L

Gt

GAFa

b

c

a

ARAA

KCGLT

1

scs

it

tN

1d

Digestive and Liver Disease 42 (2010) 571–577

Contents lists available at ScienceDirect

Digestive and Liver Disease

journa l homepage: www.e lsev ier .com/ locate /d ld

iver, Pancreas and Biliary Tract

arlic extract prevents CCl4-induced liver fibrosis in rats: The role of tissueransglutaminase

iuseppe D’Argenioa,∗, Daniela Caterina Amorusoa, Giovanna Mazzonea, Paola Vitaglioneb,ntonietta Romanoa, Maria Teresa Ribeccoa, Maria Rosaria D’Armientoc, Ernesto Mezzac,ilomena Moriscob, Vincenzo Foglianob, Nicola Caporasoa

Gastroenterology Unit, Department of Clinical and Experimental Medicine, Federico II University, Naples, ItalyDepartment of Food Science, Federico II University, Naples, ItalyDepartment of Biomorphological Science, Federico II University, Naples, Italy

r t i c l e i n f o

rticle history:eceived 24 July 2009ccepted 4 November 2009vailable online 8 December 2009

eywords:ystaminearlic extractiver fibrosisissue transglutaminase

a b s t r a c t

Background and aim: Tissue transglutaminase contributes to liver damage in the development of hepaticfibrosis. In a model of neurodegeneration, the therapeutic benefit of cystamine has been partly attributedto its inhibition of transglutaminase activity. Garlic extract contains many compounds structurally relatedto cystamine. We investigated the anti-fibrotic effect of garlic extract and cystamine as specific tissuetransglutaminase inhibitors.Methods: Rat liver fibrosis was induced by intraperitoneal injection of carbon tetrachloride (CCl4) for 7weeks. Cystamine or garlic extract was administrated by daily intraperitoneal injection, starting from theday after the first administration of CCl4. Hepatic function, histology, tissue transglutaminase immunos-taining and image analysis to quantify Red Sirius stained collagen deposition were examined. Reversetranscription-polymerase chain reaction to detect alpha-SMA, IL-1� and tissue transglutaminase expres-sion and Western blot for tissue transglutaminase protein amount were performed. Transglutaminaseactivity was assayed on liver homogenates by a radio-enzymatic method.Results: Transglutaminase activity was increased in CCl4 group and reduced by cystamine and garlicextract (p < 0.05). Treatment with cystamine and garlic extract reduced the liver fibrosis and collagendeposition, particularly in the garlic extract group (p < 0.01). Moreover, the liver damage improved and

serum alanine aminotransferase was decreased (p < 0.05). Tissue transglutaminase immunolocalised withcollagen fibres and is mainly found in the ECM of damaged liver. Alpha-SMA, IL-1�, tissue transglutami-nase mRNA and tissue transglutaminase protein were down-regulated in the cystamine and garlic extractgroups compared to controls.Conclusion: These findings concurrently suggest that transglutaminase may play a pivotal role in the

sis aof livGast

pathogenesis of liver fibrostrategy in the treatment

© 2009 Editrice

. Introduction

Liver fibrosis is a wound-healing response to a variety of chronictimuli such as viral infection (hepatitis B and C), chronic ethanolonsumption, metabolic and toxic diseases and non-alcoholic liver

teatohepatitis [1].

Hepatic fibrosis leads to cirrhosis and various complications ands associated with high morbidity and mortality [2]. Thus, it is essen-ial to develop therapeutic strategies to counteract liver fibrosis [3].

∗ Corresponding author at: Dipartimento di Medicina Clinica e Sperimentale, Gas-roenterologia, Facoltà di Medicina, Università Federico II, Via Pansini, 5, 80131,apoli, Italy. Tel.: +39 081 7462706; fax: +39 081 7462739.

E-mail address: dargenio@unina.it (G. D’Argenio).

590-8658/$36.00 © 2009 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevieroi:10.1016/j.dld.2009.11.002

nd may identify garlic cystamine-like molecules as a potential therapeuticer injury.roenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.

Liver fibrosis is due to an imbalance between synthesis anddegradation of the extracellular matrix (ECM) proteins [4], whoseexcessive deposition alters organ structure and function [5].

A major biomarker for liver fibrosis is tissue transglutaminase(tTG) or type II transglutaminase [6], a calcium-dependent enzymewhich catalyses the formation of �(�-glutamyl) lysine cross-linksor deamination of proteins. These covalent cross-links might deter-mine the oligomerisation of substrate protein(s), which acquirethe peculiar features of resistance to breakage and chemical attack[7].

Various collagen types and other extracellular proteins such asfibronectin, laminin, nidogen and transforming growth factor-� areglutamyl substrates of tTG [8].

A role for tTG has also been proposed in the pathologi-cal accumulation of insoluble protein aggregates that occurs in

Ltd. All rights reserved.

5 nd Liv

nt

[hlff

epw

uh

swC

iatl

2

2

pgw

r4suopg

2

pTo

2

pTtcSg

w(moi5

72 G. D’Argenio et al. / Digestive a

eurodegenerative diseases sharing several pathogenetical fea-ures with liver damage, such as Alzheimer and Huntington [9–11].

Recent studies have shown the ability of cystamine (CYS)�-mercaptoethanolamine disulphide] to ameliorate clinical andistopathological damage in an animal model of Huntington and

iver diseases [12,13] via inhibition of tTG. Cysteamine (the reducedorm of CYS) acts by competitive inhibition of tTG, resulting in theormation of �-(�-l-glutamyl)–cysteamine bonds [14].

The formation of such bonds would compete with the gen-ration of other transglutaminase (TG)-catalysed reactions (i.e.olyamination, protein cross-linking and deamination proteins),hich may contribute to neurodegeneration [12].

Despite its good efficacy as a TG inhibitor, CYS determines manyndesirable side effects such as nausea and trouble walking, whichighly limits the use in humans [11,12,15].

Owing to its content of organosulphur compounds, structurallyimilar to CYS and known to have many biological properties [16],e selected garlic (Allium sativum) as a potential natural source ofYS analogues.

Organosulphur compounds of alliacee are a group of phytochem-cals characterised by antioxidant properties [17]. In this study,

garlic extract (GE) was prepared, chemically characterised andested in vitro and in vivo for its ability to modulate tTG activity andiver fibrosis.

. Materials and methods

.1. Preparation of GE

Peeled garlic (500 g) was freeze dried and reduced to a fineowder in an electric mixer. The powder was transferred in a 1 Llass bottle and added with an equal volume (500 mL) of deionisedater:ethanol 50:50 (v/v).

Headspace was filled with a N2 atmosphere to reduce oxidationeaction of the extract and the bottle let in agitation for 12 h at◦C in the dark. The mixture was filtered on a paper filter and the

olution collected in a 1 L glass balloon. The GE was then evaporatednder vacuum at room temperature until its volume was reduced tone half. Successively the extract was freeze dried, reduced to a fineowder by a mixer and stored at 4 ◦C until the use. This procedureave an extraction yield of 3.4%.

.2. Chemical characterisation of GE

The proximal composition of GE (carbohydrates, dietary fibre,rotein and water) was carried out by official analytical methods.he Folin Ciocalteau method [18] was used for the quantificationf total polyphenols.

.3. Analysis of GE organosulphur compounds

Identification and quantitative determination of the organosul-hur compounds present in GE was carried out by HPLC/MS/MS.o this purpose an API 2000 triple quadrupole mass spectrome-er (Applied Biosystem Sciex), with a Turboionspray (TIS) interface,oupled with an HPLC binary micropumps (Perkin Elmer, USA, mod.eries 200) was used. All the analyses were performed using dryingas (N2) at 300 ◦C.

Chromatographic separations of organosulphur compoundsere performed on a Sphereclone 5� C18 ODS [2] (250 × 4.6 mm)

Phenomenex, Torrance, CA, USA), column using the followingobile phases, water (solvent A) and MeOH (solvent B) and a flow

f 0.5 mL/min. Initial conditions were 90% A–10% B, then the follow-ng gradient elution was applied: 0–15 min 50% A–50% B, 15–20 min0% A–50% B, 20–30 min 90% A–10% B.

er Disease 42 (2010) 571–577

The MS/MS detection was performed acquiring data in positiveion mode. The quantification was carried out in MRM (MultipleReaction Monitoring). Ions corresponding to protonated molecules[M+H]+ for organosulphur compounds were monitored. Molec-ular weights and fragments of all monitored compounds were:m/z 178 → 88, 120 for alliin, 162 m/z → 76 for S-allylcysteine and163 m/z → 88 for allicin, respectively. The ions produced in MS/MSwere obtained through fragmentation by a specific collision energy(13 V) of a selected ion precursor, applying a voltage of 4500 V.

The quantification of compounds was obtained using calibra-tion curves obtained with pure alliin, S-allylcysteine and allicin(Exrasynthèse, Genay, France). The detection limit was 10 ng/g forall the compounds.

2.4. In vitro effect of GE on TG activity

To assess the effect of GE on TG activity, GE was incubated invitro for 30 min either at room temperature or 37 ◦C with normalrat liver homogenate (10%, w/v) as source of TG activity. The opti-mal concentration of GE was calculated to obtain an inhibition ofTG activity of about 80% similar to that obtained with 1 mg/mLCYS.

2.5. Animals

Male Wistar albino rats (200–250 g) were housed in a room at amean constant temperature of 22 ◦C with a 12-h light–dark cycle,and free access to standard pellet chow and water.

The study was approved by the Federico II University School ofVeterinary Medicine Animal Care.

2.6. Experimental model of progressive fibrosis

Liver fibrosis was induced by intraperitoneal injection, twice aweek, of carbon tetrachloride (CCl4) 0.2 mL/100 g in refined oliveoil (1:1) for 7 weeks according to a well-established model [19].

A cohort of animals was injected with CCl4 for 7 weeks, afterwhich animals were killed in a box for anaesthesia under CO2 atmo-sphere and their livers were harvested. Livers were harvested atpeak fibrosis (3 days after the final injection of CCl4). After har-vesting, livers were divided with a minimum of two lobes fixedin formalin for histologic analysis and immunohistochemistry andthe remaining liver was snap-frozen for homogenisation to assayTG activity and to extract total liver RNA and total liver protein.Serum was also collected from each rat to analyse biochemicalparameters.

2.7. Animal treatment and experimental groups

Liver fibrosis was induced for 7 weeks in a set of experiments.Three experimental groups were designed as follows: (1) controlgroup receiving ip injection of vehicle alone (n = 8); (2) rats receiv-ing CYS (150 mg/kg b.w.) daily ip (n = 8); and (3) rats receiving GE(200 mg/kg b.w.) daily ip (n = 8). A group of five normal rats wasalso included in the study.

2.8. Histologic analysis

Ten percent formalin-fixed paraffin-embedded sections of livertissue were divided into 4 �m sections by using routine techniques

and mounted onto slides with coverslips. Representative sectionsof each fixed liver were stained with H&E and Sirius red/Fastgreen, according to standard protocols. All histologic analyses thenwere undertaken by an experienced histopathologist (M.R.D.) in ablinded manner.

nd Liv

2

tdehbcnopifLscbwm

2

swewmem

2

fwAG

2

dtaoi(arciaT(dp

2

s1td

G. D’Argenio et al. / Digestive a

.9. Immunohistochemistry

To detect the immunohistochemical localisation of tTG, sec-ions from formalin-fixed, paraffin-embedded specimens wereeparaffinized and rehydrated in decreasing concentrations ofthyl alcohol. All tissue sections were incubated with fresh 3%ydrogen peroxide for 20 min, and then washed with phosphate-uffered saline (PBS) (0.1 mol/L phosphate and 0.15 mol/L sodiumhloride, pH 7.4). The sections were sequentially incubated in 1%ormal blocking serum for 30 min, and then with mouse mon-clonal anti-tTG antibody (CUB 7402) (NeoMarkers, CA, USA)rimary antibody for 60 min at room temperature. Subsequent

ncubations included appropriate biotinylated secondary antibodyor 60 min and avidin–biotin–peroxidase complex (ABC kit; Vectoraboratories, Burlingame, CA, USA) for 30 min followed by the sub-trate, 3,3′diaminobenzidine (10 �g/mL in Tris–HCl buffer, pH 7.6ontaining 0.03% H2O2), for 4 min. Negative controls were obtainedy omitting the primary antibody. The sections, counterstainedith haematoxylin, were then mounted and observed under lighticroscopy by a blinded pathologist.

.10. Determination of liver collagen content

For the detection and quantification of collagen, sections weretained with picrosirius red solution. The extent of liver fibrosisas determined as the proportion of picrosirius-stained area in

ach section. For each rat, 64 fields of a constant raster of 31 mm2

ere analysed at 100-fold final magnification. For semiautomatedorphometry a Sony 3CCD (model DXC-950P) videomicroscope

quipped with motor stage and the Quantimed 500MC (Leica, Ger-any) software were used.

.11. Determination of serum biochemical parameters

Serum aspartate aminotransferase (AST), alanine aminotrans-erase (ALT), alkaline phosphatase (ALP) and total bilirubin levelsere determined to assess liver function by using a Modularutoanalyzer (Roche Diagnostics, GmbH, D-68298, Mannheim,ermany).

.12. TG activity

TG activity (tTG) in liver homogenised in RIPA lysis buffer asescribed below, was assayed according to a modified version ofhe method described by Lorand et al. [20]: 30 �L of sample wasdded to 45 �L of reaction mixture containing a final concentrationf 0.25 mM 14C-putrescine (Amersham, UK), 50 mM dithiothre-tol, 10 mM CaCl2 and 4% (w/v) dimethylcasein in Tris–HCl buffer50 mM), pH 9.0, with 0.1% Triton X-100 and incubated with shakingt 37 ◦C for 20 min. Then, 20 �L was spotted onto 3MM Whatmanound paper filters (2 cm) and immediately plunged into 10% iceold trichloroacetic acid for 15 min. Two consecutive 15-min wash-ngs were performed in 5% ice cold trichloroacetic acid followed bybrief washing in ethanol–acetone (50%, v/v) and then in acetone.he dried paper filters were counted in 6 mL of Aquasure scintillantDupont-NEN). A similar procedure was adopted for blanks, stan-ards and controls. TG units were expressed as 1 mU = 1 nmol ofutrescine in acceptor protein at 37 ◦C, pH 9.

.13. Western blotting

Liver samples were homogenised in RIPA lysis buffer (0.1%odium dodecyl sulphate (SDS), 0.5% deoxycholate, 1% Nonidet,00 mM NaCl, 10 mM Tris–HCl (pH 7.4)) containing a pro-ease inhibitor cocktail (Sigma, St Louis, Missouri, USA), 0.5 mMithiothreitol and 0.5% phenylmethylsulphonyl fluoride. Liver

er Disease 42 (2010) 571–577 573

homogenates from rats of each experimental group as well asnormal controls were run on 10% SDS/polyacrylamide gel accord-ing to Laemmli [21] and transferred to nitrocellulose using Bioradtransblot apparatus. Anti-tTG mouse monoclonal antibody (CUB7402; NeoMarkers, California, USA) diluted 1:1000 was added andincubation continued overnight at 4 ◦C. The appropriate avidinconjugated secondary antibody was added and immunoreactiveprotein was detected by development with the ABC Vectastain kit(Vector Laboratories, Burlingame, California, USA) according to themanufacturer’s instructions.

2.14. RNA extraction and reverse transcription-polymerase chainreaction (RT-PCR)

Total RNA was extracted from rat liver samples usingTrizol reagent (Gibco Reagent, USA) according to the manu-facturer’s protocol. mRNA levels of the gene under analysiswas measured by semiquantitative RT-PCR amplification. RatmRNA sequence was selected to design primer pairs for RT-PCR reactions using the Oligo 4.05 software (National BiosciencesInc., Plymouth, Minnesota, USA). Primer sequences were asfollows: for tTG, upper primer 5′-TGGCAGAGAAAGAGGAGACAGG-3′, lower primer 5′-CATACAGGGAATCAGAAAGTGG-3′; and forGAPDH, upper primer 5′-CACAGTCAAGGCTGAGAATG-3′, lowerprimer 5′-GCATTGCTGACAATCTTGAG-3′. Amplifications were car-ried out under the following conditions: 94 ◦C for 60 s, 57 ◦Cfor 60 s and 72 ◦C for 60 s. To highlight the effect of GE,we also evaluated the expression of alpha-SMA as pro-fibrogenic gene and IL-1� as pro-inflammatory gene in ratlivers. Primer sequences were as follows: for alpha-SMA,upper primer 5′-CGATAGAACACGGCATCATC-3′ lower primer 5′-CATCAGGCAGTTCGTAGCTC-3′. Amplification was carried out underthe following conditions: 94 ◦C for 60 s, 61 ◦C for 60 s and 72 ◦Cfor 60 s. For IL-1�, upper primer 5′-ACTTGGGCTGTCCAGATGAGA-3′, lower primer 5′-GCCTGCAGTGCAGCTGTCTA-3′, amplificationconditions were: 94 ◦C for 60 s, 62 ◦C for 60 s and 72 ◦C for 60 s.According to the expression level of the target genes, the numberof cycles was 35. The correct number of cycles was determined pre-viously to ensure that PCR was taking place in the linear range inorder to guarantee a proportional relationship between input RNAand densitometric readout. Each RT-PCR experiment was repeatedat least three times. Amplification products were electrophoresedon 2% agarose gel in 1× TAE. Semiquantitative analysis of mRNAlevels was carried out using the software associated with the FluorS imager (BioRad Company, Hercules, California, USA). The den-sity values of RT-PCR products were normalised with respect toendogenous GAPDH product levels.

2.15. Statistics

Groups of data were compared with an analysis of variance(ANOVA) followed by Tukey’s multiple comparison tests. Valuesof p < 0.05 were regarded as significant.

3. Results

3.1. Chemical characterisation of GE

The chemical composition of the GE was reported in Table 1. Theanalysis of total polyphenols showed a content of these bioactive

compounds of 0.3 mg gallic acid equivalent/g GE.

As shown in Fig. 1, the HPLC/MS/MS analysis of organosulphurcompounds of GE, allowed the identification of three main com-pounds. The amount of organosulphur compounds in GE was: alliin54.2 �g/g GE, S-allylcysteine 8 �g/g GE and allicin 0.4 �g/g GE.

574 G. D’Argenio et al. / Digestive and Liver Disease 42 (2010) 571–577

Table 1Chemical composition of garlic extract (g/100 g).

Water 19.8Carbohydrates 44.7Total dietary fibre 0.8Proteins 17.9

3

vaaCeaci(

3

Hbw

improved liver fibrosis showing only perivascular fibrosis with ini-

Fa

Lipids 0.8Ashes 3.9

.2. In vitro effect of GE on TG activity

To assess the effect of GE on TG activity, GE was incubated initro with normal rat liver homogenate (10%, w/v) as source of tTGctivity. The optimal concentration of GE was calculated to obtainn inhibition of TG activity similar to that obtained with 1 mg/mLYS. Results are shown in Fig. 2. The GE inhibited about 80% of thenzyme activity when incubated with liver homogenate for 30 mint room temperature at the concentration of 4.3 mg/mL. Higheroncentration of GE (8.6 mg/mL) inhibited only the 61% of tTG activ-ty. The experiment was also carried out at 37 ◦C with similar resultsnot shown).

.3. Histology and liver collagen content

All harvested livers were subjected to histologic analysis after&E and picrosirius red/fast green staining by M.R.D. and E.M. in alinded manner. Livers harvested from rats administering CCl4 for 7eeks showed a pattern of fibrosis with septa linking the centrolob-

ig. 1. HPLC/MS/MS analysis of GE. Panel A shows the Total Ion Chromatogram (TIC). Pllylcysteine and allicin as reported in Section 2.

Fig. 2. In vitro inhibition of transglutaminase (TG) activity: liver homogenate assource of TG activity was incubated at room temperature for 30 min with cystamineor garlic extract at the indicated concentration. The first column represents the TGactivity without inhibitors. Results are the mean of three different experiments.

ular veins. Sometimes livers showed the histologic appearance ofan advanced fibrosis (Fig. 3G). Red Sirius staining of livers fromrats treated with CCl4 showed a pattern of fibrosis with estab-lished septa linking hepatic veins and further new matrix bridgingthese areas to the portal tracts (Fig. 3A). CYS treatment significantly

tial septa (Fig. 3B). Livers of rats treated with GE showed evidenceof a residual perivenular fibrosis: there were no bridging septa fromthe hepatic veins to the portal tracts. In addition, the calibre of theremaining septa was reduced in width compared with peak fibrosis

eaks in panels B–D were obtained extracting the specific transitions of alliin, S-

G. D’Argenio et al. / Digestive and Liver Disease 42 (2010) 571–577 575

Fig. 3. Representative Red Sirius staining, immunohistochemistry using tTG antibody and H&E on rat liver sections: (A) Red Sirius staining 7 weeks after CCl4 shows a fibrosispattern with collagen deposition and porto-central septa (original magnification ×60). (B) Treatment with cystamine induces a significant improvement of liver fibrosiswith only perivascular fibrosis and initial septa (original magnification ×60). (C) After treatment with garlic extract only residual perivenular fibrosis without septa waspresent. (D–F) tTG immunostaining clearly overlaps the Red Sirius pattern suggesting a co-localisation with newly formed collagen fibres (original magnification ×100). (G)L osis wa

(b

3

iwifda

3

tsmrbfi

3

m

ivers harvested from rats administering CCl4 for 7 weeks showed a pattern of fibrppearance of an advanced fibrosis (original magnification ×40).

Fig. 3C). The morphometric analysis of the collagen fibres analysedy Quantimed are reported in Fig. 4A.

.4. Immunohistochemistry

To assess the distribution of tTG in the fibrotic liver andn the liver of animals treated with CYS or GE, serial sections

ere immunostained for tTG antibody. As shown in Fig. 3D, tTGmmunolocalised with collagen fibres indicating that tTG is mainlyound in the ECM of damaged liver. According to the low collageneposition following CYS or GE treatments, the ECM was invariablylmost negative for tTG staining (Fig. 3E and F).

.5. Biochemical parameters

Serum AST, ALT, ALP and total bilirubin levels were determinedo assess liver function. Rats with CCl4-induced fibrosis showederum levels of ALT and ALP higher than those of normal rats. Treat-ent with either CYS or GE reported these values within the normal

ange (Fig. 5). No statistically significant difference in total biliru-in concentration was found in the different groups compared withbrosis control (not shown).

.6. TG activity

TG activity assayed on liver homogenates by a radio-enzymaticethod was increased after CCl4 treatment, whereas CYS treat-

ith septa linking the centrolobular veins. Sometimes livers showed the histologic

ment significantly reduced TG activity levels. In a similar mannerGE treatment inhibited TG activity (Fig. 4B).

3.7. Western blotting

tTG protein levels assayed by Western blotting are shown inFig. 6A. Data revealed the increase of tTG protein expression in CCl4compared with normal rats. tTG protein levels in either CYS or GEtreated rats returned towards normal levels. These results werereproduced for all animals involved in the study in independentexperiments.

3.8. RNA extraction and RT-PCR

mRNA levels of tTG by semiquantitative RT-PCR were deter-mined to assess whether expression of this gene was modified inthe fibrotic liver. We found that tTG mRNA was more abundantin CCl4 fibrotic rats, increasing from two to four times com-pared with normal rats. mRNA levels of tTG in CYS and GE weredecreased compared to control (CCl4) group. Accordingly, Theexpression of pro-fibrogenic gene alpha-SMA was up-regulated

in CCl4 rats accounting for the increased collagen deposition.Alpha-SMA mRNA was decreased by CYS and more markedlyby GE treatments. IL-1� was increased in CCl4 rat livers anddown-regulated by CYS and GE treatments. Representative RT-PCRamplification patterns are shown in Fig. 6B–D.

576 G. D’Argenio et al. / Digestive and Liver Disease 42 (2010) 571–577

Fig. 4. (A) Percentage of collagen fibres quantified by morphometric analysis using Quangarlic extract (GE) treatments. (B) Transglutaminase (TG) activity assayed on liver homog

Fig. 5. Liver function tests (AST, ALT, ALP). The ALT level increased in rat with CCl4was reduced by cystamine (CYS) and garlic extract (GE) treatments.

Fig. 6. Effect of cystamine and garlic extract on tTG protein (A) and gene expressionof tTG (B), alpha-SMA (C) and IL-1� (D) in rat liver. Representative Immunoblot andRT-PCR from liver tissues were obtained from normal rats and CCl4, CCl4 + cystamine,CCl4 + garlic extract-treated rats.

timed. The collagen deposition was significantly reduced by cystamine (CYS) andenates by a radio-enzymatic method.

4. Discussion

The development of fibrosis, and particularly cirrhosis, is associ-ated with a significant morbidity and mortality. Thus, it is extremelyessential to develop anti-fibrotic strategies that are applicable toliver fibrosis. Most of the hypotheses relating to the pathogen-esis of liver fibrosis have focused on the increased synthesis ofthe ECM or its decreased breakdown [4]. The resistance to thematrix degradation from specific collagenases may be due to aqualitative change of the matrix itself. This is possibly mediatedby the tTG covalent stabilisation of the fibrils, incorporating �-(�-glutamyl) lysine cross-links. Eventually, this stabilisation movesthe deposition–degradation balance towards accumulation andfibrosis [22].

The therapeutic benefit of CYS has been shown in several neu-rodegenerative disease models [9,11,23] and its efficacy has beenattributed to the inhibition of tTG [12,14] and caspase-3 [24], aswell as to the stimulation of glutathione (GSH) synthesis [25]. Cys-teamine, a sulphide compound derived by CYS, has been recentlydemonstrated to be both an alternate amine-bearing substratefor tTG and an apparent competitive inhibitor of this enzyme[14]. Unfortunately, even at therapeutic doses, cysteamine givesundesirable effects that limits its use [12,15]; therefore, findingsubstitute compounds that may elicit cysteamine properties, butlimiting side effects, may be an important challenge for clinicians.

Garlic organosulphur compounds share many chemical proper-ties with cysteamine and its disulphide CYS. In in vitro preliminaryexperiments, we observed that GE containing alliin, S-allylcysteineand allicin inhibits tTG activity in liver homogenate at the sameextent as pure CYS despite the low concentration of 0.006% of GE(w/w). This suggests that the combination of these molecules mayact in a synergistic manner compared to a single compound.

Previously, the anti-fibrotic effects of garlic [26], GEs [17] andsome garlic constituents [27] in different models of liver damagehave been demonstrated by their antioxidant activity. Indeed, thisis the first study to show that the prevention of liver injury by GEmay be correlated to the direct inhibition of TG activity that in turn,reduces the levels of enzyme gene expression and transduction,possibly through a mechanism of negative feedback.

It is known that in liver tissue TGF-� stimulates the hepaticstellate cell differentiation into myofibroblasts that release colla-gen into the ECM where it is cross-linked by tTG. Furthermore,tTG activates latent TGF-�, leading to de novo synthesis of tTG in aself-maintaining system. Data obtained from this study allows us

nd Liv

toa

tdlsTaef

sTscm

nEcpsoi

baohmwo

CN

A

wf

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

G. D’Argenio et al. / Digestive a

o hypothesise that this mechanism, which leads to the formationf fibrosis, can be inhibited by garlic through the blockage of tTGctivity and expression.

The immunohistochemical study provided further insight intohe role of tTG in fibrosis, showing that the enzyme localisation inamaged liver tissue acquires structural morphologies quite simi-

ar to that obtained by collagen deposition. Both tTG and collagentaining were strongly reduced by either CYS or GE treatment.his result was consistent with the finding of a down-regulation oflpha-SMA gene expression further supporting the antifibrogenicffect of GE, while the reduced IL-1� expression provides evidenceor a role of GE also in reducing inflammation.

Several recent studies reported health benefits of garlic diallyl-ulphides in many pathologies, such as cancer and CVD [28–30].his study showed the ability of GE in reducing liver fibrosis andtrengthening the therapeutic potential of the diallyldisulphideompound and allylmercaptan, which represents the main in vivoetabolite of organosulphur compounds [31].Today, there is great public interest in natural remedies and in

on-conventional therapies for the treatment of several diseases.vidence supports the use of phytochemicals in combination withonventional therapies to reduce the side effects of pharmaceuticalroducts, making garlic a potential candidate in this scenario. Inpite of the numerous studies dealing with the beneficial effectsf garlic consumption in humans, none had reported its ability innhibiting tTG activity before our study.

In conclusion, the data shown in this paper suggest that the inhi-ition of tTG activity obtained by a tailored GE preparation mayllow recovery from liver injury and may block the progressionf liver fibrosis. Since current therapeutic approaches in treatingepatic fibrosis are limited, a comprehensive understanding of theolecular mechanisms involved in tTG function and its inhibitionould provide new therapeutic targets to inhibit the progression

f this severe disease.

onflict of interest statementone declared.

cknowledgement

D.C. Amoruso was recipient of a SIGE award, year 2008. Thisork was in part supported by a Campania Municipality grant on

unctional foods to VF.

eferences

[1] Török NJ. Recent advances in the pathogenesis and diagnosis of liver fibrosis. JGastroenterol 2008;43:315–21.

[2] Wanless IR, Nakashima E, Sherman M. Regression of human cirrhosis. Morpho-

logic features and the genesis of incomplete septal cirrhosis. Arch Pathol LabMed 2000;124:1599–607.

[3] Benyon RC, Iredale JP. Is liver fibrosis reversible? Gut 2000;46:443–6.[4] Iredale JP, Benyon RC, Pickering J, et al. Mechanisms of spontaneous resolu-

tion of rat liver fibrosis. Hepatic stellate cell apoptosis and reduced hepaticexpression of metalloproteinase inhibitors. J Clin Invest 1998;102:538–49.

[

[

[

er Disease 42 (2010) 571–577 577

[5] Iredale JP. Hepatic stellate cell behaviour during resolution of liver injury. SeminLiver Dis 2001;21:427–36.

[6] Elli L, Bergamini CM, Bardella MT, et al. Transglutaminases in inflammation andfibrosis of the gastrointestinal tract and the liver. Dig Liver Dis 2009;41:541–50.

[7] Greenberg CS, Birckbichler PJ, Rice RH. Transglutaminases: multifunc-tional cross-linking enzymes that stabilize tissues. Fed Am Soc Exp Biol J1992;5:3071–7.

[8] Kojima S, Nara K, Rifkin DB. Requirement for transglutaminase in the activationof latent transforming growth factor-beta in bovine endothelial cells. J Cell Biol1993;121:439–48.

[9] Lorand L. Neurodegenerative diseases and transglutaminase. Proc Natl Acad SciU S A 1997;93:14310–3.

10] Ikura K, Takahata K, Sasaki Y. Cross-linking of a synthetic partial length (1–28)peptide of the Alzheimer �/A4 amyloid protein by transglutaminase. FEBS Lett1993;326:109–11.

11] Lesort M, Tucholski J, Miller ML, et al. Tissue transglutaminase: a possible rolein neurodegenerative diseases. Prog Neurobiol 2000;61:439–63.

12] Dedeoglu A, Kubilus JK, Jeitner TM, et al. Therapeutic effects of cystamine in amurine model of Huntington’s disease. J Neurosci 2002;22:8942–50.

13] Qiu J-F, Zhang Z-Q, Chen W, et al. Cystamine ameliorates liver fibrosis inducedby carbon tetrachloride via inhibition of tissue transglutaminase. World J Gas-troenterol 2007;28:4328–32.

14] Jeitner TM, Delikatny EJ, Ahlqvist J, et al. Mechanism for the inhibition of trans-glutaminase 2 by cystamine. Biochem Pharmacol 2005;69:961–70.

15] Dubinsky R, Gray C. CYTE-I-HD Phase I dose finding and tolerability studyof cysteamine (Cystagon) in Huntington’s disease. Mov Disord 2006;21:530–3.

16] Tapiero H, Townsend DM, Tew KD. Organosulfur compounds from alliaceae inthe prevention of human pathologies. Biomed Pharmacother 2004;58:183–93.

17] Borek C. Antioxidant health effects of aged garlic extract. J Nutr2001;131:1010S–5S.

18] Singleton VL, Rossi Jr JA. Colourimetry of total phenolics withphosphomolybdic–phosphotungstic acid reagents. Am J Enol Viticult 1965;16:144–58.

19] Issa R, Zhou X, Trim N, et al. Mutation in collagen-1 that confers resistance to theaction of collagenase results in failure of recovery from CCl4-induced liver fibro-sis, persistence of activated hepatic stellate cells, and diminished hepatocyteregeneration. FASEB J 2003;17:47–9.

20] Lorand L, Campbell-Wilkes LK, Cooperstein L. A filter paper assay fortransamidating enzyme using radioactive amine substrate. Ann Biochem1972;50:623–31.

21] Laemmli UK. Cleavage of structural proteins during the assembly of the headof bacteriophage T4. Nature 1970;227:680–5.

22] Grenard P, Bresson-Hadni S, Alaoui SE, et al. Transglutaminase-mediated cross-linking is involved in the stabilization of extracellular matrix in human liverfibrosis. J Hepatol 2001;35:367–75.

23] Karpuj MV, BecherMW, Springer JE, et al. Prolonged survival and decreasedabnormal movements in transgenic model of Huntington disease, with admin-istration of the transglutaminase inhibitor cystamine. Nat Med 2002;8:143–9.

24] Lesort M, Lee M, Tucholski J, et al. Cystamine inhibits caspase activ-ity. Implication for the treatment of polyglutamine disorders. J Biol Chem2003;278:3825–30.

25] Chang ML, Klaidman LK, Adams Jr JD. The effects of oxidative stress on invivo brain GSH turnover in young and mature mice. Mol Chem Neuropathol1997;30:187–97.

26] Gedik N, Kabasakal L, Sehirli Ö, et al. Long-term administration of aqueous garlicextract (AGE) alleviates liver fibrosis and oxidative damage induced by biliaryobstruction in rats. Life Sci 2005;76:2593–606.

27] Amagase H, Petesch B, Matsuura H, et al. Intake of garlic and its bioactivecomponents. J Nutr 2001;131:955S–62S.

28] Powolny A, Singh SV. Multitargeted prevention and therapy of cancer by dial-lyl trisulfide and related Allium vegetable-derived organosulfur compounds.Cancer Lett 2008;269:305–14.

29] Kritchevsky D. The effect of dietary garlic on the development of cardiovasculardisease. Trends Food Sci Technol 1991;2:141–4.

30] Ackerman R, Mulrow CD, Ramirez G, et al. Garlic shows promise for improvingsome cardiovascular risk factors. Arch Int Med 2001;161:813–24.

31] Egen-Schwind C, Eckard R, Kemper FH. Metabolism of garlic constituents in theisolated perfused rat liver. Planta Med 1992;58:301–5.