Potential conflict of interest: Nothing to report.
View this article online at wileyonlinelibrary.com.
Development of hepatic steatosis and its progression to steatohepatitis may be the consequence of dysfunction of several metabolic pathways, such as triglyceride synthesis, very low-density lipoprotein (VLDL) secretion, and fatty acid β-oxidation. Peroxisome proliferator-activated receptor γ coactivator-1β (PGC-1β) is a master regulator of mitochondrial biogenesis and oxidative metabolism, lipogenesis, and triglyceride (TG) secretion. Here we generated a novel mouse model with constitutive hepatic activation of PGC-1β and studied the role of this transcriptional coactivator in dietary-induced steatosis and steatohepatitis. Selective activation of PGC-1β within hepatocytes is able to protect the liver from lipid overload and from progression to fibrosis. The protective function exerted by PGC-1β is due to its ability to induce mitochondrial oxidative phosphorylation, fatty acid β-oxidation, and citrate cycle, as well as to decrease oxidative stress and promote TG secretion in the blood stream. These findings bolster the concept that a combined hepatic specific action of PGC-1β on lipid synthesis and secretion, as well as on mitochondrial biogenesis and function, could protect against steatohepatitis. (HEPATOLOGY 2013)
Nonalcoholic fatty liver disease (NAFLD) is becoming a master component of the epidemic of obesity and metabolic syndrome worldwide due to excessive caloric intake.1 The spectrum of NAFLD ranges from simple fatty liver with benign prognosis to a potentially progressive form, nonalcoholic steatohepatitis (NASH), which may lead to liver fibrosis and cirrhosis resulting in increased morbidity and mortality. The initial phase of the steatosis or fatty liver is characterized by accumulation of fat droplets within the cytoplasm of a hepatocyte. Although simple steatosis is generally an asymptomatic syndrome with a benign course, it may progress through the inflammatory phase of steatohepatitis. Indeed, some cases develop a necroinflammatory state, hepatocytes ballooning with Mallory's hyaline, and sometimes fibrosis, features that could result in cirrhosis and, in some patients, hepatocellular carcinoma.2 Despite being potentially severe, little is known about the natural history or prognostic significance of NASH. Steatohepatitis may develop as a consequence of dysfunction of several metabolic pathways, such as triglyceride (TG) synthesis, very low-density lipoprotein (VLDL) secretion, and fatty acid β-oxidation. Indeed, one main determinant in the pathogenesis of fatty liver seems to be an increment in the serum fatty acid pool. The sources of fat contributing to fatty liver are peripheral TGs stored in white adipose tissue that are driven to the liver in the form of plasma nonesterified fatty acids (NEFAs), dietary fatty acids, and hepatic de novo lipogenesis (DNL).3 It has been recently demonstrated that, as far as TG content in the livers of patients with steatosis is concerned, 60% are synthesized from NEFAs, over 10% derive from the diet, and close to 30% arise from DNL.4 Although TGs can either be stored as lipid droplets within hepatocytes or secreted into the blood as VLDL particles, they can also be hydrolyzed to supply fatty acids for β-oxidation in the mitochondria, depending on the nutritional status of the organism.5 The metabolic partitioning of fatty acids between mitochondrial β-oxidation and TG synthesis is critically regulated. In the liver, fatty acid β-oxidation is normally inhibited by food intake through the action of insulin, which is the main regulator of DNL due to its direct activation of SREBP1c.6 In addition, when mitochondrial β-oxidation is saturated, as in the case of steatosis with a great amount of fatty acids, a negative feedback occurs due to excessive production of acetyl-CoA and reducing equivalents feeding electrons to the respiratory chain, with massive production of reactive oxygen species.7 Indeed, oxidative stress leading to lipid peroxidation may be the culprit of the necroinflammatory changes characteristic of NASH and of alcohol-induced steatohepatitis.8
Metabolic pathways controlled at the transcriptional level often depend on changes in the amounts or activities of transcription factors involved in their regulation and this represents undoubtedly a major mode of regulation. Peroxisome proliferator-activated receptor γ coactivator (PGC-1) coactivators, PGC-1α and PGC-1β, are master regulators of mitochondrial biogenesis and oxidative metabolism as well as of antioxidant defense. Hepatic PGC-1α and PGC-1β gene expression is strongly increased by fasting.9-11 PGC-1 coactivators are responsible for a complex program of metabolic changes that occur during the shift from fed to fasted state, including modifications in gluconeogenesis, fatty-acid β-oxidation, ketogenesis, heme biosynthesis, and bile-acid homeostasis. The transition between fed and fasting state-mediated by PGC-1α in liver is achieved by coactivating master hepatic transcription factors, such as HNF4α, PPARα, GR, Foxo1, FXR, and LXR.10 Both PGC-1α and PGC-1β are able to activate expression of PPARα target genes involved in hepatic fatty acid oxidation.12, 13 Conversely, only the overexpression of PGC-1β in hepatocytes stimulates TG production and secretion from the liver through the activation of genes involved in lipogenesis by way of direct coactivation of SREBP1c.14, 15 Moreover, several data link PGC-1β with the LXR pathway. PGC-1β is recruited to the promoter region of CYP7a1 and ABCA1 and activates the expression of these target genes. More recently, have it has been shown that Foxa2 interacts with PGC-1β to increase serum TG by activating Mttp and Dgat2 expression.16, 17 Additionally, it has been demonstrated that PGC-1β and its target gene apolipoprotein C3 are down-regulated by nicotinic acid that mediates TG-lowering effects and is widely used for treating hypertriglyceridemia.18, 19 Finally, whole body ablation of PGC-1β impairs hepatic lipid metabolism in response to acute high fat dietary loads, resulting in hepatic steatosis.18
Given the importance of hepatic lipid and mitochondrial metabolic dysfunctions in NASH, we wondered whether the PGC-1β regulatory network could represent a potential new therapeutic target for this hepatic disease. Thus, our aim was to address the contribution of PGC-1β constitutive activation during the development of steatohepatitis and steatosis. Here, using two nutritional (methionine choline-deficient diet [MCD] and high-fat diet) models of NASH and NAFLD in hepatic transgenic mice for PGC-1β, we show that the overexpression of this coactivator ameliorates hepatic steatosis, reduces lipid overload in the hepatocytes, and reduces the fibrotic and apoptotic phenotype. This outcome was mediated by the ability of PGC-1β to sustain the expression of target genes of several metabolic pathways that are severely compromised during steatohepatitis.
Materials and Methods
Animal Study and Transgenic Mouse Generation.
To generate pLiv.7 PGC-1β, first the hPGC-1β (3.1 kb) fragment with KpnI and XhoI restriction sites was generated by polymerase chain reaction (PCR) from pcDNA3 PGC-1β plasmid. The fragment was inserted into the KpnI and XhoI site of pLiv7, which carries the promoter, exon 1 and a fragment of exon 2 of apolipoprotein E, a protein expressed exclusively in the liver. The LivPGC-1β transgenic mice were generated by injecting into the pronuclei of the fertilized eggs of the FVB/N mice the transgene plasmid digested with SpeI. Mice carrying the transgene were identified by PCR of genomic DNA to confirm the presence of the hPGC-1β coding sequence. Stomach, liver, brain, kidney, jejunum, duodenum, ileum, and colon of transgenic mice were dissected and prepared for total RNA extraction and immunohistochemistry to evaluate the specific hepatic expression of transgene under the apolipoprotein E promoter control. For dietary protocol, wildtype and LivPGC-1β mice were randomly divided into two experimental groups and fed with MCD, high-fat diet (D12451, Research Diets), and their control diets (MCS and chow diet respectively) for 8 weeks. During the experimental period, individual body weight was recorded every 5 days.
All results are expressed as mean ± standard error of the mean (SEM). Data distribution and gene expression statistical analysis were performed using NCSS statistical and power analysis software 2007. Comparisons of two groups were performed using a Student t test followed by the Mann-Whitney U test where appropriate. P < 0.05 was considered significant.
ATPβsynt, ATPβ-synthase; Cpt-1α, carnitine palmitoyl transferase-1α; COXI, cytochrome c oxidase subunit I; cytC, cytochrome C; DNL, de novo lipogenesis; Dgat1 and 2, diacylglycerol acyltransferase 1 and 2; IL-1β, interleukin β; Idh3α, isocitrate dehydrogenase 3α; Mcad, medium-chain acyl-coenzyme A dehydrogenase; MCD, methionine and choline deficient diet; MCS, methionine and choline supplied diet; NAFLD, nonalcoholic fatty liver disease; NASH, nonalcoholic steatohepatitis; NEFAs, nonesterified fatty acids; Scd-1, stearoyl Co-A Desaturase 1; pro-col, procollagen; Tnfα, tumor necrosis factor α.
Hepatic Pgc-1β Induce Mitochondrial Metabolism and Respiration.
In order to verify that a constitutive overexpression of PGC-1β in the liver was able to induce its target genes, we first generated a mouse model in which human PGC1-β is selectively overexpressed in the liver (LivPGC-1β mice) by subcloning the hPGC1-β coding sequence under the control of the apolipoprotein E promoter. The human PGC-1β is expressed only in the liver of transgenic mice (Supporting Fig. 1). In order to characterize the tissue-specific transcriptional scenario activated by PGC-1β, we performed microarray analysis of liver samples from wildtype and LivPGC-1β mice fed a chow diet. The data showed that PGC-1β coactivator overexpression is able to induce a plethora of genes involved in several metabolic pathways (Fig. 1A). The majority of target genes whose expression is enhanced by PGC-1β (1.3-fold or more) encodes for proteins involved in the mitochondrial oxidative phosphorylation. Other pathways up-regulated by the hepatic PGC-1β overexpression were ubiquinone and protein biosynthesis, lipid metabolism, TG transport, citrate cycle, gluconeogenesis, and antioxidant systems. These results were confirmed by real-time quantitative (qPCR) analysis of the gene expression levels of cytochrome c (cytC), a component of the respiratory chain, as well as of medium-chain acyl-coenzyme A dehydrogenase (Mcad) and carnitine palmitoyl transferase-1α (Cpt-1α), two key enzymes in fatty acid β-oxidation (Fig. 1B). Moreover, real-time qPCR analysis confirmed that overexpression of PGC1-β was associated with the induction of genes involved in lipid anabolism, including Srebp1c and its target gene, Fas, both involved in fatty acid synthesis. Notably, also the expression of Stearoyl Co-A Desaturase 1 (Scd-1) that catalyzes the biosynthesis of monounsaturated fatty acids, and diacylglycerol acyltransferase 1 and 2 (Dgat1 and 2), fundamental enzymes for TG synthesis, were increased by the overexpression of hepatic PGC1-β (Fig. 1B). Protein analysis revealed that both porin and the mitochondrial DNA-encoded cytochrome c oxidase subunit I (COXI), which are well-known markers of mitochondrial biogenesis and function, were increased in livers from LivPGC-1β mice (Fig. 1C). To verify the functional impact of the induction of mitochondrial genes we measured COX and citrate synthase (a nuclear-encoded mitochondrial matrix marker of citrate cycle) enzyme activity in total liver lysates. As expected, the analyses confirmed that the specific activities of the two enzymes were increased in LivPGC1-β hepatocytes as compared with the controls (Fig. 1D). In summary, liver-specific PGC1-β overexpression is able to induce mitochondrial functions through the induction of proteins involved in citrate cycle and oxidative phosphorylation, as well as to enhance the expression of genes involved in TG biosynthesis.
Hepatic PGC-1β Affects Serum TG Homeostasis.
To test whether the ability of PGC1-β overexpression to increase the expression of genes encoding for proteins involved in TG biosynthesis could cause changes in the circulating lipids, we measured serum lipid levels of wildtype and LivPGC1-β mice fed a standard diet (chow diet). Different from previous data reporting that overexpression of PGC1-β by way of adenovirus infections dramatically increases TG levels in the blood,20 we only recorded a slight, but significant, induction in serum TG levels of LivPGC-1β mice when compared with their littermate controls (Fig. 2A). However, after the injection of Triton WR-1339 that prevents the catabolism of VLDL by inhibiting lipoprotein lipase activity, the level of TG secretion of LivPGC-1β mice fed with a normal chow diet was 3-fold higher than control littermates (Fig. 2B). Given that PGC-1β is also able to induce the expression of genes involved in fatty acid catabolism and since the serum TG levels did not reach 2-fold induction in LivPGC-1β mice despite their higher TG secretion rate, we decided to test whether PGC-1β could induce an efficient substrate switch in response to a lipid overload. For this purpose, we administered an intragastric olive oil load by gavage to wildtype and LivPGC-1β mice and followed the clearance of serum TG and NEFAs at different timepoints. Strikingly, whereas both wildtype and LivPGC-1β mice showed a massive increase in TG levels at 2 hours after administration of olive oil in LivPGC-1β mice the TG concentration returned to almost normal levels 4 hours after the fat load. In contrast, in wildtype mice TG concentrations reached basal levels only 6 hours from oil injection, suggesting that overexpression of PGC-1β in the liver promotes TG clearance from the blood under the condition of lipid overload (Fig. 2C). A similar effect was observed for the serum concentration of NEFAs, although their clearance was blunted if compared with TG clearance (Fig. 2D). Taken together, these data imply that specific overexpression of PGC-1β in the liver leads to TG and NEFA clearance by activating mitochondrial β-oxidation of fatty acids.
PGC-1β Ameliorates Steatohepatitis Induced by MCD Diet.
Since both mitochondrial functions and lipid metabolism are severely compromised during steatohepatitis, we decided to examine whether the constitutive activation of PGC-1β in the liver could affect the development of the disease. In order to cause steatohepatitis we fed LivPGC-1β mice and their control littermates a diet deficient in choline and methionine (MCD diet). Indeed, choline is an FDA-classified essential nutrient with roles in cell membrane integrity, transmembrane signaling, and phosphatidylcholine synthesis.21 The role of dietary choline deficiency in promoting hepatic steatosis and reduced plasma VLDL levels is well established in the literature. This was thought to be due to impaired synthesis of phosphatidylcholine resulting in diminished VLDL assembly and secretion and consequently reduced TG clearance. Moreover, the lack of methionine reduces glutathione synthesis, thus increasing reactive oxygen species (ROS) accumulation, mitochondrial DNA damage, and apoptotic cell death, all features of NASH.22 Indeed, mice fed a diet that is deficient in both choline and methionine develop inflammation and hepatic fibrosis in addition to simple steatosis.23 After 8 weeks of an MCD diet, the gross morphology of livers of LivPGC-1β appeared less fatty compared with that of wildtype mice clearly presenting steatotic liver engrossed with lipids (Fig. 3A). The body weight/liver ratio in wildtype mice significantly decreased if compared with the standard (MCS) diet, while the LivPGC-1β mice fed an MCD diet did not present a significant decrease in the same ratio (Fig. 3B). Furthermore, the histological analysis showed a severe macrovescicular steatosis, hepatocellular necrosis, and mixed inflammatory infiltrates in wildtype mice fed a steatogenic diet (Fig. 3C). LivPGC-1β mice presented less inflammatory infiltrates and milder steatosis, as confirmed from quantization of hepatic ballooning that was reduced by more than 50% in transgenic versus control mice fed an MCD diet (Fig. 3D). Therefore, constitutive activation of PGC-1β in the liver ameliorates steatotic phenotype, necrosis, and inflammatory infiltrates in dietary mouse models of steatohepatitis.
Role of Hepatic PGC-1β on Lipid Profile in Mice Fed an MCD Diet.
The MCD diet usually reduces TG levels in serum, as a consequence of TG retention within the hepatocytes. Thus, in order to verify whether the histological differences in the LivPGC-1β could be explained by an altered TG turnover, we measured serum and hepatic lipid levels. Indeed, the wildtype mice fed an MCD diet presented a marked reduction of serum TG compared with the MCS control diet, whereas the LivPGC-1β transgenic mice did not show significant differences in serum TG levels (Fig. 4A). Conversely, both mouse lines revealed a massive decrease in circulating cholesterol (Fig. 4A). Consistently, the MCD diet caused increased levels of intrahepatic TGs and cholesterol in wildtype, but not in LivPGC-1β mice (Fig. 4B). The hepatic steatosis and lipid content were also confirmed by Red Oil staining that revealed a massive retention of neutral lipids organized in macrovescicular lipid droplets within the hepatocytes of wildtype mice, whereas LivPGC-1β mice showed only small lipid droplets (Fig. 4C). NASH has been shown to be associated with an oxidative stress condition of the hepatocyte.24 Indeed, total hepatic TBARS, a classical marker of lipid peroxidation, were strongly increased in wildtype mice fed an MCD diet. Strikingly, the overexpression of PGC-1β in the liver almost completely prevented the accumulation of lipid peroxides in transgenic mice fed an MCD diet (Fig. 4D). Therefore, PGC-1β overexpression sustains lipid secretion into the circulation by protecting from oxidative stress, thus preventing hepatic lipid retention.
Role of PGC-1β on Fibrosis and Apoptosis in Mice Fed an MCD Diet.
The presence of hepatic stellate cells (HSC) activation and fibrosis is one of the main differences that distinguish NASH from simple steatosis. To test whether the improvement of steatotic phenotype in LivPGC-1β mice also affects HSC activation and the fibrotic state, we carried out immunohistochemistry for alpha-smooth muscle actin (α-SMA) and a Sirius staining of collagen in liver sections. Wildtype mice fed an MCD diet showed increased α-SMA staining compared with their control group (Fig. 5A,B). Likewise, wildtype mice fed and MCD diet presented a 2-fold increase in collagen content as compared with LivPGC-1β mice, suggesting a higher rate of MCD diet-induced HSC activation (Fig. 5C,D). On the other hand, both in wildtype and transgenic mice fed an MCS diet, the collagen was detectable only in the periductal cell compartment (Fig. 5C). Hepatocellular apoptosis is emerging as an important, if not critical, mechanism contributing to the progression of human liver disease. Engulfment of apoptotic bodies by HSCs stimulates their fibrogenic activity, thus likely leading to fibrosis.25 The histological analysis by terminal deoxynucleotidyl transferase mediated dUTP nick end labeling (TUNEL) assay showed no apoptotic cells in wildtype and LivPGC-1β mice fed a control diet, whereas wildtype mice fed an MCD diet displayed a 3-fold higher content of apoptotic cells if compared with the LivPGC-1β counterparts (Fig. 5E,F). Consistent with FFA-mediated hepatocyte apoptosis as a pathogenic mechanisms in NASH,26 PGC-1β was able to counteract hepatocyte cell death due to lipid accumulation by promoting TG clearance through mitochondrial β-oxidation as well as by avoiding lipid peroxidation by way of induction of free radical scavenging (Fig. 1A). Thus, PGC-1β seems to be able not only to prevent lipid accumulation in animals fed a steatogenic diet, but also to blunt fibrosis and apoptotic cell death of hepatocytes.
PGC-1β Sustains and Promotes Several Metabolic Pathways in Steatotic Livers.
PGC-1β is a coactivator of several transcription factors involved in different metabolic functions; thus, it is reasonable to presume that the protection of hepatocytes from lipid overload, ballooning degeneration, fibrosis, and cell death is due to its transcriptional potential. To test whether PGC-1β is able to induce its target genes during methionine-choline feeding, we performed microarray analysis of liver samples from wildtype and LivPGC-1β mice fed an MCS and MCD diet. Gene expression analysis in wildtype mice fed an MCD diet revealed down-regulation of fatty acid and xenobiotic metabolism, steroid and bile acid biosynthesis, as well as up-regulation of genes involved in HSC activation and fibrosis, ROS production, interleukin signaling, and phospholipid degradation (data not shown). The same analysis performed on wildtype and LivPGC-1β mice fed a steatogenic diet showed that PGC-1β was able, at the same time, to induce some metabolic pathways and to sustain the expression of genes whose transcription was compromised during steatohepatitis in wildtype mice fed an MCD diet (Fig. 6A). The majority of target genes whose expression was increased by PGC-1β (from 1.3-fold) encode for proteins that take an active part in the oxidative phosphorylation and citrate cycle. Other pathways induced by the coactivator are ubiquinone and bile acid biosynthesis, fatty acid metabolism, as well as glycolysis and gluconeogenesis (Fig. 6A). On the other hand, the overexpression of PGC-1β was able to protect hepatocytes against the MCD diet-induced up-regulation of genes involved in detrimental pathways such as cancer and apoptosis, inflammatory response, hepatic steatosis, and fibrosis (Fig. 6B). Moreover, we confirmed by real-time qPCR that the gene expression of ATPβ-synthase (ATPβsynt), cytC (oxidative phosphorylation), isocitrate dehydrogenase 3α (Idh3α) (citrate cycle), Dgat1, Scd-1 (TG synthesis), and Cyp7a1 (bile acid biosynthesis) was increased in livers from LivPGC-1β mice fed an MCD diet as compared with their wildtype controls (Fig. 6C). The sustained expression of Scd-1 is very interesting since it has been shown that inhibition of Scd-1 activity decreases triglyceride accumulation, but in turn increased lipotoxicity.27 On the other hand, the expression of procollagen (pro-col), tumor necrosis factor α (Tnfα), and interleukin β (IL-1β) was reduced in the transgenic mice (Fig. 6C). In order to confirm the effects of PGC-1β overexpression at a functional level, we measured COX and citrate synthase activity in total liver lysates. Similar to the analyses carried out on animals fed a standard (chow) diet, the activities of Complex IV and citrate synthase were increased in LivPGC1-β hepatocytes (Fig. 6D), reflecting enhanced mitochondrial biogenesis and function.
Taken together, these results suggest that the constitutive activation of PGC1-β within the hepatocytes is able to prevent the transcription of genes encoding for proteins involved in fibrosis, steatosis, and apoptosis, to sustain the expression of proteins that are greatly reduced during steatohepatitis, as well as to enhance the transcription of other proteins whose functions might be at the basis of the protective effect driven by this coactivator.
PGC-1β Protects Liver Against HFD-Induced Steatosis.
The MCD model is arguably the best-established model to study the inflammatory and fibrotic elements of the NAFLD spectrum. Despite this, there is little evidence to support the idea that this dietary model could replicate either the phenotype of metabolic syndrome usually associated with NAFLD and steatohepatitis in humans. Indeed, animals fed an MCD diet lose weight and are not insulin resistant. Moreover, PGC-1β seems to be able to influence lipid metabolism and oxidative phosphorylation, thus acting as a key player in the protection of the liver against one of the main insults that characterizes the development of steatohepatitis, represented by the lipid accumulation within hepatocytes. To test the ability of PGC-1β to ameliorate steatosis, wildtype and LivPGC-1β mice were fed a high-fat diet (HFD) containing 35% fat. Similar to MCD feeding, after 8 weeks of an HFD diet the gross morphology of livers of transgenic mice appeared healthier compared with that of wildtype mice that showed steatotic liver (Fig. 7A). Histological analysis revealed that wildtype mice challenged with HFD developed severe steatosis with macrovescicular lipid droplets, whereas LivPGC-1β mice did not show the characteristic ballooning injury of fatty liver (Fig. 7B). Hepatic lipid analysis showed a 50% increase in TG levels and a dramatic rise of cholesterol in HFD fed wildtype liver compared with the same group fed with a standard diet (chow) (Fig. 7C). Conversely, LivPGC-1β mice presented only a slight accumulation of TG in the liver and a moderate increase of cholesterol when compared with wildtype mice fed the same diet (Fig. 7C). Oil Red staining revealed the massive accumulation of neutral lipids within wildtype hepatocytes in mice fed with HFD compared with controls, while it demonstrated mild amount of lipids stored in microvesicles in LivPGC-1β mice (Fig. 7D). To gain insight into the mechanism by which the overexpression of PGC-1β leads to hepatocyte protection against lipid overload, we examined the expression of genes implicated in mitochondrial function and lipid synthesis. Messenger RNA (mRNA) levels of ATPβsynt, cytC, Idh3α, Dgat1, Scd-1, and Fas were increased in livers from LivPGC-1β mice fed an HFD diet as compared with their wildtype controls (Fig. 7E). Remarkably, PGC-1β is able to sustain the expression of Scd-1 that is strongly decreased by HFD feeding (data not shown), similar to the dietary model of steatohepatitis.
Taken together, these results demonstrate that the hepatic overexpression of PGC-1β prevents lipid accumulation within the hepatocytes during high-fat feeding, thus protecting very efficiently from simple steatosis.
This work shows that hepatic PGC-1β is able to stimulate mitochondrial functions through the induction of key enzymes involved in oxidative phosphorylation, citrate cycle, pyruvate, and lipid metabolism, as well as to induce genes involved in TG metabolism and secretion by way of VLDL in the bloodstream. Indeed, consistent with previous models of overexpression of PGC-1β in the liver by way of adenoviral administration, we showed increased TG levels in the blood at basal level and higher rate of TG secretion.14, 20 The mechanisms responsible for the development and progression of NASH in humans are complex and have not been fully delineated. Thus, given the fact that mitochondrial functions and lipid metabolism are often compromised in nonalcoholic liver disease, we wondered whether the modulation of PGC-1β in the liver could influence the progression of NASH. The MCD dietary animal model mimics the characteristic pathology of steatohepatitis found in human with mixed cell inflammatory infiltrates, hepatocellular death, and pericellular fibrosis. Lipids initially accumulate in the liver through one or more of the following mechanisms: increased fatty acid uptake, increased TG synthesis, decreased β-oxidation, or decreased hepatic export of TG by way of VLDL. Indeed, PGC-1β not only ameliorates steatotic status avoiding lipid retention, but also contributes to the protection of the hepatocytes from other insults occurring during the development of NASH. It was previously shown that NASH is associated with a dramatic increase in total lipid peroxides in the liver, this being consistent with the hypothesis that the necroinflammatory lesions might result from oxidative stress. The PGC-1β induction of the expression of several antioxidant defenses, such as the NRF2-mediated oxidative stress response, the free radical scavenging systems, and the glutathione metabolism (see Fig. 1A), together with the significant decrease in lipid peroxide content in LivPGC-1β livers after treatment with an MCD diet, clearly indicate that PGC-1β confers protection to the liver from oxidative damage. This would also prevent oxidative modification of the cytoskeleton proteins associated with impaired VLDL secretion of steatotic hepatocytes. Hepatic fibrosis attributable to HSC activation is a well-recognized end result of injury occurring from a wide variety of insults to the liver. Several studies have shown increased fibrosis in mice fed an MCD diet, with more advanced fibrosis correlating with greater steatohepatitis.1 The significant lower liver fibrosis in LivPGC-1β mice fed an MCD diet, correlating with milder steatotic phenotype, adds further proof of the protective role of PGC-1β in steatohepatitis. Moreover, some studies identified a relationship between hepatocyte apoptosis and fibrosis in NASH.28 Although enhanced hepatocyte apoptosis in fibrotic NASH may simply indicate disease severity, increasing evidence suggests a causal relationship between the two processes. The apoptotic body engulfment of HSCs is associated with further activation of these cells.25
Although originally considered a physiological event, unorchestrated and continuous apoptosis in the liver likely contributes to liver disease progression in NASH. Indeed, in contrast to physiological apoptosis occurring in organ development and in some epithelial tissues, hepatocyte apoptosis not only could be the result of lipid accumulation and inflammation, but may in turn worsen inflammation and fibrosis, establishing a vicious cycle. Whether cellular apoptosis is a primary mechanism promoting steatohepatitis or is a secondary phenomenon resulting from tissue inflammation is under investigation, but the evidence that PGC-1β seems to avoid cell death in steatotic liver suggests an important role of this coactivator in cellular survival during the development of NASH, thus avoiding the causal relationship between apoptosis and fibrosis that could lead to the progression of steatohepatitis to more severe liver diseases, such as cirrhosis and hepatocarcinoma. Taken together, our findings suggest PGC-1β coactivator as a potential player in the hepatocyte protection against steatohepatitis. Indeed, the ability of PGC-1β mice to induce mitochondrial β-oxidation and promote TG clearance in the blood, together with the ability to conserve the expression of metabolic pathways whose transcription is greatly compromised during steatohepatitis, might be the main mechanisms by which PGC-1β overexpression protects liver from steatohepatitis. In support of the theory that PGC-1β is able to protect from steatohepatitis acting on lipid accumulation through mitochondrial functions and TG clearance, its constitutive activation in mice fed an HFD diet protected also against steatosis. In contrast with previous studies that reported that the PGC-1β dependent up-regulation of mitochondrial proteins is not sufficient to prevent lipid overload in animals fed with HFD,20 in our models hepatic triglyceride and cholesterol levels are greatly reduced, leading to an improvement of steatotic phenotype. These discrepancies could be due to the different models used for this study, since our mice with constitutive overexpression of PGC-1β were challenged with a chronic high-fat feeding. Nevertheless, it could be interesting to better investigate the differences between acute and chronic overexpression of this coactivator with short- and long-term steatogenic diets.
In conclusion, this work bolsters the concept that a combined action of PGC-1β on lipid synthesis and secretion, as well as on mitochondrial β-oxidation and oxidative phosphorylation, could ameliorate liver disease in steatosis and steatohepatitis progression.
We thank S.A. Kliewer, J.M. Taylor, and A. Vidal-Puig for their tools and support.