Strain, clone and species - Journal of Medical Microbiology

Document technical information

Format pdf
Size 82.6 kB
First found May 22, 2018

Document content analysis

Category Also themed
not defined
no text concepts found


Mike Tyson
Mike Tyson

wikipedia, lookup




J. Med. Microbiol. Ð Vol. 49 (2000), 397±401
# 2000 The Pathological Society of Great Britain and Ireland
ISSN 0022-2615
Strain, clone and species: comments on three basic
concepts of bacteriology
Department of Infectious Diseases, Leiden University Medical Centre, PO Box 9600 2300 RC Leiden, The
Netherlands, Department of Biochemistry, University of Groningen, Nijenborgh 4, 9747 AG Groningen, The
Netherlands and {Department of Medical Microbiology, MalmoÈ University Hospital, Lund University, S-205 02
MalmoÈ, Sweden
Different aspects of the terms strain, clone and species are discussed. The term strain is
commonly used to denote a pure culture ± here called `the strain in the taxonomic
sense' ± but does also refer to a natural concept closely related to the clone. The term
clone on the other hand is used both in a general and in a more restricted sense, the
latter indicating a low degree of genetic exchange. The important distinction between the
de®nition of a species and the criteria for a species is emphasised and the main kinds of
criteria are considered.
Some of the bacteriological terms in daily use are
actually far from well de®ned. This is particularly the
case with laboratory jargon, but in scienti®c papers
bacteriological terms may also be used with different
meanings, depending on the context and the subjective
preferences of the author. The reason for this may be
that unambiguous de®nitions are dif®cult to formulate,
but in some cases there is disagreement about the
proper use of a term. This review attempts to analyse
the concepts behind the terms strain, clone and species.
Although the examples are taken from clinical
bacteriology, the discussion will also be relevant to
other ®elds of bacteriology.
The strain
According to the ®rst edition of Bergey's Manual of
Systematic Bacteriology `A strain is made up of the
descendants of a single isolation in pure culture and
usually is made up of a succession of cultures
ultimately derived from an initial single colony' [1].
This is a rather standard de®nition of the strain as the
basic operational unit in bacteriology and it will be
referred to below as the strain in the taxonomic sense.
The de®nition has several interesting implications.
Received 23 Feb. 1999; revised version accepted 1 Oct.
Corresponding author: Dr J. B. Ursing (e-mail: [email protected]
Present address: Torparebron, S-277 55 BroÈsarp, Sweden.
Perhaps the most important is that the strain in this
sense is not a natural concept, as the selection of the
`initial single colony' is made by decision and its
descendants are kept in arti®cial culture. Even if a
natural environment such as a sterile body site in a
laboratory animal is used for cultivation, the inoculation and recovery of a strain are controlled procedures.
A merit of the de®nition is that there can be no doubt
as to the identity of the strain, provided that it has been
correctly labelled and protected from contamination.
The starting point of the strain is another important
issue. `A single isolation in pure culture' indicates that
the strain has been isolated from a particular site at a
particular time. This con¯icts with another of the
meanings of the term strain in clinical microbiology.
Meningococci isolated at the same time from the
nasopharynx, blood and cerebrospinal ¯uid of one and
the same patient are almost certainly derived `from an
initial single colony'. The same applies to repeated
isolates of a particular streptococcus from the blood of
a patient with infective endocarditis. In both these
cases the starting point in space and time and also the
further development of bacterial spread and growth are
Thus, it seems necessary to assume that there is a
counterpart in nature to the strain in the taxonomic
sense. We will simply refer to it here as the strain in
nature. The relationship between the two concepts
becomes evident if we say that a strain in the
taxonomic sense is a sample from a strain in nature.
However, if we try to formulate a de®nition of the
Downloaded from by
On: Wed, 03 May 2017 20:15:45
strain in nature analogous to that of the strain in the
taxonomic sense, we run into the dif®culties which are
always encountered when we look for boundaries in
The strain in nature seems to have been in the minds of
Tenover et al. when they de®ned a strain as `. . . an
isolate or group of isolates that can be distinguished
from other isolates of the same genus and species by
phenotypic characteristics or genotypic characteristics
or both' [2]. This de®nition was, in all essentials,
adopted by a European study group on epidemiological
markers [3]. Neither of these groups concerned
themselves with what we have called here the strain
in the taxonomic sense.
A natural strain is rarely `pure'. Rather, it has to
compete with a number of other strains for its
existence. There are of course exceptions, such as the
clinical examples given above which concern normally
sterile sites of the body.
Bacterial strains ± both in the taxonomic sense and in
nature ± change over time. They undergo mutations
and they may lose plasmids. Strains in the taxonomic
sense retain the identity given to them even if the
phenotype is changed, which happens occasionally to
old strains. The strain in nature may also acquire
genetic material from other strains in the environment.
The decision as to whether two samples represent the
same strain in nature becomes merely a matter of
opinion. In the two clinical examples above, the
common origin of the isolates is inferred from
similarity and clinical data. We will return to the
problem when we have discussed the term clone.
The clone
This term denotes the progeny of one individual
through asexual reproduction [4] and was originally
used in botany [5]. It seems to have been introduced
rather late in bacteriology; the progenitor here is a
bacterial cell. As bacteria reproduce by ®ssion and lack
meiosis they are by de®nition clonal in this original
sense. In evolutionary terms, the clone is assumed to be
monophyletic, which means not only that all cells have
the same ancestor (opposite: polyphyletic), but also that
the clone includes all descendants of the progenitor
(opposite: paraphyletic) (Fig. 1). The latter prerequisite
is merely of theoretical interest, as in bacteriological
practice we are always dealing with samples of the
clone. And the samples ± again ± are strains in the
taxonomic sense.
The term clone has become useful in epidemiology,
particularly in the study of the relationships between
isolates representing widely separate geographical
areas. A good working de®nition was offered by
érskov and érskov: `. . . the word clone will be used
Fig. 1. A rooted evolutionary tree with the terminal
groups a, b, c and d. The unbroken line indicates
monophyly (groups a and b share an ancestor and they
are the only descendants of this ancestor). The outer
broken line indicates paraphyly (groups a, b, c and d
share an ancestor but a is excluded). The inner dashed
line indicates polyphyly (b and c do not share an
ancestor, being neither monophyletic nor paraphyletic).
to denote bacterial cultures isolated independently from
different sources, in different locations, and perhaps at
different times, but showing so many identical
phenotypic and genotypic traits that the most likely
explanation for this identity is a common origin' [6].
Tenover et al. [2] are less speci®c: `Genetically related
isolates (clones) are isolates that are indistinguishable
from each other by a variety of genetic tests . . . or that
are so similar that they are presumed to be derived
from a common parent.'
Most of what was said above about the strain in nature
is also valid for the clone. The similarity between the
two concepts is also evident from the de®nitions cited
above. However, for similar isolates recovered over
wide geographical areas, clone rather than strain is
used. Even if the clone is basically a natural concept,
the strain in the taxonomic sense may be regarded as
an arti®cial clone.
The clonal relationship may be obscured by horizontal
transfer of genetic material. This has been studied by
methods of population biology, notably multilocus
enzyme electrophoresis [7] and lately by multilocus
sequence typing [8]. Diversi®cation of clones has been
revealed and clonality has then been regarded as a
relative property of a bacterial species implying a low
frequency of horizontal transfer [9]. The opposite
behaviour is panmictic, a term also borrowed from
eukaryote biology, which denotes free interbreeding
[4]. Intermediate forms also occur. The distinction
between clonal ± in this restricted sense ± and
panmictic behaviour is useful, as an epidemiological
relationship may not be evident if horizontal gene
transfer has signi®cantly changed the genotype and
phenotype of a clone.
Thus, no rules can be formulated for the differences
Downloaded from by
On: Wed, 03 May 2017 20:15:45
allowed between samples of a clone or the strain in
nature. For each situation, several parameters have to
be considered: the discriminatory power of the typing
systems, connections in space±time, and degree of
`clonality' ± if known. The latter may differ considerably even between closely related species [9].
The species
The species is the basic category in biological
classi®cation. While the de®nition of the eukaryotic
species has focused on the biology of reproduction, the
bacterial species has mainly been discussed in terms of
similarity between strains (from now on: in the
taxonomic sense if not speci®ed).
Despite the many different opinions that have been
expressed about the bacterial species, the general
concept of species held by most bacteriologists could
probably be formulated as follows: A species consists
of strains of common origin which are more similar to
each other than they are to any other strain. The
difference between the two prerequisites of this
de®nition is smaller than it appears to be at ®rst sight,
as common origin in bacteria is largely inferred ± at
least at present ± by similarity of semantides,
information-rich macromolecules like proteins and
nucleic acids. It should further be noted that this is a
working de®nition of an arti®cial entity. It may be
monophyletic or paraphyletic but the de®nition excludes polyphyly.
If the maximum possible information is used for the
comparison of strains, this species concept will imply a
maximum achievable predictivity. From a philosophical
point of view, this principle has been regarded as the
signum of a natural or multi-purpose classi®cation;
natural referring here to what may be considered
natural to the human mind [10]. Overall similarity is of
course the basis of numerical taxonomy [11], and also
of the `polyphasic' approach, where the aim is to use
as many phenotypic and genotypic properties as
possible [12].
Although the species de®nition ± what we mean by a
species or want it to be ± may not be a major problem,
dif®culties arrive when we have to delineate species;
we then have to look for species criteria ± features that
can be used for differentiation. The species criteria
become important when it has to be decided whether or
not a bacterial group should be given a speci®c epithet,
i.e., should formally be regarded as species.
Given consensus about the de®nition, the criteria may
be regarded as a theory that has to be tested against the
de®nition. In our de®nition, it is taken for granted that
in the bacterial world there are discontinuities ± more
or less clear-cut. In other words, we suppose that there
exists in nature a counterpart to the species that we
create and name for our convenience. If not, boundaries
would be arbitrary, like the wavelengths chosen to
de®ne a speci®c colour, e.g., bluish green. The species
in nature would consist of a ®nite number of clones
and ideally would be monophyletic.
The theoretical approach to the criterion problem has
developed in an interesting way over time. Up to the
middle of this century bacterial taxa were mainly
differentiated by morphology and selected sets of
biochemical tests. A species should be characterised
by features invariably present, a view rooted in Plato's
world of ideas and put into practice in the Aristotelian
logic. The term monothetic was coined by Sneath for
groups created in this way [13]. As the species criteria
were selected by decision, identi®cation became easy
and reliable. A recent example is the Kauffmann-White
scheme for classi®cation of salmonellae, based on the
assumption that the surface antigens could be used as
species criteria [14].
The monothetic approach was succeeded by the
polythetic, used in numerical taxonomy where members
of a group have a maximum of properties in common
[13]. However, it is inherent in the polythetic view that
no single property is essential for membership of the
group. There may be situations where it is dif®cult to
®nd useful differential characteristics and identi®cation
then becomes a matter of probability.
The progress of molecular biology has changed the
situation in a profound way. New kinds of bacterial
data have become available ± cell wall composition,
whole cell protein analysis, whole cell fatty acid
analysis and gene sequences, to mention a few. The
ability to study and compare genomes has promised an
insight into evolutionary relationships and a classi®cation based on natural af®nity ± natural in a different
sense from above. This has led to the view that
genotypic data are the most important ± clearly a
retreat from the polythetic approach. A side-effect of
this development could be added; taxonomic work has
been increasingly undertaken by biochemists rather
than microbiologists ± chemical compounds are easier
to classify than organisms.
The species criterion dominating the last 30 years has
been provided by DNA±DNA pairing data. Studies of
enterobacteria had shown the existence of discontinuities between groups of data, relatedness values in
the range 50±70% being comparatively rare. Moreover,
DNA from strains of existing species were >70%
related [15]. (It should be noted that current methods
do not record re-association between DNA strands
unless they are c. 80% similar, meaning that the range
of 80±100% similarity corresponds to the range of 0±
100% re-association. About 90% similar DNAs show
only c. 50% re-association.)
A recommendation that implied that strains with a
Downloaded from by
On: Wed, 03 May 2017 20:15:45
DNA±DNA pairing value ,70% are members of
different species [16] triggered an increasing number
of proposals of new species. Recently, a comparison of
DNA±DNA pairing data and 16S rRNA similarity data
showed that strains with rRNA similarity less than c.
97% generally showed no signi®cant DNA±DNA reassociation and thus belong to different species [17].
Similarity .97% may or may not indicate close
relationship. The use of this percentage as a rule of
thumb has in many cases made rRNA sequencing
replace the more cumbersome DNA±DNA pairing
technique for the creation of new species. At present,
either the 70% or the 97% rule is used to underpin
most proposals for new species.
DNA±DNA pairing values as species criteria have the
merit of the monothetic approach: they are easily
applied and identi®cation is reliable ± although not
easily accomplished, as few routine laboratories have
access to the technique. However, when a species has
been described, species-speci®c methods based on
genomic data may present alternative identi®cation
methods. A major shortcoming is that the recommended values were a generalisation of results for a
limited spectrum of organisms; they make no provision
for bacterial groups related at different DNA±DNA
pairing values [18]. Groups that are phenotypically and
ecologically different may not merit the status of
different species by the DNA±DNA pairing criteria;
there are also situations in which genomically distinct
groups cannot be differentiated phenotypically [19].
This means that DNA±DNA pairing has failed to
provide generally applicable criteria for the bacterial
species as de®ned above. The recommendations cited
[16] include the statement that a DNA±DNA pairing
group should not be named unless it can be
`differentiated by some phenotypic property'. The use
of the singular form implies a rather modest contribution to a polyphasic approach [16]. Indeed, adherence
to the 70% and 97% rules with few or no phenotypical
or ecological requirements means that the species
de®nition above is narrowed to suit only those
interested in phylogenetic relationships.
In the early stages of our knowledge of horizontal
genetic transfer, there was a hope that capability of
gene exchange could be used as a species criterion
[20]. The idea that the bacterial species ± like the
species of higher organisms ± could be de®ned as a
gene pool has recently been revived [21]. Proteincoding genes may better recognise ecological populations than DNA±DNA pairing data, which should make
them better suited as species criteria [22]. Much work
lies ahead in this area.
Concluding remarks
We have been looking into the meaning of words. The
question of true or false is of course not relevant here.
To cite Humpty Dumpty: `When I use a word, . . . it
means just what I choose it to mean ± neither more nor
less' [23]. A de®nition cannot be wrong but it can be
unsuitable if it is contrary to general usage or
The de®nition of strain in the taxonomic sense is
probably endorsed by most bacteriologists. In any
event, they have to accept it as inherent in the type
strain concept, which is the backbone of current
nomenclature praxis [24]. Nevertheless, we also have
to live with the term strain as it refers to the strain in
nature. Of course, clone could be used here, although
this could be confusing ± e.g., if it referred to
gonococci isolated from the blood and from the genital
region of one and the same person, gonococci being
the standard example of a bacterium with a `nonclonal' population structure [9].
For groups of bacteria capable of horizontal genetic
transfer, Ravin coined the term genospecies [20]. He
also suggested taxospecies for a phenotypically circumscribed group and nomenspecies for any named
species. Genospecies has since been used in different
senses, among others to denote a DNA±DNA pairing
group. For this purpose, the terms genomic species and
genomospecies are also in use. This slum of species
has given rise to much confusion and would be better
discarded. There should be no room for more than one
kind of species in one and the same classi®cation [19].
The revived monothetic approach to species criteria,
leaving phenotypic properties behind, is a source of
frustration for bacteriologists in the ®eld [25]. Most of
them are interested in bacteria as they are now and do
not care much about how they evolved. Their feeling
that they have been abandoned by taxonomists is
aggravated by the growing number of proposals of new
species based on a single strain [26]. Knowledge of
intraspecies variation is necessary for all identi®cation
work. Nevertheless, there is hope for the future. There
is growing support for the polyphasic species concept
[27, 28]. It is to be hoped that the prevailing, rather
rigid and dogmatic approach to bacterial systematics
will be replaced by a more pragmatic view.
1. Staley JT, Krieg NR. Classi®cation of procaryote organisms: an
overview. In: Krieg NR, Holt JG (eds) Bergey's Manual of
systematic bacteriology, vol 1. Baltimore, Williams and
Wilkins. 1984: 1±4.
2. Tenover FC, Arbeit RD, Goering RV et al. Interpreting
chromosomal DNA restriction patterns produced by pulsed®eld gel electrophoresis criteria for bacterial strain typing.
J Clin Microbiol 1995; 33: 2233±2239.
3. Struelens MJ and the members of the European study group of
epidemiological markers (ESGEM) of the European Society for
Clinical Microbiology and Infectious Diseases (ESCMID).
Consensus guidelines for appropriate use and evaluation of
microbial epidemiologic typing systems. Clin Microbiol Infect
1996; 22: 2±11.
Downloaded from by
On: Wed, 03 May 2017 20:15:45
4. Rieger R, Michaelis A, Green MM. Glossary of Genetics, 5th
edn. Berlin, Springer-Verlag. 1991: 95, 368.
5. Webber HJ. New horticultural and agricultural terms. Science
1903; 18: 501±503.
6. érskov F, érskov I. Summary of a workshop on the clone
concept in the epidemiology, taxonomy, and evolution of the
enterobacteriaceae and other bacteria. J Infect Dis 1983; 148:
7. Selander RK, Musser JM. Population genetics of bacterial
pathogenesis. In: Iglewski BH, Clark VL (eds) Molecular basis
of bacterial pathogenesis. (The Bacteria, vol II.) San Diego,
Academic Press. 1990: 11±36.
8. Maiden MCJ, Bygraves JA, Feil E et al. Multilocus sequence
typing: a portable approach to the identi®cation of clones
within populations of pathogenic microorganisms. Proc Natl
Acad Sci USA 1998; 95: 3140±3145.
9. Maynard Smith J, Smith NH, O'Rourke M, Spratt BG. How clonal
are bacteria? Proc Natl Acad Sci USA 1993; 90: 4384±4388.
10. Gilmour JSL. A taxonomic problem. Nature 1937; 139: 1040±
11. Sneath PHA, Sokal RR. Numerical taxonomy. San Francisco,
WH Freeman. 1973: 1±15.
12. Colwell RR. Polyphasic taxonomy of bacteria. In: Iizuka H,
Hasegawa T (eds) Culture collections of microorganisms.
Proceedings of the International Conference on Culture
Collections, Tokyo, Oct. 7±11, 1968. Baltimore, MD, University Park Press. 1970: 421±436.
13. Sneath PHA. The construction of taxonomic groups. In:
Ainsworth GC, Sneath PHA (eds) Microbiological classi®cation. 12th Symposium of the Society for General Microbiology.
Cambridge, Cambridge University Press. 1962: 289±332.
14. Kauffmann F. Serological diagnosis of salmonella-species. Kauffmann-White-schema. Copenhagen, Munksgaard. 1972: 12±15.
15. Brenner DJ. Deoxyribonucleic acid reassociation in the
taxonomy of enteric bacteria. Int J Syst Bacteriol 1973; 23:
16. Wayne LG, Brenner DJ, Colwell RR et al. Report of the ad
hoc committee on reconciliation of approaches to bacterial
systematics. Int J Syst Bacteriol 1987; 37: 463±464.
17. Stackebrandt E, Goebel BM. Taxonomic note: a place for
DNA±DNA reassociation and 16S rRNA sequence analysis in
the present species de®nition in bacteriology. Int J Syst
Bacteriol 1994; 44: 846±849.
18. Sneath PHA. Analysis and interpretation of sequence data for
bacterial systematics: the view of a numerical taxonomist. Syst
Appl Microbiol 1989; 12: 15±31.
19. Ursing JB, RosselloÂ-Mora RA, Garcia-Valdes E, Lalucat J.
Taxonomic note. A pragmatic approach to the nomenclature of
phenotypically similar genomic groups. Int J Syst Bacteriol
1995; 45: 604.
20. Ravin AW. Experimental approaches to the study of bacterial
phylogeny. Am Nat 1963; 97: 307±318.
21. Dykhuizen DE, Green L. Recombination in Escherichia coli
and the de®nition of biological species. J Bacteriol 1991; 173:
22. Palys T, Nakamura LK, Cohan FM. Discovery and classi®cation of ecological diversity in the bacterial world. The role of
DNA sequence data. Int J Syst Bacteriol 1997; 47: 1145±1156.
23. Carroll L. Alice's adventures in wonderland=Through the
looking glass. In: The annotated Alice. Gardner M (ed) New
York, Bramhall House. 1960: 269.
24. Sneath PHA (ed). International code of nomenclature of
bacteria, 1990 revision. Washington, DC, American Society
of Microbiology. 1992: 17±20.
25. Magee JT. Forsaking the tome ± a worms' eye view of
taxonomy. J Med Microbiol 1993; 39: 401±402.
26. Frederiksen W, Magee JT, Ursing J. Proposed new bacterial
taxa and proposed changes of bacterial names published during
1997 and considered to be of interest to medical or veterinary
bacteriology. J Med Microbiol 1999; 48: 113±116.
27. Vandamme P, Pot B, Gillis M, de Vos P, Kersters K, Swings J.
Polyphasic taxonomy, a consensus approach to bacterial
classi®cation. Microbiol Rev 1996; 60: 407±438.
28. Goodfellow M, Man®o GP, Chun J. Towards a practical species
concept for cultivable bacteria. In: Claridge MF, Dawah HA,
Wilson MR (eds) Species, the units of biodiversity. (The
Systematics Association Special Volume Series 54.) London,
Chapman and Hall. 1997: 25±59.
Downloaded from by
On: Wed, 03 May 2017 20:15:45

Similar documents


Report this document