RAB25 promotes hepatitis B virus-induced liver fibrosis progression through activation of the PI3K/AKT signaling pathway.

1
Introduction
Hepatitis B infection is a worldwide epidemic, with about 257 million people living with the hepatitis B virus (HBV). The disease is mostly prevalent in Asia and Africa (Asrani et al., 2019). Most hepatitis B infections eventually evolve into cirrhosis and liver cancer without clinical intervention, which has serious implications for human health. Hepatic fibrosis is an important stage in the progression of hepatitis B to cirrhosis. Studies have shown that the progression of early liver fibrosis can be stopped or reversed after eliminating the underlying cause (Guerrieri and Levrero, 2023). Treatment regimens such as nucleotide analogs or interferon are mainly used to inhibit viral replication and slow down viral damage to the liver rather than directly treating hepatic fibrosis (Li et al., 2019, Sun et al., 2020). Therefore, interventions for HBV-induced liver fibrosis can substantially improve prognosis.
Ras-related protein 25 (RAB25), a member of the Rab subfamily of GTPases, regulates cell membrane trafficking in virtually all cell lines and plays a coordinating role in regulating intracellular vesicular transport, thereby modulating a wide range of physiological processes such as immunity, hormone secretion and neurotransmission (Homma et al., 2021). In recent years, associations between GTPases and fibrotic diseases have been revealed in the literature. For instance, activating the small G protein Rnd3 effectively alleviated myocardial fibrosis and prevented cardiac remodeling in diabetic heart disease (Zhang et al., 2022), and knockdown of RAB6 significantly inhibited PM2.5-induced pulmonary fibrosis and oxidative stress in mice (Yang et al., 2020). Most current studies on RAB25 have focused on its pro-cancer effects in malignant tumors (Jeong et al., 2019, Temel et al., 2020). In addition, it has been reported in the literature that RAB25-dependent lipoautophagy is an important target for the prevention of hepatic fibrosis (Qiu et al., 2019); however, the function and mechanism of RAB25 in HBV-associated hepatic fibrosis remain unclear.
The PI3K/AKT signaling acts as a major pathway regulating cell growth, proliferation, metabolism, survival, and angiogenesis, exerting a crucial regulatory effect on fibrotic diseases. The common denominator of fibrotic diseases is the overproduction and deposition of extracellular matrix (ECM). Targeting the PI3K/AKT signaling was found to be an effective potential therapy to treat diseases such as idiopathic pulmonary fibrosis (Wang et al., 2022) and cardiac fibrosis (Qin et al., 2021). Several recent studies have reported the correlation between the PI3K/AKT pathway and hepatic fibrosis within HBV-infected individuals. For example, according to enrichment analysis, HBV can activate several signaling pathways, such as the PI3K/AKT signaling and JAK-STAT (Patil et al., 2023). However, HBV-induced activation of hepatic stellate cells (HSCs) can adjust the motility of HSCs through the PI3K/AKT pathway (Xie et al., 2021). In addition, inhibiting the PI3K/AKT signaling effectively alleviated the symptoms of HBV-associated glomerulonephritis (Guan et al., 2022). The PI3K/AKT pathway is an underlying target for HBV-associated hepatic fibrosis.
Herein, the differentially expressed genes associated with hepatic fibrosis were screened from GSE171294 and GSE84044 through bioinformatics. Combined with cellular experiments, it was determined that RAB25 expression was up-regulated in HSCs induced by HBV hepatocellular carcinoma cells, and RAB25 knockdown alleviated the activation of TGF-β1 and HSC induced by HBV hepatocellular carcinoma cells through the inhibition of the PI3K/AKT signaling. RAB25 has been proven to be an underlying target for the clinical treatment of HBV-associated liver fibrosis.

2
Material and methods
2.1
Bioinformatic analysis
The liver fibrosis-associated expression profiles were obtained from the GEO database (https://www.ncbi.nlm.nih.gov/geo/). GSE171294 contains the hepatic gene expression profiles of liver samples from 8 patients who underwent surgical resection for the treatment of HBV-associated liver cancer in the liver fibrosis group (n=4) or hemangioma in the control group (n=4) (Wu et al., 2021). GSE84044 contains hepatic gene expression profiles of 81 liver fibrosis differentiation stage 1, 2, 3, and 4 samples and 43 normal liver samples from 124 chronic hepatitis B patients (Wang et al., 2017). The R language limma package was used to perform differential gene analysis on this dataset with the screening parameters as |logFC| > 1, adjusted P.val < 0.05. The R language ggplot2 package was used to draw volcano plots and gene expression profiles.

2.2
Metascape pathway enrichment
Metascape (https://metascape.org/) is a website tool providing functional gene annotation (Lin et al., 2019). Liver fibrosis-associated potential genes within GSE171294 and GSE84044 were subjected to pathway enrichment with a P-value cutoff of 0.01, a min overlap of 3, and a min enrichment of 1.5.

2.3
Clinical sample
A total of 45 participants were enrolled in this study after approval by the Institutional Review Board of The First Affiliated Hospital of University of South China, and written informed consent was obtained from all subjects. The cohort comprised 15 healthy controls, 15 chronic hepatitis B (CHB) patients without evidence of liver fibrosis or cirrhosis, and 15 CHB patients with histologically confirmed liver fibrosis. Healthy controls had no history of liver disease, normal liver function tests, and negative serological markers for HBV infection. CHB patients without fibrosis were defined by persistent HBsAg positivity for at least six months and normal or mildly elevated liver enzymes with no clinical or histopathological signs of fibrosis. Patients in the fibrosis group had chronic HBV infection and liver fibrosis confirmed by percutaneous liver biopsy. Liver tissue specimens from the fibrosis group were obtained under ultrasound guidance, processed for histological evaluation, and staged according to established criteria. Serum samples were collected from all participants after overnight fasting and stored at −80°C for subsequent serum RAB25 expression analysis by ELISA (Human RAB25 ELISA Kit, SLD4057Hu, Sunlong Biotech, Hangzhou, China). Demographic and clinical data, including routine laboratory tests and non-invasive fibrosis indices such as the aspartate aminotransferase-to-platelet ratio index (APRI) and fibrosis-4 (FIB-4) index, were recorded at the time of sample collection to facilitate clinical correlation analysis.

2.4
Hepatic stellate cells activation model
To elucidate the relationship between HBV infection and hepatic fibrosis, an in vitro co-culture system of HBV-associated hepatocellular carcinoma cells HepG2.2.15 (CD0258, Shanghai Qidian Biotechnology Co., Ltd, Shanghai, China) and common hepatocellular carcinoma cells HepG2 (TChu 72, Chinese Academy of Sciences Cell Bank, Shanghai, China) with LX-2 cells (SCSP-527, Chinese Academy of Sciences Cell Bank) was used in this study.
Briefly, 1 × 105 HepG2.2.15 and HepG2 cells were inoculated within the top chamber of Transwell (3640, Corning Incorporated, Corning, USA), with the same LX-2 cell number in the bottom chamber. The cells were subjected to 24-h co-culture in DMEM and added with 10% FBS. LX-2 cells were then collected for subsequent experiments.

2.5
ELISA assay
The culture supernatants from HepG2.2.15 and HepG2 cells were collected by centrifugation at 2,500 × g for 5 minutes at room temperature. The TGF-β1 concentration in the supernatants was then measured using a commercial ELISA kit (CSB-E04725h, CUSABIO, Wuhan, China), as directed by the manufacturer.

2.6
Cell transfection and treatment
RAB25 siRNA and its negative control (si-NC) were purchased from Guangzhou Ribo Biotechnology Co. (Guangzhou, China). RAB25 overexpression vector was obtained by cloning the coding region sequence of RAB25 into pLVX vector. The empty pLVX vector was set as negative control (NC). Lipofectamine 2000 reagent (Invitrogen, Carlsbad, USA) was employed as directed to transfect LX-2 cells upon reaching 70-80% fusion in six-well plates. Following 48 h of transfection, LX-2 cells were induced with TGF-β1 (4 ng/mL) or subjected to 24-h co-culture using HepG2.2.15 and HepG2 cells before being harvested for subsequent experiments. Moover, to suppress PI3K/AKT signaling, LX-2 cells were treated with the selective PI3K inhibitor LY294002 (10 μM; HY-10108, MedChem Express, Monmouth Junction, USA) for 24 h (Choi et al., 2021). Cell groupings were named according to the respective transfected siRNA or vector. SiRNA sequences and overexpression vector construction primers are listed in Table S1.

2.7
CCK8 assay
After being made into single-cell suspension and counted, the transfected or co-cultured LX-2 cells were spread onto 96-well plates at a concentration of 5,000 cells/well in 100 μL of cell culture solution. Before the cell viability detection at each time point of 0, 24, 48, and 72 h, each well was supplemented with 10 μL CCK-8 solution and placed in the incubator for 2-h further incubation. An enzyme marker was used to measure the absorbance at 450 nm (Xue et al., 2023).

2.8
Colony formation experiments
According to the experimental plan, 200 transfected or co-cultured LX-2 cells were inoculated in 6-well culture plate, repeating three times for each group. The cells were incubated by placing the culture plate in a cell culture incubator. The media was replaced every three days for fourteen days until visible cell colonies were formed in the plates. After completion of the incubation, a pre-cooled 4% paraformaldehyde solution was added for 30-min fixation treatment. Subsequently, the formaldehyde solution was removed, and 10-min staining was performed using 0.1% crystal violet solution (C0121, Beyotime, Shanghai, China) at room temperature (RT). Following staining, the culture plates were rinsed using double-distilled water and then air-dried. Finally, photographs were taken and analyzed for cell colony counting (Chang et al., 2019).

2.9
Scratch test
A suitable number of LX-2 cells in favorable conditions were inoculated into the small wells of a six-well plate and spread evenly to guarantee that each well had the same cell number. When cell fusion reached 100%, a sterilized 200 μL tip perpendicular to the plane of the six-well plate was employed to make a smooth scratch in the middle of the well. After removing the scattered cells by washing with PBS, cells in the plate were added with serum-free media and incubated in a CO2 incubator (37°C, 5% CO2), imaged, marked, and recorded. At 0 h and 24 h, a microscope was utilized to observe, photograph, and record the scratches at the same position. The changes in the width of the recorded scratches were statistically analyzed (Li et al., 2022).

2.10
qRT-PCR
LX-2 cells were collected after transfection or co-culture, and TRIZOL (Invitrogen) was used to extract total RNA. A reverse transcription kit (TaKaRa, Tokyo, Japan) was used as directed to perform reverse transcription. A LightCycler 480 qRT-PCR instrument (Roche Diagnostics, Indianapolis, USA) was used to detect gene expression, and reaction conditions were conducted as per the operating protocols of the qRT-PCR kit (SYBR Green Mix, Roche Diagnostics). The PCR temperature cycling conditions were as follows: initial denaturation at 95°C for 5 min; 45 cycles of denaturation at 95°C for 10 s, annealing at 60°C for 10 s, and elongation at 72°C for 10s. The final cycle was followed by an extension at 72°C for 5 minutes. Each qRT-PCR reaction was carried out thrice. GAPDH was used for the normalization of the relative expression of target genes, which was calculated using the 2-ΔΔCt method, ΔΔCt=experimental group (Ct target gene-Ct internal reference) -control group (Ct target gene-Ct internal reference). The amplification primer sequences for each gene and its internal control are listed in Table S2.

2.11
HBV DNA detection
HBV DNA levels were quantified by qPCR to assess viral replication in cell cultures. Total DNA was extracted from HepG2.2.15 or HepG2 cells following co‑culture using a genomic DNA extraction kit (10503027, Thermo Fisher Scientific, Waltham, USA) according to the manufacturer’s instructions. Extracted DNA was subjected to qPCR amplification targeting a conserved region of the HBV genome with specific primers and a fluorescent probe, and reactions were performed on a real‑time PCR system using a SYBR Green Mix. A series of HBV DNA standards of known concentration were run in parallel to generate a standard curve for absolute quantification, and an internal control was included to monitor for potential PCR inhibition. The cycling conditions consisted of an initial denaturation at 95 °C for 10 min, followed by 40 cycles of denaturation at 95 °C for 15 s and annealing/extension at 60 °C for 60 s. Relative HBV DNA levels were calculated using the 2-ΔΔCt method by normalizing the HBV DNA Ct value in each sample to that of a reference control group and to an internal reference gene (GAPDH), thereby expressing data as fold changes relative to the control group.

2.12
Western blot
Total protein specimens were obtained by lysing LX-2 cells with RIPA lysis solution (Beyotime). The BCA kit (Beyotime) was used to measure the protein content. The corresponding protein volume was subsequently supplemented into the sampling buffer (Beyotime) and mixed thoroughly. A 5-min boiling water bath was carried out to denature the protein. The proteins were subjected to electrophoresis (80 V, 30 min) until the bromophenol blue was migrated into the separated gel; next, a higher voltage (120 V) was used for 1-2 h. The membrane transfer was performed in an ice bath, with a current of 220 mA for 120 min. Subsequently, after 1-2 min washing in the washing solution, the membrane was put into the sealing solution (5% non-fat milk in TBST solution) to seal for 60 min at RT. The membrane was incubated with primary antibodies (GAPDH (5174S, 1:1000, Cell Signaling Technology [CST], Boston, USA), RAB25 (13189-1-AP, 1:500, Proteintech, Wuhan, China), α-SMA (#19245, 1:1000, CST), Collagen I (#72026, 1:1000, CST), MMP2 (#4022, 1:1000, CST), PCNA (#13110, 1:1000, CST), p-PI3K (ab182651, 1:200, Abcam, Cambridge, USA), PI3K (ab302958, 1:1000, Abcam), p-AKT (#4060, 1:2000, CST), and PI3K (#9272, 1:1000, CST)) for a night at 4°C on shaker. On the following day, the membrane was rinsed thrice (10 min each) with TBST washing solution and then transferred into a secondary antibody (horseradish peroxidase-conjugated goat anti-rabbit IgG, 1:5000, Beijing Cwbio Co., Ltd., Beijing, China), followed by 1-h incubation at RT. The membrane was rinsed thrice (10 min each). A chemiluminescent imaging system (Bio-rad, Hercules, USA) was used to detect protein blots on the membrane.

2.13
Data analysis
The cell experiments were performed at least three times. All experimental data were expressed in terms of the mean ± standard deviation (SD). GraphPad Prism 8.0 software was applied to perform the t-test or one-way ANOVA followed by the Tukey method. The threshold for statistical significance was P < 0.05.

3
Result
3.1
Liver fibrosis-related genes were screened by bioinformatic analysis
First, HBV-associated liver fibrosis microarrays GSE84044 and GSE171294 were used to screen for liver fibrosis-related genes. 69 differentially expressed genes (DEGs) were obtained from GSE84044, and 629 DEGs were obtained from GSE171294 (Fig. 1A). A total of 32 common differentially expressed genes were obtained from the two microarrays (Fig. 1B). Metascape was applied to perform pathway enrichment analysis of 32 common differential genes. Based on Gene Ontology (GO) analysis, these DEGs contribute to regulating cell migratory ability, neutrophil chemotaxis, and chemokine-mediated signaling pathways. Based on the Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis, these DEGs contribute to regulating interactions with viral proteins with cytokines and cytokine receptors, chemokine signaling pathways, etc. (Fig. 1C).
The key to fibrosis formation is the abnormal activation, hyperproliferation, and migration of effector cells (Ezhilarasan, 2022). 32 differential genes were analyzed using the Metascape database to further investigate the genes associated with cell migration. The findings indicated that a total of five genes might contribute to the regulation of cell migration, including AQP1, CXCL10, CCL20, SOX9, and RAB25, which were all significantly up-regulated in liver fibrosis tissues (Fig. 1D).

3.2
RAB25 is upregulated in activated HSCs induced by HBV-associated hepatocellular carcinoma cells
The activation of hepatic stellate cells (HSCs) is considered a critical event within the pathological response to hepatic tissue injury (Ezhilarasan, 2022). TGF-β1 is a key pro-fibrotic cytokine for HSCs activation. Firstly, TGF-β1 levels in HepG2 and HepG2.2.15 cells were detected by ELISA assay. The TGF-β1 level was notably increased in HepG2.2.15 cells when compared to HepG2 cells (Fig. 2A). A series of experiments were conducted to further investigate whether HBV-expressing hepatocytes cause HSC activation. As shown by CCK8, colony formation, and scratch assays, HepG2.2.15 co-culture (HepG2.2.15+HSCs) significantly promoted the cell viability (Fig. 2B), the number of cell colonies (Fig. 2C), and the migratory ability (Fig. 2D) of HSCs compared with HepG2 co-culture (HepG2+HSCs). In addition, the expression levels of fibrotic markers (α-SMA and Collagen I), matrix remodeling factor MMP2, and proliferation-associated protein PCNA were also examined. Western blot results showed that HepG2.2.15 co-culture significantly promoted the expression levels of these proteins within HSCs compared to HepG2 co-cultured HSCs (Fig. 2E).
Moreover, the expression levels of AQP1, CXCL10, CCL20, SOX9, and RAB25 were also examined in HSCs after co-culture. qPCR results showed that HepG2.2.15 co-culture significantly up-regulated the mRNA levels of these five genes compared to HSCs co-cultured with HepG2, with the largest up-regulation of RAB25 (Fig. 2F). Similarly, RAB25 protein level was notably increased in HSCs co-cultured HepG2.2.15 when compared to HSCs co-cultured with HepG2 (Fig. 2G). Moreover, in the clinical cohort composed of 15 healthy controls, 15 CHB patients without fibrosis or cirrhosis, and 15 CHB patients with histologically confirmed liver fibrosis, serum RAB25 levels were significantly elevated in patients with liver fibrosis compared with both healthy controls and non-fibrotic CHB patients, and this trend paralleled increased in established non-invasive fibrosis indices APRI and FIB-4 (Fig. S1A-B). Correlation analysis revealed that serum RAB25 expression positively correlated with both APRI and FIB-4 scores (Fig. S1C), suggesting an association between RAB25 up-regulation and the severity of HBV-related liver fibrosis. Therefore, RAB25 was selected as the focus of the subsequent experiments.

3.3
Silencing RAB25 represses TGF-β1-induced activation of HSCs
Knockdown RAB25 was used to investigate the function of RAB25 in HSCs activation. Firstly, two siRNAs targeting RAB25 (si-RAB25#1 and si-RAB25#2) were constructed and transfected into LX-2 cells. After transfection for 48 h, the cells were collected, and qPCR and Western blot assay were performed to detect the knockdown efficiency of the siRNAs. RAB25 mRNA and protein expression were markedly reduced following siRNA treatment (Fig. 3A-B). Si-RAB25#1 showed better knockdown efficiency and was selected for subsequent experiments.
TGF-β1 is a key pro-fibrotic cytokine for HSCs activation. LX-2 was transfected using si-RAB25#1 and its control product. 48 h after transfection, 4 ng/ml TGF-β1 was added to continue to induce HSC activation for 24 h. LX-2 was subsequently collected, and cell viability, proliferation, and migration levels were detected, indicating that TGF-β1 addition significantly promoted HSCs viability (Fig. 3C), colony formation (Fig. 3D) and migration ability (Fig. 3E) than the Blank control. In addition, TGF-β1 also significantly promoted α-SMA, Collagen I, MMP2, and PCNA protein contents (Fig. 3F). Taken together, TGF-β1 induces the activation of HSCs. Compared with the TGF-β1+si-NC group, TGF-β1+si-RAB25#1 significantly relieves the above effects of TGF-β1 upon HSCs (Fig. 3C-F). These findings indicate that silencing RAB25 could inhibit TGF-β1-induced activation of HSCs.

3.4
Overexpression of RAB25 promotes TGF-β1-induced activation of HSCs
HSCs were transfected with either the RAB25 overexpression vector or the NC vector to further explore the effect of RAB25 overexpression on HSC activation. As demonstrated in Fig. 4A-B, RAB25 mRNA and protein expressions were significantly elevated in the RAB25-transfected group compared to the NC group, thereby confirming successful overexpression of RAB25 in HSCs. RAB25 overexpression further increased cell viability and proliferation in TGF-β1-treated HSCs (Fig. 4C-D). RAB25 overexpression promoted faster migration of TGF-β1-stimulated HSCs in the wound healing assay (Fig. 4E). Besides, RAB25 overexpression significantly elevated the protein levels of α-SMA, Collagen I, MMP2, and PCNA (Fig. 4F). These findings suggest that RAB25 plays a critical role in promoting HSC activation and fibrogenesis in response to TGF-β1 signaling.

3.5
Silencing RAB25 inhibits the activation of HSCs induced by HBV-related hepatocellular carcinoma cells
The function of RAB25 in HSC activation caused by HBV-related hepatocellular carcinoma cells was further investigated by establishing an HBV-infected liver fibrosis cell model. Firstly, si-RAB25#1 and its control product were transfected into LX-2 and cultured for 48 h. Next, the transfected LX-2 was subjected to 24-h co-culture with HepG2.2.15 and HepG2, respectively, and the LX-2 cells were harvested for the subsequent experiments. HepG2.2.15+HSCs significantly increased HSCs viability, proliferation, and migration levels compared with HepG2+HSCs (Fig. 5A-C). Moreover, the knockdown of RAB25 was found to significantly reverse the promoting effects of HBV on HSCs, including cell viability, colony formation, and migration ability, compared with the HepG2.2.15+HSCs+si-NC group (Fig. 5A-C). The knockdown RAB25 could also inhibit HBV-associated hepatocellular carcinoma cells induced promotion upon the expression levels of proteins, including α-SMA, Collagen I, MMP2, and PCNA within HSCs (Fig. 5D). These results suggest that silencing RAB25 was able to inhibit HBV-associated hepatocellular carcinoma cell-induced HSC activation.

3.6
Overexpression of RAB25 promotes the activation of HSCs induced by HBV-related hepatocellular carcinoma cells
The role of RAB25 overexpression on the activation of HSCs induced by HBV-related hepatocellular carcinoma cells was also investigated. HSCs co-cultured with HepG2.2.15 cells exhibited increased viability compared to those co-cultured with HepG2 cells, and RAB25 overexpression further enhanced this viability (Fig. 6A). Similarly, RAB25 overexpression boosted HSCs proliferation and migration in the presence of HepG2.2.15 cells (Fig. 6B-C). MoreoverRAB25 overexpression upregulated the α-SMA, Collagen I, MMP2, and PCNA protein levels in HSCs co-cultured with HepG2.2.15 cells (Fig. 6D). These findings suggest that the overexpression of RAB25 promotes HSC activation and proliferation in the context of HBV-related hepatocellular carcinoma cells.
Given that chronic HBV infection drives liver fibrosis primarily through sustained viral replication and hepatocyte–stellate cell interactions (Yao et al., 2020), hence, whether modulation of RAB25 expression in HSCs could influence HBV replication markers detectable in the co-culture system were explored (Fig. S2). qPCR results showed that HBV DNA was significantly higher in HepG2.2.15 cells than in HepG2 controls, consistent with active viral replication in the HepG2.2.15 model. Moreover, knockdown of RAB25 in HSCs led to a reduction in HBV DNA in HepG2.2.15 cells (Fig. S2A), whereas RAB25 overexpression in HSCs correlated with increased HBV DNA levels in HepG2.2.15 (Fig. S2B). These data suggest that alterations in RAB25 expression in HSCs can indirectly influence HBV replication markers in co-cultured hepatocellular carcinoma cells.

3.7
RAB25 promotes PI3K-AKT signaling activation in HBV-associated hepatocellular carcinoma cell-induced HSC activation
The PI3K-AKT pathway exerts a pivotal effect on HSC activation and enhances fibrosis within hepatic stellate cells (Xie et al., 2021). Moreover, the knockdown of RAB25 remarkably suppressed PI3K and AKT phosphorylation within glioblastoma cells (Ding et al., 2017). Therefore, it was tested whether the knockdown of RAB25 inhibits HSC activation induced by HBV-associated hepatocellular carcinoma cells was related to the PI3K-AKT pathway. HepG2.2.15 co-culture significantly promoted PI3K and AKT phosphorylation within HSCs compared to HepG2 co-culture. The RAB25 knockdown was able to inhibit the promotion of HepG2.2.15 on PI3K and AKT phosphorylation within HSCs (Fig. 7A). Furthermore, RAB25 overexpression in the HepG2.2.15 + HSC-RAB25 group further increased the PI3K and AKT phosphorylation when compared to the HepG2.2.15 + HSC-NC group (Fig. 7B). These results suggest that HBV-infected hepatocellular carcinoma cells may promote HSCs activation by enhancing PI3K-AKT phosphorylation within HSCs.
To determine whether the pro-activating effects of RAB25 on HSCs were mediated through the PI3K/AKT signaling pathway, the rescue experiments were performed using the selective PI3K inhibitor LY294002. Western blot analysis showed that LY294002 markedly reduced the phosphorylation levels of both PI3K and AKT in RAB25-overexpressing HSCs compared with DMSO control, confirming effective pathway inhibition (Fig. 8A). Functionally, LY294002 significantly attenuated the RAB25 overexpression-induced the increasing of cell viability (Fig. 8B), cell proliferative capacity (Fig. 8C) and cell migratory ability (Fig. 8D) in HSCs. In addition, LY294002-treated group exhibited markedly lower expression of α-SMA, Collagen I, MMP2, and PCNA compared with RAB25 overexpression alone (Fig. 8E). These results indicated that pharmacological blockade of the PI3K/AKT pathway effectively reverses the RAB25-mediated activation, proliferation, migration, and fibrotic phenotype of HSCs, supporting that RAB25 drived HSCs activation predominantly through the PI3K/AKT signaling.

4
Discussion
This study reported that RAB25 exhibited abnormally high expression within HBV-associated liver fibrosis for the first time. RAB25 expression correlated with clinical indicators of liver fibrosis severity. The knockdown of RAB25 effectively inhibited HBV-associated hepatocellular carcinoma cells and TGF-β1-induced HSCs activation in vitro. In terms of mechanism, RAB25 knockdown effectively inhibited the expression levels of fibrotic markers (α-SMA and Collagen I), matrix remodeling factor MMP2, and proliferation-associated protein PCNA, and suppressed PI3K/AKT signaling activation. However, the overexpression of RAB25 exerted the opposite effect of RAB25 knockdown on HBV-associated hepatocellular carcinoma cells induced hepatic fibrosis. Besides, alterations in RAB25 expression in HSCs can indirectly influence HBV replication markers in co-cultured hepatocellular carcinoma cells. The targeting of RAB25 is expected to be an underlying therapeutic option for HBV-associated hepatocellular carcinoma cells induced hepatic fibrosis.
Since HSCs have been proven to be the major collagen-producing cells within the damaged liver, multiple experimental and clinical investigations have focused on HSCs as a treatment for hepatic fibrosis (Khomich et al., 2019). Previous studies have demonstrated the effect of RAB25 on HSC activation. RAB25 blocks the regeneration of lipid droplets induced by docosahexaenoic acid (DHA) in activated HSCs, thereby subsequently abrogating the anti-hepatic fibrosis effect of DHA (Qiu et al., 2019). In this study, the upregulation of RAB25 expression was observed in HBV-associated hepatocellular carcinoma cells and TGF-β1-induced activated HSCs. The RAB25 knockdown significantly inhibited HBV-associated hepatocellular carcinoma cells and TGF-β1-induced HSC activation. However, the overexpression of RAB25 significantly promoted HBV-associated hepatocellular carcinoma cells and TGF-β1-induced HSCs activation. This result further confirmed the possibility that targeting RAB25 knockdown could effectively alleviate liver fibrosis.
Hepatic fibrosis is a regenerative process occurring post-injury and is marked by excessive fibroblast proliferation and excessive fibroblast-dominated ECM deposition. Meanwhile, chronic damage stimulated by the long-term fibrotic process may lead to excessive matrix tissue deposition and impaired liver function (Friedman and Pinzani, 2022). In order to respond to liver injury, HSC was activated and transdifferentiated into proliferative and contractile myofibroblasts expressing α-SMA and synthesizing ECM components, including Collagen I and others (Kisseleva and Brenner, 2021). Herein, the effects of knockdown or overexpression of RAB25 upon the viability, proliferation, and migration of HSCs, as well as the expression of HSCs fibrosis markers, matrix remodeling-associated factors, and proliferation-associated proteins, were examined. It was found that the knockdown of RAB25 effectively inhibited the viability, proliferative, and migratory levels of HSCs and reduced the protein levels of α-SMA, Collagen I, MMP2, and PCNA within HSCs. However, RAB25 overexpression exerted the opposite effect to RAB25 knockdown. These results indicate that targeting RAB25 can alleviate liver fibrosis by suppressing the activation of HSCs, thereby alleviating hepatic fibrosis.
The PI3K/AKT pathway was first reported to be activated in HBV-associated liver fibrosis. The PI3K/AKT pathway is tightly associated with fibrosis-related diseases and is a potential therapeutic target for various fibrotic diseases (Wang et al., 2022, Qin et al., 2021, Zhang et al., 2021, Margaria et al., 2022). According to the literature, targeting RAB25 impairs tumor growth in nude mice by inhibiting PI3K and AKT phosphorylation (Ding et al., 2017). Herein, PI3K and AKT phosphorylation were significantly upregulated in HBV-activated HSCs; and RAB25 knockdown effectively inhibited PI3K/AKT activation; the overexpression of RAB25 notably promoted PI3K/AKT activation. Pharmacological blockade of the PI3K/AKT pathway effectively reverses the RAB25-mediated activation, proliferation, migration, and fibrotic phenotype of HSCs. Therefore, the targeting of RAB25 can inhibit HSCs activation by suppressing the PI3K/AKT pathway, thereby inhibiting HSCs activation.
Chronic HBV infection induces a sustained inflammatory and immune microenvironment that profoundly influences HSCs activation and liver fibrogenesis. During HBV infection, immune cell subsets including monocytes/macrophages, Th17 cells, NK cells, and NKT cells are recruited to the liver and secrete cytokines and chemokines such as TGF‑β, IL‑17, TNF‑α, CCL2, and CXCL10, which can directly stimulate HSC activation and signal through pathways including PI3K/AKT to promote extracellular matrix production and fibrogenesis (You et al., 2023). Additionally, immune interactions such as TGF‑β‑mediated suppression of NK cell anti‑fibrotic functions have been observed in HBV‑related cirrhosis, highlighting how immune dysregulation can promote HSCs persistence and extracellular matrix deposition (Shi et al., 2017). Given that RAB25 enhances HSCs activation via the PI3K/AKT axis in vitro model, it is plausible that inflammatory cytokines and immune cell–derived signals in the liver microenvironment could modulate RAB25 expression or function, thereby amplifying HSCs fibrogenic responses in vivo. Conversely, activated HSCs themselves can shape the hepatic immune milieu by secreting cytokines and chemokines that recruit or polarize immune cells, creating a bidirectional crosstalk that sustains chronic inflammation and fibrosis progression (Zhang et al., 2023).
One limitation of this study is the use of HepG2.2.15 cells as the sole in vitro model of HBV expression. Although HepG2.2.15 cells support continuous HBV DNA replication and virion production, this model does not recapitulate natural receptor‑mediated infection or early steps of the viral life cycle, as the HBV genome is integrated and these cells lack the NTCP entry receptor required for authentic HBV infection. HepG2.2.15 cells are also insensitive to infection with patient‑derived virions and do not fully reflect the influence of the immune microenvironment on HBV replication and host responses, limiting the physiological relevance of this system in modelling human disease. Moreover, while HSCs are key effectors in liver fibrosis, the mechanisms by which HBV interactions with HSCs occur, including how stellate cells might influence HBV replication through uptake or paracrine signaling, are not fully understood. Studies suggest that HBV can interact with HSCs and modulate their behavior but does not support classical productive replication in these cells (You et al., 2023), highlighting the need for further investigation of in vivo and more physiologically relevant in vitro models. Future studies should incorporate models that support receptor‑mediated HBV infection, such as HepG2‑NTCP cells or primary human hepatocytes, and examine secreted factors and signaling pathways that mediate communication between HBV‑infected hepatocytes and stellate cells to provide deeper mechanistic insight and greater translational relevance.

5
Conclusion
In summary, targeted therapy against RAB25 and the PI3K/AKT signaling pathway can effectively slow down the activation process of HSCs and provide a potential therapeutic option for HBV-associated liver fibrosis.

Ethical approval
This study was conducted in accordance with the principles of the Declaration of Helsinki. Ethical approval was obtained from the Institutional Review Board of The First Affiliated Hospital of University of South China.

Funding
This work has been supported by the National Natural Science Foundation of Hunan Province, Youth Project (2022JJ40398) and 10.13039/501100010451Research Project of the Health Commission of Hunan Province (202203082548).

CRediT authorship contribution statement
Jing Liu: Writing – original draft, Data curation, Conceptualization. Bingjie Liu: Resources, Methodology, Investigation. Xia Xie: Visualization, Software, Formal analysis. Xin Yang: Validation, Data curation. Qiong Liu: Writing – review & editing, Supervision, Project administration, Funding acquisition.

Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.