1521-0103/356/3/604–614$25.00 http://dx.doi.org/10.1124/jpet.115.229906THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS J Pharmacol Exp Ther 356:604–614, March 2016Copyright ª 2016 by The American Society for Pharmacology and Experimental Therapeutics

ASB14780, an Orally Active Inhibitor of Group IVA PhospholipaseA2, Is a Pharmacotherapeutic Candidate for Nonalcoholic FattyLiver Diseases

Shiho Kanai, Keiichi Ishihara, Eri Kawashita, Toshiyuki Tomoo, Kazuhiro Nagahira,Yasuhiro Hayashi, and Satoshi AkibaDepartment of Pathological Biochemistry, Kyoto Pharmaceutical University, Yamashina-ku, Kyoto, Japan (S.K., K.I., E.K., S.A.);and Asubio Pharma Co., Ltd., Chuo-ku, Kobe, Japan (T.T., K.N., Y.H.)

Received October 7, 2015; accepted December 15, 2015

ABSTRACTWe have previously shown that high-fat cholesterol diet (HFCD)–induced fatty liver and carbon tetrachloride (CCl4)–inducedhepatic fibrosis are reduced in mice deficient in group IVAphospholipase A2 (IVA-PLA2), which plays a role in inflammation.We herein demonstrate the beneficial effects of ASB14780(3-[1-(4-phenoxyphenyl)-3-(2-phenylethyl)-1H-indol-5-yl]propanoicacid 2-amino-2-(hydroxymethyl)propane-1,3-diol salt), an orallyactive IVA-PLA2 inhibitor, on the development of fatty liver andhepatic fibrosis in mice. The daily coadministration of ASB14780markedly ameliorated liver injury and hepatic fibrosis following 6weeks of treatment with CCl4. ASB14780 markedly attenuatedthe CCl4-induced expression of smooth muscle a-actin (a-SMA)protein and the mRNA expression of collagen 1a2, a-SMA, and

transforming growth factor-b1 in the liver, and inhibited theexpression of monocyte/macrophage markers, CD11b andmonocyte chemotactic protein-1, while preventing the recruit-ment of monocytes/macrophages to the liver. Importantly,ASB14780 also reduced the development of fibrosis even inmatured hepatic fibrosis. Additionally, ASB14780 also reducedHFCD-induced lipid deposition not only in the liver, but also inalready established fatty liver. Furthermore, treatment withASB14780 suppressed the HFCD-induced expression of lipo-genic mRNAs. The present findings suggest that an IVA-PLA2inhibitor, such as ASB14780, could be useful for the treatmentof nonalcoholic fatty liver diseases, including fatty liver andhepatic fibrosis.

IntroductionNonalcoholic fatty liver disease (NAFLD) is a complex liver

disease with a spectrum of hepatic histopathological changesranging from mere intracellular fat deposition to nonalcoholicsteatohepatitis (NASH), which may progress into hepaticfibrosis, cirrhosis, or hepatocellular carcinoma (Kleineret al., 2005; Clark, 2006; Marchesini et al., 2008). NAFLDand NASH are generally treated with weight reduction,exercise, and dietary changes. However, the curative effectsof these nonpharmacotherapies are minimal, owing to highrates of noncompliance. In addition to guidance about changesin lifestyle, treatment with statins, antihypertensive agents,

and b-blockers for concurrent metabolic disorders of obesity isalso known to ameliorate NAFLD. However, these treatmentshave limited effects on the morbidity andmortality of patientswith NAFLD and NASH.Recently, pioglitazone, a synthetic peroxisome proliferator-

activated receptor g activator, has had considerable successin the treatment of NAFLD and NASH as a liver-directedpharmacotherapy. However, the positive effects of pioglita-zone on NAFLD and NASH remain controversial. Althoughseveral clinical trials have demonstrated benefits for piogli-tazone in patients with NAFLD and NASH (Belfort et al.,2006; Aithal et al., 2008; Sanyal et al., 2010), long-termtreatment with pioglitazone is associated with weight gain(Sanyal et al., 2010), an increased risk of congestive cardiacfailure (Lago et al., 2007) and bladder cancer (Piccinni et al.,2011), and reduced bone density (Lecka-Czernik, 2010). Thus,alternative pharmacotherapeutic options that improve thehistologic features of NAFLD and NASH are required.Inflammation plays a central role in the development of

NAFLD and NASH (Tilg and Moschen, 2010). The lipotoxicity

This study was supported by a Grant-in-Aid for Scientific Research (C)[Project No. 26460110 to S.A.) from the Ministry of Education, Science, Cultureand Sports of Japan, and a grant from the Suzuken Memorial Foundation[Grant 13-040 to K. I.].

S.K. and K.I. contributed equally to this work.dx.doi.org/10.1124/jpet.115.229906.s This article has supplemental material available at jpet.aspetjournals.org.

ABBREVIATIONS: ACC, acetyl-CoA carboxylase; ALT, alanine aminotransferase; ANOVA, analysis of variance; ASB14780, 3-[1-(4-phenoxyphenyl)-3-(2-phenylethyl)-1H-indol-5-yl]propanoic acid 2-amino-2-(hydroxymethyl)propane-1,3-diol salt; AST, aspartate aminotransferase; CCl4, carbontetrachloride; FAS, fatty acid synthase; HFCD, high-fat cholesterol diet; HPC, hydroxypropyl cellulose; IVA-PLA2, group IVA phospholipase A2; MCP-1,monocyte chemotactic protein-1; NAFLD, nonalcoholic fatty liver disease; NASH, nonalcoholic steatohepatitis; ND, normal diet; PBS(2), phosphate-buffered saline without calcium chloride and magnesium chloride; RANTES, regulated on activation normal T cell expressed and secreted; RT-PCR,reverse-transcription polymerase chain reaction; SCD-1, stearoyl-CoA desaturase 1; a-SMA, smooth muscle a-actin; sPLA2, secreted phospholipaseA2s; SREBP1c, sterol regulatory element-binding protein 1c; TBS-T, Tris-buffered saline containing 0.05% Tween 20.

604

http://jpet.aspetjournals.org/content/suppl/2015/12/23/jpet.115.229906.DC1Supplemental material to this article can be found at:

at ASPE

T Journals on M

arch 24, 2020jpet.aspetjournals.org

Dow

nloaded from

of hepatic fats induces inflammatory and oxidative stresses(Takaki et al., 2014). Therefore, molecules that mediateinflammation may be promising candidate pharmacothera-peutic targets for these disorders. In the present study, weinvestigated the possibility of targeting group IVA phospho-lipase A2 (IVA-PLA2 or cytosolic PLA2a), which catalyzes thefirst step in the arachidonic acid cascade (Kita et al., 2006), forthe treatment of these conditions. We previously demon-strated dramatic attenuation of high-fat cholesterol diet(HFCD)–induced fatty liver and carbon tetrachloride (CCl4)–induced hepatic fibrosis in IVA-PLA2–deficient mice (Ii et al.,2009; Ishihara et al., 2012). These findings led us to hypoth-esize that IVA-PLA2 inhibitors may be promising candidatesfor the treatment of NAFLD and NASH. IVA-PLA2–specificinhibitors have already been developed by Wyeth Pharma-ceuticals, Shionogi Pharmaceuticals, Astra Zeneca, and theKokotos and Dennis groups, and include indole derivatives(McKew et al., 2003, 2006, 2008; Lee et al., 2007), pyrrolidine-based compounds (Seno et al., 2000, 2001; Ono et al., 2002;Flamand et al., 2006), propan-2-ones (Connolly et al., 2002;Ludwig et al., 2006; Hess et al., 2007; Fritsche et al., 2008),and 2-oxoamide compounds (Kokotos et al., 2002, 2004;Stephens et al., 2006; Six et al., 2007), respectively. Sincenone of these inhibitors is orally active, the prospect of usingan IVA-PLA2 inhibitor has been limited. However, we re-cently developed an indole derivative, ASB14780 (3-[1-(4-phenoxyphenyl)-3-(2-phenylethyl)-1H-indol-5-yl]propanoic acid2-amino-2-(hydroxymethyl)propane-1,3-diol salt), as a specificinhibitor of IVA-PLA2 (Tomoo et al., 2014). It has desirablebioavailability and oral efficacy, because it does not containhydrophobic long alkyl chains (Tomoo et al., 2014). The specific-ity of ASB14780 has been demonstrated, and ASB14780inhibited human IVA-PLA2 with an IC50 value of 20 nM,which was excellent compared with other known secretedphospholipase A2s (sPLA2s), such as sPLA2-IA (cobra venom),sPLA2-IIA (crotalus venom), sPLA2-III (bee venom), andsPLA2-IB (porcine pancreas), which have IC50 values of∼10 mM in vitro (Tomoo et al., 2014). Based on these findingsabout ASB14780, we examined its effects on the progression offatty liver and hepatic fibrosis in mice.

Materials and MethodsAnimals. Male 6- to 8-week-old C57BL/6N mice were purchased

from Japan SLC Inc. (Shizuoka, Japan). All mice were housed at,5/cage with a 12-hour light/12-hour dark cycle and ad libitum accessto food and water. The study protocol was approved by the Experi-mental Animal Research Committee at Kyoto Pharmaceutical Uni-versity (Kyoto, Japan).

Preparation of ASB14780 Suspension. The IVA-PLA2 inhibi-tor ASB14780 was synthesized by Asubio Pharma Co., Ltd. (Hyogo,Japan) and was suspended in 0.5% hydroxypropyl cellulose (HPC) insterile water. The inhibitor was administered daily by oral gavage inaccordance with the body weight of each animal (0.1 or 0.3 g/kg bodyweight). Mice in the control group received 0.5% HPC via the sameadministration route.

Animal Model of CCl4-Induced Hepatic Fibrosis, andTreatment with ASB14780. Mice (8 weeks old) were randomlydivided into four groups: a control group, a CCl4 alone group, a CCl4plus low-dose ASB14780 group, and a CCl4 plus high-dose ASB14780group. Liver fibrosis was induced by twice weekly intraperitonealadministration of CCl4 at 0.31 ml/kg body weight (diluted in corn oil;Sigma-Aldrich, St. Louis, MO) for 6 weeks. ASB14780 was orally

administered daily at 0.1 or 0.3 g/kg body weight, and was adminis-tered 1 hour prior to CCl4 injections. Control animals received 0.5%HPC via the same route of administration. In the experimentsperformed to elucidate the effects of ASB14780 on the progression ofestablished hepatic fibrosis, mice (8 weeks old) were randomly dividedinto the following four groups: groups A andB, treatedwith CCl4 alonefor 6 (group A) and 9 weeks (group B), respectively; group C, treatedwith CCl4 and ASB14780 (0.3 g/kg) for 9 weeks; and group D, treatedwith CCl4 for 6 weeks, followed by treatmentwith CCl4 and ASB14780for 3 weeks. The mice were sacrificed 48 hours after the last injection.

Animal Model of HFCD-Induced Fatty Liver and Treatmentwith ASB14780. The mice (6 weeks old) were randomly divided intothe following groups: a normal diet group (ND; 5.3% fat) treated with0.5% HPC (group 1), an ND group treated with ASB14780 (0.3 g/kg,once daily; group 2), an HFCD group (20% fat and 1.25% cholesterol;Research Diets Inc., Brunswich, NJ) treated with 0.5%HPC (group 3),and an HFCD group treated with ASB14780 (0.3 g/kg, daily; group 4)for 16 weeks. In another group, the mice were fed the HFCD for 16weeks, andwere administered ASB14780 (0.3 g/kg, daily) for the last 6weeks (group 5). At the end of a 6-hour fasting period, the mice weresacrificed, and parameters related to fatty liver diseasewere assessed.The food consumption of each cage was measured daily at 5 PM duringthe last week of administration.

Serum Biomarker Measurements. Blood samples were col-lected from the inferior vena cava, and the serum was prepared bycentrifugation at 10,000� g for 10 minutes at room temperature. Theactivities of serum alanine aminotransferase (ALT) and aspartateaminotransferase (AST) were measured using Transaminase C II-Test kits (Wako Pure Chemical Industries, Osaka, Japan), and thelevel of triglycerides was determined using Triglyceride E-test Wakokits (Wako Pure Chemical Industries).

CYP2E1 Activity Assay. The CYP2E1 activity was measured asdescribed previously (Chang et al., 2006). Vehicle (HPC)-treated micewith or without ASB14780 treatment were then sacrificed 48 hoursafter the last injection. After perfusion with ice-cold saline via theheart, the livers (30 mg) were homogenized in a buffer containing50mMTris-HCl (pH 7.4) and 0.15MKCl using a Dounce Tissue Grinder(Wheaton, Millville, NJ). The liver homogenate was centrifuged at9000 � g for 20 minutes, then the supernatant was further ultra-centrifuged at 105,000 � g for 60 minutes. The microsomal pelletobtained was suspended in 90 ml of assay buffer containing 50 mMKPO4 (pH 7.4). The concentration of microsomal protein was mea-sured by a Bradford protein assay. The liver microsomal samples(45 mg of protein) were incubated in reaction buffer containing 100 mMp-nitrophenol, NADPH regenerating system solution A (BD Biosci-ences, Boston, MA), and NADPH regenerating system solution B (BDBiosciences) at 37°C for 90 minutes in a water bath. To assess theeffect of ASB14780 on CYP2E1 activity in vitro, the liver microsomalsamples (45mg of protein)were incubated in reaction buffer containing100 mM p-nitrophenol, NADPH regenerating system solution A, andNADPH regenerating system solutionB at 37°C for 120minutes in thepresence of ASB14780. The enzymatic reaction was stopped by theaddition of 3.3% (v/v) trichloroacetic acid, and samples were centri-fuged at 10,000 � g for 5 minutes. The absorbance of the supernatantat 535 nmwasmeasured after the addition of 0.67MNaOH. The blankwas prepared by adding the complete incubation mixture with heat-inactivated liver microsomal samples (95°C, 5 minutes). CYP2E1activity was calculated as the p-nitrophenol hydroxylation activityand was expressed as nanomoles of product formed/(minute � mg ofmicrosomal protein).

Histologic Analysis. The mice were transcardially perfused withsaline, and excised livers were immersed in 10% buffered formalde-hyde for 1 week. The fixed livers were embedded in paraffin formicrotome slicing into 5-mm-thick sections. The tissue sections weremounted onto MAS-coated Superfrost glass slides (Matsunami Glass,Osaka, Japan), and were deparaffinized and stained with picricacid–sirius red or H&E (Wako Pure Chemical Industries). Stainedsections were photographed using a microscope (model IX71; Olympus,

Suppression of NAFLD and NASH Using an IVA-PLA2 Inhibitor 605

at ASPE

T Journals on M

arch 24, 2020jpet.aspetjournals.org

Dow

nloaded from

Tokyo, Japan) with a digital camera. Images were taken at fullresolution with a single image dimension set at 1360 � 1024 pixels.Collagen fibers were stained with sirius red, and were quantified bymeasuring the red areas using the U.S. National Institutes of HealthImageJ v1.47 software program (http://rsb.info.nih.gov/ij).

Immunohistochemistry. Frozen liver sections (10 mm thick)were stained with anti–smooth muscle a-actin (a-SMA; Sigma-Aldrich) or anti-F4/80 (CI:A3-1 clone; Abcam, Cambridge, UK)antibodies. The sections were then incubated at 90°C in preheatedRetrievagen A at pH 6.0 (BD Biosciences) for 10 minutes. Afterincubation with Mouse-on-Mouse blocking solution (Vector Laborato-ries, Burlingame, CA) for 1 hour, the slides were incubated withblocking solution containing 10% horse (for a-SMA) or rabbit (forF4/80) serum in phosphate-buffered saline without calcium chlorideand magnesium chloride [PBS(2)] containing 0.3% Triton X-100 andavidin-blocking solution (Vector Laboratories) for 1 hour at roomtemperature. After being washed with PBS(2), the sections wereincubated overnight with biotin-blocking solution (Vector Laborato-ries) and a-SMA (1:1000 dilution) or F4/80 antibodies (1:100 dilution)in a humidified chamber at 4°C. The slides were washed with PBS(2),and endogenous peroxidases were quenched by incubation with 0.3%hydrogen peroxide inmethanol for 30minutes. The sections were thenincubated with horse biotinylated mouse IgG (1:1000 dilution; VectorLaboratories) or rabbit biotinylated rat IgG (1:800 dilution; VectorLaboratories) for 30 minutes. Detection of the antibody-antigencomplexes was accomplished using a Vectastain Elite ABC kit (VectorLaboratories) and a metal-enhanced DAB substrate kit (ThermoScientific, Rockland, IL).

Western Blot Analysis. Frozen liver specimens were homoge-nized in ice-cold lysis buffer containing 20 mM Tris-HCl (pH 7.6),150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Nonidet P-40, and 1%sodium deoxycholate supplemented with a complete protease in-hibitor cocktail tablet (Roche Diagnostics, Indianapolis, IN) andphosphatase inhibitor cocktail solution (Wako Pure Chemical In-dustries). Protein extracts were collected after centrifugation at10,000 � g for 10 minutes at 4°C. The protein concentrations weredetermined using the Bradford method. Protein samples (90 mg) weredenatured and electrophoresed on 7.5% (for IVA-PLA2) and 10% (fora-SMA and a-tubulin) SDS-PAGE gels, and were then transferredonto nitrocellulosemembranes. After being blockedwith 3% skimmilkin Tris-buffered saline containing 0.05% Tween 20 (TBS-T), themembranes were incubated with antibodies against IVA-PLA2 (CellSignaling Technology, Danvers, MA; diluted 1:1000 with Toyobo CanGet Signal solution 1 (Toyobo, Osaka, Japan)), a-SMA (Sigma-Aldrich;diluted 1:1000 with blocking solution), or a-tubulin (Cell SignalingTechnology; diluted 1:2000 with blocking solution) for 12 hours at 4°C.After being washed with TBS-T, the membranes were incubated withhorseradish peroxidase–conjugated goat anti-rabbit IgG (for IVA-PLA2; Santa Cruz Biotechnology, Santa Cruz, CA; diluted 1:5000 withCan Get Signal solution 2) or goat anti-mouse IgG (for a-SMA ora-tubulin; Santa Cruz Biotechnology; diluted 1:5000 with 0.3% skimmilk in TBS-T) for 1 hour. After being washed again, immunoreactivebands were detected using Chemi-Lumi One Super (Nacalai Tesque,Kyoto, Japan) with a LAS-3000mini-image analysis system (Fujifilm,Tokyo, Japan). The band intensities were quantified using the ImageJsoftware program.

Isolation of Total RNA and Quantitative Real-TimeReverse-Transcription Polymerase Chain Reaction Analysis.Total RNA was extracted from liver specimens using RNAiso Plus(Takara Bio, Shiga, Japan) according to the manufacturer’s instruc-tions. Total RNA (2 mg) was reverse transcribed using randomhexamers and ReverTra Ace reverse transcriptase. Quantitativereal-time reverse-transcription polymerase chain reaction (RT-PCR)was then performed on cDNA samples using SYBR-Green I (TakaraBio) and an ABI 7500 Real-Time PCR system (Applied Biosystems,Foster City, CA) or a LightCycler Nano Real-Time PCR system (RocheDiagnostics, Mannheim, Germany). The specific primers used areshown in (Supplemental Table S1). PCR was performed for 40 cycles

with 5 seconds of denaturation at 95°C and annealing and extensionat 60 or 62°C for 34 seconds (Supplemental Table S1). The geneexpression was determined using the relative standard curve method,and the expression levels were normalized to those of 18S ribosomalRNA or 36B4 after parallel amplification.

Measurement of Hepatic Lipid Contents. In experiments withthe fatty liver model, hepatic lipids were extracted from livers (50 mg)with chloroform-methanol (2:1, v/v) according to the proceduresdescribed by Folch et al. (1957). The hepatic triglyceride level wasthen determined using a commercial assay kit (Wako Pure ChemicalIndustries).

Statistical Analysis. Data are indicated as themean6 S.E. Sincethe data were unpaired, differences among means were analyzedusing one-way analysis of variance (ANOVA) followed by the Tukey-Kramer post-hoc test. P , 0.05 was considered as the lowest level ofsignificance. Statistical calculations were performed using Macstatistics version. 2.0 (Excel addin) for Macintosh (ESUMI Co., Tokyo,Japan).

ResultsASB14780 Reduces CCl4-Induced Hepatotoxicity in

Mice. In a previous study, the hepatotoxicity induced bychronic administration of CCl4 was less extensive in IVA-PLA2–deficient mice than in wild-type mice (Ishihara et al.,2012). To evaluate the beneficial effects of the present orallyactive inhibitor of IVA-PLA2 on CCl4-induced liver damage(Fig. 1), mice were administered ASB14780 daily (0.1 or 0.3 g/kgbody weight) and CCl4 twice weekly (0.31 ml/kg bodyweight) for 6 weeks. The body and liver weights are summa-rized in Table 1. No obvious adverse effects were observed inmice treated with ASB14780 at 0.3 g/kg body weight. Incontrast, the liver weights and liver-to-body-weight ratioswere significantly increased in mice treated with CCl4 alone(Table 1).The serum levels of ALT and AST, markers of liver injury,

were significantly increased by chronic CCl4 administration(Fig. 1A). The increases in the serum levels of ALT and ASTinduced by CCl4 were significantly and dose-dependentlyreduced by oral administration of ASB14780 (Fig. 1A).Histologic H&E staining showed that damaged areas wereobserved surrounding the central veins in CCl4-treated mice,and this damage was dramatically and dose-dependentlyreduced by the oral administration of ASB14780 (Fig. 1B).CYP2E1 in hepatocytes plays a role in CCl4-induced liver

injury through the generation of the CCl3 radical (Noguchiet al., 1982; Avasarala et al., 2006). Similar CYP2E1 activitywas detected in the livers of control and ASB14780-treatedmice in the absence of CCl4 (Fig. 1C). To further evaluate thedirect effects of ASB14780 on CYP2E1 activity, the hepaticmicrosomal fraction was incubated in the presence ofASB14780. As shown in Fig. 1D, no inhibition of CYP2E1activity was detected in the presence of ASB14780, suggestingthat ASB14780 did not affect the generation of the CCl3radical in the treatment groups. Taken together with ourprevious observations showing a similar reduction of liverinjury in IVA-PLA2–deficient mice (Ishihara et al., 2012),these data suggest that IVA-PLA2 deficiency or inhibitionreduces CCl4-induced hepatotoxicity.ASB14780 Attenuates CCl4-Induced Hepatic Fibrosis

in Mice. In further experiments, hepatic fibrosis wasassessed in mice after chronic administration of CCl4 bystaining liver sections with picrosirius red and visualizing

606 Kanai et al.

at ASPE

T Journals on M

arch 24, 2020jpet.aspetjournals.org

Dow

nloaded from

the hepatic collagen deposition. As shown in Fig. 2A, themarked collagen deposition induced by CCl4 was reduced inthe livers of ASB14780-treated mice. Semiquantification ofthe picrosirius red–stained areas revealed that the collagenaccumulation in the livers of ASB14780-treated mice wassignificantly reduced in a dose-dependentmanner (Fig. 2C). In

addition, quantitative RT-PCR for Col1a2 mRNA alsorevealed that the CCl4-induced increase in Col1a1 mRNAwas significantly reduced inmice treatedwith ASB14780 (Fig.2E).Chronic treatment with CCl4 leads to the transformation of

hepatic stellate cells to myofibroblast-like cells expressing

Fig. 1. Inhibitory effects of the IVA-PLA2 inhibitorASB14780 on CCl4-induced liver injury. Six-week-oldmale C57BL/6N mice were intraperitoneally admin-istered CCl4 in corn oil at 0.31 ml/kg (2 times/week) for6 weeks. Some mice were concomitantly administered0.1 or 0.3 g/kg of ASB14780 by oral gavage (once a day1 hour before CCl4 injection). All of the mice weresacrificed 48 hours after the last injection. (A) Theserum levels of AST and ALT were determined usingenzymatic assays. (B) Paraffin-embedded liver sec-tions were stained with hematoxylin and eosin; scalebar, 40 mm. (C) CYP2E1 catalyzes the oxidation ofp-nitrophenol in mitochondrial fractions and is involvedin the metabolism and toxicity of CCl4. The CYP2E1activity was determined spectrophotometrically. Thedata are presented as the means 6 S.E. (n = 6–7/group), and differences were identified using anANOVA with a Tukey-Kramer post-hoc analysis(**P , 0.01). (D) CYP2E1 activity was assessed invitro in the absence or presence of ASB14780. Thedata are presented as mean activity (n = 4) relative tocontrol (0 mM) 6 S.E., and differences were identifiedusing an ANOVA with a Tukey-Kramer post-hoc anal-ysis. ASB, ASB14780; CV, central vein; Veh, vehicle.

TABLE 1The effects of ASB14780 and CCl4 on the body weight, liver weight, and liver-to-body-weight ratios of control and treatedanimalsThe data are presented as the means 6 S.E. for six to seven individuals (Tukey-Kramer post-hoc analysis).

ASB (0)/Veh ASB (0.1) /Veh ASB (0.3) / Veh ASB (0)/CCl4 ASB (0.1) /CCl4 ASB (0.3) /CCl4

Body weight (g) 27.6 6 0.9 26.0 6 0.5 26.2 6 0.5 26.2 6 0.7 24.8 6 0.2 24.4 6 0.8Liver weight (g) 1.20 6 0.06 1.15 6 0.08 1.27 6 0.05 1.63 6 0.09* 1.42 6 0.12 1.50 6 0.08Liver weight /body weight (%) 4.35 6 0.21 4.41 6 0.29 4.86 6 0.17 6.2 6 0.2* 5.76 6 0.53 6.17 6 0.32

ASB, ASB14780; Veh, vehicle.*P , 0.05 versus ASB(0)/Veh.

Suppression of NAFLD and NASH Using an IVA-PLA2 Inhibitor 607

at ASPE

T Journals on M

arch 24, 2020jpet.aspetjournals.org

Dow

nloaded from

a-SMA (Zerbe and Gressner, 1988; Tsukamoto et al., 1995). Insubsequent immunohistochemical (Fig. 2B), Western blot(Fig. 2D), and quantitative RT-PCR (Fig. 2E) analyses,a-SMA was not detectable in the liver sections from cornoil–treated control mice regardless of ASB14780 treatment,whereas CCl4 treatment resulted in extensive a-SMA expres-sion in portal areas. The treatment of mice with ASB14780reduced the expression of a-SMA induced by CCl4 exposure(Fig. 2B). These suppressive effects of ASB14780 were con-firmed in Western blot and quantitative RT-PCR analyses ofa-SMA (Fig. 2, D and E). In addition, the mRNA expression ofa potent fibrogenic cytokine, transforming growth factor-b1, inthe liver (Fig. 2E) was increased by CCl4 treatment, and wasdose-dependently suppressed by ASB14780 treatment. Theseresults are consistent with our previous study showing thatCCl4-inducedhepatic fibrosiswas reduced in IVA-PLA2–deficientmice (Ishihara et al., 2012). Thus, the oral administration ofASB14780 inhibited hepatic fibrosis.

CCl4-Induced Recruitment of Hepatic Monocytes/Macrophages Is Reduced in ASB14780-Treated Mice.Hepatic monocyte/macrophage recruitment was increased inmice injected with CCl4. Then effects of ASB14780 on CCl4-induced infiltration of monocytes/macrophages were exam-ined in the mouse livers (Fig. 3). The immunohistochemicalanalyses of the macrophage surface marker F4/80 demon-strated that CCl4-induced recruitment of hepatic macro-phages around the central veins was reduced in miceadministered ASB14780 (Fig. 3A, lower panels), whereas F4/80-positive cells detected in mice in the absence of CCl4 werestill observed in mice treated with ASB14780 alone (Fig. 3A).A quantitative RT-PCR analysis also showed significant dose-dependent reductions in the expression of CD11b, a monocyte/macrophage marker, in the livers of mice treated withASB14780 (Fig. 3B). These results suggest that CCl4-inducedrecruitment ofmacrophages in the liver was diminished by thetreatment of ASB14780. In addition, the CCl4-induced mRNA

Fig. 2. Inhibitory effects of ASB14780 on CCl4-induced collagen deposition and activation ofhepatic stellate cells. (A) Liver sections werestained with picrosirius red; the scale barcorresponds to 200 mm. (B) a-SMA–expressingcells were detected using immunohistochemistry;the scale bar corresponds to 200 mm. (C) Thecollagen fibers stained with picrosirius red in (A)were quantified using the National Institutes ofHealth ImageJ software program, and the ex-pression of the a-SMA (top blot) and a-tubulin(bottom blot) proteins (D) in liver extracts wasassessed using Western blot analyses. The bandintensities were quantified as arbitrary units (A.U.) using the National Institutes of HealthImageJ software program. (E) The expressionlevels of Col1a2, a-SMA, and transforminggrowth factor-b1 (TGF-b1) mRNA were assessedusing real-time RT-PCR, and were normalized tothat of 18S ribosomal RNA. The data arepresented as the means 6 S.E. (A.U., n = 6–7/group), and differences were identified using anANOVA with a Tukey-Kramer post-hoc analysis(*P , 0.05; **P , 0.01). ASB, ASB14780; IHC,immunohistochemistry; Veh, vehicle.

608 Kanai et al.

at ASPE

T Journals on M

arch 24, 2020jpet.aspetjournals.org

Dow

nloaded from

expression of monocyte chemotactic protein-1 (MCP-1), whichstimulates the migration of monocytes, was significantlyreduced by treatment with ASB14780 (Fig. 3B).Because T lymphocytes also participate in the development

of hepatic fibrosis (Safadi et al., 2004), we assessed the liverexpression of regulated on activation normal T cell expressedand secreted (RANTES) mRNA, which is a potent chemo-attractant and activator of T lymphocytes. The RANTESmRNA levels were significantly increased in CCl4-treatedmice. However, concomitant treatment with ASB14780 had nosignificant effect on the expression of RANTES mRNA (Fig.3B). To assess the effects of ASB14780 on the expression levelof IVA-PLA2 in the liver, real-time RT-PCR and Western blotanalyses were performed. As shown in Fig. 4, the expressionlevels of the Pla2g4a mRNA and IVA-PLA2 protein weresignificantly increased in mice treated with CCl4, whereasthese mRNA and protein levels were significantly attenuatedby treatment with ASB14780. There was little effect ofASB14780 alone on the expression levels. These observationssuggest that ASB14780 inhibited the CCl4-induced inflam-mation in the liver. Thus, ASB14780 may reduce the progres-sion of hepatic fibrosis by blocking the recruitment of hepaticmonocytes/macrophages into the liver parenchyma.ASB14780 Prevents Further Progression of Estab-

lished Hepatic Fibrosis. Patients with hepatic fibrosistypically start pharmacotherapy at later stages of the disease.Therefore, the effects of ASB14780 onmatured hepatic fibrosiswere examined in the following four treatment groups: mice

treated with CCl4 for 6 weeks (group A) or 9 weeks (group B),mice treated with CCl4 and ASB14780 for 9 weeks (group C),and mice treated with CCl4 for 6 weeks followed by treatmentwith both CCl4 and ASB14780 for 3 weeks (group D; Fig. 5A).The serumAST and ALT levels (Fig. 5B) and the areas stainedwith picrosirius red on the liver sections (Fig. 5, C andD) in themice in groups A and B were equally high, suggesting that 6weeks of exposure to CCl4 is sufficient to lead the maturedhepatic fibrosis in mice. Notably, the levels of serum AST andALT and the areas stained with picrosirius red in the mice ingroups C and Dwere significantly reduced in comparison withthose in group B (Fig. 5, B–D), suggesting that orallyadministered ASB14780 is effective even in the presence ofestablished fibrosis.Prevention of HFCD-Induced Fatty Liver following

Treatment with ASB14780. We have previously shownreduced development of fatty liver in HFCD-fed mice withIVA-PLA2 deficiency (Ii et al., 2009). To elucidate the effects ofASB14780 on the development of fatty liver induced by anHFCD, mice were divided into the following four groups: ND(group 1), ND with ASB14780 (group 2), HFCD (group 3), orHFCD with ASB1480 (group 4) for 16 weeks. Significantincreases in body weight and liver weight were observed in themice in group 3 relative to those in group 1 (Fig. 6; Table 2; P,0.00001, two-way ANOVA), indicating that an HFCD for 16weeks led to obesity and hepatic enlargement in mice.ASB14780 did not affect the body and liver weights of ND-fed mice (group 1 vs. group 2; Fig. 6; Table 2; not significant,two-way ANOVA). However, ASB14780 almost entirely pre-vented the obesity and hepatic enlargement induced byHFCD(group 3 vs. group 4; Fig. 6; Table 2; P , 0.00001, two-wayANOVA). Furthermore, we examined the effects of ASB14780on fatty liver, which had already started developing. Micewere fed an HFCD for 16 weeks and were administeredASB14780 during the last 6 weeks (group 5). The body weightsof the mice in group 5 did not increase to the level of those ingroup 3, but rather decreased to the level of those in group 1(Fig. 6B) despite the fact that there was no change in the food

Fig. 3. Reduced CCl4-induced infiltration of macrophages and Kupffercells into the liver after administration of ASB14780. (A) The CCl4-induced infiltration of macrophages and Kupffer cells was assessed byimmunohistochemical staining with an anti-F4/80 antibody; the scale barcorresponds to 40 mm. (B) The expression levels of MCP-1, CD11b, andRANTES mRNA were determined using real-time RT-PCR and werenormalized to that of 18S ribosomal RNA. The data are expressed as themeans6 S.E. [arbitrary units (A.U.), n = 6–7/group], and differences wereidentified using an ANOVA with a Tukey-Kramer post-hoc analysis (*P,0.05; **P , 0.01). ASB, ASB14780; Veh, vehicle.

Fig. 4. Reduction of CCl4-induced IVA-PLA2 levels after the administra-tion of ASB14780. (A) The expression levels of Pla2g4a mRNA wereassessed using real-time RT-PCR, and were normalized to that of 18Sribosomal RNA. (B) The levels of IVA-PLA2 (top blot) and a-tubulin(bottom blot) proteins in the liver extracts were assessed using a Westernblot analysis. The band intensities were quantified using the NationalInstitutes of Health ImageJ software program. The data are expressed asthe means 6 S.E. [arbitrary units (A.U.), n = 6–7/group], and differenceswere identified using an ANOVA with a Tukey-Kramer post-hoc analysis(**P , 0.01). ASB, ASB14780; Veh, vehicle.

Suppression of NAFLD and NASH Using an IVA-PLA2 Inhibitor 609

at ASPE

T Journals on M

arch 24, 2020jpet.aspetjournals.org

Dow

nloaded from

consumption between the groups of mice fed an HFCD(Table 2).Histologic H&E staining of the liver also showed dramati-

cally reduced microvesicular steatosis in the mice in groups 4and 5 in comparison with those in group 3 (Fig. 7A). Moreover,the HFCD-associated increases in the liver triglyceride levelswere reduced after treatment with ASB14780 (group 3 vs.groups 4 and 5; Fig. 7B). In contrast, although higher levels ofserum triglycerides were detected in HFCD-fed mice in group3 compared with the ND-fed mice in group 1, these levels werenot reduced after concomitant or delayed treatment withASB14780 (groups 4 and 5; Fig. 7B), indicating thatASB14780 affects the triglyceride level increased by HFCDin the liver, but not in the serum. The food consumption wasalmost the same between the HFCD-fed mouse groups

(Table 2). Taken together, these data indicate that ASB14780reduces the HFCD-induced triglyceride levels in the liverwithout affecting the levels in the serum or the food consump-tion of the mice, and that ASB14780 can prevent the progres-sion of fatty liver induced by HFCD, even after fatty liverstarts developing.ASB14780 Inhibits the Expression of Lipogenesis-

Related mRNAs. To identify the molecular mechanisms bywhich ASB14780 prevents the development of fatty liver, themRNA levels of genes related to lipid metabolism werequantified using real-time RT-PCR in HFCD-fed mice withand without ASB14780 treatment (Fig. 8). Although themRNA expression of CD36, which transports free fatty acidswithin cells, was markedly increased in the HFCD-fed mice(group 3) compared with ND-fed mice (group 1) (Fig. 8A), its

Fig. 5. Reduction of CCl4-induced hepatic fibrosisby the administration of ASB14780 after hepaticfibrosis had already developed. Hepatic fibrosiswas induced in 6-week-old mice following twice-weekly intraperitoneal injections of CCl4 in cornoil (0.31 ml/kg). The mice were sacrificed 48 hoursafter the last injection. (A) Group A was injectedwith CCl4 for 6 weeks, and groups B, C, and Dwere injected with CCl4 for 9 weeks. In addition,group C received 0.3 g/kg ASB14780 by oralgavage every day, administered 1 hour beforeCCl4 injection. After 6 weeks of CCl4 treatment,group D received 0.3 g/kg ASB14780 for another 3weeks. (B) The serum AST and ALT levels weredetermined using enzymatic assays. (C) Liversections were stained with picrosirius red; thescale bar corresponds to 200 mm. (D) The numberof collagen fibers was quantified by measuring thered areas using the ImageJ software program. Thedata are presented as the means 6 S.E. (n = 6–7/group), and differences were identified using anANOVA with a Tukey-Kramer post-hoc analysis(**P , 0.01; ##P , 0.01).

610 Kanai et al.

at ASPE

T Journals on M

arch 24, 2020jpet.aspetjournals.org

Dow

nloaded from

expression in the livers ofHFCD-fedmice treatedwith 0.3 g/kgASB14780 (group 4) was similar to that in the mice in group 3.However, the mRNA expression of liver fatty acid bindingprotein, which transports free fatty acids to mitochondria(Kim and Storch, 1992), and carnitine palmitoyl-transferase1a, which performs the rate-limiting step in fatty acidoxidation (McGarry and Brown, 1997), did not differ betweenthe mice in groups 1 and 3 (Fig. 8A).The expression levels of genes encoding proteins involved in

fatty acid synthesis and monounsaturated fatty acid synthe-sis, including acetyl-CoA carboxylase (ACC), fatty acid syn-thase (FAS), and stearoyl-CoA desaturase 1 (SCD-1), weresignificantly increased in the livers of the mice in group 3compared with those in group 1 (Fig. 8B). Moreover, themRNA expression of sterol regulatory element-binding pro-tein 1c (SREBP1c), which is a transcription factor leadingto the expression of ACC, FAS, and SCD-1 mRNA, was

consistently increased in the livers of the mice in group 3compared with those in group 1 (Fig. 8B). However, the livermRNA expression of several lipogenesis-related genes(SREBP1c, ACC, FAS, and SCD-1) was significantly de-creased in the mice in group 4 compared with the mice ingroup 3, indicating that ASB14780 has significant antilipo-genic activity. We further demonstrated that the mRNAexpressions of SREBP1c, FAS, SCD-1, and ACC increased inmice on an HFCD. The mice on an HFCD with delayedtreatment with ASB14780 after 10 weeks (group 5) haddecreases in body weight and liver triglycerides, but theexpression level of SREBP1c mRNA was comparable withthat in the mice on an HFCD (group 3). In contrast, mice ingroup 5 appear to have decreased FAS and SCD-1 expression.The contradictory expressions of SREBP1c, FAS, SCD-1, andACC mRNAs in the mice of group 5 might be regulated byunknown factor(s) that suppresses the transcriptions oflipogenic genes individually.In further experiments, the mRNA expression of DGAT2,

which esterifies diacylglycerol for de novo lipogenesis, wasalso reduced after treatment of HFCD-fed mice withASB14780 (group 4; Fig. 8B). However, the levels of DGAT1and microsomal triglyceride protein, which mediate very-low-density lipoprotein secretion, did not differ significantlyamong the three groups (groups 1, 3, and 4; Fig. 8C).Finally, the mRNA expression of hormone-sensitive lipase,

which catalyzes the hydrolysis of triglycerides, was signifi-cantly decreased in the livers from the mice in group 4compared with those from group 3 (Fig. 8C), indicating thatlipogenesis decreased following treatment with ASB14780

Fig. 6. ASB14780 suppressed HFCD-induced weight gain in mice. Micewere fed an HFCD for 16 weeks to induce fatty liver. The mice in thetreatment groups were administrated the IVA-PLA2 inhibitor ASB14780(ASB) at 0.1 or 0.3 g/kg by oral gavage every day. After a 6-hour fast, themice were sacrificed, and parameters related to fatty liver were assessed.(A) The schedule of the treatment with ASB14780 and the HFCD. Themice were divided into five groups according to the treatment. Broken andsolid lines with arrowheads indicate the presence and absence ofASB14780, respectively. Blue and red arrows indicate that the mice werefed an ND and the HFCD, respectively. The body weights were measuredevery week for 16 weeks for mice fed the ND or HFCD (B), and the bodyweights were measured at weeks 10 and 16 (C). The data are presented asthe means 6 S.E. (n = 4–8/group), and differences were identified using atwo-way ANOVA (*P , 0.05; **P , 0.01).

Fig. 7. Suppression of HFCD-induced fatty liver by ASB14780. (A)Paraffin-embedded liver sections were stained with hematoxylin andeosin; the scale bar corresponds to 20 mm. (B) The liver and serumtriglyceride levels were determined using enzymatic assays. The data arepresented as the means 6 S.E. (n = 4–8/group), and differences wereidentified using an ANOVA with a Tukey-Kramer post-hoc analysis (*P,0.05; **P , 0.01). CV, central vein.

Suppression of NAFLD and NASH Using an IVA-PLA2 Inhibitor 611

at ASPE

T Journals on M

arch 24, 2020jpet.aspetjournals.org

Dow

nloaded from

(Fig. 8B). In the mice fed the ND (groups 1 and 2), there waslittle effect of ASB14780 alone on the mRNA expression levelsof genes involved in lipid metabolism. Taken together, thesefindings suggest that ASB14780 suppresses HFCD-inducedde novo lipogenesis in the liver.

DiscussionFollowing our observations of suppressed HFCD-induced

fatty liver and CCl4-induced hepatic fibrosis in IVA-PLA2–deficient mice (Ii et al., 2009; Ishihara et al., 2012), inthe present study, we examined the effects of an orally activeIVA-PLA2 inhibitor, ASB14780, on the development of fattyliver and hepatic fibrosis. Daily administration of ASB14780reduced CCl4-induced hepatic fibrosis and HFCD-inducedfatty liver, and subsequent quantitative RT-PCR analysessuggested that lipogenesis was significantly reduced in thesemice. Notably, we also showed that ASB14780 was effectiveagainst established hepatic fibrosis and fatty liver. Thesefindings suggest that ASB14780 may represent a candidatefor pharmaceutical use in the treatment of NAFLD andNASH, and indicate that IVA-PLA2 is a valid therapeutictarget for the treatment of fatty liver and hepatic fibrosis.For CCl4-induced events, we have previously shown that

collagen deposit, recruitment of monocyte/macrophage intothe liver, and induction ofMCP-1mRNAwere reduced in IVA-PLA2–deficient mice (Ishihara et al., 2012). In the presentstudy, ASB14780 also reduced the development of hepaticfibrosis with suppression of recruitment of monocyte/macrophage into the liver and induction of MCP-1 mRNA. Ina similar fashion, reduced development of fatty liver wasalso observed in mice with ASB14780 similar to IVA-PLA2–deficient mice. Thus, ASB14780 is thought to exertinhibitory effects on the development of hepatic fibrosis andfatty liver through its inhibitory activity for IVA-PLA2. Incontrast, ASB14780 almost completely prevented body weightgain in mice fed the HFCD. However, the body weights of IVA-PLA2–deficient HFCD-fed mice were equivalent to those ofwild-type mice on an HFCD (Ii et al., 2009), suggesting thatthe pharmacological effects of ASB14780 on HFCD-inducedobesity may be independent of IVA-PLA2 activity.We have shown that CCl4-induced liver injury was reduced

by deficiency or inhibition of IVA-PLA2 inmice (Ishihara et al.,2012 and the current study). In contrast, Fas-induced liverinjury was reportedly reduced in transgenic mice overexpress-ing IVA-PLA2 in hepatocytes, suggesting that IVA-PLA2 pro-tects against Fas-induced hepatocyte apoptosis (Li et al., 2011).In contrast, IVA-PLA2 overexpression sensitized hepatocytes

to lipopolysaccharide/D-galactosamine–induced hepatic toxic-ity, resulting in marked increases in the serum ALT and ASTlevels in hepatic IVA-PLA2–overexpressing mice (Li et al.,2011). Thus, the IVA-PLA2–mediated hepatic responses may

Fig. 8. Suppression of hepatic lipogenesis-related mRNA expression byASB14780. The expression levels of CD36, liver fatty acid binding protein(L-FABP), and carnitine palmitoyl-transferase 1a (CPT1a ) (A); SREBP1c,ACC, FAS, SCD-1, and DGAT2 (B); and DGAT1, microsomal triglycerideprotein (MTP), and hormone-sensitive lipase (HSL) mRNA (C) wereassessed using real-time RT-PCR and were normalized to that of 36B4.The data are presented as the means 6 S.E. [arbitrary units (A.U.), n =4–8/group], and differences were identified using an ANOVA with aTukey-Kramer post-hoc analysis (*P , 0.05; **P , 0.01; #P , 0.05).

TABLE 2The effects of ASB14780 and HFCD on the tissue weights, serum biomarker levels, and food intake of control and treated animalsThe data are presented as the means 6 S.E. for four to seven individuals (Tukey-Kramer post-hoc analysis).

ASB(0)/ND (Group 1) ASB(0.3)/ND (Group 2) ASB(0)/HFCD (Group 3) ASB(0.3)/HFCD (Group 4) ASB(0→0.3)/HFCD (Group 5)

Body weight (g)a 27.5 6 1.1 28.6 6 1.3 37.3 6 0.8** 29.9 6 1.1## 30.8 6 1.0##

Liver weight (g) 1.12 6 0.06 1.28 6 0.10 1.44 6 0.06* 1.27 6 0.04 1.35 6 0.06Fat pad weight (g) 0.72 6 0.11 0.80 6 0.10 2.25 6 0.15** 1.07 6 0.06## 1.31 6 0.15##

Serum AST (IU/l) 21.54 6 0.91 22.67 6 0.91 30.42 6 2.57 31.46 6 2.09 26.71 6 1.84Serum ALT (IU/l) 6.29 6 0.52 5.57 6 0.15 9.23 6 1.43 7.16 6 0.78 8.21 6 1.02Food intake (g) 2.79 6 0.07 3.17 6 0.12* 2.68 6 0.08 2.5 6 0.07 2.40 6 0.06

ASB, ASB14780.aBody weights were measured after a 6-hour fast.*P , 0.05 versus ASB(0)/ND; **P , 0.01 versus ASB(0)/ND; #P , 0.05 versus ASB(0)/HFCD; ##P , 0.01 versus ASB(0)/HFCD.

612 Kanai et al.

at ASPE

T Journals on M

arch 24, 2020jpet.aspetjournals.org

Dow

nloaded from

vary depending on the context of liver injury. These findings inmodels with IVA-PLA2 overexpression are compatible with thepresent data demonstrating that inhibition of IVA-PLA2 pro-tects against the hepatic injury induced by chronic CCl4exposure, because lipopolysaccharide/D-galactosamine–and CCl4-induced hepatotoxicity are reportedly mediated byoxidative stress (Osakabe et al., 2002).In further experiments, ASB14780 prevented the progres-

sion of lipid accumulation and fibrosis in the liver, even after10 weeks of exposure to the HFCD to induce fatty liverdevelopment or 6 weeks of exposure to CCl4 to induce hepaticfibrosis. This is very important for the development ofpharmacotherapy for NASH and NAFLD, because mostpatients with these conditions begin to receive treatment atthe advanced stage of the diseases. A previous meta-analysisindicated that thiazolidinediones, especially pioglitazone,improved steatosis and inflammation, but were associatedwith weight increases in approximately 70% of patients andwith edema in less than 10% of patients (Musso et al., 2010). Incontrast, Cochrane reviews suggested that there is insufficientevidence to support the use of bile acids, antioxidant supple-ments, metformin, or thiazolidinediones in the treatment ofNAFLD patients without diabetes (Angelico et al., 2007;Lirussi et al., 2007; Orlando et al., 2007). Thus, pharmaco-therapy for NAFLD has not yet been established. The presentinhibitor suppressed the accumulation of lipids and collagenin the liver, even after fatty liver and hepatic fibrosis hadalready developed, indicating that ASB14780 may be apromising agent for the treatment of NAFLD.In the present study, the levels of lipogenic genes, such as

FAS, SCD-1, and ACC, were not always closely correlated tothe level of SREBP1c mRNA. In the group of mice adminis-tered ASB14780 during the last 6 weeks, the expression levelof SREBP1c was high similar to mice on HFCD, althoughexpressions of FAS and SCD-1 mRNAs were basal level.Although it has been suggested that the transcriptions ofFAS, SCD-1, and ACCmRNAs are regulated by the activity ofSREBP1c, contradictory results on expressions of SREBP1cand its downstream genes, such as FAS, ACC, and SCD-1,have also been shown in the model of hepatic steatosis bytamoxifen (Lee et al., 2010). Lelliott et al. (2005) reported thatthe level of SREBP1mRNA in the rat liver shows no change inresponse to tamoxifen treatment, whereas the levels of FAS,SCD-1, and ACC mRNAs in tamoxifen-treated rat livers arelower than those in the control group. Cole et al. (2010) haveshown that there are no differences in the levels of mouse liverSREBP1c and ACC mRNAs between control mice andtamoxifen-treated mice, but the levels of FAS and SCD-1mRNAs in tamoxifen-treated mice are significantly increasedand decreased, respectively. It has also been suggested thattamoxifen does not affect the activities and levels of SCD-1mRNA in the rat liver, but tamoxifen can decrease theactivities of rat liver ACC and FAS (Gudbrandsen et al., 2006).The removal of causative factors is the most effective

treatment of hepatic fibrosis, because liver fibrosis is re-versible (Bataller and Brenner, 2005). Accordingly, hepaticfibrosis was reversed in a patient with chronic hepatitis Bvirus infection after successful treatment with the antiviralagent lamivudine (Kweon et al., 2001). Since it is accepted thatoxidative stress is central to the pathogenesis of NASH(Angulo, 2002; Marchesini and Forlani, 2002), the antioxi-dants vitamin E, silymarin, and phosphatidylcholine all

inhibit the activation of hepatic stellate cells, protect hepato-cytes from apoptosis, and attenuate experimental hepaticfibrosis (Tome and Lucey, 2004). Because inflammationprecedes and promotes the progression of hepatic fibrosis,anti-inflammatory drug therapies have been considered, andcorticosteroids are widely used in the treatment of hepaticfibrosis in patients with autoimmune hepatitis and acutealcoholic hepatitis (Czaja and Carpenter, 2004). In addition,pentoxifylline, which acts as an anti-inflammatory agent byinhibiting tumor necrosis factor a production (Tilg, 2010), hasbeen considered for the treatment of NASH (Zein et al., 2011).Similar to corticosteroids and pentoxifylline, ASB14780 pre-vented the development of NAFLD symptoms, such as fattyliver and hepatic fibrosis, probably through its activity as aninhibitor of IVA-PLA2 (Tomoo et al., 2014).Nonetheless, the present data suggest that ASB14780, as a

selective inhibitor of IVA-PLA2, may ameliorate NAFLDand NASH as well as HFCD-associated obesity. Moreover,ASB14780 prevented hepatic lipid accumulation when admin-istered before and after the development of fatty liver andhepatic fibrosis. Hence, although further studies are neces-sary to determine the clinical efficacy of ASB14780, this orallyactive agent has potential as a pharmaceutical treatment ofNAFLD and NASH.

Acknowledgments

The authors thank Ami Takeuchi and Yuki Ito, Kyoto Pharmaceu-tical University, for their technical help with this study.

Authorship Contributions

Participated in research design: Ishihara, Nagahira, Hayashi,Akiba.

Conducted experiments: Kanai, Tomoo, Nagahira.Performed data analysis: Ishihara, Akiba.Wrote or contributed to the writing of the manuscript: Kanai,

Ishihara, Kawashita, Akiba.

References

Aithal GP, Thomas JA, Kaye PV, Lawson A, Ryder SD, Spendlove I, Austin AS,Freeman JG, Morgan L, and Webber J (2008) Randomized, placebo-controlled trialof pioglitazone in nondiabetic subjects with nonalcoholic steatohepatitis. Gastro-enterology 135:1176–1184.

Angelico F, Burattin M, Alessandri C, Del Ben M, and Lirussi F (2007) Drugs im-proving insulin resistance for non-alcoholic fatty liver disease and/or non-alcoholicsteatohepatitis. Cochrane Database Syst Rev (1):CD005166.

Angulo P (2002) Nonalcoholic fatty liver disease. N Engl J Med 346:1221–1231.Avasarala S, Yang L, Sun Y, Leung AW, Chan WY, Cheung WT, and Lee SS (2006) Atemporal study on the histopathological, biochemical and molecular responses ofCCl(4)-induced hepatotoxicity in Cyp2e1-null mice. Toxicology 228:310–322.

Bataller R and Brenner DA (2005) Liver fibrosis. J Clin Invest 115:209–218.Belfort R, Harrison SA, Brown K, Darland C, Finch J, Hardies J, Balas B, GastaldelliA, Tio F, and Pulcini J, et al. (2006) A placebo-controlled trial of pioglitazone insubjects with nonalcoholic steatohepatitis. N Engl J Med 355:2297–2307.

Chang TK, Crespi CL, and Waxman DJ (2006) Spectrophotometric analysis of humanCYP2E1-catalyzed p-nitrophenol hydroxylation. Methods Mol Biol 320:127–131.

Clark JM (2006) The epidemiology of nonalcoholic fatty liver disease in adults. J ClinGastroenterol 40 (Suppl 1):S5–S10.

Cole LK, Jacobs RL, and Vance DE (2010) Tamoxifen induces triacylglycerol accu-mulation in the mouse liver by activation of fatty acid synthesis. Hepatology 52:1258–1265.

Connolly S, Bennion C, Botterell S, Croshaw PJ, Hallam C, Hardy K, Hartopp P,Jackson CG, King SJ, and Lawrence L, et al. (2002) Design and synthesis of a noveland potent series of inhibitors of cytosolic phospholipase A(2) based on a 1,3-disubstituted propan-2-one skeleton. J Med Chem 45:1348–1362.

Czaja AJ and Carpenter HA (2004) Decreased fibrosis during corticosteroid therapyof autoimmune hepatitis. J Hepatol 40:646–652.

Flamand N, Picard S, Lemieux L, Pouliot M, Bourgoin SG, and Borgeat P (2006)Effects of pyrrophenone, an inhibitor of group IVA phospholipase A2, on eicosanoidand PAF biosynthesis in human neutrophils. Br J Pharmacol 149:385–392.

Folch J, Lees M, and Sloane Stanley GH (1957) A simple method for the isolation andpurification of total lipides from animal tissues. J Biol Chem 226:497–509.

Fritsche A, Elfringhoff AS, Fabian J, and Lehr M (2008) 1-(2-Carboxyindol-5-yloxy)propan-2-ones as inhibitors of human cytosolic phospholipase A2a: synthesis,biological activity, metabolic stability, and solubility. Bioorg Med Chem 16:3489–3500.

Suppression of NAFLD and NASH Using an IVA-PLA2 Inhibitor 613

at ASPE

T Journals on M

arch 24, 2020jpet.aspetjournals.org

Dow

nloaded from

Gudbrandsen OA, Rost TH, and Berge RK (2006) Causes and prevention oftamoxifen-induced accumulation of triacylglycerol in rat liver. J Lipid Res 47:2223–2232.

Hess M, Schulze Elfringhoff A, and Lehr M (2007) 1-(5-Carboxy- and 5-carbamoylindol-1-yl)propan-2-ones as inhibitors of human cytosolic phospholipaseA2a: bioisosteric replacement of the carboxylic acid and carboxamide moiety. Bio-org Med Chem 15:2883–2891.

Ii H, Yokoyama N, Yoshida S, Tsutsumi K, Hatakeyama S, Sato T, Ishihara K,and Akiba S (2009) Alleviation of high-fat diet-induced fatty liver damage in groupIVA phospholipase A2-knockout mice. PLoS One 4:e8089.

Ishihara K, Miyazaki A, Nabe T, Fushimi H, Iriyama N, Kanai S, Sato T, Uozumi N,Shimizu T, and Akiba S (2012) Group IVA phospholipase A2 participates in theprogression of hepatic fibrosis. FASEB J 26:4111–4121.

Kim HK and Storch J (1992) Free fatty acid transfer from rat liver fatty acid-bindingprotein to phospholipid vesicles. Effect of ligand and solution properties. J BiolChem 267:77–82.

Kita Y, Ohto T, Uozumi N, and Shimizu T (2006) Biochemical properties and path-ophysiological roles of cytosolic phospholipase A2s. Biochim Biophys Acta 1761:1317–1322.

Kleiner DE, Brunt EM, Van Natta M, Behling C, Contos MJ, Cummings OW, FerrellLD, Liu YC, Torbenson MS, and Unalp-Arida A, et al.; Nonalcoholic Steatohepa-titis Clinical Research Network (2005) Design and validation of a histologicalscoring system for nonalcoholic fatty liver disease. Hepatology 41:1313–1321.

Kokotos G, Kotsovolou S, Six DA, Constantinou-Kokotou V, Beltzner CC, and DennisEA (2002) Novel 2-oxoamide inhibitors of human group IVA phospholipase A(2). JMed Chem 45:2891–2893.

Kokotos G, Six DA, Loukas V, Smith T, Constantinou-Kokotou V, Hadjipavlou-LitinaD, Kotsovolou S, Chiou A, Beltzner CC, and Dennis EA (2004) Inhibition of groupIVA cytosolic phospholipase A2 by novel 2-oxoamides in vitro, in cells, and in vivo. JMed Chem 47:3615–3628.

Kweon YO, Goodman ZD, Dienstag JL, Schiff ER, Brown NA, Burchardt E,Schoonhoven R, Brenner DA, and Fried MW (2001) Decreasing fibrogenesis: animmunohistochemical study of paired liver biopsies following lamivudine therapyfor chronic hepatitis B. J Hepatol 35:749–755.

Lago RM, Singh PP, and Nesto RW (2007) Congestive heart failure and cardiovas-cular death in patients with prediabetes and type 2 diabetes given thiazolidine-diones: a meta-analysis of randomised clinical trials. Lancet 370:1129–1136.

Lecka-Czernik B (2010) Bone loss in diabetes: use of antidiabetic thiazolidinedionesand secondary osteoporosis. Curr Osteoporos Rep 8:178–184.

Lee KL, Foley MA, Chen L, Behnke ML, Lovering FE, Kirincich SJ, Wang W, Shim J,Tam S, and Shen MW, et al. (2007) Discovery of Ecopladib, an indole inhibitor ofcytosolic phospholipase A2a. J Med Chem 50:1380–1400.

Lee MH, Kim JW, Kim JH, Kang KS, Kong G, and Lee MO (2010) Gene expressionprofiling of murine hepatic steatosis induced by tamoxifen. Toxicol Lett 199:416–424.

Lelliott CJ, López M, Curtis RK, Parker N, Laudes M, Yeo G, Jimenez-Liñan M,Grosse J, Saha AK, and Wiggins D, et al. (2005) Transcript and metabolite analysisof the effects of tamoxifen in rat liver reveals inhibition of fatty acid synthesis inthe presence of hepatic steatosis. FASEB J 19:1108–1119.

Li G, Chen W, Han C, and Wu T (2011) Cytosolic phospholipase A(2)a protects againstFas- but not LPS-induced liver injury. J Hepatol 55:1281–1290.

Lirussi F, Azzalini L, Orando S, Orlando R, and Angelico F (2007) Antioxidantsupplements for non-alcoholic fatty liver disease and/or steatohepatitis. CochraneDatabase Syst Rev (1):CD004996.

Ludwig J, Bovens S, Brauch C, Elfringhoff AS, and Lehr M (2006) Design and syn-thesis of 1-indol-1-yl-propan-2-ones as inhibitors of human cytosolic phospholipaseA2a. J Med Chem 49:2611–2620.

Marchesini G and Forlani G (2002) NASH: from liver diseases to metabolic disordersand back to clinical hepatology. Hepatology 35:497–499.

Marchesini G, Moscatiello S, Di Domizio S, and Forlani G (2008) Obesity-associatedliver disease. J Clin Endocrinol Metab 93(11, Suppl 1)S74–S80.

McGarry JD and Brown NF (1997) The mitochondrial carnitine palmitoyltransferasesystem. From concept to molecular analysis. Eur J Biochem 244:1–14.

McKew JC, Foley MA, Thakker P, Behnke ML, Lovering FE, Sum FW, Tam S, Wu K,Shen MW, and Zhang W, et al. (2006) Inhibition of cytosolic phospholipase A2a: hitto lead optimization. J Med Chem 49:135–158.

McKew JC, Lee KL, Shen MW, Thakker P, Foley MA, Behnke ML, Hu B, Sum FW,Tam S, and Hu Y, et al. (2008) Indole cytosolic phospholipase A2 a inhibitors:discovery and in vitro and in vivo characterization of 4-3-[5-chloro-2-(2-[(3,4-dichlorobenzyl)sulfonyl]aminoethyl)-1-(diphenylmethyl)-1H-indol-3-yl]propylbenzoicacid, efipladib. J Med Chem 51:3388–3413.

McKew JC, Lovering F, Clark JD, Bemis J, Xiang Y, Shen M, Zhang W, Alvarez JC,and Joseph-McCarthy D (2003) Structure-activity relationships of indole cytosolic

phospholipase A(2)a inhibitors: substrate mimetics. Bioorg Med Chem Lett 13:4501–4504.

Musso G, Gambino R, Cassader M, and Pagano G (2010) A meta-analysis of ran-domized trials for the treatment of nonalcoholic fatty liver disease. Hepatology 52:79–104.

Noguchi T, Fong KL, Lai EK, Alexander SS, King MM, Olson L, Poyer JL, and McCayPB (1982) Specificity of a phenobarbital-induced cytochrome P-450 for metabolismof carbon tetrachloride to the trichloromethyl radical. Biochem Pharmacol 31:615–624.

Ono T, Yamada K, Chikazawa Y, Ueno M, Nakamoto S, Okuno T, and Seno K (2002)Characterization of a novel inhibitor of cytosolic phospholipase A2a, pyrrophenone.Biochem J 363:727–735.

Orlando R, Azzalini L, Orando S, and Lirussi F (2007) Bile acids for non-alcoholicfatty liver disease and/or steatohepatitis. Cochrane Database Syst Rev (1):CD005160.

Osakabe N, Yasuda A, Natsume M, Sanbongi C, Kato Y, Osawa T, and Yoshikawa T(2002) Rosmarinic acid, a major polyphenolic component of Perilla frutescens, re-duces lipopolysaccharide (LPS)-induced liver injury in D-galactosamine (D-GalN)-sensitized mice. Free Radic Biol Med 33:798–806.

Piccinni C, Motola D, Marchesini G, and Poluzzi E (2011) Assessing the association ofpioglitazone use and bladder cancer through drug adverse event reporting. Di-abetes Care 34:1369–1371.

Safadi R, Ohta M, Alvarez CE, Fiel MI, Bansal M, Mehal WZ, and Friedman SL(2004) Immune stimulation of hepatic fibrogenesis by CD8 cells and attenuation bytransgenic interleukin-10 from hepatocytes. Gastroenterology 127:870–882.

Sanyal AJ, Chalasani N, Kowdley KV, McCullough A, Diehl AM, Bass NM,Neuschwander-Tetri BA, Lavine JE, Tonascia J, and Unalp A, et al.; NASHCRN (2010) Pioglitazone, vitamin E, or placebo for nonalcoholic steatohepatitis.N Engl J Med 362:1675–1685.

Seno K, Okuno T, Nishi K, Murakami Y, Watanabe F, Matsuura T, Wada M, Fujii Y,Yamada M, and Ogawa T, et al. (2000) Pyrrolidine inhibitors of human cytosolicphospholipase A(2). J Med Chem 43:1041–1044.

Seno K, Okuno T, Nishi K, Murakami Y, Yamada K, Nakamoto S, and Ono T (2001)Pyrrolidine inhibitors of human cytosolic phospholipase A2. Part 2: synthesis ofpotent and crystallized 4-triphenylmethylthio derivative ‘pyrrophenone’. BioorgMed Chem Lett 11:587–590.

Six DA, Barbayianni E, Loukas V, Constantinou-Kokotou V, Hadjipavlou-Litina D,Stephens D, Wong AC, Magrioti V, Moutevelis-Minakakis P, and Baker SF, et al.(2007) Structure-activity relationship of 2-oxoamide inhibition of group IVA cyto-solic phospholipase A2 and group V secreted phospholipase A2. J Med Chem 50:4222–4235.

Stephens D, Barbayianni E, Constantinou-Kokotou V, Peristeraki A, Six DA, CooperJ, Harkewicz R, Deems RA, Dennis EA, and Kokotos G (2006) Differential in-hibition of group IVA and group VIA phospholipases A2 by 2-oxoamides. J MedChem 49:2821–2828.

Takaki A, Kawai D, and Yamamoto K (2014) Molecular mechanisms and newtreatment strategies for non-alcoholic steatohepatitis (NASH). Int J Mol Sci 15:7352–7379.

Tilg H (2010) The role of cytokines in non-alcoholic fatty liver disease. Dig Dis 28:179–185.

Tilg H and Moschen AR (2010) Evolution of inflammation in nonalcoholic fatty liverdisease: the multiple parallel hits hypothesis. Hepatology 52:1836–1846.

Tome S and Lucey MR (2004) Review article: current management of alcoholic liverdisease. Aliment Pharmacol Ther 19:707–714.

Tomoo T, Nakatsuka T, Katayama T, Hayashi Y, Fujieda Y, Terakawa M,and Nagahira K (2014) Design, synthesis, and biological evaluation of 3-(1-Aryl-1H-indol-5-yl)propanoic acids as new indole-based cytosolic phospholipase A2ainhibitors. J Med Chem 57:7244–7262.

Tsukamoto H, Rippe R, Niemelä O, and Lin M (1995) Roles of oxidative stress inactivation of Kupffer and Ito cells in liver fibrogenesis. J Gastroenterol Hepatol 10(Suppl 1):S50–S53.

Zein CO, Yerian LM, Gogate P, Lopez R, Kirwan JP, Feldstein AE, and McCulloughAJ (2011) Pentoxifylline improves nonalcoholic steatohepatitis: a randomizedplacebo-controlled trial. Hepatology 54:1610–1619.

Zerbe O and Gressner AM (1988) Proliferation of fat-storing cells is stimulated bysecretions of Kupffer cells from normal and injured liver. Exp Mol Pathol 49:87–101.

Address correspondence to: Dr. Satoshi Akiba, Department of PathologicalBiochemistry, Kyoto Pharmaceutical University, Misasagi Nakauchi-cho 5,Yamashina-ku, Kyoto 607-8414, Japan. E-mail: akiba@mb.kyoto-phu.ac.jp

614 Kanai et al.

at ASPE

T Journals on M

arch 24, 2020jpet.aspetjournals.org

Dow

nloaded from