Title: The role of prostaglandin E2 in acute acetaminophen hepatotoxicity in mice
Original scientific paper
Running title: PGE2 in Acute APAP Hepatotoxicity
Names of authors in order of appearance: Ivan Ćavar1, Tomislav Kelava2, Katarina Vukojević1,3,
Mirna Babić-Saraga3 and Filip Čulo1,2
Department of Physiology, School of Medicine, University of Mostar, Bosnia and Herzegovina
Department of Physiology, School of Medicine, University of Zagreb, Croatia
Department of Histology and Embryology, School of Medicine, University of Split, Croatia
Address for correspondence:
Department of Physiology,
School of Medicine, University of Mostar,
Bijeli brijeg b.b., 88000 Mostar, Bosnia and Herzegovina
e-mail: [email protected]
Prostaglandin E2 (PGE2), which is synthesized by many cell types, has a cytoprotective effect in the
gastrointestinal tract and in several other tissues and cells. On the other hand, overdose or chronic use of a
high dose of acetaminophen (Paracetamol, APAP) is a major cause of acute liver failure in the western
world. These observations prompted us to investigate whether PGE2 plays a role in host defence to toxic
effect of APAP. (CBAT6T6xC57Bl/6)F1 hybrid mice of both sexes were intoxicated with a single lethal or
high sublethal dose of APAP, which was administered to animals by oral gavage. Stabile analogue of PGE2,
16,16-dimethyl PGE2 (dmPGE2), or inhibitor of its production, CAY10526, were given intraperitoneally
(i.p.) 30 minutes before or 2 hours after APAP administration. The toxicity of APAP was determined by
observing the survival of mice during 48 hours, by measuring concentration of alanine-aminotransferase
(ALT) in plasma 20-22 hours after APAP administration and by liver histology. The results have shown that
PGE2 exhibits a strong hepatoprotective effect when it is given to mice either before or after APAP, while
CAY10526 demonstrated mainly the opposite effect. Immunohistochemical or immunofluorescent
examinations in the liver tissue generally support these findings, suggesting that PGE2 inhibited APAPinduced activation of nuclear factor kappa B (NF-κB). Similarly, PGE2 down regulated the activity of
inducible nitric oxide synthase (iNOS), which was up regulated by APAP. Thus, by these and perhaps by
other mechanisms, PGE2 contributes to the defence of the organism to noxious effects of xenobiotics on the
Key words: prostaglandin E2, acetaminophen, liver injury, NF-κB, immunohistochemistry
Acetaminophen (Paracetamol, N-acetyl-p-aminophenol, APAP), the most commonly used analgesic and
antipyretic drug, is very safe at therapeutic doses. However, overdose or chronic use of a high dose of APAP
has been shown as a major cause of acute liver failure in the western world (reviewed in Lee, 2003; Bernal
2003). After an overdose of APAP, the reactive metabolite, N-acetyl-p-benzoquinone imine (NAPQI), is
generated by hepatic cytochromes P450 (CYPs). Although, the precise biochemical mechanism of cell death
is not fully understood it is generally recognized that NAPQI triggers hepatic toxicity via mitochondria
injury, apoptosis and cell necrosis. These effects of NAPQI are mediated through its covalently binding with
nucleophilic macromolecules such as DNA or proteins, glutathione (GSH) depletion, lipid peroxidation and
generation of oxidative stress molecules (Jaeschke, 2000; James et al., 2003; Mazer and Perrone, 2008). It is
believed that inflammatory mechanisms play a role in the later phases of liver APAP toxicity (O'Connor and
Bennet, 2005). Thus, it was shown that nuclear factor kappa B (NF-B) is also activated during APAPinduced hepatotoxicity (Bauer et al., 2000; Dambach et al., 2006). Oxidative stress has been shown to
activate signal transduction pathways involving transcription factors such as NF-B and activator protein-1
(AP-1) (Jaeschke, 2000; Janssen-Heininger et al., 2000). NF-B is known to regulate the expression of
genes controlling inflammatory mediators, including inducible nitric oxide synthase (iNOS) and consequent
production of nitric oxide (NO), which is a highly reactive oxidant produced in the liver in response to
different inflammatory stimuli and has been implicated in hepatotoxicity (Gardner et al., 2002; James et al.,
2003; Kamanaka et al., 2003).
Prostaglandins (PGs) are lipid-derived autacoids generated by sequential metabolism of arachidonic acid by
the cyclooxygenase (COX) and prostaglandin synthase enzymes. PGs are ubiquitously produced and have
been implicated in a broad array of diseases, including cancer, inflammation, cardiovascular disease and
hypertension (reviewed in Hata and Breyer, 2004; reviewed in Matsuoka and Narumiya, 2007).
Prostaglandin E2 (PGE2) is produced by many cells of the body and exerts its actions by binding to one (or a
combination) of its four subtypes of receptor, EP1, EP2, EP3 and EP4 (Harris et al., 2002). PGE2 is
considered to be very important for normal physiological functions, especially for the function of
gastrointestinal tract (Dey et al., 2006) and kidney (Hao and Breyer, 2008). Thus, in the stomach of rat PGE2
was essential for conferring protection against indomethacin or ethanol-induced injury (Araki et al., 2000;
Suzuki et al., 2001). Protective effects of exogenous PGE2 and prostacyclin (PGI2) has been demonstrated in
various models of liver injury induced with lipopolysaccharide (LPS), D-galactosamine (DGalN),
concanavalin A (Con A), carbon tetrachloride (CCl4), virus and ischemia-reperfusion (reviewed in Quiroga
and Prieto, 1993; Yin et al., 2007). The protective effect of PGE2 was also observed, but not directly proven,
in APAP-induced liver injury in mice (Renić et al., 1995; Reilly et al., 2001). Various mechanisms for
cytoprotective effects of PGE2 are proposed (reviewed in Quiroga and Prieto, 1993). Data in vitro have
revealed plenty of cell signaling pathways by which this protection could be mediated. Most often
investigated is the effect of PGE2 on activation of NF-κB, by which it may have an influence on the
synthesis of inflammatory hepatotoxic cytokines and generation of oxidative radicals (Tran-Thi et al., 1995;
Laskin et al., 2001; Dambach et al., 2006; Ogawa et al., 2009), activation of iNOS (Laskin et al., 2001;
James et al., 2003), stimulation of synthesis of inhibitory (suppressive) cytokine – IL-10 (Cheon et al.,
2006), endogenous antioxidants (Enomoto et al., 2001) and cyclic adenosine monophosphate (cAMP),
which has various immunosuppressive and anti-inflammatory effects (Bourne et al., 1974; Nakano et al.,
1994). However, some of these mechanisms are far from being definitively proven, since their role in
toxicity of APAP is not firmly established and the obtained effects of PGE2 were sometimes contradictory.
This particularly relates to effects on activation of NF-κB and iNOS.
Based on these data, the present studies aimed to investigate the role of exogenously applied PGE2 and its
derivatives on APAP-induced hepatotoxicity in vivo, as well as to examine their effects on expression of
NF-κB and iNOS in liver tissue.
Materials and Methods
Male or female (CBAT6T6xC57Bl/6)F1 hybrid mice aged 12-16 weeks and weighing 20-30 g were used in
all experiments. They were raised in an animal facility unit at the Department of Physiology, School of
Medicine, University of Zagreb. The animal colony unit had regulated 12 h light/dark cycle and the
temperature and relative humidity in the animal room were 22 ± 2 °C and 50 ± 5%, respectively. The cages
were sanitized twice weekly and mice were allowed free access to tap water and standard mouse chow diet
(No. 4RF21, Diet Standard, Milano, Italy). All animal protocols were approved by the Ethics Committee of
the University of Zagreb, School of Medicine (Zagreb, Croatia).
Chemicals and treatments of animals
Pure APAP substance was a kind gift from Belupo (Koprivnica, Croatia). Phenobarbitone-sodium was
obtained from Kemika (Zagreb, Croatia). Since the PGE2 is rapidly conversed to an inactive metabolite,
13,14-dihydro-15-keto PGE2, its stable structural analog, 16,16-dimethyl PGE2 (dmPGE2), was used for in
vivo experiments. Stock solution of dmPGE2 in methyl acetate (No. 14750, Cayman Chemical, Ann Arbor,
MI, USA) was first evaporated under a gentle stream of nitrogen and the remaining substance was dissolved
(1.0 mg/mL) in phosphate buffered saline (PBS, pH=7.2). DmPGE2 was administered to animals (0.2
mg/kg, i.p.) 30 min before or 2 h after APAP. CAY10526 (an inhibitor of PGE2 production through the
selective modulation of microsomal PGE synthase-1 expression, mPGES-1) was supplied as a crystalline
solid (No. 10010088, Cayman Chemical). Since the CAY10526 is sparingly soluble in aqueous buffers, it
was first dissolved in an organic solvent, dimethyl formamide (DMF, 25 mg/mL), diluted in PBS (pH=7.2)
and finally injected (2.0 mg/kg, i.p.) into animals 30 min before or 2 h after APAP. The doses of the drugs
for application in vivo were chosen from scarce data in the literature or according to the toxicity data in our
preliminary experiments, in which the effects of the drugs on survival of mice and gross macroscopic
changes of liver and other visceral organs were observed. Animals in control groups received appropriate
vehicle. Survival of mice was followed for 48 h after APAP administration, since almost all mice either died
within this period or fully recovered thereafter. For immunohistochemical or immunofluorescent
examinations, PGE2 or vehicle was given to mice 30 min before, and CAY10526 was given 2 h after APAP
Assessment and measurement of hepatotoxicity induced with APAP
In order to induce hepatic cytochromes P450 (CYPs), mice were given phenobarbitone-sodium in drinking
water for 7 days (0.3 g/L). Thereafter, mice were fasted overnight and APAP was given by oral gavage in a
volume of 0.4 to 0.6 mL. APAP was dissolved under mild magnetic stirring in warm PBS, to which 1-2
drops of Tween 20 were added. Animals were allowed free access to food 4 h later (Guarner et al., 1988;
Renić et al., 1995). To observe the survival of the mice, APAP was administered in a dose of 250-300
mg/kg, which in our previous experiments induced 43 to 72% mortality of untreated animals. To determine
plasma alanine aminotransferase (ALT) and NO concentration in plasma, as well as for histopathological,
immunohistochemical or immunofluorescent evaluation of liver slices and measurement of PGE2 production
by liver fragments, mice were treated with high sublethal dose of APAP (150 mg/kg). Experimental and
control groups of mice contained 12-13 animals (for observation of survival) or 6-10 animals (for all other
Plasma ALT activity. ALT levels were determined 20-22 h after APAP administration. Mice were given 250
U heparin i.p. 15 min before bleeding and blood was collected by puncture of the medial eye angle with
heparinized glass capillary tubes. After centrifugation, separated plasma was stored at -80 °C for 24 h before
ALT determination. ALT concentrations were measured by standard laboratory techniques (Renić et al.,
Liver histology. Mice were sacrificed under light ether anesthesia by cervical dislocation 20-22 h after
APAP administration. Liver lobes of each animal (9-10 animals per group) were fixed in 4% buffered
paraformaldehyde, dehydrated in increasing concentrations of ethanol and embedded in paraffin. Thereafter,
sections of tissue were cut at 5 mm on a rotary microtome, mounted on clean glass slides and dried
overnight at 37 C. The sections were cleared, hydrated and stained with hematoxyllin and eosin.
Microscopically, the liver damage was classified using arbitrary scale from 0 to 5 as follows: degree 0–there
was no damage; degree 1–minimal lesions involving single to few necrotic cells; degree 2–mild lesions, 1025% necrotic cells or mild diffuse degenerative changes; degree 3–moderate lesions, 25-40% necrotic or
degenerative cells; degree 4–marked lesions, 40-50% necrotic or degenerative cells; degree 5–severe
lesions, more than 50% necrotic or degenerative cells. Sections, with scores higher than 2, were considered
to exhibit significant liver injury (Silva et al., 2001).
Influence of CAY10526 on ex vivo production of PGE2
Mice were sacrificed 6 h after APAP administration and samples of liver tissue, kept on ice, were minced in
small fragments (1-2 mm3) in PBS. After sedimentation at unit gravity, they were washed 2 times more in
fresh PBS, transferred into preweighed tubes and centrifuged at 500 g at +4 ºC for 3 min. The sediment was
quickly weighed, resuspended in Minimal Essential Medium (MEM, 5 µl MEM/mg tissue) and incubated in
a water bath at 37 ºC for 1 h. The samples were then centrifuged as above and supernatants stored at -80 ºC
until analysis. Concentration of PGE2 was determined using appropriate PGE2 EIA Kit according to the
manufacturer’s instructions (No. 514010, Cayman Chemical) (Čulo et al., 1995).
Mice were sacrificed under light ether anesthesia by cervical dislocation 6 h after APAP administration.
Immediately, liver lobes were taken from each animal (6 mice per group) and tissue samples were fixed in
4% paraformaldehyde in phosphate buffer and dehydrated in 100% ethanol. They were embedded in
paraffin wax, serially sectioned as 7 µm thick sections and mounted on glass slides. After removing paraffin
with xylene, the sections were rehydrated in ethanol and water. In order to quench endogenous peroxidase
activity, sections were incubated for 10 min in 0.3% H2O2, washed in PBS and then cooked in sodium citrate
buffer for 17 min at 95 °C. After being cooled to room temperature, sections were incubated with diluted
(1:100) polyclonal rabbit anti-NF-κB (p65 subunit) or anti-iNOS antibodies (No. sc-109 and sc-651,
respectively, Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) for 1 h at room temperature in a
humidified chamber. After being washed with PBS, sections were incubated with a biotinylated secondary
antibody (rabbit UniTect ABC Kit, Oncogene, Boston, Mass., USA) for 30 min at room temperature.
Afterwards, sections were washed in PBS and incubated with avidin biotinylated horseradish peroxidase
complex (ABC) for 30 min, washed again with PBS and stained with diaminobenzidine (DAB). Finally,
sections were rinsed in distilled water, counterstained with hematoxyllin and dehydrated in ethanol and
xylol. Cells positive to NF-κB had brown-stained cytoplasm or nuclei (depending of NF-κB activation) and
cells positive to iNOS had brown-stained cytoplasm. Positive internal controls for NF-κB and iNOS staining
were smooth muscle cells in vascular walls. Negative controls were prepared by substituting Dako
ChemMate antibody diluent for secondary antibody. Images were captured with digital camera (SPOT
Insight, Diagnostic Instruments, USA) mounted on an Olympus BX51 microscope using the SPOT software.
Liver tissue specimens of four experimental groups of mice were analyzed. The number of cells was
evaluated quantitatively by two independent investigators and classified as negative (not stained) cells,
nuclear positive for NF-κB counting and cytoplasm positive for iNOS counting. Counts were made along
the length of the liver only for DAB staining sections. DP-SOFT version 3.1 software was used to divide
each liver section into squares of 100μm×100μm at 40× magnifications. The cells below the left and upper
border of squares were not taken into account, but only those at the right and lower border. To avoid
counting the same cell twice, we used every other consecutive section. The examination was performed on
an Olympus BX51 microscope equipped with a DP11 a digital camera. Images were analyzed using DPSOFT version 3.1 software.
After deparaffinization and rehydration, the sections were treated in a microwave oven at 95 °C for 17
minutes in sodium citrate buffer (pH 6.0) for antigen retrieval. After being cooled to room temperature,
sections were washed in PBS. Sections were incubated with goat serum (Normal Goat Serum, X0907
DAKO, Glostrup, Denmark) for 1 h to block non-specific antibody binding. Sections were then incubated
with primary antibodies for 1 h at room temperature. Polyclonal rabbit anti-NF-κB p65 antibody from Santa
Cruz Biotechnology (No. sc-109), diluted 1:100, was used. After multiple washes in PBS, sections were
incubated for 1 h with diluted (1:200) secondary antibody (Texas Red, No. sc-2780, Santa Cruz
Biotechnology). Following secondary antibody incubation, the sections were washed in PBS and
counterstained with diamidino-2-phenylindole (DAPI) to stain nuclei. After final rinsing in PBS, sections
were mounted, air-dried and coverslipped (Immuno-Mount, Shandon, Pittsburgh, PA, USA). Control of
specificity included omitting primary antibody from the staining procedure. Sections were examined by
using a fluorescence microscope (Olympus BX61, Tokyo, Japan). Immunofluorescent images were obtained
with a digital camera (DP71) mounted on an Olympus BX61 microscope using the Olympus Cell software
and Adobe Photoshop.
Measurement of NO in plasma
Mice were sacrificed under light ether anesthesia and plasma samples were taken 6 h after treatment of mice
with APAP. Thereafter, samples were stored at -80 °C until determination of NO synthesis. For
quantification of total NO production, nitrate (NO3-) was first reduced into nitrite (NO2-) using coppercoated cadmium granules (NITRALYZER-II, Nitrate to Nitrite reduction kit, World Precision Instruments,
Sarasota, FL, USA). Concentration of nitrite was measured by Griess reaction according to the
manufacturer’s instructions. Finally, absorbance readings were performed with the use of an automated
microplate reader (Dynatech MR 5000, Dynatech Technology, Inc., Horsham, PA, USA) at 540 nm.
Results are expressed as mean ± SEM. Differences in survival between groups of mice were compared by
chi-square test using Yate’s correction when indicated. Statistical comparisons between two groups were
made using a Student’s t-test. Comparisons between multiple groups were carried out using one-way
analysis of variance (ANOVA) with a post hoc test of significance between individual groups. Significance
was accepted at p<0.05.
Effects of dmPGE2 and CAY10526 on APAP-induced mortality and plasma ALT concentration in
To determine the survival of animals, mice were treated with 300 mg/kg of APAP. DmPGE2 (0.2 mg/kg,
i.p.) and CAY10526 (2.0 mg/kg, i.p.) were given either 30 min before or 2 h after APAP administration.
Administration of dmPGE2 30 min before APAP significantly improved the survival of animals (Fig. 1A,
p<0.05). When given 2 h after APAP, dmPGE2 increased the survival of mice, but without statistical
significance (Fig. 1A, p>0.05). CAY10526 decreased the survival of animals when given either 30 min
before or 2 h after APAP, although, the differences did not reach statistical significance (Fig. 1B, p>0.05 for
both comparisons). To determine the plasma ALT concentration, mice were treated as in the previous
experiment, except that mice received a lower dose of APAP (150 mg/kg). Fig. 2A and B show mean ALT
levels (±SEM) obtained in 8-10 mice per group 20-22 h after APAP administration. Pretreatment or post
treatment of mice with dmPGE2 significantly reduced ALT level (Fig. 2A, p<0.05 for both comparisons).
Fig. 2B shows that CAY10526, if given 2 h after APAP, increased ALT concentration, but, due to high intra
group variability, the difference was not significant (p>0.05).
Effect of CAY10526 on ex vivo production of PGE2
PGE2 production was determined in supernatants of incubated liver fragments taken from normal mice and
mice treated with CAY10526 (2 mg/kg, i.p.) or vehicle 2 h after APAP administration. In comparison to
normal (non-treated) mice, treatment with APAP alone (vehicle group) significantly increased production of
PGE2, while treatment with CAY10526 reduced that increase in PGE2 production (Fig. 3, p<0.05 for both
Macroscopically, the whole liver surface of some APAP treated animals had a mottled appearance; dark red
hemorrhagic-necrotic spots were regularly scattered on the yellowish background. Microscopically, the liver
damage was graduated using arbitrary scale from 0 to 5 as described in Materials and Methods (Fig. 4). The
severity of necrosis was quite variable both between animals and also within different parts of the same
liver. However, dmPGE2 significantly decreased the number and size of necrotic foci in the liver, which
could be easily seen by macroscopic observation and on histological analysis. Macroscopic and microscopic
damages of the liver parenchyma appeared more pronounced in mice injected with CAY10526 (Table 1).
Immunohistochemistry to NF-κB p65
In normal (non-treated) mice, NF-κB was expressed in liver cells around blood vessels in Kiernan spaces
(afferent arterioles), where numerous small granules were diffusely scattered in the cytoplasm. NF-κB
immunostaining was also expressed in individual cells in the lobules and around the central vein (efferent
arterioles) (Fig. 5A). Almost all NF-κB immunoreactivity was seen in cytoplasm and there were only 3.7%
positive cells with nuclear positivity to NF-κB (Table 2). In mice which received APAP alone (vehicle
group), the whole liver tissue was infiltrated with reactive cells. Normal cells were rarely observed. Around
blood vessels, there were layers of cells with numerous vacuoles in cytoplasm and nuclei (Fig. 5B). In the
central part of lobules, perimembranous staining pattern and dark colored vacuoles in nuclei were observed.
Smaller cells could represent Kupffer cells. Besides cytoplasmic positivity to NF-κB, immunoreactivity was
also seen in nuclei (21.8% positive cells) of hepatocytes, indicating significantly higher activity of NF-κB in
comparison to normal mice (Table 2, p<0.001). In mice treated with PGE2, the whole tissue was also
infiltrated with reactive cells. However, a lot of cells were in the different mitotic periods, which implied
liver regeneration. Around the blood vessels, there were reactive cells with small, diffusely scattered
granules giving to individual cells a strong positivity in cytoplasm in comparison to nuclei (Fig. 5C).
Immunohistochemical localization of NF-κB, especially around the blood vessels, was mostly seen in the
cytoplasm, while its nuclear expression (8.0% cells with NF-κB positive nuclei) was significantly reduced in
comparison to mice which received APAP alone (Table 2, p<0.001). In mice treated with CAY10526,
immunohistochemical expression of NF-κB in liver cells showed mostly perimembranous staining pattern.
In the central part of lobules, NF-κB positivity was observed in the cytoplasm and in nuclei with vacuoles in
both cell compartments (Fig. 5D). Compared to mice which received APAP alone, nuclear positivity to NFκB in liver cells was higher (23.0% positive cells), but did not reach statistical significance (Table 2,
Immunofluorescence to NF-κB p65
Immunofluorescent analysis of NF-κB expression in liver cells generally confirmed previous investigation
done by immunohistochemistry. In group of normal mice, only a few cells showed nuclear positivity to NFκB (Fig. 6A). Nuclear expression of NF-κB was significantly higher in mice which received APAP alone in
comparison to normal mice (Fig. 6B), whereas treatment with PGE2 reduced that expression (Fig. 6C). In
mice treated with CAY10526, nuclear positivity to NF-κB revealed a similar pattern as in the group which
received APAP alone (Fig. 6D).
Immunohistochemistry to iNOS
Normal mice. Granular accumulations of iNOS were seen in the cytoplasm of hepatocytes organized in
small groups, while other hepatocytes were devoid of iNOS reactivity (Fig. 7A). Cells on the inner surface
of sinusoid capillaries (endothelial or Kupffer cells) also displayed reactivity to iNOS in their cytoplasm.
There were 3.9% positive cells with cytoplasmic positivity to iNOS (Table 2).
Mice which received APAP alone. iNOS was expressed in the cytoplasm of hepatocytes in the form of dense
small granules or large single vacuoles. Some endothelial (or Kupffer) cells expressed iNOS in their
cytoplasm as well (Fig. 7B). The number of hepatocytes with cytoplasmic immunoreactivity to iNOS
(41.6% positive cells) was significantly higher in comparison to normal mice (Table 2, p<0.001).
PGE2 treated mice. Granular expression of iNOS was observed in the cytoplasm of hepatocytes, particularly
of those situated around the lumen of blood vessels such as central vein (Fig. 7C). The number of liver cells
positive to iNOS (20.5% positive cells) was significantly lower in comparison with the group which
received APAP alone (Table 2, p<0.001).
CAY10526 treated mice. iNOS was expressed in the cytoplasm of hepatocytes in the form of fine granular
accumulations or large single vacuoles (Fig. 7D). The number of hepatocytes, which showed positive
immunoreactivity to iNOS (42.5% positive cells), was similar as in the group which received APAP alone
(Table 2, p>0.05).
Effects of PGE2 and CAY10526 on the plasma concentration of nitrite/nitrate (Fig. 8)
NO synthesis by hepatocytes, measured as a concentration of nitrite/nitrate in mice plasma, was almost
undetectable in normal animals. Treatment with APAP significantly elevated nitrite concentration, whereas
pretreatment with PGE2 reduced that level, but not significantly. In CAY10526 treated animals, nitrite level
was elevated in comparison to the group which received APAP alone, but again the difference was not
Drug toxicity is a severe complication to drug therapy and new drug development. Because APAP is a major
cause of acute liver failure in the western world (reviewed in Lee, 2003; Bernal, 2003), we used it as a
model of drug-induced liver injury for examination the influence of PGE2 and its derivatives on drug
Although the protective role of PGE2 and its analogues is well proven in other models of liver injury,
especially in those considered as immune system-mediated (reviewed in Quiroga and Prieto, 1993; Muntané
et al., 2000; Yin et al., 2007), the protective effect of PGE2 in a model of APAP-induced toxicity was shown
only indirectly. It was reported that a lack of PGE2 in COX-2 deficient mice (COX-2-/+ and -/- mice) and
inhibition of COX-2 by selective COX-2 inhibitory drug, celecoxib, increased the hepatotoxic effect of
APAP (Reilly et al., 2001). In our previous experiments, we demonstrated that the protective effect of
interleukin 1α (IL-1α) on APAP-induced toxicity in mice was abolished by administration of specific antiPGE2 antibodies (Renić et al., 1995). The presented results clearly demonstrate that dmPGE2 improves the
survival of mice, especially when it was injected into mice before an APAP overdose. Furthermore, hepatic
damage, as assessed by plasma ALT concentration and liver histology, was alleviated when animals
received dmPGE2 either before or after an APAP administration. These findings are in accordance with
previous studies of other authors describing beneficial protective effects of PGE2 against a variety of
hepatotoxic agents other than APAP (ethanol, CCl4, aflatoxin, Con A, DGalN, LPS alone, or combined with
DGalN) (reviewed in Quiroga and Prieto, 1993; Yin et al., 2007). CAY10526 is known as an inhibitor of
PGE2 production through the selective modulation of mPGES-1 expression. In the present experiments,
treatment of mice with CAY10526 aggravated the liver damage, as shown by the increase in mortality of
animals, elevation of serum ALT level and histopathological changes in liver morphology. As far as we
know, this is the first time that CAY10526 has been used in vivo in a model of experimental liver damage
induced by a noxious agent. The mechanism of its hepatotoxic action is most probably due to the inhibition
of PGE2 synthesis, because it significantly inhibited ex vivo production of PGE2 by liver homogenates and it
displayed the strongest effect when given to mice 2 h after APAP. This indirectly points to PGE2 as an
endogenously produced hepatoprotective agent, which is supported by our preliminary observation that
APAP alone increases synthesis of PGE2 and PGI2 in the liver (data not shown).
Our investigations demonstrated that administration of APAP resulted in a higher expression of NF-κB in
liver cells from control animals. This is in agreement with previous studies, in which it was shown that
APAP-induced liver damage was influenced by oxidative stress (Jaeschke, 2000; James et al., 2003) and that
this is dependent, in part, on NF-κB (Bauer et al., 2000; Dambach et al., 2006). Immunohistochemistry and
immunofluorescence to NF-κB revealed that administration of dmPGE2 significantly inhibited activation of
NF-κB and its translocation from the cytoplasm to the nucleus of hepatocytes, while CAY10526 had the
opposite effect. In another animal model, it was shown that selective EP4 agonist (EP4Rag) suppressed
production of proinflammatory cytokines, chemokines, adhesion molecules and NF-κB (Ogawa et al., 2009).
In LPS-treated cultured rat liver macrophages (Kupffer cells), PGE2 reduced TNF-α production due to down
activation of NF-κB (Tran-Thi et al., 1995). NF-κB is known to up regulate expression of genes controlling
inflammatory mediators including iNOS, TNFα, IL-1, IL-10 and COX-2, each of which has been shown to
be implicated in APAP-induced hepatotoxicity (Dambach et al., 2006). Although it is not yet clear whether
NF-κB plays a positive or negative role in the pathogenesis of toxicity in our and above described models,
our results just indicate that a protective effect of PGE2 could be expressed through inhibition of activation
The present results also showed that treatment with APAP high significantly increased the percentage of
iNOS-positive cells and that PGE2 significantly diminished expression of iNOS in hepatocytes and reduced
NO production, which was reflected by lower (albeit statistically not significant) nitrite/nitrate concentration
in plasma in comparison to animals receiving APAP alone. CAY10526 showed slightly inverse action on
iNOS production. Our results are in agreement with findings provided by others, which reported a
significant increase of NO synthesis in APAP toxicity (Hinson et al.,1998; Jaeschke et al., 2003) indicating
that increased synthesis of iNOS has a pathogenic role in hepatitis induced by APAP. However, data in the
literature on role of iNOS in liver damage, i.e. whether it is a part of regenerative (protective) or toxic
(pathogenic) mechanisms, are highly controversial. This was shown in various models of hepatic damage
using different experimental designs, including toxic hepatitis induced by APAP (Bourdi et al., 2002;
Gardner et al., 2002; Ito et al., 2003; Fiorucci et al., 2005), CCl4 (Morio et al., 2001), immune-mediated
hepatitis (Willuweit et al., 2001), etc. Explanation for these conflicting results might be in conception that
iNOS is pathogenic and constitutive NOS (cNOS) protective (Fiorucci et al., 2005), or by observation that
the action of NO and its derivatives depends on precise targets on which it is acting and the level of
oxidative tissue injury done by superoxide (Laskin et al., 2001). It was shown in this model that NO
scavenges superoxide anion to produce peroxynitrite, which then causes protein nitration and tissue injury
(Jaeschke 2000; Jaeschke et al., 2003). Despite its controversial role in hepatotoxicity, NO is generally
considered as a proinflammatory mediator and our results, at least in part, support that opinion.
It has been shown that the major producers of endogenous PGs in the liver are Kupffer cells and extra
hepatic inflammatory cells recruited to liver by chemoatractants (Decker, 1990). APAP-induced
hepatotoxicity was paralleled by significant elevation in synthesis of PGI2, PGE2 and TXA2 from liver
homogenates or fragments of treated animals (Guarner et al.,1988, our unpublished data). Possible
cytoprotective mechanisms of PGE2 could be, besides the above discussed pathways, due to vasodilatation
(by which it can reduce or reverse hepatic vascular congestion), increase in intracellular cAMP level and
stimulation of mitogenesis in hepatocytes (reviewed in Quiroga and Prieto, 1993). Data in vitro have
revealed that cAMP, as well as PGE2, inhibits the release of potentially hepatotoxic inflammatory cytokines:
TNF-α, IL-1 and IFN-γ (Oh-ishi et al., 1996; Schroer et al., 2002). Taken together with our previous
investigations, these findings support the view that PGE2 has a cytoprotective effect and is involved in the
defense of the organism to noxious effects of xenobiotics on liver. According to our results, this protection is
mediated, at least partially, through down activation of NF-κB and iNOS. Therefore, much remains to be
done to elucidate precise mechanisms underlying cytoprotective effects of PGs in acute liver injury.
Araki H., Ukawa H., Sugawa Y., Yagi K., Suzuki K. and Takeuchi K. (2000). The roles of prostaglandin E
receptor subtypes in the cytoprotective action of prostaglandin E2 in rat stomach. Aliment.
Pharmacol. Ther. 14, 116-124.
Bauer I., Vollmar B., Jaeschke H., Rensing H., Kraemer T., Larsen R. and Bauer M. (2000). Transcriptional
activation of heme oxygenase-1 and its functional significance in acetaminophen-induced hepatitis
and hepatocellular injury in the rat. J. Hepatol. 33, 395-406.
Bernal W. (2003). Intensive care support therapy. Liver. Transpl. 9, 15-17.
Bourdi M., Masubuchi Y., Reilly T.P., Amouzadeh H.R., Martin J.L., George J.W., Shah A.G. and Pohl
L.R. (2002). Protection against acetaminophen-induced liver injury and lethality by interleukin 10:
role of inducible nitric oxide synthase. Hepatology 35, 289-298.
Bourne R., Lichtenstein L.M., Melmon K.L., Henney C.S., Weinstein Y. and Shearer G. M. (1974).
Modulation of Inflammation and Immunity by Cyclic AMP. Science 184, 119-128.
Cheon H., Rho Y. H., Choi S. J., Lee Y. H., Song G. G., Sohn J., Won N. H. and Ji J. D. (2006).
Prostaglandin E2 augments IL-10 signaling and function. J. Immunol. 177, 1092-1100.
Čulo F., Renić M., Sabolović D., Radoš M., Bilić A. and Jagić V. (1995). Ketoconazole inhibits
acetaminophen-induced hepatotoxicity in mice. European Journal of Gastroenterology and
Hepatology 7, 757-762.
Dambach D. M., Durham S. K., Laskin J. D. and Laskin D. L. (2006). Distinct roles of NF-kappaB p50 in
the regulation of acetaminophen-induced inflammatory mediator production and hepatotoxicity.
Toxicol. Appl. Pharmacol. 211, 157-165.
Decker K. (1990). Biologically active products of stimulated liver macrophages (Kupffer cells). Eur. J.
Biochem. 192, 245-261.
Dey I., Lejeune M. and Chadee K. (2006). Prostaglandin E2 receptor distribution and function in the
gastrointestinal tract. Br. J. Pharmacol. 149, 611-623.
Enomoto A., Itoh K., Nagayoshi E., Haruta J., Kimura T., O’Connor T., Harada T. and Yamamoto M.
(2001). High Sensitivity of Nrf2 Knockout Mice toAcetaminophen Hepatotoxicity Associated with
Decreased Expression of Regulated Drug Metabolizing Enzymes and Antioxidant Genes.
Toxicological Sciences 59, 169–177.
Fiorucci A., Gerometta P.S., DeVecchi M., Guzman C., Costantino M.L. and Arena V. (2004). In vitro
assessment of the vacuum-assisted venous drainage (VAVD) system: risks and benefits. Perfusion
Gardner C. R., Laskin J. D., Dambach D. M., Sacco M., Durham S. K., Bruno M. K., Cohen S. D., Gordon
M. K., Gerecke D. R., Zhou P. and Laskin D. L. (2002). Reduced hepatotoxicity of acetaminophen in
mice lacking inducible nitric oxide synthase: potential role of tumor necrosis factor-alpha and
interleukin-10. Toxicol. Appl. Pharmacol. 184, 27-36.
Guarner F., Boughton-Smith N. K., Blackwell G. J. and Moncada S. (1988). Reduction by prostacyclin of
acetaminophen-induced liver toxicity in the mouse. Hepatology 8, 248-253.
Hao C. M. and Breyer M. D. (2008). Physiological regulation of prostaglandins in the kidney. Annu. Rev.
Physiol. 70, 357-377.
Harris S. G., Padilla J., Koumas L., Ray D. and Phipps R. P. (2002). Prostaglandins as modulators of
immunity. Trends Immunol. 23, 144-150.
Hata A. N. and Breyer R. M. (2004). Pharmacology and signaling of prostaglandin receptors: multiple roles
in inflammation and immune modulation. Pharmacol. Ther. 103, 147-166.
Hinson J. A., Pike S. L., Pumford N. R. and Mayeux P. R. (1998). Nitrotyrosine-protein adducts in hepatic
centrilobular areas following toxic doses of acetaminophen in mice. Chem. Res. Toxicol. 11, 604607.
Ito Y., Abril E.R., Bethea N.W. and McCuskey R.S. (2003). Role of nitric oxide in hepatic microvascular
injury elicited by acetaminophen in mice. Am J Physiol Gastrointest Liver Physiol 286, 60-67.
Jaeschke H. (2000). Reactive oxygen and mechanisms of inflammatory liver injury. J. Gastroenterol.
Hepatol. 15, 718-724.
Jaeschke H., Knight T. R. and Bajt M. L. (2003). The role of oxidant stress and reactive nitrogen species in
acetaminophen hepatotoxicity. Toxicol. Lett. 144, 279-288.
James L. P., Mayeux P. R. and Hinson J. A. (2003). Acetaminophen-induced hepatotoxicity. Drug. Metab.
Dispos. 31, 1499-1506.
Janssen-Heininger Y. M., Poynter M. E. and Baeuerle P. A. (2000). Recent advances towards understanding
redox mechanisms in the activation of nuclear factor kappaB. Free Radic. Biol. Med. 28, 1317-1327.
Kamanaka Y., Kawabata A., Matsuya H., Taga C., Sekiguchi F. and Kawao N. (2003). Effect of a potent
iNOS inhibitor (ONO-1714) on acetaminophen-induced hepatotoxicity in the rat. Life Sci. 74, 793802.
Laskin J. D., Heck D. E., Gardner C. R. and Laskin D. L. (2001). Prooxidant and antioxidant functions of
nitric oxide in liver toxicity. Antioxid. Redox Signal. 3, 261-271.
Lee W. M. (2003). Acute liver failure in the United States. Semin. Liver Dis. 23, 217-226.
Matsuoka T. and Narumiya S. (2007). Prostaglandin receptor signaling in disease. ScientificWorld Journal
Mazer M. and Perrone J. (2008). Acetaminophen-induced nephrotoxicity: pathophysiology, clinical
manifestations, and management. J. Med. Toxicol. 4, 2-6.
Morio L.A., Chiu H., Sprowles K.A., Zhou P., Heck D.E., Gordon M.K. and Laskin D.L. (2001). Distinct
roles of tumor necrosis factor-alpha and nitric oxide in acute liver injury induced by carbon
tetrachloride in mice. Toxicol. Appl. Pharmacol. 172, 44-51.
Muntané J., Rodríguez F.J., Segado O., Quintero A., Lozano J.M., Siendones E., Pedraza C.A., Delgado M.,
O'Valle F., García R., Montero J.L., De La Mata M. and Miño G. (2000). TNF-alpha dependent
production of inducible nitric oxide is involved in PGE(1) protection against acute liver injury. Gut
Nakano H., Monden M., Umeshita K., Murata M., Miyoshi H., Kanai T., Gotoh M. and Mori T. (1994).
Cytoprotective effect of prostaglandin I2 analogues on superoxide-induced hepatocyte injury.
Surgery 116, 883-889.
O'Connor J. A. and Bennett S. (2005). Acetaminophen toxicity revisited: is drug-induced hepatotoxity
immune mediated? J. Pediatr. Gastroenterol. Nutr. 40, 528-529.
Oh-ishi S., Utsunomiya I., Yamamoto T., Komuro Y. and Hara Y. (1996). Effects of prostaglandins and
cyclic AMP on cytokine production in rat leukocytes. Eur. J. Pharmacol. 300, 255-259.
Ogawa M., Suzuki J., Kosuge H., Takayama K., Nagai R. and Isobe M. (2009). The mechanism of
antiinflammatory effects of prostaglandin E2 receptor 4 activation in murine cardiac transplantation.
Transplantation 87, 1645-1653.
Quiroga J. and Prieto J. (1993). Liver cytoprotection by prostaglandins. Pharmacol. Ther. 58, 67-91.
Reilly T. P., Brady J. N., Marchick M. R., Bourdi M., George J. W., Radonovich M. F., Pise-Masison C. A.
and Pohl L. R. (2001). A protective role for cyclooxygenase-2 in drug-induced liver injury in mice.
Chem. Res. Toxicol. 14, 1620-1628.
Renić M., Čulo F. and Sabolović D. (1995). Protection of acute hepatotoxicity in mice by interleukin-1α –
the role of interleukin-6 and prostaglandin E2. Periodicum Biologorum 97, 55-60.
Schroer K., Ying Z., Saunders M.A., Deng W.G., Xu X.M., Meyer-Kirchrath J. and Wu K.K. (2002).
Obligatory Role of Cyclic Adenosine Monophosphate Response Element in Cyclooxygenase-2
Promoter Induction and Feedback Regulation by Inflammatory Mediators. Circulation 105, 27602765.
Silva V. M., Chen C., Hennig G. E., Whiteley H. E. and Manautou J. E. (2001). Changes in susceptibility to
acetaminophen-induced liver injury by the organic anion indocyanine green. Food Chem. Toxicol.
Suzuki K., Araki H., Mizoguchi H., Furukawa O. and Takeuchi K. (2001). Prostaglandin E inhibits
indomethacin-induced gastric lesions through EP-1 receptors. Digestion 63, 92-101.
Tran-Thi T. A., Decker K. and Baeuerle P. A. (1995). Differential activation of transcription factors NFkappa B and AP-1 in rat liver macrophages. Hepatology 22, 613-619.
Willuweit A., Sass G., Schöneberg A., Eisel U., Tiegs G. and Clauss M. (2001). Chronic inflammation and
protection from acute hepatitis in transgenic mice expressing TNF in endothelial cells. J. Immunol.
Yin H., Cheng L., Langenbach R. and Ju C. (2007). Prostaglandin I(2) and E(2) mediate the protective
effects of cyclooxygenase-2 in a mouse model of immune-mediated liver injury. Hepatology 45,