Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 Jul 6.
Published in final edited form as: Cell Metab. 2011 Jul 6;14(1):91–103. doi: 10.1016/j.cmet.2011.03.025

Heat shock transcription factor 1 is a key determinant of HCC development by regulating hepatic steatosis and metabolic syndrome

Xiongjie Jin 1,2,3, Demetrius Moskophidis 1,2,4, Nahid F Mivechi 1,2,3,4
PMCID: PMC3214631  NIHMSID: NIHMS303995  PMID: 21723507

Abstract

Hepatocellular carcinoma (HCC) occurrence and progression is linked tightly to progressive hepatic metabolic syndrome associated with insulin resistance, hepatic steatosis, and chronic inflammation. Heat shock transcription factor 1 (HSF1), a major transactivator of stress proteins, increases survival by protecting cells against environmental stressors. It has been implicated in the pathogenesis of cancer, but specific mechanisms by which HSF1 supports cancer development remain elusive. We propose a pathogenic mechanism whereby HSF1 activation promotes growth of pre-malignant cells and HCC development by stimulating lipid biosynthesis and perpetuating chronic hepatic metabolic disease induced by carcinogens. Our work shows that inactivation of HSF1 impairs cancer progression, mitigating adverse effects of carcinogens on hepatic metabolism by enhancing insulin sensitivity and sensitizing activation of AMP-activated protein kinase (AMPK), an important regulator of energy homeostasis and inhibitor of lipid synthesis. HSF1 is a potential target for the control of hepatic steatosis, hepatic insulin resistance, and HCC development.

Introduction

Cancer arises through a multi-stage process involving selection of genetic and epigenetic alterations, allowing evasion of mechanisms that limit cell proliferation or cell death (Hanahan and Weinberg, 2011; Luo et al., 2009). In addition to genetic changes, tumor formation and progression is regulated by the tumor microenvironment and extracellular signaling pathways. In particular, inflammation through aberrant secretion of cytokines and chemokines and other pro-growth factors supports proliferation and expansion of pre-malignant cells (Grivennikov et al., 2010; Mantovani et al., 2008). Moreover, metabolic changes driven by the tumor microenvironment and oncogene activation or tumor suppressor gene loss also confer a selective advantage for survival and proliferation of cancer cells (Levine and Puzio-Kuter, 2010; Siegel and Zhu, 2009). Notably, altered metabolism provides cancer cells not only with energy but also serves to support synthesis of fatty acids, membrane lipids and proteins that are required for rapidly dividing cells (Menendez and Lupu, 2007). Lastly, equally important for survival of a transformed cell and tumor formation and progression is the contribution of the proteostatic system, which involves pathways that control protein synthesis, quality control, and trafficking (Hotamisligil, 2010; Powers et al., 2009).

HCC induced within an established background of chronic liver disease and cirrhosis after long latency (20-40 years) is one of the most prevalent malignant cancers worldwide. The transformation process underlying HCC development in humans, and many risk factors, including alcoholic use, chronic infections with hepatitis B and C viruses, and nonalcoholic steatohepatitis (NASH), which is a more severe form of nonalcoholic fatty liver disease (NAFLD), are well defined (El-Serag and Rudolph, 2007; Farazi and DePinho, 2006; Siegel and Zhu, 2009). Notably, NAFLD, a hepatic manifestation of metabolic syndrome, affects nearly 25% of the US population and is a fast-growing medical condition in terms of incidence due to the fact that obesity and metabolic syndrome are becoming epidemics worldwide. Despite recent progress, HCC is still a disease with poor outcome and limited therapeutic options (Spangenberg et al., 2009).

HCC development is well defined at the cellular level, beginning with hyperplastic and dysplastic nodule formation in the cirrhotic liver, and ultimately malignant conversion of a small number of lesions. In contrast, the exact molecular mechanisms accounting for hepatitis-driven cellular transformation are less well understood. Supported by mouse genetic studies using the pro-carcinogen diethylnitrosamine (DEN) or elimination of functional NF-κB signaling, it was thought that chronic liver injury caused by oxidative stress triggers inflammation, which drives the compensatory proliferation of intact hepatocytes and liver progenitors leading to initiation and progression of HCC development (Luedde et al., 2007; Maeda et al., 2005). These observations suggest that inhibition of compensatory liver cell proliferation by modulating the inflammatory response may be a realistic approach to reduce the risk of HCC.

Another critical hallmark of the pathogenesis of cancer in general, and HCC in particular, is the contribution of metabolic processes functionally related to de novo fatty acid synthesis and glucose disposal and utilization. In fact, dysregulated metabolism and exacerbated lipid biosynthesis and accumulation emerge early in cancer development and accelerate cell growth and transformation (Menendez and Lupu, 2007; Siegel and Zhu, 2009). Recent reports indicating that lipogenesis in connection with the lipolytic release and remodeling of free fatty acids promote high level of malignancy, suggest substantial crosstalk between lipid metabolic pathways in cancer cells (Nomura et al., 2010). Elucidating metabolic pathways that are vital to the cancer pathogenesis and specifically assess to what extent lipid accumulation in the injured liver may contribute to HCC is of particular importance as an urgent need for development of novel treatments of this disease.

Here we investigated a possible causal relationship between alterations in hepatic energy and lipid metabolism, protein homeostasis, metabolic syndrome, and HCC development. We describe that HSF1, a key regulator of heat shock protein (HSP) response, which is a protective mechanism conserved among all eukaryotes and enhances survival and longevity of organisms (Akerfelt et al., 2010; Hsu et al., 2003), is also critically involved in HCC development by promoting metabolic adaptive mechanisms for cancer development, including insulin signaling and hepatic lipid accumulation.

Results

Hsf1 deletion suppresses chemical-induced liver tumorigenesis

To investigate the specific contribution of HSF1 to cancer pathogenesis we conducted studies with pro-carcinogen DEN, which induces HCC in 100% of male mice within 5 to 7 months. DEN-induced HCC is a valid model for human HCC and reproduces some of the key pathological and histological features of this cancer (Farazi and DePinho, 2006; Lee et al., 2004). Mice of three genotypes, hsf1-/-, hsf1+/-, and hsf1+/+, were injected with DEN on postnatal day 14, and the tumor burden (number and size) was determined 7 and 10 months later (Fig. 1A-B). While DEN-treated hsf1-/- mice were free of liver tumors and signs of apparent liver pathology, all hsf1+/+ control littermates, referred as wild type (WT), developed multiple typical HCCs that contained clusters of cells staining strongly positive for alpha-fetoprotein (α-FP) (Fig. 1D), an established liver tumor marker (Ariel et al., 1998). Western blot analysis at different ages confirmed α-FP re-expression in livers of WT mice bearing tumors, whereas it was absent in tumor-free hsf1-/- mice. Histological examination revealed the presence of extensive liver pathology associated with severe steatosis in DEN-treated WT mice (Fig. 1C-D). Strikingly, the number of tumors detected in hsf1+/- mice and their size were dramatically reduced compared to wild-type controls, indicating dose-dependent effect of HSF1. Thus, HSF1 is required for efficient HCC induction and growth in response to DEN treatment.

Figure 1. Deletion of hsf1 inhibits DEN-induced HCC formation.

Figure 1

Open in a new tab

(A) Representative macroscopic pictures of livers from wild-type (WT) and hsf1-/- mice 7 and 10 months after DEN injection. Arrowhead indicates tumor nodules. m, month.

(B) Quantification of liver tumors in WT (n=15), hsf1+/- (n=7), and hsf1-/- (n=16) mice 7 months after DEN injection. Bars represent mean ± SD. Statistical significance is indicated (* p<0.05, ** p<0.01, *** p< 0.001).

(C) Histological analysis (H&E) of livers from WT, hsf1+/- and hsf1-/- mice 7 months after DEN injection. Representative sections show no steatosis or HCC in hsf1-/- mice. WT and hsf1+/- mice display hepatic steatosis and HCC. Arrowheads depict HCC border (scale bar 100 μm).

(D) Immunohistochemical analysis of liver sections for alpha-fetoprotein (α-FP) expression from WT and hsf1-/- mice at the indicated time points after DEN-treatment at 14 days of age (0). Positive staining is present in livers of both genotypes during early postnatal period that decreases and disappeared around 1 month of age. α-FP was prominently re-expressed within HCCs developed in DEN-treated WT mice (arrows). Western blot analysis of α-FP expression in liver extracts from WT or hsf1-/- mice at indicated time after DEN-treatment (right panel).

Hsf1 deletion does not affect DEN-induced liver damage and compensatory proliferation during early tumor promotion

To further investigate the mechanisms for the remarkable resistance of hsf1-/- mice to HCC, we compared early (short-term) and late effects of DEN injection in WT and hsf1-/- mice. Short-term effects of DEN on liver integrity are marked by hepatocyte death, increased cytokine production and compensatory hepatocyte proliferation and are proposed as good predictors of the outcome of the disease (Farazi and DePinho, 2006; Karin and Greten, 2005). WT and hsf1-/- mice displayed comparable hepatic injury after DEN injection (short-term) showing similar levels of ALT and AST release and AST/ALT ratio, and no significant differences in the liver cell proliferation and apoptosis (Fig. 2A-C). As expected, WT mice bearing (late stage) tumors, exhibited sustained increase in cellular proliferation, apoptosis (and necrosis), and high levels of serum ALT and AST and decreased ratio of AST/ALT, a useful parameter for non-alcoholic fatty liver disease (Sorbi et al., 1999). In contrast, hsf1-/- mice recovered remarkably from the early effects of DEN injection as revealed by normalization of the liver pathology (Fig. S1A). These differential effects on liver pathology appeared at 10 days and became more pronounced at 1 and 7 months post DEN injection.

Figure 2. Comparable levels of liver damage and compensatory proliferation during early stage of DEN-induced HCC between WT and hsf1-/- mice.

Figure 2

Open in a new tab

(A, B) Ki67 and TUNEL staining of liver sections from WT or hsf1-/- mice at the indicated time points after DEN injection (left panels). Statistical analysis of Ki67+ proliferating cells or TUNEL+ apoptotic hepatocytes per high-power-field (HPF) is presented (right panels). Scale bar, 20μm (Ki67) and 10μm (TUNEL). Bars are mean ± SD; (n=5 mice per group).

(C) Serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels. Bars are mean ± SD; (n=5 mice per group).

(D) Expression of cytokines in livers of DEN-treated mice by real-time RT-PCR (upper panels). Serum cytokine levels of untreated or DEN-injected mice measured by ELISA (lower panels). Bars are mean ± SD; (n=5 mice per group).

For all panels statistical significance is indicated (* p<0.05, ** p<0.01, *** p<0.001). d, day and m, month. 0, WT or hsf1-/- mice at 14 days of age without DEN treatment.

Production of cytokines like IL-6, IL-1β, and TNF-α and hepatocyte growth factor (HGF), by macrophages and Kupffer cells triggered by substrates released from necrotic cells following DEN injection appear to have a critical role in DEN-induced HCC (Maeda et al., 2005). Indeed, activation of several cytokines in early stage of DEN treatment were dramatically induced, but overall detectable levels were comparable between WT and hsf1-/- mice (Fig. 2 D). This initial pro-inflammatory response persisted at a lower steady-state level in WT mice through the progression stage, however the cytokine production in hsf1-/- mice was not sustained and decreased to barely detectable levels at 1 and 7 months of age. Transcriptome analysis of the livers from hsf1-/- versus WT mice, without treatment, and 4 hours and 30 days after DEN injection, revealed that expression of several genes associated with inflammation and immune response were up-regulated. However, no overt differences in the overall gene expression pattern were observed between the genotypes (data not shown). This indicates that tumor inhibition because of hsf1 deficiency cannot be attributed to the altered inflammatory response during tumor initiation phase. Histological analysis also confirmed that although the early effects of DEN (1 and 2 days) with respect to macrophage infiltration detected by CD11b staining, were indistinguishable between the genotypes, hsf1-/- mice exhibited significantly fewer infiltrating cells in liver at 10 days and 1 or 7 months after DEN administration (Fig. S1B). Together, these results suggest that HSF1 function is critical during tumor progression and hsf1 deficiency does not have apparent effects on the initial DEN-induced hepatocyte death, cytokine secretion, and compensatory proliferation.

Hsf1 deletion prevents DEN-induced hepatic steatosis

A causal link between altered hepatic metabolism and cancer progression has been proposed (Siegel and Zhu, 2009). Liver damage induced by DEN administration is accompanied by hepatic steatosis characterized by lipid deposition in hepatocytes. Histological examination of livers at 1 and 2 days after DEN injection showed focal lipid accumulation by Oil-Red-O staining (Fig. 3). The extent of lipid deposition was comparable between genotypes and subsided by 10 days post-DEN injection. However, starting at 2 months, livers of WT mice displayed signs of lipid deposition that increased in intensity during tumor progression and led to the development of chronic liver disease with extensive steatosis at 7 and 10 months of age. In contrast, hsf1-/- mice showed dramatic inhibition of hepatic lipid accumulation and fatty liver disease (Fig. 3A). Consistently levels of triglycerides and non-esterified fatty acids (NEFA) in circulation and liver were significantly lower in hsf1-/- mice (Fig. 3B). Similarly, blood levels of total cholesterol, low-density lipoprotein (LDL/VLDL) and high-density lipoprotein (HDL) were reduced in hsf1-/- mice in both DEN-treated and untreated groups when compared to WT mice (Fig. 3C). However, the HDL/total cholesterol ratio was the same between the genotypes. These results suggest that HSF1 activation is critical in the development of HCC by potentiating DEN-induced hepatic lipid biosynthesis and deposition.

Figure 3. Hsf1 deletion prevents chronic hepatic steatosis induced by DEN treatment.

Figure 3

Open in a new tab

(A) Representative liver sections stained with Oil red-O at indicated times after DEN injection. Counterstained with hematoxylin. 0, indicates WT or hsf1-/- mice at 14 days of age without DEN treatment. Scale bar, 10μm.

(B) Levels of triglycerides and non-esterified fatty acids (NEFA) in circulation and livers of untreated (Control) or DEN-treated mice at 7 months of age. Bars are mean ± SD; (n=5 mice per group). Statistical significance is indicated * p<0.05, ** p<0.01.

(C) Levels of cholesterol in HDL or LDL/VLDL and total cholesterol in the sera of untreated (control) or DEN-treated mice at 7 months of age. Bars are mean ± SD; (n=5 mice per group). Statistical significance is indicated * p<0.05, ** p<0.01.

Hsf1 regulates lipid metabolism

To investigate how HSF1 ablation counteracts changes in hepatic lipid metabolism, we examined the effects of age on growth, fat content and lipid metabolism in hsf1-/- and WT mice (Fig. 4). The rate at which hsf1-/- mice increased body weight on normal chow diet (CD) was significantly reduced compared with WT mice; at 9 months of age the difference in body-weight was 25%. Moreover, WT mice displayed further increase in body-weight upon DEN treatment when compared to untreated controls. No significant change in body-weight was observed between DEN-treated or untreated hsf1-/- mice (Fig. 4A). Dissection of hsf1-/- mice revealed marked reduction in epididymal white adipose tissue (WAT). When normalized to body-weight, total WAT was significantly reduced in hsf1-/- compared to WT mice (Fig. 4B). This decrease was specific for fat, because the weight of other organs, like kidney and heart, normalized to body-weight did not show significant differences. Notably, development of hepatic tumors in WT mice accounts for the increase in liver weight in DEN-treated WT mice as compared to hsf1-/- mice. Since food intake measured at 3 and 7 months of age was comparable between the genotypes, decrease in fat accumulation was not due to hypophagia (data not shown). Additionally, histological analysis of epididymal fat cells at 7 months of age were consistently smaller with an average 70% decrease in size in hsf1-/- mice compared to WT mice. Depression of adipose tissue in hsf1-/- mice was associated with strong down-regulation of adipose-specific genes like PPARγ1, PPARγ2, C/EBP, and FABP4 (αP2) (Fig. S2).

Figure 4. Loss of HSF1 protects from DEN-induced hepatic steatosis through inhibition of lipid synthesis.

Figure 4

Open in a new tab

(A) Growth curves of untreated or DEN-treated WT and hsf1-/- mice on standard chow diet (CD). Bars are mean ± SD; (n=10 mice per group). * p<0.05 for WT-DEN-treated compare to WT-untreated and ** p<0.01 for WT-untreated compared to hsf1-/- untreated mice.

(B) Weight of epididymal adipose tissue (WAT) of 6- or 10-month-old WT and hsf1-/- mice, untreated (Control) or DEN treated mice. Bars are mean ± SD; (n=10 mice per group).

(C) Real-time RT-PCR analysis of indicated genes in the liver of untreated or DEN-treated 7-month-old WT or hsf1-/- mice. Data are presented as relative levels of mRNA expression. Bars are mean ± SD; (n=5 mice per group). Sequences of the primers used are indicated in Table S1.

(D) Western blot analysis of liver extracts from 7-month-old untreated (-DEN, control) or DEN-treated (+DEN) WT and hsf1-/- mice using antibody against PPARγ. Quantification of protein expression level normalized to β-actin loading control, Bars are mean ± SD; (n=8 mice per group) (right panels).

For panels B and C, insert in the upper corner. For all panels statistical significance is indicated (* p<0.05, ** p<0.01, *** p<0.001).

The profound reduction in DEN-induced hepatic steatosis in hsf1-/- mice may reflect the flexibility in the metabolic system to efficiently compensate for enforced lipid accumulation by altering hepatic lipid synthesis, lipolysis, uptake, secretion, or β-oxidation of fatty acids. Therefore, we investigated whether hsf1 inactivation negatively affects de novo lipogenesis in the liver and thus inhibits hepatic fat accumulation (Fig. 4C-D). Expression of the lipogenic transcription factors, e.g., SREBP1c, ChREBP, LXRα or C/EBPβ, were not significantly different between normal (without DEN treatment) hsf1-/- and WT mice when measured at 7 months of age (Fig. 4C). No significant differences in C/EBPα or C/EBPβ protein levels were detected between the genotypes (data not shown). However, mRNA and protein levels of PPARγ1 or PPARγ2 were found to be significantly lower in liver of hsf1-/- compared to age-matched WT mice (Fig. 4C-D). This result was also confirmed in 1½ months old mice (data not shown). Of interest, significant increase in mRNA levels of these lipogenic transcription factors was evident in DEN-treated WT mice, further supporting the hypothesis that such changes regulate hepatic fat accumulation. However, HSF1 ablation in DEN-treated mice inhibited mRNA expression of SREBP1c, LXR, PPARγ1, and PPARγ2. The lack of HSF1 also compromised DEN-induced mRNA levels of SREBP1c and PPARγ1 targets regulating lipid synthesis, including FAS, ACL, and SCD1, while the mRNA level of ACC1 remained unchanged. Interestingly, mRNA level of ATGL, a rate-limiting enzyme for triacylglycerol hydrolysis, was reduced in hsf1-/- mice providing evidence for inhibition of hepatic lipolysis. The hepatic expression of ApoB and ApoE genes involved in triglyceride secretion was not different between the genotypes of untreated mice. In contrast, DEN treatment induced expression of both genes that was more pronounced in hsf1-/- mice. Additionally, in WT mice, DEN robustly induced expression of CD36, a PPARγ1 target gene involved in fatty acid uptake. But lack of HSF1 entirely blunted this response. No differences in mRNA levels of the other genes related to lipid uptake, such as LPL, FABP1, and FATP, were observed under untreated or DEN-treated conditions. Finally, lack of HSF1 also impaired the expression of genes related to fatty acid oxidation such as PPARα, ACO, LCAD, or MCAD, which were robustly induced following DEN treatment in the WT mice. Particularly, expression of MCAD, a rate-determining enzyme in fatty acid oxidation, was markedly reduced by lack of HSF1, while ACO expression was slightly, but not significantly affected. Together, these data indicate that hsf1 deficiency prevents hepatic steatosis induced by DEN through down-modulation of genes regulating lipid synthesis, lipolysis and fatty acid uptake. It is probable that the observed reduction of fat synthesis is the reason for the decreased hepatic lipid oxidation in hsf1-/- mice.

Increased hepatic insulin sensitivity and glucose tolerance in hsf1-deficient mice

Insulin signaling is a critical component in glucose homeostasis and a major stimulus of hepatic metabolism (Herman and Kahn, 2006). Although no significant differences were seen in glucose tolerance test (GTT) between the genotypes (without DEN treatment) at 1 month of age, chow diet (CD) fed hsf1-/- mice assayed at 3 and 7 months of age exhibited a markedly improved glucose tolerance compared to WT mice (Fig. 5A). Of interest, DEN-treated hsf1-/- mice already exhibited a modest, but significantly improved, glucose clearance measured by GTT at 1 month of age, and this effect was more pronounced at 3 and 7 months of age (Fig. 5A). Insulin sensitivity, as measured in the insulin tolerance test (ITT), was also enhanced in hsf1-/- mice at 7 months of age. This effect was independent of DEN treatment (Fig. 5B). In addition, feeding blood glucose levels did not differ between hsf1-/- and WT mice assayed at different age. However, fasting glucose levels were slightly but reproducibly reduced in hsf1-/- mice compared to WT controls at 7 months of age (Fig. 5C). This effect was more pronounced in DEN-treated mice. Consistent with enhanced insulin sensitivity, we also noted that feeding plasma insulin levels in DEN-treated or untreated hsf1-/- mice were significantly reduced as compared to WT mice (Fig. 5D). Of interest, mRNA levels of key gluconeogenic enzymes, PEPCK and G6Pase were found to be lower in the liver of hsf1-/- mice, indicating inhibition in hepatic glucose production (Fig. 5E).

Figure 5. Insulin sensitivity and glucose tolerance is enhanced in hsf1 deficient mice.

Figure 5

Open in a new tab

(A) Glucose tolerance test (GTT) in 16-hr-fasted untreated (top panels), or DEN-treated (lower panels), wild-type (WT) (●) or hsf1-/- (○) mice. Bars are mean ± SD: (n=8 mice per group).

(B) Insulin tolerance test (ITT) in 16-hr-fasted untreated or DEN-treated WT (●) or hsf1-/- (○) mice, at 7 months of age. Bars are mean ± SD; (n= 8 mice per group).

(C) Blood glucose in fed (Fed) or 16-hr-fasted (Fast) untreated or DEN-treated mice at 1 or 7 months of age. Bars are mean ± SD; (n= 5 mice per group).

(D) Plasma insulin levels determined by ELISA in fed (Fed) or 16-hr-fasted (Fast) WT or hsf1-/- mice, untreated or DEN-treated, at the indicated ages. Bars are mean ± SD; (n= 5 mice per group).

(E) mRNA levels of G6Pase and PEPCK in the liver of CD (Fed) or fasted (Fast) mice, untreated or DEN treated, at 7 months of age, Bars are mean ± SD; (n= 4 mice per group). Sequences of the primers used are indicated in Table S1. For all panels statistical significance is indicated (* p<0.05, ** p<0.01, *** p<0.001).

To more sensitively test the positive effects of hsf1 ablation on insulin sensitivity, cohorts of hsf1-/- and WT mice were predisposed to diet-induced obesity and insulin resistance by feeding with a high fat diet (HFD) beginning at 4 weeks of age. Consumption of HFD during the 20-week feeding period was equivalent between the genotypes. As expected, WT mice maintained on HFD gained substantial weight as compared to CD-feeding. In contrast, hsf1-/- mice fed with CD or HFD had body-weight that did not significantly change after 10 weeks of age and was significantly lower as compared to WT mice (Fig. S3A-C). HFD increased fat mass and hepatic lipid accumulation in WT mice, but Hsf1-/- mice were remarkably resistant (Fig. S3D-E). In addition, CD- or HFD-fed hsf1-/- mice exhibited equivalent insulin sensitivity as indicated by glucose clearance in the GTT or ITT, which, however, was greater compared to WT mice (Fig. S3F-G). Consistent with these results, under HFD-feeding hsf1-/- mice maintained lower insulin and glucose levels compared to WT controls. In addition, while HFD produced a blunted insulin response in WT mice, the hsf1-/- mice exhibited normal insulin-stimulated AKT-signaling in liver and muscle (Fig. S3H-I). Taken together these results show that insulin sensitivity is increased as a result of hsf1 deficiency and is reflected in reduced hepatic lipid accumulation irrespective of CD or HFD feeding.

HSF1 deficiency enhances insulin-induced activation and signaling

Insulin is well known to stimulate glucose disposal by increasing glucose uptake in peripheral tissues and also driving de novo lipogenesis and steatosis, and in the fed state it is a major anti-lipolytic hormone (Duncan et al., 2007; Kersten, 2001). In this context, one surprising result of our study was that hsf1 deficiency increased insulin sensitivity but decreased de novo lipogenesis, thus preventing hepatic steatosis. To further explore this issue we investigated the activation of insulin receptor (IR) and downstream signaling pathways in the liver of hsf1-/- versus WT mice at 7 months of age. Interestingly, we detected a marked increase in mRNA and protein levels of IRβ and associated insulin receptor substrates (IRS), IRS1 and IRS2, in livers of untreated or DEN-treated hsf1-/- mice compared to WT counterparts (Fig. 6A and Fig. S4). In addition, insulin-like growth factor-I receptor (IGF-IR) level in the liver remained unaltered (Fig. S4).

Figure 6. Enhanced insulin receptor (IR) signaling in the absence of HSF1.

Figure 6

Open in a new tab

(A) Immunoblot analyses of basal and insulin-stimulated total and p-Y-IR, total and p-Ser IRS1, and total IRS2 levels in livers of 7-months old WT and hsf1-/- mice. Mice were fasted for 16 hrs, and injected with saline or insulin 0.75 mU/g i.p. Liver extracts prepared from individual mice 10 min after insulin injection and blots were quantified by densitometry. Level of phosphorylated protein normalized to total level was expressed as relative fold-increase to control (WT without insulin) that was arbitrarily set at 1 (100%). Bars are mean ± SD; (n= 5 mice per group).

(B) Liver extracts prepared 10 min after insulin or saline injection from individual WT or hsf1-/- mice were immunoprecipitated using IRS1 antibody and processed for immunoblot analysis with antibodies to p-Tyr, IRS-1, or PI3K p85. ns, non-specific antibody (negative control) (left panel). Ratio of p-Tyr (P-Y) IRS1 to total IRS1 signal intensity is indicated.

Whole liver extracts of mice in the left panel were probed with antibody to p-Tyr, IRS1, PI3K p85 or β-actin. Relative molecular sizes corresponding to p-Tyr IRS1 and p-Tyr IR are indicated (right panel).

(C) Activation of AKT and mTOR signaling in the liver extracts from panel A was examined by immunoblot analysis using indicated antibodies. Quantification of blots is presented (right panels). Bars are mean ± SD; (n= 5 mice per group).

For all panels: Statistical significance is indicated (* p<0.05, ** p<0.01, *** p<0.001). Bars indicate WT (filled) or hsf1-/- (open) mice.

Next we evaluated that insulin induced phosphorylation of IRβ tyrosine (p-Y) or IRβ subunit (Y1158/Y1162/Y1163) was enhanced in hsf1-/- mice compared to WT controls (Fig. 6A-B). In addition, insulin stimulation resulted in a significant increase in p-Y-IRS1 in livers of both genotypes, but hsf1-/- mice exhibited much higher levels of total and p-Y-IRS1 (Fig. 6B). Next, we examined the insulin stimulated phosphorylation of IRS1 (p-IRS1) at putative inhibitory sites at residue S612 and S636/639. No significant difference was observed in relative p-IRS1(S612) levels, but interestingly hsf1-/- mice exhibited a reduced basal and insulin-stimulated ratio of p-IRS1(S636/S639)/total IRS1 compared to WT controls. Consistent with the above findings, hsf1-/- mice displayed increased p-Y-IRS1 and associated p85 levels compared to WT counterparts (Fig. 6B). These results suggest that the improved insulin sensitivity in hsf1-/- mice can be ascribed primarily to increased IRβ and IRS levels rather than to a particular pattern of IRS protein modification by phosphorylation. Expectedly, insulin stimulated AKT(S473) and to lesser extent AKT(T308) phosphorylation was significantly higher in the liver of hsf1-/- mice compared to WT controls (Fig. 6C). The p-AKT increase in the liver of mutant mice was accompanied by an enhanced increase in phosphorylation and inactivation of AKT substrates Foxo1 (S256 and S319), GSK3α (S21), GSK3β (S9) and AS160 (T612) (regulator of GLUT4 trafficking predominantly expressed in muscles and adipose tissues) compared to WT controls. In comparison hsf1-/- mice displayed slightly better response to injected insulin with respect to GSK3 phosphorylation, although this did not reach to a significance. The dramatic increase in insulin-stimulated p-AKT (S473) in livers of hsf1-/- mice did not yield the expected level of mTORC1 activation revealed by reduced phosphorylation of p70SK1 (T389) and rS6 (S240/S241), which are major mTOR targets (Fig. 6C). Moreover, the level of p-mTOR (S2448) mediated by p70SK1 (Foster and Fingar, 2010) was found to be significantly lower in hsf1-/- mice than in controls, consistent with the attenuation of p70SK1 activity. Notably, inhibition of mTOR pathway in hsf1-/- mice was partial as indicated by the residual phosphorylation levels of p70SK1 and rS6 proteins. Further, insulin-induced signaling under fasting conditions did not impair phosphorylation of mTOR target 4E-BP1 (T37/46 and S65) in the livers of hsf1-/- mice compared to WT controls (data not shown), suggesting the translational levels were not dramatically impaired. Thus, the differential sensitivity to insulin stimulation in the mutant mice depends on increased IR and IRS levels. This leads to enhanced PI3K/AKT pathway activation, rapid phosphorylation and attenuation of the activity of AKT substrates (Foxo1, GSK3, AS160), which inhibit glucose production but increased utilization. Enhanced insulin-induced signaling was also observed in livers of hsf1-/- mice at 2 months of age indicating that HSF1 has central role in insulin sensitivity at early stage of HCC development (data not shown).

Loss of HSF1 decreases hepatic lipogenesis by promoting insulin signaling in association with AMP-kinase activation

AMPK, which senses cellular energy status, is a negative regulator of mTOR activation, and there is evidence that AMPK activation may prevent insulin resistance (Canto and Auwerx, 2010; Shaw, 2009; Zhang et al., 2009). We found that livers of WT mice expressed a relatively low basal level of p-AMPK(T172) that was not altered by insulin treatment (Fig. 7A). In contrast, livers of hsf1-/- mice had significantly higher basal levels of p-AMPK and insulin induced AMPK activation to a much greater level compared to WT counterparts. To evaluate the functional consequences of AMPK activation we analyzed AMPK substrates. Activated AMPK by phosphorylating mitochondrial enzyme ACC(S79) decreases its activity and inhibits fatty acid synthesis (ACC1) and promotes fatty acid oxidation (ACC2). Moreover, AMPK-mediated phosphorylation of raptor (S792) decreases mTORC1 activity and of hormone sensitive lipase (HSL) (S565) inhibits HSL-mediated lipolysis (Gwinn et al., 2008; Watt et al., 2006). Notably, during periods of increased energy demands (e.g. fasting), the rate of lipolysis is increased. Although lipolysis is regulated by a complex process involving several hormone-regulated lipases, the phosphorylation of HSL by activated AMPK that attenuates its activity is believed to play a critical role in regulating lipid homeostasis in relation to energy demand. Our results show that phosphorylation levels of ACC (S79), raptor (S792) and HSL (S565) are significantly higher in the livers of hsf1-/- mice compared to WT controls (Figs. 7A and S5). Injection of insulin in fasted animals stimulated p-HSL (S565) in both genotypes, but livers of hsf1-/- mice exhibited a higher basal level of p-HSL that was strongly induced to a much higher level compared to controls (Fig. S5). Similarly, insulin-induced p-AMPK was enhanced in muscles of hsf1-/- mice (data not shown). Of interest, insulin-stimulated signaling in liver of WT mice induced HSF1 activation, which was detected by the slower HSF1 electrophoretic mobility due to increased phosphorylation (arrows in Fig. 7A).

Figure 7. HSF1 promotes hepatic metabolic adaptation by modulating proteostasis and influences energy metabolic pathways in response to nutrient availability and insulin stimulation.

Figure 7

Open in a new tab

(A) Increased insulin sensitivity in hsf1-/- mice results in enhanced AMPK activation: Immunoblot analyses of basal, and insulin stimulated p-AMPK levels in livers of WT and hsf1-/- mice. 7 months-old mice were fasted for 16 hrs and injected with saline or insulin (0.75 mU/g i.p.). Liver extracts prepared 10 min after insulin or saline injection were examined and blots were quantified (right panels). Level of p-AMPK or p-ACC normalized to total protein level was expressed as relative fold-increase to control (WT without insulin). Bars are mean ± SD; (n=5 mice per group).

(B) Expression of HSF1, AMPK, and target genes: under nutrient available conditions: Representative Western blot of liver extracts from 7 months-old mice, fasted for 24 hrs, or refed ad libitum for 3 hours, using the indicated antibodies. The ratio of p-AMPK or p-ACC normalized to total protein level is given. Real-time RT-PCR analysis of genes that encode for HSPs in liver of hsf1-/- or WT mice (lower panels). Data are presented as relative mRNA expression. Bars are mean ± SD; (n=3-5 mice per group).

In all panels, bars indicate WT (filled) or hsf1-/- (open) mice. Statistical significance is indicated (* p<0.05, ** p<0.01, *** p<0.001).

As HSF1 regulates components of the proteostatic system and protein-folding quality control, we evaluated whether HSF1 senses nutrient availability. Physiological protein-induced stress can be created in liver by feeding the mice after a fasting period; this process not only activates cellular stressors by inducing protein synthesis in the endoplasmic reticulum (ER) and cytosol, but also activates insulin signaling, a condition that is ideal to test the potential contribution of HSF1 to proteostasis and hepatic metabolism. Under fasting conditions, HSF1 activity was present at low levels. However, under feeding, or re-feeding conditions, we observed an increase in HSF1 activity (indicated by arrows) leading to expression of hsp genes (Fig. 7B). This result suggests that the metabolic load created by re-feeding led to HSF1 activation. We next determined that the p-AMPK/AMPK ratio was significantly higher in the livers of hsf1-/- mice than control mice of 7 months of age (Fig. 7B). However, p-AMPK level did not fluctuate during nutrient cycle. Of note, basal p-AMPK level was high in fed mutant mice obscuring nutrient-induced p-AMPK under fasting or re-feeding conditions. Consistent with this result, although the p-ACC (S79) levels exhibited only modest fluctuation in livers of fasted and postprandial mice, hsf1-/- mice exhibited a significantly higher p-ACC/ACC ratio compared to WT mice indicating enhanced AMPK activity (Figs. 7B). This increased AMPK activation to nutrient availability was also observed in the liver of hsf1-/- mice at 2 months of age (data not shown).

To investigate the functional significance of AMPK activation in context of hepatic metabolism and carcinogen-induced liver tumorigenesis we examined the phosphorylation status of AMPK and AMPK substrates in untreated or DEN-treated mice. First, we noted that under CD-feeding p-AMPK levels increased in aged-mice, but the livers of hsf1-/- mice exhibited significantly higher p-AMPK/AMPK ratio than WT controls (Fig. S6A). This was accompanied by increased p-ACC, p-Raptor and p-HSL in mutant mice indicating enhanced sensitivity to energy metabolism. Of note, p-AMPK levels in aged-WT mice did not yield the expected increase in p-ACC level indicating age-dependent decline in AMPK activity as reported (Reznick et al., 2007). Further, we observed that hsf1-/- mice compared to WT controls exhibited much higher basal level of p-AKT (S473) throughout the life that is consistent with enhanced insulin signaling. In summary, the age of onset of elevated AMPK activation correlates with the timing of increased AKT-signaling in hsf1-/- mice. Next, we checked the expression levels of AMPK and AMPK substrates in the liver of CD-fed DEN-treated mice. We observed that hsf1-/- mice exhibited small but significant increase in p-AMPK level compared to WT controls, albeit p-AMPK levels were lower in DEN-treated animals (Fig. S6B). Overall, the p-ACC level was also slightly but reproducibly increased in mutant mice compared to WT controls consistent with the p-AMPK levels. The level of p-HMG-CoA reductase (S871) (HMG-CR), a rate-determining enzyme in cholesterol synthesis and AMPK target, which decreases enzyme activity, did not differ significantly between the genotypes (Fig. S6B). Compared to WT control, DEN-treated hsf1-/- mice exhibited significantly decreased level of FAS, a central molecule in lipogenesis and a known transcriptional target of SREBP-1c in response to feeding and insulin signaling. The fact that mRNA level of SREBP-1c was found to be reduced in DEN-treated hsf1-/- mice (Fig. 4C) may explain the decreased FAS protein level. Further, there was not significant difference in AMPK-mediated p-HSL level in CD-fed mice. However DEN-treated WT mice exhibited markedly decreased p-HSL that was less pronounced in hsf1-/- mice (Fig. S6). This result suggests that inhibition of HSL activity by elevated p-HSL might be responsible for the lower fed serum and hepatic NEFA levels found in DEN-treated hsf1-/- mice (Fig. 3).

Taken together, these results suggest a link between energy metabolism and proteostasis driven by HSF1 that enable metabolic adaptation to nutrient availability under physiological conditions. Hsf1 deficiency enhances insulin signaling, causing a more severe energy deficit especially under nutrient availability and promotion of AMPK activation. Activation of AMPK in absence of HSF1 changes the metabolic balance in the liver towards utilization of carbohydrates as an energy source and suppression of lipid synthesis, thus generating metabolic conditions where malignant transformation induced by carcinogens is inhibited.

Discussion

Transcriptional regulators that respond to stress also influence survival and longevity of organisms. HSF1 plays central role in regulating heat shock response, which is an adaptive response by organisms to protect against environmental insults and enhances survival and longevity (Akerfelt et al., 2010; Hsu et al., 2003). However, increased HSF1-mediated resistance to stress also has growth-promoting effects towards specific tumor types that are detrimental to the normal life span of organisms. As reported previously, loss of HSF1 negatively impacts tumorigenesis driven by p53 or Ras mutations (Dai et al., 2007; Min et al., 2007). Since HSF1 does not function as a classic oncogene or tumor repressor, it was suggested that increased resistance to proteotoxic stress induced by HSF1 may support tumor initiation and growth by enabling cells to accommodate genetic alterations that occur during malignancy. Although HSF1 impacts a large array of cellular functions that are altered during malignancy, including translation, ribosome biogenesis, and glucose metabolism (Hahn et al., 2004), specific mechanisms by which HSF1 may support growth of specific tumors are not well understood. Here, we have demonstrated that HSF1 has detrimental effects on survival of the organism by potentiating perturbations in hepatic metabolism caused by carcinogens that lead to hepatic steatosis and HCC. The data indicate that genetic inactivation of HSF1 results in the up-regulation of protective mechanisms against metabolic stress induced by risk factors for chronic liver disease and cirrhosis that can lead to HCC development after several decades.

One interesting result in the present study is that hsf1 deficiency corrects DEN-induced hepatic metabolic abnormalities associated with fat accumulation by co-activating two opposing signaling pathways: increased insulin sensitivity with improved insulin signal transduction, and activated AMPK in response to energy depletion and insulin stimulation (model in Fig. S7). In liver, insulin, through activation of IR and AKT, suppress hepatic glyconeogenesis by negatively regulating Foxo1, PCG-1a, and CRTC2 (Dentin et al., 2007; Gross et al., 2008; Li et al., 2007). Of interest we observed repression of hepatic gluconeogenesis in liver of DEN treated hsf1-/- mice that correlates with lower mRNA levels of PEPCK and G6Pase, most likely due to enhanced insulin signaling. However, insulin also stimulates lipid accumulation through activation of AKT, which positively regulates de novo anabolic lipid metabolism by promoting glucose uptake, glycolysis, and expression of genes involved in lipid biosynthesis (such as SREBP1c). In addition, insulin promotes mTORC1 activation that up-regulates ribosomal biogenesis and protein translation, but also induction of SREBP-1 and expression of target genes involved in lipid synthesis (Laplante and Sabatini, 2009). On the other hand, AMPK activation induced in response to ATP depletion in the cell can blunt the anabolic effects of insulin in lipid biogenesis, reducing hepatic lipid deposition by promoting fatty acid oxidation and impairing SREBP-1c expression and activation. Negative regulation of mTORC1 by AMPK can increase insulin sensitivity by repressing p70S6K1 phosphorylation and inactivation of IRS1 (Shaw, 2009; Zhang et al., 2009).

Based on this information and our data, we propose a model (Fig. S7) whereby loss of HSF1 increases basal level of IR and IRS1, sensitizing cells to insulin stimulation. This causes rapid glucose utilization preferentially in muscles but also in other tissues via GLUT4 translocation to cell membrane, a process negatively regulated by AS160. Limited glucose availability under insulin stimulation, particularly in the liver of mutant mice in which glucose production is impaired, may cause transient energy deficit resulting in AMPK activation. Further, enhanced AKT signaling can promote protein synthesis and protein glycosylation that can consume ATP causing transient ATP depletion and AMPK activation (Fang et al., 2010). Although this mechanism has been proposed to promote aerobic glycolysis and tumor cell growth, however hsf1-deficient cells appear to be resistant to undergo the glycolytic shift characteristic of cancer cells, thus the tumor promoting effects of AKT signaling on cellular glucose metabolism are blunted by HSF1-deficiency (Dai et al., 2007). The mechanism by which HSF1 exerts this effect remains elusive. On the other hand, AMPK activation inhibits lipogenesis and thus protect from hepatic steatosis diminishing insulin-induced effects on lipid synthesis. In principle it is possible that HSF1 may exert effects on lipid metabolism through transcriptional control of lipogenic factors such as PPARγ, a key regulator of adipogenesis, this possibility is unlikely, since both primary mouse embryonic fibroblasts from WT and hsf1-/- mice were equally capable of inducing adipogenesis under pro-differentiative conditions, giving rise to lipid accumulating cells that express markers of differentiated adipocytes including PPARγ (data not shown). Altogether we propose that enhanced insulin sensitivity in association with AMPK activation is possibly the driving force for the lean phenotype of the hsf1-/- mice and striking resistance to DEN-induced hepatic steatosis and HCC development. However, our work does not preclude the contribution of inflammation triggered by excessive lipid deposition in cancer progression in our model. Further, the contribution of other organs such as adipose tissue and muscle to altered metabolism in hsf1-/- mice cannot be excluded. This aspect requires further investigation because our data indicate that key pro-lipogenic transcription factors are also significantly reduced in adipose tissues in the absence of HSF1, indicating its broader role in lipid metabolism.

Another interesting observation in our study is the link established between proteostasis and hepatic metabolism. We anticipate that HSF1 activity regulated by nutrient availability supports malignant transformation through promoting proteostatic pathways and thus metabolic adaptations towards cancer development. In general, protein quality control pathways are efficient to cope with the anabolic effects of stimulated protein synthesis in ER and cytosol under normal nutrient conditions. We presume that the feedback mechanism in our model by which HSF1 target molecules attenuate the IR-IRS1-AKT signaling pathway after insulin stimulation helps to restore metabolic homeostasis, averting pathological over-stimulation of insulin signaling. Although the underlying mechanism is unknown, we propose that HSF1 and its major targets (HSPs) may be critical negative regulators of IR and IRS1 expression by modulating the ubiquitin-proteasome degradation system as reported (Mayer and Belsham, 2010; Zhande et al., 2002). Notably, we have shown that hsf1 deletion in cell culture inhibits this pathway and stabilizes p53, and perhaps other proteins (Jin et al., 2009). However, direct effect of HSF1 on transcriptional repression of IR and IRS1 expression cannot also be ruled out and our data supports such a mechanism. In DEN-induced HCC, this HSF1-dependent mechanism may participate to increase insulin resistance, and thus tumor formation.

Taken together, the results suggest that HSF1 may have a central role along with other signaling components in HCC development by modulating proteostasis and metabolic pathways and may be an important drug target in the treatment of this disease.

Experimental procedures

Animals, liver tumor induction, and analysis

Hsf1-/- mice were previously reported (Zhang et al., 2002). Both hsf1-/- and hsf1+/+ control littermates were of mixed C57BL/6-129SvPas F1 genetic background. Animal care and experiments were performed in accordance with the guidelines of Institutional Animal Care and Use Committee and NIH guidelines.

A single dose of DEN (25 mg/kg body weight) (Sigma) was injected into 14-day-old male mice intraperitoneally (i.p.) to initiate tumor formation, and animals were euthanized at different ages as indicated. Livers were removed, analyzed for presence of HCCs using a dissecting microscope, and individual lobes were snap frozen in OCT or paraffin-embedded and subjected to histological and immunochemical analyses. For early response analyses, mice were sacrificed at 4 hours or 1, 2, 10, or 30 days post DEN administration. For HCC analysis mice were euthanized at 7 or 10 months after DEN treatment, and mouse weight, liver weight, number of tumors per liver (% tumor incidence), maximal tumor size (mm), and % tumor area/liver were determined.

Glucose and insulin tolerance tests

Glucose and insulin tolerance tests were carried out following overnight fasting (16 hours). Animals were injected i.p. with either 2 mg/g glucose or 0.75 mU/g insulin. Blood glucose was assayed from the tail vein using an automated glucose monitor (Bayer HealthCare LLC, IN, USA).

Liver enzymes and Insulin Assays

Serum ALT and AST levels were detected using Liquid Reagent Kit according to the manufacturer’s instruction (Pointe Scientific Inc. MI, USA). Triglycerides and non-esterified fatty acid concentrations were determined using an acyl-CoA oxidase-based colorimetric kit (Wako Biochemicals USA inc., Richmond, VA). Serum insulin level was determined using ELISA kit (Calbiochem, Spring Valley, CA). Serum cholesterol in HDL or LDL/VLDL and total cholesterol levels were assayed using detection kit (BioVision, Mountain view, CA).

Serum Cytokine Assays

Concentration of serum IL-6 and TNF-α, were determined using an immunoassay kit (eBioscience, Inc. San Diego, CA).

Western blot, histology and immunohistochemistry, and real-time PCR

These assays were performed by standard procedures. For more information see the Supplemental Experimental Procedures.

Statistical consideration

All experiments were performed with at least 3-20 mice. Data are presented as mean ± SD. Statistical significance between experimental groups was assessed using Student’s t test and p<0.05 was considered significant.

Supplementary Material

01

Highlights.

  1. HSF1 deficiency impairs HCC progression

  2. Inactivation of HSF1 enhances insulin sensitivity and sensitizes activation of AMPK

  3. HSF1 is a potential target for control of hepatic steatosis and insulin resistance

  4. HSF1 is a potential target for treatment of liver cancer

Acknowledgments

The research was supported by grants from the National Institutes of Health CA062130 and CA132640 (N.F.M) and CA121951 (D.M.).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Akerfelt M, Morimoto RI, Sistonen L. Heat shock factors: integrators of cell stress, development and lifespan. Nat Rev Mol Cell Biol. 2010;11:545–555. doi: 10.1038/nrm2938. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Ariel I, Miao HQ, Ji XR, Schneider T, Roll D, de Groot N, Hochberg A, Ayesh S. Imprinted H19 oncofetal RNA is a candidate tumour marker for hepatocellular carcinoma. Mol Pathol. 1998;51:21–25. doi: 10.1136/mp.51.1.21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Canto C, Auwerx J. AMP-activated protein kinase and its downstream transcriptional pathways. Cell Mol Life Sci. 2010;67:3407–3423. doi: 10.1007/s00018-010-0454-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Dai C, Whitesell L, Rogers AB, Lindquist S. Heat shock factor 1 is a powerful multifaceted modifier of carcinogenesis. Cell. 2007;130:1005–1018. doi: 10.1016/j.cell.2007.07.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Dentin R, Liu Y, Koo SH, Hedrick S, Vargas T, Heredia J, Yates J, 3rd, Montminy M. Insulin modulates gluconeogenesis by inhibition of the coactivator TORC2. Nature. 2007;449:366–369. doi: 10.1038/nature06128. [DOI] [PubMed] [Google Scholar]
  6. Duncan RE, Ahmadian M, Jaworski K, Sarkadi-Nagy E, Sul HS. Regulation of lipolysis in adipocytes. Annu Rev Nutr. 2007;27:79–101. doi: 10.1146/annurev.nutr.27.061406.093734. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. El-Serag HB, Rudolph KL. Hepatocellular carcinoma: epidemiology and molecular carcinogenesis. Gastroenterology. 2007;132:2557–2576. doi: 10.1053/j.gastro.2007.04.061. [DOI] [PubMed] [Google Scholar]
  8. Fang M, Shen Z, Huang S, Zhao L, Chen S, Mak TW, Wang X. The ER UDPase ENTPD5 Promotes Protein N-Glycosylation, the Warburg Effect, and Proliferation in the PTEN Pathway. Cell. 2010;143:711–724. doi: 10.1016/j.cell.2010.10.010. [DOI] [PubMed] [Google Scholar]
  9. Farazi PA, DePinho RA. Hepatocellular carcinoma pathogenesis: from genes to environment. Nat Rev Cancer. 2006;6:674–687. doi: 10.1038/nrc1934. [DOI] [PubMed] [Google Scholar]
  10. Foster KG, Fingar DC. Mammalian target of rapamycin (mTOR): conducting the cellular signaling symphony. J Biol Chem. 2010;285:14071–14077. doi: 10.1074/jbc.R109.094003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Grivennikov SI, Greten FR, Karin M. Immunity, inflammation, and cancer. Cell. 2010;140:883–899. doi: 10.1016/j.cell.2010.01.025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Gross DN, van den Heuvel AP, Birnbaum MJ. The role of FoxO in the regulation of metabolism. Oncogene. 2008;27:2320–2336. doi: 10.1038/onc.2008.25. [DOI] [PubMed] [Google Scholar]
  13. Gwinn DM, Shackelford DB, Egan DF, Mihaylova MM, Mery A, Vasquez DS, Turk BE, Shaw RJ. AMPK phosphorylation of raptor mediates a metabolic checkpoint. Mol Cell. 2008;30:214–226. doi: 10.1016/j.molcel.2008.03.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Hahn JS, Hu Z, Thiele DJ, Iyer VR. Genome-wide analysis of the biology of stress responses through heat shock transcription factor. Mol Cell Biol. 2004;24:5249–5256. doi: 10.1128/MCB.24.12.5249-5256.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144:646–674. doi: 10.1016/j.cell.2011.02.013. [DOI] [PubMed] [Google Scholar]
  16. Herman MA, Kahn BB. Glucose transport and sensing in the maintenance of glucose homeostasis and metabolic harmony. J Clin Invest. 2006;116:1767–1775. doi: 10.1172/JCI29027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Hotamisligil GS. Endoplasmic reticulum stress and the inflammatory basis of metabolic disease. Cell. 2010;140:900–917. doi: 10.1016/j.cell.2010.02.034. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Hsu AL, Murphy CT, Kenyon C. Regulation of aging and age-related disease by DAF-16 and heat-shock factor. Science. 2003;300:1142–1145. doi: 10.1126/science.1083701. [DOI] [PubMed] [Google Scholar]
  19. Jin X, Moskophidis D, Hu Y, Phillips A, Mivechi NF. Heat shock factor 1 deficiency via its downstream target gene alphaB-crystallin (Hspb5) impairs p53 degradation. J Cell Biochem. 2009;107:504–515. doi: 10.1002/jcb.22151. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Karin M, Greten FR. NF-kappaB: linking inflammation and immunity to cancer development and progression. Nat Rev Immunol. 2005;5:749–759. doi: 10.1038/nri1703. [DOI] [PubMed] [Google Scholar]
  21. Kersten S. Mechanisms of nutritional and hormonal regulation of lipogenesis. EMBO Rep. 2001;2:282–286. doi: 10.1093/embo-reports/kve071. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Laplante M, Sabatini DM. An emerging role of mTOR in lipid biosynthesis. Curr Biol. 2009;19:R1046–1052. doi: 10.1016/j.cub.2009.09.058. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lee JS, Chu IS, Mikaelyan A, Calvisi DF, Heo J, Reddy JK, Thorgeirsson SS. Application of comparative functional genomics to identify best-fit mouse models to study human cancer. Nat Genet. 2004;36:1306–1311. doi: 10.1038/ng1481. [DOI] [PubMed] [Google Scholar]
  24. Levine AJ, Puzio-Kuter AM. The control of the metabolic switch in cancers by oncogenes and tumor suppressor genes. Science. 2010;330:1340–1344. doi: 10.1126/science.1193494. [DOI] [PubMed] [Google Scholar]
  25. Li X, Monks B, Ge Q, Birnbaum MJ. Akt/PKB regulates hepatic metabolism by directly inhibiting PGC-1alpha transcription coactivator. Nature. 2007;447:1012–1016. doi: 10.1038/nature05861. [DOI] [PubMed] [Google Scholar]
  26. Luedde T, Beraza N, Kotsikoris V, van Loo G, Nenci A, De Vos R, Roskams T, Trautwein C, Pasparakis M. Deletion of NEMO/IKKgamma in liver parenchymal cells causes steatohepatitis and hepatocellular carcinoma. Cancer Cell. 2007;11:119–132. doi: 10.1016/j.ccr.2006.12.016. [DOI] [PubMed] [Google Scholar]
  27. Luo J, Solimini NL, Elledge SJ. Principles of cancer therapy: oncogene and non-oncogene addiction. Cell. 2009;136:823–837. doi: 10.1016/j.cell.2009.02.024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Maeda S, Kamata H, Luo JL, Leffert H, Karin M. IKKbeta couples hepatocyte death to cytokine-driven compensatory proliferation that promotes chemical hepatocarcinogenesis. Cell. 2005;121:977–990. doi: 10.1016/j.cell.2005.04.014. [DOI] [PubMed] [Google Scholar]
  29. Mantovani A, Allavena P, Sica A, Balkwill F. Cancer-related inflammation. Nature. 2008;454:436–444. doi: 10.1038/nature07205. [DOI] [PubMed] [Google Scholar]
  30. Mayer CM, Belsham DD. Central insulin signaling is attenuated by long-term insulin exposure via insulin receptor substrate-1 serine phosphorylation, proteasomal degradation, and lysosomal insulin receptor degradation. Endocrinology. 2010;151:75–84. doi: 10.1210/en.2009-0838. [DOI] [PubMed] [Google Scholar]
  31. Menendez JA, Lupu R. Fatty acid synthase and the lipogenic phenotype in cancer pathogenesis. Nat Rev Cancer. 2007;7:763–777. doi: 10.1038/nrc2222. [DOI] [PubMed] [Google Scholar]
  32. Min JN, Huang L, Zimonjic DB, Moskophidis D, Mivechi NF. Selective suppression of lymphomas by functional loss of Hsf1 in a p53-deficient mouse model for spontaneous tumors. Oncogene. 2007;26:5086–5097. doi: 10.1038/sj.onc.1210317. [DOI] [PubMed] [Google Scholar]
  33. Nomura DK, Long JZ, Niessen S, Hoover HS, Ng SW, Cravatt BF. Monoacylglycerol lipase regulates a fatty acid network that promotes cancer pathogenesis. Cell. 2010;140:49–61. doi: 10.1016/j.cell.2009.11.027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Powers ET, Morimoto RI, Dillin A, Kelly JW, Balch WE. Biological and chemical approaches to diseases of proteostasis deficiency. Annu Rev Biochem. 2009;78:959–991. doi: 10.1146/annurev.biochem.052308.114844. [DOI] [PubMed] [Google Scholar]
  35. Reznick RM, Zong H, Li J, Morino K, Moore IK, Yu HJ, Liu ZX, Dong J, Mustard KJ, Hawley SA, et al. Aging-associated reductions in AMP-activated protein kinase activity and mitochondrial biogenesis. Cell Metab. 2007;5:151–156. doi: 10.1016/j.cmet.2007.01.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Shaw RJ. LKB1 and AMP-activated protein kinase control of mTOR signalling and growth. Acta Physiol (Oxf) 2009;196:65–80. doi: 10.1111/j.1748-1716.2009.01972.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Siegel AB, Zhu AX. Metabolic syndrome and hepatocellular carcinoma: two growing epidemics with a potential link. Cancer. 2009;115:5651–5661. doi: 10.1002/cncr.24687. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Sorbi D, Boynton J, Lindor KD. The ratio of aspartate aminotransferase to alanine aminotransferase: potential value in differentiating nonalcoholic steatohepatitis from alcoholic liver disease. Am J Gastroenterol. 1999;94:1018–1022. doi: 10.1111/j.1572-0241.1999.01006.x. [DOI] [PubMed] [Google Scholar]
  39. Spangenberg HC, Thimme R, Blum HE. Targeted therapy for hepatocellular carcinoma. Nat Rev Gastroenterol Hepatol. 2009;6:423–432. doi: 10.1038/nrgastro.2009.86. [DOI] [PubMed] [Google Scholar]
  40. Watt MJ, Holmes AG, Pinnamaneni SK, Garnham AP, Steinberg GR, Kemp BE, Febbraio MA. Regulation of HSL serine phosphorylation in skeletal muscle and adipose tissue. Am J Physiol Endocrinol Metab. 2006;290:E500–508. doi: 10.1152/ajpendo.00361.2005. [DOI] [PubMed] [Google Scholar]
  41. Zhande R, Mitchell JJ, Wu J, Sun XJ. Molecular mechanism of insulin-induced degradation of insulin receptor substrate 1. Mol Cell Biol. 2002;22:1016–1026. doi: 10.1128/MCB.22.4.1016-1026.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Zhang BB, Zhou G, Li C. AMPK: an emerging drug target for diabetes and the metabolic syndrome. Cell Metab. 2009;9:407–416. doi: 10.1016/j.cmet.2009.03.012. [DOI] [PubMed] [Google Scholar]
  43. Zhang Y, Huang L, Zhang J, Moskophidis D, Mivechi NF. Targeted disruption of hsf1 leads to lack of thermotolerance and defines tissue-specific regulation for stress-inducible Hsp molecular chaperones. J Cell Biochem. 2002;86:376–393. doi: 10.1002/jcb.10232. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

01
NIHMS303995-supplement-01.doc (4.5MB, doc)

RESOURCES