biol. 268,
2473–2478.
Broekhuizen, S., 1983. Habitat use of beech marten (Martes
foina) in relation to landscape elements in a Dutch
agricultural area. In: Proceedings of the XVI Congress
International Union of Game Biologists, CSSR,
pp. 614–624.
418
A. Balestrieri et al. / Mamm. biol. 75 (2010) 412–419
Buskirk, S.W., 1992. Conserving circumboreal forests for
martens and fishers. Conserv. Biol. 6, 318–320.
Canova, L., Rossi, S., 2008. First records of the northern
raccoon Procyon lotor in Italy. Hystrix Ital. J. Mammal.
(n.s.) 19 (2), 179–182.
Carroll, C., Noss, R.F., Paquet, P.C., 2001. Carnivores as
focal species for conservation planning in the Rocky
Mountain region. Ecol. Appl. 11, 961–980.
Creel, S., Creel, N.M., 1996. Limitation of African wild dogs
by competition with larger carnivores. Conserv. Biol. 10,
526–538.
Crooks, K.R., 2002. Relative sensitivities of mammalian
carnivores to habitat fragmentation. Conserv. Biol. 16,
488–502.
Davison, A., Birks, J.D.S., Brookes, R.C., Braithwaite, T.C.,
essenger, J.E., 2002. On the origin of faeces: morphological versus molecular methods for surveying rare
carnivores from their scats. J. Zool. (London) 257,
141–143.
Delibes, M., 1983. Interspecific competition and the habitat of
the stone marten Martes foina (Erxleben 1777). Eur. Acta
Zool. Fenn. 174, 229–231.
Marinis, A.M., Massetti, M., 1993. Distribution of the pine
marten Martes martes L., 1758 (Mammalia, Carnivora) on
the island of Elba, northern Tyrrhenian sea. Suppl. Ric.
Biol. Selvaggina 21, 263–267.
Durant, S.M., 1998. Competition refuges and coexistence: an
example from Serengeti carnivores. J. Anim. Ecol. 67,
370–386.
Fitch, H.S., Goodrum, P., Newman, C., 1952. The armadillo
in the southwestern United States. J. Mammal. 33, 21–37.
Frey, J.K., 1992. Response of a mammalian faunal element to
climatic changes. J. Mammal. 73, 43–50.
Geluso, K., 2004. Westward expansion of the Eastern fox
squirrel (Sciurus niger) in northeastern New Mexico and
southeastern Colorado. Southwest. Nat. 49, 111–116.
Geluso, K., Hoffman, J.D., Ashe, V.A., White, J.A., Bogan,
M.A., 2005. Westward expansion of the Tawny-Bellied
Cotton Rat (Sigmodon fulviventer) in West-Central New
Mexico. Southwest. Nat. 50 (2), 73–77.
Genovesi, P., 1993. Strategie di sfruttamento delle risorse e
struttura sociale della faina (Martes foina Erxleben 1777) in
ambiente forestale e rurale. Ph.D. Thesis, Università di
Roma ‘‘La Sapienza,’’ Roma.
Genovesi, P., De Marinis, A.M., 2003. Martes martes. In:
Boitani, L., Lovari, S., Vigna Taglianti, A. (Eds.), Fauna
d’Italia. Mammalia III, Carnivora – Artiodactyla. Bologna,
Ed. Calderini, p. 108.
Gloor, S., 2002. The rise of urban foxes (Vulpes vulpes) in
Switzerland and ecological and parasitological aspects of a
fox population in the recently colonised city of Zurich.
Ph.D. Thesis, University of Zurich, Switzerland.
Gómez-Moliner, B.J., Cabria, M.T., Rubines, J., Garin, I.,
Madeira, M.J., Elajalde, A., Aihartza, J., Fournier, P.,
Palazón, S., 2004. PCR-RFLP identification of mustelid
species: European mink (Mustela lutereola), American mink
(Mustela vison) and polecat (Mustela putorius) by analysis
of excremental DNA. J. Zool. London 262, 311–316.
Gorman, M.L., Mills, M.G., Raath, J.P., Speakman, J.R.,
1998. High hunting costs make African wild dogs vulner-
able to kleptoparasitism by hyaenas. Nature (London) 391,
479–481.
Goszczynski, J., 1976. Composition of the food of martens.
Acta Theriol. 21, 527–534.
IUCN, 2000. Guidelines for the Prevention of Biodiversity
Loss caused by Alien Invasive Species. IUCN, Gland.
Halliwell, E.C., 1997. The ecology of red squirrels in Scotland
in relation to pine marten predation. Ph.D. Thesis,
University of Aberdeen.
Jedrzejewski, W., Jedrzejewska, B., Szymura, L., 1995. Weasel
population response, home range, and predation on rodents
in a deciduous forest in Poland. Ecology 76, 179–195.
Jedrzejewski, W., Zalewski, A., Jedrzejewska, B., 1993.
Foraging by pine marten Martes martes in relation to
variable food resources in Bialowieza National Park. Acta
Theriol. 38, 405–426.
Jensen, A., Jensen, B., Husmaren (Martes foina) og marjagten
in Danmark 1967–1968. Danske Vildtundersogelser 15,
Vildtbiol, 1970, Station, Kalo-Ronde.
Kawecki, T.J., 1995. Demography of source-sink populations and the evolution of ecological niches. Evol. Ecol. 9,
38–44.
Kelly, M.J., Laurenson, M.K., FitzGibbon, C.D., Collins,
D.A., Durant, S.M., Frame, G.W., Bertram, B.C.R., Caro,
T.M., 1998. Demography of the Serengeti cheetah population: the first twenty-five years. J. Zool. (London) 244,
473–488.
Kerr, J.T., Currie, D.J., 1995. Effects of human activity on
global extinction risk. Conserv. Biol. 9, 1528–1538.
King, C.M., 1989. The natural history of weasels and stoats.
Christopher Helm.
Kirkpatrick, M., Barton, N.H., 1997. Evolution of a species
range. Am. Nat. 150, 1–23.
Kruger, H.H., 1990. Home ranges and patterns of distribution
of stone and pine martens. In: Myberget, S. (Ed.),
Transactions 19th International Congress of Game Biologists. Trondheim, Norwegian Institute for Nature Research, pp. 348–349.
Lanszki, J., 2003. Feeding habits of stone martens in a
Hungarian village and its surroundings. Folia Zool. 52,
367–377.
Lindstrom, E.R., Brainerd, S.M., Helldin, J.O., Overskaug,
K., 1995. Pine marten–red fox interactions: a case of
intraguild predation? Ann. Zool. Fenn. 32, 123–130.
Linnell, J.D.C., Strand, O., 2000. Interference interactions, coexistence and conservation of mammalian carnivores.
Diversity Distrib. 6, 169–176.
O’ Mahony, D., O’Reilly, C., Turner, P., 2006. National Pine
Marten Survey of Ireland 2005. Coford Connects, Environment No 7, pp. 1–8.
Marchesi, P., 1989. Écologie et comportement de la martre
(Martes martes L.) dans le Jura Suisse. Ph.D. Thesis,
Université de Neuchâtel.
Marchesi, P., Lachat, N., Leinhard, R., Debieve, P., Mermod,
C., 1989. Comparaison des régimes alimentaires de la
fouine (Martes foina Erxl.) et de la martre (Martes martes
L.) dans une région du Jura suisse. Rev. Suisse Zool. 96,
281–296.
Martinoli, A., 2001. Martes martes. In: Prigioni, C., Cantini,
M., Zilio, A. (Eds.), Atlante dei Mammiferi della Lombar-
A. Balestrieri et al. / Mamm. biol. 75 (2010) 412–419
dia. Regione Lombardia e Università degli Studi di Pavia,
pp. 236–238.
Novikov, G.A., 1962. K ekologii kamennoi kunicy v
lesostepnych dubravach. Bjull. Mosk. Obsc. Ispit. Prir.
Otd. Biol. 47 (6), 5–16.
Palomares, F., Caro, T.M., 1999. Interspecific killing among
mammalian carnivores. Am. Nat. 153, 492–508.
Palomares, F., Gaona, P., Ferreras, P., Delibes, M., 1995.
Positive effects on game species of top predators by
controlling smaller predator populations: an example with
lynx, mongooses and rabbits. Conserv. Biol 9, 295–305.
Parmesan, C., Yohe, G., 2003. A globally coherent fingerprint
of climate change impacts across natural systems. Nature
421, 37–42.
Pigliucci, M., 2001. In: Phenotypic Plasticity: Beyond Nature
and Nurture. Johns Hopkins University Press, Baltimore.
Pilot, M., Gralak, B., Goszczynski, J., Posluszny, M., 2007.
A method of genetic identification of pine marten (Martes
martes) and stone marten (Martes foina) and its application
to faecal samples. J. Zool. London 271, 140–147.
Pittiglio, C., 1996. Analisi comparativa di uso e selezione del
habitat della faina e della martora in condizioni di
simpatria. Degree Thesis, Università di Roma, ‘‘La
Sapienza’’, Roma.
Posluszny, M., Pilot, M., Goszczynski, J., Gralak, B., 2007.
Diet of sympatric pine marten (Martes martes) and stone
marten (Martes foina) identified by genotyping of DNA
from faeces. Ann. Zool. Fenn. 44, 269–284.
Powell, R.A., Zielinski, W.J., 1983. Competition and coexistence in mustelid communities. Acta Zool. Fenn. 174,
223–227.
Prigioni, C., Balestrieri, A., Remonti, L., 2007. Decline and
recovery in otter Lutra lutra populations in Italy. Mamm.
Rev. 37 (1), 71–79.
Prigioni, C., Balestrieri, A., Remonti, L., Cavada, L., 2008.
Differential use of food and habitat by sympatric carnivores in the eastern Italian Alps. Ital. J. Zool. 75 (2),
173–184.
Prigioni, C., Cantini, M., Zilio, A., 2001. Atlante dei
Mammiferi della Lombardia. Regione Lombardia e Università degli Studi di Pavia.
Prigioni, C., Sommariva, A., 1997. Ecology of the stone
marten (Martes foina) in the urban habitat of Cavalese
(Trento, Italy). Report Centro di Ecologia Alpina 11, 1–26.
Proulx, G., Aubry, K.B., Birks, J., Buskirk, S.W., Fortin, C.,
2004. World distribution and status of the genus Martes in
2000. In: Harrison, D.J., Fuller, A.K., Proulx, G. (Eds.),
Martens and Fishers (Martes) in Human Altered Environments: an International Perspective. Springer, New York,
pp. 21–76.
Robson, N., 1997. The evolution of the common agricultural
policy and the incorporation of environmental considerations. In: Pain, D., Pienckowski, M.W. (Eds.), Farming and
Birds in Europe. Academic Press, London, pp. 43–78.
Rosellini, S., Osorio, E., Ruiz-González, A., Piñeiro, A., Barja,
I., 2008. Monitoring the small-scale distribution of sympatric European pine martens (Martes martes) and stone
419
martens (Martes foina): a multievidence approach using
faecal DNA analysis and camera-traps. Wildl. Res. 35,
434–440.
Ruiz-González, A., Rubines, J., Berdión, O., Gómez-Moliner,
B.J., 2008. A non-invasive genetic method to identify the
sympatric mustelids pine marten (Martes martes) and stone
marten (Martes foina): preliminary distribution survey on
the northern Iberian Peninsula. Eur. J. Wildl. Res. 54 (2),
253–261.
Sacchi, O., Meriggi, A., 1995. Habitat requirements of the
stone marten (Martes foina) on the Tyrrhenian slopes of the
northern Apennines. In: Prigioni, C. (Ed.), Proc. II It.
Symp. on Carnivores. Hystrix (n.s.) 7(1–2), pp. 99–104.
Sadlier, L.M.J., Webbon, C.C., Baker, P.J., Harris, S., 2004.
Methods of monitoring red foxes Vulpes vulpes and badgers
Meles meles: are field signs the answer? Mamm. Rev. 34 (1),
75–98.
Savoldelli, P., Sindaco, R., 2008. Grandi e piccoli predatori.
Collana ‘‘La nostra fauna’’. Osservatorio Faunistico per la
Fauna Selvatica, Regione Piemonte.
Sax, D.F., Stachowicz, J.J., Gaines, S.D., 2005. In: Species
Invasions: Insights into Ecology, Evolution, and Biogeography. Sinauer, Sunderland, MA.
Sindaco, R., 2006. Segnalazioni faunistiche piemontesi e
valdostane. Riv. Piem. St. Nat. 27, 443–459.
Skogland, T., 1991. What are the effects of predators on large
ungulate populations? Oikos 1, 401–411.
Smedshaug, C.A., Selas, V., Lund, S.E., Sonerud, G.A., 1999.
The effect of a natural reduction of red fox Vulpes vulpes on
small game hunting bags in Norway. Wildl. Biol. 5,
157–166.
Storch, I., Lindstrom, E., Jounge, J., 1990. Diet and habitat
selection in pine martens in relation to competition with the
red fox. Acta Theriol. 35, 311–320.
Stout, V., 1986. What is happening to PCBs? Elements of the
environmental monitoring as illustrated by an analysis of
PCB trends in terrestrial and aquatic organisms. In: Waid,
J.S. (Ed.), PCBs and the Environment, vol. 2. CRC Press,
Boca Raton, FL, pp. 164–205.
Strachan, C., Jefferies, D.J., 1996. Otter Survey of England
1991–94, a Report on the Decline and Recovery of
the Otter in England and on its Distribution, Status
and Conservation in 1991–94. Vincent Wildlife Trust,
London.
T.C.I., 1957. Conosci l’Italia, vol., I, L’Italia fisica. Touring
Club Italiano, Milano, 318pp.
Waechter, A., 1975. Ecologie de la fouine en Alsace. Terre et
Vie 29, 399–457.
Walther, G.R., Post, E., Convey, P., Menzel, A., Parmesan,
C., Beebee, T.J., 2002. Ecological responses to recent
climate change. Nature 416, 389–395.
West-Eberhard, M., 2003. In: Developmental Plasticity and
Evolution. Oxford University Press, New York.
Zalewski, A., Jedrzejewski, W., 2006. Spatial organisation and
dynamics of the pine marten Martes martes population in
Bialowieza Forest (E Poland) compared with other
European woodlands. Ecography 29, 31–43.
Paper IIIc
Appendix c
Appendix C
PAPER IIIc
Balestrieri A, Remonti L, Ruiz-González A, Capelli E,
Vergara M, Gómez-Moliner BJ, Prigioni C.
Food habits of genetically identified pine marten (Martes
martes) expanding in agricultural lowlands (NW Italy).
Acta Theriologica. In press
- 229 -
Paper IIIc
Appendix c
- 230 -
Paper IIIc
Appendix c
PAPER IIIc
Food habits of genetically identified pine marten (Martes martes)
expanding in agricultural lowlands (NW Italy).
ABSTRACT
We assessed the diet of pine marten (Martes martes) expanding in the heavily human-altered agricultural
plain of the River Po, northern Italy. Between February 2008 and November 2009, surveys were carried
out twice a week during seasonal sampling sessions of 3-4 consecutive weeks. To distinguish the faecal
samples from those of sympatric carnivores, a PCR-RFLP method was applied. The availability of small
mammals was assessed by the analysis of 59 barn owl (Tyto alba) pellets. A total of 109 pine marten faeces
was analysed. Its diet consisted of fruit, rodents, lagomorphs and birds. Seasonal variation occurred for
fruit, which prevailed in summer, and rodents, which were more preyed upon in autumn. In winter, the
diet of the pine marten was almost totally based on vertebrates, lagomorphs were the main source of
protein in summer. Use of small mammals differed significantly from their availability, voles, particularly
bank vole Myodes glareolus, being preferred to mice (Apodemus sp.) and rats (Rattus sp.). Medium-size
mammals formed about 18% of pine marten diet, a value generally reported for high latitude habitats. In
agricultural areas the pine marten proved to be an opportunistic predator, able to face the reduced
availability of small mammals by preying upon medium-size prey and fruit.
Key words: PCR-RFLP method, non-invasive sampling, rodent availability, diet analysis, range expansion
- 231 -
Paper IIIc
Appendix c
INTRODUCTION
The pine marten (Martes martes) occurs throughout much of Europe and northern and central Asia, from
northern Portugal to western Siberia (Stubbe 1993, Proulx et al. 2004). Although in the Mediterranean
area the species has been reported in insular maquis and coppices (De Marinis and Masseti 1993a, Murgia
et al. 1995) and cultivated land with woodland fragments (Pittiglio 1996, Pereboom et al. 2008), the pine
marten is generally associated with forest habitats, mainly mature coniferous and mixed forests (Delibes
1983, Buskirk 1992, Proulx et al. 2000). Deforestation and forest fragmentation have been reported to
affect the distribution and density of pine marten (Brainerd et al. 1994, Kurki et al. 1998), which are
believed to need a minimum woodland area to survive (Zalewski and Jędrzejewski 2006) and tend to avoid
treeless areas (Storch et al. 1990, Brainerd and Rolstad 2002, Pereboom et al. 2008). Accordingly, in the
Republic of Ireland, afforestation and increasing habitat connectivity have been suggested as the main
factors favouring the ongoing expansion of the pine marten (O’Mahony et al. 2006).
Nonetheless, in the last years the pine marten has progressively expanded in the western part of the
intensively cultivated plain of the River Po (NW Italy). In the core of this area, woods consist of isolated,
small fragments within an agricultural landscape matrix (Balestrieri et al. 2010).
Food availability is considered a main environmental factor influencing the population dynamics of small
mustelids (King 1989, Jędrzejewski et al. 1995). Like other Martes species, the pine marten is a generalist
predator, consuming a wide range of resources according to their local and seasonal availability (reviews in:
De Marinis and Masseti 1995, Zalewski 2004). Small mammals are its main prey throughout Europe,
representing up to 81% of all prey (on average ca. 45%; Zalewski 2004). Fruit (Marchesi and Mermod
1989), birds (Helldin 2000) and invertebrates (Clevenger 1993a) may sometimes represent the most
frequent food item, while medium- and large-size mammals (lagomorphs and ungulate carrion) generally
constitute secondary prey at high latitudes (Zalewski 2004). Amphibians (Reig and Jędrzejewski 1988),
fish (Lockie 1961) and crustaceans (Ruiz-Olmo and Nadal 1991) are eaten only occasionally.
In agricultural landscapes, the reduction of both habitat quality and diversity have been claimed as the
main causes of biodiversity loss and affect the availability of food resources for predators (Benton et al.
2003, Firbank 2005). In particular, small mammals are confined to field margins and non-cropped areas
scattered in cultivated fields (Fitzgibbon 1997) and in agricultural areas rodent communities show lower
diversity (Millán de la Peña et al. 2003) and biomass (Michel et al. 2006) than those in forested habitats.
The plasticity of pine marten food habits should allow the mustelid to cope with the shortage of its main
prey in sub-optimal habitats. Nonetheless, while pine marten diet has been largely studied in boreal and
- 232 -
Paper IIIc
Appendix c
deciduous forests (Marchesi and Mermod 1989, Jędrzejewski et al. 1993, Helldin 2000), in field-forest
mosaics it has still been poorly studied (Posluszny et al. 2007). The recent expansion of the pine marten in
north-western Italy offers the opportunity to investigate the feeding adaptability of a typical forestdwelling predator in heavily human-altered agricultural lowlands. The main aim of our study was to draw
a first picture of the food habits of the pine marten in the River Po plain by means of scat analysis. As pine
marten faeces cannot be distinguished from those of the stone marten (Martes foina), which is widespread
in the whole plain (Genovesi and De Marinis 2003), and can be also easily confused with those of other
carnivores (Davison et al. 2002, Harrington et al. 2010), molecular techniques were applied for the
identification of faecal samples.
To examine seasonal variation in the species diet, sampling for faeces was carried out over a 20-month
period. Small mammals being considered their preferred prey, use by pine marten of this resource was
compared to its availability in the study area.
We predicted that pine marten would compensate for the possible reduced availability of small mammals
with respect to forested habitats by relying on alternative food items according to their seasonal
availability.
Study area
Surveys were carried out in two sites of the western plain of the River Po (Fig. 1), each covering about 4
km2 and consisting of small patches of hygrophilous woods merged into an agricultural landscape matrix.
For both study sites mean altitude is 90 m a.s.l. and climate is sub-continental temperate, with an annual
average temperature of 12.4 °C and average rainfall of about 1000 mm.
The first site partly coincided with the Natural Reserve "Garzaia di Valenza" (45° 01’ N, 8° 64’ W;
hereafter: NRGV), on the left bank of the River Po. The whole territory is flat, extensively covered by
cultivated rice and maize fields (ca. 57%) and poplar (Populus sp.) plantations (ca. 19%). Woods (ca.
13%) consist of willows (Salix cinerea, S. alba), oak (Quercus robur), poplars (Populus alba and various
hybrids), alder (Alnus glutinosa) and black locust (Robinia pseudoacacia), bordering an abandoned river
meander and three naturalized artificial lakes.
The second site was included in the Natural Reserve “San Massimo” (45° 18’ N, 8° 99’ W; hereafter:
NRSM) 5 km to the west of the River Ticino. Alder woods and mixed hygrophilous assemblages (ca.
30%), where alders are joined by willows (S. alba) and poplars (P. alba and P. nigra), are surrounded by
rice and maize fields (ca. 54%). There are a few farms scattered throughout the site.
- 233 -
Paper IIIc
Appendix c
Figure 1. Study area. 1: Natural Reserve "Garzaia di Valenza"; 2: Natural Reserve "S. Massimo”.
MATERIALS AND METHODS
Field surveys
Sampling was carried out between February 2008 and November 2009, along two linear transects in each
study site (2.2 and 2.3 km in NRGV and 2.6 and 1.8 km in NRSM), coinciding with paths, country
roads and wood/field margins and covering both open and forested habitats.
As climate, vegetation cover and land use are very similar in the two study sites, we assumed that both
ecological conditions and food availability for pine marten were the same. Thus diet data were merged as
to provide an overall picture of the species food habits in the agricultural landscape.
To reduce the risk for faecal DNA degradation, surveys were carried out twice a week during seasonal
sampling sessions of 3-4 consecutive weeks (N = 56).
Faecal samples were initially assigned to the genus Martes according to their size (diameter < 10 mm) and
shape (Bang and Dahlström 1974). A portion (about 20%) of each faecal sample was picked up with sticks
and preserved in 96% ethanol and by freezing for DNA extraction, while the rest was retained for dietary
analysis.
- 234 -
Paper IIIc
Appendix c
Genetic analyses
DNA was isolated using the QIAamp DNA Stool Mini Kit (Qiagen) according to the manufacturer’s
instructions. The specific identification of faecal samples was accomplished by a polymerase chain reaction
– restriction fragment length polymorphism (PCR-RFLP) method, providing for an effective genetic
identification of sympatric marten species (Ruiz-González et al. 2008). The procedure consists of PCR
amplifying a mitochondrial D-loop region by two specifically designed primers - Mm_L1 (5’CCCAAAGCTGACATTCTAAC-3’) and Mm_H1 (5’-ATGGGCCCGGAGCGAGAAGAGGTACAC3’). These primers amplify the mtDNA from Martes martes, M. foina and four Mustela species, of which
only M. putorius had been previously reported for our study area. Other sympatric carnivore species, in
particular the red fox, whose scats can be easily mistaken for those of martens (Davison et al. 2002), render
no amplicons. The resulting 276-bp-long amplicons were digested by two restriction enzymes
simultaneously - HaeIII and RsaI – allowing the differentiation of the two Martes species and both of
them from the other mustelids whose mtDNA is amplified by the selected primers (see Ruiz-González et
al. 2008 for further details).
Diet analysis
Faecal samples were washed through three sieves of 1.5, 0.3 and 0.1 mm mesh and food remains inspected
to count or estimate the total numbers of each kind of food.
Mammal hairs were compared at 20x and 40x magnifications with the keys of Debrot et al. (1982),
Teerink (1991) and a personal collection of hair photos. Bird feathers were identified with reference to
Day (1966). The undigested remains of insects (wings, legs and cuticle parts) and wild or cultivated fruits
(seeds) were identified using personal collections. Since marten are known to consume earthworms (Lynch
and McCann 2007), sediment remaining in the sieve with the smallest mesh was examined under a
binocular microscope to detect the presence of chaetae.
Although conversion factors (e.g.: Zielinski 1986, Jędrzejewska and Jędrzejewski 1998) make it possible to
assess biomass consumed from the dry weight of prey remains, we agree with Zielinski and Duncan (2004)
on the difficulty of separating the undigested remains effectively. As a consequence, the contribution of
each food item in terms of volume was assessed according to Kruuk and Parish’s method (1981), which
provides for quantifications of food volumes as reliable as those obtained by the analysis of stomach
contents (Balestrieri et al. in press). For each faecal sample, the method entails the estimate by eye of the
bulk of each item ‘as ingested‘. This estimate presents little difficulty for small prey, which are ingested in
- 235 -
Paper IIIc
Appendix c
their entirety, whilst for larger prey the relative volume is computed as 100 minus the score given to the
other remains. To reduce the equating of occurrence bias (Kelly 1991, Ciucci et al. 1996), undigested
remains occurring in negligible proportions were not considered in the analysis. Prey were identified to the
lowest possible level.
Results were expressed as per cent mean volume (%mV = total estimated volume of each food item as
ingested / total number of examined spraints), which represents the proportional contribution of each
food item to the overall diet (Kruuk and Parish 1981). To allow the comparison with previous studies,
data were also expressed as per cent frequency of occurrence (%FO: number of faecal samples containing a
specific food item / total number of faecal samples x 100) and per cent relative frequency of occurrence
(%RFO = number of occurrences of each item / total number of items x 100).
Seasonal trophic niche breadth was estimated by Levins’ index B = 1
(∑ p ) (Levins 1968), using the
2
i
proportions of occurrence (pi = %mV/100) of five main food categories (fruit, insects, birds, lagomorphs
and rodents). The index ranges between 1, when only one food item is used, and 5, when all items are
consumed in equal proportions.
Rodent availability
The diet of the barn owl is considered an accurate reflection of the relative abundance of small mammals
(Libois 1984, Taberlet 1986, Clark and Bunck 1991, Taberlet and Fumagalli 1995, Love et al. 2000). To
assess rodent availability for pine marten in the study area, the relative abundance of rodent species in barn
owl Tyto alba diet was investigated through the analysis of 59 pellets collected in winter (N = 39) and
summer (N = 20) 2009 under two barn owl shelters (one in NRGV and one in NRSM). Pellets were
stored individually until analysis. Each pellet was then soaked in water and then teased apart using
tweezers and a needle. Prey remains were identified on the basis of reference keys (Chaline et al. 1974,
Amori et al. 1986) and personal collections of rodent skulls. Data were expressed as %RFO (= number of
occurrences of each species / total number of small mammals preyed x 100).
Statistical analysis
Spearman’s rank correlation test (rs) was used to check for any relationship between the proportion of use
(%mV) of the main food categories.
To analyse seasonal variation in pine marten diet, data were split as follows: winter: Jan-Mar; spring: AprJun; summer: Jul-Sep; autumn: Oct-Dec. As the proportions (%mV/100) describing diet composition
sum to one, to overcome the ‘unit sum constraint’ seasonal variation was assessed by Compositional
- 236 -
Paper IIIc
Appendix c
Analysis (Aitchison 1986, Aebischer et al. 1993). Ratios were calculated using the proportional data for the
five main food items. As a first step, raw seasonal volumetric data were compared by Kruskal-Wallis’ test
and the group for which the null hypotheses had the lowest probability of being correctly rejected (insects:
χ2 = 2.1, 3 d.f., P = 0.54) was used as the denominator in the transformation. Logarithms (ln) of the
resulting four ratios were then calculated to normalize their distribution (Aitchison 1986). As in the logratio transformation zero proportions cannot be computed, zeros were replaced with 0.01 (Aebischer et al.
1993). Data were analysed using Multivariate Analysis of Variance (MANOVA) and Games-Howell posthoc test, which does not rely on homogeneity of variance. Variance being significantly different, Welch
and Brown-Forsythe statistics were used to test the results of separate univariate ANOVAs before rejecting
the null hypothesis. To account for multiple tests on repeated data the level of significance of post-hoc
tests was calculated by Bonferroni’s sequential technique (Rice 1989).
The chi-squared (χ2) test and Bonferroni’s confidence intervals for the proportion of use were used to
compare the use by pine marten of six species/groups of small mammals (insectivores, Gliridae, Myodes
glareolus, Microtus sp., Rattus sp. and Apodemus sp.) to their availability as assessed by the analysis of barn
owl pellets (p = number of occurrences of each group / total number of small mammals in the diet of both
the pine marten and barn owl).
RESULTS
Eighty-one per cent of faecal samples was successfully genotyped, yielding 109 pine marten faeces. The
diet of the pine marten included fruit (%mV = 32.5), small rodents (%mV = 28.3), medium-size
mammals (lagomorphs; %mV = 17.7) and birds (%mV = 15.5; Tab. 1). Wild cherries (Prunus sp.),
blackberries (Rubus sp.) and figs (Ficus carica) were the main fruit, while voles, particularly bank voles
(Myodes glareolus), were a more frequent prey than both mice (Apodemus sp.) and other voles (Microtus
savii). Coypu (Myocastor coypu) were found in 2.7% of faecal samples. Birds included mainly small
passerines and pigeons (Columba sp.), while lagomorphs consisted of nearly equal proportions of hares
(Lepus europaeus) and introduced Eastern cottontails (Sylvilagus floridanus). Invertebrates, mainly imagos
and larvae of Coleoptera, were eaten in negligible proportions and did not include earthworms (Tab. 1).
The %mV of fruit was inversely correlated with those of rodents (rs = -0.42, P < 0.0001) and lagomorphs
(rs = -0.29, P < 0.01), while the %mV of rodents was inversely correlated with that of lagomorphs (rs = 0.32, P < 0.001)
- 237 -
Paper IIIc
Appendix c
Table 1. Diet of the pine marten as assessed by the analysis of 109 faeces (%mV = per cent mean volume;
%FO = per cent frequency of occurrence; %RFO = per cent relative frequency of occurrence; N = number
of identified items).
Food items
N
%mV
%FO
%RFO
Fruit
46
32.5
41.3
30.1
Undetermined fruit
8
5.6
7.3
3.4
Phytolaccaceae
1
0.2
0.9
0.7
1
0.2
0.9
0.7
10
6.5
9.2
6.8
Phytolacca americana
Moraceae
10
6.5
9.2
6.8
29
20.1
26.6
19.9
Rubus sp.
12
9.1
11.0
8.2
Prunus sp.
14
8.6
12.8
9.6
Ficus carica
Rosaceae
Pirus communis
Vitaceae
Vitis vinifera
Insects
4
2.4
3.7
2.7
1
0.1
0.9
0.7
1
0.1
0.9
0.7
6
2.7
5.5
4.1
Coleoptera imagos
2
1.4
1.8
1.4
Coleoptera larvae
2
1.0
1.8
1.4
Hymenoptera
1
0.3
0.9
0.7
Orthoptera
1
0.1
0.9
0.7
Birds
25
15.5
22.9
17.1
Undetermined birds
1
0.9
0.9
0.7
Galliformes
2
1.2
1.8
1.3
Passeriformes
16
8.6
14.7
11.0
Columbiformes
6
4.8
5.5
4.1
61
48.4
56.0
41. 8
Small mammals
39
28.7
26.6
Sorex araneus
1
0.4
31.2
0.9
Glis glis
1
0.9
0.9
0.7
Muscardinus avellanarius
4
1.8
3.7
2.7
Myodes glareolus
19
15.3
17.4
13.0
Microtus savii
4
3.7
3.7
2.7
Rattus sp.
1
0.9
0.9
0.7
Apodemus sp.
4
2.8
2.7
Undetermined rodents
5
2.9
3.7
3.4
3.4
15.1
Mammals
0.7
Medium-size mammals
22
17.7
20.2
Sylvilagus floridanus
11
9.0
7.5
7.5
Lepus europaeus
9
7.1
6.2
Undetermined lagomorphs
2
1.6
6.2
1.4
Large mammals
Myocastor coypus
4
4
- 238 -
2.9
2.9
2.7
2.7
1.4
2.7
2.7
Paper IIIc
Appendix c
The proportion of use of the main food categories varied among seasons (Wilks’ Lambda = 0.525, F = 6.2,
12 d.f., P < 0.0001; Tab. 2). Significant variation occurred for fruit (F = 12.2, 3 d.f., P < 0.0001) and
rodents (F = 8.4, 3 d.f., P < 0.0001). Games-Howell post-hoc tests showed that fruit was eaten more in
summer than in autumn and winter (P < 0.01 for both comparisons), while rodents were more preyed
upon in autumn than in summer (P < 0.01). On the whole, fruit formed the bulk of marten diet in
summer and rodents in autumn (mV% ≈ 70 for both items), while in the other two seasons the overall
share of lagomorphs and birds ranged between 36% and 57% of the overall diet. Accordingly, niche
breadth was the lowest in summer-autumn (B = 2.0 for both seasons) and increased progressively in winter
(B = 3.2) and spring (B = 3.9). In winter, the diet of the pine marten was almost totally based on
vertebrates, whilst lagomorphs were the main source of protein in summer.
Table 2. Seasonal variation in both per cent relative frequency of occurrence (%RFO) and per cent mean
volume (%mV) of the main food items in pine marten diet. Number of faecal samples analysed in
brackets.
Food items
Fruit
Insects
Birds
Lagomorphs
Rodents
Spring (32)
%RFO
%mV
31.1
33.9
3.34
13.9
22.0
26.9
4.4
15.6
17.8
24.4
Summer (31)
%RFO %mV
59.5
4.8
9.5
11.9
7.1
- 239 -
67.7
4.8
7.4
14.2
5.8
Autumn (21)
%RFO
%mV
13.8
6.9
13.8
3.4
58.6
12.1
1.9
12.1
4.8
69.0
Winter (25)
%RFO %mV
3.3
0.0
33.3
26.7
33.3
4.0
0.0
30.4
27.2
38.4
Paper IIIc
Appendix c
Remains of 95 small mammals were recovered from barn owl pellets. Wood mice (Apodemus sylvaticus)
were the most common prey, followed by rats (Tab. 3). Use of small mammals by the pine marten differed
significantly from their availability (χ2 = 73.8, 4 d.f., P < 0.001). Pine marten preyed upon bank voles
more than expected, while the opposite was true for mice and rats (Fig. 2).
Table 3. Per cent relative frequency of occurrence (%RFO) of small mammal species in barn owl pellets.
Small mammals
N
%RFO
Sorex araneus
Crocidura suavolens
Crocidura leucodon
Muscardinus avellanarius
Rattus sp.
Rattus norvegicus
Apodemus sylvaticus
Myodes glareolus
Microtus arvalis
Microtus savii
Arvicola terrestris
1
2
2
1
14
12
38
14
1
10
1
1.0
2.1
2.1
1.0
14.6
12.5
39.6
14.6
1.0
10.4
1.0
Figure 2. Use of small mammals by the pine marten vs. their availability (*: P < 0.001, Bonferroni’s
confidence intervals; p = number of occurrences of each group / total number of small mammals in the
diet of both the pine marten - use -, and barn owl - availability).
- 240 -
Paper IIIc
Appendix c
DISCUSSION
The misidentification of scats can lead to the misrepresentation of the diet of the species being investigated
(Reed et al. 2004). The PCR-RFLP method used, having been specifically designed for faecal material
(Ruiz-González et al. 2008), allowed us to distinguish pine marten samples from those of sympatric
carnivore species and, consequently, to draw a reliable picture of the diet of this mustelid in agricultural
areas. Genotyping success was rather high with respect to other approaches (e.g.: 58%, Lucentini et al.
2007; 53.4%, Pilot et al 2006) and similar to those obtained, by the same method, in the Iberian
peninsula (Ruiz-González et al. 2008, Rosellini et al. 2008), suggesting that the method is robust. The
shorter time interval to scat collection (3-4 vs. 7-8 days), with respect to previous applications of the
method in the study area, probably played a major role in the improvement of our success rate (81% vs.
21.4%, Balestrieri et al. 2008; 30.3%, Balestrieri et al. 2010), although the effect of the increased
experience of the surveyors cannot be ruled out.
As reported for several studies throughout Europe, in the study area small mammals and fruit were the
main food resources for pine marten. Nonetheless, fruit consumption in winter and that of rodents in
summer were much lower than those previously reported for other Mediterranean areas (on average,
%RFO = 52.7 and 27.5, respectively; Zalewski 2004). The low consumption of small mammals in
summer may depend on the availability of more profitable food resources, namely fruit (see Cavallini and
Volpi 1996), which in the same season were exploited by pine marten more heavily than in most study
areas (Zalewski 2004), and lagomorphs, which represented their main vertebrate prey in summer. The
composition of the small mammal community assessed by the analysis of barn owl pellets agreed with that
obtained by live-trapping in the central plain of the River Po (Giordano and Meriggi 2009). The
dominant rodent, the wood mouse, is a widespread and opportunistic species, which can occur in a wide
variety of habitats (Ouin et al. 2000), whilst the bank vole, although a typical forest-dwelling species, may
also live in the small wood-patches scattered in agricultural landscapes (Fitzgibbon 1997, Butet et al.
2006).
The preference shown by pine marten for voles is in agreement with previous studies, which demonstrated
that microtines are the most important food resource for the species throughout its range (De Marinis and
Masseti 1995, Zalewski 2004). In accordance with the findings from continental Europe (Jędrzejewski et
al. 1993, Pulliainen and Ollinmäki 1996), the bank vole seemed to be preferred to Microtus voles,
although the last ones are considered to be more profitable prey (Buskirk and Macdonald 1984) and are
selected by British pine marten (Gurnell et al. 1994, Putman 2000). Both wood mice and bank voles are
- 241 -
Paper IIIc
Appendix c
typical forest species, although they can be found in a wide variety of habitats (Canova and Fasola 1991,
Caryl 2008), while Microtus voles are generally restricted to open grassland and hedgerows (Miklós and
Žiak 2002, Caryl 2008). Pine marten preference for bank voles may thus depend on the behaviour of the
predator rather than the profitability of the prey, pine marten being more inclined to search for prey under
the tree canopy than in open fields (Jędrzejewski et al. 1993). The low interest in wood mice shown by
pine marten has been considered as a consequence of their greater agility compared to voles (Jędrzejewski
et al. 1993), while rats, although they can represent the main prey of insular pine marten (De Marinis and
Masseti 1993b), are seldom preyed upon on the continent (De Marinis and Masseti 1995). In contrast,
pine marten fed on coypu, probably carrions or young individuals, which are much more vulnerable to
predation that adults (see also Rosell and Hovde 1998 about Eurasian beavers).
Medium-size mammals generally form less than 10% of the total diet, both in northern (e.g. Helldin
2000, Caryl 2008) and southern Europe (e.g. Marchesi and Mermod 1989, Clevenger 1993b), while, in
our study area, hares and Eastern cottontails totalled about 18% of pine marten diet (%RFO ≈ 15).
According to the generalized model of latitudinal variation in %RFO of the main food items in winter
diet constructed by Zalewski (2004), such values are predicted for the northern range of the species (> 60°
N).
In both study sites, lagomorph availability for pine marten is probably high (NRGV: 15-20
cottontails/km2 and 1-2 hares/km2, Balestrieri et al. 2005, NRSM: 80-100 cottontails/km2 , 8-10
hares/km2 and 8-10 rabbits/km2; D. Massignani, pers. comm.), mainly as a consequence of the rapid
spread of introduced cottontails (Vidus Rosin et al. 2008). As previously reported for the red fox
(Balestrieri et al. 2005), cottontails, which select wood-field margins as resting sites (Swihart and Yahner
1982, Althoff et al. 1997), may represent a more profitable prey for pine marten than hares.
As in forested habitats (Jędrzejewski et al. 1993, Pulliainen and Ollinmäki 1996, Lynch and McCann
2007), the pine marten proved to be a generalist predator in the recently colonised plain of the River Po.
The greater importance of fruit in summer and lagomorphs throughout the year with respect to previous
studies may be explained as an adaptation to the reduced availability of small mammals. This feeding
flexibility is likely to play a major role in enhancing the species success in the colonization of the heavy
human-altered agricultural landscape of northern Italy.
- 242 -
Paper IIIc
Appendix c
ACKNOWLEDGEMENTS
A. Ruiz-González holds a Ph.D. fellowship awarded by the Dept. of Education Universities and Research of the
Basque Government (Ref. BFI09.396). This study has been partially supported by the Basque Country Government
to the Research group “Systematic, Biogeography and Population Dynamics” (Ref. IT317-10;GIC10/76) and by the
River Po Park (Vercelli and Alessandria provinces). We are grateful to P. Chanin, for suggestions and the revision of
the English style, and L. Gola (NRGV) and D. Massignani (NRSM) for their help during field surveys. Two
anonymous referees improved the manuscript with their suggestions.
REFERENCES
Aebischer N.J., Robertson P.A. and Kenward R.E. 1993. Compositional analysis of habitat use from animal radiotracking data. Ecology 74: 1313–1325.
Aitchison J. 1986. The Statistical Analysis of Compositional Data. Chapman & Hall, New York: 1-416.
Althoff D.P., Storm G.L., Dewalle D.R., 1997. Daytime habitat selection by cottontails in central Pennsylvania.
Journal of Wildlife Management 61: 450-459.
Amori G., Cristaldi M. and Contoli L. 1986. [On rodents (Gliridae, Arvicolidae, Muridae) from peninsular and
insular Italy in relation to the Mediterranean climate]. Animalia 11: 217-269. [In Italian]
Balestrieri A., Remonti L. and Prigioni C. 2005. Local feeding specialization of the red fox (Vulpes vulpes) in response
to eastern cottontail (Sylvilagus floridanus) introduction (NW Italy). Hystrix Italian Journal of Mammalogy
16 (2): 113-126.
Balestrieri A., Remonti L. and Prigioni C. (in press). Assessing red fox diet by faecal samples and stomach contents:
an Alpine case study. Central European Journal of Biology.
Balestrieri A., Remonti L., Ruiz-González A., Gómez-Moliner B.J., Vergara M. and Prigioni C. 2010. Range
expansion of the pine marten (Martes martes) in an agricultural landscape matrix (NW Italy). Mammalian
Biology 75: 412-419.
Balestrieri A., Ruiz-González A., Remonti L., Gómez-Moliner B.J., Genovese S., Gola L. and Prigioni C. 2008. A
non-invasive genetic survey of the pine marten (Martes martes) in the western River Po plain (Italy):
preliminary results. Hystrix It. J. Mamm. (n.s.) 19 (1): 77-80.
Bang P. and Dahlström P. 1974. Animal tracks and signs. London: Collins: 1-235.
Benton T.G., Vickery J.A. and Wilson J.D. 2003. Farmland biodiversity: is habitat heterogeneity the key? Trends in
Ecology and Evolution 18 (4): 182-188.
Brainerd S.M. and Rolstad J. 2002. Habitat selection by Eurasian pine martens Martes martes in managed forests of
southern boreal Scandinavia. Wildlife Biology 8: 289-297.
Brainerd S.M., Helldin J.-O., Lindström E. and Rolstad J. 1994. Eurasian pine martens and old industrial forest in
southern boreal Scandinavia. [In: Martens, sables, and fishers: biology and conservation. S. W. Buskirk, A.
S. Harestad, M. G. Raphael, and R. A. Powell, eds]. Cornell University Press, Ithaca, New York, USA:
343–354.
Buskirk S.W. 1992. Conserving circumboreal forests for martens and fishers. Conservation Biology 6: 318-320.
Buskirk S.W. and MacDonald S.O. 1984. Seasonal food habits of marten in south central Alaska. Canadian Journal
of Zoology 62: 944-950.
Butet A., Pallat G. and Delettre Y. 2006. Seasonal changes in small mammal assemblages from field boundaries in an
agricultural landscape of western France. Agricultural Ecosystem and Environment 113: 364–369.
Canova L. and Fasola M. 1991. Communities of small mammals in six biotopes of northern Italy. Acta theriologica
36: 73-86.
Caryl F.M. 2008. Pine marten diet and habitat use within a managed coniferous forest. PhD Thesis, School of
Biological & Environmental Sciences, University of Stirling: 1-306.
Cavallini P. and Volpi T. 1996. Variation in the diet of the red fox in a Mediterranean area. Revue d’Ecologie (Terre
et Vie) 51: 173-189.
- 243 -
Paper IIIc
Appendix c
Chaline J., Baudvin H., Jammot D. and Saint Girons M.C. 1974. Les proies des rapaces. Doin Ed., Paris: 1-141.
Ciucci P., Boitani L., Raganella Pelliccioni E., Rocco M. and Guy I. 1996. A comparison of scat-analysis methods to
assess the diet of the wolf Canis lupus. Wildlife Biology 2: 37-48.
Clark-Jr D.R. and Bunck CM. 1991. Trends in North American small mammals found in common barn-owl (Tyto
alba) dietary studies. Canadian Journal of Zoology 69(12): 3093-3102.
Clevenger A.P. 1993a. The European pine marten Martes martes in the Balearic Islands, Spain. Mammal Review 23:
65-72.
Clevenger A.P. 1993b. Pine marten comparative feeding ecology in an island and mainland population of Spain.
Zeitschrift für Säugetierkunde 58: 212-224.
Davison A., Birks J.D.S., Brookes R.C., Braithwaite T.C. and Messenger J.E. 2002. On the origin of faeces:
morphological versus molecular methods for surveying rare carnivores from their scats. Journal of Zoology
of London 257: 141–143.
Day M.G. 1966. Identification of hair and feather remains in the gut and faeces of stoats and weasels. Journal of
Zoology of London 148: 201-217.
Debrot S., Fival G., Mermod C. and Weber J.M. 1982. Atlas des poils des mammifères d’Europe. Institut de
Zoologie, Université de Neuchâtel: 1-208.
Delibes M. 1983. Interspecific competition and the habitat of the stone marten Martes foina (Erxleben 1777) in
Europe. Acta Zoologica Fennica 174: 229-231.
De Marinis, A.M. and Massetti M. 1993a. Distribution of the pine marten Martes martes L., 1758 (Mammalia,
Carnivora) on the island of Elba, northern Tyrrhenian sea. Supplementi di Ricerca di Biologia della
Selvaggina 21, 263-267.
De Marinis A.M. and Massetti M. 1993b. Pine marten Martes martes on the island of Elba. Small Carnivore
Conservation 8: 13.
De Marinis A.M. and Masseti M. 1995. Feeding habits of the pine marten Martes martes L., 1758, in Europe: a
review. [In: Proceedings II Italian Symposium On Carnivores. C. Prigioni ed]. Hystrix 7: 143-150.
Firbank L.G. 2005. Striking a new balance between agricultural production and biodiversity. Annals of Applied
Biology 146: 163–175.
Fitzgibbon C.D. 1997. Small mammals in farm woodlands: the effects of habitat, isolation and surrounding land-use
patterns. Journal of Applied Ecology 34 (2): 530-539.
Genovesi P. and De Marinis A.M. 2003. Martes foina – Distribuzione geografica. [In: Fauna d’Italia. Mammalia III,
Carnivora – Artiodactyla. L.Boitani, S.Lovari, A.Vigna Taglianti eds]. Calderini: 113-123.
Giordano M. and Meriggi A. 2009. Use by small mammals of short-rotation plantations in relation to their structure
and isolation. Hystrix Italian Journal of Mammalogy (n.s.) 20 (2): 127-135.
Gurnell J., Venning T., MacCaskill B. and MacCaskill D. 1994. The food of pine martens in west Scotland. Journal
of Zoology of London 234: 680-683.
Harrington L.A., Harrington A.L., Hughes J., Stirling D. and McDonald D.W. 2010. The accuracy of scat
identification in distribution surveys: American mink, Neovison vison, in the northern highlands of
Scotland. European Journal of Wildlife Research 56: 377-384.
Helldin J.O. 2000. Seasonal diet of pine marten Martes martes in southern boreal Sweden. Acta Theriologica 45:
409-420.
Jędrzejewska B. and Jędrzejewski W. 1998. Predation in vertebrate communities: the Białowieza Primeval Forest as a
case study. Ecological Studies vol. 135. Springer-Verlag, Berlin, Heidelberg, and New York, 450 pp.
Jędrzejewski W., Jędrzejewska B. and Szymura L. 1995. Weasel population response, home range, and predation on
rodents in a deciduous forest in Poland. Ecology 76: 179-195.
Jędrzejewski W., Zalewski A. and Jędrzejewska B. 1993. Foraging by pine marten Martes martes in relation to food
resources in Bialowieża National Park, Poland. Acta Theriologica 38: 405-426.
Kelly B.T. 1991. Carnivore scat analysis: an evaluation of existing techniques and the development of predictive
models of prey consumed. M.Sc. Thesis, University of Idaho, Moscow: 1-200.
King C. M. 1989. The natural history of weasels and stoats. Christopher Helm: 1-253.
Kruuk H. and Parish T. 1981. Feeding specialization of the European badger (Meles meles) in Scotland. Journal of
Animal Ecology 50: 773-788.
Kurki S., Nikula A., Helle P. and Lindén H. 1998. Abundances of red fox and pine marten in relation to the
composition of boreal forest landscapes. Journal of Animal Ecology 67: 876– 886.
- 244 -
Paper IIIc
Appendix c
Levins R. 1968. Evolution in changing environments. Princeton University Press, Princeton: 1-120.
Libois R.M. 1984. Essai sinécologique sur le micromammifères d’Europe atlantique et ouest méditerranéenne. Etude
par analyse du régime alimentaire de la Chouette effraie, Tyto alba (Scopoli). Cahiers d’Ethologie Appliquée
4: 1-202.
Lucentini L., Vercillo F., Palomba A.,Panara F. and Ragni B. 2007. A PCR-RFLP method on faecal samples to
distinguish Martes martes, Martes foina, Mustela putorius and Vulpes vulpes. Conservation Genetics 8: 757–
759.
Lynch A.B. and McCann Y. 2007. The diet of the pine marten (Martes martes) in Killarney National Park. Biology
& Environment: Proceedings of the Royal Irish Academy 107 (2): 67-76.
Lockie J.D. 1961. The food of the Pine marten Martes martes in West Ross-shire, Scotland. Proceedings of the
Zoological Society of London 136: 187-195.
Love R.A., Webbon C., Glue D.E. and Harris S. 2000. Changes in the food of British barn owls (Tyto alba) between
1974 and 1997. Mammal Review 30 (2): 107-129.
Marchesi P. and Mermod C. 1989. Regime alimentaire de la marte (Martes martes) dans de Jura Suisse (Mammalia:
Mustelidae). Revue Suisse de Zoologie 96: 127-146.
Michel N., Burel F. and Butet A. 2006. How does landscape use influence small mammal diversity, abundance and
biomass in hedgerow networks of farming landscapes? Acta Oecologica 30 : 11-20.
Miklós P. and Žiak D. 2002. Microhabitat selection by three small mammal species in oak-elm forest. Folia
Zoologica 51: 275-288.
Millán De La Peña N., Butet A., Delettre Y., Paillat G., Morant P. and Burel F. 2003. Response of the small
mammal community to changes in western French agricultural landscapes. Landscape Ecology 18: 265–
278.
Murgia C., Secci E. and Deiana A.M. 1995. Preliminary research on some ecological and biometric aspects of the
Sardinian pine marten (Martes martes). [In: Proceedings II Italian Symposium On Carnivores. C.Prigioni
ed]. Hystrix 7: 151-154.
O’Mahony D., O’Reilly C. and Turner P. 2006. National Pine Marten Survey of Ireland 2005. Coford Connects,
Environment No 7, pp. 1–8.
Ouin A., Paillat G., Butet A. and Burel F. 2000. Spatial dynamics of wood mouse (Apodemus sylvaticus) in an
agricultural landscape under intensive use in the Mont Saint Michel Bay (France). Agricultural Ecosystem
and Environment 78: 159-165.
Pereboom V., Mergey M., Villerette N., Helder R., Gerard J. and Lóde T. 2008. Movement patterns, habitat
selection and corridor use of a typical woodland-dweller species, the European pine marten (Martes martes),
in fragmented landscape. Canadian Journal of Zoology 86: 983-991.
Pilot M., Gralak B., Goszczyński J. and Posłuszny M. 2006. A method of genetic identification of pine marten
(Martes martes) and stone marten (Martes foina) and its application to faecal samples. Journal of Zoology
(London) 271: 140-147.
Pittiglio C. 1996. Analisi comparativa di uso e selezione del habitat della faina e della martora in condizioni di
simpatria. Degree thesis, Università di Roma ‘‘La Sapienza’’, Roma.
Posluszny M., Pilot M., Goszczynski J. and Gralak B. 2007. Diet of sympatric pine marten (Martes martes) and stone
marten (Martes foina) identified by genotyping of DNA from faeces. Annales Zoologici Fennici 44: 269284.
Proulx G., Aubry K., Birks J., Buskirk S., Fortin C., Frost H., Krohn W., Mayo L., Monakhov V., Payer D., Saeki
M., Santos-Reis M., Weir R. and Zielinski W. 2004. World distribution and status of the genus Martes in
2000. [In: Martens and Fishers (Martes) in Human-altered Environments. D.J. Harrison, A.K. Fuller, G.
Proulx eds]. Springer, London, UK: 21-76.
Pulliainen E. and Ollinmäki P. 1996. A long-term study of the winter food niche of the pine marten Martes martes
in northern boreal Finland. Acta Theriologica 41 (4): 337-352.
Putman R.J. 2000. Diet of pine martens Martes martes L. in west Scotland. Journal of Natural History 34: 793-797.
Reed J.E., Baker R.J., Ballard W.B. and Kelly B.T. 2004. Differentiating Mexican grey wolf and coyote scats using
DNA analysis. Wildlife Society Bulletin 32: 685-692.
Reig S. and Jedrzejewski W. 1988. Winter and early spring food of some carnivores in the Bialowieza National Park,
eastern Poland. Acta Theriologica 33: 57-65.
Rice W.R. 1989. Analysing tables of statistical tests. Evolution 43: 223-225.
- 245 -
Paper IIIc
Appendix c
Rosell F. and Hovde B. 1998. Pine marten, Martes martes, as a Eurasian beaver, Castor fiber, lodge occupant and
possible predator. The Canadian Field Naturalist 112: 535-536.
Rosellini S., Osorio E., Ruiz-González A., Pineiro A. and Barja I. 2008. Monitoring the small-scale distribution of
sympatric European pine martens (Martes martes) and stone martens (Martes foina): a multievidence
approach using faecal DNA analysis and camera-traps. Wildlife Research 35: 434–440.
Ruiz-González A., Rubines J., Berdión O. and Gómez-Moliner B. J. 2008. A non-invasive genetic method to
identify the sympatric mustelids pine marten (Martes martes) and stone marten (Martes foina): preliminary
distribution survey on the northern Iberian Peninsula. European Journal of Wildlife Research 54 (2): 253261.
Ruiz-Olmo J. and Nadal J. 1991. Régime alimentaire de la martre (Martes martes L., 1758) en hiver et taille des
portèes à Ménorca, Iles Baléares. Mammalia 55 : 639-642.
Storch I., Lindstrom E. and De Jounge J., 1990. Diet and habitat selection in pine martens in relation to
competition with the red fox. Acta Theriologica 35: 311-320.
Stubbe M. 1993. Martes martes (Linné, 1758) – Baum., Edelmarten. [In: Handbuch der Säugetiere Europas. Band 5:
Raubsäuger - Carnivora (Fissipedia). Teil I: Canidae, Ursidae, Procyoinidae, Mustelidae. J. Niethammer, F.
Krapp eds]. Weisbaden, Aula Verlag: 374–426.
Swihart R.K. and Yarner R.H. 1982. Habitat features influencing use of farmstead shelterbelts by the Eastern
cottontail (Sylvilagus floridanus). American Midland Naturalist 107: 411-416.
Taberlet P. 1986. Etude de l’écologie des micromammifères a partir des pelotes de réjection de Tyto alba (Scopoli,
1769). Application au Bas-Chablais (Haute-Savoie, France). Revue d’Ecologie (La Terre et la Vie) 41: 193217.
Taberlet P. and Fumagalli L. 1995. Owl pellets as a source of DNA for genetic studies of small mammals. Molecular
Ecology 5: 301-305.
Teerink B.J. 1991. Hair of West-European mammals. Atlas and identification key. Cambridge, Cambridge
University Press: 1-232.
Vidus Rosin A., Gilio N. and Meriggi A. 2008. Introduced Lagomorphs as a threat to “native” Lagomorphs: The
case of the Eastern cottontail (Sylvilagus floridanus) in northern Italy. [In: Lagomorph Biology, Evolution,
Ecology and Conservation P.C. Alves, N. Ferrand, K. Hacklander eds]. Springer, Eidenberg: 153-165.
Zalewski A. 2004. Geographical and seasonal variation in food habits and prey size of European pine martens. Pages
77-98 [In: Martens and Fishers (Martes) in Human-altered Environments. D.J. Harrison, A.K. Fuller, G.
Proulx eds]. Springer, London, UK: 77-98.
Zalewski A. and Jedrzejewski W. 2006. Spatial organisation and dynamics of the pine marten Martes martes
population in Bialowieza Forest (E Poland) compared with other European woodlands. Ecography 29: 3143.
Zielinski W.J. 1986. Relating marten scat contents to prey consumed. California Fish and Game 72: 110–116.
Zielinski W.J. and Duncan N.P. 2004. Diets of sympatric populations of American martens (Martes americana) and
fishers (Martes pennanti) in California. Journal of Mammalogy 85(3): 470–477.
- 246 -

Similar documents

×

Report this document