Liver-Specific Deletion of Augmenter of Liver ... - Gastroenterology

Gastroenterology 2015;148:379–391

BASIC AND TRANSLATIONAL—LIVER Liver-Specific Deletion of Augmenter of Liver Regeneration Accelerates Development of Steatohepatitis and Hepatocellular Carcinoma in Mice Chandrashekhar R. Gandhi,1,2,3,4 J. Richard Chaillet,5 Michael A. Nalesnik,4,6 Sudhir Kumar,1 Anil Dangi,2,3 A. Jake Demetris,4,6 Robert Ferrell,7 Tong Wu,8 Senad Divanovic,9 Traci Stankeiwicz,9 Benjamin Shaffer,5 Donna B. Stolz,10 Stephen A. K. Harvey,11 Jiang Wang,12 and Thomas E. Starzl4 1

Department of Pediatrics, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio; 2Department of Surgery, University of Cincinnati, Cincinnati, Ohio; 3Cincinnati VA Medical Center, Cincinnati, Ohio; 4Thomas E. Starzl Transplantation Institute, Department of Surgery; 5Department of Microbiology and Molecular Genetics; 6Department of Pathology; 7School of Public Health, University of Pittsburgh, Pennsylvania; 8Department of Pathology and Laboratory Medicine, Tulane University School of Medicine, New Orleans, Louisiana; 9Division of Immunobiology, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio; Department of 10Cell Biology; 11Department of Ophthalmology, University of Pittsburgh, Pennsylvania; and 12Department of Pathology and Laboratory Medicine, University of Cincinnati, Cincinnati, Ohio

BACKGROUND & AIMS: Augmenter of liver regeneration (ALR, encoded by GFER) is a widely distributed pleiotropic protein originally identified as a hepatic growth factor. However, little is known about its roles in hepatic physiology and pathology. We created mice with liver-specific deletion of ALR to study its function. METHODS: We developed mice with liver-specific deletion of ALR (ALR-L-KO) using the albumin-Cre/LoxP system. Liver tissues were collected from ALR-L-KO mice and ALRfloxed/floxed mice (controls) and analyzed by histology, reverse-transcription polymerase chain reaction, immunohistochemistry, electron microscopy, and techniques to measure fibrosis and lipids. Liver tissues from patients with and without advanced liver disease were determined by immunoblot analysis. RESULTS: Two weeks after birth, livers of ALR-L-KO mice contained low levels of ALR and adenosine triphosphate (ATP); they had reduced mitochondrial respiratory function and increased oxidative stress, compared with livers from control mice, and had excessive steatosis, and hepatocyte apoptosis. Levels of carbamyl-palmitoyl transferase 1a and ATP synthase subunit ATP5G1 were reduced in livers of ALR-L-KO mice, indicating defects in mitochondrial fatty acid transport and ATP synthesis. Electron microscopy showed mitochondrial swelling with abnormalities in shapes and numbers of cristae. From weeks 2–4 after birth, levels of steatosis and apoptosis decreased in ALR-L-KO mice, and numbers of ALR-expressing cells increased, along with ATP levels. However, at weeks 4–8 after birth, livers became inflamed, with hepatocellular necrosis, ductular proliferation, and fibrosis; hepatocellular carcinoma developed by 1 year after birth in nearly 60% of the mice. Hepatic levels of ALR were also low in ob/ob mice and alcohol-fed mice with liver steatosis, compared with controls. Levels of ALR were lower in liver tissues from patients with advanced alcoholic liver disease and nonalcoholic steatohepatitis than in control liver tissues. CONCLUSIONS: We developed mice with liver-specific deletion of ALR, and

showed that it is required for mitochondrial function and lipid homeostasis in the liver. ALR-L-KO mice provide a useful model for investigating the pathogenesis of steatohepatitis and its complications.

Keywords: Augmenter of Liver Regeneration; Mouse Model; NASH; ALD.

A

ugmenter of liver regeneration (ALR) (encoded by Gfer [growth factor ERV1 homolog of Saccharomyces cerevisiae]) protein was originally identified and purified from weanling and regenerating rat livers, and its gene cloned.1–4 Expression of ALR in unmodified adult rat liver5 suggested that it might also be functionally significant in this setting. ALR is known to function as a sulfhydryl oxidase,6 cytochrome c reductase,7,8 and inducer of cytosolic protein Fe/S maturation9 and has additional effects such as suppression of hepatic natural killer cell (NK) cytotoxicity10 and Kupffer cell activation.11 Inhibition of ALR synthesis in cultured hepatocytes leads to mitochondrial dysfunction/damage and cell death.12 In order to dissect the roles that ALR plays in normal hepatocyte physiology, we generated liver-specific conditional ALR knockout (ALR-L-KO) mice. The data indicate that ALR is critical for mitochondrial function, lipid homeostasis, and cell survival, and abnormality in ALR gene function may be an important determinant in the development of steatohepatitis and its complications. Abbreviations used in this paper: ACACA, acetyl-CoA carboxylase; AdenoCre, adenovirus containing Cre recombinase; ALD, alcoholic liver disease; ALR, augmenter of liver regeneration; ASH, alcoholic steatohepatitis; ATP, adenosine triphosphate; CPT, carbamoyl palmitoyl transferase; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; HPC, hepatic progenitor cell; KO, knockout; mRNA, messenger RNA; NASH, nonalcoholic steatohepatitis; NK, natural killer cell; ROS, reactive oxygen species; WT, wild type. © 2015 by the AGA Institute 0016-5085/$36.00 http://dx.doi.org/10.1053/j.gastro.2014.10.008

BASIC AND TRANSLATIONAL LIVER

See editorial on page 285.

380

Gandhi et al

Gastroenterology Vol. 148, No. 2


Materials and Methods Animal protocols were approved by Institutional Animal Care and Use Committees according to National Institutes of Health guidelines.

Generation of Liver-Specific Conditional Augmenter of Liver Regeneration Knockout Mouse Details are described in Supplementary Material and Supplementary Figure 1. Briefly, ALR-targeting vector with LoxP sites placed between exons 1 and 2 and after exon 3 of the Gfer gene was constructed, and standard molecular techniques were employed to generate ALRfloxed/floxed mouse. To ALR-L-KO mice, hemizygous Alb-Cre transgenic mice were first crossed with ALRfloxed/floxed mice. F1 Alb-Cre;ALRfloxed/þ offspring were then crossed with ALRfloxed/floxed mice to generate mice with following genotypes: Alb-Cre; ALRfloxed/floxed (homozygous ALRL-KO); Alb-Cre; ALRfloxed/þ ; ALRfloxed/floxed; and ALRfloxed/null. The mice were maintained on a mixed B6.SV129 background. No difference between characteristics of the wild-type (WT) ALRþ/þ and ALRfloxed/floxed mice was observed from birth until more over 1 year. All of the other procedures are established standard techniques and are described in the Supplementary Material.

Statistical Analysis All data are presented as mean ± SD. Statistical significance was determined by Student t test using Graph Pad Prism. A P value of 70% developing tumors and approximately 60% demonstrated hepatocellular carcinoma (HCC) (Figure 7A). Perhaps notably, ALR mRNA and protein expression were similar in WT and ALR-L-KO livers at 1 year (Figure 7B). Histologically, a clear distinction between


tumor and nontumor regions was apparent. Tumor cells were pleomorphic and often steatotic, with frequent mitotic figures and variably anaplastic nuclei (Figure 7C). Ultrastructural examination showed cytosolic and autophagosomal lipid accumulation and dilated endoplasmic reticulum (endoplasmic reticulum stress) (typical of solid organ tumors, including HCC) in the neoplastic hepatocytes (Figure 7D).


Figure 7. Liver tumor development in ALR-L-KO mice and ALR in murine and human fatty liver disease. (A) Macroscopically, there were multiple tumors in ALR-L-KO livers at 1 year. Bar graph shows liver to body weight ratio in ALR-L-KO and WT mice (*P < .01). (B) ALR mRNA and protein expressions as determined by semi-quantitative reverse transcription polymerase chain reaction and Western blot analysis, respectively, were similar in WT and ALR-L-KO livers. (C) The tumor (HCC) and nontumor regions of the ALR-L-KO liver are shown. HCC is characterized by the loss of liver structure with mitotic figures and anaplastic nuclei. (D) Transmission electron micrograph shows lipid accumulation and autophagosomes (arrows) as well as endoplasmic reticulum dilation (arrowheads) in ALR-L-KO liver at 1 year. Bar ¼ 500 nm. (E) Western analysis shows hepatic ALR expression in Ob/Ob mice (8 weeks of age), mice fed Lieber-De Carlie isocaloric (IC) or alcoholic (EtOH) diet for 5 weeks that resulted in fatty liver, and (F) humans with advanced ALD or NASH. Bar graphs in (E) show ratio of ALR vs b-actin expression (n ¼ 5 each) ± SD. In the graph for (F), ALR vs b-actin ratio was plotted individually.

Association of Decreased Augmenter of Liver Regeneration With Hepatic Steatosis and Human Diseases Significant resolution of hepatic steatosis with reemergence of ALR in the ALR-L-KO mouse suggested a relationship between hepatic ALR deficiency and fatty liver disease. Therefore, we measured hepatic ALR in alcoholtreated mice with fatty livers, Ob/Ob mice that develop fatty livers spontaneously (kindly provided by Dr Marsha Cole, Department of Pharmacology, University of Pittsburgh), and humans with advanced alcoholic liver disease (ALD), nonalcoholic steatohepatitis (NASH) and hepatitis C virus (HCV)–induced liver disease or control livers (deidentified samples from the Health Sciences Tissue Bank, University of Pittsburgh Medical Center). The livers of Ob/ Ob and alcohol-treated mice had significantly lower ALR as compared with controls (Figure 7E). More interestingly, hepatic ALR expression was also lower than the mean value for control livers in humans with ALD (4/5) and NASH (4/4) (Figure 7F), but not with HCV-infection and comparable fibrosis (Supplementary Figure 9).

Discussion We report for the first time that ALR is critical for both mammalian embryonic development (failure to obtain homozygous ALRnull/null mouse) and for the postnatal survival and physiologic functions of hepatocytes in vivo. Depletion of hepatic ALR shortly after birth led to profound steatosis, mitochondrial degeneration, and apoptosis of hepatocytes by 2 weeks, followed by continued cell death and regeneration, persistent inflammation, ductular proliferation, and eventually HCC. Hepatocyte loss at 2 weeks is consistent with our previous in vitro finding showing progressive loss of ALR leading to ATP depletion, release of cytochrome c and hepatocyte death.12 In this study,12 we found that hepatocyte death via apoptosis ensues with >30%–40% cellular and mitochondrial ALR depletion. This calculation may not be extended to whole ALR-L-KO livers in vivo because of variable depletion of ALR from individual cells at any given point. ALR plays a critical role as a component of the mitochondrial disulfide relay system that facilitates disulfide bond formation and transport of newly synthesized proteins across the outer mitochondrial membrane.22 In this system, Mia40 (mitochondrial intermembrane space and assembly pathway 40 kDa) oxidizes incoming proteins and is reoxidized by ALR, which in turn transfers its electrons to cytochrome c, which is then acted on by cytochrome c oxidase to convert oxygen to water.23–25 The central role of ALR in this respiratory chain prevents the generation of pathological levels of ROS, shown for example by the deletion of ALR gene in ALRfloxed/floxed hepatocytes causing oxidative stress and mitochondrial damage. The critical importance of ALR in hepatic mitochondrial survival and function is also underscored by the recovery of ALR-L-KO mice upon regaining ALR. This functional in vivo role of ALR is reminiscent of an analogous situation in Saccharomyces

ALR Deficiency and Accelerated Steatohepatitis

389

cerevisiae, in which disruption of ERV-1, the yeast ortholog of ALR, leads to complete loss of the mitochondrial genome with subsequent cell death.26,27 The appearance of ALR-expressing hepatocytes after excessive apoptosis and almost complete loss of ALR in ALRL-KO mice by 2 weeks is interesting and likely reflects the complexity of the in vivo condition. Deletion of target gene expression using AlbCre-LoxP technology occurs progressively beginning at about 1 week post birth and is nearly complete by 2–12 months.28–30 Liver regeneration after profound loss as seen during fulminant liver failure (in contrast to replication of existing hepatocytes after partial hepatectomy) depends on HPCs in bile ductules and canals of Hering.31–33 The existence of robust ductular proliferation and presence of Ki67-positive hepatocytes emerging from these sites in ALR-L-KO mice are most consistent with HPCassociated regeneration. This interpretation is supported by strongly increased a-fetoprotein expression at 2 and 4 weeks in ALR-L-KO livers, and this protein normally declines rapidly after birth.34 It may be that the profound loss of hepatocytes due to ALR depletion is followed by preferential regeneration of progenitor cells that either lack Cre recombinase or have an inactive enzyme. We note that ALR in ALR-L-KO mice did not reach WT levels until almost 1 year, which could suggest the continued death of hepatocytes with replacement by ALRexpressing cells. Excessive accumulation of triglycerides in ALR-L-KO livers at 2 weeks could be related to mitochondrial damage, reduced ATP content, decreased mitochondrial metabolic capacity, and/or impaired fatty acid b-oxidation.35,36 Hepatic expression and activity of CPT1a, that regulates free fatty acid transport into mitochondria for b-oxidation, was strongly decreased in the ALR-L-KO mice. Reduced expression of peroxisome proliferator–activated receptor–a, which regulates both peroxisomal and mitochondrial fatty acid b-oxidation, provides additional support to our contention that ALR deficiency has a strong negative effect on fatty acid b-oxidation. Unaltered fatty acid synthase expression and lower ACACA expression in ALR-L-KO mice than in WT mice at 2 weeks likely exclude accumulation of fatty acids due to increased synthesis. Inhibition of CPT1a activity by high levels of malonyl CoA can also be ruled out because expression of ACACA that catalyzes malonyl CoA synthesis was reduced. These results and the decrease in Atp5g1 and TFAM expression in ALR-L-KO liver suggest an interesting possibility that ALR, which is also present in the nucleus,6,37 may influence expression of genes involved in lipid and energy metabolism at the transcriptional level. It is worth noting that an ontologic survey showed that the CPT1a, TFAM, and ACACA gene expression is targeted by microRNAs-370, -122, and -33 raising the speculation that ALR might modulate the expression of these and/or other functional miRNAs. Despite the incremental increase in ALR that did not reach WT levels, hepatic steatosis was reversed by 4 weeks and only a few cells with lipid accumulation were seen at 8 weeks in ALR-L-KO livers. This observation suggests that a threshold concentration of ALR, especially in mitochondria, is required for hepatocyte function. Based on the Western


February 2015

390

Gandhi et al


analysis, we predict that the high molecular weight ALR (40 kDa) might be required for this function. Despite reversal of steatosis and regeneration with ALRexpressing cells, there was a strong inflammatory response with neutrophil infiltration and increased expression of proinflammatory cytokines. Continued death of hepatocytes due to mediators released by infiltrating cells (notably neutrophils, NKT and CD8þ T cells) likely sustained the inflammatory environment throughout the life of the ALR-L-KO mice. Continued inflammation, hepatocyte death, and regeneration (from oval cells/HPCs) precede dysplasia and tumor development in the ALR-L-KO mouse. In this regard, oval cells have been detected in human livers in conditions associated with increased risk of neoplastic development.38–40 Indeed, A6- and keratin 19–positive cells continued to be found in ALR-L-KO liver until 1 year (Supplementary Figure 3C), suggesting that detrimental/carcinogenic progenitors might persist and give rise to the tumors. However, the possibility that individual ALRdeficient hepatocytes persisted in this environment and contributed to neoplasia cannot be excluded. Our results might therefore have direct implications for analysis of the complications of advanced alcoholic (ASH) and NASH, as both diseases involve severe mitochondrial dysfunction and can progress to inflammation, fibrosis/cirrhosis, and liver cancer.41–43 Although hepatic ALR levels were lower in subjects with ASH and NASH, ALR levels were similar in the ALR-L-KO and WT mice at 1 year when they developed HCC. Another study reported greater levels of hepatic ALR in patients with liver cirrhosis, HCC, and cholangiocarcinoma,44 and hepatic ALR expression was similar in normal and HCV livers (Supplementary Figure 9). It is also plausible that increased ALR might support survival and replication of neoplastic hepatocytes. Although a definitive conclusion cannot be drawn from the human data due to limited sample size, we found 48 sequence variants, 17 of which were novel (not in SNPdb), in a preliminary study in which the ALR gene from 100 human samples was sequenced (not shown). This raises the speculation that some of these variants might be present among the population vulnerable to ASH and/or NASH. Our current understanding of the precise mechanisms of advanced ASH and NASH, which occur in only a subset of the population exposed to associated risk factors, is inadequate. Based on the results of these studies, we propose that the ALR-L-KO mouse can be an important model to investigate mechanisms relevant not only to the clinical development of ALD or NASH, but also to the fibroinflammatory and neoplastic consequences that arise from these diseases. The full role of this ubiquitous protein in human physiology remains to be elucidated.

Supplementary Material Note: To access the supplementary material accompanying this article, visit the online version of Gastroenterology at www.gastrojournal.org, and at http://dx.doi.org/10.1053/ j.gastro.2014.10.008.


References 1. Francavilla A, Hagiya M, Porter KA, et al. Augmenter of liver regeneration: Its place in the universe of hepatic growth factors. Hepatology 1994;20:747–757. 2. Gandhi CR. Augmenter of liver regeneration. Fibrogenesis Tissue Repair 2012;5:10. 3. Hagiya M, Francavilla A, Polimeno L, et al. Cloning and sequence analysis of the rat augmenter of liver regeneration (ALR) gene: expression of biologically active recombinant ALR and demonstration of tissue distribution. Proc Natl Acad Sci U S A 1994;91:8142–8146. 4. Giorda R, Hagiya M, Seki T, et al. Analysis of the structure and expression of the augmenter of liver regeneration (ALR) gene. Mol Med 1996;2:97–108. 5. Gandhi CR, Kuddus R, Subbotin VM, et al. A fresh look at augmenter of liver regeneration in rats. Hepatology 1999; 29:1435–1445. 6. Tury A, Mairet-Coello G, Lisowsky T, et al. Expression of the sulfhydryl oxidase ALR (augmenter of liver regeneration) in adult rat brain. Brain Res 2005;1048:87–97. 7. Farrell SR, Thorpe C. Augmenter of liver regeneration: a flavin-dependent sulfhydryl oxidase with cytochrome c reductase activity. Biochemistry 2005;44:1532–1541. 8. Allen S, Balabanidou V, Sideris DP, et al. Erv1 mediates the Mia40-dependent protein import pathway and provides a functional link to the respiratory chain by shuttling electrons to cytochrome c. J Mol Biol 2005;353:937–944. 9. Lange H, Lisowsky T, Gerber J, et al. An essential function of the mitochondrial sulfhydryl oxidase Erv1p/ ALR in the maturation of cytosolic Fe/S proteins. EMBO Rep 2001;2:715–720. 10. Francavilla A, Vujanovic NL, Polimeno L, et al. The in vivo effect of hepatotrophic factors augmenter of liver regeneration, hepatocyte growth factor, and insulin-like growth factor-II on liver natural killer cell functions. Hepatology 1997;25:411–415. 11. Gandhi CR, Murase N, Starzl TE. Cholera toxin-sensitive GTP-binding protein-coupled activation of augmenter of liver regeneration (ALR) receptor and its function in rat Kupffer cells. J Cell Physiol 2010;222:365–373. 12. Thirunavukkarasu C, Wang LF, Harvey SA, et al. Augmenter of liver regeneration: an important intracellular survival factor for hepatocytes. J Hepatol 2008; 48:578–588. 13. Dyer MR, Walker JE. Sequences of members of the human gene family for the c subunit of mitochondrial ATP synthase. Biochem J 1993;293:51–64. 14. Li HS, Zhang JY, Thompson BS, et al. Rat mitochondrial ATP synthase ATP5G3: cloning and upregulation in pancreas after chronic ethanol feeding. Physiol Genom 2001;6:91–98. 15. Tiranti V, Rossi E, Ruiz-Carrillo A, et al. Chromosomal localization of mitochondrial transcription factor A (TCF6), single-stranded DNA-binding protein (SSBP), and endonuclease G (ENDOG), three human housekeeping genes involved in mitochondrial biogenesis. Genomics 1995;25:559–564. 16. Lemasters JJ, Thurman RG. Hypoxia and reperfusion injury to liver. Prog Liver Dis 1993;11:85–114.

17. Kaplowitz N, Tsukamoto H. Oxidative stress and liver disease. Prog Liver Dis 1996;14:131–159. 18. Kanvah S, Joseph J, Schuster GB, et al. Oxidation of DNA: damage to nucleobases. Acc Chem Res 2010; 43:280–287. 19. Cheng KC, Cahill DS, Kasai H, et al. 8-Hydroxyguanine, an abundant form of oxidative DNA damage, causes G/T and C/A substitutions. J Biol Chem 1992; 267:166–172. 20. Chalmers RA, Stanley CA, English N, et al. Mitochondrial carnitine-acylcarnitine translocase deficiency presenting as sudden neonatal death. J Pediatr 1997;131:220–225. 21. Rakhshandehroo M, Knoch B, Müller M, et al. Peroxisome proliferator-activated receptor alpha target genes. PPAR Res 2010;2010:ID612089. 22. Riemer J, Bulleid N, Herrmann JM. Disulfide formation in the ER and mitochondria: two solutions to a common process. Science 2009;324:1284–1287. 23. Bihlmaier K, Mesecke N, Terziyska N, et al. The disulfide relay system of mitochondria is connected to the respiratory chain. J Cell Biol 2007;179:389–395. 24. Banci L, Bertini I, Calderone V, et al. Molecular recognition and substrate mimicry drive the electron-transfer process between MIA40 and ALR. Proc Natl Acad Sci U S A 2011;108:4811–4816. 25. Dabir DV, Hasson SA, Setoguchi K, et al. A small molecule inhibitor of redox-regulated protein translocation into mitochondria. Dev Cell 2013;25:81–92. 26. Lisowsky T. Dual function of a new nuclear gene for oxidative phosphorylation and vegetative growth in yeast. Mol Gen Genet 1992;232:58–64. 27. Lisowsky T. ERV1 is involved in the cell-division cycle and the maintenance of mitochondrial genomes in Saccharomyces cerevisiae. Curr Genet 1994;26:15–20. 28. Postic C, Shiota M, Niswender KD, et al. Dual roles for glucokinase in glucose homeostasis as determined by liver and pancreatic beta cell-specific gene knock-outs using Cre recombinase. J Biol Chem 1999;274: 305–315. 29. Gu J, Weng Y, Zhang QY, et al. Liver-specific deletion of the NADPH-cytochrome P450 reductase gene: impact on plasma cholesterol homeostasis and the function and regulation of microsomal cytochrome P450 and heme oxygenase. J Biol Chem 2003;278:25895–25901. 30. Xu Z, Chen L, Leung L, et al. Liver-specific inactivation of the Nrf1 gene in adult mouse leads to nonalcoholic steatohepatitis and hepatic neoplasia. Proc Natl Acad Sci U S A 2005;102:4120–4125. 31. Michalopoulos GK. Liver regeneration: alternative epithelial pathways. Int J Biochem Cell Biol 2011;43: 173–179. 32. Best J, Dollé L, Manka P, et al. Role of liver progenitors in acute liver injury. Front Physiol 2013;4:258.

ALR Deficiency and Accelerated Steatohepatitis

391

33. Riehle KJ, Dan YY, Campbell JS, et al. New concepts in liver regeneration. J Gastroenterol Hepatol 2011;26(Suppl 1):203–212. 34. Tilghman SM, Belayew A. Transcriptional control of the murine albumin/alpha-fetoprotein locus during development. Proc Natl Acad Sci U S A 1982;79:5254–5257. 35. Boison D, Scheurer L, Zumsteg V, et al. Neonatal hepatic steatosis by disruption of the adenosine kinase gene. Proc Natl Acad Sci U S A 2002;99:6985–6990. 36. Fromenty B, Pessayre D. Inhibition of mitochondrial boxidation as a mechanism of hepatotoxicity. Pharmacol Ther 1995;67:101–154. 37. Lu C, Li Y, Zhao Y, et al. Intracrine hepatopoietin potentiates AP-1 activity through JAB1 independent of MAPK pathway. FASEB J 2002;16:90–92. 38. Libbrecht L, Roskams T. Hepatic progenitor cells (HPCs) in human liver tumor development. Semin Cell Dev Biol 2002;13:389–396. 39. Lowes KN, Brennan BA, Yeoh GC, et al. Oval cell numbers in human chronic liver diseases are directly related to disease severity. Am J Pathol 1999;154:537–541. 40. Roskams TA, Libbrecht L, Desmet VJ. Progenitor cells in diseased human liver. Semin Liver Dis 2003;23:385–396. 41. Powell EE, Cooksley WG, Hanson R, et al. The natural history of nonalcoholic steatohepatitis: a follow-up study of forty-two patients for up to 21 years. Hepatology 1990; 11:74–80. 42. Day C, Saksena S. Non-alcoholic steatohepatitis: definitions and pathogenesis. J Gastroenterol Hepatol 2002; 17:S377–S384. 43. Brunt EM. Nonalcoholic steatohepatitis. Semin Liver Dis 2004;24:3–20. 44. Thasler WE, Schlott T, Thelen P, et al. Expression of augmenter of liver regeneration (ALR) in human liver cirrhosis and carcinoma. Histopathology 2005;47:57–66.

Received May 21, 2014. Accepted October 7, 2014. Reprint requests Address requests for reprints to: Chandrashekhar R. Gandhi, PhD, Department of Surgery ML0558, University of Cincinnati, 231 Albert Sabin Way, Cincinnati, Ohio 45267-0558. e-mail: [email protected]; fax: 513-558-8677. Acknowledgments The authors thank Dr George Michalopoulos for his advice and comments, Mr John Prelich, and Ms Rachel Stewart for excellent technical assistance, and Dr Berthony Deslouches for his help during the initial stages of this work. A6 rat anti-mouse Ab was a kind gift from Dr Valentina Factor, National Institutes of Health, National Cancer Institute. Conflicts of interest The authors disclose no conflicts. Funding This work was supported by a VA Merit Review Award (1IO1BX001174) and grants from National Institutes of Health (DK 54411, PO1AIO81678, and R21AA020846) to CRG, as well as support from Cincinnati Children’s Hospital Medical Center and University of Cincinnati.


February 2015

391.e1

Gandhi et al

Supplementary Methods Generation of Liver-Specific Alrnull/null Mouse To generate liver-specific knockout of ALR, we first constructed a conditional knockout of the Alr gene (also known as Gfer). For targeted mutagenesis of the mouse Alr locus on chromosome 17, a replacement vector was constructed from a single clone isolated from a 129/Sv strain mouse genomic l phage library. Standard plasmid cloning and site-directed mutagenesis techniques were used to generate the targeting plasmid in pBluescript. The methods used for gene targeting in mouse embryonic stem cells (ES) are described schematically in Supplementary Figure 1. Accordingly, the structure of the Alr targeting vector, a linearized version of the targeting plasmid is shown in which the Alr gene is flanked by loxP sites (floxed). The 50 loxP site was placed in the first intron instead of the 50 end of the gene in order to avoid possible adverse effects of a loxP insert on cis-acting regulatory sequences in the 50 end of Alr (and possibly 50 regulatory sequence of the adjacent, oppositely transcribed gene Noxo1). The 2 loxP sites recombined in the presence of Cre recombinase removing exons 2 and 3 of Alr to generate a AlrNull allele. Although exon 1, which codes for the first 79 amino acids of mouse ALR, would remain in the AlrNull allele, it was unlikely that mRNA transcripts through this region from the Alr promoter would be processed correctly, and therefore highly unlikely that any truncated forms of ALR protein would be made. The linearized targeting construct containing the neor gene flanked by FRT sites was transfected into mouse ES cells and a number of homologous recombinant ES cell clones were obtained. These clones were identified by Southern blot analysis of DNA samples from transfected G418-resistant colonies (data not shown). Chimeric mice were generated from 3 heterozygous mutant ES clones that contained a targeted Alr allele, and heterozygous mutant mice were obtained from a chimeric male generated with the mutant ES clone designated Alr25. Heterozygous AlrFloxed mice were obtained by crossing these heterozygous mutant mice with transgenic mice expressing FLP recombinase under control of the b-actin promoter. In the presence of FLP recombinase, FRT sites recombine and neor gene sequences are deleted. Heterozygous Alrnull mice were produced, in turn, by crossing AlrFloxed mice with a transgenic mouse strain that expresses the Cre recombinase protein from the adenovirus EIIa promoter.1 Supplementary Figure 1B shows the Southern blot demonstrating the presence of various Alr alleles. Accordingly, Alr gene is deleted in the liver and not in other organs (Southern data for the kidney are shown). Hepatocytes isolated from an 8-week-old ALR-L-KO mouse show absence of the WT allele and deletion of the Floxed allele with formation of the null allele. We first attempted to generate ALRnull/null mice from ALRþ/null mice as breeding pairs. However, no ALRnull/null mice could be obtained, indicating the absolute requirement of ALR during embryonic development. No homozygous null/null E9.5 embryos were detected in crosses between


heterozygous null/þ parents, suggesting that null/null embryos die early in embryogenesis due to lack of Alr protein. To generate ALR-L-KO mice, hemizygous Alb-Cre transgenic mice were crossed first with ALRfloxed/floxed mice. F1 Alb-Cre;ALRfloxed/þ offsprings were then crossed with ALRfloxed/floxed mice to generate mice with the following genotypes: Alb-Cre; ALRfloxed/floxed (homozygous ALR-L-KO); Alb-Cre; ALRfloxed/þ ; ALRfloxed/floxed; and ALRfloxed/null. The mice were maintained on a mixed B6.SV129 background. No difference between characteristics of the WT ALRþ/þ and ALRfloxed/floxed mice was observed from birth until over 1 year.

Determination of Histopathology Mice were weighed at various times after birth and anesthetized. After drawing blood, lthe iver was excised, weighed, and portions were fixed in 10% buffered formalin or 2.5% glutaraldehyde in phosphate-buffered saline (PBS) at 4 C, or snap frozen in liquid nitrogen. Formalin-fixed and paraffin-embedded liver sections were used for H&E staining (histopathologic analysis).

Augmenter of Liver Regeneration Immunostaining Paraformaldehyde-fixed liver sections were permeabilized with 0.1% Triton X-100 in PBS, followed by blocking in 2% bovine serum albumin (BSA) and incubation in affinity-purified anti-ALR antibody (Ab).2 Sections were then incubated in anti-rabbit secondary Ab for 1 hour and blue Hoechst reagent.

Identification of Hepatic Progenitor Cells or Oval Cells Paraformaldehyde-fixed sections were blocked with 5% donkey serum in 0.1% BSA in PBS, followed by application of A6 rat anti-mouse Ab (kind gift from Dr Valentina Factor) (1:200 in PBS/1% BSA). Sections were then washed and incubated in secondary donkey anti-rat Ab, followed by nuclear labeling with 40 ,6-diamidino-2-phenylindole.

Terminal Deoxynucleotidyl Transferase dUTP Nick End Labeling Staining Terminal deoxynucleotidyl transferase dUTP nick end labeling assay was performed as per the manufacturer’s instructions (Roche Pharmaceuticals, Nutley, NJ).

Ki67 and a–Smooth Muscle Actin Staining Formalin-fixed liver sections were deparaffinized. AntiKi67 Ab (AbCam, Cambridge, MA)3 or anti–a-smooth muscle actin Ab (Dako, Carpinteria, CA)4 in conjunction with relevant secondary Abs were employed to detect proliferating cells and activated stellate cells or myofibroblasts, respectively.

Sirius Red Staining Sirius Red staining was performed as described previously.5 Briefly, paraffin-embedded liver section was deparaffinized/hydrated and series of washes were performed,

February 2015

followed by staining with 0.1% Sirius Red (Direct Red 80; Sigma-Aldrich, St Louis, MO) in saturated aqueous picric acid.

Oil Red-O Staining Sections of the frozen tissue were allowed to dry on glass slides, then fixed in ice-cold 10% neutral buffered formalin. The sections were then covered successively with absolute propylene glycol, pre-warmed Oil Red-O, 85% propylene glycol, then rinsed in water, covered with Mayer’s hematoxylin, rinsed in water, and cover-slipped with gelvatol.

Determination of Inflammatory Cells Hepatic accumulation of inflammatory cells was identified in paraffin-embedded sections immunohistochemically using Naphthol AS-D Chloroacetate (Specific Esterase) Kit (Sigma-Aldrich).

Keratin 19 Staining Paraffin-embedded liver sections were deparaffinized followed by microwave heating in citrate buffer (pH 6.0). The sections were then incubated with goat polyclonal keratin (CK)19 Ab (sc33119; Santa Cruz Biochemicals, Santa Cruz, CA) at 1:50 dilution overnight at 4 C, followed by incubation with anti-goat IgG-horseradish peroxidase (Sigma) for 2 hours at room temperature, then reacted with diaminobenzidine (Thermo Scientific, Logan, UT) to develop a brown color.

Protein and RNA Analysis Protein lysates were prepared from the frozen tissue, and Western blotting was performed to assess ALR and apoptosis-related proteins as described previously.2,4 RNA was prepared from the frozen liver tissue and the various mRNA transcripts were quantified by quantitative reverse transcription polymerase chain reaction, as described previously6,7 using primers listed in Supplementary Table 1.

Caspase-3 Assay Frozen livers from WT and ALR-L-KO mice were homogenized in cold lysis buffer (50 mmol/L HEPES [pH 7.4], 150 mmol/L NaCl, 20 mmol/L ethylene diaminetetraacetic acid, 0.2% Triton X-100) containing protease inhibitor cocktail (Sigma) and incubated on ice for 30 minutes. Cell lysates were centrifuged (13,000g, 15 minutes, 4 C) to obtain supernatants for the assay. Caspase-3 fluorimetric assay were performed using a commercial caspase-3 assay Kit (Sigma). In brief, enzyme activity was measured by caspase-3–mediated hydrolysis of the peptide substrate acetyl-Asp-Glu-Val-Asp-7-amido-4-methylcoumarin (AcDEVD-AMC) and the release of the fluorescent 7-amino-4methylcoumarin.

Transmission Electron Microscopy After overnight fixation in 2.5% glutaraldehyde in PBS at 4 C, several 1-mm3 cubes were removed from the liver, washed (3 times) in PBS, and post-fixed in aqueous 1%

ALR Deficiency and Accelerated Steatohepatitis 391.e2

OsO4, 1% K3Fe(CN)6 for 1 hour. After 3 PBS washes, the pellet was dehydrated through a graded series of 30%– 100% ethanol, 100% propylene oxide, and infiltrated in 1:1 mixture of propylene oxide:Polybed 812 epoxy resin (Polysciences, Warrington, PA) overnight at room temperature. After several changes of 100% resin over 24 hours, the pellet was embedded in molds, cured at 37 C overnight, followed by additional hardening at 65 C for 2 more days. Ultrathin (70 nm) sections were collected on 200 mesh copper grids, stained with 2% uranyl acetate in 50% methanol for 10 minutes, followed by 1% lead citrate for 7 minutes. Sections were imaged using a JEOL JEM 1011 transmission electron microscope (Peabody, MA) at 80 kV fitted with a side mount AMT 2k digital camera (Advanced Microscopy Techniques, Danvers, MA).

Mitochondrial Isolation Liver tissue was minced in ice-cold liver isolation buffer (containing 210 mM D-mannitol, 70 mM sucrose, 5 mM HEPES, 1 mM ethylene glycol-bis-tetraacetic acid, and 0.5% fatty acid-free BSA [pH 7.4]), homogenized, and centrifuged (700g; 10 minutes). The supernatant was centrifuged (10,000g; 10 minutes) and the pellet containing mitochondria was washed in liver isolation buffer and centrifuged (10,000g; 10 minutes) 3 times. Mitochondria were then suspended in mitochondrial assay solution 1 (buffer containing 70 mM sucrose, 220 mM D-mannitol, 5 mM potassium dihydrogen phosphate, 5 mM magnesium chloride, 2 mM HEPES, 1 mM EDTA, and 0.2% fatty acid-free BSA [pH 7.4]) to yield high-density mitochondrial suspension. Mitochondrial protein concentration was determined with a bicinchoninic acid protein assay kit (Thermo Scientific Pierce).

Determination of Mitochondrial Oxygen Consumption Rate Isolated liver mitochondria were used to determine oxygen consumption rate (OCR) in XF24 Analyzer (Seahorse Bioscience, Copenhagen, Denmark). Briefly, an XF24 extracellular flux assay cartridge was hydrated overnight at 37 C according to manufacturer’s instruction. Mitochondria (10 mg per well) were plated in mitochondrial assay solution 1 buffer onto a polyethylenimine precoated XF Cell Culture Microplate, centrifuged (200g/12 minutes) and incubated at 37 C in a non-CO2 incubator for 10 minutes. OCR was quantified after sequential injections of 5 mM succinate, 5 mM succinate/1 mM adenosine diphosphate, 2 mM oligomycin, and 4 mM carbonyl cyanide p-trifluoromethoxyphenylhydrazone. The OCR values were normalized to the maximum respiratory rate induced by carbonyl cyanide p-trifluoromethoxyphenylhydrazone stimulation.

Determination of Carbamyl Palmitoyl Transferase 1 Activity CPT1 enzyme activity was determined as described previously.8 Briefly, liver homogenates of ALR-L KO and

391.e3

Gandhi et al

age-matched WT mice, prepared in a buffer containing 0.25 M sucrose, 1 mM EDTA, 0.1% ethanol, and protease inhibitors cocktail at a 1:5 (w/v) ratio on ice, were centrifuged at 300g for 10 minutes at 4 C, and the supernatant was centrifuged at 12,000g for 5 minutes at 4 C. Ten microliters of each of the supernatant and Tris-HCL DCNB buffer (116 mM Tris-HCl [pH 8.0], 2.5 mM EDTA, 2 mM DTNB, and 0.2% Triton X-100) were mixed in a 96-well plate and incubated for 5 minutes at 30 C. Ten microliters of 1 mM palmitoylCoA was added, and the reaction was then started by adding 2 mL L-carnitine solution (1.2 mM dissolved in 1 M Tris-HCl pH 8.0). Optical density was recorded at 412 nm for 3 minutes. The enzyme activity was defined as nmol CoA-SH released/min/mg protein or U/mg protein.

Biochemical Analysis and Oxidative Stress Serum alanine aminotransferase was determined using Infinity ALT reagent (Thermo DMA, Louisville, CO). Liver triglycerides and cholesterol were determined using kits from Stanbio Laboratories, Boerne, TX. Lipid peroxidation was quantified by measuring thiobarbituric acid-reactive substance. The liver tissue was homogenized in 20 mM phosphate buffer (pH 7.4) containing 5 mM butylated hydroxytoluene. The homogenate was mixed at a 1:2 ratio with a solution containing 15% trichloroacetic acid, 0.37% thiobarbituric acid, 0.2M HCl, and 0.03% butylated hydroxytoluene, and boiled for 15 minutes. Thiobarbituric acid-reactive substance was measured spectrophotometrically at 535 nm.

Adeno-Cre Infection of Gferfloxed/floxed Hepatocytes

Hepatocytes isolated from 6- to 8-week-old Gferfloxed/floxed mice were cultured on gelatin-coated 6-well plates in William’s medium E supplemented with 5% fetal bovine serum and human insulin (7 mg/mL) at 37 C with 5% CO2, as described previously.2,4 After renewal of the medium at 3 hours and overnight culture, the cells were incubated in 1 mL culture medium containing Adeno-Cre (4 108 viral particles). To assess adequate transfection, hepatocytes were also transfected with Adeno-GFP; all cells were found to be GFP-positive after 8 hours. Oxidative stress, mitochondrial injury and DNA damage, and fatty acid uptake/accumulation were determined at 9, 12, and 15 hours.

Determination of Oxidative Stress Hepatocytes were washed with PBS 3 time and placed in PBS containing 20 ,70 -dichlorofluorescin diacetate (1 mM in dimethyl sulfoxide; Sigma) 5 minutes before analysis. 20 ,70 Dichlorofluorescin diacetate is deacetylated by cellular esterases and then oxidized by reactive oxygen species to 20 ,70 -dichlorofluorescin, which was detected by fluorescence microscopy with maximum excitation and emission spectra of 495 nm and 529 nm respectively.


Determination of Mitochondrial Injury and DNA Damage Hepatocytes were incubated culture medium containing mitotracker (15 mM) (Cell Signaling, Danvers, MA) in the for 30 minutes, washed twice with PBS, and fixed in ice-cold methanol at –20 C for 20 minutes. Hepatocytes were then incubated in blocking buffer (3% BSA, 0.1% Triton X100, and 5% donkey serum in PBS) and labeled with anti8-Oxoguanin Ab (1:400) (Abcam, Cambridge, MA) overnight at 4 C, followed by incubation with rabbit anti-mouse fluorescein isothiocyanate, and 40 ,6-diamidino-2-phenylindole for nuclear labeling.

Determination of Fatty Acid Uptake/ Accumulation Oleic or palmitic acid (800 mM in 2 mL isopropanol) was added to the medium at the time of Adeno-Cre transfection. The cells were washed with PBS and fixed with 10% formalin for 5 minutes. Oil Red O staining was performed as described here.

References 1. Lakso M, Pichel JG, Gorman JR, et al. Efficient in vivo manipulation of mouse genomic sequences at the zygote stage. Proc Natl Acad Sci U S A 1996;93:5860–5865. 2. Gandhi CR, Kuddus R, Subbotin VM, et al. A fresh look at augmenter of liver regeneration in rats. Hepatology 1999; 29:1435–1445. 3. Lunz JG 3rd, Tsuji H, Nozaki I, et al. An inhibitor of cyclindependent kinase, stress-induced p21Waf-1/Cip-1, mediates hepatocyte mito-inhibition during the evolution of cirrhosis. Hepatology 2005;41:1262–1271. 4. Thirunavukkarasu C, Yang Y, Subbotin VM, et al. Endothelin receptor antagonist TAK-044 arrests and reverses the development of carbon tetrachloride-induced cirrhosis in rats. Gut 2004;53:1010–1019. 5. Sweat F, Puchtler H, Rosenthal S. Sirius red F3BA as a stain for connective tissue. Arch Pathol Lab Med 1964; 78:69–72. 6. Harvey SAK, Dangi A, Tandon A, et al. The transcriptomic response of rat hepatic stellate cells to endotoxin: implications for hepatic inflammation and immune regulation. PLoS ONE 2013;8:e82159. 7. Stewart R, Dangi A, Huang C, et al. A novel mouse model of depletion of stellate cells clarifies their role in ischemia/reperfusion- and endotoxin-induced acute liver injury. J Hepatol 2014;60:298–305. 8. Bieber LL, Abraham T, Helmrath T. A rapid spectrophotometric assay for carnitine palmitoyltransferase. Anal Biochem 1972;50:509–518. Author names in bold designate shared co-first authorship.

February 2015

ALR Deficiency and Accelerated Steatohepatitis 391.e4

Supplementary Table 1.Quantitative Polymerase Chain Reaction Primer Sequences ACACA ALR ATP5G1 bactin Bcl2 Bax CD4 CD8 CPT1a Collagen 1 FAS a-fetoprotein GAPDH IFNg IL-1b IL6 IL10 IL33 MMP13 NKG2D PPARa SREBP1 TFAM TIMP1 TNFa

Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Reverse Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse

50 -50 -AGCCAGAAGGGACAGTAGAA-30 50 -CTCAGCCAAGCGGATGTAAA-30 50 -AATTGGGTCGCCACACCTG-30 50 -GGCCAGCGTATGGAGGAAA-30 50 -AGTTGGTGTGGCTGGATCA-30 50 -GCTGCTTGAGAGATGGGTTC-30 50 -AGAGGGAAATCGTGCGTGAC-30 50 -CAATAGTGATGACCTGGCCGT-30 50 -TGAGTACCTGAACCGGCATCT-30 50 -GCATCCCAGCCTCCGTTAT-30 50 -AAGGCCTCCTCTCCTACTT-30 50 -CAGCCCATCTTCTTCCAGAT-30 50 -TCCCACTCACCCTCAAGATA-30 50 -ATCACCACCAGGTTCACTTC-30 50 -GGGATCACCAGCATGCTTTA-30 50 -CACGTCTAGTCTGCCTCTTTAC-30 50 -ACTCTTGGAAGAAGTTCA-30 50 -AGTATCTTTGACAGCTGGGAC-30 50 -AGACCTGTGTGTTCCCTACT-30 50 -GAATCCATCGGTCATGCTCTC-30 50 -CTGCGGAAACTTCAGGAAATG-30 50 -GGTTCGGAATGCTATCCAGG-30 50 -CTTGGTGAAGCAAAAGCCTGAA-30 50 -GGACCCTCTTCTGTGAAACAGACT-30 50 -TGTTGAAGTCACAGGAGACAACCT-30 50 -AACCTCCCAAGTATGATGACATCA-30 50 -CTCTTCCTCATGGCTGTTTCT-30 50 -TTCTTCCACATCTATGCCACTT-30 50 -GGTGTGTGACGTTCCCATTA-30 50 -ATTGAGGTGGAGAGCTTTCAG-30 50 -CCGGAGAGGAGACTTCACAG-30 50 -TCCACGATTTCCCAGAGAAC-30 50 -CCCTGGGTGAGAAGCTGAAG-30 50 -CACTGCCTTGCTCTTATTTTCACA-30 50 -CCTACTCCCTCAGCTTTCTTTC-30 50 -GCAGGGTAAAGACAGTGGAATA-30 50 -GAC ACA GCA AGC CAG AAT AAA G-30 50 -GGAAAGCAGAGAGGGATTAACA-30 50 -GCA GGG TAA AGA CAG TGG AAT A-30 50 -TGTGACACCACTTCCGAATC-30 50 -CGGGAAAGACCAGCAACAAC-30 50 -ATAGCAGCCACAAACAGGGA-30 50 -AGCCCTCCACCAGGTAATAA-30 50 -GGGTTCCCAGTCTACTCACTAA-30 50 -GCTCTACACGCCCCTGGTTTCTGG-30 50 -TCGCTGTAGTGCCTGCTGCTCCTG-30 50 -CGGTGGGTGGATGAGTAATG-30 50 -GGCTGCACAGTGGAGAATAA-30 50 -CCCAGGTATATGGGCTCATACC-30 50 -GCCGATTTGCTATCTCATACCAGG-30