Rev-erbα mediates steatosis in alcoholic fatty liver through regulating autophagy

AFL is a liver disease caused by long-term excessive drinking, it is characterized by steatosis. Understanding the regulatory mechanism of steatosis is crucial for the treatment of AFL. Rev-erbα has been implicated in regulation of lipid metabolism. However, the role and the underlying mechanisms of Rev-erbα in AFL remains unknown. In this study, the antagonists or agonists of Rev-erbα as well as Rev-erbα shRNA were applied in vitro and vivo . Triglyceride and lipid droplets accumulation were measured by using TG kit and ORO staining. Lipid synthesis related factor Srebp1c and lipid β-oxidation regulatory factor Pparα were measured by using Western blot, qRT-PCR and immunohistochemistry analysis. Autophagy activity was measured by western blot and electron microscope, and lysosomal probe was used to labeled lysosomal acidity. We observed that the expression of Rev-erbα was significantly increased in vivo and vitro, and Rev-erbα activation mediated steatosis in L-02 cells. then, inhibition/down-expression of Rev-erbα improved the triglyceride and lipid droplets accumulation and the abnormal expression of Pparα and Srebp1c through enhancing the autophagy activity. Furthermore, down-expression of Rev-erbα up-regulated the nuclear expression of Bmal1, which regulated the autophagy activity in vitro . Collectively, these findings indicate that Rev-erbα induces liver steatosis and leads to the progression of AFL. Our study reveals a novel steatosis regulatory mechanism in AFL and suggest that Rev-erbα might be a potential therapeutic target for AFL.


Introduction
Chronic alcohol consumption is a crucial factor contributing to alcoholic liver disease (ALD). ALD includes a broad spectrum of liver disorders, ranging from alcoholic fatty liver (AFL), alcoholic steatohepatitis (ASH), alcoholic fibrosis (AF), alcoholic cirrhosis (AC) to alcoholic hepatocellular carcinoma (AHCC) [1][2][3]. Now, alcoholic fatty liver disease (AFLD) has become a global healthcare problem. Among AFLD, AFL is the earliest phase characterized by ballooning of hepatocytes, lipid droplets deposition and inflammatory cells infiltration. Hepatic steatosis is a progression of excessive triglyceride accumulation caused by the imbalance between lipids synthesis and oxidation [4]. As the main pathological factor, lipid accumulation plays a pivotal role in the occurrence and progression of AFL [5,6]. Although mild fatty liver can be alleviated by exercise and diet, the regulatory mechanism that causes steatosis remains to be supplemented.
Rev-erbα is an orphan nuclear receptor, which belongs to nuclear receptor family.
It is associated with many diseases, including hyperlipidemia, hyperglycemia and liver diseases such as liver fibrosis and cancer [7][8][9]. Among this, Rev-erbα is closely related to lipid metabolism, this is supported by the fact that Rev-erbα can promote the differentiation of 3T3-L1 preadipocytes and enhance lipid storage [10]. Moreover, Rev-erbα regulates lipoproteinase and triglyceride by directly inhibiting the activity of ApoC-III [11,12]. As a transcriptional repressor that is expressed in a circadian manner, Rev-erbα has a dual role in regulating lipid metabolism. Dan Feng et al. has pointed out that lipid deposition increased significantly in mice lacking Rev-erbα. [13]. Consistent with this view, Sitaula S et al. also thought that Rev-erbα could repress transcription of cholesterol biosynthesis genes by promoting recruitment of NCoR and HDAC3, resulting in reducing cholesterol levels and biosynthesis in mice. [14]. However, the function of Rev-erbα in the pathogenesis of AFL and whether Rev-erbα regulates hepatic steatosis conversion remains unclear. Hence, it is important to understand the potential molecular mechanisms underlying the control of lipid metabolism in AFL.
Autophagy is a protective mechanism for removing lipid droplets, protein aggregates, and damaged organelles from hepatocytes [15]. Growing evidence highlights the involvement of autophagy in regulating hepatic lipid metabolism. It has been reported that inhibition of autophagy in hepatocyte could increase the storage of triglyceride in lipid conjoined as well as inhibit the degradation of lipid droplets [16].
Furthermore, lack of autophagy activity alters fatty liver and liver injury condition induced by alcohol [17]. Attractively, it was pointed out that Rev-erbα can represses autophagosome formation and lysosomal biogenesis directly in skeletal muscle [18,19]. Grimaldi B et al. have pointed out that Rev-erbα can inhibit the formation of autophagy, which blocked the source of cancer nutrition [20]. More importantly, the expression of autophagy genes in zebrafish is under the control of Rev-erbα [21].
Given the critical role of autophagy on lipid metabolism as well as the intimate relationship between Rev-erbα and autophagy, we want to explored whether Rev-erbα has an effect on accelerating lipid accumulation in AFL by impacting on the activity of autophagy.
In this study, we found that the expression of Rev-erbα was significantly increased both in vitro and in vivo. Following inhibition of Rev-erbα, hepatic steatosis was ameliorated with the improvement of autophagy activity. Mechanistic studies suggest that Rev-erbα inhibited autophagy by regulating the expression of the liver circadian clock gene Bmal1. Taken together, our results have elucidated that Rev-erbα accelerated lipid deposition by inhibiting autophagy, and we sought to defined the potential roles of Rev-erbα in hepatic steatosis and the molecular mechanisms underlying this regulation in AFL. and humidity (50±5%) environment with a 12 h light/dark cycle with ad lib access to food and water. Mice were randomly divided into Control diet (CD-fed), ethanol diet (EtOH-fed). Ethanol diet feeding and binge was performed with the protocol described by Gao-Binge [22]. According to protocol, mice were fed the regular Lieber-DeCarli normal diet or ethanol diet (Nantong troffi, CAS number: TP4030) containing increasing (1%-5% vol/vol) ethanol for the adaptation period (5 days) and modeling (10 days) with 5% vol/vol ethanol liquid diet at 5 o'clock every afternoon.

Reagents and antibodies
Half an hour after feeding, SR8278 group were injected with SR8278 (dissolve in

RNA extraction and Quantitative Real-Time PCR
Total RNA was isolated from liver tissue or L-02 cells using the TRIzol reagent

Immunofluorescence
Cells cultured in 6-well were washed by PBS 3 times and fixed in 4% formaldehyde for 15 min. Then washed by PBS 3 times and blocked by BSA for 30 min at room temperature. Next, cells were washed with PBS 3 times and permeabilized with 1% Triton X-100 solution for 5 min. After washed, cells were incubated with Rev-erbα primary antibody (1:50) overnight at 4 ℃. Next day, cells were incubated with secondary antibody at dark for 1 hand counterstained with DAPI for 5 min. At last stained sections were examined by using confocal microscopy (Zeiss, Germany).

Serum biochemical analysis
The activities of serum alanine aminotransferase (ALT), triglyceride (TG) and total cholesterol (T-CHO) in serum were measured using commercial assay kits (Jiancheng, Nanjing) by microplate reader (Biotek, USA ) at appropriate wavelength.

ORO staining
Cells in the 6-well plates were washed 3 times by PBS, then fixed with 4% Paraformaldehyde for 15 min, After washed 3 times by PBS, sections were stained with working solution ORO ( prepared freshly at 25 ℃) for 30 min. Finally, sections were washed with 60% dimethylcarbinol and double distilled water. The fat drops were observed by inverted fluorescent microscope. The same operation was introduced to liver tissue after fixing.

Morphological assessment
Liver tissues were fixed with 4% paraformaldehyde for 24 h, then embedded in paraffin blocks and stained with hematoxylin and eosin (H&E). Immunohistochemistry (IHC) was performed according to a standard procedure. The

Transmission Electron microscope
Liver tissue sample at grain size was fixed in 2.5% glutaraldehyde at 4 ℃ and placed in 1% osmium tetroxide for 4 h on ice. Next sections were dehydrated in a graded series of ethanol and embedded in LR White resin after washed by 0.1 M sodium cacodylate (pH 7.4). Embedded samples were detected by a transmission electron microscope (Hitachi-7800, Japan).

Lysosomal acid analysis
Lyso Tracker Green DND-26 was selectively labeled for acidic lysosome in living cells. After cell slides and certain treatment, discard the medium and add 50 nM Lyso Tracker Green DND-26, cells were incubated for 5 min under growth, then switched to fresh medium and detected by laser confocal microscope. The intensity of fluorescence intensity represents the acidity of lysosomes.

Statistical analysis
All data presented were representative of at least 3 repeat experiments and expressed as mean ± SEM. Statistical analyses were performed with SPSS 17.0 software (Statistical Program for Social Sciences). One-way analysis of variance (ANOVA) was used to evaluate differences between each groups. The differences between two groups were determined by unpaired two-tailed t-test. Results were considered statistically significant with p value < 0.05.

Rev-erbα was up-regulated in the liver of EtOH-fed mice in vivo
We adopted the chronic EtOH plus a single EtOH binge feeding on mice described by the NIAAA model protocol to create Murine model of AFL. As displayed in Fig.1A, compared to CD-fed mice, apparent lipid deposition was showed in the liver of EtOH-fed mice. H&E staining showed that mice in the EtOH-fed group developed steatosis; and, fat droplets fill the hepatocytes, especially those hepatocytes located around the central vein. ORO staining indicated aggravated steatosis in the EtOH-fed mice compared to control mice. The ratio of liver weight to body, serum ALT, TG and T-CHO level were all increased in EtOH-fed mice compared to CD-fed mice (Fig.1B).
Results of western blot and qRT-PCR showed that Pparα was decreased and Srebp1c was increased in EtOH-fed mice compared to CD-fed mice (Fig.1C and 1D).
To explore whether Rev-erbs was involved in the pathogenesis of AFL, we detected the expression of two subtypes of Rev-erbs. qRT-PCR analysis showed that the level of Rev-erbα was higher than Rev-erbβ in CD-fed group and Rev-erbα was enhanced but Rev-erbβ had no obvious change in EtOH-fed group compared to CD-fed group (Fig.1F). The higher expression of Rev-erbα in EtOH-fed group was further confirmed by western blot and immunohistochemistry analysis ( Fig.1E and 1G).
These results indicated that it may be Rev-erbα rather than Rev-erbβ that plays a critical role in AFL.

Rev-erbα was up-regulated in EtOH-treated L-02 cells and mediated liver steatosis
In vitro, L-02 cells were treated with EtOH (150 mM, 48 h). As shown in S1A and S1B, lipid droplets and the level of TG were significantly increased in EtOH-treated cells compared to L-02 cells. Then, qRT-PCR analysis revealed that the expression of Pparα was decreased and Srebp1c was increased in EtOH-treated L-02 cells compared to L-02 cells, this result was further confirmed by western blot analysis (S1C and S1D). The above evidence showed that 150 mM EtOH could cause disorder of lipid metabolism in L-02 cells. Consistenting with in vivo results, the mRNA expression of Rev-erbα was higher than Rev-erbβ in L-02 cells, and that Rev-erbα was increased while Rev-erbβ had no obvious changed in EtOH-treated L-02 cells (Fig. 2B).
Western blot analysis further confirmed that Rev-erbα was increased in EtOH-treated L-02 cells ( Fig. 2A). Furthermore, immunofluorescence was used to detect intracellular distribution of Rev-erbα in L-02 cells, the result indicated that Rev-erbα was significantly elevated in the nucleus but almost unchanged in the cytoplasm after treatment with EtOH 48 h in L-02 cells, this founding was further demonstrated by western blot analysis (Fig. 2C and Fig. 2D).
Steatosis is the main pathological process of AFL [5,6]. Studies have shown that Rev-erbα can regulate lipid metabolism [10,11]. To investigate whether Rev-erbα was related to liver steatosis, L-02 cells were treated with Rev-erbα agonist GSK4112 at a fixed time of the day, at 5 p.m (10 μM, 24 h) [23,24]. Compared with L-02 cells, lipid droplets and TG level were significantly increased, and the protein of Pparα was down-regulated but Srebp1c was up-regulated in GSK4112-treated L-02 cells ( Fig.2E-G). These results indicated that activation of Rev-erbα can promote disorder of lipid metabolism and Rev-erbα may be involved in the pathological process of AFL.

SR8278 attenuates steatosis in the liver of EtOH-fed mice and EtOH-treated L-0 2 cells
To better understand the function of Rev-erbα in EtOH-induced liver injury and steatosis, the Rev-erbα antagonist SR8278 (2 mg/kg) was injected in EtOH-fed mice via tail vein, half an hour after feeding [25,26]. As shown in Fig. 3A, the fatty liver was significantly alleviated in EtOH-fed mice after injecting SR8278 for 3 days. H&E and ORO staining revealed interlobular space of liver, inflammatory cell infiltration and lipid droplets were improved after treatment with SR8278. The ratio of liver weight to body was increased and serum ALT, TG, and T-CHO level in EtOH-fed mice were decreased by SR8278 (Fig. 3B-E). Moreover, increased of Pparα and decreased of Srebp1c were showed by immunohistochemistry analysis (Fig. 3F).
Additionally, SR8278 (the antagonist of Rev-erbα, 10 μM) reduced the level of TG and intracellular lipid droplets in EtOH-treated L-02 cells for 24 h (S1E and S1F), and higher expression of Pparα and lower expression of Srebp1c were demonstrated by western blot analysis (S1G).

Down-regulated Rev-erbα attenuates steatosis in EtOH-treated L-02 cells
To further verify the effect of Rev-erbα on lipid metabolism, Rev-erbα was knocked down by transfecting Rev-erbα shRNA in L-02 cells. The results of qRT-PCR and western blot showed that Rev-erbα was knocked down by Rev-erbα shRNA in EtOH treatment L-02 cells (S1H and Fig. 4A). Then, the results of ORO staining, TG assay and the level of Pparα and Srebp1c showed that Rev-erbα shRNA improved lipid metabolism disorder and reduced lipid deposition ( Fig. 4B and E). In summary, inhibition or silencing of Rev-erbα may attenuate steatosis in vivo and in vitro.

Rev-erbα regulated lipid metabolisnm by enhancing autophagy activity in vivo and in vitro
It is well known that autophagy is involved in the degradation of lipid droplets and regulation of lipid metabolism [16][17]. Rev-erbα has been reported to regulated autophagy in skeletal muscle [18][19]. Therefore, we hypothesized that Rev-erbα could ameliorate EtOH-induced lipid steatosis by regulating autophagy. As shown in Fig.5A, electronic microscopy showed that autophagosome and lysosomal were significantly decreased by EtOH, and SR8278 could increase the number of autophagosome and lysosomal in EtOH-fed mice. Immunohistochemistry analysis further showed that SR8278 increased the expression of Lc3 and decreased the expression of P62 in EtOH-fed mice (Fig.5B).
In vitro, Lyso Tracker Green DND-26 analysis showed that lysosomal acidity was decreased by EtOH in L-02 cells while it was increased in EtOH-treated L-02 cells by SR8278 and Rev-erbα shRNA (S2A and Fig. 5C). Next, western blot analysis confirmed that compared to L-02 cells, the ratio of LC3II/Ⅰ and the level of Beclin1 were decreased but P62 was increased in EtOH-treated L-02 cells, however, the results were reversed after treatment with SR8278 and Rev-erbα shRNA (S2B and S2C). Above experimental results indicated that autophagy can be negatively regulated by Rev-erbα in AFL.

Rev-erbα inhibits the activity of autophagy through regulating Bmal1
As presented in S3A and S3B, the level of Bmal1 was decreased in the liver of EtOH-fed mice and EtOH-treated L-02 cells. Moreover, Bmal1 protein decreased prominently in the nucleus and it had no distinctly changed in the cytoplasm in EtOH-treated L-02 cells compared to L-02 cells (Fig. 6A). Knocked down of Rev-erbα up-regulated the expression of Bmal1 detected by using western blot and qRT-PCR analysis (Fig. 6B and S3C). The above experimental results showed that Rev-erbα might play a critical role in regulating the expression of Bmal1 in AFL.
To further confirm whether Rev-erbα inhibit autophagy dependent Bmal1 in AFL, Rev-erbα shRNA and Bmal1 siRNA were co-transfected into EtOH-treated L-02 cells.
First, The resluts of qRT-PCR and western blot showed that Bmal1 was knocked down by Bmal1 siRNA in EtOH treatment L-02 cells (S3D and S3E). As illustrated in Fig. 6C, Bmal1 siRNA have reversed the acidity of lysosomes which was increased by Rev-erbα shRNA in EtOH treatment L-02 cells. What's more, western blot analysis showed that Rev-erbα shRNA and Bmal1 siRNA co-transfection decreased the ratio of LC3 II/Ⅰ, increased P62 level, and down regulated Pparα in EtOH treatment L-02 cells but Srebp1c has no significantly changed (Fig. 6D). These data indicated that Rev-erbα may inhibit the activity of autophagy by Bmal1 in EtOH-induced lipid steatosis.

Discussion
Lipid accumulation in hepatocytes is a typical morphological characteristic of AFL [27,28]. According to the initial 'two hit hypothesis,' on the base of liver steatosis, inflammation as the second hit promotes the transformation from AFL to ASH [29]. As an important node in the development of AFLD, the improvement of hepatic lipid metabolism in AFL can prevent the occurrence of ASH and reduce the incidence of ALD. The nuclear receptor Rev-erbs are known to regulate multiple downstream genes involved in diverse cellular functions including metabolism. It has been widely participate in the physiological process of energy, glucose and lipid metabolism [30][31][32][33][34]. However, the effect of Rev-erbs on lipid regulation of alcoholic fatty liver remains to be studied. Rev-erbs has two subtypes with high homology, and their distribution are different. In this study, we found that Rev-erbα has a more abundant distribution compared to Rev-erbβ in the liver of mice and L-02 cells, this result was consistent with the research that Rev-erbα was higher expressed in liver, meanwhile Rev-erbβ was lower in physiological system but higher in CNS in mice [35]. In addition, we showed higher expression of Rev-erbα in the liver of EtOH-fed mice and EtOH-treated L-02 cells accompanying with severe steatosis, characterized by the increase of lipid droplets and triglycerides. Moreover, L-02 cells showed significant steatosis when cells were treated with GSK4112. These results indicate that Rev-erbα is closely related to liver lipid metabolism. Importantly, steatosis were ameliorated in the liver of EtOH-fed mice after the mice were treated with SR8278, the same result was found when EtOH-treated L-02 cells were treated with SR8278 or transfected with Rev-erbα ShRNA. Rev-erbα is a member of nuclear receptor superfamily, which plays an important role in transcriptional inhibition. And we found that Rev-erbα was mainly up-regulated in the nucleus in EtOH-treated L-02 cells.
This suggests that Rev-erbα may plays an important role in the transcription of lipid metabolism genes in the nucleus. Coincidentally, Srebp1c expression decreased and pparα expression increased at both mRNA and protein levels in SR8278 treated EtOH-fed mice and SR8278 / Rev-erbα ShRNA treated EtOH-treated L-02 cells. To sum up, Rev-erbα may be a vital mediator of lipid metabolism in vivo and vitro.
Autophagy is a wide range of cells and lysosomal-dependent degradation pathway. It is a mechanism that delivers cytoplasmic cargo into acidic compartments of the cell known as lysosomes. The acidic environment in lysosomes is the key to autophagy [36][37][38][39][40][41][42]. Rajat Singh et al. had identified that autophagy was required for lipid droplets breakdown and inhibition of autophagy increases lipid storage [43,44].
Other studies, for example, Martinez-Lopez N et al. reported that cold induced autophagy which triggered lipolysis in mouse liver [45]. In this study, the ratio of autophagosomes membrane protein LC3II/I and Beclin1 expression was down-expressed. Further, the acidity in lysosomes was decreased as well as the production of p62 increased significantly in the liver of EtOH-fed mice and Previous studies have found that Bmal1 (Aryl-hydrocarbon nuclear translocator-like 1) drives the cyclic expression genes involved in lipid metabolism [46,47]. Zhang D et al. has studied that mice with Bmal1 depletion were more susceptible to ethanol induced fatty liver and liver injury while Bmal1 over-expression protects EtOH-fed mice from fatty liver and liver injury [48]. Taking into consideration that Rev-erbα plays a negative adjustment function through inhibits Bmal1 transcription by target a ROR-response element in the promoter of the Bmal1 gene in the biological clock system [49,50]. We speculate that Rev-erbα may interact with Bmal1 in cell metabolism. Indeed, we observed that Bmal1 was decreased in the nucleus in EtOH-treated L-02 cells, this result is consistent with the increase of Rev-erbα in nucleus. Further, inhibiting of Rev-erbα resulted in an increase of Bmal1 in EtOH-treated L-02 cells. It suggested that there may be a crosstalk between Rev-erbα and Bmal1 in the nucleus respond to ethanol treatment. Importantly, Bmal1 depletion inhibited the activity of autophagy, and the promotion of Rev-erbα on Pparα expression was abolished by knocking out of Bmal1. It is likely that the modest suppression of Rev-erbα promoted Pparα-dependent β-oxidation pathway by up-regulating Bmal1 through regulating autophagy in EtOH-treated L-02 cells. It is well known that Pparα and Srebp1c were pivotal lipid metabolism-associated transcription factors [51,52]. However, Bmal1 knockout did not change the positive regulatory effect of Rev-erbα on Srebp1. Previous study has demonstrated Rev-erbα inhibits transcription mainly by recruiting HDAC3 and NCoR. But Srebp1c expression is not HDAC3-sensitive, Berthier A et al. has proved that Rev-erbα directly bound to genomic regions in the vicinity or within the Srebf1 gene [53,54].
This suggest that Rev-erbα may be directly involved in the expression of lipid synthesis genes, which is independent of Bmal1 expression.
In conclusion, we presented evidence supporting that Rev-erbα was higher expressed and reducing Rev-erbα could improve the accumulation of lipid in vivo and vitro. Impaired autophagy function in the liver of EtOH-fed mice and EtOH-treated L-02 cells was enhanced by limiting the activity of Rev-erbα. Further researches promulgated a regulation of Rev-erbα on autophagy by Bmal1 thus influencing hepatic fatty acid oxidation pathway in vitro (Fig.7). Our findings identified a critical role of the protein Rev-erbα in ameliorating lipid metabolism through autophagy by impacting on Bmal1, suggesting that focusing on Rev-erbα function in lipid metabolism may offer a therapeutic approach to AFL. The biggest flaw in our research is that Rev-erbα and Bmal1 were only used as the regulatory factors of lipid metabolism, the changes of their biological rhythm were not studied, which needs to be improved.