Pathological mechanisms in NAFLD


Maternal obesity promotes offspring non-alcoholic fatty liver disease (NAFLD) through disruption of endoplasmic reticulum homeostasis

Junpei Soeda1, Angelina Mouralidarane1, Esra Asilmaz1, Shuvra Ray1, Joaquim Pombo2, Lucilla Poston2, Paul D. Taylor2, Jude A. Oben1,3

1Institute for Liver and Digestive Health, University College London, London, United Kingdom; 2Division of Women's Healh, King's College London, London, United Kingdom; 3Gastroenterology and Hepatology, Guy's and St Thomas' Hospital, London, United Kingdom

We previously showed that maternal obesity (MO) programs offspring obesity and non-alcoholic fatty liver disease (NAFLD) with involved mechanisms unclear. Accumulating evidence suggests that endoplasmic reticulum (ER) stress induced unfolded protein response (UPR) plays a central role in the pathogenesis of steatosis and non-alcoholic steatohepatitis (NASH). It has recently been shown that one of the UPR pathways (IRE1α) follows a 12 hour period rhythmic activation in normal liver but demonstrates constant activation in obese, leptin deficient, ob/ob mice. However, little is known about the role of UPR in developmentally programmed NAFLD. AIMS & METHODS: C57BL6 mice were fed standard chow (SC) or an obesogenic diet (OD) for 6 weeks prior to pregnancy, throughout pregnancy and lactation. Litters were weaned onto standard or an OD to generate 4 groups. Animals were sacrificed at 4-hourly intervals over a 12: 12hr light- dark cycle periods at 6 months. We initially studied UPR pathway at one specific time point and then further characterised rhythmic expression of specific UPR markers at all time points. RESULTS: Offspring exposed to MO and a post-weaning OD (OffOb-OD) developed profound NAFLD compared to those exposed to post-partum OD alone (OffCon-OD) or the control group (OffCon-SC), as assessed by raised ALT (p<0.001) and NAFLD Activity Score (p<0.01). At a single time point, phospho eIF-2alpha was specifically increased in Offob-OD (p<0.05) compared to OffCon-SC. ATF6 cleavage and the spliced form of XBP-1 were most abundantly expressed in Offob-OD. Also, Phopho SAPK/JNK, and Lc3BII protein expression were significantly increased in Offob-OD compared to OffCon-SC. In parallel CHOP expression was significantly higher in OffOb-OD compared to OffCon-Sc and Offob-OD. Furthermore, hepatocyte apoptosis was detected in Offob-OD. These results indicate that unresolved UPR is significantly activated in OffCon-OD. However, GRP78, a major ER chaperone and central regulator for ER stress, was significantly downregulated in Offob-OD. UPR induced chaperone (GRP94) and ER-associated protein degradation (ERAD) related genes (HERP and EDEM) were downregulated in OffCon-OD and Offob-OD. Further analysis at all time points showed that all 3 proximal sensors of UPR were continuously activated in Offob-OD while 12h rhythmic expression of GRP78 was blunted in Offob-OD. Finally, UPR downstream ERAD genes showed either a 12h or 24h rhythmic expression which was attenuated in Offob-OD. CONCLUSION: MO and a post-natal OD profoundly disrupted ER homeostasis in offspring. We propose that disrupted ER homeostasis may be involved in the propagation of programmed of NAFLD.


The following people have nothing to disclose: Junpei Soeda, Angelina Mouralidarane, Esra Asilmaz, Shuvra Ray, Joaquim Pombo, Lucilla Poston, Paul D. Taylor, Jude A. Oben


Hepatic Gap Junction Inhibition Limits Liver Injury and Inflammation in Non-alcoholic Steatohepatitis (NASH)

Suraj J. Patel1,3, Jay Luther2, Kevin R. King1, Stefan Bohr1, 3, John Scichilone1,3, John Garber2, Lee F. Peng2, Raymond T. Chung2, Martin L. Yarmush1,3

1Surgery, Massachusetts General Hospital, Boston, MA; 2Gastroenterology, Mossochusetts General Hospital, Boston, MA; 3Research, Shriners Burns Hospital, Boston, MA

Non-alcoholic fatty liver disease (NAFLD) is the leading cause of chronic liver disease in the US. While the simple steatosis seen in NAFLD does not correlate with increased morbidity or mortality, progression of this condition to non-alcoholic steatohepatitis (NASH) dramatically increases the risk of cirrhosis, liver failure, and hepatocellular carcinoma. However, treatment options are limited due to an incomplete understanding of the inflammatory mechanisms underlying the development of NASH. We have previously shown that inhibition of a key hepatic gap junction protein, Connexin 32 (Cx32), protects against acute liver injury by limiting oxidative stress and the associated inflammatory response (Patel et al., Nat Biotechnol 2012). In this study, we investigated the role of hepatic gap junction communication in modulating inflammation in NASH. We report that Cx32 is an important mediator of NASH by showing that mice deficient in Cx32 exhibit limited liver injury and inflammation in response to the classic methionine choline deficient (MCD) diet for inducing NASH. Compared to wildtype mice, Cx32 deficient mice on the MCD diet had 2.5- to 3-fold lower serum ALT/AST levels and reduced histological evidence of hepatocyte ballooning and lobular inflammation, as represented by a significantly lower NAFLD activity score. Furthermore, we demonstrated that Cx32 deficient mice on the MCD diet have significantly reduced hepatic expression of inflammatory cytokines, such as TNFα and IL-6.These cytokines are known to increase intestinal permeability, and have been recently implicated in the pathogenesis of NASH (Henao-Mejia et al., Nature 2012). We found that Cx32 deficient mice on the MCD diet had significantly lower portal serum levels of LPS and 4 kDa FITC-dextran (orally gavaged) compared to wildtype mice, suggesting reduced intestinal microbial translocation and paracellular permeability, respectively. Immunohistochemistry staining for intestinal tight junction proteins also revealed decreased tight junction disruption in the Cx32 deficient mice compared to wild-type. Lastly, we identified a selective small molecule inhibitor of Cx32 that limits liver injury and inflammation in NASH when administered during the MCD diet. Together these findings reveal that hepatic gap junction communication plays a significant role in establishing NASH, and that inhibiting Cx32, either genetically or pharmacologically, reduces liver injury, inflammation, and downstream intestinal permeability in NASH. As such, our findings suggest a potentially promising pathway upon which to build an experimental therapy for limiting NASH.


Suraj J. Patel - Stock Shareholder: Heprotech Kevin R. King - Patent Held/Filed: Heprotech Inc

Raymond T. Chung - Advisory Committees or Review Panels: Idenix; Consulting: Enanta; Grant/Research Support: Gilead, Merck, Mass Biologic, Gilead

The following people have nothing to disclose: Jay Luther, Stefan Bohr, John Scichilone, John Garber, Lee F. Peng, Martin L. Yarmush


A lipotoxic mechanism of NASH: Free cholesterol causes hepatocyte apoptosis and necrosis by JNK-mediated mitochondrial injury, and resultant release of hepatocyte-derived HMGB1 and microparticles activate Kupffer cells

Lay Gan1, Derrick M. Van Rooyen1, Mark Koina2, Robert S. McCuskey3, Narci Teoh1, Geoffrey C. Farrell1

1Liver Research Group, Australian National University at The Canberra Hospital, Woden, ACT, Australia; 2Anatomical Pathology, The Canberra Hospital, Woden, ACT, Australia; 3Cellular and Molecular Medicine, University of Arizona, Tucson, AZ

Free cholesterol (FC) accumulates in livers of non-alcoholic steatohepatitis (NASH) in humans and mice with obesity, diabetes and metabolic syndrome. Cholesterol-loaded livers are sensitized to cytokine-mediated mitochondrial injury, but no direct evidence links FC lipotoxicity to hepatocyte cell death. We loaded primary murine hepatocytes with FC to characterise the mechanisms of resultant apoptosis and necrosis, and then test the hypothesis that c-Jun N-terminal kinase (JNK) activation and mitochondrial injury are essential steps in FC hepatocellular lipotoxicity. Further, we explored how FC-induced hepatocyte injury could promote Kupffer cell (KC) activation. Methods: We determined subcellular site of hepatocyte FC in NASH livers by co-localising filipin fluorescence with organelle markers. Primary hepatocytes (C57B6 wild type [WT] or JNK1-/-) were incubated with LDL (0-40μM) to load with FC. Pathways of FC-mediated cell death were determined by western blot, immunofluorescence and pathway-specific inhibitors. Separately, supernatants from FC-injured hepatocytes were used to assay high mobility group box 1 (HMGB1) and microparticles (MPs). Supernatant or MPs were added to KC cultures. Ultrastructure was assessed by electron microscopy (EM). Results: In NASH livers, FC co-localised to plasma membrane (PM), mitochondria and endoplasmic reticulum (ER). This pattern was replicated in hepatocytes incubated with LDL to dose-dependent increase hepatocyte FC. FC loading caused dose-dependent LDH leakage, apoptosis and necrosis with release of HMGB1.At 40μM LDL, cell death associated with JNK1 activation, mitochondrial membrane pore transition resulting in cyt c release into cytoplasm, cellular oxidative stress (increased GSSG) and ATP depletion. JNK inhibition (CC-401, CC-930) ameliorated apoptosis and necrosis, while JNK–1–/hepatocytes were refractory to FC-induced injury. Cyclosporine A and caspase-3 inhibition abrogated FC-mediated hepatocellular cell death, but 4-phenylbutyric acid did not; there was no increase of ER stress proteins (GRP78, CHOP) in vitro or in vivo. FC deposition in PM reduced fluidity to cause surface blebbing and release of MPs, evident on EM. Addition of HMGB1-enriched culture medium or MPs from FC-loaded hepatocytes activated KCs, assessed by increased nuclear NF-kB (p65), release of IL-1β, TNF-α and ultrastructural changes. Conclusions: These findings demonstrate that FC deposition in mitochondria and PM causes hepatocyte cell death, confirm JNK-1 activation is important for hepatocyte lipotoxic injury, revealing links between HMGB1 and MPs with lipotoxicity and engagement of KC activation in the transition of steatosis to NASH.


The following people have nothing to disclose: Lay Gan, Derrick M. Van Rooyen, Mark Koina, Robert S. McCuskey, Narci Teoh, Geoffrey C. Farrell


Degradation of Keap1 activates BH3-only proteins Bim and PUMA during hepatocyte lipoapoptosis

Sophie C. Cazanave1, Xuan Wang2, Huiping Zhou2, Curtis D. Klaassen3, Arun J. Sanyal1

1Division of Gastroenterology, Hepatology and Nutrition, Department of Internal Medicine, Virginia Commonwealth University, Richmond, VA; 2Department of Microbiology and Immunology, Virginia Commonwealth University, Richmond, VA; 3Department of Pharmacology, Toxicology and Therapeutics, University of Kansas Medical Center, Kansas City, KS

Background: Non-alcoholic steatohepatitis (NASH) is characterized by hepatic steatosis, elevated levels of circulating free fatty acids (FFA) and hepatocyte lipoapotosis. This lipoapoptosis requires activation of the pro-apoptotic BH3-only proteins Bim and PUMA. Keap1 is a BTB-kelch protein that can regulate the expression of Bcl-2 protein and control apoptotic cell death. Yet, a role for Keap1 in mediating hepatocyte lipotoxicity is unknown. In vivo, keap1 deletion worsened insulin resistance and increased hepatocyte injury in diet-induced and genetic obesity, suggesting a protective role of Keap1 regarding these parameters. Thus, our aim was to determine if Keap1 was dysregulated during lipotoxicity by FFA. Methods: Hepatocarcinoma cell lines Hep3B and Huh-7, or mouse primary hepatocytes were treated with saturated FFA palmitate (PA) (400-600 microM). Keap1, PUMA, Bim expression and JNK activation were examined by real-time PCR and/or immunoblot analysis. Keap1 expression was selectively knocked-down using shRNA. Cell death was assessed by trypan blue exclusion assay, DAPI staining and caspase 3/7 activation using a fluorogenic assay. Results: PA is toxic to liver cells and induces significant cell death by 8h and 16h after treatment. Interestingly, Keap1 protein underwent rapid cellular elimination within 2 to 4 hours after treatment with PA. PA-induced decrease in Keap1 protein was associated with JNK activation and upregulation of Bim and PUMA protein levels. In contrast, no alteration in Keap1 expression was noted following incubation with oleic acid, a non-toxic FFA. PA did not alter Keap1 mRNA expression, excluding a transcriptional regulation of Keap1 during this process. Keap1 degradation was not affected by either proteasome inhibition with MG132, or by pan-caspase inhibition with QVD-OPh. In contrast, disruption of the autophagy pathway, by silencing of the autophagy-related protein p62, prevented Keap1 decrease by PA, indicating that PA-induced decrease in Keap1 is due to autophagy degradation. Stable knockdown of Keap1 expression in Hep3B or Huh-7 cells resulted in increased JNK phosphorylation and downstream upregulation of Bim and PUMA protein expression with subsequent increased cell death. Keap1 knockdown also significantly enhanced PA-mediated cell death and caspase 3/7 activity. Finally, primary hepatocytes isolated from liver-specific keap/- mice, which express higher Bim and PUMA protein levels, displayed increased sensitivity to PA-induced apoptosis than WT mouse hepatocyte. Conclusion: These results implicate p62dependent autophagic degradation of Keap1 by palmitate as a mechanism promoting hepatocyte lipoapoptosis.


Arun J. Sanyal - Advisory Committees or Review Panels: Gore, Gilead, Abbott, Ikaria; Consulting: Salix, Immuron, Exhalenz, Bayer-Onyx, Genentech, Norgine, GalMed, Novartis, Echosens, Takeda; Grant/Research Support: Salix, Genentech, Genfit, Intercept, Ikaria, Takeda, Gilead; Independent Contractor: UpToDate

The following people have nothing to disclose: Sophie C. Cazanave, Xuan Wang, Huiping Zhou, Curtis D. Klaassen


Role of the IL-17 axis in the development and progression of non-alcoholic fatty liver disease

Daniel Giles1, Traci Stankiewicz1, Isaac T. Harley1, Monica Cappeiletti1, Samir Softic2, Stavra A. Xanthakos2, Rohit Kohli2, Christopher L. Karp1, Senad Divanovic1

1Cellular and Molecular Immunology, Cincinnati Children's Hospital Medical Center, Cincinnati, OH; 2Gastroenterology, Hepatology, & Nutrition, cincinnati Children's Hospital Medical Center, Cincinnati, OH

Obesity is a primary risk factor for the development of non-alcoholic fatty liver disease (NAFLD), a spectrum of disorders ranging from steatosis to steatohepatitis to cirrhosis. NAFLD has become the most common chronic liver disease in the developed world. The twinned observations that obesity is associated with increased activation of the IL-17 axis and that this axis can regulate liver damage in diverse contexts prompted us to address the role of IL-17RA signaling in the progression of NAFLD. IL-17RA-deficient mice were subjected to obesogenic diet stress (a standard model of diet-induced obesity and NAFLD) or a regular diet as a control. Development of obesity, pro-inflammatory cytokine production, glucose dysmetabolism, hepatic triglyceride accumulation and inflammation and hepatocellular damage were analyzed. Additionally, by colonizing or depleting an intestinal commensal, known to drive IL-17 production, we examined the role of intestinal microbe-driven IL-17 induction in progression of NAFLD in WT and Leptin receptor mutant mice (Leprdb/db). Notably, our data indicate that IL-17RA-/- mice respond to obesogenic diet stress with significantly greater weight gain, visceral adiposity, and hepatic steatosis than wild type controls. However, obesity-driven lipid accumulation was uncoupled from its end organ consequences in IL-17RA-/- mice, which exhibited decreased steatohepatitis, NADPH-oxidase enzyme expression and hepatocellular damage and were protected from glucose dysmetabolism. Further, antibody-mediated neutralization of IL-17A significantly reduced obesity associated hepatocellular damage in wild type mice. Lastly, colonization of mice with segmented filamentous bacteria (SFB), a commensal that induces Th17 differentiation, elevated systemic IL-17A production and exacerbated obesityinduced hepatocellular damage. Similarly, selective (though not specific) SFB depletion suppressed IL-17A production and protected from obesity-induced hepatocellular damage. These data indicate that obesity-driven activation of the IL-17 axis is central to the development and progression of NAFLD and identify the IL-17 pathway as a novel therapeutic target in NAFLD. Ongoing studies aim at defining the biologically-relevant IL-17RA expressing cell types, IL-17RA ligands and molecular pathways central to the IL-17 axis mediated progression of NASH.


Rohit Kohli - Grant/Research Support: Johnson and Johnson, Johnson and Johnson

The following people have nothing to disclose: Daniel Giles, Traci Stankiewicz, Isaac T. Harley, Monica Cappelletti, Samir Softic, Stavra A. Xanthakos, Christopher L. Karp, Senad Divanovic


Short Heterodimer Partner (SHP) is necessary for the Improvement in NASH after Sleeve Gastrectomy in Obese Mice

Andriy Myronovych1, Rosa-Maria Salazar-Gonzalez1, Lili Miles1, Karen K. Ryan2, Randy J. Seeley2, Rohit Kohli1

1Department of Gastroenteroloqy/ Cincinnati Children's Hospital Medical Center, Mason, OH; 2Internal Medicine, University of Cincinnati, Cincinnati, OH

Elevated serum bile acids suppress bile acid synthesis and lipogenesis after vertical sleeve gastrectomy (VSG) is performed in obese mice. Bile acids bind to the transcription factor farnesoid X receptor (FXR) that induces short heterodimer partner (SHP) leading to downregulation of these hepatic bile acid synthesis and lipogenic genes. We hypothesized that SHP activation is necessary for the bile acid induced improvement of hepatic steatosis after VSG. Methods: To induce obesity adult male SHP knockout (SHP KO)mice and their wild type littermates were fed a high saturated fat diet (HFD, 60kcal%) for 8 weeks. Mice were randomized into four groups (n=4-8) viz.; VSG surgery SHP KO (SHP KO VSG), VSG wild type (WT VSG), Sham surgery SHP KO (SHP KO Sham), Sham wild type (WT Sham). Mice were on liquid diet for three days post-surgery and then back on the HFD. Animals were sacrificed 8 weeks post-surgery. Results: SHP KO mice were obese (>30g) but had lower body weight compared to their wild type littermates before surgery (p<0.001). Both KO and WT VSG mice lost more body weight and had lower body weight compared to their respective Sham groups at 8 weeks post-surgery (p<0.001). Serum bile acid levels were not different between groups pre-surgery but were higher in both WT and SHP KO VSG groups compared to the Sham operated mice at 2, 4, 6 and 8 weeks postsurgery (fasting) and 8 weeks (post-prandial; p<0.05). As seen before in WT mice, Cyp7a1 (bile acid synthesis) gene expression in SHP KO VSG mice was suppressed compared to SHP KO Sham mice (p<0.001). Liver triglyceride levels were lower in WT VSG group compared to WT Sham (p<0.001) but no difference was observed between SHP KO VSG and SHP KO Sham groups. Further histological steatosis scores were also not different between SHP KO VSG and SHP KO Sham mice. Plasma ALT levels were lower in WT VSG mice compared to WT Sham (54.86±4.9 U/L vs. 113.7±17.5 WT Sham; p<0.05) while no difference was observed between SHP KO VSG and SHP KO Sham mice (102.6±2.0 and 99.60±10.2). Further the NAFLD Activity Score was higher in SHP KO mice that underwent VSG compared to Sham operated SHP KO mice (p<0.05). Conclusions: SHP KO and WT mice both lost more body weight and had increased serum total bile acid levels after VSG surgery. Despite weight loss and bile acid synthesis gene suppression, SHP KO VSG mice had no reduction in hepatic steatosis, triglyceride accumulation or plasma ALT levels. We conclude that having an intact SHP transcription factor is necessary for the improvement in NASH seen after VSG surgery in obese mice though the suppression of bile acid synthesis seen in SHP KO VSG mice maybe SHP independent.


Randy J. Seeley - Advisory Committees or Review Panels: J&J, Eli Lilly, Zafgen, Novo Nordisk, J&J, Eli Lilly, Zafgen, Novo Nordisk, J&J, Eli Lilly, Zafgen, Novo Nordisk, J&J, Eli Lilly, Zafgen, Novo Nordisk; Consulting: Merck, Amylin, Ethicon Endo-Surgery, Merck, Amylin, Ethicon Endo-Surgery, Merck, Amylin, Ethicon Endo-Surgery, Merck, Amylin, Ethicon Endo-Surgery; Grant/Research Support: Amylin, Zafgen, Ethicon Endo-Surgery, Novo Nordisk, Pfizer, Amylin, Zafgen, Ethicon Endo-Surgery, Novo Nordisk, Pfizer, Amylin, Zafgen, Ethicon Endo-Surgery, Novo Nordisk, Pfizer, Amylin, Zafgen, Ethicon Endo-Surgery, Novo Nordisk, Pfizer; Speaking and Teaching: Eli Lilly, Merck, Novo Nordisk, Eli Lilly, Merck, Novo Nordisk, Eli Lilly, Merck, Novo Nordisk, Eli Lilly, Merck, Novo Nordisk; Stock Shareholder: Zafgen, Zafgen, Zafgen, Zafgen

Rohit Kohli - Grant/Research Support: Johnson and Johnson, Johnson and Johnson

The following people have nothing to disclose: Andriy Myronovych, Rosa-Maria Salazar-Gonzalez, Lili Miles, Karen K. Ryan