Impact of hepatic immunoreactivity of angiotensin-converting enzyme 2 on liver fibrosis due to non-alcoholic steatohepatitis

(1)

Availableonlineat

ScienceDirect

www.sciencedirect.com

ORIGINAL

ARTICLE

Impact

of

hepatic

immunoreactivity

of

angiotensin-converting

enzyme

2 on

liver

ﬁbrosis

due

to

non-alcoholic

steatohepatitis

Mustafa

Cengiz

a,∗

_,

_Seren

_Ozenirler

b

_,

_Guldal

_Yılmaz

c

_,

Gulbanu

Erkan

d

a_Dr._A.Y._Ankara_Oncology_Training_and_Research_Hospital,_Department_of_{Gastroenterology,}

06200Ankara,Turkey

b_Gazi_University_Faculty_of_Medicine,_Department_of_{Gastroenterology,}_Ankara,_Turkey c_Gazi_University_Faculty_of_Medicine,_Department_of_Pathology,_Ankara,_Turkey d_Ufuk_University_Faculty_of_Medicine,_Department_of_{Gastroenterology,}_Ankara,_Turkey

Availableonline14April2015

Summary

Background: We aimedto evaluate thehepatic immunoreactivity ofangiotensin-converting enzyme2(ACE2) innon-alcoholic steatohepatitis (NASH) patients,elucidate its association withtheclinicopathologicalcharacteristicsandalsodetermineitsroleinfibrosisprogression. Methods:TheconsecutivebiopsyprovenNASHpatientsweresubdividedintotwogroups accord-ingtotheir fibrosisscore.Fibrotic stages<3inmild fibrosisgroup andfibroticstages≥3in advancedfibrosisdependingonthepresenceofbridgingfibrosis.Liverbiopsyspecimenswere immunohistochemicallystained for ACE2immunoreactivity.Demographics andclinical prop-ertieswere compared between the groups. Univariate and multivariate analysis were also performedtoevaluatetheindependentpredictingfactorsforthepresenceofadvancedliver fibrosiscausedbyNASH.

Results:Onehundredandeightpatientswereenrolledinthestudy.Outofthis,ninety-four patients representing87% were classifiedas mildfibrosis group, whilstfourteen represent-ing13%wereinadvancedfibrosisgroup.WecomparedhighhepaticimmunoreactivityofACE2 betweenmildandadvancedfibrosisgroupsandfoundastatisticallysignificantdifference65.9% vs28.5%,respectivelyandP=0.008.HepaticACE2immunoreactivitywasinverselycorrelated withthefibrosisscore(r:—0.337;P<0.001).Thesignificantvariablesintheunivariateanalysis werethenevaluatedinmultivariatelogisticregressionanalysisandhepaticACE2 immunore-activitywas anindependent predictingfactor ofliver fibrosis [oddsratio (OR): 0.194; 95% confidenceinterval(CI):0.082—0.897,P=0.036].

∗_{Corresponding}_author._Mehmet_Akif_Ersoy_Mahallesi._13._Cadde_No₅₆_Demetevler,_Yenimahalle,_Ankara,_Turkey.

Tel.:+903123360909;fax:+903123340352.

E-mailaddress:drmustafacen@gmail.com(M.Cengiz).

http://dx.doi.org/10.1016/j.clinre.2015.02.010

(2)

Conclusion:Hepaticimmunoreactivity ofACE2wasinverselycorrelatedwith theliver ﬁbro-sisamong biopsyprovenNASHpatients anditwasalso anindependent predictingfactorof advancedﬁbrosis.

Introduction

Non-alcoholic fatty liver disease (NAFLD) constitutes a pathologicalillness, ranging fromsimple hepaticsteatosis tosteatohepatitis,fibrosis andcirrhosis[1]. Non-alcoholic steatohepatitis (NASH) is undoubtedly an enhanced step of NAFLD described as the accumulation of hepatic fats, inflammationandvariousdegreesofliverfibrosis. NASHis emerging asa main publichealth issue [2,3]. The patho-genesis ofNASHis notreallycomprehended andthemost acknowledgedconcepttodescribethisparticularillnessmay bethe‘‘two-hithypothesis’’[4].Accordingtothistheory, first,theexcessfat is depositedinthe liverandthen the particularhepaticdamageiscausedbyoxidativestressand other cytokines [5]. Subsequently, excessive cytokines in theliverparticipateinthehistopathologicalprogressionof NASH.

Angiotensin (Ang) II, the principal effector in the renin—angiotensinsystem(RAS) hasbeenidentifiedfor its critical roles in the development, progression of NASH, steatosis,inflammation andliverfibrosis[6].Ang IIis cre-ated from Ang I through the catalyzing of ACE. ACE2, a homologueofACE, degradesAngII togenerateAng-(1—7)

[7].ACE2influencestheAngIIamountanditsphysiological activities[8],especiallyintheinsulinsensitivetissues:liver, adiposetissues andskeletalmuscle[9—11].The discovery ofthehomologueofACE2haschangedourunderstandingof theprinciplesofRASandguidedresearchesconcerningthis enzymeanditsimpactonchronicliverdiseasesand treat-ment approaches. Various studies have documented that hepaticACE2 immunoreactivitywasupregulatedinhuman cirrhotic livers or in theinjury model of mice [10,12,13]. There are contradictory results concerning ACE2. While somestudies showedincreasedeffects,someof the stud-iesconcludeddecreasingeffectsonliverfibrosis[10,14,15]. Thisissueneedsclarificationbecauseoftheimportanceof ACE2intheliver.

We aimedtoevaluatethe hepaticimmunoreactivityof ACE2 in NASH patients, elucidate its association with the clinicopathologicalcharacteristicsandalsoexamineitsrole intheprogressionofliverﬁbrosis.

Materials

and

methods

Patients

The consecutive biopsy proven NASH patients who were admittedtoourhospitalbetweenMarch2009andDecember 2013 had persistent increase liver enzymes and hep-atosteatosis on ultrasonography were enrolled into the study.Allof them werealso olderthan 18years andwith noreasons for the elevated liver enzymes. The following

groupsofindividualsandpatientswereexcludedfromthe study:viralhepatitis,sclerosingcholangitis,primarybiliary cirrhosis,autoimmunehepatitis,hemochromatosis,Wilson’s disease, alpha-1-antitrypsin deficiency, malignancy, usage of steatogenic drugs and drug-induced liver diseases and alcoholconsumptions morethan20g/d.Patientswhohad clinical, laboratory and histologically determined decom-pensated cirrhosis were scrutinized as ineligible for the study and excluded. None of the individuals with NASH possesses any co-existing acute or chronic inflammatory diseases,includinghypertension,sarcoidosis,renalfailure, cardiac failure, diabetes mellitus or any type of other enclosed diseases that might influence ACE. None of the individualswasreceivingACEinhibitors/angiotensin recep-torblockersoranyothermedicationsthatcouldaffectthe RAS.

Clinicalandlaboratoryassessment

Thedemographics,clinical,laboratory,andliverbiopsydata ofpatientswereregisteredinadatabasebyanuninformed cliniciantopreventrecallbias.Anthropometricevaluation ofheightandweightweremeasured,bodymassindex(BMI) wascalculated by the formula of dividing the weight by the square of height in meter (kg/m2_). _Before _the _liver

biopsy,venousbloodsampleswereobtainedfromthe ante-cubitalveinbetween8.30and10.00am,after10—12hours fast.Complete blood count (CBC)containing,hemoglobin (Hb), white blood cell (WBC), neutrophils, lymphocytes and platelets was performed by using Beckman Coulter Gen-S automated analyzer (High Wycombe, UK) in 2h afterobtainingtheblood samples.Fastingplasmaglucose andlipid parameters[total cholesterol (TC),triglycerides (TG), and high-density lipoprotein cholesterol (HDL-C)] weremeasuredusingcommercially availablereagent kits. Low-densitylipoproteincholesterol(LDL-C)levelswere cal-culated using Freidewald’s equation [16]. Roche Modular SystemautoanalyzerRocheCobasIntegra800(Indianapolis, USA)wasusedtoperformalbumin,creatinine,bloodurea nitrogen(BUN),alanine aminotransferase(ALT), aspartate aminotransferase(AST),gammaglutamylaminotransferase (GGT)andalkalinephosphatase(ALP)levelsinthe labora-toryofourhospital.

Histopathologicalevaluationoftheliver

Percutaneousliverbiopsywasperformedusinga16-G Hep-aﬁx(BraunMelsungenAG,Germany),non-reusableneedle, by the same skilled clinician. All of liver biopsy speci-mensincludedatleast12completeportaltractsandwere longerthan20mm.All livertissuesamples wereassessed bythesameexperiencedhepatopathologistblindedtothe

(3)

Figure1 A.Mildﬁbrosisinnon-alcoholicsteatohepatitis.B.Advancedﬁbrosisinnon-alcoholicsteatohepatitis.

patients’ clinical, demographics and laboratory records. The stains of Hematoxylin and eosin (H—E) and Masson trichromewereusedtoevaluatethehistopathological diag-nosesofformalin-fixedandparaffin-embeddedlivertissues. ThediagnosisofNASHhadbeendeterminedbyBrunt’s Crite-ria [17]. Histological features were graded in accordance withthe NAFLD scoring system (NAS) and in concordance withtheNationalInstituteofDiabetesandDigestiveand Kid-neyDiseasesNASHClinicalResearchNetwork[18].Hepatic steatosis has been graded from 1 to 3 depending on the steatosis percentage: 5—33%, 33—66% and>66% denoting score 1, 2 and 3, respectively. Lobular inflammation was identified as the entire assessment of all inflammations: absolutelynofociasscore 0,lower than2 fociper ×200 fieldasscore1,two-fourfociper×200fieldasscore2,more than4fociper×200fieldasscore3.Ballooningscoringwas termedas:score0ifnobalooning,score1forfewandscore 2iftherewerenumerousballooningofhepatocytes.Liver fibrosiswasstagedasfollow: stage0;nofibrosis,stage1; perisinusoidalor periportalfibrosis,stage2;perisinusoidal andportal/periportalfibrosis,stage3;bridgingfibrosisand stage4ascirrhosis.Histopathologically,thetotalNASwas calculatedasasumofsteatosis(1—3),lobularinflammation (0—3)and ballooning (0—2).Individuals were classified as mildandadvancedfibrosis groups.While mildfibrosis was characterizedasfibrosisscore<3(Fig.1A),advancedfibrosis wascharacterizedasfibrosisscore≥3(Fig.1B).

HepaticACE2immunoreactivityassessement

Liver tissue specimens were 10% formalin fixated and paraffin-embedded.Theywerecut into4␮m sectionsand placed on slides before deparaffinisation and rehydrata-tionprocesses.ACE2antibodyIGGwasused(dilution1:50, Santa Cruz Biotechnology, California, USA) with antigen retrieval via Tris—EDTA. The immunohistochemical stain-ingwas proceededmanually by using streptavidine-biotin tripleindirectimmunoperoxidasemethod.The chromogen usedwas3-amino9-ethylcarbazole(AEC)and counterstain-ing was done by using Mayer’s haematoxylin. All biopsy specimenswereanalyzedbythesameskilled hepatopathol-ogistin a blinded manner. Immunoreactivityof ACE2 was evaluated by calculating the percentage of cytoplasmic immunostainedhepatocytesforeachcase.Theintensityof cytoplasmicstainingwasdefinedaslow(stainedcells,<10%)

orhigh(stainedcells,≥10%)expression.Fortheevaluation ofACE2immunoreactivity,thecasesweredividedintotwo groupsshowingfocalornostaining(<10%)aslow,andliver immunoreactivitystaining(≥10%)ashigh.Accordingtothe datasheetoftheproduct,thecytoplasmicpositivestaining of thekidney tubularepithelial cells wasusedaspositive control. An independent investigator assessed the posi-tivityof ACE2 semiquantitativelywithout prior knowledge oftheclinicalstudy.AmongallACE2positivecases,nuclear ACE2immunoreactivitypercentageanddegreeswere calcu-latedbycountingat least1000cellsat randomlyselected 10 ﬁelds under high magniﬁcation. According to percents anddegreesofthehistologicalstainthatrecordedblindly, ACE2 immunoreactivity severity was considered as a low immunoreactivity(<10%)ACE2(Fig.2A)andhigh immunore-activity(≥ 10%)(Fig.2B).

Ethics

This particularresearchstudy wascarriedout in concord-ancewiththe1975DeclarationofHelsinkithatwasupdated in 2008 and approved by the Local Institutional Review Board.

Statisticalanalysis

All the statistical analysis wasperformed using SPSS ver-sion 21(SPSS Inc.,Chicago, IL, USA) andMedCalc version 12.7(Mariakerke,Belgium).Thevariableswereinvestigated using visual (histogram, probability plots) and analyti-calmethods(Kolmogorov—Smirnov/Shapiro—Wilk’stest)to identify whether or not they are normally distributed. Descriptiveanalysiswaspresentedusingmedians±standard error (SE) for non-normally distributed variables and mean±standard deviation (SD) for normally distributed ones.Thefrequencieswerecalculatedforcategorical varia-bles.Thecorrelationcoefficientsandtheirsignificancewere calculated using Spearman/Pearson tests where appropri-ate. The univariate analysis to identify variables among patientswithadvancedfibrosiswasinvestigatedusingChi2_,

Fisherexact,Student’sTandMann-WhitneyUtests.For mul-tivariateanalysis,theactualfactorsrecognizedalongwith the univariate studies had been additionally subjected to the actual logistic regression evaluation toﬁnd out inde-pendentpredictorsfortheadvancedﬁbrosis.Forassessing

(4)

Figure2 A.LowhepaticACE2immunoreactivity.B.HighhepaticACE2immunoreactivity.

the model fit, Hoshmer—Lemeshow goodness of fit statis-tics wereused. A5% type-1error level wasused toinfer statisticalsignificance.

Results

Accordingtotheinclusionandexclusioncriteria,one hun-dred and eight consecutive biopsy proven NASH patients were enrolledin the study of whom43.5% (47/108) were female.While ninety-fourpatientsrepresenting 87% were classifiedasmildfibrosisgroup,therewerefourteen repre-senting13%inadvancedfibrosisgroup.Inthemildfibrosis group, 41.5% of patients were female while in advanced fibrosis group 57.1% were female. There was no statisti-cally significant differencedepending ongender (P=0.39) between fibrosis groups. In the advanced fibrosis group, median age was 51 (37—70) while it was 48 (22—71) in mild fibrosis group and P=0.05. There was no statisti-callysignificantdifferencedependingonBMIbetweenmild andadvanced fibrosis groups29.35±0.61vs29.41±1.73, respectivelyandP=0.99.Wecomparedhighliver immunore-activity of ACE2 between the mild and advanced fibrosis subgroups of NASH and found a statistically significant difference65.9%vs28.5%,respectively,P=0.008.The com-parisons of clinical, demographics and histopathological featuresofgroupsaresummarizedinTable1.

Histopathological features of liver biopsies of patients aresummarizedinTable2.Fibrosiswasfoundas1.44±0.97 andsteatosiswas2.37±0.72amongpatients.Themeanof NASamongpatientswas5.27±1.33.Interfacehepatitisor otherfeaturessuggestiveofautoimmunehepatitisandother chronicliverdiseaseswerenotpresentinanyofthecases. HepaticACE2immunoreactivitywasinverselycorrelated with the fibrosis scores of biopsy proven NASH patients. This inverse correlation was confirmed statistically (r: —0.337; P<0.001). Age, gender and BMI were not cor-related withhepatic immunoreactivity of ACE2 (P>0.05). While there was a correlation between liver fibrosis and hepatic immunoreactivityof ACE2 therewere no correla-tionsbetweentheotherhistopathologicalfeaturesofNASH suchassteatosis, inflammationandbalooning(P>0.05)as seeninTable3.

Univariate analysisfor thepresence of advanced ﬁbro-sisduetoNASHindicatedthathepaticimmunoreactivityof

ACE2,age,AST,hemoglobin,whitebloodcellcount,platelet countwerestatistically signiﬁcant predictingfactors. The resultsofunivariateanalysisareshowninTable4.

Thosevariablesthatwerestatisticallysignificantinthe univariate logistic regression analysis were further sub-jected to multivariate logistic regression analysis, and liver ACE2 immunoreactivity continued to be statistically significantandanindependentpredictingfactorofthe fibro-sis, odds ratio (OR)=0.194, 95% confidence interval (CI): 0.042—0.897,P=0.036,asseeninTable5.

Discussion

The purpose of this specific study was to assess hepatic ACE2immunoreactivityinNASHpatientsandtheparticular hypothesiswasthatlocalACE2 intheliver mightindicate ongoingpathologicalactivitiesinthemicroenvironmentof theliverfibrosis. Explanationofthe associationsbetween ACE2 and NASH could givea better understanding of the enigmaticpathogenesisofNASH.Remarkably,the informa-tiongainedfromthisresearchdemonstratedthatindividuals withhighACE2immunoreactivityinliverpossessdecreased fibrosisscores,indicatingthatACE2isactuallyanovel pro-gnosticfactor inNASH andtothe bestof ourknowledge, thisisthefirststudyconductedamonghumanbeings.ACE2 immunoreactivitywasinverselycorrelatedwithliver fibro-sis and after analysis, it was an independent predicting factor ofadvanced liver fibrosis. Alltheseresults declare thatACE2immunoreactivitycorrelateswithNASH develop-mentandprogressionofthedisease.Recently,ithasbeen concludedthat hypertensive patients with NAFLD on RAS blockertreatment hadless advanced hepatic fibrosis sug-gestinganindirectbeneficialeffectofRASblockersinNAFLD

[19]. Ithasbeen shown thatACE2 takesa protectiverole in hepatic ﬁbrosis and also the increased ACE2 improves hepaticinjury.As wefound thatthehepatic immunoreac-tivityofACE2wasdecreasedinadvancedliverﬁbrosis,this wascompatiblewiththeresultsofprecedingexperimental models[12—14].

ACE2,ahomologueofACE,candegradeAngIIandcreate Ang-(1—7),whichactsasthemaincounter-regulatorofthe traditionalACE-AngII-AT1axis.Pereiraetal.concludedthat blockingendogenousAng-(1—7)pharmacologically acceler-atesliverﬁbrosis [20].KnockoutoftheACE2geneinmice

(5)

Table1 Thecomparisonofdemographic andclinicalfeaturesbetweenmildandadvancedﬁbrosisgroups ofbiopsyproven NASH.

Factors Mildﬁbrosisa_(n₌₉₄₎ _Advanced_ﬁbrosisb_(n₌₁₄₎ _P_value

Ageyears(min—max) 48(22—71) 51(37—70) 0.049

Gender(female)% 31(44.9%) 16(41%) 0.273 BMI(kg/m2₎ _29.35_±_0.61 _29.41_±_1.73 _0.99 ACE2immunoreactivity(%) 65.9% 28.5% 0.008 Hemoglobin(g/dL) 14.2±0.29 13.85±0.52 0.04 Platelet(×103_/mm3₎ ₂₄₉_±_50.9 ₁₆₁_±_26.9 _0.000 Leucocyte(109_/L) ₇₅₅₀_±_319.2 ₆₁₁₀_±_662.03 _0.07 INR 0.93±0.02 1.02±0.02 0.000 ALT(U/L) 61_±7.5 43_±31.7 0.678 AST(U/L) 39±6.8 40±38.6 0.087 ALP(U/L) 82_±11.2 116_±13.6 0.007 GGT(U/L) 53±14.83 77±27.74 0.075 Albumin(g/dL) 4.4±0.07 4.30±2.96 0.065 Totalcholesterol(mg/dL) 194±4.19 184±11.9 0.249 Totalbilirubin(mg/dL) 0.59±0.09 0.89±0.08 0.016 LDL(mg/dL) 119_±4.06 111.5_±9.75 0.578 HDL(mg/dL) 42±1.17 40±2.25 0.429 Triglyceride(mg/dL) 161_±9.31 116.5_±17.05 0.183 Na 140±0.32 138.5±0.74 0.067 K 4.41_±0.05 4.24_±0.08 0.327 Ca 9.6±0.06 9.55±0.13 0.322 BUN(mg/dL) 13±0.44 12.6±1.28 0.939 Creatinine(mg/dL) 0.82±0.02 0.75±0.08 0.142

ACE2:angiotensin-convertingenzyme2,ALT:alanineaminotransferase,AST:aspartateaminotransferase,ALP:alkalinephosphatase, GGT:gammaglutamyltransferase,BMI:bodymassindex,BUN:bloodurinenitrogen,HDL:high-densitylipoprotein,INR:international normalisedratio,LDL:low-densitylipoprotein.Variablesareskeweddistributedandexpressedasmedian_±standarderror(SE).

a_Mild_fibrosis_is_defined_as_fibrotic_score_<_3. b _Advanced_fibrosis_is_defined_as_fibrotic_score_≥_3.

Table2 HistopathologicalfeaturesofNASHpatients.

n=108 Mean±SD Fibrosisscore 1.44±0.97 NAS 5.27±1.33 Steatosis 2.37±0.72 Inﬂammation 1.47±0.63 Blooning 1.41_±0.55

NAS:non-alcoholicfattyliverdiseasescore.

Table3 Thecorrelationbetweenhepatic immunoreactiv-ityofACE2andcliniccharacteristicsandhistopathological featuresofNASH.

Factors rvalue Pvalue

Age 0.06 0.538 Gender —0.011 0.913 BMI —0.122 0.378 Fibrosis —0.337 <0.001 Inﬂammation —0.092 0.345 Steatosis —0.011 0.911 Balooning —0.086 0.381 NAS —0.078 0.424

BMI:bodymassindex, NAS:non-alcoholicfattyliverdisease score,rvalue:correlationcoefﬁcient.

Table4 Theunivariateanalysisofpredictingliverﬁbrosis inNASH. Factors OR CI95% Pvalue Age 1.062 1.004—1.122 0.035 Gender 0.532 0.171—1.655 0.276 BMI 1.039 0.891—1.211 0.625 ALT 1.004 0.997—1.011 0.291 AST 1.009 1.001—1.018 0.027 ALP 1.005 0.997—1.012 0.22 GGT 1.003 0.998—1.008 0.288 Hemoglobin 0.745 0.556—0.999 0.05

Whitebloodcell 0.999 0.999—1.00 0.013

INR 0.978 0.492—1.941 0.948 Platelets 0.984 0.975—0.993 0.001 Albumin 1.140 0.927—1.402 0.215 HighACE2 immunoreactivity 0.206 0.06—0.710 0.012 ACE2: angiotensin-converting enzyme 2, ALT: alanine aminotrensferase,AST:aspartataminotransferase,ALP: alka-line aminotransferase, BMI: body mass index, GGT: gamma glutamyltransferase,INR:internationalnormalizedratio.

(6)

Table5 Themultivariateanalysisofpredictingliver ﬁbro-sisinNASH. Factors OR CI95% Pvalue Age 1.068 0.994—1.148 0.072 AST 1.009 0.999—1.019 0.094 WBC 1.00 0.999—1.00 0.174 HB 1.025 0.685—1.535 0.903 PLT 0.986 0.974—0.998 0.02 ACE2 0.194 0.042—0.897 0.036

ACE2: angiotensin-converting enzyme 2, ALP: alkaline phosphatase,GGT:gammaglutamyltransferase.

exacerbates liver fibrosis following bile duct ligation and chroniccarbontetrachloride(CCL4).Ithasbeenconcluded that recombinantACE2 plays aprotective role in liverby decreasing experimental liver fibrosis in a well-designed mousemodel[15].Ourresultssupportthisideaandlarge population-basedprospectiveclinicaltrialsneedtobe car-riedouttoconfirmthisasascientificfact.

In a previous research, the authors have shown that ACEIscouldcertainlyenhancehepaticsteatosisamong dia-beticrats[21].Typically,theupregulationofhepaticACE2 immunoreactivity mayplay a criticalrole in the develop-ment of hepatic steatosis. The tissue-specific ACE2 could possibly be an additional potential target for enhancing hepatic steatosis. But, we could not find a correlation between hepatic ACE2 immunoreactivity and steatosis in NASH patients and our result is not compatible with the results gained fromthe animal models. It maybe due to the role of ACE2 in hepatic stellate cell (HSC) in fibro-sis.AT-IIinduces typicallythe fibroticimpactoftriggered HSCs throughstimulating TGF-b1 expression and boosting collagen synthesisin liver throughtheactivation of AT-1R receptor[6,22,23].

ACE2, a promising component of RAS, has the ability of reducingliverdamage. Moreresearches regarding liver fibrosismustbeperformedtoensuretheparticularroleof localACE2.Ithasbeen arguedthatimproving thehepatic immunoreactivityofACE2andalsotheactivityofACE2could beapowerfulwayofthetreatmentofliverfibrosis[24].In the present study,we observed thatthere is asignificant negativecorrelationbetweenhepaticimmunoreactivityof ACE2andfibrosisstages.ThecorrelationsbetweenACE2and liverfunctionsindirectlysupportthehypothesisthatACE2 playsanimportantroleinliverinjury.

In experimental models of hypertension and diabetes, the expression of tissue ACE2 is reduced, contributing to AngII-mediated tissueinjury[25,26]. Inarecent study,it hasbeenconcludedthatthetissue-specificACE2couldbe anotherpotentialtargetforimprovinghepaticsteatosis.It hasbeen concludedthatamongtheratswithhigh-fatdiet (HFD)-induced NASH,theupregulationof ACE2 immunore-activityimprovedhepaticstetatosisandliverdamage[27]. Asdiabetes,hypertensionanddrugsmightaffecttheACE2, weexcludedthesefactorsandwe foundtheexacteffects ofhepaticACE2immunoreactivityontheliverfibrosis.High hepatic ACE2 immunoreactivity was inversely correlated withadvancedfibrosis.

Coreyetal.researchedtheeffectofAng-blockingagents onliverfibrosis in patientswithhepatitis Cand also con-cludedthatmuchlessfibrosiswasrelatedtotheprotective roleofACEinhibitorsamongindividuals[28].Furthermore, areview performed by Koh etal. have emphasized some potential benefits of the inhibition of the RAS system in liverinjury and liver fibrosis [29]. In this specific review, theauthorsidentifiedthattheimpedingoftheRASprocess incitesliverrestorationandpreventstumor development. Additionally,somesurprisingeffectsonthefunctionofthe RASsystemandliverdiseaseshavediscoveredthattheAng IIinducesvesselshrinkageandworksasaproliferativeagent forHSCsbyAT1receptors[30,31].Allthesestudies extrapo-latedthespecializedmedicalprocesstothedevelopmentof anumberofprobabledrugtreatments,whichoftenimpact theinitializedHSCsevokingtheimpedimentinvolvingAngII throughintrahepaticflow.Cirrhosismaybethefinalresultof mostpersistentdiseasesintheliver.Forthisreason, focus-ingontheactualHSCsisreallyanencouragingalternative forpreventingtheadvancementofliverfibrosis.According toourresults, hepaticACE2 immunoreactivitymay affect theHSCsactivitiesandlowertheliverfibrosis.

Todate,thereisalack ofrandomized, controlled clin-icaltrials regarding the preferable treatment approaches forNASH.Moreexperimentalandclinicalscientificstudies areessentialinlookingfortheprecisestatusofcirculating andlocal ACE2 in the pathobiologyof the widespectrum of various chronic liver diseases, such asNASH. Focusing ontherelationship between circulatingand localACE2 in thehepatic microenvironment wouldlikelyilluminate the particularbiologicalfoundationandclinicalmanagementof liverdiseasesandfibrosis.Forthisreason,additional exper-imentalandmedicalresearchesmustbecarriedouttolook forthelocalACE2asatreatmentmodalityinthe pathophy-siologyofliverdiseases.Upcomingresearchesregardingthe correlationbetweenACE2andliverdiseases aredefinitely warrantedand ourstudy will open a newavenue for this topic.

Thereweresomelimitationsofourstudy.Relatively,the smallnumberofpatientswiththesameethnicitywasa lim-itation.Inthefuture,population-basedrandomizedclinical trialsin differentethnicitieswillbe of beneﬁttoconﬁrm theresults ofthestudy.The lackofserumACElevelswas anotherlimitationofthestudy aswell. Wewerenotable toevaluatethe serumlevelsbecauseof theretrospective designofthestudydependingontheclinicopathologic eval-uation. In the future, studies evaluating the association betweenhepaticimmunoreactivityandserumlevelsofACE2 inNASHwillgivemoreinformationaboutthediseaseandits progression.

Inconclusion,highhepaticACE2immunoreactivity pos-sesses decreased ﬁbrosis scores, indicating that ACE2 is actually a novel prognostic factor in NASH. This study declaredthatACE2immunoreactivitycorrelateswithNASH developmentinadditiontoprogressionofdiseaseand sug-gestanoveltherapeutictargetforliverﬁbrosisduetoNASH.

Disclosure

of

interest

Theauthorsdeclarethattheyhavenoconﬂictsofinterest concerningthisarticle.

(7)

References

[1]CohenJC,HortonJD,HobbsHH.Humanfattyliverdisease:old questionsandnewinsights.Science2011;332:1519—23.

[2]Pascale A, Pais R, Ratziu V. An overview of nonalcoholic steatohepatitis:past,present and futuredirections, Journal ofgastrointestinalandliverdiseases.JGastrointestinLiverDis 2010;19:415—23.

[3]Hashimoto E,TokushigeK.Prevalence,gender,ethnic varia-tions, andprognosisofNASH.JGastroenterol2011;46(Suppl 1):63—9.

[4]DayCP,JamesOF.Steatohepatitis:ataleoftwo‘‘hits’’? Gas-troenterology1998;114:842—5.

[5]BrowningJD,HortonJD.Molecularmediatorsofhepatic steato-sisandliverinjury.JClinInvest2004;114:147—52.

[6]Matthew Morris E, FletcherJA, ThyfaultJP, Rector RS.The roleofangiotensinIIinnonalcoholicsteatohepatitis.MolCell Endocrinol2013;378:29—40.

[7]Donoghue M, Hsieh F, Baronas E, Godbout K, Gosselin M, Stagliano N, et al. A novel angiotensin-converting enzyme-related carboxypeptidase (ACE2) converts angiotensin I to angiotensin1-9.CircRes2000;87:1—9.

[8]SantosRA,SimoeseSilvaAC,MaricC,SilvaDM,MachadoRP, deBuhrI,etal.Angiotensin-(1-7)isanendogenousligandfor theGprotein-coupledreceptorMas.ProcNatlAcadSciUSA 2003;100:8258—63.

[9]GembardtF,Sterner-KockA,ImbodenH,SpalteholzM, Reib-itzF,SchultheissHP,etal.Organ-speciﬁcdistributionofACE2 mRNAandcorrelatingpeptidaseactivityinrodents.Peptides 2005;26:1270—7.

[10]PaizisG,TikellisC,CooperME,SchembriJM,LewRA,Smith AI, et al. Chronic liver injury in rats and humans upregu-latesthenovelenzymeangiotensinconvertingenzyme2.Gut 2005;54:1790—6.

[11]FernandesT,Hashimoto NY,OliveiraEM.Characterizationof angiotensin-convertingenzymes1and2inthesoleusand plan-tarismusclesofrats.BrazJMedBiolRes2010;43:837—42.

[12]Herath CB, Warner FJ, Lubel JS, DeanRG, Jia Z, Lew RA, etal.Upregulationofhepaticangiotensin-convertingenzyme 2(ACE2)andangiotensin-(1-7) levelsinexperimentalbiliary ﬁbrosis.JHepatol2007;47:387—95.

[13]HuangQ,XieQ,ShiCC,XiangXG,LinLY,GongBD,etal. Expres-sionofangiotensin-convertingenzyme2inCCL4-inducedrat liverﬁbrosis.IntJMolMed2009;23:717—23.

[14]Huang ML, Li X, Meng Y, Xiao B, Ma Q, Ying SS, et al. Upregulation of angiotensin-converting enzyme (ACE) 2 in hepaticﬁbrosisbyACEinhibitors.ClinExpPharmacolPhysiol 2010;37:1—6.

[15]Osterreicher CH, Taura K, De Minicis S, Seki E, Penz-Osterreicher M, Kodama Y, et al. Angiotensin-converting-enzyme2inhibitsliverﬁbrosisinmice.Hepatology 2009;50:929—38.

[16]Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentrationoflow-densitylipoproteincholesterolinplasma, without use of the preparative ultracentrifuge. Clin Chem 1972;18:499—502.

[17]BruntEM,JanneyCG,DiBisceglieAM,Neuschwander-TetriBA, BaconBR.Nonalcoholicsteatohepatitis:aproposalfor grad-ingand stagingthe histologicallesions. Am JGastroenterol 1999;94:2467—74.

[18]Kleiner DE, Brunt EM, Van Natta M, Behling C, Contos MJ, CummingsOW,etal. Designand validationofa histological scoringsystemfornon-alcoholicfattyliverdisease.Hepatology 2005;41:1313—21.

[19]GohGB,PagadalaM,DasarathyJ,Unalp-AridaA,SargentR, HawkinsC,etal.Reninangiotensinsystemandﬁbrosisin non-alcoholicfattyliverdisease.LiverInt2014;35:979—85.

[20]PereiraRM,DosSantosRA,TeixeiraMM,LeiteVH, CostaLP, daCostaDiasFL,etal.Therenin-angiotensinsysteminarat modelof hepaticﬁbrosis: evidencefor aprotective role of Angiotensin-(1-7).JHepatol2007;46:674—81.

[21]ZhangX, Li ZZ,Liu DF,Xu X,Mei ZC, Shen W. Angiotensin-converting enzyme inhibitors improve hepatic steatosis by modulating expression of tumour necrosis factor-alpha, interleukin-6andadiponectin receptor-2inratswithtype2 diabetes.ClinExpPharmacolPhysiol2009;36:631—6.

[22]BatallerR,Sancho-BruP,GinesP,BrennerDA.Liver ﬁbrogene-sis:anewrolefortherenin-angiotensinsystem.AntioxidRedox Signal2005;7:1346—55.

[23]NabeshimaY,TazumaS,KannoK,HyogoH,IwaiM,HoriuchiM, etal.Anti-ﬁbrogenicfunctionofangiotensinIItype2receptor inCCl4-inducedliverﬁbrosis.Biochem BiophysResCommun 2006;346:658—64.

[24]ZhangW,MiaoJ,LiP,WangY,ZhangY.Up-regulationof com-ponentsoftherenin-angiotensinsysteminliverﬁbrosisinthe ratinducedbyCCL(4).ResVetSci2013;95:54—8.

[25]CrackowerMA,SaraoR,OuditGY,YagilC,KozieradzkiI,Scanga SE,etal.Angiotensin-convertingenzyme2isanessential reg-ulatorofheartfunction.Nature2002;417:822—8.

[26]Tikellis C, Johnston CI, Forbes JM, Burns WC, Burrell LM, Risvanis J, et al. Characterization of renal angiotensin-convertingenzyme 2 in diabeticnephropathy. Hypertension 2003;41:392—7.

[27]ZhangW,XuYZ,LiuB,WuR,YangYY,XiaoXQ,etal. Pioglita-zoneupregulatesangiotensinconvertingenzyme2expression ininsulin-sensitivetissuesinratswithhigh-fat diet-induced nonalcoholicsteatohepatitis.SciWorldJ2014;2014:603409.

[28]Corey KE, Shah N, Misdraji J, Abu Dayyeh BK, Zheng H, BhanAK,etal.Theeffectofangiotensin-blockingagentson liverﬁbrosisinpatientswithhepatitis C.LiverInt2009;29: 748—53.

[29]KohSL,AgerEI,ChristophiC.Liverregenerationandtumour stimulation:implicationsoftherenin-angiotensinsystem.Liver Int2010;30:1414—26.

[30]BatallerR,GinesP,NicolasJM,GorbigMN,Garcia-RamalloE, GasullX, et al. Angiotensin IIinduces contractionand pro-liferationof human hepatic stellatecells. Gastroenterology 2000;118:1149—56.

[31]BatallerR,GabeleE,ParsonsCJ,MorrisT,YangL,Schoonhoven R, et al. Systemic infusion of angiotensin II exacerbates liver ﬁbrosis in bile duct-ligated rats. Hepatology 2005;41: 1046—55.