Toxicology and Applied Pharmacology 286 (2015) 102–111

Contents lists available at ScienceDirect

Toxicology and Applied Pharmacology

j ourna l homepage: www.e lsev ie r .com/ locate /ytaap

Inhibition of soluble epoxide hydrolase attenuates hepatic fibrosis andendoplasmic reticulum stress induced by carbon tetrachloride in mice

Todd R. Harris a, Ahmed Bettaieb b, Sean Kodani a, Hua Dong a, Richard Myers c,Nipavan Chiamvimonvat c, Fawaz G. Haj b,d, Bruce D. Hammock a,⁎a Department of Entomology and Comprehensive Cancer Center, University of California, Davis, CA 95616, USAb Department of Nutrition, University of California, Davis, CA 95616, USAc Department of Internal Medicine: Cardiovascular, University of California, Davis, CA 95616, USAd Department of Internal Medicine: Endocrinology, Diabetes and Metabolism, University of California, Davis, CA 95616, USA

⁎ Corresponding author at: Department of EntomoloCenter, University of California, Davis, 1 Shields Avenue, D

E-mail address: bdhammock@ucdavis.edu (B.D. Hamm

http://dx.doi.org/10.1016/j.taap.2015.03.0220041-008X/© 2015 Elsevier Inc. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 3 January 2015Revised 18 March 2015Accepted 20 March 2015Available online 28 March 2015

Keywords:Soluble epoxide hydrolaseEndoplasmic reticulum stressCarbon tetrachlorideFibrosis

Liver fibrosis is a pathological condition in which chronic inflammation and changes to the extracellular matrixlead to alterations in hepatic tissue architecture and functional degradation of the liver. Inhibitors of the enzymesoluble epoxide hydrolase (sEH) reduce fibrosis in the heart, pancreas and kidney in several disease models. Inthis study, we assess the effect of sEH inhibition on the development of fibrosis in a carbon tetrachloride(CCl4)-inducedmousemodel bymonitoring changes in the inflammatory response, matrix remolding and endo-plasmic reticulum stress. The sEH inhibitor 1-trifluoromethoxyphenyl-3-(1-propionylpiperidin-4-yl) urea(TPPU) was administered in drinking water. Collagen deposition in the liver was increased five-fold in theCCl4-treated group, and this was returned to control levels by TPPU treatment. Hepatic expression of Col1a2and 3a1 mRNA was increased over fifteen-fold in the CCl4-treated group relative to the Control group, and thisincrease was reduced by 50% by TPPU treatment. Endoplasmic reticulum (ER) stress observed in the livers ofCCl4-treated animals was attenuated by TPPU treatment. In order to support the hypothesis that TPPU is actingto reduce the hepatic fibrosis and ER stress through its action as a sEH inhibitor we used a second sEH inhibitor,trans-4-{4-[3-(4-trifluoromethoxy-phenyl)-ureido]-cyclohexyloxy}-benzoic acid (t-TUCB), and sEH null mice.Taken together, these data indicate that the sEHmay play an important role in the development of hepatic fibro-sis induced by CCl4, presumably by reducing endogenous fatty acid epoxide chemical mediators acting to reduceER stress.

© 2015 Elsevier Inc. All rights reserved.

Introduction

Liver fibrosis is a pathological condition that results from a normallybeneficial wound healing process (Hernandez-Gea and Friedman,2011). Damage to the liver triggers the inflammatory response, whichinitiates recruitment of macrophages and fibrocytes and remodeling ofthe extracellular matrix (ECM) in order to clear damaged cells andpreserve the architecture of the liver (Bataller and Brenner, 2005). How-ever, repeated injury, such as occurs in chronic alcoholism, exposure toenvironmental toxins, and viral infection can lead to a cascade of inflam-matory system-related changes in the liver that result in abnormaldeposition and composition of collagens and other elements of theECM, as well as extensive damage to the tissue (Iredale, 2007). Thispathological condition is called fibrosis.

Fibrosis is facilitated by a number of signaling molecules thatmodulate the immune response such as the cytokines IL-6 and IL-1β

gy and Comprehensive Canceravis, CA 95616, USA.ock).

and the growth factor TGFβ (Wynn and Ramalingam, 2012). Alongsidethese commonly studied protein inflammatory mediators, bioactivelipids play an important role (Funk, 2001; Stables and Gilroy, 2011).There is growing evidence of the importance of lipid signaling in liverfibrosis. The eicosanoids are a class of lipid mediators that include theleukotrienes, prostaglandins and epoxyeicosatrienoic acids (EETs),produced through the activity of lipoxygenase (LOX), cyclooxygenase(COX) and cytochrome P450 enzymes, respectively (Funk, 2001;Stables and Gilroy, 2011). The leukotrienes and prostaglandins havebeen implicated in the progression of liver fibrosis in studies employingleukotriene receptor antagonists and COX inhibitors. A leukotrienereceptor antagonist, montelukast, attenuates liver damage and fibrosisin a bile duct ligation (BDL) and resection rat model (El-Swefy andHassanen, 2009). A COX inhibitor, meloxicam, reduces fibrosis in thissame model (Kim et al., 2008), although contradictory results havebeen reported using celecoxib (Yu et al., 2009).

A third class of arachidonic acid-derivedmediators, the EETs, displayanti-inflammatory and anti-fibrotic properties (Stables and Gilroy,2011). The EETs are metabolized by soluble epoxide hydrolase (sEH),producing dihydroxymolecules that are less lipophilic andmore readily

103T.R. Harris et al. / Toxicology and Applied Pharmacology 286 (2015) 102–111

conjugated, leading to their removal from the site of action (Morisseauand Hammock, 2013). sEH inhibitors modulate the levels of epoxy fattyacids (EpFAs) in tissues and dramatically reduce acute systemic inflam-mation in LPS-inducedmodels aswell as attenuatefibrosis and inflamma-tion in the heart, kidney and pancreas (Li et al., 2008; Iyer et al., 2012;Kompa et al., 2013; Sirish et al., 2013). In diabetic models, sEH inhibitorsdisplay organ protective properties, reducing islet cell apoptosis in thepancreas and tissue damage in the liver (Iyer et al., 2012). It has beenhypothesized that these inhibitors act by blocking the major route ofmetabolism of organ protective and anti-inflammatory EpFA (Iyer et al.,2012). Given these reported findings, we evaluated a sEH inhibitor in amurine model of liver fibrosis.

Once activated by metabolism, carbon tetrachloride (CCl4) causesdamage to the liver through lipid peroxidation, resulting in an upregu-lation of pro-inflammatory pathways and eventual fibrosis (Iredale,2007). EpFAs have been implicated in this model. Both COX-2 and 5-LOX inhibitors have been found to reduce fibrosis after CCl4 treatment,and when used in combination, reduce both inflammation and necrosis(Horrillo et al., 2007), though there are contradictory results whendifferent COX-2 inhibitors are employed, as in the BDL model of fibrosis(Hui et al., 2006). Dietary supplementation with arachidonic acid andthe omega-3 docosahexaenoic acid (DHA) has been found to attenuatefibrosis in an ethanol-inducedmodel of liver damage (Song et al., 2008).

A growing body of evidence indicates that endoplasmic reticulum(ER) stress is a contributor to fibrotic diseases (Mollica et al., 2011;Lenna and Trojanowska, 2012; Tanjore et al., 2012). Directly relevantto this study, recent reports have demonstrated an increase in ER stressin rodents injected with CCl4 (Ji et al., 2011; Lee et al., 2011; Jin et al.,2013). However, the link between the development of hepatic fibrosisand ER stress is still poorly understood. The ER is highly responsive tonutrients and to the energy state of the cell. It plays an important rolein folding of newly synthesized proteins. When the folding capacity ofthe ER is exceeded, misfolded proteins accumulate and lead to ER stress(Schroder and Kaufman, 2005). Cells use adaptivemechanisms to coun-ter the deleterious effects of ER stress known as the unfolded proteinresponse (Kaufman et al., 2002). The unfolded protein response consistsof three major branches that are controlled by the ER transmembraneproteins PKR-like ER-regulated kinase (PERK), inositol requiring protein1α (IRE1α), and activating transcription factor 6 (ATF6) (Hotamisligil,2010; Hummasti and Hotamisligil, 2010; Ron and Walter, 2007). Inparticular, the IRE1α sub-arm of ER stress signaling is critical for theunfolded protein response in fibrotic tissues (Chiang et al., 2011;Martino et al., 2013). The various unfolded protein response sub-armssynergize to attenuate stress by increasing the folding capacity of theER (Schroder andKaufman, 2005). However, if the compensatorymech-anisms fail to facilitate the adaptation of cells to ER stress, induction ofthe unfolded protein response can lead to elimination of stressed cellsby apoptosis (Zinszner et al., 1998; Nishitoh et al., 2002). We recentlydemonstrated that sEH deficiency or pharmacological inhibition inmice attenuated chronic high fat diet-induced ER stress in liver andadipose tissue in a cell autonomous manner (Bettaieb et al., 2013).

Based on the above informationwe examinedmarkers of ER stress inourmurinemodel of liver fibrosis and asked if they could be reduced bytreatment with sEH inhibitors.

Materials and methods

Reagents. TPPU (1-trifluoromethoxyphenyl-3-(1-propionylpiperidin-4-yl) urea) (Rose et al., 2010) and t-TUCB (trans-4-{4-[3-(4-trifluoromethoxy-phenyl)-ureido]-cyclohexyloxy}-benzoic acid),(Hwang et al., 2007) were synthesized and the physical propertieswere assessed as previously reported. Antibodies for pPERK (Thr980),PERK, peIF2α (Ser51), eIF2α, sXBP1, ATF6, IRE1α, and BiP were pur-chased from Santa Cruz Biotechnology (Santa Cruz, CA) while phospho-p38 (Thr180/Tyr182), p38, pJNK (Thr183/Tyr185), JNK and cleaved cas-pase3 antibodies were from Cell Signaling (Beverly, MA). Antibodies for

pIRE1α (Ser724) were purchased from Abcam (Cambridge, MA). Horse-radish peroxidase (HRP)-conjugated secondary antibodies were pur-chased from BioResources International (Carlsbad, CA). Unlessotherwise indicated, all other chemicals were purchased from Sigma(St. Louis, MO).

Animals. All animal studies were approved by the University of CaliforniaDavis AnimalUse andCare Committee andwere performed in accordancewith the National Institutes of Health guide for the care and use of labora-tory animals. Male C57BL (20 g) were obtained from Charles River Labo-ratories (Wilmington, MA). A colony of C57BL mice with a disruption inthe sEH gene resulting in a functional knockout is maintained by the UCDavis Mouse Biology Program (Luria et al., 2007). Mice were dividedinto four groups of 8: The Control group was given intraperitoneal injec-tions (I.P.) of sterile filtered Neobee M-5 (Fisher Scientific, Houston, TX)everyfive days forfiveweeks, for a total of 7 injections. They also receiveddrinking water containing 1% PEG-400 (Fisher Scientific). The CCl4-onlygroup was given I.P. injections of 197 mg/kg CCl4 (Sigma) mixed 1:7 insterile filtered Neobee M-5 every five days for five weeks, for a total of 7injections. This dose and route of administration was chosen since it hasbeen shown to induce robust fibrosis in C57BL mice (Constandinou,et al., 2005). They received drinkingwater containing 1% PEG-400 (FisherScientific). The CCl4 + TPPU group was treated with CCl4 similar to theCCl4-only group and received drinking water with 10 mg/L TPPU and 1%PEG-400, prepared as follows. TPPU was dissolved in PEG-400 to give a1000mg/L clear solution. This solutionwas then added towarm drinkingwater with rapid stirring to give the 10mg/L solution of TPPU in 1% PEG-400 in drinking water. The water was provided ad lib. The TPPU-onlygroup was given I.P. injections of sterile filtered Neobee M-5 like theControl group, and received drinking water containing TPPU similar tothe CCl4+ inhibitor groups. In the experimentwith the second sEH inhib-itor, t-TUCBwas solubilized in PEG-400 and then added to drinkingwater,resulting in a final concentration of 10mg/L TPPU and 1% PEG-400. Basedon estimated daily water consumption, a concentration of 10 mg/Linhibitor in drinking water will result in a dose of approximately1.7 mg/kg/day. The mice were euthanized three days after the finalinjection.

Histology. Liver samples were embedded, sectioned and stained by theUC Davis Veterinary Medical Teaching Hospital Anatomic PathologyService (Davis, CA). The slides were imaged using a Nikon Diaphotinverted microscope and quantified using ImageJ (NIH, Bethesda, MD).

Protein analysis. Mouse tissues were dissected and immediately frozenin liquid nitrogen. For zymography, liver samples were homogenizedin cold 25 mM Tris–HCl buffer using 3 × 30 s bursts of a roto-statorhomogenizer with one minute intervals in between. Zymography wasperformed using Novagen pre-cast gelatin zymography gels (Invitrogen,Carlsbad, CA) according to manufacturer's instructions and quantifiedusing ImageJ (NIH). For Western blots, tissues were ground in the pres-ence of liquid nitrogen and lysed using radio-immunoprecipitationassay (RIPA) buffer (10mMTris–HCl, pH 7.4, 150mMNaCl, 0.1% sodiumdodecyl sulfate [SDS], 1% Triton X-100, 1% sodium deoxycholate, 5 mMEDTA, 1mMNaF, 1mM sodium orthovanadate and protease inhibitors).Lysates were clarified by centrifugation at 10,000 ×g for 10 min andprotein concentrationswere determined using a bicinchoninic acid pro-tein assay kit (Thermo Scientific,Waltham,MA). Proteinswere resolvedby SDS-PAGE and transferred to PVDF membranes. Immunoblots wereperformed with the relevant antibodies. Proteins were visualized usingLuminata™ Forte (Millipore, Billerica, MA). For quantitation purposes,pixel intensities of immuno-reactive bands from blots that were in thelinear range of loading and exposure were quantified using FluorChem9900 (Alpha Innotech, San Lenardo, CA).

104 T.R. Harris et al. / Toxicology and Applied Pharmacology 286 (2015) 102–111

Quantitative real time PCR. Fresh liver samples were immersed inRNALater (Qiagen, Carol Stream, IL) and left at 4 °C overnight beforestorage at −80 °C. Liver tissue was homogenized by roto-stator in asingle 45 s burst and the lysates were passed through a QIAshreddercolumn according to manufacturer's suggestion (Qiagen). Total RNAwas isolated using the RNeasy miniprep kit (Qiagen) and first strandcDNA synthesized using the RT2 First Strand Kit (Qiagen). Gene expres-sionwas assessed by the RT2 Profiler PCR Array (Mouse Fibrosis) for ABI7500 Fast Block (Invitrogen). For qRT-PCR mRNA of BiP, sXBP1, ATF4,ATF6 and CHOP were assessed by reverse transcription PCR (iCycler,BioRad, Hercules, CA) and normalized to TATA-Box binding protein.For RT-PCR, Absolute blue qPCR premix (Fisher Scientific) was mixedwith each primer. ATF4 primers: 5′-GACCTGGAAACCATGCCAGA-3′(For.), 5′-TGGCCAATTGGGTT CACTGT-3′ (Rev.); ATF6 primers: 5′-TGTATTACGCCTCCCCTGGA-3′ (For.), 5′-ACACAGCAGAACCAACACCA-3′(Rev.); CHOP primers: 5′-CCCTGCCTTTCACCTTGG-3′ (For.), 5′-CCGCTCGTTCTCCTGCTC-3′ (Rev.); BiP primers: 5′-ACTTGGGGACCACCTATTCCT-3′ (For.), 5′-ATCGCCAATCAGACGCTCC-3′ (Rev.); sXBP-1primers: 5′-GGTCTGCTGAGTCCGCAGCAGG-3′ (For.), 5′-AGGCTTGGTGTATA CATGG-3′ (Rev.); TATA binding protein primers: 5′-TTGGCTAGGTTTCTGCG GTC-3′ (For.), 5′-TGCTGTAGCCGTATTCATTG-3′ (Rev.).

LC-MS/MS analysis of plasma oxylipins. Plasma oxylipins were extractedand analyzed by LC-MS/MS as previously described (Yang et al., 2009).Blood levels of inhibitors were determined by LC-MS/MS as previouslydescribed (Liu et al., 2013). Retention time, selected reaction monitor-ing and internal standard information used to quantify all analytes arein the supplementary materials Table S1.

LC-MS/MS analysis of hydroxyproline. The quantification of hepatic tissuehydroxyproline by LC-MS/MS was performed using the method devel-oped by Kindt et al. (2003) with the following changes. The analysiswas carried out using aMicromass Quattro Ultima triple quadrupole tan-demmass spectrometer (Micromass, Manchester, UK) equipped with anelectrospray ionization interface. The HPLC system consisted of a Watersmodel 2790 separations module (Waters, Milford, MA) equipped witha Waters model 2487 dual-wavelength absorbance detector. The massspectrometer was coupled to the outlet of the HPLC (Pursuit C18,2.0 × 150 mm, 3 μm column (Agilent, Englewood, CO)).

Statistical analysis. All data are expressed asmean± standard error. TheStudent's t-test was used to test for differences in groups in all experi-ments. Statistical analyses were performed using Excel (Microsoft,Redmond, WA). Differences were considered significant with P-value b 0.05 or lower. Sample size “N” refers to the number of animals.

Results

Administration of the sEH inhibitor TPPU reduced CCl4-induced collagendeposition and collagen mRNA expression

Animals were treated with CCl4 for five weeks to induce liver fibro-sis. To potentially reduce development of this fibrosis, the sEH inhibitorTPPUwas chosen due to its potency on and selectivity for the sEH targetand because of its ADME properties in mice (Liu et al., 2013). TPPU wasadministered in drinkingwater startingwith thefirst CCl4 injection, andblood levels of the compound were assessed at 2.5 weeks and at thetermination of the experiment. At both time points, taken at 8 AMfrommice on a 7 AM to7 PM light cycle, the TPPU-treated concentrationof the inhibitor was approximately 500 nM (2.5 wk = 511 ± 37 nM,5 wk = 543 ± 17 nM). All experimental groups gained weight duringthe six week experiment (supplementary materials). The CCl4-onlyand TPPU-only groups displayed a greater weight gain than theCCl4 + TPPU and Control groups. Three days after the last injection ofCCl4, the livers were removed and prepared for protein, mRNA, andhistological analysis. Sections were stained with picrosirius red (Fig. 1A).

The percent area stained by picrosirius red relative to the Controlgroup increased five-fold in the CCl4-only group, while the CCl4 + TPPUgroup returned to the level of the Control group (Fig. 1B). The CCl4-onlygroup displayed a roughly six-fold increase in total hepatic tissue hy-droxyproline as determined by LC-MS/MS (Fig. 1C). The CCl4 + TPPUgroupwas not statistically different (P-value N 0.05) from either the Con-trol or the CCl4-only groups (Fig. 1C). Staining for alpha smooth muscleactin (α-SMA) revealed an increase in percent area stained in the CCl4-only group relative to the Control group that was not significantly alteredby TPPU treatment (supplementarymaterials Fig. S2). A PCR array exper-iment (described in detail below) determined that themessages for colla-gens 1a2 and 3a1 were upregulated approximately sixteen fold in theCCl4-only group (Fig. 3). The upregulation was reduced by 50% in theCCl4 + TPPU group. This indicates that TPPU treatment has an effect onboth the expression and deposition of collagen due to CCl4 treatment.

Administration of the sEH inhibitor TPPU modulated the mRNA expressionand activity of MMPs and the mRNA expression of other components ofthe ECM

In order to assess the signalingpathways involved in the anti-fibroticeffect of TPPU treatment, a PCR array was run that included growthfactors, cytokines, chemokines and components of the ECM involvedin the progression of fibrosis. When compared to the Control group,the CCl4-only group displayed 33 genes upregulated 1.5 fold or morewith a P-value of 0.5 or lower (Fig. 2). The CCl4 + TPPU group displayed20 genes upregulated 1.5 fold or more with a P-value of 0.5 or lowercompared to the Control group (Fig. 2). The thirteen genes that differen-tiated the CCl4-only and CCl4 + TPPU groups with a P-value of 0.5 orlower are displayed in Fig. 3. To validate these results, themRNA expres-sion of three genes included in the array were independently deter-mined by RTPCR and found to display the same trend as reportedby the array (supplementary materials Table S3). Compared to theCCl4-only group, the CCl4 + TPPU group displayed a decrease in mRNAexpression of MMP-2, -8, -9, -13, and -14 (Fig. 3). The largest reductionoccurring forMMP-8 andMMP-9, whichwere decreased approximatelyfour-fold with respect to the CCl4-only group.MMP-1awas increased byapproximately two-fold in the CCl4-only and CCl4 + TPPU groups com-pared to control (supplementarymaterials Table S2). Of the tissue inhib-itors of metalloproteases (TIMPs), only TIMP-2 showed a statisticallysignificant (P-value b 0.05) modulation in the CCl4-only group relativeto the Control group (supplementary materials Table S2). In theCCl4-only and CCL4 + TPPU groups, TIMP-2 expression was upregulat-ed 3.5 and 2.5 fold, respectively. The modulation of the expression ofseveral MMPs prompted us to examine MMP activity in the hepatictissues. Using a randomly selected subset of samples from the experimen-tal groups, we performed gelatin zymography, which assays MMP-1, -2and -9 activity, primarily (Snoek-van Beurden and Von den Hoff, 2005).Intracellular pools of MMP-1a, -2, and -9 contain both pro- and active-forms of the enzymes, the active form is produced when a roughly10 kDa peptide is cleaved from the pro-form. In zymography, theenzymes are separated by native PAGE and then both forms of theenzymes are activated. We detected an increase in gelatinase activity inthe CCl4-only group in the 90–107 kDa range relative to the Controlgroup, which spans the molecular weight of the 92 kDa pro-MMP-9(Fig. 4). This activity was reduced in the CCl4 + TPPU group. We alsodetected a large increase in gelatinase activity in the 45 –54 kDa rangein the CCl4 + TPPU group relative to the Control group, correspondingto the molecular weight of active MMP-1a, which has a roughly 57 kDapro-form and a 48 kDa active form (Balbin et al., 2001). This activitywas only slightly increased in the CCl4-only group. We also detected amoderate increase in gelatinase activity in the 62–69 kDa molecularweight range in the CCl4-only and CCl4 + TPPU groups, correspondingto active MMP-2, which is roughly 62 kDa (supplementary materialsFig. S3).

Fig. 1.TPPU attenuates collagen deposition due to CCl4 treatment.Micewere injected (I.P.)with CCl4 for 5weeks as described inMaterials andmethods. TPPUwas administered indrinkingwater startingwith CCl4 treatment. A) Representative slides of liver sections stainedwith picrosirius red (40×). B) Quantification of staining expressed in percent area. C) Quantification ofhydroxyproline levels in liver tissue. Error bars represent standard error. *P-value vs. Control group b 0.05. ^P-value vs. CCl4-only group b 0.05. N = 6–8 animals per group.

105T.R. Harris et al. / Toxicology and Applied Pharmacology 286 (2015) 102–111

TPPU treatment significantly modulated the mRNA expressionof two other components of the ECM, integrin α2 (Itgα2) andThrombspondin 2. The three-fold induction in Itgα2 between theCCl4-only and Control group was reduced by almost 50% in theCCl4 + TPPU group (Fig. 3). Thrombspondin 2 was increased almostsix-fold in the CCl4-only group relative to the Control group. Thisincrease was attenuated by 33% in the CCl4 + TPPU group (Fig. 3).

Administration of TPPU modulated the inflammatory response

Of the chemokines and chemokine receptors in the PCR array, tworeceptors, C-C chemokine receptor 2 and C-X-C motif chemokinereceptor-4were significantly reduced in the CCl4+ TPPU group relativeto the CCl4-only group (Fig. 3). C-C chemokine receptor 2was increasedover five-fold in the CCl4-only group relative to the Control group, andthis increase was almost halved in the CCl4 + TPPU group. Similarly,an over 4-fold increase in C-X-C motif chemokine receptor-4 mRNAexpression in the CCl4-only group relative to the Control group wasreduced by 50% in the CCl4 + TPPU group. The PCR array also revealedthat an almost 2 fold increase in TGFβ1 in the CCl4-only group relativeto the Control group was decreased to control level in the CCl4 + TPPUgroup (Fig. 3).

Since sEH inhibitors have been shown tomodulate the inflammatoryEpFA in other disease models, we examined blood levels of oxylipids(supplementary materials Table S4, Figs. S4, and S5). We did not see astatistically significant difference in most classes, including the EETs.However, one lipid mediator, the 12-LOX metabolite 12-HETE, wasincreased over 3-fold in the CCl4 group. Levels of this EpFAwere reducedin the CCl4 + TPPU group, but did not reach statistical significance.

Administration of TPPU attenuated CCl4-induced ER stress

ER stress is an emerging factor in fibrotic diseases and a number ofstudies have identified significant correlation between hepatic fibrosisand ER stress (Mollica et al., 2011; Lenna and Trojanowska, 2012;Tanjore et al., 2012), although the precise mechanism is not clear. Inthis study we investigated whether exposure to CCl4 induces hepaticER stress in vivo and evaluated the effects of sEH pharmacological inhi-bition on CCl4-induced hepatic ER stress. In line with published studiesexposure to CCl4 induced ER stress (Lewis and Roberts, 2005; Lee et al.,2011; Jin et al., 2013), as evidenced by increased PERK (Thr980), eIF2α(Ser51) and IRE1α (Ser724) phosphorylation and increased sXBP1expression in liver lysates (Fig. 5A). Importantly, pharmacological inhi-bition of sEH attenuated hepatic ER stress in CCl4-treated mice as evi-denced by decreased PERK (Thr980), eIF2α (Ser51) and IRE1α (Ser724)phosphorylation and decreased sXBP1 expression (Fig. 5A). In linewith these findings, hepatic mRNA of BiP, sXBP1 ATF4, ATF6 and CHOPwas decreased upon pharmacological inhibition of sEH compared withCCl4-treated mice, indicating attenuated ER stress (Fig. 5B).

Excessive ER stress leads to the induction of inflammatory responsesand eventually cell death (Zha and Zhou, 2012). In addition, CCl4-induced liver injury has been shown to involve TNF-α- and TGF-β-mediated activation of JNK and cell death (Hong et al., 2013; Ma et al.,2014; Wu and Cederbaum, 2013). Thus, we investigated the effects ofpharmacological inhibition of sEH on CCl4-induced inflammation. Con-sistent with previous reports, induction of liver fibrosis correlatedwith increased hepatic JNK and p38 phosphorylation (Fig. 5C). On theother hand, sEH pharmacological inhibition reduced CCl4-inducedphosphorylation of JNK and p38 (Fig. 5C). After exposure to apoptoticstimuli, cells activate initiator caspases that proteolytically cleave and

Fig. 2. TPPU treatment reduces the upregulation of pro-inflammatory and pro-fibrotic genesassociatedwith fibrosis. PCR array analysis of hepatic tissues after 5weeks of CCl4 treatment.TPPUwas administered in drinkingwater startingwith CCl4 treatment. A volcano plot of PCRarray data fromhepatic tissue. The log of the fold regulation of an experimental group relativeto control is plotted versus the log base 10 of the P-value. A P-value of 0.05 is indicated by thehorizontal line. The vertical lines on either side of the axis represent 1.5-fold inductionrelative to the Control group. A) The change in CCl4-only group versus the Control group.B) The change in CCl4 group relative to the CCl4 + TPPU group. N = 6–8 animals pergroup. Full results in supplementary data Table S2.

Fig. 3. TPPU treatment significantly downregulates the hepatic mRNA expression of thirteengenes compared to the CCl4-only group. PCR array analysis of hepatic tissues after 5weeks ofCCl4 treatment. TPPU was administered in drinking water starting with CCl4 treatment. Thegenes displayed a 1.5 fold upregulation or higher in the CCl4-only group relative to the Con-trol group, and a concomitant downregulation in theCCl4+TPPUversus the CCl4-only groupwith a P-value of 0.05 or lower. *P-value vs. Control group ≤ 0.05; **P-value vs. Controlgroup ≤ 0.01; ~P-value vs. CCl4-only group ≤ 0.05; ~ ~ P-value vs. CCl4-only group ≤ 0.01.N = 6–8 animals per group.

106 T.R. Harris et al. / Toxicology and Applied Pharmacology 286 (2015) 102–111

activate effector caspases (such as caspase-3) to dismantle the dyingcell. Caspase3 is implicated in ER stress-induced cell death (Zhanget al., 2011a,b). Accordingly, we determined ER stress-induced expres-sion of active caspase-3 in CCl4-treatedmice versusmicewith combinedtreatment (CCl4 + TPPU). Consistent with published reports (Chanet al., 2013; Moran-Salvador et al., 2013), treatment with CCl4 causeda significant increase in the expression of the active form of caspase3whereas sEH inhibition significantly decreased CCl4-induced caspase-3activation (Fig. 5D). Together, these results indicate that the pharmaco-logical inhibition of sEH can provide a protective mechanism againstCCl4-induced ER stress in the liver.

Disruption of the sEH gene or administration of an alternate sEH inhibitorreduced collagen deposition

In order to support the hypothesis that TPPU is acting to reducethe hepatic fibrosis and ER stress through its action as a sEH inhibi-tor, wild-type mice were treated with a second sEH inhibitor witha different structure, trans-4-{4-[3-(4-trifluoromethoxy-phenyl)-ureido]-cyclohexyloxy}-benzoic acid (t-TUCB). Blood levels of the com-pound were determined using the procedure described above for TPPUand found to be 440 ± 36 nM midway through the experiment

(2.5 week time point) and 481 ± 23 nM at the end (5 week timepoint.) The t-TUCB treated animals displayed a roughly 25% reductionin collagen deposition as determined by picrosirius red staining (Figs.6A and B).

Quantification by LC-MS/MS revealed that the total tissue hydroxy-proline levels in the livers from t-TUCB treated animals were not sta-tistically different from either the Control group or the CCl4-onlygroup (P-value N 0.05) (Fig. 6C). To further test the involvement ofthe sEH, C57BL mice with a disruption in the gene for sEH were treatedwith CCl4, TPPU, or both. The hydroxyproline levels in the livers ofthe CCl4-treated Null animals were not statistically different (P-value N 0.05) than the Null Control group, wild-type Control, or thewild-type CCl4-only group (Fig. 6C). However, the CCl4-treated Nullanimals displayed significantly less (P-value b 0.05) collagen depositionas judged by picrosirius red staining, comparable to the CCl4 + t-TUCBgroup (Figs. 6A and B). The Null + CCl4 + TPPU group did not showsignificant differences (P-value N 0.05) in the hydroxyproline levels orthe collagen deposition from the Null + CCl4 group (Fig. 6).

Discussion

In this study, the inhibition of sEH, the major route of metabolism forEpFA such as the EETs, resulted in a dramatic attenuation of manymarkers of liver fibrosis and ER stress associated with CCl4 treatment.TPPU has an IC50 of 8.3 nM when assayed with the recombinant murineenzyme and cyano(2-methoxynaphthalen-6-yl)methyl(3-phenyloxiran-2-yl)methyl carbonate (CMNPC) as the substrate (Liu et al., 2013), sothe plasma levels of TPPU in this studywerewell above the concentration

Fig. 4. TPPU modulates metalloprotease activity in hepatic tissues following CCl4 treatment. A) Gelatin zymography. 30 ug/lane of hepatic tissue lysates from Control and treated groups.B) Analysis of five major bands on the gels by densitometry.

107T.R. Harris et al. / Toxicology and Applied Pharmacology 286 (2015) 102–111

required to inhibit the enzyme. TPPU treatment not only altered collagendensity, but also the composition of the ECM, reducing collagen deposi-tion and collagen 1a2 and 3a1 mRNA expression in the CCl4 + TPPUgroup compared to the CCl4-only group. These changes were accompa-nied by dramatic shifts in MMP activity and expression, although TPPUdid not alter the upregulation of TIMP-2 observed in the CCl4-onlygroup. Taken together, the decrease in collagen deposition and differ-ences in expression patterns of collagen isoforms and MMPs betweenthe CCl4 + TPPU and CCl4-only groups indicate that the ECM in theCCl4+ TPPU groupmore closely resembles the control tissue. The shiftsin MMP expression, especially, may have implications for tissue regen-eration and recovery from fibrosis, where there is evidence that sEHplays a role. A recent study reported that treatmentwith a sEH inhibitorincreased liver to body weight ratio after partial hepatectomy and sEHnull mice displayed accelerated skin wound healing (Panigrahy et al.,2013). Regenerative physiological processes like these are facilitatedby ECM reorganization and construction, which require changes inMMP activity.

Our zymography yielded unexpected results. We detected a modestincrease in pro-MMP-9 activity in the CCl4-only group that was attenu-ated by TPPU treatment and no change inMMP-2 activity. UpregulationofMMP-9 expression and increasedMMP-9 activity have been associat-ed with the progression of liver fibrosis in both human and rodent(Kurzepa et al., 2014), and therefore reduction inMMP-9 activity shouldattenuate pro-fibrotic matrix remolding. By far, the most striking resultwas the dramatic increase in gelatinase activity in the 45–54 kDa rangein the TPPU + CCl4 group relative to all other groups. This molecularweight range corresponds to active MMP-1a (Balbin et al., 2001).MMP-1a, first characterized in 2001, has been primarily investigatedin the context of cancer, where knockout of the gene causes an increase

in tumor growth and angiogenesis in a Lewis lung carcinoma mousemodel (Fanjul-Fernández, et al., 2013; Foley et al., 2014). The role thatMMP-1a may play in the observed anti-fibrotic effects of sEH inhibitionis currently under investigation.

While the change in most inflammatory cytokines and growth fac-tors did not pass the statistical test, TGFβ1 was substantially reducedin the CCl4 + TPPU group relative to the CCl4-only group. TGFβ impactsfibrosis through multiple mechanisms including the modulation ofMMP and TIMP expression and the stimulation of the production ofmatrix components such as the collagens and integrin α2 (Baghyet al., 2012). The reduction of TGFβ in the TPPU + CCl4 relative to theCCl4-only group is further indirect evidence of the changes to the ECMelicited by sEH inhibition. Our observation parallels a result obtainedin a rat diabetic model of kidney fibrosis, in which the elevation ofTGFβ in the fibrotic kidney was reduced by administration of a sEHinhibitor (Elmarakby et al., 2013). It should be noted that these mRNAexpression data may not reflect hepatic protein or enzymatic activitylevels.

Because sEH inhibition changes oxylipid metabolism, we examinedplasma levels of EpFA. Despite a high degree of variation, 12-HETE wasidentified as a potential marker of CCl4-induced liver fibrosis. 12-HETEhas been previously identified in the sputum of patients with cystic fibro-sis (Yang et al., 2012). It is generated by 12-lipoxygenase (12-LOX),whichhas been studied in the context of cardiac hypertrophy and fibrosis.Overexpression of 12-LOX in rat cardiac fibroblasts results in increasedcollagen and fibronectin secretion (Wen et al., 2003).

Recent studies provided evidence that ER stress can affect produc-tion of the extracellular matrix and consequently the progression offibrotic diseases (Mollica et al., 2011; Lenna and Trojanowska, 2012;Tanjore et al., 2012), including hepatic fibrosis that results from the

Fig. 5. sEH pharmacological inhibitionmitigates CCl4-induced ER stress in vivo. Mice were treatedwith sEH inhibitor (TPPU) or vehicle control (PEG 400) as detailed inMethods. (A) Liverlysates were immunoblotted for pPERK (Thr980), PERK, peIF2α (Ser51), eIF2α, pIRE1α (Ser724), IRE1α, spliced X-box binding protein 1 (sXBP1), BiP and Tubulin. Each lane represents asample from a separate animal. Bar graphs represent data expressed as arbitrary units (A.U.) for pPERK/PERK, peIF2α/eIF2α, pIRE1α/IRE1α and sXBP1/Tubulin, from at least six mice andpresented as means + SEM. (B) BiP, sXBP1, ATF4, ATF6 and CHOP mRNA from liver was measured by quantitative real-time PCR and normalized against TATA binding protein. Datarepresent means ± SEM of six mice. (C–D) Immunoblots of pJNK (Thr183/Tyr185), JNK, pp38 (Thr180/Tyr182), p38, caspase3 and Tubulin. Bar graphs represent normalized dataexpressed as arbitrary units (A.U.) for pJNK/JNK, pp38/p38 and caspase3/Tubulin from at least six mice. (*) indicates significant difference between CCL4-treated and non-treated mice,(#) indicates significant difference between TPPU-treated and non-treated mice.

108 T.R. Harris et al. / Toxicology and Applied Pharmacology 286 (2015) 102–111

injection of CCl4 (Ji et al., 2011; Lee et al., 2011; Jin et al., 2013). In a non-alcoholic steatohepatitis methionine–choline-deficient diet-inducedmodel, liver fibrosis correlated with the increased expression of BiPand activation of caspases 12 and 7 (Mu et al., 2009). Moreover, soluble

mediators secretedmainly by the ER in steatotic hepatocytes and activat-ed Kupffer cells mediate their trans-differentiation into myofibroblasts,which secrete fibrogenic cytokines and matrix components that triggerfibrosis (Yang et al., 2007; Mollica et al., 2011). Indeed, fibrogenic activity

Fig. 6. t-TUCB and sEH null background attenuate collagen deposition due to CCl4 treatment. Mice were injected (I.P.) with CCl4 for 5weeks as described inMaterials andmethods. t-TUCBwas administered in drinking water starting with CCl4 treatment. C57BL mice with a sEH gene disruption (Null) were treated with CCl4, TPPU or both. A) Representative slides of liversections stainedwith picrosirius red (40×). B) Quantification of staining expressed in percent area. C) Quantification of hydroxyproline levels in liver tissue. Error bars represent standarderror. †P-value vs. wild-type Control group b 0.001. ‡P-value vs. wild-type CCl4-only group b 0.001. *P-value vs. wild-type Control group b 0.05. ^P-value vs. wild-type CCl4-onlygroup b 0.05. N = 6–8 animals per group.

109T.R. Harris et al. / Toxicology and Applied Pharmacology 286 (2015) 102–111

in hepatic stellate cells increases upon induction of ER stress (Hernandez-Gea et al., 2013). Recently, we identified sEH as a regulator of ER stress inmurine models of high fat diet-induced ER stress. sEH deficiency orpharmacological inhibition attenuated high fat diet-induced ER stress inliver and adipose tissues (Bettaieb et al., 2013). In addition, sEH inhibitionattenuates high fat diet-induced hepatic steatosis due to reduced inflam-matory state (Liu et al., 2012). sEH inhibition also prevents renal fibrosisthrough reductionof TGF-β andfibronectin expression in diabetic sponta-neously hypertensive rats (Elmarakby et al., 2013). Together, these datasuggest that sEH contributes to ER stress-mediated TGF-β regulation ofECM production and trafficking. Genetic and chemical knockouts of thesEH as well as administration of EpFA alter a diversity of pathologicalstates generally toward improved health. In several of these situations,now including CCl4-induced hepatic fibrosis, reduction in ER stressappears to be a common underlying mechanism.

In order to address if TPPU was targeting the sEH as its primarymechanism of reducing fibrosis and ER stress, two experimental strate-gies were employed. First, animals were treated with an alternative sEHinhibitor with a substantially different structure than TPPU, t-TUCB,which has an IC50 of 11 nMwhen assayedwith the recombinantmurineenzyme and CMNPC as the substrate (Liu et al., 2013). Second, sEH null

mice were used. Both the CCl4+ t-TUCB and Null + CCl4 groupsdisplayed a reduction in collagen deposition compared to the CCl4-only group, as judged by picrosirius red staining. Importantly, TPPUtreatment in the null background did not yield a further reduction infibrosis. Taken together, these data indicate that TPPU is producing theobserved anti-fibrotic effects through inhibition of sEH. Hopefully, abetter understanding of sEH function will provide new insight into thedevelopment of hepatic fibrosis due to exposure to carbon tetrachlorideand other industrial and environmental chemicals.

Abbreviations

ATF6: Activating transcription factor; BiP: Immunoglobulin-heavy-chain-binding protein; CCl4: Carbon tetrachloride; CHOP: 4 chemokinereceptor 2; COX: Cyclooxygenase; ECM: Extracellular matrix; EpFA:Epoxy fatty acids; ER: Endoplasmic reticulum; JNK: Jun N-terminalkinase; LOX: Lipoxygenase; MMP: Matrix metalloprotease; PERK:PKR-like ER kinase; sEH: Soluble epoxide hydrolase; TIMP: Tissue in-hibitors of metalloproteases; TPPU: 1-trifluoromethoxyphenyl-3-(1-propionylpiperidin-4-yl) urea; t-TUCB: trans-4-{4-[3-(4-trifluoromethoxy-phenyl)-ureido]-cyclohexyloxy}-benzoic acid.

110 T.R. Harris et al. / Toxicology and Applied Pharmacology 286 (2015) 102–111

Conflict of interest statement

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.taap.2015.03.022.

Transparency document

The Transparency document associated with this article can befound, in the online version.

Acknowledgments

This work was supported by NIEHS grant R37 ES02710 (to B.D.H.)and NIEHS Superfund P42 ES04699 (to B.D.H.). Partial support wasprovided by the West Coast Comprehensive Metabolomics ResourcesCore NIH/NIDDK U24 DK097154 (to B.D.H.). T.R.H. is also supportedby NIH T32 training grant in basic and translational cardiovascularscience (T32 HL86350). Research in the F.G.H. laboratory is supportedby R01DK090492. A.B is supported by K99/R00 (1K99DK100736-01).Research in the N.C. laboratory is supported by the NIH grant HL85727and HL85844 and Veterans Administration Merit Review Grant I01BX000576. Imaging of αSMA was conducting using the CAMI corefacility at UC Davis Center for Health and the Environment.

References

Baghy, K., Iozzo, R.V., Kovalszky, I., 2012. Decorin-TGFbeta axis in hepatic fibrosis and cir-rhosis. J. Histochem. Cytochem. 60, 262–268.

Balbin, M., Fueyo, A., Knauper, V., Lopez, J.M., Alvarez, J., Sanchez, L.M., Quesada, V.,Bordallo, J., Murphy, G., Lopez-Otin, C., 2001. Identification and enzymatic character-ization of two diverging murine counterparts of human interstitial collagenase(MMP-1) expressed at sites of embryo implantation. J. Biol. Chem. 276, 10253–10262.

Bataller, R., Brenner, D.A., 2005. Liver fibrosis. J. Clin. Invest. 115, 209–218.Bettaieb, A., Nagata, N., AbouBechara, D., Chahed, S., Morisseau, C., Hammock, B.D., Haj, F.G.,

2013. Soluble epoxide hydrolase deficiency or inhibition attenuates diet-induced endo-plasmic reticulum stress in liver and adipose tissue. J. Biol. Chem. 288, 14189–14199.

Chan, C.C., Lee, K.C., Huang, Y.H., Chou, C.K., Lin, H.C., Lee, F.Y., 2013. Regulation by resver-atrol of the cellular factors mediating liver damage and regeneration after acute toxicliver injury. J. Gastroenterol. Hepatol. 29, 603–613.

Chiang, C.K., Hsu, S.P., Wu, C.T., Huang, J.W., Cheng, H.T., Chang, Y.W., Hung, K.Y., Wu, K.D.,Liu, S.H., 2011. Endoplasmic reticulum stress implicated in the development of renalfibrosis. Mol. Med. 17, 1295–1305.

Constandinou, C., Henderson, N., Iredale, J.P., 2005. Modeling liver fibrosis in rodents.Methods Mol. Med. 117, 237–250.

Elmarakby, A.A., Faulkner, J., Pye, C., Rouch, K., Alhashim, A., Maddipati, K.R., Baban, B.,2013. Role of haem oxygenase in the renoprotective effects of soluble epoxidehydrolase inhibition in diabetic spontaneously hypertensive rats. Clin. Sci. (Lond.)125, 349–359.

El-Swefy, S., Hassanen, S.I., 2009. Improvement of hepaticfibrosis by leukotriene inhibition incholestatic rats. Ann. Hepatol. 8, 41–49.

Fanjul-Fernández, M., Folgueras, A.R., Fueyo, A., Balbín, M., Suárez, M.F., Fernández-García,M.S., Shapiro, S.D., Freije, J.M., López-Otín, C., 2013. Matrix metalloproteinase Mmp-1a is dispensable for normal growth and fertility in mice and promotes lung cancerprogression bymodulating inflammatory responses. J. Biol. Chem. 288, 14647–14656.

Foley, C.J., Fanjul-Fernández, M., Bohm, A., Nguyen, N., Agarwal, A., Austin, K., Koukos, G.,Covic, L., López-Otín, C., Kuliopulos, A., 2014. Matrix metalloprotease 1a deficiencysuppresses tumor growth and angiogenesis. Oncogene 33, 2264–2272.

Funk, C.D., 2001. Prostaglandins and leukotrienes: advances in eicosanoid biology. Science294, 1871–1875.

Hernandez-Gea, V., Friedman, S.L., 2011. Pathogenesis of liver fibrosis. Annu. Rev. Pathol.6, 425–456.

Hernandez-Gea, V., Hilscher, M., Rozenfeld, R., Lim, M.P., Nieto, N., Werner, S., Devi, L.A.,Friedman, S.L., 2013. Endoplasmic reticulum stress induces fibrogenic activity inhepatic stellate cells through autophagy. J. Hepatol. 59, 98–104.

Hong, I.H., Park, S.J., Goo, M.J., Lee, H.R., Park, J.K., Ki, M.R., Kim, S.H., Lee, E.M., Kim, A.Y.,Jeong, K.S., 2013. JNK1 and JNK2 regulate alpha-SMA in hepatic stellate cells duringCCl4-induced fibrosis in the rat liver. Pathol. Int. 63, 483–491.

Horrillo, R., Planaguma, A., Gonzalez-Periz, A., Ferre, N., Titos, E., Miquel, R., Lopez-Parra,M., Masferrer, J.L., Arroyo, V., Claria, J., 2007. Comparative protection against liverinflammation and fibrosis by a selective cyclooxygenase-2 inhibitor and a nonredox-type 5-lipoxygenase inhibitor. J. Pharmacol. Exp. Ther. 323, 778–786.

Hotamisligil, G.S., 2010. Endoplasmic reticulum stress and the inflammatory basis ofmetabolic disease. Cell 140, 900–917.

Hui, A.Y., Leung,W.K., Chan, H.L., Chan, F.K., Go, M.Y., Chan, K.K., Tang, B.D., Chu, E.S., Sung,J.J., 2006. Effect of celecoxib on experimental liver fibrosis in rat. Liver Int. 26,125–136.

Hummasti, S., Hotamisligil, G.S., 2010. Endoplasmic reticulum stress and inflammation inobesity and diabetes. Circ. Res. 107, 579–591.

Hwang, S.H., Tsai, H.J., Liu, J.Y., Morisseau, C., Hammock, B.D., 2007. Orally bioavailablepotent soluble epoxide hydrolase inhibitors. J. Med. Chem. 50, 3825–3840.

Iredale, J.P., 2007. Models of liver fibrosis: exploring the dynamic nature of inflammationand repair in a solid organ. J. Clin. Invest. 117, 539–548.

Iyer, A., Kauter, K., Alam, M.A., Hwang, S.H., Morisseau, C., Hammock, B.D., Brown, L., 2012.Pharmacological inhibition of soluble epoxide hydrolase ameliorates diet-inducedmetabolic syndrome in rats. Exp. Diabetes Res. 2012, 758614–758624.

Ji, C., Kaplowitz, N., Lau, M.Y., Kao, E., Petrovic, L.M., Lee, A.S., 2011. Liver-specific loss ofglucose-regulated protein 78 perturbs the unfolded protein response and exacerbatesa spectrum of liver diseases in mice. Hepatology 54, 229–239.

Jin, C.F., Li, B., Lin, S.M., Yadav, R.K., Kim, H.R., Chae, H.J., 2013.Mechanism of the inhibitoryeffects of eucommia ulmoides oliv. cortex extracts (EUCE) in the CCl4-induced acuteliver lipid accumulation in rats. Int. J. Endocrinol. 2013, 751854–751864.

Kaufman, R.J., Scheuner, D., Schroder, M., Shen, X., Lee, K., Liu, C.Y., Arnold, S.M., 2002. Theunfolded protein response in nutrient sensing and differentiation. Nat. Rev. Mol. CellBiol. 3, 411–421.

Kim, S.M., Park, K.C., Kim, H.G., Han, S.J., 2008. Effect of selective cyclooxygenase-2inhibitor meloxicam on liver fibrosis in rats with ligated common bile ducts. Hepatol.Res. 38, 800–809.

Kindt, E., Gueneva-Boucheva, K., Rekhter, M.D., Humphries, J., Hallak, H., 2003. Determina-tion of hydroxyproline in plasma and tissue using electrospray mass spectrometry.J. Pharm. Biomed. Anal. 33, 1081–1092.

Kompa, A.R., Wang, B.H., Xu, G., Zhang, Y., Ho, P.Y., Eisennagel, S., Thalji, R.K., Marino Jr.,J.P., Kelly, D.J., Behm, D.J., Krum, H., 2013. Soluble epoxide hydrolase inhibition exertsbeneficial anti-remodeling actions post-myocardial infarction. Int. J. Cardiol. 167,210–219.

Kurzepa, J., Madro, A., Czechowska, G., Kurzepa, J., Celinski, K., Kazmierak, W., Slomka, M.,2014. Role of MMP-2 andMMP-9 and their natural inhibitors in liver fibrosis, chronicpancreatitis and non-specific inflammatory bowel diseases. Hepatobiliary Pancreat.Dis. Int. 13, 570–579.

Lee, G.H., Bhandary, B., Lee, E.M., Park, J.K., Jeong, K.S., Kim, I.K., Kim, H.R., Chae, H.J., 2011.The roles of ER stress and P450 2E1 in CCl(4)-induced steatosis. Int. J. Biochem. CellBiol. 43, 1469–1482.

Lenna, S., Trojanowska, M., 2012. The role of endoplasmic reticulum stress and the unfoldedprotein response in fibrosis. Curr. Opin. Rheumatol. 24, 663–668.

Lewis, M.D., Roberts, B.J., 2005. Role of CYP2E1 activity in endoplasmic reticulumubiquitination, proteasome association, and the unfolded protein response. Arch.Biochem. Biophys. 436, 237–245.

Li, J., Carroll, M.A., Chander, P.N., Falck, J.R., Sangras, B., Stier, C.T., 2008. Soluble epoxidehydrolase inhibitor, AUDA, prevents early salt-sensitive hypertension. Front. Biosci.13, 3480–3487.

Liu, Y., Dang, H., Li, D., Pang, W., Hammock, B.D., Zhu, Y., 2012. Inhibition of soluble epox-ide hydrolase attenuates high-fat-diet-induced hepatic steatosis by reduced systemicinflammatory status in mice. PLoS One 7, e39165.

Liu, J.Y., Lin, Y.P., Qiu, H., Morisseau, C., Rose, T.E., Hwang, S.H., Chiamvimonvat, N.,Hammock, B.D., 2013. Substituted phenyl groups improve the pharmacokineticprofile and anti-inflammatory effect of urea-based soluble epoxide hydrolase inhibitorsin murine models. Eur. J. Pharm. Sci. 48, 619–627.

Luria, A., Weldon, S.M., Kabcenell, A.K., Ingraham, R.H., Matera, D., Jiang, H., Gill, R.,Morisseau, C., Newman, J.W., Hammock, B.D., 2007. Compensatory mechanism forhomeostatic blood pressure regulation in Ephx2 gene-disrupted mice. J. Biol. Chem.282, 2891–2898.

Ma, J.Q., Ding, J., Zhang, L., Liu, C.M., 2014. Hepatoprotective properties of sesamin againstCCl induced oxidative stress-mediated apoptosis in mice via JNK pathway. FoodChem. Toxicol. 64, 41–48.

Martino, M.B., Jones, L., Brighton, B., Ehre, C., Abdulah, L., Davis, C.W., Ron, D., O'Neal,W.K.,Ribeiro, C.M., 2013. The ER stress transducer IRE1beta is required for airway epithelialmucin production. Mucosal Immunol. 6, 639–654.

Mollica, M.P., Lionetti, L., Putti, R., Cavaliere, G., Gaita, M., Barletta, A., 2011. From chronicoverfeeding to hepatic injury: role of endoplasmic reticulum stress and inflammation.Nutr. Metab. Cardiovasc. Dis. 21, 222–230.

Moran-Salvador, E., Titos, E., Rius, B., Gonzalez-Periz, A., Garcia-Alonso, V., Lopez-Vicario,C., Miquel, R., Barak, Y., Arroyo, V., Claria, J., 2013. Cell-specific PPARgamma deficiencyestablishes anti-inflammatory and anti-fibrogenic properties for this nuclear receptorin non-parenchymal liver cells. J. Hepatol. 59, 1045–1053.

Morisseau, C., Hammock, B.D., 2013. Impact of soluble epoxide hydrolase andepoxyeicosanoids on human health. Annu. Rev. Pharmacol. Toxicol. 53, 37–58.

Mu, Y.P., Xi, X.H., Chen, X.R., 2009. Mechanism of hepatocyte regeneration inhibition inrats with liver fibrosis induced by lipogenic methionine-choline- deficient diet.Zhonghua Yi Xue Za Zhi 89, 3430–3436.

Nishitoh, H., Matsuzawa, A., Tobiume, K., Saegusa, K., Takeda, K., Inoue, K., Hori, S.,Kakizuka, A., Ichijo, H., 2002. ASK1 is essential for endoplasmic reticulum stress-induced neuronal cell death triggered by expanded polyglutamine repeats. GenesDev. 16, 1345–1355.

Panigrahy, D., Kalish, B.T., Huang, S., Bielenberg, D.R., Le, H.D., Yang, J., Edin, M.L., Lee, C.R.,Benny, O., Mudge, D.K., Butterfield, C.E., Mammoto, A., Mammoto, T., Inceoglu, B.,Jenkins, R.L., Simpson, M.A., Akino, T., Lih, F.B., Tomer, K.B., Ingber, D.E., Hammock,B.D., Falck, J.R., Manthati, V.L., Kaipainen, A., D'Amore, P.A., Puder, M., Zeldin, D.C.,Kieran, M.W., 2013. Epoxyeicosanoids promote organ and tissue regeneration. Proc.Natl. Acad. Sci. U. S. A. 110, 13528–13533.

Ron, D., Walter, P., 2007. Signal integration in the endoplasmic reticulum unfoldedprotein response. Nat. Rev. Mol. Cell Biol. 8, 519–529.

Rose, T.E., Morisseau, C., Liu, J.Y., Inceoglu, B., Jones, P.D., Sanborn, J.R., Hammock, B.D.,2010. 1-Aryl-3-(1-acylpiperidin-4-yl)urea inhibitors of human and murine solubleepoxide hydrolase: structure-activity relationships, pharmacokinetics, and reductionof inflammatory pain. J. Med. Chem. 53, 7067–7075.

111T.R. Harris et al. / Toxicology and Applied Pharmacology 286 (2015) 102–111

Schroder, M., Kaufman, R.J., 2005. The mammalian unfolded protein response. Annu. Rev.Biochem. 74, 739–789.

Sirish, P., Li, N., Liu, J.Y., Lee, K.S., Hwang, S.H., Qiu, H., Zhao, C., Ma, S.M., Lopez, J.E.,Hammock, B.D., Chiamvimonvat, N., 2013. Unique mechanistic insights into thebeneficial effects of soluble epoxide hydrolase inhibitors in the prevention of cardiacfibrosis. Proc. Natl. Acad. Sci. U. S. A. 110, 5618–5623.

Snoek-van Beurden, P.A., Von den Hoff, J.W., 2005. Zymographic techniques for theanalysis of matrix metalloproteinases and their inhibitors. Biotechniques 38, 73–83.

Song, B.J., Moon, K.H., Olsson, N.U., Salem Jr., N., 2008. Prevention of alcoholic fatty liverand mitochondrial dysfunction in the rat by long-chain polyunsaturated fatty acids.J. Hepatol. 49, 262–273.

Stables, M.J., Gilroy, D.W., 2011. Old and new generation lipid mediators in acute inflam-mation and resolution. Prog. Lipid Res. 50, 35–51.

Tanjore, H., Blackwell, T.S., Lawson, W.E., 2012. Emerging evidence for endoplasmicreticulum stress in the pathogenesis of idiopathic pulmonary fibrosis. Am. J. Physiol.Lung Cell. Mol. Physiol. 302, L721–L729.

Wen, Y., Gu, J., Peng, X., Zhang, G., Nadler, J., 2003. Overexpression of 12-lipoxygenase andcardiac fibroblast hypertrophy. Trends Cardiovasc. Med. 13, 129–136.

Wu, D., Cederbaum, A.I., 2013. Inhibition of autophagy promotes CYP2E1-dependenttoxicity in HepG2 cells via elevated oxidative stress, mitochondria dysfunction andactivation of p38 and JNK MAPK. Redox Biol. 1, 552–565.

Wynn, T.A., Ramalingam, T.R., 2012. Mechanisms of fibrosis: therapeutic translation forfibrotic disease. Nat. Med. 18, 1028–1040.

Yang, L., Jhaveri, R., Huang, J., Qi, Y., Diehl, A.M., 2007. Endoplasmic reticulum stress,hepatocyte CD1d and NKT cell abnormalities in murine fatty livers. Lab. Invest. 87,927–937.

Yang, J., Schmelzer, K., Georgi, K., Hammock, B.D., 2009. Quantitative profiling method foroxylipin metabolome by liquid chromatography electrospray ionization tandemmass spectrometry. Anal. Chem. 81, 8085–8093.

Yang, J., Eiserich, J.P., Cross, C.E., Morrissey, B.M., Hammock, B.D., 2012. Metabolomicprofiling of regulatory lipid mediators in sputum from adult cystic fibrosis patients.Free Radic. Biol. Med. 53, 160–171.

Yu, J., Hui, A.Y., Chu, E.S., Go, M.Y., Cheung, K.F., Wu, C.W., Chan, H.L., Sung, J.J., 2009. Theanti-inflammatory effect of celecoxib does not prevent liver fibrosis in bile duct-ligated rats. Liver Int. 29, 25–36.

Zha, B.S., Zhou, H., 2012. ER stress and lipid metabolism in adipocytes. Biochem. Res. Int.2012, 312943.

Zhang, K., Wang, S., Malhotra, J., Hassler, J.R., Back, S.H., Wang, G., Chang, L., Xu, W., Miao,H., Leonardi, R., Chen, Y.E., Jackowski, S., Kaufman, R.J., 2011a. The unfolded proteinresponse transducer IRE1alpha prevents ER stress-induced hepatic steatosis. EMBOJ. 30, 1357–1375.

Zhang, Y., Dong, L., Yang, X., Shi, H., Zhang, L., 2011b. alpha-Linolenic acid preventsendoplasmic reticulum stress-mediated apoptosis of stearic acid lipotoxicity on primaryrat hepatocytes. Lipids Health Dis. 10, 81.

Zinszner, H., Kuroda, M., Wang, X., Batchvarova, N., Lightfoot, R.T., Remotti, H., Stevens,J.L., Ron, D., 1998. CHOP is implicated in programmed cell death in response toimpaired function of the endoplasmic reticulum. Genes Dev. 12, 982–995.

TABLE S1. Retention time, selected reaction monitoring and internal standard information used to quantify analytes.

Peak Precursor ion

(m/z) Product ion

(m/z) Surrogate Internal

Standards Retention time

(min) 6-keto-PGF1α 369.3 163.2 d4-6-keto-PGF1a 3.54 d4-6-keto-PGF1 α 373.3 167.1 Instrumental standard (IS)* 3.56 Resolvin E1 349.3 195 d4-PGE2 3.58 d4-TXB2 373.3 173.2 IS 4.28 TXB2 369.2 169.1 d4-TXB2 4.29 PGE3 349.3 269.2 d4-PGE2 4.37 PGD3 349.3 269.2 d4-PGE2 4.61 PGF2 α 353.2 309.2 d4-PGE2 4.75 d4-PGE2 355.2 275.3 IS 4.97 PGE2 351.2 271.3 d4-PGE2 4.99 PGD2 351.2 271.3 d4-PGE2 5.29 17,18-DiHETE 335.3 247.2 d11-14,15-DHET 8.34 14,15-DiHETE 335.3 207.2 d11-14,15-DHET 8.84 Instrumental Standard (IS) 339.2 214.3 None 8.86 11,12-DiHETE 335.2 167.1 d11-14,15-DHET 9.01 12,13-DiHOME 313.2 183.2 d11-14,15-DHET 9.18 8,9-DiHETE 335.2 127.1 d11-14,15-DHET 9.31 9,10-DiHOME 313.2 201.2 d11-14,15-DHET 9.54 d11-14,15-DHET 348.2 207.1 IS 10.0 19,20-DiHDPE 361.2 273.2 d11-14,15-DHET 10.12 14,15-DHET 337.2 207.1 d11-14,15-DHET 10.13 16,17-DiHDPE 361.2 233.2 d11-14,15-DHET 10.64 11,12-DHET 337.2 167.1 d11-14,15-DHET 10.79 13,14-DiHDPE 361.2 193.2 d11-14,15-DHET 10.86 10,11-DiHDPE 361.2 153.2 d11-14,15-DHET 11.17 8,9- DHET 337.2 127.1 d11-14,15-DHET 11.31 d6-20-HETE 325.2 281.2 d11-14,15-DHET 11.95 20-HETE 319.2 275.1 d6-20-HETE 12.03 15-HEPE 317.2 219.2 d8-12-HETE 12.05

8-HEPE 317.2 155.2 d8-12-HETE 12.32 12-HEPE 317.2 179.2 d8-12-HETE 12.51 5-HEPE 317.2 115.1 d8-12-HETE 12.92 d4-9-HODE 299.2 172.3 IS 13.13 13-HODE 295.2 195.2 d4-9-HODE 13.17 9-HODE 295.2 171.1 d4-9-HODE 13.24 15-HETE 319.2 219.2 d8-12-HETE 13.82 17(18)-EpETE 317.2 215.2 d11-11(12)-EET 13.99 11-HETE 319.2 167.2 d8-12-HETE 14.32 d8-12-HETE 327.2 184.2 IS 14.56 9-oxo-ODE 293.2 185.1 d4-9-HODE 14.57 15-oxo-ETE 317.2 113.1 d11-11(12)-EET 14.57 14(15)-EpETE 317.2 207.2 d11-11(12)-EET 14.66 8-HETE 319.2 155.2 d8-5-HETE 14.68 12-HETE 319.2 179.2 d8-12-HETE 14.75 11(12)-EpETE 317.2 167.2 d11-11(12)-EET 14.83 8(9)-EpETE 317.2 127.2 d11-11(12)-EET 14.98 9-HETE 319.2 167.2 d8-12-HETE 15.05 15(S)-HETrE 321.2 221.2 d8-12-HETE 15.12 d8-5-HETE 327.2 116.1 IS 15.13 12-oxo-ETE 317.2 153.1 d11-11(12)-EET 15.28 5-HETE 319.2 115.2 d8-5-HETE 15.31 19(20)-EpDPE 343.2 241.2 d11-11(12)-EET 15.98 12(13)-EpOME 295.2 195.2 d4-9(10)-EpOME 16.06 d4-9(10)-EpOME 299.2 172.2 IS 16.16 14(15)-EET 319.2 219.3 d11-11(12)-EET 16.24 9(10)-EpOME 295.2 171.1 d4-9(10)-EpOME 16.26 16(17)-EpDPE 343.2 233.2 d11-11(12)-EET 16.46 13(14)-EpDPE 343.2 193.2 d11-11(12)-EET 16.55 10(11)-EpDPE 343.2 153.2 d11-11(12)-EET 16.66 d11-11(12)-EET 330.2 167.2 IS 16.68 11(12)-EET 319.2 167.2 d11-11(12)-EET 16.81 8(9)-EET 319.2 167.2 d11-11(12)-EET 16.98 *1-Adamantan-1-yl-3-decyl-urea is used as an additional (instrumental) standard. IS was used to quantify the surrogate internal standards and to calculate extraction recovery after SPE extraction of the oxylipins. All the analytes were quantified using the deuterated standards.

TABLE S2. Fold change and P-values from PCR array.

Versus Control Group Versus CCl4 Group

CCl4 Group CCl4 + TPPU

Group TPPU only

Group Control Group

CCl4 + TPPU Group

TPPU only Group

F.C P-

Value F.C P-

Value F.C P-

Value F.C P-

Value F.C P-

Value F.C

P-Valu

e

Actin, alpha 2, smooth muscle, aorta 1.57 0.03 1.64 0.01 1.72 0.49 0.64 0.03 1.05 0.90 1.10 0.49 Angiotensinogen (serpin peptidase inhibitor, clade A, member 8) 0.70 0.03 0.75 0.17 0.84 0.20 1.43 0.03 1.07 0.55 1.20 0.20 Thymoma viral proto-oncogene 1 1.10 0.30 0.78 0.02 1.08 0.85 0.91 0.30 0.71 0.01 0.98 0.85 B-cell leukemia/lymphoma 2 1.66 0.01 1.22 0.13 1.19 0.07 0.60 0.01 0.74 0.04 0.72 0.07 Bone morphogenetic protein 7 1.29 0.31 1.12 0.76 1.68 0.10 0.78 0.31 0.87 0.45 1.30 0.10 Caveolin 1, caveolae protein 0.97 0.86 0.78 0.09 1.44 0.20 1.03 0.86 0.80 0.14 1.48 0.20 Chemokine (C-C motif) ligand 11 0.20 0.07 1.52 0.53 0.63 0.25 4.99 0.07 7.59 0.01 3.13 0.25 Chemokine (C-C motif) ligand 12 1.75 0.52 0.41 0.14 1.96 0.35 0.57 0.52 0.23 0.03 1.12 0.35 Chemokine (C-C motif) ligand 3 3.45 0.00 3.04 0.00 1.28 0.00 0.29 0.00 0.88 0.67 0.37 0.00 Chemokine (C-C motif) receptor 2 5.27 0.00 2.70 0.00 1.22 0.00 0.19 0.00 0.51 0.00 0.23 0.00 CCAAT/enhancer binding protein (C/EBP), beta 0.83 0.46 0.57 0.02 0.97 0.54 1.21 0.46 0.69 0.09 1.17 0.54 Collagen, type I, alpha 2 15.99 0.00 8.69 0.00 1.81 0.00 0.06 0.00 0.54 0.02 0.11 0.00 Collagen, type III, alpha 1 16.16 0.00 7.59 0.00 2.00 0.00 0.06 0.00 0.47 0.04 0.12 0.00 Connective tissue growth factor 1.02 0.90 0.93 0.66 1.72 0.01 0.98 0.90 0.91 0.72 1.68 0.01 Chemokine (C-X-C motif) receptor 4 4.53 0.00 1.95 0.03 1.35 0.00 0.22 0.00 0.43 0.00 0.30 0.00 Decorin 2.05 0.00 1.78 0.00 1.61 0.01 0.49 0.00 0.87 0.24 0.79 0.01 Endothelin 1 1.98 0.00 1.73 0.02 1.94 0.96 0.50 0.00 0.87 0.46 0.98 0.96 Epidermal growth factor 0.88 0.50 1.20 0.31 2.21 0.02 1.13 0.50 1.36 0.11 2.51 0.02 Endoglin 1.24 0.10 1.20 0.16 1.33 0.56 0.81 0.10 0.97 0.80 1.08 0.56 Fas ligand (TNF superfamily, member 6) 1.13 0.95 0.89 0.83 1.42 0.54 0.89 0.95 0.78 0.78 1.25 0.54 Gremlin 1 0.44 0.25 4.33 0.29 0.16 0.20 2.26 0.25 9.78 0.01 0.36 0.20 Hepatocyte growth factor 1.66 0.00 1.20 0.45 1.26 0.04 0.60 0.00 0.72 0.00 0.76 0.04 Interferon gamma 0.56 0.19 0.64 0.30 1.84 0.05 1.79 0.19 1.14 0.72 3.29 0.05 Interleukin 10 1.43 0.12 0.76 0.63 1.15 0.88 0.70 0.12 0.53 0.09 0.80 0.88 Interleukin 13 1.14 0.55 0.74 0.40 1.86 0.29 0.88 0.55 0.65 0.23 1.63 0.29 Interleukin 13 receptor, alpha 2 1.74 0.08 1.51 0.28 0.45 0.00 0.57 0.08 0.87 0.56 0.26 0.00 Interleukin 1 alpha 1.39 0.07 1.04 0.99 1.17 0.30 0.72 0.07 0.75 0.02 0.84 0.30 Interleukin 1 beta 0.91 0.52 0.31 0.03 0.19 0.02 1.10 0.52 0.34 0.00 0.21 0.02 Interleukin 4 0.37 0.12 0.95 0.74 0.53 0.63 2.68 0.12 2.54 0.14 1.42 0.63

Interleukin 5 0.36 0.07 1.31 0.34 0.47 0.38 2.79 0.07 3.66 0.01 1.32 0.38 Integrin linked kinase 1.11 0.20 1.01 0.86 1.30 0.10 0.90 0.20 0.91 0.34 1.17 0.10 Inhibin beta E 1.50 0.07 1.18 0.52 1.01 0.36 0.67 0.07 0.79 0.27 0.68 0.36 Integrin alpha 1 1.45 0.00 1.06 0.44 1.18 0.04 0.69 0.00 0.73 0.00 0.81 0.04 Integrin alpha 2 3.08 0.01 1.73 0.48 1.58 0.05 0.32 0.01 0.56 0.03 0.51 0.05 Integrin alpha 3 0.91 0.32 0.99 0.82 0.88 0.69 1.10 0.32 1.10 0.28 0.97 0.69 Integrin alpha V 1.40 0.00 0.90 0.24 1.01 0.00 0.71 0.00 0.64 0.00 0.72 0.00 Integrin beta 1 (fibronectin receptor beta) 1.14 0.06 0.94 0.41 1.15 0.85 0.88 0.06 0.82 0.01 1.01 0.85 Integrin beta 3 0.88 0.41 0.77 0.31 0.70 0.05 1.14 0.41 0.87 0.19 0.79 0.05 Integrin beta 5 1.77 0.00 1.26 0.02 1.24 0.00 0.57 0.00 0.71 0.01 0.70 0.00 Integrin beta 6 0.29 0.11 2.28 0.93 0.37 0.70 3.47 0.11 7.90 0.01 1.27 0.70 Integrin beta 8 1.59 0.02 1.06 0.81 1.17 0.63 0.63 0.02 0.67 0.04 0.74 0.63 Jun oncogene 2.06 0.01 1.06 0.81 0.23 0.00 0.49 0.01 0.51 0.00 0.11 0.00 Lysyl oxidase 4.37 0.00 3.34 0.00 1.18 0.00 0.23 0.00 0.76 0.11 0.27 0.00 Latent transforming growth factor beta binding protein 1 2.34 0.00 1.92 0.03 1.18 0.00 0.43 0.00 0.82 0.12 0.50 0.00 Matrix metallopeptidase 13 10.85 0.00 5.20 0.00 0.72 0.00 0.09 0.00 0.48 0.00 0.07 0.00 Matrix metallopeptidase 14 (membrane-inserted) 1.63 0.02 0.90 0.42 0.89 0.00 0.61 0.02 0.55 0.00 0.55 0.00 Matrix metallopeptidase 1a (interstitial collagenase) 2.48 0.00 1.82 0.01 1.49 0.32 0.40 0.00 0.73 0.16 0.60 0.32 Matrix metallopeptidase 2 13.06 0.00 7.44 0.00 1.88 0.00 0.08 0.00 0.57 0.04 0.14 0.00 Matrix metallopeptidase 3 0.89 0.75 1.13 0.96 2.44 0.17 1.12 0.75 1.27 0.70 2.75 0.17 Matrix metallopeptidase 8 5.60 0.00 1.23 0.97 1.04 0.00 0.18 0.00 0.22 0.00 0.19 0.00 Matrix metallopeptidase 9 1.58 0.51 0.42 0.11 0.52 0.00 0.63 0.51 0.27 0.00 0.33 0.00 Myelocytomatosis oncogene 1.59 0.18 1.41 0.30 0.24 0.57 0.63 0.18 0.89 1.00 0.15 0.57 Nuclear factor of kappa light polypeptide gene enhancer in B-cells 1, p105 1.21 0.07 1.02 0.77 1.14 0.50 0.83 0.07 0.85 0.20 0.94 0.50 Platelet derived growth factor, alpha 1.64 0.00 1.12 0.23 1.48 0.42 0.61 0.00 0.68 0.03 0.90 0.42 Platelet derived growth factor, B polypeptide 1.41 0.08 0.83 0.22 1.52 0.60 0.71 0.08 0.59 0.00 1.07 0.60 Plasminogen activator, tissue 8.37 0.00 6.32 0.00 1.56 0.00 0.12 0.00 0.75 0.19 0.19 0.00 Plasminogen activator, urokinase 2.80 0.00 1.98 0.04 0.59 0.11 0.36 0.00 0.70 0.14 0.21 0.11 Plasminogen 0.75 0.03 0.76 0.04 0.61 0.46 1.33 0.03 1.01 0.96 0.81 0.46 Serine (or cysteine) peptidase inhibitor, clade A, member 1a 0.72 0.00 0.85 0.01 0.93 0.00 1.38 0.00 1.18 0.00 1.28 0.00 Serine (or cysteine) peptidase inhibitor, clade E, member 1 2.06 0.96 1.52 0.53 3.74 0.03 0.49 0.96 0.74 0.19 1.82 0.03 Serine (or cysteine) peptidase inhibitor, clade H, member 1 2.84 0.00 2.04 0.00 1.72 0.00 0.35 0.00 0.72 0.03 0.61 0.00

MAD homolog 2 (Drosophila) 1.24 0.03 0.99 0.94 1.47 0.04 0.81 0.03 0.80 0.03 1.19 0.04 MAD homolog 3 (Drosophila) 1.40 0.01 0.96 0.71 1.57 0.38 0.72 0.01 0.68 0.02 1.13 0.38 MAD homolog 4 (Drosophila) 1.03 0.65 1.03 0.70 1.11 0.47 0.97 0.65 1.00 0.89 1.07 0.47 MAD homolog 6 (Drosophila) 1.48 0.01 1.12 0.30 1.68 0.43 0.67 0.01 0.76 0.08 1.13 0.43 MAD homolog 7 (Drosophila) 1.69 0.01 1.36 0.20 1.88 0.38 0.59 0.01 0.81 0.06 1.11 0.38 Snail homolog 1 (Drosophila) 0.24 0.07 0.99 0.98 0.21 0.70 4.11 0.07 4.08 0.02 0.86 0.70 Trans-acting transcription factor 1 1.15 0.07 0.91 0.32 1.32 0.08 0.87 0.07 0.80 0.03 1.15 0.08 Signal transducer and activator of transcription 1 1.16 0.34 0.89 0.32 1.39 0.16 0.86 0.34 0.77 0.03 1.20 0.16 Signal transducer and activator of transcription 6 0.94 0.49 0.89 0.34 1.34 0.00 1.06 0.49 0.95 0.65 1.43 0.00 Transforming growth factor, beta 1 1.81 0.00 1.10 0.64 1.18 0.00 0.55 0.00 0.61 0.00 0.65 0.00 Transforming growth factor, beta 2 2.77 0.00 2.11 0.00 1.76 0.11 0.36 0.00 0.76 0.12 0.63 0.11 Transforming growth factor, beta 3 3.32 0.00 2.76 0.00 1.86 0.09 0.30 0.00 0.83 0.29 0.56 0.09 Transforming growth factor, beta receptor I 1.32 0.00 0.92 0.37 0.99 0.00 0.76 0.00 0.70 0.00 0.75 0.00 Transforming growth factor, beta receptor II 1.87 0.00 1.41 0.02 1.30 0.02 0.53 0.00 0.75 0.02 0.69 0.02 TGFB-induced factor homeobox 1 1.27 0.14 1.36 0.03 1.23 0.83 0.79 0.14 1.07 0.68 0.97 0.83 Thrombospondin 1 2.09 0.00 1.55 0.09 0.71 0.00 0.48 0.00 0.74 0.03 0.34 0.00 Thrombospondin 2 5.83 0.00 3.74 0.00 1.93 0.00 0.17 0.00 0.64 0.01 0.33 0.00 Tissue inhibitor of metalloproteinase 1 0.48 0.14 1.34 0.87 0.21 0.27 2.10 0.14 2.83 0.10 0.43 0.27 Tissue inhibitor of metalloproteinase 2 3.52 0.00 2.47 0.00 1.60 0.00 0.28 0.00 0.70 0.01 0.46 0.00 Tissue inhibitor of metalloproteinase 3 1.62 0.08 1.38 0.35 1.72 0.62 0.62 0.08 0.85 0.13 1.06 0.62 Tissue inhibitor of metalloproteinase 4 1.07 0.85 1.46 0.63 1.23 0.35 0.93 0.85 1.36 0.47 1.15 0.35 Tumor necrosis factor 1.37 0.10 0.64 0.05 1.50 0.57 0.73 0.10 0.46 0.01 1.10 0.57 Vascular endothelial growth factor A 1.20 0.33 1.00 0.91 1.39 0.37 0.83 0.33 0.83 0.26 1.16 0.37

TABLE S3. Validation of PCR array by RT-PCR.

*Values are mean of fold change ± standard error.

COL1A2 MMP2 TNF RTPCR CCl4 only 24.4 ± 3.7* 13.8 ± 2.3 1.3 ± 0.2 CCl4 + TPPU 11.9 ± 2.1 7.7 ± 0.8 0.7 ± 0.1 TPPU only 1.7 ± 0.4 1.9 ± 0.5 1.2 ± 0.2 PCR ARRAY CCl4 only 16.0 13.1 1.4 CCl4 + TPPU 8.7 7.4 0.6 TPPU only 1.8 1.9 1.5

TABLE S4. Plasma Oxylipin levels.

Metabolite Control (nM)

(n=5) CCl4  (nM)(n=7) 

CCl4 + TPPU (nM)(n=6) 

TPPU (nM)(n=6) 

12,13‐DiHOME   24.5 ± 4.8  14.3 ± 3.0  27.3 ± 3.6  26.7 ± 1.6 9,10‐DiHOME   7.7 ± 1.0  5.4 ± 1.0  8.4 ± 0.9  8.1 ± 0.4 14,15‐DiHETrE   0.7 ± 0.1  0.2 ± 0.0  0.4 ± 0.1  0.5 ± 0.0 11,12‐DiHETrE  0.3 ± 0.0  0.2 ± 0.0  0.4 ± 0.0  0.3 ± 0.0 8,9‐DiHETrE  0.5 ± 0.1  0.3 ± 0.0  0.5 ± 0.0  0.6 ± 0.0 5,6‐DiHETrE   0.4 ± 0.1  0.1 ± 0.0  0.3 ± 0.0  0.5 ± 0.0 9(10)‐EpOME  18.4 ± 3.0  30.1 ± 14.6  29.6 ± 3.4  24.1 ± 1.5 12(13)‐EpOME  23.8± 3.3  81.9 ± 17.1  97.6 ± 5.7  48.8 ± 2.5 14(15)‐EpETrE   2.3 ± 0.5  3.2 ± 0.7  4.1 ± 0.4  3.3 ± 0.2 11(12)‐EpETrE   1.9 ± 0.4  4.7 ± 0.8  4.4 ± 0.4  3.5 ± 0.2 8(9)‐EpETrE  0.7 ± 0.2  1.4 ± 0.4  1.7 ± 0.2  1.4 ± 0.1 5(6)‐EpETrE   4.5 ± 1.0  7.0 ± 2.4  9.5 ± 1.1  8.3 ± 0.5 PGE2  0.2 ± 0.0  0.1 ± 0.0  0.3 ± 0.1  0.3 ± 0.0 PGD2  0.4 ± 0.1  0.3 ± 0.0  0.6 ± 0.1  0.5 ± 0.0 LTB4  0.1 ± 0.0  0.1 ± 0.0  0.1 ± 0.0  0.1 ± 0.0 TXB2  1.4 ± 0.4  1.2 ± 0.3  3.1 ± 1.5  1.3 ± 0.7 6‐keto‐PGF1a  4.5 ± 0.9  1.8 ± 1.4  4.6 ± 0.6  3.5 ± 0.3 PGF2a  3.5 ± 2.2  0.8 ± 0.4  0.7 ± 0.2  0.7 ± 0.1 5‐HETE  1.2 ± 0.1  1.9 ± 0.2  2.8 ± 0.4  2.8 ± 0.2 8‐HETE  0.7 ± 0.0  0.6 ± 0.1  1.0 ± 0.2  1.5 ± 0.1 11‐HETE  0.0 ± 0.0  0.0 ± 0.0  0.9 ± 0.6  0.2 ± 0.3 12‐HETE  27.2 ± 5.0  47.0 ± 26.1  70.2 ± 30.0  91.1 ± 13.4 15‐HETE  1.3 ± 0.2  1.0 ± 0.5  2.3 ± 0.9  2.1 ± 0.4 9,12,13‐TriHOME  3.2 ± 1.5  2.1 ± 0.7  2.8 ± 0.3  2.5 ± 0.1 

Values are Data are mean ± standard error.

9,10,13‐TriHOME  0.9 ± 0.2  0.5 ± 0.2  0.7 ± 0.1  0.6 ± 0.1 LXA4  0.0 ± 0.0  0.0 ± 0.0  0.0 ± 0.0  0.0 ± 0.0 15‐Deoxy‐PGJ2  0.3 ± 0.1  0.1 ± 0.1  0.2 ± 0.0  0.2 ± 0.0 13‐HODE  29.8 ± 6.2  56.3 ± 11.3  46.0 ± 7.5  53.2 ± 3.4 9‐HODE  11.7 ± 1.8  17.7 ± 4.4  15.2 ± 2.9  18.2 ± 1.3 13‐oxo‐ODE  1.5 ± 0.3  2.7 ± 0.4  3.2 ± 0.2  2.7 ± 0.1 15‐oxo‐ETE  0.3 ± 0.1  0.5 ± 0.2  0.8 ± 0.1  0.4 ± 0.1 9‐oxo‐ODE  12.1 ± 2.5  27.1 ± 8.8  27.1 ± 3.3  20.5 ± 1.5 20‐HETE  0.1 ± 0.0  0.1 ± 0.0  0.1 ± 0.0  0.1 ± 0.0 14,15 EET/DHET  3.6 ± 1.0  12.7 ± 0.9  12.4 ± 0.4  8.4 ± 0.2 11,12‐EET/DHET  6.6 ± 1.7  23.6 ± 1.5  13.8 ± 0.6  11.6 ± 0.3 8,9‐EET/DHET  1.5 ± 0.3  4.6 ± 0.5  3.4 ± 0.2  2.9 ± 0.1 5,6‐EET/DHET  11.5 ± 2.5  47.8 ± 3.5  37.7 ± 2.2  23.4 ± 1.0 sum EET  9.5 ± 2.0  16.2 ± 4.1  19.6 ± 2.0  16.4 ± 0.9 sum DHET  1.9 ± 0.2  0.9 ± 0.1  1.5 ± 0.2  1.9 ± 0.1 sum EET/DHET  5.1 ± 1.1  18.1 ± 1.6  13.9 ± 0.7  10.6 ± 0.3 sum EpOME  42.2 ± 6.0  112.0 ± 31.7  127.2 ± 8.8  72.9 ± 3.9 sum DiHOME  32.2 ± 5.8  19.7 ± 4.0  35.7 ± 4.5  34.8 ± 2.0 sum EpOME/DiHOME  1.3 ± 0.1  5.7 ± 0.5  6.2 ± 0.2  2.5 ± 0.1 

22

23

24

25

26

27

28

29

30

31

32

0 5 10 15 20 25 30 33

Control

CCl4 only

CCl4+TPPU

TPPU only

Days post injection

Body

 weight (g)

Figure S1. Weight gain continued through CCl4 treatment. Mice were treated with CCl4 and given TPPU in driven water on day 0 as described on Materials and Methods.  

Figure S1

Control CCl4

TPPU

Control CCl4 CCl4 + TPPU TPPU

% A

rea

Stai

ned

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

TPPU + CCl4

A

B

* *

*

Figure S2. TPPU does not modulate CCl4-induced alpha smooth muscle actin (αSMA) expression in the liver. (A) Representative liver sections stained using αSMA antibodies. (B) Quantification of staining expressed in percent area. Error bars represent standard deviation. *P-value vs. Control group≤ 0.05. N=6-8 animals per group.

Figure S2

98

64

50

36

TPPU

CCl 4+

TPPU

CCl 4

Control

TPPU

CCl 4+

TPPU

CCl 4

Control

} pro‐MMP9} active MMP 9} pro‐MMP2} active MMP2} pro‐MMP1a} active MMP1a

Expected Migration andOrder of MMPs

Figure S3. Expected migration of metalloproteases (MMPs.) MMP migration in native PAGE is altered by glycosylation and will vary depending on context (disease state and tissue.)

Figure S3

0.00

20.00

40.00

60.00

80.00

100.00

120.00

140.00

160.00

180.00

200.00

Control

CCl4

sEHI

Plasma Co

ncentration (nM)

Figure S4. Modulation of oxylipin levels in the plasma of CCl4 treated mice did not reach statistical significance. Blood was collected by heart puncture and the plasma analyzed by LC‐MS/MS as described in Materials and Methods. The differences between groups did not reach statistical significance (P > 0.05.) N=6‐8 animals per group.

Figure S4

0.00

5.00

10.00

15.00

20.00

25.00

14,15 EET/DHET 11,12‐EET/DHET 8,9‐EET/DHET 5,6‐EET/DHET sum EET/DHET

Control

CCl4

sEHI

Ratio

 of P

lasm

a Co

ncen

trations

Figure S5. Modulation of oxylipin levels in the plasma of CCl4 treated mice did not reach statistical significance. Blood was collected by heart puncture and the plasma analyzed by LC‐MS/MS as described in Materials and Methods. The differences between groups did not reach statistical significance (P > 0.05.) N=6‐8 animals per group.

Figure S5