advanced glycation endproducts induce fibrogenic activity in nonalcoholic steatohepatitis by...

Advanced Glycation Endproducts Induce FibrogenicActivity in Nonalcoholic Steatohepatitis by Modulating

TNF-a-Converting Enzyme Activity in MiceJoy X. Jiang,1* Xiangling Chen,1* Hiroo Fukada,1 Nobuko Serizawa,1 Sridevi Devaraj,2 and Natalie J. T€or€ok1

Advanced glycation endproducts (AGEs) accumulate in patients with diabetes, yet thelink between AGEs and inflammatory and fibrogenic activity in nonalcoholic steatohe-patitis (NASH) has not been explored. Tumor necrosis factor alpha (TNF-a)-convertingenzyme (TACE) is at the center of inflammatory processes. Because the main naturalregulator of TACE activity is the tissue inhibitor of metalloproteinase 3 (Timp3), wehypothesized that AGEs induce TACE through nicotinamide adenine dinucleotide phos-phate reduced oxidase 2 (NOX2); and the down-regulation of Sirtuin 1 (Sirt1)/Timp3pathways mediate fibrogenic activity in NASH. The role of NOX2, Sirt1, Timp3, andTACE was evaluated in choline-deficient L-amino acid defined (CDAA) or Western diet(WD)-fed wild-type (WT) and NOX22/2 mice. To restore Timp3, mice were injectedwith adenovirus (Ad)-Timp3. Sirt1 and Timp3 expressions were studied in livers fromNASH patients, and we found that their levels were significantly lower than in healthycontrols. In WT mice on the CDAA or WD, Sirt1 and Timp3 expressions were lower,whereas production of reactive oxidative species and TACE activity significantlyincreased with an increase in active TNF-a production as well as induction of fibrogenictranscripts. Ad-Timp3 injection resulted in a significant decline in TACE activity, pro-collagen a1 (I), alpha smooth muscle actin (a-SMA) and transforming growth factorbeta (TGF-b) expression. NOX22/2 mice on the CDAA or WD had no significantchange in Sirt1, Timp3, and TACE activity or the fibrosis markers assessed. In vitro,AGE exposure decreased Sirt1 and Timp3 in hepatic stellate cells by a NOX2-dependentpathway, and TACE was induced after exposure to AGEs. Conclusion: TACE activationis central to the pathogenesis of NASH and is mediated by AGEs through NOX2 induc-tion and down-regulation of Sirt1/Timp3 pathways. (HEPATOLOGY 2013; 00:000–000)

Diabetes mellitus (DM) is a major risk factorfor disease progression with necroinflamma-tion and fibrosis advancing to cirrhotic-stage

nonalcoholic steatohepatitis (NASH).1,2 The factorsimplicated in this progression are poorly understood,specifically the effects of diabetes or insulin resistance(IR) on fibrogenesis. Advanced glycation endproducts

(AGEs) are produced by a nonenzymatic glycation ofserum proteins, and these modifications significantlyinfluence the structure and function of key protein tar-gets.3,4 AGEs are implicated in diabetic nephropathy,vascular complications, and retinopathy,5,6 but theirrole in inducing inflammatory or fibrogenic changes inthe liver have not been adequately explored. Sirtuin 1

Abbreviations: Ad, adenovirus; AGEs, advanced glycation endproducts; ALT, alanine aminotransferase; BSA, bovine serum albumin; CDAA, choline-deficient L-amino acid defined; CSAA, choline-supplemented L-amino acid-defined; DM, diabetes mellitus; FBS, fetal bovine serum; GA, glycolaldehyde; GFP, green fluores-cent protein; HCs, healthy controls; HSCs, hepatic stellate cells; IP, immunoprecipitation; IR, insulin resistance; KCs, Kupffer cells; KO, knockout; NADPH oxidase,nicotinamide adenine dinucleotide phosphate reduced oxidase; NASH, nonalcoholic steatohepatitis; NOX2, NADPH oxidase 2; OS, oxidative stress; PBS, phos-phate-buffered saline; PKC, protein kinase C; pfu, plaque-forming units; qPCR, quantitative polymerase chain reaction; RAGE, receptor of advanced glycation end-products; ROS, reactive oxidative species; siRNAs, small interfering RNAs; Sirt1, Sirtuin 1; a-SMA, alpha smooth muscle actin; TACE, TNF-a-convertingenzyme; TG, triglyceride; TGF-b, transforming growth factor beta; Timp3, tissue inhibitor of metalloproteinase 3; TNF-a, tumor necrosis factor alpha; WD, West-ern diet; WT, wild type.

From the 1Department of Gastroenterology and Hepatology, University of California Davis, Sacramento, CA; 2Texas Children’s Hospital, Houston, TX.Received August 17, 2012; accepted April 23, 2013.Supported by the NIDDK DK 083283 to NJT.*These authors contributed equally.

1

(Sirt1) belongs to the class III family of histone deace-tylases, and its decreased activity was shown to belinked to the development of NASH.7 Accumulatingevidence suggests that, in NASH, hepatic lipid metab-olism pathways are affected by Sirt18 and low levels ofSirt1 are implicated in the development of steatosis inanimals.9 Moreover, in the heterozygous Sirt1 knock-out (KO) model, when chronically challenged with a40% fat diet, mice became obese and insulin resistant,displaying increased serum cytokine levels and develop-ing hepatomegaly.10 Hepatic metabolomic analysesrevealed that Sirt1 heterozygous mice had elevated glu-coneogenesis and oxidative stress (OS).11 Among thetargets of Sirt1 is tissue inhibitor of metalloproteinases3 (Timp3),12 a key regulator and inhibitor of the tu-mor necrosis factor alpha (TNF-a)-converting enzyme(TACE; also called A Metalloprotease and Disintegrin17) activity. Timp32/2 mice have been described todevelop vascular inflammation by increased TNF-a13

as well as hepatic steatosis.14

In this study, we showed that exposure to AGEs leadsto down-regulation of Sirt1 and Timp3 in hepatic stel-late cells (HSCs) by activation of NADPH oxidase (nic-otinamide adenine dinucleotide phosphate reducedoxidase) 2 (NOX2) and the release of reactive oxidativespecies (ROS). Accordingly, wild-type (WT) mice fedthe choline-deficient L-amino acid defined (CDAA) orwestern diet (WD) showed decreasing Sirt1, Timp3expression, and an increase in TACE activity, TNF-aproduction, and increased fibrogenic activity. Correctionof the low TIMP3 levels by injection of adenoviral(Ad)-Timp3 into CDAA or WD-fed mice reducedTACE activity, TNF-a, expression of the receptor ofadvanced glycation endproducts (RAGE), and fibro-genic response. NOX22/2 mice on NASH diets did notdevelop increased TACE activity or fibrogenic response.In summary, these data suggest an important role ofAGEs in NOX2-mediated induction of TACE, TNF-a,and fibrogenic activity during NASH progression.

Materials and Methods

Liver Biopsy Samples. Liver biopsy samples wereobtained from the University of California Davis

Cancer Center Biorepository (Sacramento, CA) fundedby the National Cancer Institute (Bethesda, MD).Samples from 6 different patients and 6 normal liverswere tested. All patients had IR or DM.

Animal Studies. C57BL/6 mice (The Jackson Lab-oratory, Bar Harbor, ME) or B6.129S-Cybbtm1Din/J(NOX22/2; Jackson Lab) were given choline-supple-mented L-amino acid-defined (CSAA) or CDAA diet(Dyets Inc., Bethlehem, PA) for 10 weeks or controlchow. Both diets contain higher calories, fat, and car-bohydrates than standard chow.15 The fast-food(“Western”) diet contains 40% energy as fat (12% sat-urated fatty acids, 2% cholesterol; AIN-76, Westerndiet; TestDiet, St. Louis, MO) and is supplementedwith high-fructose corn syrup (42 g/L).16 To test forthe role of Timp3 in TACE regulation, Ad-Timp3 wasinjected on week 8 through the tail vein (2 3 107 pla-que-forming units [pfu]/200 uL of phosphate-bufferedsaline [PBS]; Applied Biological Materials Inc., Rich-mond, British Columbia, Canada). As control, Ad-GFP (green fluorescent protein; 2 3 107 pfu/200 uL;Vector Biolabs, Philadelphia, PA) was used. In a groupof mice, GdCl3 (10 mg/kg in saline; Sigma-Aldrich,St. Louis, MO) was injected intraperitoneally everyother day throughout the experiment to inhibit macro-phages in both models. Serum and liver tissue werecollected and alanine aminotransferase (ALT) and bili-rubin were tested. Tissue was processed for the furtherassays. Animals were housed in facilities approved bythe National Institutes of Health (Bethesda, MD). Allprocedures were reviewed and approved by the animalwelfare committee of the University of CaliforniaDavis.

Cell Culture. Primary HSCs were isolated eitherfrom C57BL/6 mice or Sprague-Dawley rats asdescribed previously.17 Cells were cultured in medium199/20% fetal bovine serum (FBS). Primary HSCswere used for experiments within the first 3 days afterisolation. Cells were exposed to AGEs (50 lg/mL) orbovine serum albumin (BSA) in serum-free mediumfor 16 hours and/or transfected with small interferingRNAs (siRNAs).

For siRNA transfection, primary rat HSCs were cul-tured as described above for 1 day, then the medium

Address reprint requests to: Natalie Torok, M.D., Department of Gastroenterology and Hepatology, University of California Davis Medical Center, 4150 VStreet, Suite 3500, Sacramento, CA 95817. E-mail: [email protected]; fax: 916-734-7908.

Copyright VC 2013 by the American Association for the Study of Liver Diseases.View this article online at wileyonlinelibrary.com.DOI 10.1002/hep.26491Potential conflict of interest: Nothing to report.Additional Supporting Information may be found in the online version of this article.

2 JOY, XIANGLING, ET AL. HEPATOLOGY, Month 2013

was changed to Dulbecco’s modified Eagle’s medium,0.5% FBS. HSCs were transfected with the siRNA toSirt1 or Timp3 (Santa Cruz Biotechnology Inc., SantaCruz, CA) or scrambled siRNA using the RiboJuicetransfection reagent (EMD Chemicals Inc., Darmstadt,Germany), according to the manufacturer’s instructions.

Results

Exposure to AGEs Induces Sirt1 and Timp3Down-Regulation by NOX2 in Primary HSCs. Ac-tivation of NOXs is one of the main sources of ROSin activated HSCs. We have previously shown thatNOX2 is an important factor in HSC transdifferentia-tion.18 To study the role of AGEs in ROS production,primary cells were transfected with NOX2 orscrambled siRNA and incubated in serum-free condi-tions with glycolaldehyde (GA)-derived AGEs for 16hours. Superoxide production was significantlyincreased in control (untransfected) and scrambledsiRNA-transfected HSCs (Fig. 1A; 1.74- 6 0.13-fold;P < 0.05), whereas this was attenuated in NOX2siRNA-transfected cells (Fig. 1A). To study the effectsof AGEs/ROS on Sirt1 and Timp3 expression, real-time quantitative polymerase chain reaction (qPCR)

was done on cells from the above-described experi-ment. Sirt1 and Timp3 were down-regulated inresponse to AGEs in a NOX2-dependent manner (Fig.1B; 0.38- 6 0.14-fold; P < 0.05; 0.58- 6 0.10-fold;P < 0.05, respectively). To test whether Sirt1 directlytargets Timp3 in primary HSCs, cells were transfectedwith Sirt1 siRNA, and Timp3 expression was found tobe down-regulated in response to inhibition of Sirt1(0.36- 6 0.16-fold; N 5 3; P < 0.05; Fig. 1C).

AGEs Induce TACE Activity in HSCs. BecauseTimp3 is the main natural inhibitor of TACE activity,we next examined whether TACE could be induced byAGEs. This was studied by two different methods: flu-orometry (Fig. 2A) and immunoprecipitation (IP) withwestern blotting to assess TACE tyrosine phosphoryla-tion, which correlates with the active state of theenzyme (Fig. 2B). TACE activity was induced byAGEs in scrambled siRNA-transfected HSCs (1.48- 6

0.06-fold; P < 0.01; Fig. 2A), and this was signifi-cantly attenuated in NOX2 siRNA-transfected HSCs(1.10- 6 0.09-fold; P < 0.05). In addition, inhibitingSirt1 by siRNA also resulted in an increase in TACEactivity (1.43- 6 0.10-fold; P < 0.05), which was fur-ther increased by AGEs (2.29- 6 0.23-fold; P <0.05). TACE was phosphorylated after exposure of

Fig. 1. AGEs induce ROS production and down-regulate Sirt1 and Timp3 expression by NOX2. Primary HSCs were treated with eitherscrambled siRNA or NOX2 siRNA for 48 hours, followed by GA-derived AGE exposure or BSA (control) for 16 hours. Lucigenin assay showed thatsuperoxide production was significantly increased by AGEs in both untransfected (NT) (*P < 0.05) or scrambled siRNA-transfected (Scr) HSCs(*P < 0.05), whereas knockdown of NOX2 attenuated superoxide production (mean 6 SE; *P < 0.05; N 5 4) (A). Sirt1 and Timp3 mRNAexpression were assessed by real-time qPCR. After treating with AGEs for 16 hours, Sirt1 and Timp3 mRNA levels decreased in NT and ScrsiRNA-transfected HSCs (shown in fold expression; *P < 0.05; N 5 4), but not in NOX2 siRNA-transfected HSCs (B). Timp3 mRNA expressiondecreased in response to siRNA knockdown of Sirt1 (*P < 0.05; N 5 4), as assessed by real-time qPCR (C). SE, standard error of the mean;mRNA, messenger RNA.

HEPATOLOGY, Vol. 00, No. 00, 2013 JOY, XIANGLING, ET AL. 3

HSCs to AGEs, indicating the activated state of theenzyme (Fig. 2B). To discern whether the other majorcell types in the liver contribute to AGE-inducedTACE activation, primary hepatocytes, HSCs, andKupffer cells (KCs) were incubated either with BSA orAGEs. TACE activity of hepatocytes was very low andnot induced by AGEs (Fig. 2C). Only HSCsresponded with a significant increase (1.73- 6 0.14-fold; P < 0.05), and whereas the baseline activity ofKCs was higher than that of hepatocytes; no inductionwas noted after incubation with AGEs.

Sirt1 and Timp3 Are Down-Regulated in Liversof Patients With NASH. To study Sirt1 and Timp3in human livers with NASH, real-time qPCR was per-formed on liver biopsy samples from different patientswith grade 2-3, stage 3-4 NASH, and healthy controls(HCs; Fig. 3). The patients represented here had eitherIR with fasting hyperglycemia or type II DM. Expres-sion of both Sirt1 and Timp3 were significantlydecreased in NASH livers, compared to normal HCs(expressed as fold over control: 0.47- 6 0.11-fold; P< 0.05; and 0.24- 6 0.07-fold; P < 0.01,

respectively; N 5 6; Fig. 3). Expression of RAGE wasalso significantly increased in patients with NASH(3.93- 6 0.2-fold; P < 0.05).

NOX2-Dependent ROS Production Is Increasedin Two Different Diet Models of NASH and IsInvolved in the Regulation of Sirt1 and Timp3. Torecapitulate our in vitro findings, WT or NOX22/2 micewere fed either with CDAA, CSAA, or WD. The CDAAmodel was shown to recapitulate the features of humanNASH with IR, inflammatory cell infiltration, hepatocytedeath, and liver fibrosis,19,20 whereas mice on the CSAAdiet mainly develop steatosis, but no necroinflammationor fibrosis. The advantage of using the CSAA/CDAAmodel is that we can correlate our findings to a modelresulting in simple steatosis or progressive NASH. Toconfirm our findings, we also performed experiments inmice fed the WD supplemented with high fructose.16

This diet was shown to have higher fidelity to humanpathophysiology, with mice exhibiting significant weightgain, metabolic syndrome, and diabetes and histologicallydeveloping ballooning, lipoapoptosis, necroinflamma-tion, and fibrosis. Body weight has increased similarlyboth in WT and NOX22/2 mice on both NASH diets(Fig. 4A). WT mice on CDAA or WD had significantlyincreased ALT (P < 0.01), compared to mice on theCSAA or control chow diet. Mice on CDAA and WDhad increased glucose levels (216.0 6 36.2 mg/dL and239.4 6 12.8 mg/dL, respectively), and these diets werepreviously shown to induce impaired glucose tolerance ordiabetes.15,16 AGEs were increased in both diet modelsin WT but not in the NOX22/2 mice (Supporting Fig.1). The increase in bilirubin in WT CDAA mice mayreflect impaired synthetic function, because these micehad more-severe steatohepatitis and fibrosis. In NOX22/2

mice, ALT showed an increasing trend on NASH diets,

Fig. 2. AGEs induce TACE activity in HSCs. After transfection withscrambled (Scr) siRNA, NOX2 siRNA, or SirT1 siRNA for 48 hours, pri-mary HSCs were exposed to AGEs or BSA for 16 hours, and TACE ac-tivity was studied by fluorometry. TACE activity increased after AGEtreatment in both nontransfected (mean 6 SE; *P < 0.05; N 5 4)and Scr siRNA-transfected (**P < 0.01; N 5 4) groups. This was sig-nificantly attenuated in NOX2 siRNA-transfected cells (*P < 0.05, N5 4). TACE activity was also induced in response to knockdown ofSirt1 (*P < 0.05; N 5 4) (A). TACE activity was also analyzed by IPand western blotting with anti-p-tyrosine. TACE was phosphorylated af-ter exposure to AGEs (B). TACE activity was tested in isolated primaryhepatocytes, HSCs, and KCs after incubation with BSA or AGEs. OnlyHSC TACE activity was significantly induced by AGEs (mean 6 SE; *P< 0.05; N 5 3). Hepatocytes had a very low TACE activity, whereasKCs did not exhibit induction after AGEs (C). SE, standard error of themean.

Fig. 3. Sirt1 and Timp3 were down-regulated, whereas RAGEexpression was induced in NASH patients. Real-time qPCR was per-formed on liver samples from patients with NASH and HCs. Sirt1 andTimp3 were significantly down-regulated in livers with NASH (dataexpressed as fold over values from HCs, which was set as 1; mean 6SE; *P < 0.05; **P < 0.01, respectively; N 5 6). Expression ofRAGE was significantly increased in NASH patients (**P < 0.01). SE,standard error of the mean.


Fig. 4. NOX2-dependent ROS production was increased in two diet models of NASH and was involved in the regulation of Sirt1 and Timp3.WT and NOX22/2 mice were fed with chow, CSAA, CDAA, or WD. Compared to WT mice on chow or CSAA diet, WT CDAA- and WD-fed mice dis-played an increase in weight, a significant increase of serum ALT (**P < 0.01, N 5 6), and, in CDAA mice, total bilirubin. In NOX22/2 mice,there was no significant increase in ALT or bilirubin. (A). Hematoxylin and eosin staining showed that WT and NOX22/2 mice on CDAA and WDhad increased steatosis (B). Lucigenin assay demonstrated significantly increased superoxide production in WT mice on CDAA and WD, comparedto those on chow and CSAA diets (*P < 0.05; N 5 6). No increase was noted in NOX22/2 mice (*P < 0.05; N 5 6) (C and D). SirT1 and-Timp3 messenger RNA expression was assessed by real-time qPCR in WT and NOX22/2 mice. SirT1 and Timp3 expression significantlydecreased in WT CDAA- (E) and WD-fed mice (F) (**P < 0.01; *P < 0.05, respectively; N 5 6). In contrast, in NOX22/2 mice, Sirt1 and-Timp3 expression remained unchanged.


compared to baseline, albeit not significant, and belowthe level of liver injury observed in WT mice, whereasbilirubin had not changed. On histology, all mice onCDAA or WD displayed increased steatosis (Fig. 4B).

ROS production had significantly increased in WTmice on CDAA or WD (Fig. 4C,D; 1.57- 6 0.16-fold; P < 0.05; and 1.75- 6 0.01-fold; P < 0.05),whereas this was attenuated in NOX22/2 mice inboth models (to 0.94- 6 0.11-fold and 0.89- 6 0.26-fold; P < 0.05). To correlate the data to humans withNASH, qPCR was performed in all experimental con-ditions (Fig. 4E,F). WT CDAA- or WD-fed miceshowed a significant decrease in Sirt1 (to 0.10- 6

0.04-fold; P < 0.01; and 0.11- 6 0.031-fold; P <0.05, respectively; N 5 6) and Timp3 expression (to0.21- 6 0.04-fold; P < 0.05; and 0.43- 6 0.01-fold;P < 0.05, respectively; N 5 6). In contrast, inNOX22/2 mice, Sirt1 and Timp3 expression was notaffected by CDAA or WD.

Correction of Low Timp3 Expression in WT Miceon CDAA or WD by Ad-Timp3 Ameliorates TACEand TNF-a Activity. To confirm the causal linkbetween low Timp3 expression leading to an unop-posed increase in TACE activity and consequentinduction of TNF-a activity in vivo, we injected Ad-Timp3 or Ad-GFP into WT mice on CDAA or WD.Correcting Timp3 levels resulted in a significantdecrease in TACE activity (Fig. 5A,B) and TNF-a ac-tivity (Fig. 5C,D) in both diet models. TACE andTNFa activity showed no significant changes inNOX22/2 mice on either diet. Because KCs alsoexpress NOX2, a group of mice in both models werealso injected by GdCl3 to inhibit the function of thesecells and elucidate the respective role of HSCs andKCs in AGEs-induced TACE activation in NASH.Macrophage inhibition did not change TACE activityin either diet model, but showed a decreasing trendfor active TNF-a. Triglyceride (TG) content of theliver has significantly increased on both diets in bothgenotypes, and Ad-Timp3 did not affect TG content(Fig. 5E,F; P < 0.05; N 5 6 for each group).

Interestingly, expression of RAGE showed anincrease in livers of WT mice on the CDAA diet(2.34- 6 0.17-fold; P < 0.05; Supporting Fig. 2), andthis was attenuated by Ad-Timp3 (0.86- 6 0.14; P <0.05). No increase in RAGE was noted in NOX22/2

mice.CDAA and WD Induce Fibrogenic Activity in

WT Mice, Which Is Attenuated After Ad-Timp3Injection and in NOX22/2 Mice. Correction ofTimp3 may affect fibrogenesis. Therefore, procollagena1(I), alpha smooth muscle actin (a-SMA), and

transforming growth factor beta (TGF-b) were tested byreal-time qPCR, and liver tissues were examined afterPicro Sirius Red staining and hydoxyproline assay. Pro-collagen a1(I), a-SMA, and TGF-b expression were sig-nificantly elevated both in CDAA-fed (5.91- 6 1.85-fold; P < 0.01; 1.41- 6 0.07-fold; P < 0.01; and 3.37-6 0.76-fold, respectively; P < 0.05; N 5 6; Fig. 6A) andWD-fed mice (1.58- 6 0.12-fold; P < 0.05; 1.75- 6

0.001-fold; P < 0.05; and 1.51- 6 0.13-fold; P < 0.05).Expression of these fibrogenic transcripts significantlydecreased after Ad-Timp3 injection in both mouse mod-els: in CDAA-diet–fed mice procollagen a1(I) (to 1.54-6 0.93-fold; P < 0.05); a-SMA (to 1.05- 6 0.31-fold; P< 0.05); and TGF-b (to 0.96- 6 0.1-fold; P < 0.05). InWD-fed mice after Ad-Timp3 transduction, procollagena1(I) decreased (to 0.76- 6 0.1-fold; P < 0.05), as dida-SMA (to 0.62- 6 0.09; P < 0.05) and TGF-b (to0.63- 6 0.1-fold; P < 0.05). These data were confirmedalso by the Picro Sirius Red staining and morphometry(Fig. 6C-E) and hydroxyproline assay (Fig. 7). Injectionof GdCl3 caused a decrease in expression of a-SMA andTGF-b in the CDAA model, but not in the WD. InNOX22/2 mice on the NASH diets, fibrosis was less pro-nounced with attenuated procollagen a1(I) and a-SMAexpression as well as a significant decrease in Picro SiriusRed positive area (Figs. 6 and 7; P < 0.05) and hydroxy-proline quantity (Fig. 7; P < 0.01).

Discussion

IR and DM are major risk factors in the progressionof NASH. Several important aspects of the pathogene-sis are well defined, but how diabetes creates a proin-flammatory and fibrogenic milieu is less clear. Here,we have shown that AGEs/NOX2/ROS-induceddecrease in Sirt1 expression in HSCs culminates in aninduction of TACE and, consequently, TNF-a activity.Loss of Sirt1 was shown to be associated with meta-bolic diseases, such as type 2 diabetes, atherosclerosis,21

and NASH.21 On the other hand, hepatic Sirt1 defi-ciency was described to induce IR, and an accumula-tion of ROS was also detected in the liver.10 Treatingthese mice with antioxidants reversed the phenotype,suggesting the key role of ROS in this process. Defin-ing the source of ROS and analyzing the downstreameffects of Sirt1 down-regulation are of greatimportance.

AGEs are formed at a highly accelerated rate duringhyperglycemia by a nonenzymatic glycation of serum pro-teins. The process begins with the conversion of reversibleSchiff base adducts to more stable, covalently bound prod-ucts and, eventually, the irreversibly bound moieties known


as AGEs are formed.22,23 AGE levels in serum or liver werefound to be elevated in patients with steatohepatitis, com-pared to HCs or patients with simple steatosis.24 Interactionof AGEs with the associated cell-surface receptor, RAGE,has been linked to induction of OS in the liver,25 and itwas postulated that activation of one of the NOXs was thesource of ROS production. However, the specific NOX hasnot been identified. Here, we described the key role ofNOX2 activation in AGE-induced ROS production and

the down-regulation of Sirt1/Timp3 pathways. NOX2 is aphagocytic NADPH oxidase, and we and others haveshown that it is highly expressed and enzymatically active inHSC during liver fibrosis.18,26 NOX2 directly inducesHSC activation and production of collagen I by inducingpromoter activity by H2O2, and NOX22/2 mice developsignificantly less fibrosis in bile duct ligation18 and CCl4

26

models. AGEs have been described to induce ROS forma-tion by NOXs in other systems,27,28 and the postulated

Fig. 5. Correction of low Timp3 expression by Ad-Timp3 in WT mice on CDAA and WD ameliorated TACE and TNF-a activity. WT mice on CDAAor WD were injected with Ad-Timp3 or Ad-GFP, and TACE activity in the liver was tested. TACE activity significantly increased in WT mice on theCDAA diet, compared to the CSAA group (mean 6 SE; *P < 0.05; N 5 6), and in WD-fed mice (mean 6 SED; *P < 0.05; N 5 6). This wassignificantly attenuated in Ad-Timp3-injected mice in both diets (*P < 0.05; N 5 6, each group). NOX22/2 mice did not exhibit an increase inTACE activity on either diet (*P < 0.05; N 5 6) (A and B). Gadolinium chloride (Gad) was injected into a group of mice on both diets and ge-notypes to assess the role of KCs. TACE activity in the liver was not significantly affected by inhibiting the macrophages. TNF-a level wasassessed by enzyme-linked immunosorbent assay in the above-described conditions. Compared to the mice on the chow and CSAA diets, WTCDAA-fed or WD-fed mice had significantly higher TNF-a production (*P < 0.05; N 5 6), and Ad-Timp3 prevented this induction (*P < 0.05).In NOX22/2 mice on either diet, no increase in TNF-a production was noted (C and D). In Gad-treated mice, there was a trend toward lowerTNF-a, albeit not significant. TG content was tested in all groups and both genotypes (E and F). Both diets induced a significant accumulation ofTG (*P < 0.05; N 5 6); however, this was not improved by lack of NOX2, correcting Timp3 levels, or Gad injection. SE, standard error of mean.


mechanism could involve activation of protein kinase C(PKC)-a in the kidney28 or PKC-d in neuronal tissue29 byAGEs. Whether or not in active HSCs other NOXs canelicit a similar response to AGEs could be further

investigated in the future. Because p47phox KO HSCs hada decrease in ROS production after incubation with AGEs,it is possible that NOX1 also plays a role as a regulator ofSirt1,25 because p47phox is a common subunit to both

Fig. 6. CDAA-induced and WD-induced fibrogenic activity is attenuated by Ad-Timp3 injection and also in NOX22/2 livers. Procollagen a1(I),a-SMA, and TGF-b expression was analyzed by real-time qPCR. Messenger RNA levels of these transcripts were significantly increased in CDAA-fed and WD-fed mouse livers injected with the control vector (**P < 0.01; *P < 0.05, respectively; N 5 6) and attenuated by Ad-Timp3 injec-tion (*P < 0.05; N 5 6). No significant induction of these transcripts was noted in the NOX22/2 group on either diet. Gadolinium injectionattenuated the increase in a-SMA and TGF-b in CDAA mice (*P < 0.05; ***P < 0.001), but not in the WD-fed mice (A and B). After picrosir-ius staining (C) and ImageJ analysis (D) in WT CDAA-fed or WD-fed and control vector-injected mice, the fibrotic area (red, pericellular fibrosis)was significantly increased (*P < 0.05; **P < 0.01; N 5 6). In mice injected with Ad-Timp3, the picro sirius–positive area decreased (*P <0.05). Also, in NOX22/2 mice on both diets, the fibrotic area was significantly lower than in WT mice (expressed as fold compared to samplesafter chow diet; *P < 0.05; **P < 0.01; N 5 6).


NOX1 and 2. In our study, we demonstrated that, inNOX22/2 mice, there was a significant attenuation of fi-brosis on both diets. Because KCs also express NOX2, wetreated mice with gadolinium to inhibit their function. Wefound that this did not interfere with TACE activity in theliver. TACE activity of KCs was not induced by AGEs invitro and therefore their profibrogenic role during NASHcould be attributed to other, more dominant mechanisms,such as Toll-like receptor 4–mediated induction of HSCs.30

Steatosis was not affected by the lack of NOX2, corre-sponding to earlier data from the methionine- and choline-deficient model,31 nor was it influenced by correctingTimp3. This suggests that AGEs mainly induced oxidativeand inflammatory pathways in these models. The role ofSirt1 down-regulation in HSCs during NASH progressionhas not been studied in detail. Because the main source ofTACE activity in the liver are activated HSCs,32 and thenatural regulator of TACE is Timp3, we sought to deter-mine whether low Sirt1 could translate into a down-regula-tion of Timp3 and consequent induction of TACE. First,we confirmed that active HSCs are a potent source ofTACE production in the liver and that exposure to AGEsdirectly induces pathways leading to TACE activation,whereas hepatocytes and KCs do not respond to AGEinduction. Second, AGEs induced a NOX2-dependent

down-regulation of Sirt1 and Timp3, and both of thesetranscripts were down-regulated in humans with NASHand also in mice on both diets. Correlating to this, TACEactivity was increased by AGEs and also in WT mice onCDAA and WD, but not on the CSAA diet (steatosis). Incontrast, no increase in TACE activity was noted inNOX22/2 mice on CDAA and WD, implying thatNOX2-derived ROS plays a key a role in TACE activation.To confirm the role of Timp3 in NASH, we injected agroup of mice on the CDAA diet with Ad-Timp3. Thisresulted in a decline in TACE and TNF-a activity as well asRAGE expression. In addition, Ad-Timp3 transductionalso improved fibrosis with down-regulation of profibro-genic transcripts and lowering of hydroxyproline content.Because TACE targets multiple other cytokines and chemo-kines and their receptors,33 reducing its activity may haveled to a decreased production of fibrogenic mediators. Forinstance, TACE is known to induce ectodomain sheddingof fractalkine (CX3CL1)34; thus, decrease in fractalkinecould have also contributed to the improvement in fibro-genic activity. Mice deficient in Timp3 demonstrated ele-vated levels of TNF-a and developed IR and hepaticsteatosis, mediated by increased TACE activity.10,21

In conclusion, we have demonstrated the centralrole of the Sirt1/Timp3/TACE cascade in AGE-induced proinflammatory and fibrogenic activity inNASH. Modulation of Timp3 or TACE activity couldthus become a successful approach to halt disease pro-gression in NASH.

References

1. Chiang DJ, Pritchard MT, Nagy LE. Obesity, diabetes mellitus, andliver fibrosis. Am J Physiol Gastrointest Liver Physiol 2011;300:G697-G702.

2. Prospective Studies Collaboration, Whitlock G, Lewington S, SherlikerP, Clarke R, Emberson J, Halsey J, et al. Body-mass index and cause-specific mortality in 900 000 adults: collaborative analyses of 57 pro-spective studies. Lancet 2009;373:1083-1096.

3. Vlassara H. Advanced glycation in health and disease: role of the mod-ern environment. Ann N Y Acad Sci 2005;1043:452-460.

4. Vlassara H, Striker GE. AGE restriction in diabetes mellitus: a para-digm shift. Nat Rev Endocrinol 2011;24:526-539.

5. Stitt AW, Jenkins AJ, Cooper ME. Advanced glycation end products anddiabetic complications. Expert Opin Investig Drugs 2002;11:1205-1223.

6. Zong H, Ward M, Stitt AW. AGEs, RAGE, and diabetic retinopathy.Curr Diab Rep 2011;11:244-252.

7. Herranz D, Serrano M. SIRT1: recent lessons from mouse models. NatRev Cancer 2010;10:819-823.

8. Tao R, Wei D, Gao H, Liu Y, DePinho RA, Dong XC. Hepatic FoxOsregulate lipid metabolism via modulation of expression of the nicotina-mide phosphoribosyltransferase gene. J Biol Chem 2011;286:14681-14690.

9. Schug TT, Li X. Sirtuin 1 in lipid metabolism and obesity. Ann Med2011;43:198-211.

10. Wang RH, Kim HS, Xiao C, Xu X, Gavrilova O, Deng CX. HepaticSirt1 deficiency in mice impairs mTorc2/Akt signaling and results in

Fig. 7. Hydroxyl proline content decreased after Ad-Timp3 in theCDAA-fed and WD-fed mice. OH-proline incorporation assay was per-formed to assess the amount of collagen in the liver. Compared tomice on the chow or CSAA diet, CDAA-fed (A) and WD-fed mice (B)displayed a significant increase in OH-proline incorporation (*P <0.05; N 5 6). Ad-Timp3 lowered OH-proline content in WT mice inboth models. In NOX22/2 mice, OH-proline incorporation was alsosignificantly lower in both models (*P < 0.05; N 5 6).


hyperglycemia, oxidative damage, and insulin resistance. J Clin Invest2011;121:4477-4490.

11. Purushotham A, Xu Q, Li X. Systemic SIRT1 insufficiency results indisruption of energy homeostasis and steroid hormone metabolismupon high-fat-diet feeding. FASEB J 2012;26:656-667.

12. Cardellini M, Menghini R, Martelli E, Casagrande V, Marino A, RizzaS, et al. TIMP3 is reduced in atherosclerotic plaques from subjects withtype 2 diabetes and increased by SirT1. Diabetes 2009;58:2396-2401.

13. Federici M, Hribal ML, Menghini R, Kanno H, Marchetti V, PorzioO, et al. Timp3 deficiency in insulin receptor-haploinsufficient micepromotes diabetes and vascular inflammation via increased TNF-alpha.J Clin Invest 2005;115:3494-3505.

14. Menghini R, Menini S, Amoruso R, Fiorentino L, Casagrande V, Mar-zano V, et al. Tissue inhibitor of metalloproteinase 3 deficiency causeshepatic steatosis and adipose tissue inflammation in mice. Gastroenter-ology 2009;136:663-672.e4.

15. Kodama Y, Kisseleva T, Iwaisako K, Miura K, Taura K, De Minicis S,et al. c-Jun N-terminal kinase-1 from hematopoietic cells mediates pro-gression from hepatic steatosis to steatohepatitis and fibrosis in mice.Gastroenterology 2009;137:1467-1477.e5.

16. Charlton M, Krishnan A, Viker K, Sanderson S, Cazanave S, McCo-nico A, et al. Fast food diet mouse: novel small animal model ofNASH with ballooning, progressive fibrosis, and high physiological fi-delity to the human condition. Am J Physiol Gastrointest Liver Physiol2011;301:G825-G834.

17. Geerts A, Niki T, Hellemans K, De Craemer D, Van Den Berg K,

Lazou JM, et al. Purification of rat hepatic stellate cells by side scatter-

activated cell sorting. HEPATOLOGY 1998;27:590-598.

18. Jiang JX, Venugopal S, Serizawa N, Chen X, Scott F, Li Y, et al.Reduced nicotinamide adenine dinucleotide phosphate oxidase 2 playsa key role in stellate cell activation and liver fibrogenesis in vivo. Gas-troenterology 2010;139:1375-1384.

19. Miura K, Kodama Y, Inokuchi S, Schnabl B, Aoyama T, OhnishiH, et al. Toll-like receptor 9 promotes steatohepatitis by inductionof interleukin-1beta in mice. Gastroenterology 2010;139:323-334.e7.

20. Wang B, Majumder S, Nuovo G, Kutay H, Volinia S, Patel T, et al.

Role of microRNA-155 at early stages of hepatocarcinogenesis induced

by choline-deficient and amino acid-defined diet in C57BL/6 mice. HE-

PATOLOGY 2009;50:1152-1161.

21. Purushotham A, Schug TT, Xu Q, Surapureddi S, Guo X, Li X. Hepa-tocyte-specific deletion of SIRT1 alters fatty acid metabolism andresults in hepatic steatosis and inflammation. Cell Metab 2009;9:327-338.

22. Glomb MA, Monnier VM. Mechanism of protein modification by gly-oxal and glycolaldehyde, reactive intermediates of the Maillard reaction.J Biol Chem 1995;270:10017-10026.

23. Bucala R, Cerami A. Advanced glycosylation: chemistry, biology, andimplications for diabetes and aging. Adv Pharmacol 1992;23:1-34.

24. Hyogo H, Yamagishi S, Iwamoto K, Arihiro K, Takeuchi M, Sato T,et al. Elevated levels of serum advanced glycation end products inpatients with non-alcoholic steatohepatitis. J Gastroenterol Hepatol2007;22:1112-1119.

25. Guimar~aaes EL, Empsen C, Geerts A, van Grunsven LA. Advancedglycation end products induce production of reactive oxygen species viathe activation of NADPH oxidase in murine hepatic stellate cells. JHepatol 2010;52:389-397.

26. Paik YH, Iwaisako K, Seki E, Inokuchi S, Schnabl B, Osterreicher CH,et al. The nicotinamide adenine dinucleotide phosphate oxidase (NOX)homologues NOX1 and NOX2/gp91(phox) mediate hepatic fibrosis inmice. HEPATOLOGY 2011;53:1730-1741.

27. Wautier MP, Chappey O, Corda S, Stern DM, Schmidt AM, WautierJL, et al. Activation of NADPH oxidase by AGE links oxidant stress toaltered gene expression via RAGE. Am J Physiol Endocrinol Metab2001;280:E685-E694.

28. Thallas-Bonke V, Thorpe SR, Coughlan MT, Fukami K, Yap FY, Sour-ris KC, et al. Inhibition of NADPH oxidase prevents advanced glyca-tion end product-mediated damage in diabetic nephropathy through aprotein kinase C-alpha-dependent pathway. Diabetes 2008;57:460-469.

29. Nitti M, Furfaro AL, Traverso N, Odetti P, Storace D, Cottalasso D,et al. PKC delta and NADPH oxidase in AGE-induced neuronal death.Neurosci Lett 2007;416:261-265.

30. Seki E, De Minicis S, Osterreicher CH, Kluwe J, Osawa Y, BrennerDA, Schwabe RF. TLR4 enhances TGF-beta signaling and hepatic fi-brosis. Nat Med 2007;13:1324-1332.

31. De Minicis S, Seki E, Paik YH, Osterreicher CH, Kodama Y, Kluwe J,et al. Role and cellular source of nicotinamide adenine dinucleotidephosphate oxidase in hepatic fibrosis. HEPATOLOGY 2010;52:1420-1430.

32. Bourd-Boittin K, Basset L, Bonnier D, L’helgoualc’h A, Samson M,Th�eret N. CX3CL1/fractalkine shedding by human hepatic stellatecells: contribution to chronic inflammation in the liver. J Cell MolMed 2009;13:1526-1535.

33. Cerretti DP. Characterization of the tumour necrosis factor alpha-con-verting enzyme, TACE/ADAM17. Biochem Soc Trans 1999;27:219-223.

34. Garton KJ, Gough PJ, Blobel CP, Murphy G, Greaves DR, DempseyPJ, Raines EW. Tumor necrosis factor-alpha-converting enzyme(ADAM17) mediates the cleavage and shedding of fractalkine(CX3CL1). J Biol Chem 2001;276:37993-38001.


advanced glycation endproducts induce fibrogenic activity in nonalcoholic steatohepatitis by...

Documents