MEK1/2 Inhibitors Induce Class 1 Alcoholic Dehydrogenase (ADH1) Expression by Regulating Farnesoid X Receptor


 Background: Alcohol is mainly catabolized by class 1 alcohol dehydrogenase (ADH1) in liver. ADH deficiency can aggravate ethanol-induced tissue injury. Extracellular signal-regulated kinases 1/2 (ERK1/2) is involved in alcohol metabolism. However, the relationship between ERK1/2 and ADH1 remains unclear. Methods and Results: To inhibit ERK1/2, HepG2 and BNL cells were treated with mitogen-activated protein kinases 1/2 (MEK1/2) inhibitors (U0126 and PD98059), and C57BL/6J mice were fed U0126. After treatment, the protein and mRNA expression of ADH1 were determined by Western blot and quantitative real time-PCR. The activity of ADH1 promoter was detected using luciferase assay. The results showed MEK1/2 inhibitors significantly increased ADH1 protein expression by inducing its transcription activity. Then we found a putative farnesoid X receptor (FXR) response element (FXRE) sequence in ADH1 promoter by completing a sequence alignment analysis. To test whether FXR mediates the induction of MEK1/2 inhibitors on ADH1, HepG2 cells were transfected with FXR siRNA or ADH1 promoters with FXRE mutation. We found both FXR siRNA and FXRE mutation in ADH1 promoter abolished MEK1/2 inhibitors-induced ADH1 expression, indicating the activation of MEK1/2 inhibitors on ADH1 depends on FXR.Conclusions: Our findings revealed inhibition of ERK1/2 can significantly increase ADH1 expression, indicating MEK1/2 inhibitors may possess potential application in alcohol-related diseases.


Introduction
Alcohol consumption plays a traditional role in many countries. However, numerous studies have demonstrated alcohol is harmful to human body. It can cause cognitive de cits, such as di culties in problem-solving, memory and learning (1). Alcohol also induces dysfunction of immune system and increases the risk of cancers (2,3). Thus, to avoid alcohol-induced toxicity, accelerating alcohol metabolism is quite important. It is reported that only 2-10% of alcohol in body is excreted through the lungs, urine and sweat, while most of the remainder is catabolized in liver (4). These data indicate liver bears a heavy burden of alcohol metabolism. Correspondingly, alcohol pose a great threat to liver. About 48% of liver cirrhosis deaths was attributed to alcohol. The mortality from liver cancer secondary to alcohol for 2010 was estimated to be 149,000 deaths (5).
Alcohol dehydrogenase (ADH), a zinc-containing enzyme, is highly expressed in liver. It is the major pathway for alcohol metabolism (6). 75%-80% of alcohol in liver proceeds via ADH (5). ADH catalyzes ethanol into acetaldehyde which is then quickly converted into acetate by aldehyde dehydrogenase. There are several forms of ADH, such as class 1 ADH (ADH1), class 2 AHD (ADH4) and class 3 ADH (ADH5).
Among these forms, ADH1 plays the most important role in alcohol metabolism (6). Therefore, induction of ADH1 is bene cial to alcohol elimination. Extracellular signal-regulated kinases 1 and 2 (ERK1/2) are involved in a variety of biological processes, including cell survival, metabolism and transcription (7). Mitogen-activated protein kinases 1/2 (MEK1/2) is required to phosphorylate and activate ERK1/2. Several studies have demonstrated that phosphorylation of ERK1/2 in liver is increased after alcohol consumption in rats and mice (8,9). ERK1/2 has also been reported to be associated with ethanol-induced liver injury (10). These papers suggest ERK1/2 activation may hinder ethanol metabolism. Therefore, we hypothesis that inhibition of ERK1/2 by MEK1/2 inhibitors can active ADH1 expression in hepatocytes, thereby increasing ethanol metabolism.

In vivo study
The protocol for in vivo study was granted by the Ethics Committee of Nankai University and conforms to the Guide for the Care and Use of Laboratory Animals published by NIH. The mice were housed in SPF units of the Animal Center at Nankai University (with a 12-h light cycle from 8 a.m. to 8 p.m., 23 ± 1°C, 60-70% humidity) with free access to normal chow and water. Up to ve mice were kept per ventilated cage with corn cob bedding material. Male C57BL/6J mice (~6-week old, ~22 g) were randomly divided into two groups (5 mice/group) and fed normal chow (control group) or normal chow containing U0126 (1.5 mg/day/kg bodyweight, U0126 group) for ve weeks. At the end of experiment, all mice were anesthetized and euthanized in a CO 2 chamber followed by collection of liver samples.

Cell culture
HepG2 (a human hepatic cell line) and BNL (a murine hepatic cell line) cells were purchased from ATCC (Rockville, MD, USA) and cultured in DMEM medium containing 10% fetal bovine serum, 50 mg/mL penicillin/streptomycin and 2 mM glutamine.

Data analysis
All the experiments were repeated at least 3 times, and the representative results were presented. Values are presented as mean ± SD. GraphPad Prism software (version 8) was used for one-way ANOVA with post hoc Tukey's test or unpaired t test (two-tailed) whenever appropriate. The signi cant difference was considered if P<0.05.

MEK1/2 inhibitors induce ADH1 expression in human hepatic cells
Alcohol is harmful to various tissues because of its toxic effects. Liver is a major organ for alcohol metabolism. Ethanol can promote phosphorylation of ERK1/2 and activated ERK1/2 is involved in alcohol-induce liver injury (12,13). So, we hypothesized inhibition of ERK1/2 may protect hepatocytes against ethanol-induced cytotoxicity. MEK1/2 is the upstream kinase of ERK1/2. U0126 and PD98059 are two speci c MEK1/2 inhibitors. Both of them can signi cantly reduce pi-ERK1/2 (the active form of EKR1/2) in HepG2 cells (Fig. 1a, b). Therefore, we used U0126 and PD98059 to inhibit ERK1/2 in this study. To determine our hypothesis, we treated HepG2 cells with ethanol, or ethanol plus U0126 and PD98059. As shown in Fig. 1c (4), so we speculated that U0126 and PD98059 may ameliorate ethanol-induced cytotoxicity by regulating ADH1 expression. As we expected, the results in Fig. 1e showed that U0126 induced ADH1 protein levels in a concentration dependent manner [F 2,6 =81.6, P<0.001]. The time course study with 5 mM U0126 demonstrated the induction occurred quickly (3 h after treatment) and lasted for 24 h of the treatment [F 5, 12 =99, P<0.001] (Fig. 1f). Similarly, we found PD98059 also can increase ADH1 protein expression [F 3, 8 =121, P<0.001] (Fig. 1e).
Human ADH1 contains three genes, ADH1A, ADH1B and ADH1C which encode a, b, and g subunits respectively. These subunits can combine and function as homodimers or heterodimers (6). To determine if MEK1/2 inhibitors regulate ADH1 protein expression at the transcriptional level, we detected mRNA expression of ADH1A, ADH1B and ADH1C in response to U0126 treatment. The results in both concentration and time course studies demonstrate that U0126 greatly increased ADH1A, ADH1B and ADH1C mRNA levels [Fig. 2a (Fig. 2a, b) (Fig. 2c). To further con rm the effects of ERK1/2 inhibition on ADH1 transcription, we constructed ADH1A, ADH1B and ADH1C promoters and determined the effects of PD98059 on their activities. As shown in Fig. 2d (Fig. 3a). These results suggest MEK1/2 inhibitors can also activate mouse ADH1 expression and transcription. To further de ne the regulatory effects of ERK1/2 on ADH1 in vivo, we fed C57BL/6J mice a normal chow (control group) or normal chow containing U0126 (U0126 group). After treatment, mouse liver samples were used to determine protein expression of pi-ERK1/2, ERK1/2 and ADH1. As shown in Fig. 3d, associated with reduced pi-ERK1/2, U0126 signi cantly increased ADH1 protein level [t 8 =12.35, P<0.001].
These results indicate ERK1/2 inhibition can induce ADH1 expression in vivo.

The activation of MEK1/2 inhibitors on ADH1 expression depends on farnesoid X receptor (FXR)
Both in vitro and in vivo experiments demonstrate MEK1/2 inhibitors can increase ADH1 expression. Then we intended to further investigate the underlying mechanism. By completing a sequence alignment analysis, we found a putative FXR response element (FXRE) sequence in ADH1A, ADH1B and ADH1C promoter respectively. The sequence of the conserved FXRE motif is RGGTCAnTGACCY (R=A/G, N=A/C/G, Y=C/T). Several studies have demonstrated activation of FXR can attenuate alcohol-induced liver injury (14,15). Therefore, we hypothesis MEK1/2 inhibitors may regulate ADH1 via FXR. To determine it, we initially detected protein expression of FXR in response to U0126 treatment. Indeed, both concentration and time course studies show U0126 signi cantly upregulated FXR protein expression, indicating the regulatory effect of U0126 on FXR [ Fig. 4a: F 3, 8 =50.3, P<0.001], [Fig. 4b: F 5, 12 =96.6, P<0.001] (Fig. 4a, b). To investigate the involvement of FXR in ADH1 expression, we constructed ADH1A, ADH1B and ADH1C promoters with FXRE mutation and evaluated the effect of MEK1/2 inhibitor on activities of these promoters. Fig. 4c indicated mutation of FXRE abolished the induction effects of PD98059 on ADH1A, ADH1B and ADH1C promoters. Then we transfected cells with FXR siRNA to investigate the effects of FXR on ADH1 protein expression. Fig. 4d shows FXR siRNA signi cantly decreased ADH1 protein expression [F 5, 12 =72.3, P=0.0048]. Furthermore, the induction of U0126 on ADH1 protein was notably blocked by FXR siRNA. Taken together, MEK1/2 inhibitor activates ADH1 expression in an FXR-dependent manner.

Discussion
Because of low molecular weight and water solubility, ethanol can distribute from the blood into all tissues quickly after consumption (6). Ethanol is harmful to various tissues. It induces neuroin ammation and neuronal apoptosis in brain (16). Chronic ethanol consumption can stimulate hepatocyte necroptosis and accelerate development of alcohol liver disease (17). Ethanol treatment can also induce oxidative stress and impair mitochondrial biogenesis in pancreatic acinar cells (18). Therefore, increasing the elimination rate of ethanol is important for body to avoid ethanol toxicity.
ADH is the rate-limiting enzyme in alcohol metabolism. Mice lacking ADH showed elevated blood alcohol concentration, plasma alanine aminotransferase, and hepatic steatosis when fed ethanol (19). Increased apoptotic bodies was also observed in pancreata of ADH de cient mice (20). Compared to wild type mice, ethanol exposure signi cantly increased mortality of ADH de cient mice (20). These studies indicate ADH can protect mice against ethanol toxicity. Among several ADH isozymes, ADH1 plays the most important role in alcohol elimination (6). Therefore, induction of ADH1 may accelerate alcohol metabolism and function as a therapeutic strategy for alcohol-related diseases.
Li et al have shown ERK1/2 is involved in alcohol liver disease (10). They found ethanol induced NF-κBinducing kinase (NIK) expression. Activated NIK recruited ERK1/2 and formed a complex to suppress PPARα signaling. As a result, liver lipid accumulation increased. This paper indicates ERK1/2 aggravated alcohol-induced liver steatosis by inhibiting PPARα-induced lipid oxidation. However, whether blockage of ERK1/2 can protect liver from alcohol toxicity by promoting ethanol metabolism remains unclear.
In current study, we determined MEK1/2 inhibitors induced ADH1 expression in human and murine hepatic cell lines by promoting its transcription. In mice, administration of mice with U0126 increased ADH1 expression in liver. Mechanistically, we found FXR siRNA and FXRE mutation in ADH1 promoters abolished MEK1/2 inhibitors-induced ADH1 expression, indicating MEK1/2 inhibitors activated ADH1 mRNA and protein expression in an FXR dependent manner. Taken together, MEK1/2 inhibitors may be potential drugs for alcohol-related diseases treatment. Figure 1 MEK1/2 inhibitors induce ADH1 expression in HepG2 cells a, b: HepG2 cells were treated with U0126 (a) and PD98059 (b) for the indicated time followed by total protein extraction and determination of phosphorylated ERK1/2 (pi-ERK1/2) and ERK1/2 by Western blot (n=3); c, d: HepG2 cells were treated with 2% ethanol (V/V) plus U0126 (c) or PD98059 (d) at the indicated concentrations overnight followed by determination of cell survival rate using the MTT assay kit (n=5); e, f: HepG2 cells were treated with U0126 or PD98059 at the indicated concentrations for 16 h (e) or with 5 M U0126 for the indicated time (f). Total cellular protein was used to determine the expression of ADH1 by Western blot with quantitative analysis (n=3). *P<0.05, **P<0.01, ***P<0.001 vs. control or as indicated. Expression of ADH1A, ADH1B and ADH1C mRNA were determined by qRT-PCR; d: HepG2 cells were transfected with pGL4.10 vector or human ADH1A, ADH1B, ADH1C promoters and Renilla luciferase. After treatment with PD98059 at the indicated concentrations for 16 h, promoter activity of ADH1A, ADH1B and ADH1C were determined using dual-luciferase assay kit. **P<0.01, ***P<0.001 vs. control (n=5).