basic studies …library.tasmc.org.il/ studies preventionoflivercirrhosisinratsbycurcumin rafael...

BAS IC STUDIES

Preventionof liver cirrhosis in ratsbycurcuminRafael Bruck1,5, Michal Ashkenazi2, Sigal Weiss1, Ilana Goldiner1, Haim Shapiro1,3, Hussein Aeed4,Olga Genina6, Zamir Helpern1,5 and Mark Pines6

1 The Department of Gastroenterology, Tel-Aviv Sourasky Medical Center, Tel-Aviv, Israel

2 The Unit of Clinical Nutrition, the E. Wolfson Medical Center, Holon, Israel

3 Hypnosis, the E. Wolfson Medical Center, Holon, Israel

4 Gastroenterology, the E. Wolfson Medical Center, Holon, Israel

5 The Sackler School of Medicine, Tel-Aviv University, Israel

6 Institute of Animal Science, Agriculture Research Organization, the Volcani Center, Bet Dagan, Israel

Keywords

cirrhosis – curcumin – hepatic stellate cells –

thioacetamide

Abbreviations

ROS, reactive oxygen species; NFkB, nuclear

factor kappa B; TAA, Thioacetamide;

TNF-a, tumor necrosis factor alpha;

TGF, transforming growth factor; HSC,

hepatic stellate cells; MMP, matrix

metalloproteinase.

Correspondence

Rafael Bruck, MD, Deptartment of

Gastroenterology and Liver Diseases, Tel-Aviv

Sourasky Medical Center, 6 Weizman street,

Tel-Aviv 64239, Israel.

Tel: 1972 3 6947302

Fax: 1972 3 6966286

e-mail: [email protected]

Received 15 April 2006

accepted 26 November 2006

DOI:10.1111/j.1478-3231.2007.01453.x

AbstractBackground and Aim: Curcumin, the major polyphenolic compound in turmeric,

has been shown to attenuate hepatic damage in several animal models of liver

injury. The aim of the present study was to examine the efficacy of curcumin in

preventing thioacetamide-induced cirrhosis and to unravel the mechanism of

curcumin’s effect on hepatic fibrosis in rats. Methods: Liver cirrhosis was induced

by thioacetamide (TAA; 200 mg/kg, i.p.) twice weekly for 12 weeks. One group of

rats concomitantly received curcumin (300 mg/kg/day, by gavage for 12 weeks);

the control group received the solvent at identical amounts and duration.

Results: TAA administration induced liver cirrhosis, which was inhibited by

curcumin. Liver histopathology, hydroxyproline levels and spleen weights were

significantly lower in the rats treated with TAA1curcumin compared with TAA

only (Po 0.001). Immunohistochemical studies and in situ hybridization demon-

strated inhibition of hepatic stellate cell (a smooth muscle actin-positive) activa-

tion and collagen aI (I) gene expression in the livers of the TAA1curcumin-treated

rats. Curcumin reduced oxidative stress as shown by the decreased hepatic

nitrotyrosine staining in the curcumin1TAA-treated rats. Curcumin treatment

had no effect on pre existing liver cirrhosis. As determined by in vitro studies using

the rat HSC-T6 cell line, curcumin had no direct inhibitory effect on collagen a1

(I) messenger RNA expression. Further studies in these cells using reverse

transcriptase-polymerase chain reaction demonstrated that curcumin had no

effect on the expression of PDGF-induced TIMP-1 and TIMP-2, TGFb1, TGFb2

and MCP-1 but significantly inhibited tumor necrosis factor alpha expression.

Curcumin had no effect on hepatic stellate cells proliferation. Zymography showed

that curcumin had no effect on matrix metalloproteinase-2 activity. Conclusion-

s: Curcumin inhibited the development of TAA-induced liver cirrhosis mainly due

to its anti-inflammatory activities and not by a direct anti-fibrotic effect. As

curcumin ingestion is safe in humans, it may be reasonable to assess in clinical

studies the beneficial effect of curcumin in slowing the development of liver

cirrhosis.

Certain forms of hepatic injury lead to a chronic,partially self-perpetuating inflammation and an at-tempt at tissue regeneration and wound healing pro-gressing in some patients to liver cirrhosis. Chronictissue injury and inflammation lead to activation ofhepatic stellate cells (HSC), which become contractileand deposit collagen in the space of Disse. This caneventually lead to disruption of the functional struc-

ture of hepatic lobules and to increased resistance toportal blood flow (1). Increased collagen depositionbetween hepatocytes and sinusoids and the diminu-tion of the size of endothelial fenestrae lead to thecapillarization of sinusoids. Constriction of sinusoidsby contractile HSC and capillarization increase resis-tance to blood flow and contribute to the developmentof portal hypertension (1, 2). Complications of portal

Liver International (2007)c� 2007 The Authors. Journal compilation c� 2007 Blackwell Munksgaard 373

Liver International ISSN 1478-3223

hypertension are the main causes of mortality inpatients with cirrhosis (2). Although progress has beenmade in the management of liver cirrhosis and itscomplications, preventive strategies should be benefi-cial in reducing the burden of this disease (1, 3). Therhizome of turmeric (Curcuma longa), a plant belong-ing to the ginger family, is widely used as a foodcolouring and is one of the principal ingredients incurry powder. Turmeric has long been used in Indo-Chinese medicine as an anti-inflammatory to treatdigestive and liver disorders, and for the treatment ofskin diseases and wound healing (4). The main activeingredient in turmeric is curcumin (diferuloyl-methane) (4), administration of which is therapeuticin rodent models of a number of intestinal, pancreaticand liver diseases (5–7). Anti-inflammatory, anti-oxidant and nuclear factor-kB (NF-kB) inhibitingproperties may underlie curcumin’s beneficial effectin these conditions (8). It has been reported thatcurcumin prevents collagen type I formation byactivated HSC (9) and reduces hepatic fibrosisand inflammation in rodent models of steatohe-patitis (10, 11).

Curcumin ingestion is safe in humans, and issufficiently bioavailable to produce beneficial systemicand hepatic effects (12). Utilizing the chronic admin-istration of thioacetamide (TAA) to rats as a model,our aim in this study is to unravel the mechanism ofcurcumin’s effect on hepatic fibrosis: is it a direct effecton HSC and collagen formation or an indirect effect bythe prevention of inflammation and necrosis

Materials and methods

Materials

TAA, curcumin, cremophore and glycerol formal wereobtained from Sigma Chemicals Co. (St. Louis, MO).Curcumin was dissolved in glycerol formal, cremohoreand water (5:2:2).

A 1600- bp rat collagen a1 (I) probe was a generousgift from B.E.Kream, University of Connecticut, Farm-ington, CT. Smooth muscle actin monoclonal antibo-dies, used at 1:1200 dilution, were obtained from DakoA/S (Glostrup, Denmark). As a second antibody, theHistomouse SP kit was used (Zymed Laboratories Inc,South San Francisco, CA).

Animals

Male Wistar rats (250–300 g), obtained from Tel AvivUniversity Animal Breeding Center, were fed a Purinachow ad libitum. Animals were kept with a 12 hlight–dark cycle at constant temperature and humid-

ity, and the rats had free access to tap water during thestudy period. Animals received humane care and weretreated according to institutional guidelines.

Experimental design

Prevention of cirrhosis

For the induction of liver cirrhosis, rats were givenintraperitoneal (i. p.) injections of TAA, 200 mg/kg,twice a week for 12 weeks, as previously described(13). Two groups of six rats each were exposed to (a)bi-weekly i.p. TAA (200 mg/kg) and daily i.g. solventadministration for 12 weeks; (b) bi-weekly i.p. TAA(200 mg/kg) and daily intragingival curcumin(300 mg/kg, dissolved in solvent as described aboveand given in a final volume of 2 ml per rat) adminis-tration for 12 weeks. A third group received only thesolvent and served as control.

Treatment of cirrhosis

This experiment was performed to determine whethercurcumin would have a beneficial effect on establishedliver cirrhosis. Such an effect may support a mechan-ism that includes direct anti-fibrotic or fibrinolyticeffects. Five groups of rats were studied (a) TAA 12weeks, (b) and (c) TAA 12 weeks followed for another4 or 6 weeks by 300 mg/kg/day curcumin given bygavage after TAA discontinuation, (d) and (e) TAA 12weeks followed by solvent administration (2 ml/day)for 4 or 6 weeks after TAA discontinuation.

At the end of the study, the rats were sacrificed, theirlivers were removed and the spleen weights weremeasured.

Analysis of liver histopathology

Midsections of the left lobes of the livers were pro-cessed for light microscopy. This processing consistedof fixing the specimens in a 5% neutral formolsolution, embedding the specimens in paraffin, mak-ing 5 mm thick sections and staining the sections withhematoxylin & eosin. The tissue slices were scannedand scored blindly by two experts. The degree offibrosis, classified on a scale of 0–3 according to Mulleret al.(14), was expressed as the mean of 10 differentfields in each slide. Pathological alterations consistentwith fibrosis, nodule formation and the presence offibrotic septa were included in the fibrosis score (0–3).Inflammatory infiltration, hepatocyte apoptosis andbreaking up of the hepatocellular limiting plates in theportal tracts consisted of the inflammation score(0–3).

Liver International (2007)374 c� 2007 The Authors. Journal compilation c� 2007 Blackwell Munksgaard

Prevention of liver cirrhosis in rats by Curcumin Bruck et al.

Measurement of hepatic hydroxyproline levels

Quantitative determination of hepatic hydroxyprolinecontent was performed as previously described (15).

Assessment of hepatic nitrotyrosine

Nitrotyrosine, the product of protein oxidation, wasassessed in paraffin blocks of the different treatmentgroups. Sections were stained with mouse antinitro-tyrosine (Zymed Laboratories, San Francisco, CA) at1:250 dilution for 1 h. As second antibodies, we usedUniversal HRP-polymer kit (BioCare Medical, Con-cord, CA) according to the manufacturer instructions.

Preparation of sections, in situ hybridization andimmunohistochemistry

Liver samples were collected into phosphate-bufferedsaline (PBS) and fixed overnight in 4% paraformalde-hyde in PBS at 4 1C. Serial 5 mm sections were preparedafter the samples had been dehydrated in gradedethanol solutions, cleared in chloroform and em-bedded in Paraplast. Differential staining of collage-nous and non-collagenous proteins was performedwith 0.1% sirius red and 0.1% fast green as a counterstain, in saturated picric acid. By this procedure,collagen is stained red (16). For hybridization, thesections were deparaffinized in xylene, rehydratedthrough a graded series of ethanol solutions, rinsed indistilled water (5 min) and incubated in 2� SSC at70 1C for 30 min. The sections were then rinsed indistilled water and treated with pronase (0.125 mg/mlin 50 mM Tris-HCl, 5 mM ethylenediaminetetraaceticacid, pH 7.5) for 10 min. After digestion, the slideswere rinsed with distilled water, postfixed in 10%formalin in PBS, blocked in 0.2% glycine, rinsed indistilled water, rapidly dehydrated through gradedethanol solutions, and air dried for several hours.Before hybridization, the 1600 bp rat collagen a1(I)insert was cut out from the original plasmid (pUC18)and inserted into Safyre. The sections were thenhybridized with digoxigenin-labeled collagen a1(I)probe (17). For immunohistochemistry, smooth mus-cle actin antibodies were used, and the detection wasperformed with the Histomouse SP kit according tothe manufacturer’s instructions.

Effect of curcumin on collagen a1(I) gene expression

SV40-immortalized rat HSC-T6 cell line (18), gener-ously provided by Dr. S. L. Friedman (Mount SinaiMedical Center, NY), were incubated with curcumin(5 or 10mg/ml) for 24 h. Collagen a1(I) gene expres-

sion was determined by Northern blot analysis usingthe collagen a1(I) probe.

Northern blots and probes

Total RNA from HSC-T6 cell line was isolated with TRIreagent (Molecular Research Center Inc., Cincinnati,OH) according to the manufacturer’s instructions. ForNorthern blots, 10mg of total RNA was resolved underdenaturating conditions on 1.5% agarose/formalde-hyde gel and transferred onto Nytran N nylon mem-branes by capillary transfer. The membranes were bakedat 80 1C for 2 h and prehybridized at 68 1C for 2 h withEZ-hybridization buffer. Following prehybridization,32P-labeled cDNA probe were added into the hybridiza-tion solution and hybridized with the membranes at68 1C overnight. Membranes were then washed twicewith 2� SSC 0.1% sodium dodecyl sulphate (SDS) at42 1C for 15 min and twice with 0.1� SSC/0.1% SDSat 65 1C for 30 min and for 1.5 h. Autoradiography wasperformed on Kodak X-ray films with intensifyingscreens at � 70 1C. The probes used for the Northernblot hybridization were generated by reverse transcrip-tase-polymerase chain reaction (RT-PCR) amplifica-tion using the following primers pairs for humancollagen type I: 50-AGCAGAAACATCGGATTTGG-30,50-GGAGGaGGGTTTCAGAGGAG-30. The 18S and28S ribosomal RNA were used as control for thequantity of the RNA loaded in each lane.

Cell culture conditions

The immortalized rat hepatic stellate cell line HSC-T6were grown in Dulbecco’s modified Eagle medium(DMEM; Biological Industries, Bet-Haemek, Israel)supplemented with 10% fetal calf serum (FCS) (v/v),2 mM L-glutamine, and 100 mg/ml of penicillin–strep-tomycin. 24 h later the medium was changed tohormonally defined medium (HDM). HDM consistedof DMEM supplemented with 100 mg/ml penicillin–-streptomycin, 2 mg/ml bovine serum albumin,610 mg/ml nicotinamide, 740 ng/ml ZnSO4x7H2O,20 ng/ml CuSO4x5H2O, 5 mM glutamine, 5 mg/mlinsulin, 5 mg/ml iron-saturated transferrin, 5 ng/mlselenous acid and 10�7 M dexamethasone (all materi-als were obtained from Sigma).

Proliferation assay

Rat HSC were incubated in a 24-well plate(30 000 cells/well). The cells grew in DMEM culturemedium supplemented with 10% FCS until the loga-rithmic growth phase, and then incubated for 48 h inHDM supplemented with or without curcumin (0.5,


Bruck et al. Prevention of liver cirrhosis in rats by Curcumin

1, 5, 10 or 15 mM). After 48 h, cells were trypsinizedand counted using hemocytometer.

Gelatin zymography assay

Rat HSC were plated on six-well (80 000 cells per well).After 24 h in DMEM 10% FCS media was replacedwith HDM supplemented with or without 5 mMcurcumin. After 48 h the supernatant was collectedand cell number was evaluated using hemocytometer.The supernatant was mixed with nonreducing samplebuffer (0.4 M Tris, pH. 6.8%, 5% SDS, 20% glycerol,0.03% bromophenol blue). Proteins were separated byelectrophoresis through 8% polyacrylamide gel-con-taining gelatin (25mg/ml). Following electrophoresis,gels were washed in 2.5% Triton X-100 with gentleshaking for 30 min, and then incubated for 30 min indeveloping buffer (40 mM Tris-HCl, 0.2 mM NaCl,6.67 mM CaCl2, 0.1% Triton X-100, pH. 7.8). Thedeveloping buffer was replaced with fresh buffer andthe gel incubated overnight at 37 1C. The gels werestained with 0.5% Coomasie blue G-250 in 10% aceticacid and 40% methanol and destained in water. MMP-2 activity was visualized as bands of lysis, which appearwhite on dark background.

RNA extraction and RT-PCR

Total RNA was extracted from rat hepatic stellate cellline (HSC-T6) using the EZ-RNA kit. Its quantity andquality was assessed by measuring the optical densityat 260 and 280 nm. cDNA was synthesized from �2 mgtotal RNA using the M-MLV Reverse Transcriptase kit(Promega, Madison, WI, USA), according to manu-facturer’s instructions.

The cDNAs were amplified by PCR using REDTaqDNA polymerase. Primers were designed to detectTNF-a sense 50-ATGAGCACAGAAAGCATGATC-30,antisense 50-CAGAGCAATGACTCCAAAGTA-30), TGF-b1 (sense 50-TGGCGTTACCTTGGTAACC-30, antisense50-GGTGTTGAGCCCTTTCCAG-30), TGF-b2 (sense-50

ATCGATGGCACCTCCACATATG-30, antisense 50- GC-GAAGGCAGCAATTATGCTG-30), TIMP-1 (sense 50-ACAGCTTTCTGCAACTCG-30, antisense 50-CTATA-

GGTCTTTACGAAGGCC- 30), TIMP-2 (sense 50-ATT-TATCTACACGGCCCC-30, antisense 50- CAAGAACCA-TCACTTCTCTTG-30) and MCP-1 (sense 50- CCT-GTTGTTCACAGTTGCTGCC-30, antisense 50- TCTA-CAGAAGTGCTTGAGGTGGTTG-30). m-RNA expres-sion levels were normalized to the expression ofglyceraldehyde-3-phosphatedehydrogenase (GAPDH)(sense 50- ACCACAGTCCATGCCATCAC-30. The condi-tions for amplification were 5 min at 94 1C, 30 cycles of30 s at 94 1C, 30 s at 56 1C and 30 s at 72 1C, followed byextension for 5 min at 72 1C. PCR products were resolvedby electrophoresis through 1% agarose gels.

Statistical analysis

Results of in vivo studies are presented as themean� SD. The significance of differences amongdifferent groups was determined by ANOVA andfollowed by a post hoc test.

The data of the in vitro studies are presented asmean� SE. Experiments were replicated at least threetimes. Statistical analysis was performed with unpairedtwo-tailed Student’s t-test. P-values less than 0.05 wereconsidered significant.

Results

Liver histopathology

Histopathologic examination of liver specimens, fixedafter 12 weeks of treatment, showed significantly lessnodule formation, fibrotic septa (Table 1 and Fig. 1,Po 0.001), inflammatory infiltration and hepatocytenecrosis and apoptosis (Fig. 2, Po 0.01) in the liversof TAA1curcumin-treated rats relative to TAA only.

Quantitative analysis of hydroxyproline

Liver fibrosis was quantitated by the measurement ofhepatic levels of hydroxyproline. The mean hydro-xyproline levels of the TAA-treated group weresignificantly higher than those of both groups of ratsco-treated with curcumin (11.7� 3.1 vs. 4.4� 0.6,Po 0.001, Table 1).

Table 1. Effect of curcumin on prevention of TAA-induced liver cirrhosis

Treatment Hydroxyproline (mg/g protein) Spleen weight (g) Fibrosis score (0–3)w Inflammation (0–3)w

TAA1vehicle 11.7� 3.1 1590� 123 2.7� 0.7 2.2� 0.7TAA1curcumin 4.4� 0.6� 970� 118� 0.5� 0.3� 0.4� 0.3�

Vehicle only 1.8� 0.5� 870� 126� 0.0 0.0

Means� SD, n = 6 in each group,�Po 0.001 compared with TAA only. TAA, thioacetamide, 200 mg/kg i.p. twice weekly for 12 weeks. Curcumin, 300 mg/kg, five times/week.

wA scale of 0–3: no change, 0; slight changes, 1; stronger changes, 2; and intense changes, 3.



Spleen weight

Characteristic hemodynamic changes of liver cirrhosishave previously been shown after 3 months of TAAadministration, including portal hypertension andhyperdynamic circulation that are accompanied by asignificant increase in spleen weight (13). In our study,after 12 weeks of treatment, the mean spleen weightof rats receiving TAA1curcumin was 39% lowercompared with those receiving TAA only (Po 0.001,Table 1).

Effect of curcumin on the regression of pre-established cirrhosis

This experiment was performed to determine whethercurcumin would have a beneficial effect on establishedliver cirrhosis. Such an effect may support a mechan-ism that includes direct anti-fibrotic or fibrinolyticeffects. As shown in Table 2, curcumin administrationfor 4 or 6 weeks to rats that already received TAA for12 weeks had no effect on liver histology, hepatichydroxyproline levels or the spleen weights, indicatingthat curcumin has no effect on pre-established livercirrhosis.

Hepatic content of nitrotyrosine

As shown in Fig. 3, nitrotyrosine staining was in-creased in the livers of the rats treated with TAA for12 weeks and significantly reduced in the livers ofthose treated concomitantly with curcumin. This con-

firms that curcumin treatment reduced oxidative stressin the livers of the TAA-treated rats, and that this effectcould be observed not only after acute TAA adminis-tration, as shown in our previous stuids (19), but alsoafter prolonged treatment with TAA that causes livercirrhosis.

Activation of HSC (a smooth muscle actin) andcollagen a1 (I) gene expression in vivo

Liver sections of the control rats were devoid ofcollagen. In the control livers, when stained with asmooth muscle actin antibodies, no stellate cells weredetected, suggesting that the latter were in theirquiescent state. No cells expressing the collagen a1(I)gene were detected by in situ hybridization (Fig. 1) orimmunohistochemistry (not shown). When treatedwith TAA, the livers exhibited a marked increase inextracellular matrix (ECM) content and displayedbundles of collagen surrounding the lobules, whichresulted in large fibrous septa and distorted tissuearchitecture. These septa were populated by smoothmuscle actin-positive cells expressing high levels of thecollagen a1(I) gene, all of which are characteristic ofadvanced fibrosis and cirrhosis. Cotreatment withcurcumin prevented the activation of most of theHSC and only traces of smooth muscle-positive cellswere detected. The remaining HSC expressed lowlevels of collagen a1(I) gene that resulted in low levelsof collagen as determined by Sirius red staining(Fig. 1).

Fig. 1. Histologic analysis of liver sections. Liver samples were taken from rats treated with TAA (200 mg/kg twice weekly), orTAA1curcumin (300 mg/kg/day). The sections were stained with hematoxylin and eosin, and with Sirius red for collagen. ActivatedHSC were detected by immunohistochemistry with monoclonal antibodies for a smooth muscle actin. Collagen a1(I) gene expressionwas evaluated by in situ hybridization. Note the increase in the collagen deposition shown by Sirius red staining, in smooth muscleactin-positive stellate cells, and in collagen a1(I) gene expression in the livers of the rats treated only with TAA. These increases werecompletely abolished in the rats treated with TAA1curcumin.



Effect of curcumin on collagen a1(I) gene expressionin hepatic stellate cells

Incubation of HSC-T6 cells with curcumin (5 or10 mg/ml) for 24 h did not decrease collagen a1(I) geneexpression as determined by Northern blot analysisusing the collagen a1(I) probe (Fig. 4).

Effect of curcumin on cytokine expression in HSCin vitro

We assessed whether curcumin influenced cytokineexpression in HSC-T6 cells. Total m-RNA was ex-tracted from HSC-T6 cells 24 h following incubationin HDM supplemented with or without 5 mM curcu-min. We have shown using RT-PCR that curcuminsignificantly inhibited HSC expression of TNF-a, butnot that of TIMP-1, TIMP-2, TGFb1, TGFb2 andMCP-1 in HSC-T6 cells following stimulation byplatelet-derived growth factor (PDGF) (Fig. 5). Cur-cumin did not cause any significant change in cytokineexpression in unstimulated cells and had no effect onHSC proliferation (Fig. 6).

Effect of curcumin on MMP-2 activity in HSC-T6cells

We examined the effect of curcumin on MMP-2activity using gelatine zymography assay. Culturedhepatic stellate cells were incubated in HDM media inthe presence or absence of 5 mM curcumin for 24 h.There was no difference in MMP-2 activity in HSC-T6cells in the presence of curcumin (Fig. 7).

Fig. 2. Histologic analysis of liver sections. Liver samples weretaken from rats treated with TAA (200mg/kg twice weekly), orTAA1curcumin (300mg/kg/day). (A) Normal; (B) Liver sectionsshowed increased inflammatory cell infiltration (polymorphonuclearand mononuclear leukocytes) and hepatocytes necrosis andapoptosis in the rats treated with TAA for 12 weeks. Thisinflammatory infiltration and hepatocyte apoptosis was significantlyattenuated in the livers of the rats treated with TAA1curcumin (C),indicating that anti inflammatory activity of curcumin significantlycontributes to the prevention of liver cirrhosis in this model.

Table 2. No effect of curcumin on regression of pre-estab-lished cirrhosis

Treatment

Hydroxy-proline(mg/g protein)

Spleenweight(g)

Fibrosisscore(0–3)�

TAA 12 weeks 10.6� 0.6 1470� 140 2.5� 0.5TAA 12 weekscurcumin 4 weeks

10.8� 0.4 1510� 125 2.5� 0.5

TAA 12 weeks1

Vehicle 4 weeks11.1� 0.6 1520� 100 2.5� 0.4

TAA 12 weeks1

Curcumin 6 weeks11.4� 0.7 1420� 230 2.6� 0.5

TAA 12 weeks1

Vehicle 6 weeks11.0� 0.8 1370� 180 2.7� 0.6

Means� SD, n = 9 in each group. TAA, thioacetamide, 200 mg/kg i.p.

twice weekly for 12 weeks. Curcumin (300 mg/kg, five times per

week) administered for either 4 or 6 weeks, after 12 weeks of TAA

administration.�A scale of 0–3: no change, 0; slight changes, 1; stronger changes, 2;

and intense changes, 3.



Discussion

TAA, although not toxic itself, is metabolized intopotent hepatotoxins by hepatic cytochromes (20).These can produce liver injury by the formation ofhighly reactive compounds (21, 22) and possibly alsoby activating NFkB (23). The form of hepatic damageproduced by TAA depends on the dosage and durationof its administration: high doses can cause fulminant

hepatic failure (FHF) and early death (24), whereaslower dose administered over a prolonged period oftime culminates in liver cirrhosis (17). We recentlydemonstrated that pre-treatment with curcumin pro-vided hepatoprotection in a rodent model of TAA-induced FHF (19). In that study, curcumin inhibitedlipoperoxidation and activated NFkB and iNOS ex-pression in the liver. Survival rates were higher incurcumin-treated rats (19). Curcumin does not

Control

TAA

TAA+

curcumin

x40 x200

Fig. 3. Effect of curcumin on hepatic nitrotyrosine. In the control, untreated rats very low level of nitrotyrosine was observed. Livers ofrats treated with TAA exhibited a high number of stellate cells of which many exhibited nitrotyrose. After curcumin treatment amarked decrease in nitrotyrosine content was observed.



significantly alter the activity of hepatic CYP2E1 (25),the major isoform involved in the formation of toxicTAA metabolites (26, 27). Its beneficial effects in thepresent and latter study can therefore be attributed tothe amelioration of TAA metabolites’ toxicity, ratherthan their formation. The objective of the presentstudy was to assess whether dietary curcumin couldattenuate the development of full-blown cirrhosis andportal hypertension (manifested as splenomegaly) inrats due to chronic TAA administration, and to furtherexplore the mechanisms responsible for curcumin’santi-fibrotic effect. Rats that received bi-weekly, i.p.injections of TAA for 12 weeks developed hepatic

15µM Curcumin

25ng/ml PDGF 25ng/ml PDGF

GA

PDH

GA

PDH

TGF-β2

TGF-β1

TGF-β2

TGF-β1

TNF-α

1.4

1.2

1

0.8

0.6

0.4

0.2

0

fold

incr

ease

control

curcumin

TMC

P1

MC

P1

TIMP1

TIMP2

TIMP1

TIMP2

TGF-β2

TGF-β1

MC

P

TIMP1

TIMP2

TNF-α

TNF-α

control

cur

controlcurcumin

A

B

Fig. 5. (A) Curcumin inhibits TNFa expression by stimulated HSC-T6 cell line. Total RNA was extracted from rat hepatic stellate cell line(HSC-T6) incubated in hormonally defined medium supplemented with 25 ng/ml platelet-derived growth factor with or withoutcurcumin (5 or 15 mM). cDNA was synthesized from �2mg total RNA and amplified using primers designed to detect TNF-a, TGF-b1,TGF-b2, TIMP-1, TIMP-2 and MCP-1. m-RNA expression levels were normalized to the expression of GAPDH. Results represent theexpression level in curcumin treated cells compared with the expression level in untreated cells. (B) Quantitation of the bandscorresponding to the effect of curcumin on cytokine expression in HSC in vitro was performed by densitomety. Results represent theexpression level in curcumin-treated cells compared with the expression level in untreated cells. The graph shows representative resultsfrom at least 3 independent experiments (means� SE, �P valueo 0.05).

Fig. 4. Effect of curcumin on collagen a1(I) gene expression inHSC in vitro. Incubation of HSC-T6 cells with curcumin (5 or10mg/ml = 15 or 30 mM) for 24 h had no effect on collagen a1(I)gene expression as determined by Northern blot analysis usingthe collagen a1(I) probe. The 18S and 28S ribosomal RNA wereused as control for the quantity of the RNA loaded in each lane.



cirrhosis, manifested by gross macroscopic appear-ance, liver histopathology, hepatic hydroxyprolinecontent and increased spleen weight. Their livers alsorobustly exhibited the primary components of HSCactivation, i.e. increased expression of a smooth

muscle actin and type I collagen gene. In these in vivostudies, these indices and mediators of fibrosis andcirrhosis were significantly lower in the rats that wereadministered curcumin concomitantly with TAA. Infact, the pathological changes were virtually preventedby curcumin.

In in vitro studies, collagen a1(I) mRNA levels werenot influenced by incubation with curcumin. Theinhibition of collagen a1(I) expression shown in vivoprobably represents an effect on early events leading toactivation of collagen a1(I) expression such as necro-inflammation and oxidative stress. We also showedusing RT-PCR that curcumin inhibited HSC expres-sion of TNF-a, but not that of TIMP-1 TIMP- 2,TGFb1, TGFb2 and MCP-1 in unstimulated stateand following stimulation with PDGF, and had noeffect on HSC proliferations. Thus, curcumin didnot inhibit HSC proliferation or expression of profi-brogenic genes, despite its NFkB-inhibitory proper-ties (10). Failure of targeted NFkB inhibition toprevent phenotype-related gene expression or induceapoptosis in activated HSC has previously been re-ported (28).

Why does the anti-inflammatory effects of curcu-min inhibit hepatic fibrogenesis? A complex interplaybetween resident and infiltrating cell types takes placeduring the development of fibrosis and cirrhosis of theliver, but activation of HSC is invariably the finalcommon pathway (29, 30). The metabolically activehepatocytes are the main target of toxins such as TAAand carbon tetrachloride (30, 31) that lead to hepato-cyte dysfunction and release of reactive oxygen species(ROS), inflammatory and fibrogenic mediators, aswell as to necrosis and apoptosis. Circulating whiteblood cells are recruited and activated, and, in cohortwith Kupffer cells secrete growth factors that activateHSC, as well as other components of the pro-inflam-matory and fibrogenic cascade. In addition, apoptotichepatocytes activate HSC, and also upregulate theinflammatory state apart from increasing the produc-tion of ECM proteins. Thus, a vicious cycle in whichinflammatory and fibrogenic cells stimulate each otheris likely to occurs (1–3, 29–31).

Strategies for reducing the burden of cirrhosisinclude removal of the injurious stimulus, limitinghepatic injury and inflammation, and inhibition/re-versal of HSC activation and fibrosis (1). For instance,agents that reduce oxidative stress and inhibit theactivation of NFkB are effective in attenuating TAA-induced cirrhosis (21–23). Although not assessed inthis study of a chronic model, curcumin does mini-mize in vivo liver NF-kB activation and oxidative stressin acute TAA toxicity (19), as well as in rodent models

0

20

40

60

80

100

120

140

160

180

1510510.5DMSONone

Curcumin (µM)

Cel

l nu

mb

er (%)

Fig. 6. No effect of curcumin on HSC proliferation. HSC-T6 cellswere incubated in a 24-well plate (30 000 cells/well). The cellsgrew in Dulbecco’s modified eagle medium culture mediumsupplemented with 10% fetal calf serum until the logarithmicgrowth phase, and then incubated for 48 h in hormonallydefined medium supplemented with or without curcumin (0.5,1, 5, 10 or 15 mM). After 48 h cells were trypsinized and countedusing hemocytometer. The graph shows representative resultsfrom 3 independent experiments (means� SE).

140.00

120.00

100.00

80.00

60.00

40.00

20.00

0.005µM cur

MM

P2

acti

vity

(%

)

No cur

Fig. 7. No effect of curcumin matrix metalloproteinase (MMP) –2 activity in HSC-T6 cells in vitro. HSC-T6 cells were incubated inhormonally defined medium media supplemented with orwithout 5mM curcumin. After 24 h supernatant was collectedand the number of cells was evaluated using hemocytometer.Proteins were separated using sodium dodecyl sulfate-polyacrylamide gel electrophoresis gels and MMP-2 activity wasevaluated. MMP-2 activity is represented as percentage fromMMP-2 activity in control cells (no curcumin). The graph showsrepresentative results from three independent experiments(means� SE).



of steatohepatitis (10, 11). It has been shown thatpretreatment with curcumin inhibited NF-kB bindingactivity and TNF-a expression in Kupffer cells exposedto endotoxin, as well as in vivo in livers of rats-fed withtoxic doses of ethanol and fish oil (10). These geneproducts promote inflammation, tissue injury andHSC activation (1, 29–32).

To what can we attribute curcumin’s protectiveeffect against hepatic cirrhosis in our study? It isreasonable to suggest that inhibition of NFkB inKupffer and recruited inflammatory cells in curcu-min-treated rats may have contributed to the reducedtissue injury and activation of HSC. We have shown ina previous study using a model of acute TAA-hepato-toxicity that curcumin inhibited the formation ofhepatic lipo-peroxides in rats acutely treated withTAA, as evident by reduced TBARS level, indicatingan attenuation of oxidative stress (25). Curcumin alsoreduced hepatic lipoperoxide formation in both acuteand chronic carbon tetrachloride injury (6, 33). It hasbeen suggested that curcumin inhibits formation ofROS and enhances endogenous anti-oxidant activitybeyond its free radical scavenging property (8). In theliver, curcumin induces and/or activates a number ofhepatic antioxidant enzymes, including catalase,superoxide dismutase and the glutathione system (8,34, 35). As ROS and lipoperoxides are potent inducersof HSC activation (30), curcumin’s anti-oxidativeproperties may have attenuated HSC-mediated fibro-genesis in this study. Therefore, it appears that itsprotective role in chronic TAA administration is invirtue of its anti-inflammatory and possibly its anti-oxidative capacity. By attenuating the vicious cycle ofcell injury and production of ROS, chemokines, cyto-kines and profibrotic molecules, curcumin may bothprotect hepatocytes from demise and dysfunction(19), and also prevent the formation of the moleculesinvolved in transforming normally quiescent HSC intoa proliferative, fibrogenic and contractile myofibro-blast. The resultant development of cirrhosis andportal hypertension is thus prevented, as clearly shownin this study. This is supported by the attenuation ofoxidative stress after acute TAA administration, asshown by decreased malondialdehyde (MDA) forma-tion in the curcumin-treated rats after 48 h in ourprevious studies (19). In the present study of chronicTAA administration we used nitrotyrosine, a productof protein oxidation, and observed a marked decreasein nitrotyrosine staining in the hepatocytes and HSCof the TAA1curcumin-treated rats compared withTAA only. These results confirm that the inhibition ofoxidative stress in the liver by curcumin could bedemonstrated not only after acute TAA administration

as shown previously (19), but also after prolongedtreatment with TAA that causes liver cirrhosis. Hence,curcumin’s protective effect against liver fibrosis maybe associated with its ability to inhibit NF-kB inKupffer cells and infiltrating macrophages, and thusprevent hepatic inflammation, necrosis and apoptosis.This is represented by the significant inhibition (inaddition to hepatic fibrosis) of inflammatory infiltra-tion and hepatocytes apoptosis in the curcumin-compared with the vehicle-treated rats. To furtherexplore potential anti-fibrotic actions of curcumin,we examined whether curcumin would have a bene-ficial effect on established liver cirrhosis. Such an effectmay support a mechanism that includes direct anti-fibrotic or fibrinolytic activity. However, curcuminadministration for either 4 or 6 weeks to rats thatalready received TAA for 12 weeks and were confirmedto be cirrhotic did not improve liver histology, hepatichydroxyproline levels or the spleen weights, indicatingthat curcumin had no effect on pre-established livercirrhosis. These data are consistent with the results inHSC that showed no effect of curcumin on MMP-2activity. The failure of curcumin to inhibit HSCproliferation and profibrogenic activity in vitro, ob-served in our study, is not in accordance with aprevious study (9). The reason for this discrepancy isnot apparent, apart from the use of different HSC cellline.

In summary, curcumin treatment prevented thedevelopment of hepatic cirrhosis in rats that wereadministered the hepatotoxin TAA. As curcumin issafe for consumption by humans, it may have abeneficial role in chronic liver diseases characterizedby ongoing hepatic necroinflammation.

References

1. Iredale JP. Cirrhosis: new research provides a basis for

rational and targeted treatments. BMJ 2003; 327(7407):

143–7.

2. McGuire BM, Bloomer JR. Complications of cirrhosis. Why

they occur and what to do about them. Postgrad Med 1998;

103(2): 209–12, 217–8, 223–4.

3. Karsan HA, Rojter SE, Saab S. Primary prevention of

cirrhosis. Public health strategies that can make a difference.

Postgrad Med 2004; 115(1): 25–30.

4. Grant KL, Schneider CD. Turmeric. Am J Health Syst Pharm

2000; 57(12): 1121–2.

5. Gukovsky I, Reyes CN, Vaquero EC, Gukovskaya AS, Pandol

SJ. Curcumin ameliorates ethanol and nonethanol experi-

mental pancreatitis. Am J Physiol Gastrointest Liver Physiol

2003; 284(1): G85–95.



6. Park EJ, Jeon CH, Ko G, Kim J, Sohn DH. Protective effect of

curcumin in rat liver injury induced by carbon tetrachloride.

J Pharm Pharmacol 2000; 52(4): 437–40.

7. Ukil A, Maity S, Karmakar S, Datta N, Vedasiromoni JR, Das

PK. Curcumin, the major component of food flavour tur-

meric, reduces mucosal injury in trinitrobenzene sulphonic

acid-induced colitis. Br J Pharmacol 2003; 139(2): 209–18.

8. Joe B, Vijaykumar M, Lokesh BR. Biological properties of

curcumin-cellular and molecular mechanisms of action. Crit

Rev Food Sci Nutr 2004; 44(2): 97–111.

9. Kang HC, Nan JX, Park PH, et al. Curcumin inhibits collagen

synthesis and hepatic stellate cell activation in-vivo and in-

vitro. J Pharm Pharmacol 2002; 54(1): 119–26.

10. Nanji AA, Jokelainen K, Tipoe GL, Rahemtulla A, Thomas P,

Dannenberg AJ. Curcumin prevents alcohol-induced liver

disease in rats by inhibiting the expression of NF-kappa B-

dependent genes. Am J Physiol Gastrointest Liver Physiol 2003;

284(2): G321–7.

11. Leclercq IA, Farrell GC, Sempoux C, dela Pena A, Horsmans

Y. Curcumin inhibits NF-kappaB activation and reduces the

severity of experimental steatohepatitis in mice. J Hepatol

2004; 41(6): 926–34.

12. Sharma RA, Euden SA, Platton SL, et al. Phase I clinical trial

of oral curcumin: biomarkers of systemic activity and

compliance. Clin Cancer Res 2004 Oct 15; 10(20): 6847–54.

13. Hori N, Okanoue T, Sawa Y, Mori T, Kashima K. Hemody-

namic characterization in experimental liver cirrhosis in-

duced by thioacetamide administration. Digest Dis Sci 1993;

38: 2195–202.

14. Muller A, Machnik F, Zimmermann T, Schubert H. Thioace-

tamide-induced cirrhosis-like lesions in rats – usefulness and

reliability of this animal model. Exp Pathol 1988; 34:

229–36.

15. Woessner JF. The determination of hydroxyproline in tissue

and protein samples containing small proportions of this

imino acid. Arch Biochem Biophys 1961; 93: 440–7.

16. Gascon-Barre M, Huet PM, Belgiorno J, Plourde V, Coulomb

PA. Estimation of collagen content of liver specimens. Varia-

tion among animals and among hepatic lobes in cirrhotic

rats. J Histochem Cytochem 1989; 37: 377–81.

17. Pines M, Knopov V, Genina O, Lavelin I, Nagler A. Halofu-

ginone, a specific inhibitor of collagen type I synthesis,

prevents dimethylnitrosamine-induced liver cirrhosis. J He-

patol 1997; 27: 391–8.

18. Nieto N, Friedman SL, Greenwel P, Cederbaum AI. CYP2E1-

mediated oxidative stress induces collagen type I expression

in rat hepatic stellate cells. Hepatology 1999; 30(4): 987–96.

19. Shapiro H, Ashkenazi M, Weizmann N, Shahmurov M, Aeed

H, Bruck R. Curcumin ameliorates acute thioacetamide-

induced hepatotoxicity. J Gastro Hepatol 2005 (in print).

20. Hunter AL, Holscher MA, Neal RA. Thioacetamide-induced

hepatic necrosis. I. Involvement of the mixed-function oxi-

dase enzyme system. J Pharmacol Exp Ther 1977; 200(2):

439–48.

21. Balkan J, Dogru-Abbasoglu S, Kanbagli O, Cevikbas U,

Aykac-Toker G, Uysal M. Taurine has a protective effect

against thioacetamide-induced liver cirrhosis by decreasing

oxidative stress. Hum Exp Toxicol 2001; 20(5): 251–4.

22. Bruck R, Shirin H, Aeed H, et al. Prevention of hepatic

cirrhosis in rats by hydroxyl radical scavengers. J Hepatol

2001; 35(4): 457–64.

23. Bruck R, Schey R, Aeed H, Hochman A, Genina O, Pines M.

A protective effect of pyrrolidine dithiocarbamate in a rat

model of liver cirrhosis. Liver Int 2004; 24(2): 169–76.

24. Bruck R, Aeed H, Schey R, et al. Pyrrolidine dithiocarbamate

protects against thioacetamide-induced fulminant hepatic

failure in rats. J Hepatol 2002; 36(3): 370–7.

25. Oetari S, Sudibyo M, Commandeur JN, Samhoedi R, Ver-

meulen NP. Effects of curcumin on cytochrome P450 and

glutathione S-transferase activities in rat liver. Biochem

Pharmacol 1996; 51(1): 39–45.

26. Ramaiah SK, Apte U, Mehendale HM. Cytochrome P4502E1

induction increases thioacetamide liver injury in diet-re-

stricted rats. Drug Metab Dispos 2001; 29(8): 1088–95.

27. Wang T, Shankar K, Ronis MJ, Mehendale HM. Potentiation

of thioacetamide liver injury in diabetic rats is due to induced

CYP2E1. J Pharmacol Exp Ther 2000; 294(2): 473–9.

28. Lang A, Schoonhoven R, Tuvia S, Brenner DA, Rippe RA.

Nuclear factor kappa B in proliferation, activation, and

apoptosis in rat hepatic stellate cells. J Hepatol 2000; 33(1):

49–58.

29. Punithavathi D, Venkatesan N, Babu M. Protective effects of

curcumin against amiodarone-induced pulmonary fibrosis

in rats. Br Jl Pharmacol 2003; 139: 1342–450.

30. Bataller R, Brenner DA. Liver fibrosis. J Clin Invest 2005;

115(2): 209–18.

31. Higuchi H, Gores GJ. Mechanisms of liver injury: an over-

view. Curr Mol Med 2003; 3(6): 483–90.

32. Mehendale HM. Tissue repair: an important determinant of

final outcome of toxicant-induced injury. Toxicol Pathol

2005; 33(1): 41–51.

33. Kamalakkannan N, Rukkumani R, Varma PS, Viswanathan P,

Rajasekharan KN, Menon VP. Comparative effects of curcu-

min and an analogue of curcumin in carbon tetrachloride-

induced hepatotoxicity in rats. Basic Clin Pharmacol Toxicol

2005; 97(1): 15–21.

34. Sharma RA, Ireson CR, Verschoyle RD, et al. Effects of dietary

curcumin on glutathione S-transferase and malondialde-

hyde-DNA adducts in rat liver and colon mucosa: relation-

ship with drug levels. Clin Cancer Res 2001; 7(5): 1452–8.

35. Iqbal M, Sharma SD, Okazaki Y, Fujisawa M, Okada S.

Dietary supplementation of curcumin enhances antioxidant

and phase II metabolizing enzymes in ddY male mice:

possible role in protection against chemical carcinogenesis

and toxicity. Pharmacol Toxicol 2003; 92(1): 33–8.

36. Buss H, Chan TP, Sluis KB, Domigan NM, Winterbourn C.

Protein carbonyl measurement by a sensitive ELISA method.

Free Rad Biol Med 1997; 23: 361–6.



basic studies …library.tasmc.org.il/ studies preventionoflivercirrhosisinratsbycurcumin rafael...

Documents