Acid-sensing ion channel 1a promotes alcohol-associated liver disease in mice via regulating endoplasmic reticulum autophagy

Abstract

Alcohol-associated liver disease (ALD) is a hepatocyte dysfunction disease caused by chronic or excessive alcohol consumption, which can lead to extensive hepatocyte necrosis and even liver failure. Currently, the pathogenesis of ALD and the anti-ALD mechanisms have not been fully elucidated yet. In this study, we investigated the effects of endoplasmic reticulum autophagy (ER-phagy) in ALD and the role of acid-sensing ion channel 1a (ASIC1a) in ER stress-mediated ER-phagy. A mouse model of ALD was established using the Gao-Binge method and the AML12 cell line treated with alcohol was used as an in vitro model. We showed that ASIC1a expression was significantly increased and ER-phagy was activated in both the in vivo and in vitro models. In alcohol-treated AML12 cells, we showed that blockade of ASIC1a with PcTx-1 or knockdown of ASIC1a reduced alcohol-induced intracellular Ca²⁺ accumulation and ER stress. In addition, inhibition of ER stress with 4-PBA reduced the level of ER-phagy. Furthermore, knockdown of the ER-phagy receptor family with sequence similarity 134 member B (FAM134B) alleviated alcohol-triggered hepatocyte injury and apoptosis. In conclusion, this study demonstrates that alcohol activates ER stress-induced ER-phagy and liver injury by increasing ASIC1a expression and ASIC1a-mediated Ca²⁺ influx, providing a novel strategy for the treatment of ALD.

Introduction

The liver is an important organ responsible for coordinating the different metabolic pathways in the host, performing several important tasks on a daily basis. After it is absorbed into the bloodstream, alcohol is metabolized in the liver and other organs (e.g. gut) [1] followed by damaging the liver. Long-term drinking can cause the majority of people to suffer from different degrees of fatty liver, which will further develop into alcoholic hepatitis, liver fibrosis, cirrhosis, and eventually even hepatocellular carcinoma [2]. Analysis by the Global Burden of Disease Study shows that cirrhosis accounted for 1.32 million deaths worldwide in 2017: alcohol was recorded as a contributing factor in 48% of these deaths, whereas it was recorded as a causative factor in 332,268 deaths [3]. Alcohol-associated liver disease (ALD) is emerging as a worldwide public health threat.

Autophagy is a tightly regulated and stress-induced catabolic pathway [4]. Previous studies have found that alcohol and its metabolites can increase protein misfolding and damage the function of the endoplasmic reticulum (ER) by inducing endoplasmic reticulum stress, which can exacerbate ALD [5–7]. Recent studies have revealed that autophagy can selectively degrade the ER, and they named this novel process as ER autophagy (ER-Phagy) [8, 9]. At present, six ER-phagy receptors, FAM134B [10], SEC62 [11], RTN3L [12], CCPG1 [13], ATL3 [14] and TEX264 [15, 16], have been identified in mammals, which are distributed in different subdomains of the ER. They all contain at least one LIR interaction region (LC3 interacting region, LIR) or GABARAP-interaction motif (GIM), which enables them to directly interact with the LC3B protein on the autophagosome membrane to mediate targeted degradation. ER-phagy effectively removes damaged endoplasmic reticulum and misfolded proteins to prevent the occurrence of disease [17]. In contrast, when the ER-phagy is disordered in some pathological states, the radicalized ER-phagy over-clears the normal intracellular ER, causing normal cellular damage or exacerbating the disease process by transmitting signals that activate other cell responses [18]. Therefore, the inhibition or activation of ER-phagy could be a potential target for the prevention and treatment of endoplasmic reticulum related diseases.

Acid-sensing ion channels (ASICs) are members of the epithelial sodium channel/degenerin (ENaC/DEG) superfamily of ion channels, which are trimeric, proton-gated cation channels expressed throughout vertebrate central and peripheral nervous systems [19]. So far, at least six ASIC subunit proteins (ASIC1a, ASIC1b, ASIC2a, ASIC2b, ASIC3, and ASIC4) have been discovered [20]. ASICs have a large size of the ASIC extracellular domain containing 14 conserved cysteine residues, while approximately 318 of the 528 residues of ASIC1a are in the extracellular domain, reminiscent of an antenna protruding above the channel pore [19]. Wemmie et al. speculated that this complex structure should not only sense protons, but other ligands may exist [20]. A variety of reports have shown that many other factors also interact with the extracellular structural domains to regulate H⁺ -dependent channel activity, including Ca²⁺ [21], Zn²⁺[22, 23], lactate [24], proteases [25], arachidonic acid [26], lead and redox reagents [27]. In contrast to other ASICs subunits, ASIC1a not only permeates Na⁺, but also mediates Ca²⁺ influx. Zhou et al. found that acute treatment of neuronal cells with ethanol for more than 3 h reduced ASIC1a protein expression, ASIC currents, and acid-induced [Ca²⁺]_i elevation [28]. Early studies in our group revealed that ASIC1a also plays an important role in liver disease. Elevated expression of ASIC1a in vivo and in vitro fibrosis models mediated the activation of hepatic stellate cells and increased the progression of liver fibrosis, which is associated with autophagy [29], ER stress [30] and m⁶A [31] regulation of non-coding RNA. However, how it plays a role in ALD is still unknown. The aim of this study was to investigate whether ER-phagy is involved in alcohol-associated liver disease, and the role of ASIC1a in this and its possible mechanism.

Materials and methods

Animals and ALD models

C57BL/6 J mice (8–10 weeks, male) were obtained from the Experimental Animal Center of Anhui Medical University, Hefei, China. Animal experiments were carried out at the Animal Experiment Center of Anhui Medical University, license number: SYXK (Anhui) 2017-006. All mice were allowed to acclimate for at least 1 week under constant temperature (23 ± 2 °C) and humidity (60% ± 5%) with a 12-h light-dark cycle. All procedures were in strict accordance with the Chinese legislation on the animals and guidelines established by the Laboratory Animal Center of Anhui Medical University. All animal experiments were supervised and approved by the Animal Ethics Committee of Anhui Medical University.

Adult mice were caudal vein injected with AAV9 adeno-associated virus (ASIC1a KD, HH20201203WY-AAV01, Hanbio Biotechnology, Shanghai, China) vectors (100 μL/mouse) expressing GFP-tagged for 2 weeks. Mice injected with AAV9 vector were used as control. ALD modeling method followed the National Institute on Alcohol Abuse and Alcoholism (NIAAA) protocol. The NIAAA protocol is now referred as “Gao-binge model” [32]. The “Gao-binge model” included a 5 days diet adaptation period, 10 days of modeling (5% v/v ethanol liquid diets), and a one-time gavage (5 g/kg, body weight, 20% ethanol) on the last day for a total of 16 days. The control group was fed with the same calorie liquid diet for 15 days, and on the 16th day, 0.2 mL/20 g body weight was given by intragastric administration of 0.45 g/mL concentration of dextrin. All mice were anesthetized nine hours after alcohol administration and liver tissue and serum were extracted for subsequent experiments.

Serum analysis

The biochemical markers, including serum alanine transaminase (ALT), aspartate aminotransferase (AST), and triglyceride (TG), were measured by a fully automatic biochemical analyzer (BS-600, Mindray, Shenzhen, China) in the Inflammation and Immune Mediated Diseases Laboratory, Anhui Medical University.

Histology and immunohistochemical analysis

Liver sample was fixed in 4% paraformaldehyde (P0099, Beyotime, Shanghai, China), embedded in paraffin, sectioned, and stained with hematoxylin and eosin (H&E, G1004; G1001, Servicebio, Wuhan, China). For immunohistochemistry (IHC), paraformaldehyde-fixed paraffin-embedded liver sections were deparaffinized, hydrated, and stained with antibody against ASIC1a (1:200, 27235-1-AP, Proteintech, Wuhan, China) and FAM134B (1:200, bs-13136R, Bioss, Beijing, China). Subsequently, the slides were further processed using corresponding secondary antibodies, followed by counterstaining with hematoxylin. Lipid droplet accumulation in the liver was observed using Oil red O (G1015, Servicebio, Wuhan, China) staining. Frozen liver sections were stained with Oil red O staining reagent for 30 min and were counterstained with hematoxylin.

Immunofluorescence analysis

Liver sections were fixed by 4% paraformaldehyde at room temperature for 10 min, and blocked in 3% BSA in PBS for 1 h. Primary antibodies (LC3 [1:200, 2775, Cell Signaling Technology, Boston, MA, USA]; Calreticulin [1:1000, ab22683, Abcam, UK]; FAM134B [1:200, bs- 13136 R, Bioss, Beijing, China]; ASIC1a [1:300, bs-2586R, Bioss, Beijing, China]) were diluted in 3% BSA (A2153, Sigma-Aldrich, Germany) for overnight incubation. Sections were washed with PBS and subsequently incubated with secondary antibodies (Alexa Fluor 488 [Rabbit, ZF-0511, ZSGB-bio, Beijing, China]; Alexa Fluor 594 [Rabbit, ZF-0516, ZSGB-bio, Beijing, China]; Alexa Fluor 488 [Mouse, ZF-0512, ZSGB-bio, Beijing, China]; Alexa Fluor 594 [Mouse, ZF-0513, ZSGB-bio, Beijing, China]). The nucleus was labeled with DAPI Staining Solution (C1005, Beyotime, Shanghai, China) at room temperature for 10 min. An Olympus fluorescence microscope was used to image the cells.

Cell culture

AML12 cells were grown in DMEM/F12 supplemented with 10% BI fetal bovine serum (FBS, 04-001-1ACS, Biological Industries, Israel), and then were infected with 50 nM plasmid (ASIC1a KD) or FAM134B siRNA. After gene transfection, AML-12 cells were treated with 200 mM ethanol (200 mM) for 8 h for ALD in vitro model. For other experiments, AML-12 cells were also treated with psalmotoxin-1 (PcTx-1; 10 nM, ab120483, Abcam, UK), benzenebutyric acid (4-PBA; 2 mM, T5886, TargetMol, Shanghai, China), nimodipine (5 μM, T0343, TargetMol, Shanghai, China), 3-methyladenine (3-MA; 5 mM, GC10710, GLPBIO, Montclair, California, USA).

Cell transfection

After the AML12 cells cultured in DMEM/F12 medium reached about 30%–50%, cells were replaced in Opti-MEM (Gibco) medium, and then Lipofectamine™ 3000 (Invitrogen, Carlsbad, California, USA) was used for transfection with ASIC1a knockdown plasmids (1000 ng/mL, Gene Pharma, Shanghai, China), FAM134B siRNA (1000 ng/mL, Hanbio Biotechnology, Shanghai, China). After incubating at 37 °C and 5% CO₂ for 6–8 h, cells were replaced in fresh medium for 24–48 h, and then collected for the next experiment. The plasmids and siRNA sequences are listed Table 1.

Table 1.

Sequences of target plasmids and siRNA.

Gene	Plasmids and siRNA sequence (5ʹ-3ʹ)
ASIC1a	Sense: CACCGCCAAGAAGTTCAACAAATCGTTCAAGAGACGATTTGTTGAACTTCTTGGCTTTTTTG Antisense: GATCCAAAAAAGCCAAGAAGTTCAACAAATCGTCTCTTGAACGATTTGTTGAACTTCTTGGC
FAM134B	Sense: CCACAGAGCUCAAGAGAAATT Antisense: UUUCUCUUUGAGCUCUGUGGTT

Cell viability and cell death determination

The viability of AML12 cells was determined using the Cell count kit 8 (CCK-8 kit; CK04, Dojindo, Japan). Appropriate number of AML12 cells were plated in 96 well-plate (3×10³–5×10³ per well) and subjected to different treatments. After the treatment, the medium was refreshed by a mixture of 10 μL CCK-8 solution and 90 μL serum-free DMEM/F12, incubated in an incubator for 1–4 h, and then the cell viability could be read at 450 nm by a microplate reader.

Cell death was measured by FDA/PI Cytotoxicity Assay Kit (BB-4217, Best Bio, Shanghai, China) in accordance with the manufacturer’s instructions. A double-blinded cell counting was performed for live (green) and dead cells (red) using an inverted fluorescent microscope (U-HGLGPS, OLYMPUS, Japan).

Total RNA extraction and Western blotting

Total protein was extracted from AML12 cells and mouse liver tissues with RIPA lysis buffer (P0013B, Beyotime, Shanghai, China). BCA Protein Assay Kit (P0012S, Beyotime, Shanghai, China) was used to measure protein concentrations. Protein samples were combined with SDS loading buffer (P0015L, Beyotime, Shanghai, China) and transferred onto a PVDF membrane (Millipore Corp, Billerica, MA, USA). After blocked with 5% skim milk, the PVDF membrane was incubated with primary antibodies at 4 °C overnight, followed by incubation with the corresponding secondary antibodies at room temperature for 1 h. The primary antibodies were as follows: ASIC1a (27235-1-AP, 1:1000, Proteintech, Wuhan, China), IRE1U (#3294, 1:1000, Cell Signaling Technology, Boston, MA, USA), GRP78 (#3177, 1:1000, Cell Signaling Technology, Boston, MA, USA), FAM134B (ab151755, 1:1000, Abcam, UK), LC3B (#2775, 1:1000, Cell Signaling Technology, Boston, MA, USA), p62 (#5114, 1:1000, Cell Signaling Technology, Boston, MA, USA), Bax (bs-0127R, 1:1000, Bioss, Beijing, China), Bcl-2 (26593-1-AP, 1:1000, Proteintech, Wuhan, China) and β-actin (bs-0061R, 1:1000, Bioss, Beijing, China). Protein brands were visualized using Omni ECL reagent (SQ201, EpiZyme, Shanghai, China). Each experiment was replicated at least three times, and the acquired images were quantified by ImageJ software (NIH, Bethesda, MD, USA).

Total RNA extraction and quantitative real-time PCR (qRT-PCR)

Total RNA was isolated with the TRIzol reagent (15596026, Invitrogen, USA) from liver tissues and cells, and the first-stand cDNA was synthesized using Evo M-MLV RT Premix for qPCR kit (AG11706, Accurate Biology, Changsha, China). qRT-PCR was performed with SYBR^® Green Premix Pro Taq HS qPCR Kit (AG11701, Accurate Biology, Changsha, China), and CFX-96 real-time PCR system (CFX Connect, Bio-Rad, Hercules, California, USA) was used to detect the mRNA fold changes in each sample. The primer sequences are listed Table 2.

Table 2.

Sequences of qRT-PCR primer.

Gene	Forward (5ʹ-3ʹ)	Reverse (5ʹ-3ʹ)
GAPDH	GGCCCCTCTGGAAAGCTGTG	CCGCCTGCTTCACCACCTTCT
ASIC1a	GGCCAACTTCCGTAGCTTCA	ACACATGAATAACGGGGTTTCC
GRP78	CTGTCAGCAGGACATCAAGTTC	TGTTTGCCCACCTCCATTATCA
IRE1α	ACACCGACCACCGTATCTCA	CTCAGGATAATGGTAGCCATGTC
FAM134B	AAACAGCAGAGTCCTGGCAAG	AGGTAGCTGAGTATGACCCCA
SEC62	TGATTGCAGTAATAGCAGCCAC	GCCCACACTGAGGTAATAAACAC
RTN3L	AGGTGCCCCTACGATGTCTC	GGTTTGCTTGAGTTTTCCTCCA
ATL3	CTGGACTTTATGCTGCGATACTT	AGCCTCCTCGCCATGAAAATC
CCPG1	AAGCGGCAACTGAAAAGACAA	TCCAGCACATTCAAACATACCTG
TEX264	CCAATCCGCAACATAACTGTGG	GGGTTGTCATAGTAGACAGCGAT
ATG5	TGTGCTTCGAGATGTGTGGTT	ACCAACGTCAAATAGCTGACTC
ATG7	TCTGGGAAGCCATAAAGTCAGG	GCGAAGGTCAGGAGCAGAA
Bax	AGACAGGGGCCTTTTTGCTAC	AATTCGCCGGAGACACTCG

Transmission electron microscopy (TEM)

The AML12 cells were plated in T-25 flasks at 1.5 × 10⁶ cells per flask. Cells were washed with 0.1 mol/L PBS, harvested with a cell scraper, and then centrifuged at 1200 rpm for 5 min. Cells were fixed with 2.5% glutaraldehyde at 4 °C for 2 h, and then the samples were subjected to acetone gradient dehydration, Epon812 embedding, semithin section optical positioning, and ultrathin sectioning. The sections were double-stained with uranyl acetate and lead citrate. Ultrastructure was examined using a TEM (Talos L120C G2, Thermo Fisher, Waltham, Massachusetts, USA). More than three images per group were obtained randomly with the TEM.

Calcium imaging

The AML12 cells were seeded in laser confocal-specific dishes, allowed to adhere, and washed with Hank’s solution three times. Then, 150 μL of a mixture of Fluo-4 AM (5 μM) and F-127 (0.02%) (s1060, beyotime, Shanghai, China) was added to each dish. The cells were incubated in the dark for 30 min. After incubation, the cells were washed three times with Hank’s solution and balanced with 2 mL of Hank’s solution containing 5 μM nimodipine. The Hank’s solution in the control group dishes was replaced with D-Hank’s solution to observe the changes in the intracellular Ca²⁺ concentration in the absence of extracellular Ca²⁺ influx. The other groups were treated accordingly, and the fluorescence signals were measured using laser scanning confocal microscopy. Fluo-4 AM was excited at 488 nm, and the emission was measured at 510 nm.

Statistical analysis

The results are presented as the mean ± standard deviation (SD), and all statistical analysis and graphics were performed using GraphPad Prism 9.0 (GraphPad Software, Inc, San Diego, CA, USA). One-way ANOVA or Student’s t test was used to analyze statistical differences. A value of P < 0.05 was considered as statistically significant.

Results

ASIC1a expression is upregulated in alcohol-associated liver disease

We established murine model of alcohol-associated liver disease followed Gao-Binge model. First, histopathological changes in the liver were analyzed. Liver samples from alcohol-fed animals showed that the structure of the liver lobule was deformed, the arrangement of hepatocyte cords was disordered, and the infiltration of inflammatory cells increased (Fig. 1a). As expected for mice consuming the Lieber-DeCarli liquid diet, Oil red O staining demonstrated steatosis in livers from all 2 groups. However, there were significantly larger and more numerous lipid droplets (LDs) in livers of alcohol-fed mice, whereas LDs in control-fed livers were quite small and few (Fig. 1b). Subsequently, ALT, AST, and TG levels, which are indicators of liver injury, were also measured. As shown in Fig. 1c, compared with control treatment, alcohol stimulation increased the contents of ALT, AST, and TG in the serum. The above results indicated the successful establishment of the alcohol-induced liver injury model in vivo.

Fig. 1. ASIC1a expression is upregulated in alcohol-associated liver disease.

a, b Representative hematoxylin and eosin (H&E) staining and Oil red O staining of liver tissues in ALD. c Serum markers, including AST, ALT and TG, were measured in ALD. d The protein and (e) mRNA levels of ASIC1a in primary hepatocyte in ALD. f The protein and (g) mRNA levels of ASIC1a in liver tissue in ALD. h Immunohistochemistry analysis of ASIC1a expression in mouse liver tissues in ALD. Western blot bands are not from same membrane. *P < 0.05, **P < 0.01 vs CD-fed group.

The specific mechanism of ASIC1a involvement in alcohol-associated liver disease is not clear, therefore, we first investigated ASIC1a expression in the murine model of alcohol-associated liver disease. Western Blot and qRT-PCR were used to detect the expressions of ASIC1a protein and mRNA in the mouse primary hepatocytes. As shown in Fig. 1d, e, ASIC1a protein and mRNA in the mouse primary hepatocytes of EtOH-fed mice were obviously higher than CD-fed mice. The changes of ASIC1a protein and mRNA expression in mouse liver tissue were consistent with those mentioned above (Fig. 1f, g). In addition, we used immunohistochemical technology to detect the expression of ASIC1a in liver tissues, and there was obvious deep staining of ASIC1a protein (Fig. 1h). Taken together, these results indicate that ASIC1a in hepatocytes is significantly elevated in ALD.

Recombinant adeno-associated virus-mediated ASIC1a-shRNA protects against ALD in vivo

To further explore the role of ASIC1a in alcohol-associated liver disease in mice. we used AAV9 adeno-associated virus to coat the ASIC1a silencing plasmid (rAAV9-shASIC1a) to intervene in mice with ALD, and the adeno-associated virus empty vector (rAAV9-Vector) was used as control. The mice were injected with adeno-associated virus by caudal vein (100 μL/mouse) in advance two weeks before feeding with alcohol liquid feed. To detect the liver infection efficiency of adeno-associated virus, we used an inverted fluorescent microscope to image enhanced green fluorescent protein (EGFP) in frozen slices of mouse liver tissue that had been injected with the adeno-associated virus tail vein.

The experimental results showed that both the control group and the silent group showed high fluorescence, indicating that the adeno-associated virus has successfully enriched in the liver (Fig. 2a). The results of immunohistochemistry showed that the staining of ASIC1a was reduced in the rAAV9-shASIC1a group compared with the control group (Fig. 2b). As shown in Fig. 2c–f, the expressions of ASIC1a protein and mRNA were decreased in primary hepatocytes and liver tissues, which were consistent with the results of immunohistochemistry. In addition, the level of AST (P = 0.0883), ALT (P = 0.0950) had a slight decrease, but it was not enough to have a significant statistical difference, and the triglyceride TG level was significantly reduced (Fig. 2g). H&E staining results showed that the rAAV9-shASIC1a treatment resulted in a significant decrease in hepatic steatosis, ballooning of hepatocytes, and inflammation compared to rAAV9-Vector treated mice (Fig. 2h). Oil red O staining further confirmed that, in contrast, rAAV9-shASIC1a treatment showed minimal steatosis, with only a few small foci of steatosis observed in the centrilobular area (Fig. 2i). These results suggested that silencing ASIC1a in the ALD mice could alleviate the pathological changes in the mouse liver to some extent.

Fig. 2. Recombinant adeno-associated virus-mediated ASIC1a-shRNA protects against ALD in vivo.

a The efficiency of rAAV9 adeno-associated virus-mediated ASIC1a knockdown in the liver of mice was examined by observing EGFP. b Immunohistochemistry analysis of ASIC1a expression in mouse liver tissues after ASIC1a knockdown in ALD. c The protein level of ASIC1a in mouse primary hepatocytes after ASIC1a knockdown in ALD. d The protein level of ASIC1a in mouse liver tissues after ASIC1a knockdown in ALD. e The mRNA level of ASIC1a in mouse primary hepatocytes after ASIC1a knockdown in ALD. f The mRNA level of ASIC1a in mouse liver tissues after ASIC1a knockdown in ALD. g Serum markers, including AST, ALT and TG, were measured after ASIC1a knockdown in ALD. h, i Representative hematoxylin and eosin (H&E) staining and Oil red O staining of liver tissues after ASIC1a knockdown in ALD. Western blot bands are not from same membrane. *P < 0.05, ***P < 0.001, ****P<0.0001 vs EtOH-fed+rAAV9-Vector group.

Downregulation of ASIC1a inhibits the activation of ER-phagy

ER-phagy, a form of ER stress-mediated selective autophagy, modulates in ER quality control by removing dysfunctional ER. when the ER-phagy is disordered in some pathological states, the radicalized ER-phagy over-clears the normal intracellular ER, exacerbating the disease process. In order to explore whether alcohol triggered degradation of ER portions within autophagic compartments, electron microscopy of EtOH-treated AML12 cells was performed (Fig. 3a). As forecasted, an increased formation of degradative compartments, comprising autophagosomes as well as autolysosomes, was observed in EtOH-treated cells compared to control cells. Strikingly, about 25% of autophagosomes formed in EtOH-treated cells contained ER fragments, few of which were observed in control cells. This ultrastructural morphological analysis therefore validates and confirms that alcohol induces a type of reticulophagy. Meanwhile, in the liver of mice with ALD, we further detected the expression of six receptors for mammalian ER-phagy which include FAM134B, TEX264, RTN3L, SEC62, ATL3, and CCPG1 (Fig. 3b). FAM134B is an important ER-phagy receptor. It regulates ER-phagy and participates in some ER-phagy-related processes [33]. The results of qRT-PCR showed that FAM134B, RTN3L, ATL3, CCPG1 were statistically different, and the FAM134B was the most obvious. These experiments are analyzed and confirmed that alcohol induces a type of reticulophagy, which is consistent with our previous hypothesis. We observed that alcohol can induce apoptosis in the liver tissue and primary liver cells of mice with ALD (Supplementary Fig. S1a–d). In vitro, alcohol can induce apoptosis in AML12 cells (Supplementary Fig. S2a–c).

Fig. 3. Downregulation of ASIC1a inhibits the activation of ER-phagy.

a Transmission electron microscopy images showing ER fragmentation in EtOH-treated AML12 cells. b The mRNA levels of FAM134B, TEX264, RTN3L, SEC62, ATL3, and CCPG1 in mouse liver tissues. c The mRNA levels of FAM134B, LC3B, ATG5 and ATG7 in mouse primary hepatocytes after ASIC1a knockdown. d The mRNA levels of LC3B, ATG5 and ATG7 in mouse liver tissues after ASIC1a knockdown. e Protein levels of FAM134B, LC3BI, LC3BII and p62 in mouse primary hepatocytes after ASIC1a knockdown. f Protein levels of FAM134B, LC3BI, LC3BII and p62 in mouse liver tissues after ASIC1a knockdown. g Immunohistochemistry analysis of FAM134B expression in mouse liver tissues after ASIC1a knockdown. h Immunofluorescence staining analysis of the expression and localization of FAM134B and LC3B in mouse liver tissues after ASIC1a knockdown. Western blot bands are not from same membrane. *P < 0.05, **P < 0.01, ***P < 0.001 vs CD-fed group. ^#P < 0.05, ^##P < 0.01 vs EtOH-fed+rAAV9-Vector group. ns non-significant.

Next, we detected the effect of ASIC1a knockdown on ER-phagy in the liver of ALD mice. The qRT-PCR results showed that the mRNA expression levels of FAM134B, LC3B, ATG5 and ATG7 were increased significantly in EtOH-fed group, but the silencing ASIC1a group had the opposite results (Fig. 3c, d). Knockout of ASIC1a in vivo can reduce alcohol induced apoptosis (Supplementary Fig. S3a, b). Silencing or blocking ASIC1a in vitro can also inhibit alcohol induced apoptosis in AML12 cells (Supplementary Fig. S4a–c). We continued to detect the expression of the protein of ER-phagy related indicators in primary hepatocytes and liver tissues of ALD mice. The results showed that the level of ER-phagy was increased in EtOH-fed mice and decreased in rAAV9-shASIC1a treated mice (Fig. 3e, f). Then we used immunohistochemical technology to detect the expression of FAM134B in liver tissues, and mice treated with rAAV9-Vector exhibited deep staining of FAM134B protein, while minimal staining for FAM134B was seen in rAAV9-shASIC1a treated mice (Fig. 3g). Moreover, immunofluorescence double staining for FAM134B and LC3B showed reduced expression and co-localization of both FAM134B (green) and LC3B (red) in the liver of mice after silencing ASIC1a (Fig. 3h). It was suggested that downregulation of ASIC1a inhibited the activation of ER-phagy.

ASIC1a mediates the activation of ER-phagy through triggering ER stress in alcohol-induced AML12 cells

Growing evidence has indicated that ER stress is closely linked to the activation of ER-phagy, but will alcohol-induced ER stress activate endoplasmic reticulum autophagy? The AML12 cells were treated with or without alcohol in the presence or absence of thapsigargin (TG) for control analysis. The results showed that in Fig. 4a, both alcohol and TG could induce ER stress in AML12 cells, while TG triggered more severe endoplasmic reticulum stress. In addition, PcTx-1 reduced the degree of ER stress in alcohol-treated AML12 cells. An interesting discovery was shown in Fig. 4b, TG activated autophagy more intensely than alcohol. Meanwhile the expression of ER-phagy receptor FAM134B decreased in TG-treated AML12 cells, but increased in alcohol-treated AML12 cells. Moreover, PcTx-1 inhibited the activation of ER-phagy in alcohol-treated AML12 cells. It is suggested that ER stress could trigger the activation of ER-phagy in alcohol-treated AML12 cells, which is related to ASIC1a.

Fig. 4. ASIC1a mediates the activation of ER-phagy through triggering ER stress in alcohol-induced AML12 cells.

a The protein levels of IRE1α and GRP78 in AML12 cells after PcTx-1 and/or TG treatment. b The protein levels of FAM134B, LC3BI and LC3BII in AML12 cells after PcTx-1 and/or TG treatment. c The protein levels of FAM134B, LC3BI, LC3BII and p62 in AML12 cells after 4-PBA treatment. d Representative images of AML12 cells transfected with ptfLC3-mCherry-GFP plasmid after 4-PBA treatment. e Immunofluorescence staining analysis of the expression and localization of LC3B and Calreticulin in AML12 cells after 4-PBA treatment. Western blot bands are not from same membrane. *P < 0.05, **P < 0.01, ***P < 0.001. ^#P < 0.05, ^##P < 0.01.

To further confirm this discovery, AML12 cells were treated with alcohol combined with an ER stress inhibitor, 4-phenylbutyrate (4-PBA). As expected, inhibition of ER stress by 4-PBA weakened the activation of ER-phagy (Fig. 4c). 4-PBA inhibited alcohol induced endoplasmic reticulum stress and cell apoptosis (Supplementary Fig. S5a, b). Various researches revealed that induction of ER-phagy required the core autophagy machinery. Next, we used ptfLC3 plasmid to detect autophagic flux. After AML12 cells were transiently transfected with ptfLC3 plasmid, 4-PBA was administered two hours before alcohol treatment, as shown in Fig. 4d, inhibition of ER stress by 4-PBA significantly decreased the portion of autolysosomes. In addition, we observed that alcohol treatment induced obvious colocalization of LC3 with calreticulin (ER marker) (Fig. 4e). Collectively, these results demonstrated that alcohol activated ER-phagy through triggering ER stress, which is related to ASIC1a.

ASIC1a-mediated intracellular Ca²⁺ dysregulation contributes to ER stress in alcohol-induced AML12 cells

Next, we set out to investigate the mechanism underlying alcohol-induced ER stress. The ER is an important Ca²⁺ storage station, and intracellular calcium disorders affect ER functions, which triggers ER stress. We hypothesized that intracellular calcium overload mediated by alcohol-induced ASIC1a over-expression was a potential inducer of ER stress. ASIC1a is solely permeable to Ca²⁺ among ASICs subunits, and our previous studies verified that ASIC1a has a crucial role in liver-related diseases.

To ascertain our hypothesis, we examined ASIC1a expression following alcohol treatment, and found increased expression of ASIC1a protein and mRNA in alcohol-treated AML12 cells (Fig. 5a, b). Then, psalmotoxin-1 (PcTx-1) was used to specifically block ASIC1a in AML12 cells. PcTx-1 reduced alcohol-induced elevation of intracellular Ca²⁺ concentration in AML12 cells (Fig. 5c, Supplementary Fig. S5c).

Fig. 5. ASIC1a-mediated intracellular Ca²⁺ dysregulation contributes to ER stress in alcohol-induced AML12 cells.

a, b The protein and mRNA level of ASIC1a in EtOH-treated AML12 cells. c The changes of [Ca²⁺]_i concentration visualized by Fluo‐4 AM in AML12 cells after PcTx-1 treatment. d The protein levels of IRE1α and GRP78 in AML12 cells after PcTx-1 treatment. e The protein levels of ASIC1a, IRE1α and GRP78 in AML12 cells after the knockdown of ASIC1a. f The mRNA levels of ASIC1a, IRE1α and GRP78 in AML12 cells after the knockdown of ASIC1a. Western blot bands are not from same membrane. *P < 0.05, ***P < 0.001 vs Control group. ^#P < 0.05, ^##P < 0.01 vs EtOH group. ^&P < 0.05, ^&&P < 0.01 vs EtOH+shScramble group.

Meanwhile, we observed a decrease in the expression of alcohol-induced ER stress markers (Fig. 5d). Furthermore, knockdown of ASIC1a by ASIC1a-silencing plasmid also significantly countered alcohol-induced ER stress (Fig. 5e, f). These data suggested that ER stress is a downstream molecular event of ASIC1a expression in alcohol-induced AM12 cells.

Inhibition of ER-phagy attenuates alcohol susceptibility to AML12 cells

To investigate whether ER-phagy was involved in the ALD. Firstly, AML12 cells were treated with alcohol combined with autophagy inhibitors 3-MA. Transient transfection of ptfLC3 plasmid was applied to observe autophagic flux changes. As shown in Fig. 6a, 3-MA led to remissive autophagy intensity of alcohol-treated AML12 cells. In addition, we observed a decrease in ER-phagy following 3-MA treatment in alcohol-treated AML12 cells, as indicated by Western Blot and qRT-PCR (Fig. 6b). Moreover, recovery of cell viability and a decrease in cytotoxicity were also observed in AML12 cells after co-treatment with alcohol and 3-MA (Fig. 6c, d). Furthermore, siRNA-mediated silencing of FAM134B resulted in a decrease in both ER-phagy and apoptosis in alcohol-treated AML12 cells (Fig. 6e). Together, these data indicated that ER-phagy promotes alcohol-induced AML12 cell death, and inhibition of autophagy/ER-phagy results in diminished sensitivity of AML12 cells to alcohol in vitro.

Fig. 6. Inhibition of ER-phagy attenuates alcohol susceptibility to AML12 cells.

a The representative images of AML12 cells transfected with ptfLC3-mCherry-GFP plasmid after 3-MA treatment. b The protein levels of FAM134B, LC3I, LC3II and p62 in AML12 cells after 3-MA treatment. c The apoptosis of AML12 cells treated with 3-MA was assessed by FDA/PI staining. d The cell viability of AML12 cells treated with 3-MA detected by CCK8. e The protein levels of FAM134B, Bax and Bcl-2 in AML12 cells after the knockdown of FAM134B. Western blot bands are not from same membrane. *P < 0.05, **P < 0.01, ***P < 0.001 vs Control group. ^#P < 0.05, ^##P < 0.01 vs EtOH group. ^&P < 0.05, ^&&P < 0.01 vs EtOH+ si-Scramble group.

Discussion

ALD includes alcoholic steatosis, hepatitis, liver fibrosis and cirrhosis, with ALD being the initial stage in the development of the disease [2]. Currently, abstinence is the only effective prevention and treatment for alcohol-related liver disease [34]. Exploring targeted therapies for ALD has become increasingly important. In this experiment, an in vivo model of Gao-Binge model and an in vitro model of alcohol-treated AML12 cell line were established to study the effects of endoplasmic reticulum autophagy in alcohol-associated liver disease, and mechanistic studies also demonstrated the role of ASIC1a in endoplasmic reticulum stress-mediated endoplasmic reticulum autophagy.

Early studies identified that the pathogenesis of alcohol-related liver disease was closely linked to oxidative stress, glutathione depletion, organism malnutrition, alcohol-mediated leakage of intestinal endotoxins and subsequent activation of hepatic Kupffer cells associated [35]. Recent studies have found that autophagy plays an important role in intrahepatic homeostasis, suggesting that the regulation of autophagy may be relevant to the pathogenesis of alcoholic liver disease [36]. The involvement of autophagy in alcoholic liver disease is complex. In general, acute ethanol exposure activates autophagy, while chronic ethanol exposure does the opposite, but there is some controversy. Earlier, Ding et al. discovered increased levels of LC3II protein in mouse primary hepatocytes as well as significantly enhanced GFP-LC3 fluorescent spots in mouse liver tissues in an acute ethanol exposure model, indicating autophagy activation [37]. Later, Chen et al. showed that acute ethanol primarily stimulates autophagy through induction of enzyme systems such as CYP2E1 and NOX4 and increased ROS generation such as H₂O₂ [38]. Another study found that chronic alcohol consumption caused autophagy inhibition. And Chao et al. revealed that chronic ethanol exposure impaired the function of transcription factor EB (TFEB), decreased lysosome biogenesis, and resulting in insufficient autophagy [39]. However, some research has revealed that autophagy is activated in both acute and chronic ethanol exposure [40]. It has been reported that selective autophagy can be significantly activated in models of acute ethanol exposure, selectively clearing damaged mitochondria and accumulated lipid droplets through mitophagy and lipophagy [37]. Pang et al. showed that when HepG2 cells were treated with 400 μM oleic acid (OA), increased ER-phagy was induced 8 h after treatment, which was associated with an anti-apoptotic response as shown by the activation of the PI3K/AKT pathway, an increase in Bcl-2 expression, and the downregulation of OA-induced lipotoxicity [17]. When the cells were treated with OA for 24 h, damage-regulated autophagy modulator (DRAM) expression-dependent mitophagy resulted in increased apoptosis in HepG2 cells. ER-phagy influences the pathology of non-alcoholic fatty liver disease (NAFLD). However, the role of ER-phagy in ALD remains unclear. Therefore, this experiment is dedicated to investigate whether ER-phagy influences the pathological progression and the possible mechanisms of ALD, to find new therapeutic possibilities.

We established murine model of ALD followed Gao-Binge model by chronic alcohol feeding and acute alcohol gavage. Our results showed that the expressions of ER-phagy related proteins were significantly upregulated, along with increased liver injury and hepatocyte apoptosis in primary hepatocytes and liver tissue of the EtOH-fed mice. Meanwhile, the expression of FAM134B in the ER-phagy receptors was the most different. We treated AML12 cells with 200 mM EtOH for 8 h in vitro. The results showed that EtOH upregulated the expression of FAM134B and LC3B, which together mediated excessive ER-phagy, characterized by forming elevated numbers of autophagosomes with larger sizes detected by transmission electron microscopy. The use of small molecule inhibitors or siRNA to inhibit ER-phagy significantly reduces hepatocyte apoptosis and liver damage, contrary to the common belief that ER-phagy is a protective mechanism. In recent years, there have been reports proving that ER-phagy can be a non-protective role involved in the process of disease. In Z36 treatment induced the HeLa cells, the excessive ER-phagy accelerated ER degradation and impaired ER homeostasis, and thereby triggered ER stress and the UPR, as well as ER-phagy dependent cell death [41]. Zhang et al. revealed that berberine exerts protecting effects on steatotic livers undergoing transplantation by inhibiting ER stress-mediated ER-phagy [42]. In the ER stress model of H9c2 cardiomyoblasts induced by thapsigargin (TG) or tunicamycin (TM), ER stress mediated excessive ER-phagy, thereby inducing cell injury, which was related to the PERK/Nrf2 pathway. Recently, loperamide has been reported to induce ER stress, ER-phagy, and autophagic cell death by triggering the expression of the transcription factor ATF4 [18]. In summary, we determined that ER-phagy is involved in the pathological process of ALD, and that inhibition of ER-phagy can alleviate alcohol-induced liver injury.

Numerous studies have proven that ER stress not only induces apoptosis, but also mediates the onset of ER-phagy with the formation of autophagosomes containing ER fragments. Pang et al. [17] uniquely elucidated the anti- and pro-apoptotic functions of autophagy, which seem to be manifested through ER-phagy and mitophagy. There is always a crosstalk between autophagy and apoptosis, and the components of the autophagy process may directly play a role in regulating apoptosis. The VPS34-Beclin1 complex is essential in the formation of autophagic vesicles. Previous studies have identified Beclin1 as a BH3-only protein, carrying a Bcl-2-homology-3 (BH3) domain. Beclin1 interacted with anti-apoptotic proteins of the Bcl-2 family (particularly Bcl-2 and its homolog Bcl-XL) by the BH3 domain. It was shown that the Beclin1-Bcl-2/Bcl-XL complexes, which commonly inhibit autophagy, are specifically located in the ER and point to an organelle-specific regulation of autophagy [43]. Rubio et al. found that ER photodamage induced ER-phagy and transmitted ROS to the nearby mitochondria, triggering mitochondrial damage, which in turn amplified oxidation-related signals and activated caspase-3, causing apoptosis [44]. We inhibited ER-phagy by knocking down FAM134B, and found that EtOH-induced AML12 cell death was reduced, suggesting that inhibition of excessive ER-phagy may help alleviate alcohol-induced hepatocyte injury. And autophagy inhibitor 3-MA had the similar effect. Targeting ER-phagy is more specific than targeting autophagy and has the significant advantage of avoiding the potential side effects associated with autophagy inhibition. This experiment only explored the effect of endoplasmic reticulum autophagy on hepatocyte injury and apoptosis in ALD, but the detailed relationship and interaction mechanism between ER- phagy and apoptosis are still unknown and need to be further investigated.

We explored possible mechanisms of alcohol-induced ER stress, and revealed that ASIC1a-mediated disturbance of intracellular Ca²⁺ homeostasis is a key factor in activating the ER stress of hepatocytes. The ER is the cellular Ca²⁺ storage, which plays an important role in the production and regulation of Ca²⁺ signaling [45]. Disturbance in intracellular Ca²⁺ homeostasis affects ER function, including protein translation, synthesis and folding, triggering the UPR and ER stress [46]. Research has shown that there is expression of alcohol metabolizing enzymes ADH and ALDH2 in AML12 cells [47, 48]. Furthermore, succinylation of ALDH2 at lysine 385 abolished its protective effect against APAP-induced acute liver injury. Alcohol stimulation can induce endoplasmic reticulum stress in AML12, while ethanol can induce ALD and liver endoplasmic reticulum stress in mice. Research has shown that ethanol and its nonoxidative metabolites, FAEEs, not acetaldehyde, promoted acute alcohol-induced liver injury by inducing ER stress, adipocyte death, and lipolysis [49]. ASIC1a belongs to the acid-sensitive ion channel (ASICs) family, which is the main subtype of ASICs mediating Na⁺ and Ca²⁺ influx [50]. In addition to hepatocytes, hepatic stellate cells (HSCs) also express ASIC1a in the liver. HSCs are one of the main pathogenic cell types that cause liver fibrosis. When activated, they accumulate extracellular matrix, leading to liver fibrosis, further developing into cirrhosis and liver failure, and ultimately leading to death [51, 52]. Our previous research has shown that ASIC1a is involved in regulating the activation of hepatic stellate cells and liver fibrosis, and PDGF‐activated ASIC1a by the PI3K/AKT pathway, induced ER stress and promoted the development of liver fibrosis [31] and we found that there is an interaction between ASIC1a and ER stress. Then, we demonstrated that ASIC1a and ER stress are involved in insulin deficiency-induced neuronal apoptosis, and when ASIC1a was blocked, the expression of ER stress-related protein CHOP was decreased [53]. Interestingly, ASIC1a expression was also reduced after CHOP knockdown. It further proves that ASIC1a and endoplasmic reticulum stress interact with each other, and whether the mechanism is related to Ca²⁺ homeostasis requires further research. Many reports have shown that ethanol affects ligand-gated and voltage-gated ion channels, including glutamate-NMDA receptor ion channels, GABA receptor ion channels and voltage-gated Ca²⁺ and K⁺ channels. Zhou et al. found that acute treatment with moderate ethanol could reduce ASIC1a protein expression, ASIC currents, and acid-induced [Ca²⁺]_i elevation, which protected neuronal cells against acidosis-induced cytotoxicity [29]. However, the effect of ethanol on the activity/expression of ASIC1a channels in ALD remains unclear. We performed verification in CD-fed and EtOH-fed mice. The results showed that the expression of ASIC1a was significantly increased, and ER stress was activated in EtOH-fed mice. After ASIC1a-specific knockdown in the liver of ALD mice, the intensity of ER stress and the serum ALT and AST levels decreased. When ASIC1a was blocked or knocked down, increased expression of ASIC1a, ASIC1a-mediated Ca²⁺ influx and elevated levels of ER stress by acute induction with 200 mM alcohol were inhibited in AML12 cells. However, the molecular basis underlying these interactions still needs further investigation.

In summary, our experimental results show that ER-phagy is involved in the development of ALD. Ethanol mediates excessive ER-phagy by activating ER stress, thereby promoting hepatocyte injury and apoptosis. Silencing the ER-phagy receptor FAM134B inhibits AML12 cell apoptosis. We then investigated the possible mechanisms of ethanol-induced endoplasmic reticulum stress and found ethanol activates ER stress by increasing ASIC1a expression and ASIC1a-mediated Ca²⁺ influx. In vitro and in vivo silencing or blockade of ASIC1a inhibits ER stress activation and its mediated ER-phagy, thereby alleviating the progression of ALD.

Author contributions

All the authors contributed to the study conception and design. YQZ, LLW and ZHL conceived and carried out the experiments, analyzed the data, and wrote the manuscript. SSQ, ZX and JZ fed the mice and collected the data. YHS, XSP and ND conducted molecular biology experiments. AAE, LJT, JRZ and MXL conducted data organization and analysis. YXS and YH conceived the study and revised the manuscript. All authors have read and approved the final submitted manuscript.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41401-024-01423-4.