COMPARATIVE STUDY ON THERAPEUTIC POTENTIAL OF CAFFEIC ACID AND SILYMARIN IN PARACETAMOL-INDUCED HEPATOTOXICITY: EFFECT ON HO-1, OXIDATIVE STRESS, HEPATIC INFLAMMATION AND NEUTROPHILS INFILTRATION

Paracetamol is a widely used analgesic and antipyretic drug, but at high doses it leads to undesirable side effects, mainly hepatotoxicity. The hepatoprotective effect of caffeic acid has been previously reported, however the mechanism of this protective effect is not fully explored. The current study aimed to investigate the therapeutic potential of caffeic acid vs. silymarin in paracetamol-induced hepatic damage. To better understand the mechanisms by which these phenolic compounds confer their protective effects, parameters of lipid peroxidation, hepatic inflammation and neutrophils infiltration were evaluated. In addition, liver toxicity markers and hepatic antioxidants (enzymatic as well as non-enzymatic) were measured. Daily administration of paracetamol (700 mg/kg, p.o.) caused hepatic injury that was manifested as increased hepatic levels of malonaldehyde (MDA), tumor necrosis factor-alpha (TNF-α) and heme oxygenase (HO-1) with marked increase in myeloperoxidase (MPO) activity, depletion of liver reduced glutathione, and diminished catalase and super-oxide dismutase (SOD) activities. Furthermore, serum liver enzymes (ALT, AST, GGT) were significantly increased along with hepatocellular degeneration and steatosis. Co-treatment with caffeic acid or silymarin alleviated paracetamol-induced oxidative stress, blunted TNF-α levels and MPO activity, while enhanced HO-1 levels with restoration of antioxidant enzymes activities. This was accompanied by normalization of liver transaminases and improvement of hepatic architecture. Interestingly, the effect of caffeic acid on HO-1, TNF-α, and MPO was more prominent when compared to silymarin. The present study provide an evidence on the multiple mechanisms by which caffeic acid as well as silymarin grant their hepatoprotective activities mainly through the induction of HO-1 in addition to the reduction of oxidative stress burden in hepatocytes. The powerful anti-oxidant effect was accompanied by a significant anti-inflammatory activity.


INTRODUCTION
Paracetamol is a commonly used analgesic and antipyretic agent which is usually safe and well tolerated when used at therapeutic doses.However, acute paracetamol overdose causes severe and fatal hepatotoxicity (Maddrey, 2005).Paracetamol toxicity is one of the most common causes of poisoning worldwide, mainly causing liver injury and acute liver failure (Larson et al., 2005;Ryder and Beckingham, 2001).Damage to the liver results from one of paracetamol metabolites, that is N-acetyl-p-benzoquinoneimine (NAPQI).NAPQI is a highly reactive intermediate that depletes hepatic glutathione and binds covalently to intracellular proteins including mitochondrial proteins (James et al., 2003).Consequently, enormous hepatocellular necrosis occurs as a result of mitochondrial dysfunction, oxidative and nitrative stress, and inflammatory reactions (Jaeschke and Bajt, 2006).
Several studies have documented the effective protection afforded by antioxidants and anti-inflammatory agents against paracetamol-induced hepatotoxicity (Girish et al., 2009;Nagi et al., 2010;Yan et al., 2009).However most of these studies reported the protective effect of various agents only when administered prior to paracetamol insult.Hence, there is a need to find agents that can stop or even reverse the hepatotoxicity when administered following paracetamol-induced injury.
The main objectives of the current study were to evaluate the protective effects of caffeic acid concomitant treatment in comparison with silymarin against paracetamol-induced hepatic injury in rats and to investigate the possible mechanism(s) underlying these effects.Therapeutic potential was assessed by monitoring liver injury through estimation of serum alanine aminotransferase (sALT), aspartate aminotransferase (sAST), and gamma-glutamyl transferase (GGT) as well as hepatic lipid peroxidation expressed as malondialdehyde (MDA).In addition, the activities of hepatic endogenous antioxidants including reduced glutathione (GSH), catalase, and superoxide dismutase (SOD), hepatic levels of the pro-inflammatory marker TNF-α, and the neutrophils infiltration marker myeloperoxidase (MPO) were evaluated.

Animals
Male Wister rats weighing 120-150g were obtained from the Egyptian Organization for Biological Products and Vaccines (Egypt) and housed under controlled temperature (25 ± 1°C) on a 12 h light/dark cycle.Rats were allowed to acclimatize for one week before starting the experiment.Food and water were allowed ad libitum during the study period.

Experimental design
Thirty two rats were randomly divided into four groups of eight rats each and treated as follows:  1 st group: rats that served as the untreated (normal) control and were administered single daily dose of 5 ml/kg body weight of distilled water orally for 14 days. 2 nd group: rats that served as the model control and were administered single daily dose of paracetamol suspension at 700 mg/kg body weight for 14 days. 3 rd group: rats were co-treated daily with single oral dose of caffeic acid at 40 mg/kg body weight 1 hour after oral administration of paracetamol suspension (700 mg/kg) for 14 days. 4 th group: rats were co-treated daily with single oral dose of silymarin at 100 mg/kg 1 hour after oral administration of paracetamol suspension (700 mg/kg) for 14 days.

Sample preparation
At the end of experiment, 6 rats from each group were selected randomly and sacrificed by cervical dislocation, blood samples were collected, left for 60 min to clot, and then centrifuged for 10 min at 5000 rpm to obtain clear sera which were stored at −20°C.
The livers were excised, divided into two portions, one portion was washed with ice-cold saline, and subsequently homogenized in cold potassium phosphate buffer (0.05 M, pH 7.4).The homogenates were centrifuged at 14000 rpm for 15 min at 4°C and the resulting supernatant was stored at −80°C.The other portion was kept for histopathological evaluation.

Biochemical determination of liver enzymes
Liver enzymes (ALT, AST, and GGT) were determined in serum samples spectrophotometrically using appropriate kits.Protocols used were according to the recommended manufacturer procedures.

Measurements of lipid peroxidation and antioxidants in liver
Liver homogenate from different groups were used for measuring MDA, GSH, and the activities of SOD and catalase using standard spectrophotometric assays.Levels of MDA as thiobarbituric acid-reactive substances (TBRS) were measured by the method of Ohkawa et al. (1979).GSH was determined based on the use of Ellman's reagent (Beutler et al., 1963).SOD and catalase activities were measured as described previously (Aebi, 1984;Nishikimi et al., 1972).

Determination of liver TNF-α
Levels of TNF-α in liver homogenate were determined using ELISA following the manufacturer's protocol (R&D systems).In brief, to a 96-well microplate pre-coated with polyclonal antibody specific for rat TNF-α, 50 μl assay diluent and 50 μl samples were added.Recombinant rat TNF-α was used to set up the corresponding standard curve.After incubation for 2h at room temperature, the wells were washed and polyclonal anti-rat TNF-α antibody conjugated to horseradish peroxidase was added.Incubation was continued for 2h, plates were washed; substrate solution was added to each well and incubated for 30 min.The enzyme reaction yielded a blue product that turned yellow when the stop solution was added.The O.D at 450nm (correction wavelength set at 570 nm) was measured using microplate reader (Metertech, M960).

Determination of liver HO-1
Levels of HO-1 in liver homogenate were determined using ELISA following the manufacturer's protocol (Stressgen).Briefly, to a 96-wellmicroplate pre-coated with monoclonal antibody specific for rat HO-1, 100 ml standards, sample diluent, or samples were added to appropriate wells.Recombinant rat HO-1was used to setup the standard curve.After incubation for 1h at room temperature, the wells were washed and polyclonal antibody specific for rat HO-1 was added followed by incubation for another 1h.After washing, 100 ml of rat HO-1 horseradish peroxidase conjugate was added.Incubation was continued for 30min at room temperature, plates were washed; tetramethylbenzidine substrate solution was added to each well and incubated for 15min in dark.The enzyme reaction yielded a blue product that turned yellow when the stop solution was added.The O.D at 450nm was measured using microplate reader (Metertech, M960).

Determination of MPO activity in liver
Myeloperoxidase (MPO) activity was determined as described by Weiss and coworkers (1982).Briefly, hypochlorous acid (HOCl) was formed from MPO catalyzed reaction between chloride and hydrogen peroxide.HOCl was rapidly trapped by amino acid taurine to form a stable oxidant taurine chloramine.After incubation for 30min, the MPO catalyzed reaction was stopped by adding catalase to eliminate hydrogen peroxide.Taurine chloramine was then allowed to react with the chromophore 5-thio-2-nitrobenzoic acid (TNB) to give a colorless reaction product 5-5-dithiobis(2-nitrobenzoic acid) or DTNB.By following the decrease in absorbance at 412 nm, MPO activity was measured.One unit is the amount of MPO that can produce 1.0 nmole of taurine chloramine (hypochlorous acid) at pH 6.5 and 25°C during 30 minutes in the presence of 100 mM chloride and 100 µM of hydrogen peroxide.

Histopathological examination of liver tissue
The other liver portion from each animal was fixed in 10% formalin, dehydrated in ascending grades of alcohol and embedded in paraffin.Sections were cut at 4 µm, stained with hematoxylin and eosin (H&E) and examined under a light microscope for histopathologic evaluations.

Statistical analysis
All data were expressed as mean ± S.E.M. Statistical significance was tested by one way analysis of variance (ANOVA) followed by Bonferroni post hoc analysis.A P value <0.05 was considered statistically significant.

Effect of caffeic acid vs. silymarin on hepatic toxicity markers
Treatment with paracetamol (700 mg/kg, daily) for 2 weeks disrupted liver function as reflected by marked (2-to 6-fold) rises in serum alanine aminotransferase (Pestka et al., 2004), aspartate aminotransferase (Camacho-Barquero et al., 2007), and gamma-glutamyl transferase (GGT) (Fig. 1).The treatment with caffeic acid concomitantly with paracetamol significantly (P ≤ 0.05) inhibited paracetamol-induced increases in serum aminotransferases and kept their levels close to normal values.The effect of silymarin, the known hepatoprotective agent, on ALT and AST was comparable to caffeic acid.However, the level of GGT after silymarin treatment stayed slightly higher than normal levels.

Effect of caffeic acid vs. silymarin on lipid peroxidation and antioxidants in liver
Oxidative stress and endogenous antioxidant levels were evaluated to gain insights into the hepatic molecular and cellular changes occurring following paracetamol treatment.

Effect of caffeic acid vs. silymarin on HO-1 levels in liver
Following treatment with paracetamol for 2 weeks, a marked (15-fold) increase in heme oxygenase-1 (HO-1) protein was observed in liver cell homogenate suggesting a stress response by hepatocytes.The co-treatment with caffeic acid resulted in a further 5-fold increase in HO-1 expression while the concomitant treatment with silymarin caused only 3.5-fold increase in HO-1 compared to paracetamol treated group (Fig. 3).

Effect of caffeic acid vs. silymarin on paracetamol-induced hepatic inflammation
To gain more insights into the mechanisms by which caffeic acid and/or silymarin confer their hepatoprotective effects against paracetamol-induced hepatotoxicity, TNF-α and MPO were evaluated as markers for hepatic inflammation and neutrophils infiltration, respectively.Paracetamol significantly (P ≤ 0.05) increased the hepatic TNF-α level (3.6-fold) compared with control rats.Caffeic acid or silymarin co-administration with paracetamol significantly (P ≤ 0.05) reduced paracetamol-induced hepatocytic production of TNF-α by 66 % and 50 %, respectively (Fig. 4A).Similarly, paracetamol markedly (7-fold) increased MPO activity in liver homogenate compared with normal control group.Co-administration of caffeic acid or silymarin significantly (P ≤ 0.05) reduced paracetamol-induced MPO activity in liver cell homogenates by 78 % and 68%, respectively, compared with paracetamol control group (Fig. 4B).

Histopathological evaluation
The microscopic examination of liver of control rats showed normal architecture without any signs of vascular or inflammatory changes (Fig. 2A).Histopathological examination revealed that paracetamol induced diffuse cytoplasmic vacuolation, centrilobular necrosis, vascular congestion and nuclear pyknosis of the hepatocytes which are indicative of acute hepatocellular injury (Fig. 2B) that came in line with the increased levels of serum aminotranferases.Whereas, caffeic acid or silymarin co-treatment preserved the liver histology close to the normal architecture with minimal small focal necrosis and no evidence of degeneration or steatosis (Fig. 2C-D) which is consistent with the amelioration of pathologic anomalies by both compounds.

DISCUSSION
Caffeic acid, one of the most widely distributed flavonoids, has been demonstrated to have potent antioxidant ability (Jayanthi and Subash, 2010; Rice-Evans and Miller, 1996) and has been reported to protect cells from oxidative damages caused by a diverse range of oxidative stresses, including paracetamol-mediated toxicity (Janbaz et al., 2004).However, to the best of our knowledge, the present study is the first to investigate the therapeutic potential of "posttreatment" with caffeic acid against paracetamol-induced hepatotoxicity in rats.
Cancer patients on chemotherapy generally receive higher doses of paracetamol to relieve pain (Nassar et al., 2009).Hepatic damage as a result of paracetamol overdoses in human is fairly common (Nelson, 1995).At therapeutic doses, paracetamol is metabolized primarily in the liver via glucuronidation and sulfation reactions, and results in water-soluble metabolites that are excreted via kidney.As a result of the metabolic conversion of paracetamol by the microsomal CYP-450 enzyme system, a highly reactive intermediate, N-acetyl-p-benzoquinoneimine (NAPQI) is produced (Mitchell et al., 1973).NAPQI directly reacts with GSH and at overdoses of paracetamol, depletion of cellular GSH occurs allowing NAPQI to bind to cellular proteins especially in the mitochondria.The resulting mitochondrial oxidative and nitrative stress leads to mitochondrial dysfunction and eventually hepatocellular necrosis (Jaeschke et al., 2011).

B A D C
Indeed, histopathological evaluation confirmed the hepatic damage associated with high doses of paracetamol in the current model as was evident by hepatic necrosis and steatosis.In addition, the elevated levels of serum transaminases observed in this study support the previous observation.Our results are in accordance with previous reports showing that hepatic damage resulting from paracetamol overdoses is always associated with increased serum liver enzymes as marker for hepatotoxicity (Heard, 2008;Larson et al., 2005) Paracetamol also induces the release of numerous cytokines and signaling molecules from activated Kupffer cells, including TNF-α which in turn up-regulates the iNOS enzyme (Gardner et al., 2002;Morris and Billiar, 1994) resulting in increased hepatic production of nitric oxide (Jaeschke et al., 2011).Excess nitric oxide depletes intracellular GSH that increases the susceptibility to oxidative stress (Clancy and Abramson, 1995).Moreover, excess nitric oxide reacts with superoxide anion to generate peroxynitrite radical that causes further cell damage by oxidizing and nitrating cellular macromolecules.In agreement with previous studies (Girish et al., 2009;Nagi et al., 2010;Oz and Chen, 2008;Yan et al., 2009), current results demonstrated that oxidative stress accompanied by increased lipid peroxidation, undermined antioxidant defenses and increased release of pro-inflammatory cytokines which are the hallmarks of paracetamol hepatotoxicity.The antioxidant effect of caffeic acid was evident in the this study through the restoration of catalase and SOD activities as well as GSH with blunting of lipid peroxidation.Our results came in line with previous reports that demonstrated similar effect of caffeic acid in different models (Jayanthi and Subash, 2010; Kus et al., 2004).Silymarin produced a comparable effect to caffeic acid in the current study.The antioxidant activities of both polyphenols were associated with an improved liver architecture and normalized serum transaminases.
There is an emerging evidence that caffeic acid, as well as other polyphenols, may exert modulatory actions on the endogenous antioxidative defense system by interactions with intracellular signaling cascades in addition to their ability of quenching ROS (Williams et al., 2004).Heme oxygenase-1 (HO-1), a redox-sensitive inducible protein that provides efficient cytoprotection against oxidative stress, has attracted much attention for its up-regulation in a variety of stress-related circumstances (Dore, 2002;Ryter et al., 2007).Growing evidences indicate that HO-1 plays a key role in defense mechanisms against oxidative damages (Nakahira et al., 2003;Ryter et al., 2007).The induction of the HO-1 gene has been proposed to be a beneficial adaptive response to oxidative stress (Applegate et al., 1991).Further, Alum et al. (1995) have shown HO-1 gene to be induced by reactive oxygen species.It is known that both biliverdin and bilirubin, the byproducts of heme metabolism by HO-1, exhibit antioxidant properties.In the light of these observations, our findings suggest that the reactive oxygen species released from the damaged liver as a result of paracetamol overdose induced HO-1 expression.Indeed, activation of HO-1 at the transcriptional level has been previously reported in acetaminophen-induced hepatitis which confirm the current finding (Bauer et al., 2000;Ishida et al., 2006).On the other hand, preceding findings showed that caffeic acid markedly increased heme oxygenase activity and HO-1 protein in astrocytes (Scapagnini et al., 2002).Similarly, silymarin was reported to up-regulate HO-1 mRNA in human hepatoma cells (Bonifaz et al., 2009).However, as far as we know, the current study is the first to investigate the effect of caffeic acid and silymarin on liver HO-1 in the paracetamol-induced hepatotoxicity.Both caffeic acid and silymarin enhanced the production of HO-1 by the liver far beyond the effect of paracetamol alone suggesting an additional mechanism by which these polyphenols confer a protective effect against paracetamol-induced toxicity.TNF-α is a pro-inflammatory cytokine that is associated with liver injury in many experimental models (Camacho- Barquero et al., 2007;Cover et al., 2006) and human diseases (Felver et al., 1990).In the current study, treatment of rats with paracetamol resulted in a 3.6fold increase in hepatic TNF-α.Previous reports showed that paracetamol-induced liver injury resulted in a significant increase in TNF-α which is in accordance with our results (Cover et al., 2006;Sener et al., 2006;Teng et al., 2012).However, the treatment of paracetamol-intoxicated rats with caffeic acid or silymarin significantly inhibited the rise in TNF-α indicating antiinflammatory activity of the currently studied polyphenols.Preceding studies showed the inhibitory effect of caffeic acid or silymarin on TNF-α expression in different models support our finding (Chao et al., 2010;Song et al., 2006).
It is well established that free radicals play a role in the recruitment of neutrophils by injured tissues as a potential source of free radicals and activating cytotoxic enzymes, namely MPO (Lau et al., 2005), which in turn maintain the vicious cycle of generating more free radicals resulting in lipid peroxidation.The infiltration of polymorphnuclear leukocytes that generates oxygen radicals and induces direct tissue injury has been previously reported with indomethacin treatment (Tanaka et al., 2001).This effect can result in increased lipid peroxidation that may exacerbate the already existing damage.In the present study, the elevated hepatic MDA concentrations along with increased hepatic MPO activity following paracetamol intoxication indicated the contribution of neutrophils infiltration and the impact of free radical generation on oxidative hepatic damage.While the co-treatment with caffeic acid or silymarin abrogated hepatic neutrophils infiltration as was evident by the significant reduction in MPO activity.This was consistent with our histopathological findings that showed normal architecture in the liver tissues from caffeic acid-or silymarin-treated animals and this could be attributed to the known antioxidant and anti-inflammatory activities of both compounds (Jayanthi and Subash, 2010; Lieber, 2004;Muriel, 2009).
It is noteworthy that the HO system abates inflammation through several mechanisms including the suppression of macrophage-infiltration and abrogation of oxidative/inflammatory transcription factors like NF-κB, JNK and activating protein-1 (Ndisang, 2010).HO-1 is stimulated by a wide variety of different stimuli including oxidative and inflammatory insults (Jonas et al., 2003;Mohri et al., 2006), hyperglycemia (Jonas et al., 2003), hypertension (Ndisang et al., 2003), and dyslipidaemia (Landar et al., 2006).Therefore, HO-1 may be considered a sensitive index that is triggered in the onset of pathophysiological changes.However, in most cases the pathophysiological activation of HO-1 results only in a transient or marginal increase of HO-1 that falls below the threshold necessary to activate the downstream signaling components of the HO system (Ndisang et al., 2004;Ndisang et al., 2003) through which the HO system elicits its effects of restoring tissue homeostasis (Farombi and Surh, 2006;Ndisang et al., 2003).Therefore, a more vigorous enhancement of HO-1 would be needed to surmount that threshold, an effect that could be achieved by pharmacological agents capable of inducing HO (Ndisang and Jadhav, 2009).This may explain the results observed in the current study that although hepatic level of HO-1 was induced after paracetamol insult, it wasn't sufficient to subside the inflammatory response elicited by paracetamol reactive metabolites.On the other hand, the co-administration of caffeic acid or silymarin in this study abrogated the paracetamol-induced inflammation and neutrophils infiltration as was evident by significant reduction of hepatic TNF-α levels and MPO activities through the robust enhancement of hepatic HO-1levels and subsequent resume of liver homeostasis.
Fig 2A demonstrates that paracetamol resulted in 4-fold increase in MDA levels in liver cell homogenates indicating lipid peroxidation.This was accompanied by 63% reduction in GSH levels (Fig 2B).In addition, the activities of antioxidant enzymes, catalase and SOD, were depleted in liver following paracetamol treatment as evident by 69% and 47% reduction, respectively (Fig 2C-D).Treatment with caffeic acid or silymarin significantly (P ≤ 0.05) suppressed hepatic lipid peroxidation and prevented the reductions of GSH level and catalase as well as SOD activities resulting from paracetamol administration.

Fig 5 .
Fig 5. Photomicrographs of rat liver (H&E, 400×) from: (A) normal control group showing normal hepatic architecture; (B) paracetamol group showing cytoplasmic vacuolization and ballooning degeneration of hepatocytes; (C and D) acetaminophen plus caffeic acid or silymarin, respectively, showing a histological picture comparable to that of the control group.