SREBP2: A link between insulin resistance ... - Wiley Online Library

doi:10.1111/j.1440-1746.2011.06704.x

EDITORIALS

jgh_6704

789..795

SREBP-2: A link between insulin resistance, hepatic cholesterol, and inflammation in NASH Derrick M Van Rooyen and Geoff C Farrell Australian National University Medical School, Australian National University at the Canberra Hospital, ACT, Canberra, Australia

See article in J. Gastroenterol. Hepatol. 2011; 26: 875–883. In recent years, the prevailing two-“hit” model of non-alcoholic steatohepatitis (NASH) pathogenesis has been challenged and gradually replaced by a model of lipotoxicity, which envisages multiple interactive connections between the metabolic and inflammatory determinants of NASH. The original, widelyaccepted model, theorized that a first “hit,” namely hepatic steatosis, was caused by metabolic factors (obesity, type 2 diabetes [T2D], dyslipidemia), and sensitized the liver to multiple second “hits” that cause hepatocellular injury and liver inflammation. Injury mechanisms are clearly operative in NASH; they include oxidant stress and immunomodulation via cytokines and innate immunity, culminating in hepatocellular injury/cell death and liver fibrosis.1 The more embracing lipotoxicity hypothesis, however, is based on the premise that metabolic and injury domains of steatohepatitis are interactive, not separate. Specifically, one or more (yet to be elucidated), “toxic/pro-inflammatory” lipid species accumulate in the liver in some cases of steatosis, and these molecules are what subsequently lead to hepatic inflammation, cell death (“hepatitis”), and fibrosis.2 The search for the key specific lipid mediators of liver injury in fatty liver disease has sparked a myriad of clinical and experimental studies. Clinical investigations, however, are difficult, in light of the ethical and logistical constraints surrounding the availability of suitable liver tissue from lean (healthy) patients and those with either simple steatosis (who are rarely biopsied for clinical indications) or NASH. On the other hand, much data generated from experimental studies have been in reductionist systems, such as primary hepatocytes or non-physiological (cancer) cell lines, or have employed small animal models of non-alcoholic fatty liver disease (NAFLD) that either do not exhibit steatohepatitis, or show steatohepatitis, but the pathology is not caused by obesity/T2D/ metabolic syndrome, the risk factors for human NAFLD/NASH.3 Despite these challenges and limitations, studies in animal models, particularly those using molecular inhibition of triglyceride synthesis,4 and available small human lipidomic studies, have ruled

Accepted for publication 16 February 2011. Correspondence Professor Geoff C Farrell, Australian National University Medical School, Australian National University at The Canberra Hospital, Yamba Drive, Garran, ACT 2605, Australia. Email: geoff.farrell@ anu.edu.au

out triglycerides (TG) as the major lipotoxic mediator of NASH.5 The focus now falls on other lipid species, particularly free fatty acids (FFA), diacylglycerides, toxic phospholipids (ceramides, sphingolipids),5 and most recently, cholesterol. What is the evidence that cholesterol is associated with NASH, and how does it accumulate in the liver? Puri et al. reported the first lipidomic study of NASH patients and found no difference in TG or FFA between the small numbers with NASH versus simple steatosis.6 Instead, they identified increased hepatic free cholesterol (FC) in livers of NASH patients versus lean controls and patients with simple steatosis.6 This finding was subsequently corroborated by Caballero et al., who not only identified increased FC, but also found increased hepatic sterol regulatory elementbinding protein (SREBP)-2 transcript expression in NASH patients.7 In order to fully understand the role of cholesterol in NASH, the origin of increased hepatic cholesterol needs to be addressed. As a key transcription factor regulating cellular cholesterol uptake, synthesis, biotransformation, and excretion, SREBP-2 may hold the key to understanding how cholesterol fits into the bigger scheme of NASH. SREBP-2 was discovered in 1993 by the Nobel Prize-winning Goldstein and Brown research group, who identified it as the third SREBP (the others are SREBP-1a and SREBP-1c). SREBP-2 is expressed as a 125-kDa inactive precursor protein, comprised of a –NH2 transcription factor domain and a –COOH regulatory domain.8 Nascent SREBP-2 localizes to the endoplasmic reticulum (ER) membrane, with both terminal domains facing the cytosol in a hairpin fashion. It binds to SREBP cleavage activating protein (SCAP) via the –COOH terminal regulatory region. Under conditions of low intracellular cholesterol, SCAP acts as a chaperone responsible for translocating SREBP-2 to the Golgi apparatus, where two proteases, site-1 serine protease and site-2 metalloproteinase, cleave off a 68-kDa SREBP-2 transcriptionregulatory fragment. The latter truncated protein enters the nucleus, where it is responsible for transcriptional regulation of a number of target genes involved in cholesterol biosynthesis, the most important being 3-hydroxy-3-methyl-glutaryl-CoA reductase (HMGR) (the rate-limiting step), and low-density lipoprotein (LDL) receptor (LDLR), the plasma membrane transporter responsible for receptor-mediated endocytosis of serum LDL cholesterol. Active SREBP-2 fragments are also able to increase the expression of SREBP-2, resulting in a feed-forward mechanism. Conversely, in response to heightened cellular cholesterol levels, the sterol-sensing domain of SCAP changes conformation and binds to insulin-induced gene-1 and -2; this retains the SREBP-2/ SCAP complex within the ER.9 Impaired translocation to Golgi inhibits SREBP-2 cleavage, leaving the parent protein inactive. SREBP-2 therefore functions as a cholesterol-sensitive critical regulatory checkpoint, responsible for controlling intracellular cholesterol homeostasis. More recently, our understanding of SREBP-2 function has been expanded by the identification of a genetic locus within the SREBP-2 encoding region, which codes for a highly-conserved microRNA (miR), miR-33.10 miR-33 functionally inhibits cellular cholesterol export via ATP-binding cassette protein-A1, as well as mitochondrial FFA b-oxidation through the suppression of several enzymes; the latter include hydroxyacyl-CoA dehydrogenase, 3-ketoacyl-CoA, thiolase/enoyl-CoA hydratase bsubunit, carnitine palmitoyltransferase 1A, and carnitine O-

Journal of Gastroenterology and Hepatology 26 (2011) 789–795 © 2011 Journal of Gastroenterology and Hepatology Foundation and Blackwell Publishing Asia Pty Ltd

789

Editorials

DM Van Rooyen and GC Farrell

octanoyltransferase.10 These findings indicate that SREBP2expression has pleiotropic effects on cellular lipid homeostasis, affecting FFA oxidation, as well as the well-known effects on cholesterol turnover. Physiologically, and from a teleological point of view, the activation of SREBP-2 under low sterol conditions perfectly suits the retention of intracellular cholesterol, while decreased b-oxidation of FFA increases the availability of longchain fatty acids needed to form cholesterol esters (CE); CE are the preferred and “safer” storage form of cholesterol within cells and cell membranes. However, if as demonstrated in human NASH studies (and our own unpublished data),11 SREBP-2 is activated within lipid-laden hepatocytes, this would constitute a

highly inappropriate time to promote cholesterol influx, or to inhibit cholesterol turnover/efflux and FFA catabolism. The net effect of these events would further exacerbate the accumulation of at least three potentially lipotoxic hepatic lipids (FC, FFA, diacylglycerol), potentially contributing to the pathogenic mechanism of the NASH phenotype. miR-122, the most abundant hepatic miR, has been found to be strongly downregulated in NASH patients.12 Further, replicating miR-122 suppression in mice significantly increased SREBP-2 and HMGR expression in both in vivo and in vitro systems.12 It has subsequently been hypothesized that miR-122 acts as a negative post-transcriptional regulator of SREBP-2 activation, by stabiliz-

Figure 1 Activation of hepatocellular sterol regulatory element binding protein-2 (SREBP-2) through inflammatory and insulin responsive pathways: implications for non-alcoholic steatohepatitis. Nuclear SREBP-2 is upregulated by inflammatory cytokines (interleukin [IL]-1b, IL-6, tumor necrosis factor-a), as well as hyperinsulinemia and low microRNA (miR)-122 levels. Increased nuclear SREBP-2 induces the expression of low-density lipoprotein receptor (LDLR) and HMG-CoA reductase (HMGR), which increases low-density lipoprotein (LDL) cholesterol uptake and nascent cholesterol biosynthesis. SREBP-2 also increases miR-33, leading to the suppression of both mitochondrial b-oxidation of free fatty acids (FFA) and cholesterol efflux. Collectively, these pathways induce hepatic cholesterol accumulation, which may trigger hepatocellular death and hepatic inflammation, resulting in macrophage and neutrophil recruitment, culminating in cytokine release, further exacerbating SREBP-2 activation. CE, cholesterol esters; FC, free cholesterol.

790


DM Van Rooyen and GC Farrell

ing the inactive SREBP-2 protein. Therefore, the suppression of miR-122 likely results in increased SREBP-2 activation, which constitutes an additional pathway of increased SREBP-2 activation in human NASH. In this edition of The Journal of Gastroenterology and Hepatology, Zhao and colleagues describe the effect of inflammation on hepatic SREBP-2, and subsequent LDLR and HMGR expression in both in vivo and in vitro systems.13 Chronic low-grade inflammation and cytokine stimulation (interleukin [IL]-1b and IL-6) triggered SREBP-2 activation and increased hepatic LDLR receptor expression, in addition to enhancing HMG-CoA reductase activity, culminating in increased hepatic cholesterol (free and esterified) levels. Furthermore, chronic low-grade inflammation was found to interfere with the feedback mechanism responsible for decreasing nuclear SREBP-2 in the presence of heightened intracellular FC levels, thereby deregulating cholesterol homeostasis. Subcutaneous casein injections in C57Bl/6J mice induced both IL-6 and serum amyloid A protein (similar to human C-reactive protein), simultaneously increasing LDLR and HMGR expression, in addition to nuclear SREBP-2 levels. Similar results were obtained in human HepG2 cells. Following IL-1b and IL-6 exposure, expression of LDLR and HMGR increased. Both in vivo and in vitro experimental approaches were found to increase hepatic intracellular cholesterol and esterified lipids (triglycerides and cholesteryl esters), confirming that the molecular changes in cholesterol uptake and biosynthesis did lead to changes in lipid molecule accumulation. Further compounding SREBP-2 involvement in NASH, hyperinsulinemia, a common component of the pathophysiology of NASH that stems from insulin resistance, also increases hepatic SREBP-2 in mice by an intracellular signaling pathway, in a manner that is thought to involve extracellular signal-regulated kinases (ERK)-1 and ERK-2.14 These findings provide a potential explanation for the close relationship between insulin resistance (and resulting hyperinsulinemia), dyslipidemia, and atheroma, and therefore, why NASH is such a strong risk factor for cardiovascular disease. In addition, there is a deepening relationship between metabolic factors and cytokines, as mediators of both insulin resistance and hepatic lipid metabolism. For instance, macrophage inflammatory protein-1, which may be related to adipose and liver inflammation in metabolic syndrome, is known to activate the lipogenic transcription factor SREBP-1, at least in vitro, thereby potentially aggravating hepatic steatosis. Further, tumor necrosis factor-a and IL-6, circulating products of adipose macrophages, have been implicated in the pathogenesis of hepatic insulin resistance, for example, by induction of suppressor of cytokine signaling-3, which blocks insulin receptor signaling pathways. Interestingly, IL-1b and IL-6, among others, have all been shown to activate the ERK1/2 pathway (which is also activated by the insulin receptor), suggesting that this pathway may be involved in the inflammation-dependent activation of SREBP-2.15 Collectively, these findings suggest that numerous factors pertinent to NASH, namely inflammatory cytokine signaling, hyperinsulinemia, and miR dysregulation, are capable of increasing hepatic SREBP-2 (Fig. 1). In the process, this increases hepatic intracellular cholesterol levels. As genetic and environmental determinants of NAFLD and NASH pathogenesis are currently being clarified, it will be important to understand how such factors

Editorials

influence expression of molecules that sit at the cross-roads of NASH as an inflammatory, metabolic disorder. SREBP-2 is clearly now one such “suspect of interest.” The question still remains as to whether increased hepatic cholesterol levels, either or both FC and oxysterol metabolites of cholesterol, actually contribute to NASH pathogenesis, although in LDLR gene-deleted mice, cholesterolladen macrophages are quite capable of causing liver inflammation when passaged into naïve animals.16 We have noted that rodents with metabolic syndrome “appear to develop NASH as a result of hepatic cholesterol accumulation” (Mr Derrick Van Rooyen, unpubl. data, 201111). Whatever the outcome of these and similar studies, the delineation of how SREBP-2 fits into the broader context of NASH should contribute importantly to understanding the convergence of metabolic and inflammatory pathways in this pathologically- and clinically-progressive form of NAFLD.

Acknowledgments This research was supported by project grant 585411 of the Australian National Health and Medical Research Council (NHMRC), and DVR is supported by an NHMRC scholarship 585539.

References 1 Day CP. NASH-related liver failure: one hit too many? Am. J. Gastroenterol. 2002; 97: 1872–4. 2 Neuschwander-Tetri BA. Hepatic lipotoxicity and the pathogenesis of nonalcoholic steatohepatitis: the central role of nontriglyceride fatty acid metabolites. Hepatology 2010; 52: 774–88. 3 Larter CZ, Yeh MM. Animal models of NASH: getting both pathology and metabolic context right. J. Gastroenterol. Hepatol. 2008; 23: 1635–48. 4 Yamaguchi K, Yang L, McCall S et al. Inhibiting triglyceride synthesis improves hepatic steatosis but exacerbates liver damage and fibrosis in obese mice with nonalcoholic steatohepatitis. Hepatology 2007; 45: 1366–74. 5 Neuschwander-Tetri BA. Nontriglyceride hepatic lipotoxicity: the new paradigm for the pathogenesis of NASH. Curr. Gastroenterol. Rep. 2010; 12: 49–56. 6 Puri P, Baillie RA, Wiest MM et al. A lipidomic analysis of nonalcoholic fatty liver disease. Hepatology 2007; 46: 1081–90. 7 Caballero F, Bataller R, Lacy A, Fernandez-Checa JC, Caballeria J, Garcia-Ruiz C. Enhanced free cholesterol, SREBP-2 and StAR expression in human NASH. J. Hepatol. 2009; 50: 789–96. 8 Hua X, Yokoyama C, Wu J et al. SREBP-2, a second basic-helix-loop-helix-leucine zipper protein that stimulates transcription by binding to a sterol regulatory element. Proc. Natl. Acad. Sci. USA 1993; 90: 11603–7. 9 Radhakrishnan A, Goldstein JL, McDonald JG, Brown MS. Switch-like control of SREBP-2 transport triggered by small changes in ER cholesterol: a delicate balance. Cell Metab. 2008; 8: 512–21. 10 Horie T, Ono K, Horiguchi M et al. MicroRNA-33 encoded by an intron of sterol regulatory element-binding protein 2 (Srebp2) regulates HDL in vivo. Proc. Natl. Acad. Sci. USA 2010; 107: 17321–6. 11 Van Rooyen DM, Larter CZ, Yeh MM, Haigh WG, Teoh N, Farrell GC. Dietary cholesterol modulates severity of liver injury in mice with metabolic-syndrome related NASH. J. Gastroenterol. Hepatol. 2010; 25: A5. 12 Cheung O, Puri P, Eicken C et al. Nonalcoholic steatohepatitis is associated with altered hepatic MicroRNA expression. Hepatology 2008; 48: 1810–20.


791

Editorials

JO Park and MM Yeh

13 Zhao L, Chen Y, Tang R et al. Inflammatory stress exacerbates hepatic cholesterol accumulation via increasing cholesterol uptake and de novo synthesis. J. Gastroenterol. Hepatol. 2011; 26: 875–83. 14 Xie X, Liao H, Dang H et al. Down-regulation of hepatic HNF4alpha gene expression during hyperinsulinemia via SREBPs. Mol. Endocrinol. 2009; 23: 434–43. 15 Gao B, Calhoun K, Fang D. The proinflammatory cytokines IL-1beta and TNF-alpha induce the expression of Synoviolin, an E3 ubiquitin ligase, in mouse synovial fibroblasts via the Erk1/2-ETS1 pathway. Arthritis Res. Ther. 2006; 8: R172. 16 Bieghs V, Wouters K, van Gorp PJ et al. Role of scavenger receptor A and CD36 in diet-induced nonalcoholic steatohepatitis in hyperlipidemic mice. Gastroenterology 2010; 138: 2477–86, 2486.e1–3.

Clinical significance and implication of neoangiogenesis in hepatocellular carcinoma James O Park* and Matthew M Yeh

jgh_6681

792..798

†

Departments of *Surgery and †Pathology, University of Washington School of Medicine, Seattle, Washington, USA

See article in J. Gastroenterol. Hepatol. 2011; 26: 866–874. Hepatocellular carcinoma (HCC), the most common primary liver malignancy, with over 660 000 new cases and 630 000 resultant deaths estimated in 2009, is the sixth most prevalent cancer and the third leading cause of cancer-related deaths worldwide.1 The lethality of HCC is apparent in its equal annual incidence and mortality, and the dismal 6–8-month median survival without treatment.2 Sorafenib, a small molecule tyrosine kinase inhibitor of Raf and intracellular vascular endothelial growth factor (VEGF) receptor,3 as the only current “standard therapy” for non-surgical, non-ablatable cases, has a 2% objective response, and improves overall survival by a mere 2–3 months.4 When detected early, surgical resection, liver transplantation, and ablative therapies can be employed, achieving 5-year survival rates up to 75%.5 However, only 10–15% of patients present with localized disease, limiting broad application. Catheter-based chemoembolization and radioembolization therapies prolong survival in selected cases, but their role is largely palliative.6 Thus, novel effective therapies are needed for this global health crisis. HCC is one of the most vascular solid tumors, characterized by its propensity for vascular invasion and high metastatic potential. It is responsible for the high rate of early postoperative recurrence in the liver remnant or distant sites following resection. Angiogenesis plays a crucial role in all stages of tumor development: Accepted for publication 24 January 2011. Correspondence Dr Matthew M Yeh, Department of Pathology, University of Washington School of Medicine, 1959 NE Pacific Street, NE140D, Box 356100, Seattle, WA 98195, USA. Email: [email protected]

792

growth, invasion, and metastasis. For this reason, assessment of tumor neovascularity might provide additional prognostic information regarding the tumor biology, risk of metastasis or recurrence, and resultant patient survival.7 Thus, it might be useful in guiding the selection of HCC patients at high risk of treatment failure for neo/adjuvant therapies. In this issue of the Journal of Gastroenterology and Hepatology, Chen et al. report that the morphological patterns of microvessel formation in HCC correlate with patient outcome following surgical resection.8 They observed that the sinusoidal type of neovasculature, rather than the capillary type, was predominantly encountered in patients with larger tumors, and predicted significantly worse disease-free and overall survivals in these patients, to whom adjuvant therapy would be most applicable., The most established reported indices of angiogenic activity in HCC are: (i) measurement of microvessel density; (ii) quantification of tumor and circulating levels of angiogenic factors; and (iii) evaluation of tumor vascularity demonstrated on angiographic imaging. Microvessel density (MVD) is determined by counting neovasculature in tumor tissue sections after immunohistochemical staining with one or more endothelial cell markers, such as CD34, CD31, von Willebrand factor, and CD105. Controversy remains regarding which marker is most effective and comprehensive, exact methods of assessment, and at what size or stage of the tumor MVD is predictive of outcome. However, there appears to be general consensus that a correlation between high MVD and poor outcomes exists, and that MVD is an independent prognostic factor of overall and disease-free survival after the resection of HCC.9,10 As there is considerable variation in outcomes in patients of the same stage based on the current American Joint Committee on Cancer staging system for HCC, this information might be used to guide the addition of adjuvant therapy in patients with high MVD tumors. An inherent issue in MVD measurements is subjective variability and bias, even if the panel of immunohistochemical markers to apply, and the tumor region to analyze, and techniques used for measurement are standardized. MVD is most accurately employed on the whole-tumor surgical specimen after resection, limiting its predictive value to the adjuvant setting. Circulating levels of various angiogenic factors, for example, VEGF,11 basic fibroblast growth factor (FGF),12 and angiopoietin213 have been found to be significantly elevated in HCC patients compared with normal control or patients with benign chronic liver disease. Further, correlation has been found with higher levels of such factors and advanced tumor stage, as well as poor prognostic features, such as vascular invasion, with resultant early recurrence following resection. These serum or plasma measurements have been demonstrated to be reasonable surrogate markers of tumor overexpression of these factors. Among them, most attention has been focused on VEGF, with suggestions of ELISA-based detection of plasma VEGF levels as a biomarker to complement current screening protocols.14 Compared to MVD measurements, circulating angiogenic factor levels are less subject to observer interpretation and can be assessed prior to planned surgical resection, affording an opportunity for consideration of neoadjuvant therapy. Another non-invasive method of assessing tumor angiogenesis is angiographic imaging of tumor vascularity. Enhancement patterns on dynamic contrast-enhanced magnetic resonance imaging have been demonstrated to be influenced by tumor angiogenesis,
