Potential conflict of interest: Nothing to report.
This study was supported by the National Institute on Alcoholism and Alcohol Abuse (grants AA-015951 and AA-013623; to M.Y.) and the National Institute of Diabetes and Digestive and Kidney Diseases (grant DK-078187; to B.F.).
Lipin-1 is a protein that exhibits dual functions as a phosphatidic acid phosphohydrolase enzyme in the triglyceride synthesis pathways and a transcriptional coregulator. Our previous studies have shown that ethanol causes fatty liver by activation of sterol regulatory element-binding protein 1 (SREBP-1) and inhibition of hepatic AMP-activated protein kinase (AMPK) in mice. Here, we tested the hypothesis that AMPK-SREBP-1 signaling may be involved in ethanol-mediated up-regulation of lipin-1 gene expression. The effects of ethanol on lipin-1 were investigated in cultured hepatic cells and in the livers of chronic ethanol-fed mice. Ethanol exposure robustly induced activity of a mouse lipin-1 promoter, promoted cytoplasmic localization of lipin-1, and caused excess lipid accumulation, both in cultured hepatic cells and in mouse livers. Mechanistic studies showed that ethanol-mediated induction of lipin-1 gene expression was inhibited by a known activator of AMPK or overexpression of a constitutively active form of AMPK. Importantly, overexpression of the processed nuclear form of SREBP-1c abolished the ability of 5-aminoimidazole-4-carboxamide ribonucleoside to suppress ethanol-mediated induction of lipin-1 gene-expression level. Chromatin immunoprecipitation assays further revealed that ethanol exposure significantly increased the association of acetylated histone H3 at lysine 9 with the SRE-containing region in the promoter of the lipin-1 gene. Conclusion: In conclusion, ethanol-induced up-regulation of lipin-1 gene expression is mediated through inhibition of AMPK and activation of SREBP-1. (Hepatology 2012)
Lipin-1, a mammalian Mg2+-dependent phosphatidate phosphatase type (PAP), has recently been identified as a key regulator of lipid metabolism in several organs, including the liver.1 The gene encoding lipin-1 (LPIN1) was first identified by positional cloning of the mutant gene underlying lipodystrophy in the fatty liver dystrophy (fld) mouse in 2001.1, 2
Lipin-1 exhibits two distinct functions in regulating lipid metabolism according to subcellular localization studies. In the cytoplasm, lipin-1 functions as a Mg2+-dependent PAP enzyme involved in the biosynthesis of triacylglycerol (TAG) and phospholipids by converting phosphatidate (PA) to diacylglycerol (DAG) at the endoplasmic reticulum (ER).1 In the nucleus, lipin-1 acts as a transcriptional coactivator to increase the capacity of the liver for fatty acid oxidation by interacting with peroxisome proliferator-activated receptor alpha (PPARα) and PPARγ coactivator-1 alpha (PGC-1α).1, 3 The nuclear-localized lipin-1 also suppresses the functions of sterol regulatory element-binding protein 1 (SREBP-1), a master regulator of lipid metabolism.4 The subcellular localization of lipin-1 is highly regulated by post-translational modifications. Specifically, sumoylation promotes nuclear retention and transcriptional activity.5 Second, though there are numerous putative phosphorylation sites that may have accessory effects, serine phosphorylation promotes nuclear export and translocation to the ER membrane, whereas dephosphorylation promotes its cytosolic distribution.1, 5
Clinically, alcoholic fatty liver disease (AFLD) is characterized by increased accumulation of fat in the livers of patients who have consumed excessive amounts of alcohol for prolonged periods. Considerable evidence has shown that increased fat accumulation in the liver can progress to more harmful forms of liver injury, such as fibrosis and cirrhosis, in humans. The molecular and cellular mechanisms by which ethanol causes AFLD are multiple and still incompletely understood. Previously, we and several other groups have shown that ethanol induces lipid synthesis by activation of SREBP-1 in the livers of animals.6-9 Moreover, ethanol's effect on SREBP-1 results partially from inhibition of AMP-activated protein kinase (AMPK).9 Hence, ethanol-mediated dysregulation of the AMPK-SREBP-1–signaling pathway contributes to the development of AFLD.
Before the identification of lipin-1, PAP activity was shown to be increased in the livers of human alcoholics and patients with AFLD in several studies.10-12 In parallel, ethanol-mediated activation of PAP was closely associated with the development of fatty liver in rodents and humans.10-12 Consistent with these studies, we recently reported that chronic ethanol feeding significantly increased lipin-1 messenger RNA (mRNA) and its cytosolic protein levels in the livers of mice, supporting the concept that up-regulation of lipin-1 by ethanol contributes to enhanced PAP activity and hepatic lipid accumulation in ethanol-fed mice.13 Nevertheless, the molecular mechanisms and signaling pathways affected by ethanol, which result in altering the gene and protein expression of lipin-1, are not fully understood. The present study was undertaken to investigate the underlying mechanisms by which ethanol regulates lipin-1, with a focus on the role of AMPK-SREBP-1 signaling.
The immortalized mouse hepatocyte cell line, AML-12, was purchased from the American Type Culture Collection (Manassas, VA). Various in vitro assays, using AML-12 cells exposed to ethanol or other reagents, were performed as described in the Supporting Materials.
Male C57BL/6J mice (6-8 weeks old) were purchased from Jackson Laboratory (Bar Harbor, ME). Mice were fed a modified Lieber-DeCarli ethanol-containing diet or a pair-fed control diet, as described in the Supporting Materials.
Data are presented as means ± standard deviation (SD). All data were analyzed by two-way analysis of variance, followed by Tukey's multiple comparison procedure, with P < 0.05 being considered significant.
Additional materials and methods are described in the Supporting Materials.
Up-regulation of Lpin1 Promoter by Ethanol in Mouse AML-12 Hepatocytes.
Mouse AML-12 hepatocytes express sufficient levels of class I (low Km) alcohol dehydrogenase (ADH) and aldehyde dehydrogenase 2 (ALDH2) proteins and efficiently metabolize ethanol (data not shown). However, AML-12 cells lack detectable immunoreactive protein cytochrome P450 2E1 (CYP2E1) (Supporting Fig. 1A).
AML-12 cells were transfected with a reporter gene (mouse Lpin1-luciferase) and an internal control plasmid (β-galactosidase) and exposed to various concentrations of ethanol (20-100 mM), then harvested for assay of reporter enzymes. The Lpin1 reporter activity was significantly increased in a concentration-dependent manner by incubation with ethanol in AML-12 hepatocytes (Fig. 1A).
We determined whether ethanol metabolism was required for ethanol-induced Lpin1 promoter activity by use of inhibitors of ethanol metabolism. We used the ADH inhibitor, 4-methylpyrazole (4-MP), and the ALDH2 inhibitor, cyanamide (Cya). Treatment with each of these inhibitors alone had no effect on baseline Lpin1-luciferase levels; however, when cells were exposed to ethanol, the inhibitors virtually abolished the ethanol-dependent induction of Lpin1-luciferase (Fig. 1B). Moreover, acetate (10 mM), one of ethanol's major metabolites, shared its ability to increase Lpin1 promoter activity in AML-12 cells (Fig. 1C).
Taken together, the results suggest that ethanol metabolism is necessary for the ability of ethanol to activate Lpin1 promoter activity.
Ethanol Promotes Cytoplasmic Localization of Lipin-1 in Mouse AML-12 Hepatocytes.
In AML-12 cells, ethanol increased total lipin-1 protein levels (Fig. 2A). We assessed the subcellular localization of lipin-1 in response to ethanol treatment in AML-12 cells. AML-12 cells were cultured in the presence of ethanol, and extracts of these cells were fractionated into cytosol and nuclei, followed by western blotting analysis. The increase in endogenous lipin-1 protein induced by ethanol was observed strictly in the cytosolic fractions (Fig. 2B).
Immunofluorescent staining of nuclei (4′,6-diamidino-2-phenylindole [DAPI] staining) and lipin-1 confirmed that endogenous lipin-1 was present predominantly in the cytoplasm (Fig. 2C). Furthermore, treatment with ethanol dramatically increased the intensity of lipin-1 staining in the cytoplasm, as compared to controls, suggesting an increase of lipin-1 protein expression.
Cellular PAP activity was significantly induced by ethanol treatment, compared with the control, in AML-12 cells (Fig. 2D).
Collectively, these results suggest that ethanol promotes lipin-1 cytoplasmic accumulation and induces its PAP activity.
Ethanol-Induced Triglyceride Accumulation Is Mediated Through Lipin-1 Induction in Mouse AML-12 Hepatocytes.
The effect of ethanol on the development of steatosis in AML-12 cells was determined by measuring cellular triglyceride (TG) content with enzymatic assays using a triglyceride kit.14 There was a significant, concentration-dependent increase in TG content of cells exposed to ethanol (Supporting Fig. 2). Moreover, treatment with either 4-MP or cyanamide (Cya) essentially blocked the ability of ethanol to increase TG levels, indicating that ethanol metabolism is necessary for ethanol-mediated cellular lipid accumulation (Fig. 3A).
To determine whether the increased lipin-1 induced by ethanol would affect cellular TG synthesis, we examined TG accumulation in AML-12 cells transfected with Lpin1 short interfering (siRNA) or control siRNA in response to ethanol exposure. Knocking down lipin-1 with Lpin1 siRNA largely eliminated the capacity of ethanol to induce TG accumulation (Fig. 3B). Note that lipin-1 protein levels were decreased ∼70% after transfection with Lpin1 siRNA (Supporting Fig. 1B). These results suggest that ethanol metabolism increases lipin-1 enzymatic activity and, subsequently, promotes TG accumulation.
Ethanol Exposure Increased Association of Acetylated Histone H3/Lysine 9 and Nuclear Factor Y With Lpin1 Promoter In Vitro.
SREBP-1 functions together with nuclear factor Y (NF-Y) to transactivate the LPIN1 promoter through sterol regulatory element (SRE) and NF-Y-binding sites.15 The effect of ethanol on the binding of acetylated histone H3/Lys9 (lysine 9), SREBP-1, or NF-Y to the Lpin1-SRE binding site was determined using chromatin immunoprecipitation (ChIP) assays. After chromatin had been cross-linked in control and ethanol-treated AML-12 cells, it was sheared by sonication. DNA-protein complexes were immunoprecipitated with antibodies directed against acetylated histone H3/Lys9, SREBP-1, or NF-Y. Immunoprecipitated DNA was amplified by real-time quantitative reverse-transcription polymerase chain reaction (qRT-PCR), with primers for the Lpin1 promoter region containing SRE-binding elements. Ethanol significantly increased the interaction of acetylated histone H3/Lys9 and of NF-Y with the Lpin1-SRE promoter (Fig. 4A). The association of SREBP-1 with the Lpin1 promoter was not affected by ethanol. This may have been the result of rapid proteasomal degradation of nuclear SREBP-1 protein.16
SREBP-1 siRNA was found to be an effective inhibitor of SREBP-1 expression in AML-12 cells (Supporting Fig. 1C). Knocking down SREBP-1 with SREBP-1 siRNA partially abrogated the ability of ethanol to stimulate Lpin 1 promoter activity (Fig. 4B).
Pharmacological or Genetic Activation of AMPK Prevents Enhanced Expression of Lipin-1 and Elevated Lipid Accumulation Induced by Ethanol.
We further explored the role of AMPK-SREBP-1 signaling in the ethanol-mediated increase of Lpin1.9 Though ethanol robustly increased Lpin1 promoter activity and mRNA, pretreatment with 5-aminoimidazole-4-carboxamide ribonucleoside (AICAR) or overexpression of a constitutively active form of AMPK (AMPKα1312) largely prevented ethanol-dependent increases in Lpin1 promoter activity and mRNA levels (Fig. 5A; Supporting Fig. 3). Conversely, pharmacological inhibition or epigenetic silencing of AMPK with either compound C or AMPKα siRNA slightly augmented the effect of ethanol on Lpin1.
To determine whether SREBP-1 is involved in regulating the effects of AMPK on lipin-1, we stimulated SREBP-1 activity by overexpression of the active nuclear form of SREBP-1c (nSREBP-1) in AML-12 cells. Overexpression of nSREBP-1c abolished the ability of AICAR to suppress ethanol-mediated induction of lipin-1 gene expression (Fig. 5B). Conversely, inhibition of SREBP-1 expression by SREBP-1 siRNA further augmented the effect of AICAR on Lpin 1 in AML-12 cells exposed to ethanol.
Collectively, these results suggest that inhibition of AMPK and activation of SREBP-1 by ethanol may be involved, at least in part, in the up-regulation of lipin-1.
It is important to note the effect of transfection with AMPKα312 and AMPKα siRNA on the levels of AMPKα protein, as determined by western blotting analysis (Supporting Fig. 1D). Expression of AMPKα312 or AMPKα siRNA significantly increased or inhibited AMPK activity, respectively, in cultured hepatic cells.9 The alteration of AMPKα activity was accompanied by altered phosphorylation status of acetyl-CoA carboxylase (ACC), a downstream indicator of AMPK activity (Supporting Fig. 1D).
Effect of Chronic Ethanol Feeding on Lipin-1 in Mouse Livers.
Feeding mice ethanol (29% of the total calories) via a modified Lieber-DeCarli liquid diet for 4 weeks led to the development of fatty liver (Supporting Table 1).
Ethanol feeding markedly increased total mRNA expression of hepatic lipin-1 in by nearly 4.5-fold, compared to pair-fed controls (Fig. 6A).17 Note that there was no significant change in mRNA levels for lipin-2 and -3 in the livers of ethanol-fed mice, compared to controls (data not shown).
Acetylated histone H3/Lys9 was drastically increased by ethanol feeding, whereas histone H3 protein level was not affected by ethanol (Fig. 6B). Accordingly, the association of acetylated histone H3/Lys9 with the Lpin1-SRE promoter was significantly enhanced ∼2.4-fold by ethanol feeding, compared with pair-fed control mice (Fig. 6C). Note that ChIP assays demonstrated that the association of acetylated histone H3/Lys9 or glucocorticoid receptor (GR) with the Lpin 1-GRE site was not significantly affected by ethanol administration to mice, compared with controls (data not shown).
Ethanol feeding to mice significantly reduced sumoylation levels of hepatic lipin-1, while at the same time markedly increasing its level of acetylation (Fig. 6D; Supporting Fig. 4). More important, ethanol feeding robustly increased the amount of lipin-1 in the cytoplasm and dramatically decreased it in the nucleus in the mouse livers (Fig. 7A,B). Accordingly, hepatic PAP activity was significantly increased in ethanol-fed mice, compared with the pair-fed controls (Fig. 7D).
Taken together, our results clearly indicate that ethanol feeding increased hepatic lipin-1 gene expression and stimulated the cytoplasmic localization of lipin-1 in mouse livers.
In the present study, we investigated the effects of ethanol on lipin-1 in cultured hepatic cells and in animal tissues and explored the underlying mechanisms. In cultured AML-12 hepatocytes, chronic ethanol exposure robustly enhanced the activity of a mouse Lpin1 promoter and increased cytosolic lipin-1 protein levels as well as PAP activity. The ethanol-dependent up-regulation of lipin-1 was associated with elevated cellular TG accumulation in AML-12 cells. We also showed that acetate alone, a product of ethanol metabolism, produces many of these effects in AML-12 cells. Interestingly, ethanol-induced activation of the Lpin1 promoter and enhancement of lipin-1 mRNA levels were each inhibited by a known activator of AMPK (AICAR), as well as by overexpression of a constitutively active form of AMPK. Importantly, overexpression of nSREBP-1c largely abolished the ability of AICAR to suppress ethanol-mediated up-regulation of lipin-1, suggesting that AMPK lies upstream of the SREBP-1/lipin-1 axis. Consistent with in vitro findings, feeding mice an ethanol-containing liquid diet resulted in a robust increase in lipin-1 mRNA and cytosolic protein levels. Moreover, ChIP assays revealed that ethanol exposure significantly increased the association of acetylated histone H3/Lys9 with the SRE-containing region in the promoter of the lipin-1 gene, both in vitro and in vivo. We also demonstrated, for the first time, that acetylation and sumoylation of lipin-1 displayed reciprocal patterns in livers of chronically ethanol-fed mice. Taken together, our findings suggest that chronic ethanol exposure up-regulates hepatic lipin-1 and that this effect may contribute to the development of AFLD. Importantly, we have shown that this effect is mediated, at least in part, by modulating AMPK-SREBP-1 signaling (Fig. 8).
Our current data clearly suggest that ethanol metabolism through both ADH and ALDH2 are required for the effect of ethanol on lipin-1 in AML-12 cells. Paradoxically, our previous results suggest that acetaldehyde generated from ethanol is largely responsible for the ability of ethanol to activate SREBP-1 in hepatoma cell lines.6 This discrepancy may, in part, be explained by the use of different cell lines. Indeed, we have failed to establish cellular alcoholic steatosis models in two hepatoma cell lines (H4IIEC3 and McA-RH7777). On the other hand, though lipin-1 and SREBP-1 are both targets of ethanol, the underlying mechanisms for the observed effects could be entirely different between them. Interestingly, we and several other groups have recently shown that both acetaldehyde and acetate, two major metabolites of ethanol, are involved in alcoholic liver injury.18, 19
Our current findings suggest that ethanol increased lipin-1 gene expression largely through activation of SREBP-1 and NF-Y. Conceivably, additional molecular mechanisms are also involved. For instance, several putative glucocorticoid (GC) response elements (GREs) in the LPIN1 promoter have been identified. Indeed, lipin-1 expression is directly regulated by GCs in liver and adipose tissue.20, 21 This effect requires the GR and is mediated by binding of the receptor to GRE sites upstream of the LPIN1 gene. The GC-mediated effects are specific to lipin-1 (i.e., not lipin-2 or -3). The involvement of GC in ethanol-induced increases in lipin-1 is supported by a previous study showing that ethanol-mediated PAP activity was attenuated in adrenalectomized rats.10 Surprisingly, we found that the in vivo association of acetylated histone H3/Lys9 or GR with the Lpin1-GRE site in response to ethanol exposure was not significantly induced. We recently demonstrated that lipin-1 exhibits reciprocal patterns of gene expression in livers and adipose tissues of chronically ethanol-fed mice, suggesting a mechanism largely independent of GC effects.13 The definitive involvement of GCs in ethanol-mediated up-regulation of lipin-1 may need to be further studied through use of genetically modified animal models—such as liver-specific GR knockout mice. It is also tempting to speculate that ethanol may stabilize lipin-1 protein via enhanced lipin-1 acetyaltion and, subsequently, inhibition of lipin-1 degradation.
The molecular role of lipin-1 is dependant upon its subcellular localization. Nuclear compartmentalization of lipin-1 ensures that its role as a transcriptional coactivator predominates over its role as a PAP enzyme.1-5 Sumoylation of lipin-1α is required for its nuclear localization and transcriptional coactivator activity toward PGC-1α in cultured neuronal cells.5 Our current in vivo findings show that ethanol feeding markedly reduced hepatic lipin-1 sumoylation levels, which correlates with the observed dramatic reduction in its nuclear localization. In addition to inhibition of lipin-1 sumoylation, ethanol feeding also robustly increased the acetylation of lipin-1 in mouse livers. Acetate, the principal hepatic metabolite of ethanol, may contribute to hepatic lipin-1 hyperacetyaltion in ethanol-fed mice. Moreover, a lysine acetylation/deacetylation-sumoylation switch has been implicated in the functional regulation of several important molecules.22, 23 Ethanol inhibits sirtuin 1 (SIRT1), an nicotinamide adenine dinucleotide-positive–-dependent class III protein deacetylase, both in cultured hepatic cells and in animals.13, 24 It is possible that this ethanol-mediated hyperacetylation/hyposumoylation of lipin-1 may be a consequence of the inhibition of SIRT1 by ethanol. Whether acetylation/sumoylation would serve as a molecular switch to control the nuclear localization and coactivator activity of lipin-1 in the liver and how ethanol affects the functional relationship of SIRT1 and lipin-1 are currently under investigation in our laboratory.
Lipin-1 localizes to the nucleus and is a component of a transcriptional complex with PPARα/PGC-1α, which stimulates fatty acid oxidation in the liver.3 Ethanol-mediated dysregulation of the hepatic PPARα/PGC-1α axis and subsequent incomplete stimulation of PPARα/PGC-1α target genes involved in fatty acid oxidation contributes to the development of alcoholic liver steatosis.17 Taken together with a recent study demonstrating that a high-fat-diet–induced fatty liver is partially mediated by impairment of the PGC-1α/nuclear lipin-1/PPARα axis and fatty acid oxidation in mice, our current findings suggest that depletion of nuclear lipin-1 is likely to lead to impairment of the PPARα/PGC-1α axis and fatty acid oxidation in the livers of chronically ethanol-fed animals.25 Furthermore, lipin-1 subcellular localization regulates SREBP-1 signaling and governs the assembly and secretion of very-low-density lipoproteins (VLDLs).1, 4 It is tempting to speculate that ethanol-induced nucleocytoplasmic shuttling may activate SREBP-1 and impair VLDL secretion and, subsequently, contribute to hepatic fat accumulation.
Another major novel finding of the present study is that ethanol up-regulates lipin-1 largely through inhibition of AMPK and activation of SREBP-1. Our study provides evidence, for the first time, to our knowledge, that AMPK is involved in the regulation of lipin-1 gene expression. However, the exact mechanism by which AMPK inhibition by ethanol leads to activation of SREBP-1, and subsequent inhibition of Lpin1, remains to be determined. Our earlier work showed that ethanol selectively increases hepatic SREBP-1 activity in rodent models through inhibition of AMPK.6, 9 AMPK directly phosphorylates SREBP-1 and suppresses SREBP-1 activity in hepatocytes exposed to high glucose.26 Conceivably, ethanol-mediated inhibition of AMPK may cause reduced phosphorylation of SREBP-1, which, in turn, results in activation of proteolytic processing and transcriptional activity of SREBP-1 and, ultimately, increased lipin-1 gene expression. Moreover, several lines of evidence have suggested functional connections between SIRT1 and AMPK.27, 28 Ethanol-mediated impairment of the SIRT1/AMPK axis, and the ensuing enhancement of SREBP-1 activity, has emerged as a central mechanism in the development of AFLD. Therefore, it is tempting to speculate that ethanol may up-regulate lipin-1 through these metabolic modulators.
In summary, we have demonstrated that ethanol feeding up-regulates hepatic lipin-1 mRNA and protein and promotes lipin-1 cytoplasmic localization, which, in turn, may lead to increased lipogenesis, impaired fatty acid oxidation, and the development of steatosis in mice. Furthermore, we have identified lipin-1 as a vital downstream target of AMPK-SREBP-1 signaling in the action of ethanol in the liver. Our study sheds new light on the multifactorial pathogenesis of AFLD, and suggests that future studies to investigate the effects of nutritional or pharmacological inhibitors of SREBP-1, lipin-1, and/or activators of AMPK on its development are clearly warranted.
The authors thank Dr. David N. Brindley (University of Alberta, Edmonton, Canada) and Drs. R. Kennedy Keller and Laura Flatow (University of South Florida, Tampa, FL) for their outstanding technical and intellectual contributions.