PERK/ATF4/LAMP3-arm contributes to the aggressive phenotype of hepatocellular carcinoma cells in response to the chronic ethanol exposure

Abstract

Hepatocellular carcinoma (HCC), the predominant form of liver cancer, is the fifth most prevalent cancer worldwide and second leading cause of cancer-related deaths, underscoring its grim prognosis. Key risk factors for HCC include HBV, HCV, cirrhosis, inherited metabolic diseases, vitamin supplementation, heavy alcohol use, obesity etc. Despite the well-documented impact of alcohol on HCC, there remains a significant gap in understanding alcohol-associated HCC (A-HCC) compared to viral hepatitis associated HCC. So, in this study we tried to elucidate the role of UPR pathway in exacerbating HCC prognosis under the condition of A-HCC. Notably, our RT-qPCR and western blot analysis showed significant upregulation of PERK-ATF4-LAMP3 arm along with CHOP, VEGF-A and nuclear translocation of NF-KB in both HepG2 and Hep3B cell lines. Increased extracellular & intracellular cholesterol and triglyceride levels obtained can be related with the higher expression of SREBP-2 and SREBP-1c respectively. Ethanol exposure also enhanced the invasive and migratory properties of HCC, reduced apoptosis with increased stemness in a PERK dependent manner. Moreover, orally available PERK inhibitor (GSK2606414) successfully relieved the effects caused by ethanol in HepG2 and Hep3B cell lines. In summary, HCC cells gain aggressiveness due to ethanol exposure via PERK/ATF4/LAMP3 pathway, and targeting PERK could serve as a promising therapeutic strategy for A-HCC, mitigating several cancer hallmarks.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-37114-9.

Introduction

Hepatocellular carcinoma (HCC) presents a major global health issue, being the fifth most prevalent malignant tumors in humans¹. It’s the primary form of liver cancer and ranks second in cancer-related mortality worldwide, highlighting its dire prognosis². Major risk factors for HCC include HBV and HCV but cirrhosis, inherited metabolic diseases, vitamin supplementation, heavy alcohol use, obesity, and tobacco use also pose significant risks for HCC³. While the global impact of alcohol consumption on the onset, severity, and fatality of HCC is widely acknowledged, there remains a significant gap in our understanding of alcohol-associated HCC (A-HCC) when compared to the more extensively studied viral hepatitis-associated HCC. Furthermore, the divide between basic research and clinical application persists mainly due to the absence of robust preclinical models for A-HCC, hindering the translation of findings into clinical practice⁴. Also, the disparities between A-HCC and HCCs stemming from alternative causes have not received adequate exploration concerning clinical prognosis, genetic and epigenetic profiles, and its molecular mechanisms⁵. Prolonged and excessive alcohol consumption triggers a range of pathological stress responses, including the induction of the endoplasmic reticulum (ER) stress response which occurs due to the build-up of misfolded proteins in the ER, prompting the activation of the unfolded protein response (UPR) via three membrane sensors: inositol requiring enzyme 1 alpha (IRE-1α), activating transcription factor (ATF-6), and protein kinase R-like kinase (PERK)⁶. Usually, these sensor proteins remain deactivated by binding to glucose-regulated protein (GRP78). However, during the heightened ER stress, association with GRP78 is disrupted, initiating UPR signal transduction. Activation of the IRE-1α arm further causes the splicing of XBP1, leading to the generation of the spliced form of XBP1⁷. On the other hand, activation of ATF4, the key regulator of the UPR pathway, occurs in a PERK-dependent manner, which subsequently inhibits the global translation process. ATF4 can further increase the expression of CHOP (C/EBP homologous protein) and other stress-related genes⁸. PERK activation also intensifies the ROS production, which subsequently induces the NF-κB expression and hence inflammation and proliferation, which can further exacerbate ER stress creating a feedback loop that maintains both UPR and NF-κB signaling⁹. Previous literature suggests a potential feed-forward loop between oxidative stress and ER stress sensors, but this was not explored in our study. In addition, ATF4 can directly or indirectly regulate the VEGF-A expression, a key regulator of angiogenesis¹⁰. Our previous study showed the engagement of the PERK-ATF4 arm in causing intensified metastatic activity by activating LAMP3 in folate-deficient HCC cells¹¹. In case of A-HCC, ethanol metabolism generates ROS which further affects the transcription of SREBP-1c and SREBP-2. The enhanced activation of SREBPs due to ethanol-induced oxidative stress, ER stress leads to the activation of hepatic lipogenic genes which contributes to the development of liver diseases such as hepatic steatosis and steatohepatitis¹². In contrast to the activation of CHOP gene and hence apoptosis by the PERK-ATF4 arm, it also enhances the transcription of LC-3 and other autophagy related genes facilitating the autophagic process during ER stress¹³. The intensity and duration of UPR dictate whether the cell adopts a pro-survival or pro-apoptotic state¹⁴. Moderate ER stress can foster cell proliferation, angiogenesis, and metastasis, while severe ER stress can lead to cell death¹⁵.

This study seeks to investigate the role of activation of the UPR pathway under the conditions of chronic ethanol exposure in the adverse prognosis of HCC. The results of the study demonstrate the actuation of the PERK-ATF4 branch of the UPR pathway, accompanied by an increase in the cancer hallmarks indicative of more aggressive phenotype. Blocking the PERK pathway successfully decreased the aggressiveness of carcinoma in the case of A-HCC. Consequently, these findings suggest that inhibiting the PERK arm of the UPR pathway in A-HCC could yield suboptimal outcomes for HCC, thereby pointing towards potential improvements in therapeutic strategies.

Methodology

Cell culture and ethanol treatment

Three distinct cell lines associated with hepatocellular carcinoma viz. HepG2, hepatoma cell line; Hep3B, originating from Hepatitis B virus; and Huh-7 cells derived from Hepatitis C virus were employed in this study and were attained from the National Centre for Cell Science (NCCS), Pune, India. All the three cell lines were authenticated using STR profiling. All three cell lines were nurtured in a growth medium consisting of minimal essential medium (MEM) and 10% Fetal bovine serum (FBS) sourced from HiMedia Laboratories. The growth medium was supplemented with gentamycin, penicillin, sodium bicarbonate, and streptomycin. For chronic ethanol exposure, each of the three cell lines was subjected to four different concentrations of molecular-grade ethanol as determined by the previous literature i.e., 50mM, 100mM, 150mM, and 200mM. The exposure to ethanol lasted for five days without any intervening passages. To mitigate ethanol evaporation, T-25 flasks were placed within the containers containing an equivalent concentration of ethanol. Additionally, the containers were sealed tightly with parafilm to prevent any potential evaporation. 2.5µM concentration of GSK2606414 (PERK inhibitor, Tocris, India) was prepared in ethanol-containing media and normal media from the stock solution of 100µM prepared in dimethyl sulfoxide (DMSO).

MTT assay

To assess the impact of sealed container storage and ethanol treatment on cell viability, cells were cultured in 96-well plates from three different cell lines: HepG2, Hep3B, and Huh-7cell lines were subjected to treatment with ethanol at four distinct concentrations i.e., 50mM, 100mM, 150mM and 200mM. Following treatment, 10µL of MTT solution (5 mg/mL) was introduced into each well and incubated for 4 h. Subsequently, dissolution of formazan crystals was done with the addition of 50µL of DMSO, followed by measuring the absorbance at 540 nm.

Cholesterol, triglycerides (TG), and lactate dehydrogenase (LDH) measurement

For the measurement of intracellular levels of cholesterol (colorimetric method), TG (enzymatic colorimetric method), and LDH (UV-based assay), the cell pellet was suspended in RIPA lysis buffer, and media of the different cells was used for the measurement of extracellular levels using the Cobas.

Quantitative real-time PCR

RNA extraction from HepG2, Hep3B and Huh-7 cells was conducted using TRIzol Reagent, followed by cDNA synthesis employing IScript™ cDNA synthesis kit from Bio-Rad, India. SYBR green master mix from TAKARA, USA was used to execute qRT-PCR. The expression of GAPDH gene expression was utilized as an internal control for normalization process. Primer sequences for GAPDH, GRP-78, PERK, ATF-4, LAMP-3, CHOP, LC-3, VEGF-A, ATF-6, SREBP-1c, SREBP-2, IRE-1, XBP-1, NF-κB are provided in Table 1. Relative gene expression for these genes was determined using the standard 2^− ΔΔct method.

Table 1.

Primer sequences for different genes.

Gene	Forward primer (5′–3′)	Reverse primer (5′–3′)
GAPDH	ACATCGCTCAGACACCATG	TGTAGTTGAGGTCAATGAAGGG
GRP-78	TCAGGCCAAGCCCAATACAG	TCCACGGTAGTGAGAGCCTT
PERK	ACGATGAGACAGAGTTGCGAC	ATCCAAGGCAGCAATTCTCCC
ATF4	CCTTCACCTTCTTACAACCT	GTAGTCTGGCTTCCTATCTC
LAMP3	AGCAAGCACCTCACCAAACTTT	AATTTTTACTGTGGCCGCTGTT
CHOP	GCTCAGGAGGAAGAGGAGGA	TCCTGCTTGAGCCGTTCATT
LC-3	AGCAGCATCCAACCAAAATC	CTGTGTCCGTTCACCAACAG
VEGF-A	CGAGGGCCTGGAGTGTGT	CCGCATAATCTGCATGGTGAT
ATF-6	ACCGTATTCTTCAGGGTGC	CACTCCCTGAGTTCCTGCTG
SREBP-1c	GGAGCCATGGATTGCACATT	GGCCCGGGAAGTCACTGT
SREBP-2	CGGTAATGATCACGCCAACA	CGGTAATGATCACGCCAACA
IRE-1	GCCGAAGTTCAGATGGAATC	ATCTGCAAAGGCCGATGA
XBP-1	TGGTTCAGCCTCTTAACTCGG	TGCACGTAGTCTGAGTGCTG
NF-κB	ACCTTCAAATATTAGAGCAACCTAAACA	CATGGGATGGGCCTTCAC

Semi-quantitative RT-PCR

For the amplification of both spliced and unspliced XBP1 mRNA, XBP1 primers (forward: 5′TTACGAGAGAAAACTCATGGCC-3′, reverse: 5′GGGTCCAAGTTGTCCAGAATGC-3′) were utilised. The PCR products formed were subjected to electrophoresis on 2.5% agarose gel. Notably, a size difference of 26 nucleotides distinguishes the spliced form (263 bp) from the unspliced XBP-1 fragments (289 bp).

Western blot

For protein extraction, cultured cells were harvested and sonicated with RIPA buffer (HiMedia laboratories) with the addition of protease and phosphatase inhibitors. BCA protein kit from Thermo Fisher Scientific Inc., Rockford, USA was employed to determine the concentration of protein. Equal amounts of the samples were seperated using10% sodium dodecyl sulfate-polyacrylamide gel for electrophoresis which were then transferred onto a PVDF membrane. Primary monoclonal antibodies against GRP-78 (1:500; Affinity Biosciences), ATF-4 (1:500; Affinity Biosciences), ATF-6 (1:400; Affinity Biosciences), CHOP (1:1000; Cell Signaling Technology), LAMP-3 (1:1000; Affinity Biosciences), eIF-2α (1:1000; Cell Signaling Technology), p- eIF-2α (1:800, Cell Signaling Technology), PERK (1:500; Affinity Biosciences), p-PERK (1:1000; Cell Signaling Technology), and β-actin (1:5000, Cell signaling Technology) were used. This was followed by the incubation with the secondary antibody conjugated to horseradish peroxidase (1:10,0000; Cell Signaling Technology). A chemiluminescent signal was detected using ECL detection reagents (BioRad, India). Densitometric analysis of blots was done using Image J software.

Immunofluorescent staining

After washing the cultured cells with PBS, they were fixed using 4% paraformaldehyde for 15 min. Subsequently, 0.1% Triton X-100 was used to permeabilize the cells at room temperature. Cells were incubated with a blocking buffer (2% BSA in PBST). After washing, cells were treated with primary antibody for NF-κB (1: 1000; Cell Signaling Technology) at 4 °C overnight. After washings, cells were treated with Alexa Flour 594 (red, 1:1000; Cell Signaling Technology) for 1 h. DAPI (Thermofisher) and phalloidin iflour 488 reagents (Abcam) were used to counterstain the nucleus and actin filaments, respectively. Images were visualized under a confocal microscope (Nikon, Japan) and analyzed using FIJI software. Three fluorescent fields were captured using consistent settings, including identical exposure time and magnification. The average fluorescent intensity per pixel was then measured for each image without any adjustment, utilizing Hue values of 188 and 215 to identify the red and blue colors, respectively.

Scratch assay/migration assay

HepG2 cells and Hep3B cells were grown in 6-well plates under the specific medium conditions for the control group, 150mM concentration for the ethanol-treated-group and 150mM ethanol with the 2.5µM concentration of GSK2606414 for the PERK inhibition. A sterile 10µL pipette tip was utilized to create wounds, followed by washing with PBS. Microscopic examination involved photographing at least three fields at 6 different time points until the wound gets closed. In addition, ten distances were measured within each field using a phase contrast microscope.

Invasion assay

HepG2 and Hep3B cells were grown in a transwell chamber coated with Matrigel with the addition of 100 µL of serum-free medium. The lower chamber was filled with complete media containing FBS. The cells that had invaded were stained with crystal violet after the incubation of 24-hours. The stain was then extracted using glacial acetic acid, and their absorbance was quantified at 540 nm.

Sphere formation assay

In order to establish low attachment conditions, 6-well plates were prepared by coating them with 1% agarose, which was solidified and sterilized under UV for 6 h. Subsequently, 3000 cells were seeded into each well along with the specific conditions and allowed to incubate for 15 days. Proliferation of single stem cell gives rise to one sphere. After the incubation period, number of spheres were counted under microscope from each well which was further divided into nine fields.

Annexin/PI assay

Apoptosis was evaluated utilizing the annexin V-FITC apoptosis detection kit (BD, Pharmigen, USA) according to the manufacturer’s instructions. HepG2 and Hep3B cells were exposed to ethanol for four days in T-25 flasks. On the fourth day of ethanol treatment, cells were treated with PERK inhibitor for 24 h. After treatment, trypsin was used to detach the cells followed by the washing with PBS. The collected cells were then incubated with annexin V (1µL) and PI (5µL) dyes at room temperature for 15 min in dark. Ultimately, flow cytometry analysis of stained cells was conducted using a Beckman Coulter (CA, USA).

ROS assay

To measure cellular ROS levels, DCFH-DA (2′7′-dichlorodihydroflourescein diacetate) from Sigma, USA was incubated at a concentration of 10µM in dark at room temperature for 30 min. It then exhibits a green color fluorescence at 485 nm excitation and 530 nm emission wavelength. Images from every well were clicked under confocal microscope and the percentage of DCFH-DA positive cells were measured using flow cytometry.

Statistical analysis

The results were represented as the mean± standard error of the mean (SEM), with each experiment repeated thrice or biological replicates. For comparison between two groups and more than two groups, statistical significance was assessed using a t-test and one-way ANOVA or two-way ANOVA respectively, using GraphPad Prism software version 9. A value p < 0.05 was considered to indicate statistical significance. All values represent the mean±SEM of three independent experiments, with significance levels indicated as follows: *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001(n = 3 compared to control), *p < 0.05; ^%p < 0.001; ^$p < 0.01; ^#p < 0.0001 for results of RT-qPCR.

Results

Effect of chronic ethanol exposure on cell morphology and cell viability

Following the treatment with three different concentrations of ethanol as outlined in prior literature, all the experimental groups exhibited epithelial-like morphology, with normal nuclei observed in all three cell lines: HepG2 (Fig. 1A), Hep3B (Fig. 1B) and Huh-7 (Fig. 1C). Nonetheless, a slight decrease in growth rate was noted in ethanol-treated cells compared to the control group.

Fig. 1.

Effects of chronic ethanol exposure in HepG2, Hep3B, and Huh-7 cell lines. (A) Representative images of the cells at 20X magnification treated with 50mM, 100mM, and 150mM concentration of ethanol at day 1 and day 5 vs. control (A) HepG2 cells (B) Hep3B cells (C) Huh-7 cells. (D) Cell viability of HepG2, Hep3B, and Huh-7 cells after chronic treatment of 5 days at 50mM, 100mM, 150mM, and 200mM concentrations of ethanol. (E) Intracellular and extracellular levels of triglycerides, cholesterol, and LDH in HepG2, Hep3B and Huh-7 cell lines. Data taken from three independent experiments (n = 3) is displayed as mean±SEM. Ordinary one-way ANOVA, and Dunnett’s multiple comparison was employed to check statistical significance and to compare various groups with the control group respectively. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 vs. control. Scale bar = 200 μm.

In order to check the effect of chronic ethanol exposure for 5 days on cell viability, HepG2, Hep3B, and Huh-7 cells were exposed with 50mM, 100mM, 150mM, and 200mM concentrations of ethanol. Treatment of the cells with a 200mM concentration of ethanol caused a significant reduction (p < 0.0001) in the cell viability in all three cell lines (< 50%). However, cell viability was more than that of 50% in comparison to the control cells when they were treated with 50mM, 100mM and 150mM concentrations of ethanol in all the cells (Fig. 1D). Hence, for further experiments, we proceeded with these three concentrations, and 200mM was excluded.

mRNA expression of different genes of the UPR pathway under the conditions of chronic ethanol exposure

Effect of chronic ethanol exposure on GRP-78 expression in HepG2, Hep3B and Huh-7 cells

To study the effect of chronic ethanol exposure on HepG2, Hep3B, and Huh-7 cells, cells were administered with 50mM, 100mM, and 150mM concentrations of ethanol, and mRNA expression of different genes of the UPR pathway was measured using RT-qPCR.

GRP-78, an ER stress sensor, showed a significant increase (5-fold, ^$p < 0.01; 19-fold, ^#p < 0.0001respectively) in the 100mM and 150mM ethanol-treated HepG2 cells whereas 50mM ethanol did not show any considerable alteration in comparison to control cells (Fig. 2A).In the case of Hep3B cells, only the 150mM ethanol-treated group showed a noteworthy elevation (6-fold, ^$p < 0.01) in the expression of GRP78, whereas no substantial change was observed at the lower concentrations studied (Fig. 2B).Huh-7 cells also exhibited a substantial elevation in the expression of GRP78 levels, with a noteworthy 2.5-fold increase (^%p < 0.001) in the group treated with 50mM ethanol, 1.7-fold increase in 100mM (^$p < 0.01) and 1.5-fold in 150mM ( *p < 0.05) ethanol-treated groups (Fig. 2C).

Fig. 2.

Gene expression of different genes of the UPR pathway after chronic ethanol exposure. RT-qPCR of GRP-78, PERK, ATF-4, LAMP-3, NF-κB, CHOP, VEGF-A, LC-3, ATF-6, IRE-1, XBP, SREBP-1c and SREBP-2 genes in (A) HepG2 cells (B) Hep3B cells (C) Huh-7 cells. GAPDH served as an internal control. Data (n = 3) is represented as mean±SEM from three independent experiments. One-way ANOVA was employed to determine the statistical significance, and Dunnett’s multiple comparison test was used to compare different treatment groups to control group. *p < 0.05; ^%p < 0.001; ^$p < 0.01; ^#p < 0.0001 vs. control. Gel images of the product of XBP-1 and its spliced product after the semi-quantitative PCR in (D) HepG2 cells and (E) Hep3B cells.

mRNA expression of IRE-1α arm of UPR pathway

Impact on IRE-1 expression after ethanol exposure in HepG2, Hep3B and Huh-7 cells

IRE-1 mRNA expression showed a notable rise (3.6-fold, *p < 0.05) in the 100mM ethanol-treated HepG2 cells, whereas no significant change was seen in the 50mM and 150mM treated HepG2 cells (Fig. 2A). There was no significant alteration in IRE-1 expression in Hep3B cells following the treatment with varying concentrations of ethanol (Fig. 2B). In contrast to this, a concentration-dependent increase was observed in the Huh-7 cells. An increasing trend was seen in the ethanol-treated Huh-7 cells as a significant increase (2-fold, *p < 0.05) was seen in both 100mM and 150mM ethanol-treated Huh-7 cells (Fig. 2C).

To check the effect of ethanol exposure on XBP-1 splicing as the downstream effect of IRE-1 activation

mRNA expression of XBP-1 revealed no significant change in HepG2 and Huh-7 cells (Fig. 2A and C). However, an increase of the tune of 2-fold (*p < 0.05) was observed in the Hep3B cells when they were treated with the 100mM and 150mM concentrations of ethanol in comparison to the control untreated cells (Fig. 2B). To further check the spliced product of XBP-1, i.e. XBP-1s (263 bp), semiquantitative RT-PCR was done, but no spliced product was observed on the 2.5% agarose gel in both HepG2 (Fig. 2D) and Hep3B cells (Fig. 2E).

mRNA expression of ATF-6 arm of UPR pathway

Regulation of ATF-6 expression by ethanol in HepG2, Hep3B, and Huh-7 cells

RT-qPCR data revealed no notable variation in the expression of ATF-6 in HepG2 cells when treated with three different concentrations of ethanol (Fig. 2A). Similarly, 50mM and 100mM ethanol-treated groups of Hep3B cells also showed no significant change, but the treatment of these cells with higher concentration i.e., 150mM led to an elevation (1.5-fold, *p < 0.05), in the expression of ATF-6 (Fig. 2B). In the case of Huh-7 cells, a significant decrease (2-fold, *p < 0.05) was seen in the 100mM and 150mM ethanol-treated cells, but no change was seen in the 50mM ethanol group (Fig. 2C).

mRNA expression PERK arm of UPR pathway

Ethanol-induced changes in PERK expression in HepG2, Hep3B, and Huh-7 cells

A significant surge in PERK expression was seen only in 150mM ethanol-treated HepG2 cells (1.8- fold, ^%p < 0.001) (Fig. 2A) and Hep3B cells (5-fold, *p < 0.05) (Fig. 2B). 50mM and 100mM ethanol-treated group did not show any notable change in comparison to the control cells. Huh-7 cells depicted a significant elevation of 3-fold (^$p < 0.01) in the 100mM group and 8-fold upregulation (^#p < 0.0001) in the 150mM ethanol-treated group (Fig. 2C).

To check the effect of PERK activation on the downstream gene ATF-4 after chronic ethanol exposure

ATF4, which is a downstream gene of PERK, unveiled a substantial upregulation in a concentration-dependent manner i.e., a 2-fold increase (^$p < 0.01), observed in the 100mM group, was further increased up to 4-fold (^#p < 0.0001) in the HepG2 cells when treated with 150mM ethanol (Fig. 2A). In addition to this, Hep3B cells depicted a surge in the expression of ATF4 as evident by the 3-fold (*p < 0.05) and 9-fold (^#p < 0.0001) increase of the expression in 100mM and 150mM ethanol-treated group, respectively (Fig. 2B).

However, no discernible alteration was seen in the 50mM group in both the cell lines. Concordantly, Huh-7 cells also showed a significant rise in 100mM (6.6-fold, ^%p < 0.001) and 150mM group (7.3-fold, ^%p < 0.001) when compared to the control cells (Fig. 2C).

Expression profile of LAMP-3 in response to ethanol treatment in HepG2, Hep3B, and Huh-7 cells

RT-qPCR of LAMP-3 revealed that it was significantly elevated in the 100mM (7-fold, and 150mM (27-fold, ^#p < 0.0001) ethanol-treated group in comparison to the control group of HepG2 cells (Fig. 2A). LAMP-3 expression of Hep3B cells depicted a concentration-dependent increasing trend, which achieved significance in the 150mM group (10.5-fold, ^%p < 0.001) (Fig. 2B). Conversely, LAMP-3 expression was significantly declined upon the treatment of Huh-7 cells with 100mM (1.6-fold, ^$p < 0.01) and 150mM (1.4-fold, *p < 0.05) ethanol (Fig. 2C).

Ethanol-induced modulation of CHOP expression in HepG2, Hep3B, and Huh-7 cells.

A substantial increase was seen in the relative mRNA expression of CHOP at 50mM (1.6-fold, ^$p < 0.01) and 100mM (1.5-fold, *p < 0.05) ethanol-treated group of HepG2 cells, which was further intensified on increasing the concentration to 150mM (3.2-fold, #p < 0.0001) (Fig. 2A). In the case of Hep3B cells, an increase in the expression was seen in only the 150mM group (5-fold, *p < 0.05 (Fig. 2B). In comparison to both HepG2 and Hep3B cells, treatment of ethanol did not cause any alteration of CHOP expression in Huh-7 cells (Fig. 2C).

Impact of ethanol treatment on LC-3 expression in HepG2, Hep3B, and Huh-7 cells

As depicted from the results of the RT-qPCR, mRNA expression of LC-3 (which is known to have its role in autophagy) indicated no significant alteration in the expression of HepG2 and Huh-7 cells after the treatment with the different concentrations of ethanol (Fig. 2A and C). However, Hep3B cells showed an increased expression of LC-3 but only at 100mM concentration (2-fold, *p < 0.05) of ethanol (Fig. 2B).

mRNA expression of genes involved in triglyceride and cholesterol synthesis

SREBP-2 caused increased cholesterol synthesis in HepG2, Hep3B and Huh-7 cells

mRNA expression unveiled a notable surge in the levels of SREBP-2 (responsible for cholesterol synthesis) in all three cell lines viz. HepG2 (2.5-fold, ^%p < 0.001) (Fig. 2A), Hep3B (10-fold, ^$p < 0.01) (Fig. 2B), and Huh-7 (15-fold, ^#p < 0.0001) (Fig. 2C) when treated with the highest concentration of ethanol, i.e. 150mM whereas no significant change was seen upon treatment with 50mM and 100mM ethanol concentrations. Moreover, increased extracellular and intracellular cholesterol levels observed in all cell lines further bolster these results (Fig. 1E).

SREBP-1c increased the TG synthesis in all the three cell lines viz. HepG2, Hep3B and Huh-7 cells

RT-qPCR demonstrated a substantial elevation in the SREBP-1c levels, with the noteworthy 4-fold (^$p < 0.01) and 7.5-fold increase (^#p < 0.0001) in 100mM and 150mM ethanol-treated HepG2 cells, respectively (Fig. 2A). Huh-7 cells also showed a similar kind of trend in the expression levels of SREBP-1c with a significant 4-fold (^$p < 0.01) increase in the 100mM group and a 9-fold (^#p < 0.0001) surge in the 150mM group (Fig. 2C). However, significant alteration was observed only in the 150mM group of the Hep3B cells with an increase of 81-fold (^#p < 0.0001) (Fig. 2B). Conversely, 100mM and 50mM ethanol-treated Hep3B cells did not show any change when compared to the untreated cells. Collectively, this demonstrates that there is a concentration-dependent elevation of SREBP-1c gene, upon ethanol treatment, which was further corroborated by the elevated intracellular and extracellular triglycerides levels in all three cell lines (Fig. 1E).

Effect of ethanol exposure on VEGF-A and NF-κB

VEGF-A elevated the angiogenesis, as the crucial feature of HCC after the ethanol exposure in cell lines

Results from the RT-qPCR indicated a significant rise in the VEGF-A expression (well-known role in the angiogenesis) in all three cell lines i.e. HepG2 cells (2.5 fold, ^$p < 0.01) (Fig. 2A), Hep3B cells (1.6-fold, ^$p < 0.01) (Fig. 2B), and Huh-7 cells (7.6-fold, ^#p < 0.0001) (Fig. 2C) at 150mM concentration of ethanol. However, the other two concentrations did not prove to cause any significant change in its expression in any of these cell lines.

Ethanol-mediated modulation of NF-κB expression and LDH levels in HepG2, Hep3B, and Huh-7 cells

Analysis of mRNA expression unveiled substantial upregulation in NF-κB levels, with a significant 14-fold increase (^#p < 0.0001) observed in the group treated with 100mM ethanol in HepG2 cells. Notably, this elevation intensified in the cells when treated with 150mM ethanol, demonstrating an even more pronounced 48-fold increase (^#p < 0.0001). In contrast, treatment with 50mM ethanol concentration did not result in significant alteration in NF-κB expression compared to control cells. These findings emphasize the concentration-dependent influence of ethanol on the NF-κB in HepG2 cells (Fig. 2A). A remarkable increase was observed in Hep3B cells treated with 150mM ethanol, showing a 37-fold increase (^$p < 0.01) (Fig. 2B). In contrast to the HepG2 and Hep3B cells, Huh-7 cells showed a significant decline in the expression of the NF-κB gene in a concentration-dependent manner. A decrease of 2-fold (^$p < 0.01) was seen in the 50mM ethanol-treated Huh-7 cells, which was further enhanced on treating the cells with higher concentrations of ethanol i.e. 100mM (10-fold, ^#p < 0.0001) and 150mM (33-fold, ^#p < 0.0001) (Fig. 2C). Furthermore, LDH, the inflammatory marker, showed a concentration-dependent increase in both the intracellular and extracellular levels, supporting the activation of NF-κB after ethanol treatment in all the three cell lines (Fig. 1E).

Collectively, there is an upregulation of PERK-ATF4-LAMP3 or CHOP arm of the UPR pathway in HepG2 and Hep3B cell lines along with the increase in expression of genes involved in cholesterol and TG synthesis, which is in accordance with the higher levels of extracellular and intracellular TG and cholesterol obtained due to the ethanol treatment whereas Huh-7 did not show any regular trends in the gene expression (Fig. 1E). Based upon the systematic trend in fold change and elevation of gene expression of PERK arm in HepG2 cells and Hep3B cells, further experiments were carried out on HepG2 and Hep3B cells with the higher concentrations of ethanol, i.e., 100mM and 150mM.

Effect of chronic ethanol exposure on protein expression of UPR pathway

To further validate the activation of PERK-ATF4-LAMP3 or CHOP arm of the UPR pathway, protein expression of the upregulated genes was measured using western blot or immunofluorescence staining.

Western blot analysis indicated a significant concentration-dependent increase in the GRP-78 expression (p < 0.05 at 150mM) in Hep3B cells (Fig. 3B and D), whereas HepG2 did not show any significant change in the GRP-78 protein expression (Fig. 3A and C). All the proteins involved in the PERK arm of the UPR pathway, viz. p-PERK, p-eIF2α, ATF4, CHOP, and LAMP-3 unveiled a substantial rise in their expression in a concentration-dependent manner in HepG2 cells while a significant increase (*p < 0.05) was evident at the highest concentration of ethanol exposure i.e., 150mM (Fig. 3A and C). Regarding the ATF-6 arm of the UPR pathway, ATF-6 did not show any significant change upon ethanol treatment at different concentrations (Fig. 3A and C). In case of Hep3B cells, an unexpected outcome was seen in the p-PERK protein i.e., a significant reduction (*p < 0.05) was observed at 100mM concentration, whereas no significant change was observed at 150mM (Fig. 3B and D). However, the downstream genes of the PERK pathway viz. ATF4 and CHOP unveiled a substantial elevation (*p < 0.05) in their protein expression. Additionally, LAMP-3 also showed an increase in the 150mM ethanol-treated group; however, it was not significant (Fig. 3B and D). Notably, ethanol exposure in Hep3B cells did not prove to cause any significant change in the protein expression of p-eIF2α and ATF-6.

Fig. 3.

Protein expression of dysregulated genes of the UPR pathway. Western blot analysis of p-eIF-2α, t-eIF-2α, LAMP-3, GRP-78, ATF-6, ATF-4, CHOP, p-PERK, and PERK protein in (A) HepG2 cells and (B) Hep3B cells. Quantification of the western blot images done using Image J software in (C) HepG2 cells and (D) Hep3B cells. Data (n = 3) is presented as mean±SEM. Statistical significance was evaluated using one-way ANOVA and for comparison between experimental group and control group, Dunn’s multiple comparison test was done. *p < 0.05 vs. control. Representative images of the (E) HepG2 cells and (F) Hep3B cells when co-stained with phalloidin (green), Alexa flour 594 (red), and DAPI (blue). Quantification of nucleocytoplasmic ratio was done using FIJI software (G). Data (n = 3) is presented as mean±SEM. Two-way ANOVA was done to evaluate the statistical significance, which was followed by Sidak’s multiple comparison test to compare the experimental group with the control group. ****p < 0.0001 vs. control.

Moreover, co-staining experiments with markers for specific cellular compartments i.e., DAPI for the nucleus and phalloidin for actin filaments with the specific antibody against NF-κβ revealed the predominantly cytoplasmic localized pattern of NF-κB in the untreated HepG2 (Fig. 3E) and Hep3B (Fig. 3F) cells. Following the ethanol treatment, there was a noticeable shift in the localization, resulting in the nuclear translocation of NF-κB. Fluorescence intensity quantification exhibited a significant rise in expression of NF-κB in the 100mM (****p < 0.0001; Pearson’s coefficient = 0.42) and 150mM (****p < 0.0001; Pearson’s coefficient = 0.56) ethanol-treated HepG2 cells in comparison to control (Pearson’s coefficient = 0.27) (Fig. 3G). Ethanol-treated Hep3B cells also showed similar results at 150mM ethanol concentration (****p < 0.0001, Pearson’s coefficient = 0.50), whereas no significant results were found at 100mM concentration when compared to control (Pearson’s coefficient = 0.36) (Fig. 3G).

The treatment with a 150mM concentration of ethanol showed significant changes in the expression of the studied genes and proteins of the PERK arm of the UPR pathway in comparison to the untreated control. Therefore, further experiments were carried out only at this concentration (150 mM ethanol).

Effect of GSK2606414 (PERK inhibitor) treatment on the PERK-ATF4 arm of the UPR in ethanol-treated cells

To check the effect of GSK2606414 treatment on the protein expression of eIF-2α using western blot, HepG2 and Hep3B cells were exposed with 150mM ethanol and 2.5µM GSK2606414 (represented in Fig. 4A and B)¹¹. Treatment of PERK inhibitor led to slower growth, and cells showed disintegrated nuclei (Fig. 4A). Usage of GSK2606414 in ethanol-treated HepG2 cells significantly reduced (^$$p < 0.01) the levels of phosphorylated eIF-2α which was elevated upon ethanol- treatment (**p < 0.05) (Fig. 4C and D).

Fig. 4.

Effect of usage of PERK inhibitor on PERK arm of UPR pathway. Representative phase-contrast images of the HepG2 and Hep3B cells treated with the 2.5µM concentration of PERK inhibitor under a 20X microscope (A). Western blot analysis of p-eIF-2α, t-eIF-2α, ATF-4, and LAMP-3 after GSK2606414 treatment in HepG2 and Hep3B cells (B). Image J quantification of protein expression in HepG2 cells (C) and Hep3B cells (D). Data (n = 3) is presented as mean±SEM. Two-way ANOVA was employed to evaluate the statistical significance, followed by Sidak’s multiple comparison test to compare the experimental group with the control group. *p < 0.05; **p < 0.01; ****p < 0.0001 vs. control, ^$p < 0.05; ^$$p < 0.01; p < 0.0001 vs. 150mM ethanol-treated group. Scale bar = 200 μm.

Fig. 8.

Effect of ethanol and GSK treatment on stemness of HepG2 and Hep3B cells.

Representation of the number of spheres formed in different groups of HepG2 cells under 4X and 10X magnification (A) and quantification of at least 9 fields in each well which were further in triplicates (n = 3) (B). No. of spheres in each field of Hep3B cells under 4X and 10X magnification of light microscope (C) and spheres quantification formed in 9 fields of each well (D). Statistical analysis was performed using two-way ANOVA followed by Tukey’s post-hoc test for multiple comparisons. Results were deemed significant at *p < 0.05; ****p < 0.0001 vs. control, p < 0.0001 vs. 150mM ethanol-treated group. Scale bar = 1000 μm.

In line with our hypothesis, 2.5µM PERK inhibitor diminished the ethanol-induced elevation in the protein levels of ATF-4 and LAMP-3 (p < 0.0001) in HepG2 cells (Fig. 4C and D). A successful inhibition of phosphorylated eIF-2α (^$$p < 0.01) and its downstream protein viz. ATF-4 (^$p < 0.05) and LAMP-3 (not significant) were also seen due to the usage of PERK inhibitor in ethanol-treated Hep3B cells (Fig. 4C and E). Hence, usage of 2.5µM PERK inhibitor was successful in inhibiting the phosphorylation of eIF-2α protein, consequently inhibiting the protein expression of its downstream genes i.e., ATF-4 and LAMP-3 (Fig. 4C and E).

Impact of chronic ethanol exposure on ROS production

To measure the production of reactive oxygen species, DCFH-DA dye was used which gives a green fluorescent signal. An increased production of ROS was seen upon the 150mM ethanol treatment in HepG2 cells, as evident in the images from the confocal microscope (Fig. 5A). Additionally, quantification using flow cytometry also represented that there was an apparent increase up to 50% from 20% in the DCFH-DA + cells upon ethanol treatment.

Fig. 5.

Effect of ethanol treatment and inhibition of PERK arm on ROS production. Representative confocal microscope images of cells stained with DCFH-DA dye and grayscale image. The merged image shows the DCFH-DA positive cells in HepG2 cells (A) and Hep3B cells (C). Representation of the flow cytometry analysis of the number of cells that were positive with the DCFH-DA dye in HepG2 cells (B) and Hep3B cells (D). Quantification of % of DCFH-DA positive cells in comparison to control in HepG2 cells (E) and Hep3B cells (F). Data (n = 3) is presented as mean±SEM. Statistical significance was evaluated using one-way ANOVA, and experimental groups were compared to the control group using Tukey’s multiple comparison test. ***p < 0.001; ****p < 0.0001 vs. control, ^$$$p < 0.001; p < 0.0001 vs. 150mM ethanol-treated group.

(p < 0.0001). This increased production of ROS was relieved with the usage of PERK inhibitor to almost 20% (p < 0.0001) in comparison to the only ethanol-treated group (Fig. 5B and E).

In concordant with these results, similar results were obtained in the Hep3B cells where treatment with the 150mM concentration of ethanol almost doubled the percentage of DCFH-DA + cells (p < 0.001) (Fig. 5C). Upon GSK treatment, the percentage of DCFH-DA + cells (p < 0.001) were significantly reduced when compared to the ethanol-treated group (Fig. 5D and F).

Determination of effect on different cancer hallmarks after ethanol exposure and PERK inhibition

Further, to investigate the effect of ethanol treatment and PERK inhibition on different hallmarks of cancer, cell invasion (transwell invasion), cell migration (scratch assay), stemness (sphere formation), and apoptosis (annexin/PI) assays were performed.

Effect of chronic ethanol exposure and GSK2606414 on the invasion of HepG2 and Hep3B cells

Upon staining the HepG2 cells with crystal violet, it was observed that the 150mM ethanol-.

treated group exhibited a significant elevation in the cell count when compared to the control group. Specifically, there was an increase of 200% (p < 0.0001) in the invaded cells, which was reversed with the usage of PERK inhibitor to the 5% (p < 0.0001 vs. control and ethanol-treated) (Fig. 6A and B).

Fig. 6.

Effect of ethanol exposure and PERK inhibitor usage on invasiveness of HCC cells. Illustrative images of the invasive HepG2 cells after staining with crystal violet and visualised under a light microscope under 20X magnification (A) and quantification of the invaded cells (B). Representation of invasive Hep3B cells under 20X magnification (C) and quantification of these invaded Hep3B cells (D). Significance was determined using one-way ANOVA followed with Tukey’s multiple comparison test. **p < 0.01; ****p < 0.0001 vs. control, p < 0.0001 vs. 150mM ethanol-treated group. Scale bar = 400 μm.

Hep3B cells also showed similar results as evident by the increase of 200% in the percentage of invaded cells (p < 0.0001). Moreover, a significant reduction was seen with the use of PERK inhibitor to 70% (p < 0.01vs control, p < 0.0001 vs. ethanol-treated) (Fig. 6C and D). However, normal cells did not show any significant change in the invasive properties upon usage of PERK inhibitor in both the cell lines.Altogether, this suggests an increase in the invasive capacity of both HepG2 and Hep3B cells due to the ethanol-treatment, was curtailed by the usage of GSK2606414 as an inhibitor to the PERK arm of the UPR pathway.

Impact of prolonged ethanol exposure and PERK inhibition on the migrating capacity of HepG2 and Hep3B cells

Moreover, measurement of the migrating capacity of HepG2 and Hep3B cells using scratch assay revealed a substantial rise in the migrating capacity upon the ethanol treatment could be reversed by GSK2606414 successfully. Wound closure time for the untreated group was 72 h and 36 h in HepG2 and Hep3B cells untreated group, respectively which was increased to 48 h (p < 0.0001 vs. control, p < 0.0001 vs. ethanol-treated group) and 36 h (p < 0.0001 vs. control, p < 0.0001 vs. ethanol-treated group) in 150mM ethanol-treated groups of HepG2 (Fig. 7A) and Hep3B cells (Fig. 7C), respectively. In addition to this, wound closure took longer time in both the cell lines when the PERK arm was inhibited using an inhibitor, i.e., more than 72 h and 36 h in HepG2 (Fig. 7A and B) and Hep3B cells (Fig. 7C and D) respectively. In line with the results obtained in the invasion assay, migratory capacity also remains unaffected upon the PERK inhibitor treatment in normal HepG2 and Hep3B depicting the activation of PERK pathway to be ethanol-dependent.

Fig. 7.

Effect of ethanol and GSK treatment on migratory property of HCC cells. In-vitro scratch assay, phase contrast images under 20X magnification at 0 h, 12 h, 36 h, 48 h, 72 h (A), and % migration for different treatment groups in HepG2 cells (B). Phase contrast representative images under 20X magnification at 0 h, 12 h, 24 h, 36 h, and 48 h (C) and %migration determined by the wound closure in Hep3B cells (D). Ten different measurements from each field were measured, and three fields from each well were imaged. Data (n = 3) is presented as mean±SEM. Two-way ANOVA and Tukey’s multiple comparison tests were employed to determine statistical significance among various groups. ****p < 0.0001 vs. control, p < 0.0001 vs. 150mM ethanol-treated group. Scale bar = 200 μm.

Investigation of cancer stemness after ethanol treatment and usage of GSK2606414

Cancer stem cells are known for their capacity to proliferate even under adverse conditions. To assess the number of stem cells, sphere formation assay was conducted using agarose-coated plates to minimize the cell adhesion to the surface. In line with previous findings, a remarkable increase was found in the number of spheres to 900 and 300 in 150mM ethanol-treated HepG2 (p < 0.0001) and Hep3B cells (p < 0.0001) respectively (Fig. 8B and D). Much to our astonishment, no spheres were formed in both HepG2 (Fig. 8A) and Hep3B cells (Fig. 8C) when PERK arm was inhibited using GSK2606414, the observation that further enhanced the robustness of our findings whereas the PERK inhibitor did not show any independent effects, as there was no significant change in the no. of spheres in control group treated with inhibitor.

Examination of effects of chronic ethanol treatment alongside GSK2606414 treatment on the apoptotic behavior of HepG2 and Hep3B cells

Further, to corroborate our findings, an annexin/PI assay was done to check the effect of chronic ethanol exposure on apoptotis of the cells. Results from the flow cytometry experiments showed that there was a significant decline in the number of early apoptotic (by 10 times, p < 0.0001) and late apoptotic cells (30 times, p < 0.01) upon ethanol exposure to HepG2 cells, which was increased upon GSK treatment (Fig. 9A and B).

Fig. 9.

Effect of GSK treatment on apoptosis. Representative images of flow cytometry analysis for viable, early apoptotic, late apoptotic, and necrotic cells in HepG2 cells (A) and Hep3B cells (C) when stained with annexin V and PI dyes. Quantification of the number of cells in different quadrants found in flow cytometry in HepG2 cells (B) and Hep3B cells (D). To determine the statistical significance among various treated groups, 2-way ANOVA was followed by Tukey’s multiple comparison test. *p < 0.05; **p < 0.01; ****p < 0.0001 vs. control group. ^$$p < 0.01; p < 0.0001 vs. 150mM ethanol-treated group.

Apart from this, Hep3B cells also showed a 10% decline in the early apoptotic cells upon 150mM ethanol treatment (p < 0.05). In addition to this, a decrease (p < 0.0001) was also observed in the late apoptotic cell number, which was relieved upon the GSK treatment in 150mM treated group (p < 0.0001) in comparison to the ethanol treatment alone (Fig. 9C and D). Also, no change was seen in the apoptotic behavior of the cells upon GSK treatment in the control group.

Discussion

Tumor cells typically grow in a hypoxic and oxidative conditions, which are associated with sustained ER stress. However, it remains unclear how tumor cells manage to evade apoptosis induced by ER stress and persist in challenging environments. Moreover, prolonged and excessive alcohol consumption triggers various pathological stress responses, including ER stress, but there remains a notable gap in the understanding of molecular mechanisms underlying the ER stress-induced aggressiveness in the case of A-HCC.

In this study, we identified the induction of the PERK arm of the UPR pathway under chronic ethanol exposure, which shields the tumor cells from the apoptosis triggered by ER stress, thus promoting more aggressive hepatocellular carcinoma. The involvement of the PERK arm was further confirmed by observations that the inhibition of PERK alleviated the ethanol-induced aggressiveness in HCC cells, suggesting that targeting the PERK arm could hold promise as a therapeutic strategy for A-HCC.

In order to achieve this, HepG2, Hep3B and Huh-7 cell lines were exposed to different concentrations of ethanol viz. 50mM, 100mM and 150mM based upon previous literature¹⁶. Our results revealed that both HepG2 and Hep3B cells, upon chronic ethanol exposure, exhibited significant elevation in the ER stress sensor: GRP78, along with a significant rise in the expression of PERK, ATF4, LAMP-3, CHOP, VEGF-A, and NF-κB notably at higher concentrations. These findings were consistent with the previous study by Magne et al., which reported the activation of ATF4/CHOP as a result of ethanol exposure¹⁷. Also, the activation of NF-κB due to chronic ethanol exposure was in line with the previous literature that reports the NF-κB pathway as the mechanism for the HCC progression and aggressiveness¹⁸.

However, no significant change was seen in the expression of IRE-1α, XBP-1, ATF-6, and LC-3B genes, depicting the activation of only the PERK arm in A-HCC, which is also supported by the no change in the XBP-1 splicing. An earlier study, however, showed the activation of ATF-6 as a necessary and self-sufficient factor for the alcoholic fatty liver in zebrafish¹⁹.

Apart from this, SREBP-1c and SREBP-2 were also elevated upon chronic ethanol exposure which is supported by the increased TG and cholesterol levels in a concentration-dependent manner. You et al., also found the role of activation of SREBP in the fatty acid synthesis induced by ethanol in rats²⁰. In the case of Huh-7 cells, the activation of the genes of the PERK arm was not consistent upon ethanol exposure. Such differential effect of ethanol on Huh-7 and other cell lines can be attributed to the genetic, metabolic and biochemical differences in the cell lines due to their inherent characteristics of different origin.

To further validate this data, protein expression of the genes involved in the PERK arm of the UPR pathway was measured in both HepG2 and Hep3B cells. The results depicted the significant elevation in the downstream proteins of the PERK arm of the UPR pathway, i.e., ATF-4, CHOP, and LAMP-3 in HepG2 cells, along with the augmentation in the phosphorylation of PERK and eIF2-α protein. These observations depicting the activation of PERK-ATF4-CHOP protein levels were consistent with the relevant findings that demonstrated the activation of this pathway for the HCC progression²¹. LAMP-3 has already been recognized as the regulator of hepatic lipid metabolism in the PI3K/AKT pathway, but our study is the first to reveal the dependency of LAMP3 on the UPR pathway in A-HCC²². In addition to this, NF-κB was also translocated to the nucleus from cytoplasm upon ethanol exposure. The role of NF-κB in HCC progression has been well explored in the previous literature²³. Similar observations were noted in Hep3B cells, albeit with no significant alteration in the phosphorylation of PERK protein, which posed an intriguing finding which might be due to the relatively high basal expression of PERK or other stress mechanisms could be compensating for the lack of increased PERK phosphorylation. While the previous literature has highlighted the role of PERK-ATF4-CHOP pathway in the aggressiveness or the progression of HCC, our study pioneers the investigation of its involvement in A-HCC, an area in which molecular mechanisms have been minimally explored till now. Discussing a little about the clinical setting, for the chronic alcoholics, a blood alcohol concentration of 300 mg/dL (65mM) is considered the intoxicated level. However, such high concentrations as 150mM cannot be achieved in plasma as it is toxic. In the case of in vitro experiments, high concentrations of ethanol are required because of its volatile nature and evaporates rapidly. In addition, in vitro systems lack the physiological metabolism, antioxidant defence, and clearance mechanisms present in vivo, which further necessitate higher concentrations to achieve comparable stress responses.

Furthermore, to check the effect of the PERK arm on cancer hallmarks and ROS production, HepG2 and Hep3B cells exposed to 150mM of ethanol were treated with 2.5µM concentration of GSK2606414 based upon our lab findings¹¹. Upon its usage, there was a decrease in the phosphorylation of eIF-2α along with a decrease in the protein expression of ATF-4 and LAMP-3 in both the cell lines, depicting the successful inhibition of this pathway. In alignment with previous research, an increase in the ROS production in both HepG2 and Hep3B cells was seen upon the ethanol exposure, which was relieved with the usage of PERK inhibitor²⁴. Moreover, the aggressiveness of HCC cells due to ethanol exposure was marked by the increase in migration capacity, invasiveness, and stemness. In this context, a study reported the activation of EMT as well in the ethanol-induced stemness and metastasis in HCC²⁵. To further corroborate these findings, apoptosis was measured using Annexin/PI staining, which showed a noteworthy decrease in the number of early apoptotic and late apoptotic cells in both HepG2 and Hep3B cells when exposed to 150mM ethanol concentration. There are contradictory reports in the literature in context to the effect of ethanol exposure on apoptosis. These findings were in line with the literature findings that reported a decrease in the pro-apoptotic protein and an elevation in the anti-apoptotic protein in pig liver tissue, whereas another study showed the activation of pro-apoptotic genes Bcl-2 with the ethanol exposure^26,27. We also found the activation of CHOP with no significant change in LC-3 expression with the ethanol treatment in both cell lines but apoptosis was inhibited under these conditions. This could be due to the fact that the activation of the UPR pathway is totally duration-dependent and PERK-eIF2α plays a crucial role in switching signaling from pro-survival to the pro-death type.

The aggressiveness of the HCC cells was alleviated with the usage of GSK2606414 as evident by the decrease in the invasive and migratory capacity of ethanol treated-HCC cells. Interestingly, no cancer stem cells were formed in the ethanol-treated HepG2 and Hep3B cells upon treatment with the PERK inhibitor, which reiterated the involvement of the PERK arm of the UPR pathway causing the increased stemness in A-HCC. However, GSK2606414 did not show any independent effects as the control group treated with PERK inhibitor did not show any significant change in all the cancer hallmarks which was also consistent with our previous findings¹¹. Also, an increase was observed in the number of early and late apoptotic cells when the PERK inhibitor was used as a rescue to ER stress caused by ethanol exposure to HCC cells, which was similar to the results by Rozpedek et al. which showed the inhibition of eIF-2 using GSK2606414 human neuroblastoma and human colorectal adenocarcinoma cells lines as a switch from the pro-adaptive state to the pro-apoptotic branch and concluding it as the promising treatment for these cancer types²⁸. It is important to highlight that longer exposure of ethanol e.g., for 2–4 weeks will lead to PERK desensitization or attenuation of the PERK pathway’s initial protective response, shifting the cellular balance towards cell death, which is not explored this study. In this study, we aimed to establish a mechanistic understanding of the role of the PERK/ATF4/LAMP3 pathway in hepatocellular carcinoma (HCC) progression under chronic ethanol exposure. Cell lines were chosen as an initial model system due to their controlled and reproducible nature, which allows us to dissect molecular pathways with high precision. Although ethanol needs to be metabolised into acetaldehyde and acetate in our body, which has stronger carcinogenic effects but all of these cell lines exhibit very low ADH/ALDH activity, depicting that most of the effects are produced from the chronic ethanol exposure made directly to the cell lines. In addition, reproducing physiologically relevant acetaldehyde levels in vitro is technically challenging, often leads to rapid volatilization and instability of acetaldehyde, making controlled and reproducible exposure difficult. To enhance the robustness of the results obtained, we employed multiple cell lines representative of different HCC subtypes and confirmed the reproducibility of our findings across these models. While our in vitro system provides valuable mechanistic insights into ethanol-induced cellular stress responses under controlled conditions, it does not capture the complex physiological metabolism, immune microenvironment, and multicellular interactions that characterize chronic alcoholic liver disease in vivo. Furthermore, dynamic systemic factors, such as immune regulation, hormonal signaling, and tissue-level remodeling present in patients, cannot be fully reproduced using cell-based models. However, it is important to mention the lack of further validation of activation of the PERK pathway in in vivo models of HCC is a limitation in order to be conclusive. Although the activation of the PERK pathway has been observed in the HepG2 and Hep3B cells but HepG2 is the hepatoma cell line which prompts us to validate our findings in other liver cancer cell line models like PLC/PRF/5 or SK-Hep-1 in the future studies.

Taken together, these observations suggest the induction of the PERK-ATF4 arm of the UPR pathway leads to increased expression of CHOP and LAMP-3 protein expression in association with an increase in the aggressiveness of HCC under ethanol exposure. This underscores LAMP-3’s potential as a pro-metastatic gene in A-HCC. Additionally, UPR- dependent activation of VEGF-A and NF-κB indicates their role in A-HCC. Furthermore, activation of TG synthesis and cholesterol synthesis due to the activation of SREBP-1c and SREBP-2, respectively, strengthens our assertion regarding the involvement of PERK arm of UPR pathway in A-HCC aggressiveness. Importantly, inhibiting this pathway effectively mitigated these effects across various cancer hallmarks, suggesting a promising therapeutic strategy for A-HCC. Notably, the reduction in cancer stem cells upon PERK ablation supports this hypothesis, although further validation in an in vivo model is warranted to confirm these findings.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

H.G.-conception and design, methodology, data acquisition, interpretation and analysis and original draft writing, J.K.-conception and design, study supervision, review and revision of manuscript.

Funding

The work was supported by Indian Council of Medical Research, New Delhi through a project grant. However, no funding was specifically provided for the publication.

Data availability

The data that support these findings has been submitted in the manuscript or shared as the part of the supplementary information.

Declarations

Competing interests

The authors declare no competing interests.

Ethical approval

Ethical committee approval was not required for this work as it did not involve human subjects or animal experiments. For transparency and reproducibility, cell line authentication including STR profiling and contamination screening was done.