SPG21, a potential oncogene targeted by miR-128-3p, amplifies HBx-induced carcinogenesis and chemoresistance via activation of TRPM7-mediated JNK pathway in hepatocellular carcinoma

Abstract

Purpose

Chronic hepatitis B virus (HBV) infection is the primary risk factor for the malignant progression of hepatocellular carcinoma (HCC). It has been reported that HBV X protein (HBx) possesses oncogenic properties, promoting hepatocarcinogenesis and chemoresistance. However, the detailed molecular mechanisms are not fully understood. Here, we aim to investigate the effects of miR-128-3p/SPG21 axis on HBx-induced hepatocarcinogenesis and chemoresistance.

Methods

The expression of SPG21 in HCC was determined using bioinformatics analysis, quantitative real-time PCR (qRT-PCR), western blotting, and immunohistochemistry (IHC). The roles of SPG21 in HCC were elucidated through a series of in vitro and in vivo experiments, including real-time cellular analysis (RTCA), matrigel invasion assay, and xenograft mouse model. Pharmacologic treatment and flow cytometry were performed to demonstrate the potential mechanism of SPG21 in HCC.

Results

SPG21 expression was elevated in HCC tissues compared to adjacent non-tumor tissues (NTs). Moreover, higher SPG21 expression correlated with poor overall survival. Functional assays revealed that SPG21 fostered HCC tumorigenesis and invasion. MiR-128-3p, which targeted SPG21, was downregulated in HCC tissues. Subsequent analyses showed that HBx amplified TRPM7-mediated calcium influx via miR-128-3p/SPG21, thereby activating the c-Jun N-terminal kinase (JNK) pathway. Furthermore, HBx inhibited doxorubicin-induced apoptosis by engaging the JNK pathway through miR-128-3p/SPG21.

Conclusion

The study suggested that SPG21, targeted by miR-128-3p, might be involved in enhancing HBx-induced carcinogenesis and doxorubicin resistance in HCC via the TRPM7/Ca²⁺/JNK signaling pathway. This insight suggested that SPG21 could be recognized as a potential oncogene, offering a novel perspective on its role as a prognostic factor and a therapeutic target in the context of HCC.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13402-024-00955-5.

Introduction

Primary liver cancer ranks as the sixth most commonly diagnosed cancer (906,000 new cases) and the third leading cause of cancer-related deaths (830,000 deaths) worldwide, as reported in global cancer statistics for 2020 [1]. Notably, approximately half of the new cases and deaths occur in China alone [2]. Hepatocellular carcinoma (HCC) constitutes 75–85% of primary liver cancer cases, and chemoresistance remains a significant factor contributing to HCC-related mortality [3].

Chronic hepatitis B virus (HBV) infection is the dominant risk factor for the high incidence of HCC in many regions of Asia and Sub-Saharan Africa [4]. The HBV X protein (HBx), encoded by the X region of the HBV genome, is a 154-amino acid polypeptide with a molecular weight of 17 kDa and has been identified as an oncogenic factor [5]. Previous research has also demonstrated that HBx acts as a tumor inducer, promoting the tumorigenic transformation of normal cells [6]. Although the role of HBx in HCC chemoresistance has been reported [7–9], the underlying mechanism remains unclear.

Recent studies have indicated that the endosomal/trans-Golgi network is involved in the trafficking of HBV envelope proteins [10], protecting HBV from lysosomal degradation and maintaining HBV replication [11]. The trans-Golgi network has also been linked to intracellular processing of cisplatin and mediating resistance in osteosarcoma and ovarian carcinoma [12]. However, its role in HCC chemoresistance has not been documented. Interestingly, spastic paraplegia 21 (SPG21), a protein associated with a complicated form of hereditary spastic paraplegia (HSP) [13], localizes in both the cytoplasm and membrane of the endosomal/trans-Golgi network [14, 15].

MiRNAs are a class of endogenous, short, non-coding RNAs that can regulate gene expression post-transcriptionally by binding to the 3′-UTR complementary sequence of the target gene [16]. They also modulate chemoresistance in HCC [17]. In our previous study, we have found that miR-135a-5p induces resistance of HCC cells to doxorubicin [18]. Furthermore, it has been documented that miR-128-3p is lowly expressed in HCC tissues and acts as a tumor suppressor in HCC progression [19, 20]. However, the relationship between miR-128-3p and HBV, as well as the molecular mechanism of miR-128-3p in chemotherapy resistance of HCC, remains undefined.

The c-Jun N-terminal kinase (JNK), a member of the mitogen-activated protein kinase (MAPK) family, promotes chemoresistance by positively regulating autophagy to counteract apoptosis [21]. Inhibition of the JNK pathway sensitizes HCC cells to cisplatin [22]. Nevertheless, more studies are needed to determine whether HBV is associated with chemotherapy resistance through the JNK pathway.

This study utilized both bioinformatics predictions and clinical sample data to observe that SPG21 was expressed at higher levels in HCC tissues compared to adjacent non-tumor tissues (NTs). The results of a series of experiments, including Cell Counting Kit-8 (CCK-8) assay, real-time cellular analysis (RTCA), transwell migration assay, matrigel invasion assay, wound healing assay, foci formation assay, soft agar colony formation assay, and xenograft mouse model indicated that SPG21 might play a significant role in promoting the development and invasion of HCC. MiR-128-3p, which targeted SPG21, showed high expression in NTs, and HBx reduced its expression by inhibiting its promoter activity through interaction with the transcription factor Snail, resulting in increased SPG21 expression. Further investigation revealed that HBx elevated transient receptor potential melastatin-subfamily member 7 (TRPM7) expression through miR-128-3p/SPG21 to promote calcium influx, thereby activating the JNK pathway. Additionally, HBx inhibited doxorubicin-induced apoptosis through this pathway. In conclusion, we discovered that SPG21 could be involved in promoting malignant transformation and might play a role in HBx-induced carcinogenesis and chemoresistance in HCC, particularly through the TRPM7/Ca²⁺/JNK signaling pathway. These findings provided insights into a possible mechanism of HCC chemoresistance and proposed SPG21 as a potential biomarker for HCC, warranting further investigation.

Methods

Bioinformatics analysis

Data for gene expression profiling of HCC samples were obtained from the GEO datasets (https://www.ncbi.nlm.nih.gov/geo/). Differentially expressed genes (DEGs) between HCC tissues and NTs were identified using the online tool GEO2R, with the standard criteria of|logFC| > 1 and adjusted P < 0.05. The gene ontology (GO) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichments of DEGs were performed using DAVID (https://david.ncifcrf.gov/). GO enrichment analysis consisted of molecular functions (MF), biological processes (BP), and cellular components (CC).

Clinical samples

A total of 84 pairs of HCC tissues and adjacent non-tumor tissues were collected from Renmin Hospital of Wuhan University (Wuhan, China) for quantitative real-time PCR (qRT-PCR), western blotting, and immunohistochemistry (IHC) analysis. The samples were classified using the American Joint Committee on Cancer/Union for International Cancer Control (AJCC/UICC) tumor-node-metastasis (TNM) staging system. Sample collection was conducted with consensus and was approved by the Ethics Committee of Wuhan University, School of Basic Medical Sciences (Wuhan, China).

Cell culture and transfection

HepG2, HepG2.2.15, HCCLM3, NIH3T3, and HEK293T cell lines were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA). The Huh7 cell line was obtained from the Japanese Collection of Research Bioresources Cell Bank (Osaka, Japan). The Bel-7402 cell line was originally purchased from the ATCC and was authenticated using Short Tandem Repeat (STR) profiling. All cell lines were cultured at 37 °C in a humidified atmosphere with 5% CO₂ in Dulbecco’s modified Eagle’s medium (DMEM, 11,965,092, Gibco, USA) containing 10% fetal bovine serum (FBS, 16,140,071, Gibco, USA) and 1% penicillin/streptomycin (15,140,122, Gibco, USA). HepG2.2.15 cell line was additionally maintained in a medium containing 400 µg/mL G418 (11,811,023, Invitrogen, CA, USA). Transfection was performed using Lipofectamine 2000 (11,668,019, Invitrogen, CA, USA) following the manufacturer’s instructions.

Dual-luciferase reporter assay

To validate miRNA-mRNA interaction, the SPG21-3’UTR-wild type (WT) containing the miR-128-3p binding recognition site was cloned into the pmirGLO vector. The promoter sequence was cloned into the pGL3-Basic vector and the internal control vector pRL-TK was simultaneously transfected to detect promoter activity. The Dual-Luciferase Reporter Assay System (E1960, Promega, USA) was used to detect firefly and Renilla luciferase activities following the manufacturer’s instructions. The transfection efficiency was normalized by Renilla luciferase activity.

Co-immunoprecipitation (Co-IP) assay

HEK293T cells were lysed in Lysis Buffer (20 mmol/L Tris-Cl [pH 7.5], 150 mmol/L NaCl, 1% Triton X-100) containing 1 × protease inhibitor cocktail (P8340, Sigma-Aldrich, MO, USA). After centrifugation, the supernatant was incubated with the indicated antibodies and then Protein A/G Agarose (20,421, Thermo Scientific, USA) was added. The immunoprecipitated proteins were analyzed by western blotting.

Apoptosis assay

Approximately 1 × 10⁵ resuspended cells were collected and treated with the Annexin V-FITC/propidium iodide (PI) Apoptosis Assay Kit (ZP327-1, Zomanbio, Beijing, China) according to the manufacturer’s instructions. Apoptosis rates were detected by flow cytometer (FACS Aria III, BD, USA) and the results were analyzed using FlowJo v10 software (Leonard Herzenberg, USA).

RTCA

The RTCA (xCELLigence, Roche, Basel, Switzerland) was used as previously described to monitor cell proliferation [23]. 5 × 10³ cells were seeded in an E-plate 16 and then treated under different conditions. The software supplied with the instrument monitored and recorded the cell index (indicating cell activity in the assay well) every 5 min and continuously monitored for up to 48 h.

CCK-8 assay

Cells were seeded in 96-well plates and subjected to the CCK-8 assay (ZP328-1, Zomanbio, China) following the manufacturer’s instructions. Absorbance at 450 nm was measured using a Microplate Reader (Multiskan FC 357, Thermo Scientific, USA).

QRT-PCR

Total RNA was extracted using TRIzol reagent (15,596,018, Invitrogen, USA) and cDNA was synthesized using the RT Master Mix with gDNA Remover (FSQ-301, TOYOBO, Japan). QRT-PCR was performed on an iCycler system (C1000, Bio-Rad, USA) using the SYBR Green PCR Master Mix (04913914001, Roche, Switzerland). β-Actin or U6 was used as an internal control to normalize the levels of other genes. The primers were listed in Supplementary Table 9.

Wound healing assay

The ability of cell migration was evaluated by the wound healing assay. After serum starvation for 24 h, the cells in a 6-well plate were scratched using a 10 µL pipette tip and washed with serum-free medium to remove detached cells. The migration of cells to the wound area was monitored every 24 h using an inverted phase contrast microscope (Olympus CH-40, Japan). The area of wound closure was calculated using the following formula: migration area (%) = (M₀ - M_n)/M₀ × 100, where M₀ represented the initial wound area at 0 h, and M_n represented the remaining wound area at the measurement time point.

Transwell migration assay and matrigel invasion assay

For the transwell migration assay, cells in serum-free media were seeded into 24-well transwell chambers (3422, Corning, USA) at approximately 5 × 10⁴ cells per well. Medium containing 10% FBS in the lower chamber served as a chemotactic agent. The cells were then incubated at 37 °C for 48 h. After incubation, the cells adhered to the lower chamber were fixed, stained with crystal violet, and counted under a microscope (Olympus CH-40, Japan). For the matrigel invasion assay, matrigel (356,234, BD Biosciences, USA) diluted to 200 µg/mL was used to coat the transwell chambers. The rest of the procedure was the same as described above.

Foci formation assay and soft agar colony formation assay

For the foci formation assay, cells were seeded into a 6-well plate at approximately 1,000 cells/well. After incubation at 37 °C for 2 weeks, the colonies were stained with crystal violet, and those consisting of more than 50 cells were counted. For the soft agar colony formation assay, first, 2 mL of semi-solid medium containing 1.2% low gelling agar (A9045, Sigma-Aldrich, MO, USA) was laid in a 6-well plate and placed at room temperature for 20 min. Then, 2 mL of semi-solid medium containing about 1,000 cells and 0.7% low gelling agar was poured onto the upper layer, and the plate was placed at 4 °C for 5 min. After incubation at 37 °C for 3 weeks, the colonies were stained with crystal violet and examined under a microscope (Olympus CH-40, Japan). Clusters with a diameter greater than 0.2 mm were used as the counting standard.

Xenograft mouse model

5 × 10⁶ cells were subcutaneously injected into the dorsal flank of 4-6-week-old BALB/c nude mice. Tumor volume (mm³) was measured every 5 days after injection up to 30 days, and the volume was calculated according to the following formula: volume = ( Length × Width² )/2. Mice were sacrificed under anesthesia and the tumors were harvested. The tumor samples were evaluated by hematoxylin and eosin (H&E), terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL), and proliferating cell nuclear antigen (PCNA) staining, following the manufacturer’s instructions. All animal care and handling procedures were carried out under the National Institutes of Health’s Guide for the Care and Use of Laboratory Animals. Animal experiments were approved by the Animal Ethics Committee of Wuhan University, Wuhan University Center for Animal Experiment/A3 Laboratory.

ELISA and western blotting

The levels of IL-6 (1:5, RX106126H), TNF-α (1:5, RX104793H), VEGF (1:5, RX105003H), MMP-2 (1:5, RX105775H), and MMP-9 (1:5, RX105770H) in culture supernatants were measured by ELISA kits (Warner Bio, Wuhan, China) following the manufacturer’s instructions.

Western blotting was performed using the standard method. The following antibodies were purchased from Abcam (Cambridge, UK): SPG21 (1:1000, ab220229), TRPM7 (1:200, ab135817), MYC (1:1000, ab32072), Bcl-xL (1:1000, ab32370), Cyclin D1 (1:1000, ab134175), p-JNK (1:1000, ab131499), JNK (1:1000, ab179461), and HBx (1:1000, ab2741). Snail (1:1000, A5243) and β-Actin (1:20,000, AC028) were purchased from ABclonal (Wuhan, China). β-Actin was used as an internal control to normalize the levels of other proteins.

IHC

IHC staining was carried out according to the standard procedure using the following antibodies: SPG21 (1:200, HPA040436, Atlas Antibodies, Sweden), and TRPM7 (1:100, ab135817, Abcam, UK) were used as primary antibodies. The secondary antibody was HRP-conjugated goat anti-rabbit IgG (1:2000, ab205718, Abcam, UK). The staining intensity was scored as 0 (negative), 1 (weak), 2 (moderate), 3 (strong), and 4 (very strong). Samples with a score up to 2 + were considered positive, while scores of 0 and 1 + were considered negative.

Pharmacologic treatment

SP600125 (HY-12,041, MCE, USA), a highly selective JNK inhibitor, was used to prevent the activation of the JNK pathway [24]. Anisomycin (HY-18,982, MCE, USA), a potent JNK agonist, was used to activate the JNK pathway [25]. Cells were treated with 10 µmol/L SP600125 or 10 µmol/L Anisomycin in a humidified incubator for 24 h.

Bradykinin (HY-P0206, MCE, USA), a specific TRPM7 agonist, was used to activate TRPM7 channels [26]. 2-APB (HY-W009724, MCE, USA), a non-specific TRPM7 inhibitor, was used to block TRPM7 channels [27].

Doxorubicin (D8740, Solarbio, Beijing, China) is a chemotherapeutic agent used to treat a variety of cancers. Doxorubicin was added to the culture media for 48 h in serial dilutions or at the stated concentrations.

Measurement of intracellular calcium

The HCC cells were incubated with 5 µmol/L Fluo 8-AM (21,083, AAT Bioquest, USA) and 0.04% Pluronic F-127 (P3000MP, Invitrogen, USA) at 37 °C for 30 min, followed by 3 washes and additional incubation in Hank’s Balanced Salt Solution (HBSS, SH30268.01, HyClone, UT, USA) at 37 °C for 10 min. Fluorescent signals were captured by a fluorescence microscope (AF6000, Leica, Wetzlar, Germany). The fluorescence intensity of calcium was recorded over time and fluorescence images were collected every 1 s. The ΔF/F₀ ratio was calculated as (F - F₀)/F₀. F indicated fluorescence intensity, F₀ was the baseline fluorescence before 100 µmol/L Bradykinin treatment, and ΔF was the difference between F and F₀.

The fluorescence was measured by a flow cytometer (FACS Aria III, BD, USA) with an excitation wavelength of 490 nm and an emission wavelength of 525 nm. The intracellular calcium concentration ([Ca²⁺]_i) was calculated by the following formula: [Ca²⁺]_i = K_d × [(F - F_min)/(F_max - F)]. The dissociation constant (K_d) of the calcium probe was 389 nmol/L. F was the fluorescence of the probe at experimental calcium levels, F_min was the fluorescence in the absence of calcium, and F_max was the fluorescence at calcium saturation.

Statistical analysis

All experiments were performed with at least three replicates. Spearman’s correlation analysis was used to assess the correlation between rank variables. One-way or two-way ANOVA with Tukey correction or with Sidak correction was used for multiple comparisons. Cox proportional hazard regression models were performed for the univariate and multivariate analyses of survival. Data were analyzed using GraphPad Prism 8 software (GraphPad Software Inc., USA) and SPSS Statistics 20 software (IBM Corp., USA). Results were presented as mean ± standard error of the mean (SEM), or mean ± standard deviation (SD). P < 0.05 was considered statistically significant. All experiments were carried out blindly.

Results

Elevated SPG21 expression is closely associated with advanced TNM stage and poor prognosis in HCC

To determine the DEGs of HBV-related HCC, the online tool GEO2R was used to analyze the GSE62232 dataset, which contained 10 HBV-related HCC tissues and 10 NTs. The results showed 910 up-regulated genes and 786 down-regulated genes (Supplementary Fig. S1A and B). According to GO enrichment analysis, the up-regulated genes associated with HBV-related HCC were involved in the positive regulation of MAPK activity, positive regulation of release of sequestered calcium ion into cytosol, cell migration, and other processes (Fig. 1A; Supplementary Table 1). KEGG and Gene Set Enrichment Analysis (GSEA) found that up-regulated genes were involved in the cell cycle, pathways in cancer (such as JAK-STAT, JNK/MAPK, Calcium, Wnt, etc.), viral carcinogenesis (such as HBx), basal transcription factors, and more (Supplementary Fig. S1C and D; Supplementary Tables 2 and 3).

Fig. 1.

High expression of SPG21 is an indicator of poor prognosis in HCC. A GO-BP enrichment analysis of 910 up-regulated genes in HBV-related HCC. B SPG21 was highly expressed in pan-cancer, including LIHC in the TIMER2.0 database. C SPG21 mRNA expression was significantly higher in HBV-related HCC tissues than in NTs in the GSE3500 and GSE14520 datasets. D-F SPG21 expression was significantly higher in HCC tissues than in NTs detected by qRT-PCR, western blotting (WB), and IHC, respectively. G-I The expression of SPG21 in stage III-IV HCC tissues was higher than that in stage I-II HCC tissues, as shown by qRT-PCR, western blotting, and IHC, respectively. J ROC curve analysis of SPG21 based on western blotting experimental data from 84 pairs of HCC tissues and NTs. K Kaplan-Meier analysis of overall survival was performed according to SPG21 levels. L and M The expression of SPG21 was significantly higher in HCC cell lines than in NT, as determined by qRT-PCR and western blotting, respectively. *P < 0.05, **P < 0.01, ***P < 0.001

To clarify the expression of SPG21 in pan-cancer, we utilized the Tumor Immune Estimation Resource 2.0 (TIMER2.0) database (http://timer.cistrome.org/) and observed that SPG21 was highly expressed in various tumor tissues, such as Liver hepatocellular carcinoma (LIHC), compared to NTs (Fig. 1B). The UALCAN database (https://ualcan.path.uab.edu/index.html) also indicated a significant elevation of SPG21 in HCC tissues compared to NTs (Supplementary Fig. S2A). Further analysis using the GSE62232 dataset showed that SPG21 was markedly higher in HBV-related HCC tissues compared to NTs. This pattern of expression was distinctively associated with HBV-related HCC, as it was not observed in HCC tissues related to HCV and alcohol (Supplementary Fig. S1E). The GSE3500 dataset, which included 90 HBV-related HCC tissues and 56 NTs, and the GSE14520 dataset, comprising 212 HBV-related HCC tissues and 220 NTs, both consistently demonstrated higher SPG21 expression in HBV-related HCC tissues compared to NTs (Fig. 1C). In contrast, analysis of the GSE6764 dataset, containing 35 HCV-related HCC tissues and 10 NTs, showed no difference in SPG21 expression between HCV-related HCC tissues and NTs (Supplementary Fig. S2B). These findings collectively suggested a possible link between HBV infection and the upregulation of SPG21 expression.

We collected HCC samples and their corresponding NTs to validate the results of bioinformatics analysis. QRT-PCR indicated that SPG21 was highly expressed in 58 of 84 (69.05%) HCC tissues and only in 20 of 84 (23.81%) NTs (Fig. 1D; Supplementary Table 4). Western blotting also demonstrated increased expression of SPG21 in 59 of 84 (70.24%) HCC samples, but only in 18 of 84 (21.43%) NTs (Fig. 1E; Supplementary Table 5). IHC further showed that 61 of 84 (72.62%) HCC samples were positively stained, whereas only 17 of 84 (20.24%) NTs exhibited positive staining for SPG21 (Fig. 1F; Supplementary Table 6).

We re-analyzed the data from HCC tissues to further investigate the role of SPG21 in the development of HCC. QRT-PCR and western blotting revealed that the expression of SPG21 in stage III-IV HCC tissues was higher than that in stage I-II HCC tissues (Fig. 1G and H). In IHC, only 19 of 34 (55.88%) HCC tissues at stage I-II were positively stained for SPG21, whereas 42 of 50 (84%) HCC tissues at stage III-IV showed positive staining for SPG21 (Fig. 1I; Supplementary Table 7). The data also indicated that the level of SPG21 was positively correlated with tumor size, vascular invasion, alpha-fetoprotein (AFP), and HBsAg. However, there was no significant correlation between the expression of SPG21 and age/gender (Supplementary Table 7). Our clinical sample set, comprising 73 HBV-related HCC tissues and 11 Non-HBV-related HCC tissues, revealed that SPG21 expression was substantially higher in HBV-related HCC tissues compared to Non-HBV-related HCC tissues (Supplementary Fig. S3A and B).

We performed receiver operator characteristic (ROC) curve analysis to assess the predictive value of SPG21 in the diagnosis of HCC patients and identified SPG21 as a potential diagnostic marker for HCC (Fig. 1J). Furthermore, survival analysis of our clinical samples indicated that high expression of SPG21 was positively associated with poor overall survival (Fig. 1K). Cox proportional hazard regression models were performed to further assess the prognostic potential of SPG21 expression. High SPG21 and AFP levels, as well as advanced TNM stage, were associated with worse overall survival in univariate analysis (Supplementary Table 8). Further multivariate analysis identified high SPG21 levels as an independent risk factor for worse overall survival (Supplementary Table 8).

The expression of SPG21 was also detected in five HCC cell lines and NT by qRT-PCR (Fig. 1L) and western blotting (Fig. 1M). Its expression level was significantly higher in HCC cell lines than in NT, and the level was higher in HepG2.2.15 cells than in HepG2 cells which might indicate a role of HBV infection in the expression of SPG21.

In summary, SPG21 mRNA and protein levels were significantly upregulated in both HCC tissues and HCC cell lines, and high expression of SPG21 in HCC was associated with poor clinical outcomes.

SPG21 promotes HCC tumorigenesis and invasion

Alterations in tissue mechanics and metabolism are defining features of cancer, affecting not only proliferation but also migration and invasion [28]. To determine whether SPG21 can promote cell proliferation, migration, and invasion in HCC, we performed RTCA, CCK-8 assay, transwell migration assay, matrigel invasion assay, and wound healing assay.

The RTCA results showed that the proliferation rate of NIH3T3 cells transfected with SPG21 (NIH3T3-SPG21 cells) was significantly increased compared to the control (Supplementary Fig. S4A and B; Fig. 2A), whereas knockdown of SPG21 in HepG2.2.15 cells (HepG2.2.15-shSPG21 cells) had the opposite effects (Supplementary Fig. S4C; Fig. 2B). CCK-8 assay data also confirmed the above results (Fig. 2C and D). Both transwell migration assay and matrigel invasion assay results showed that NIH3T3-SPG21 cells could significantly promote cell penetration from the upper chamber to the lower chamber (Fig. 2E). In contrast, HepG2.2.15-shSPG21 cells showed the opposite effects (Fig. 2F). Moreover, NIH3T3-SPG21 cells significantly promoted wound closure (Fig. 2G), while HepG2.2.15-shSPG21 cells inhibited this effect (Fig. 2H). These results suggested that SPG21 might enhance the proliferation and metastatic potential of HCC cells.

Fig. 2.

SPG21 enhances cell proliferation, migration, and invasion of HCC cells. A and C RTCA and CCK-8 assay showed that NIH3T3-SPG21 cells exhibited an increased proliferation rate. B and D RTCA and CCK-8 assay demonstrated that HepG2.2.15-shSPG21 cells reduced proliferation rate. E Transwell migration assay and matrigel invasion assay showed that SPG21 enhanced migration and invasion in NIH3T3 cells. F Transwell migration assay and matrigel invasion assay showed that knockdown of SPG21 attenuated migration and invasion in HepG2.2.15 cells. G Wound healing assay showed that SPG21 promoted wound closure in NIH3T3 cells. H Wound healing assay showed that knockdown of SPG21 inhibited wound closure in HepG2.2.15 cells. **P < 0.01, ***P < 0.001

The direct evidence used to determine tumorigenicity includes the foci formation assay, soft agar colony formation assay, and xenograft mouse model [29]. The Foci formation assay and soft agar colony formation assay were utilized to assess the effect of SPG21 on the malignant transformation ability of cells. The number of colony formations significantly increased in NIH3T3-SPG21 cells compared to the control (Fig. 3A), while HepG2.2.15-shSPG21 cells exhibited a decrease in the number of colony formations (Fig. 3B). The data from the soft agar colony formation assay also confirmed the above results (Fig. 3C and D).

Fig. 3.

SPG21 enhances cell tumorigenicity. A and C Foci formation assay and soft agar colony formation assay demonstrated that NIH3T3-SPG21 cells increased the number of colony formations. B and D Foci formation assay and soft agar colony formation assay showed that knockdown of SPG21 reduced the number of colony formations in HepG2.2.15 cells. E NIH3T3-SPG21 cells induced a larger tumor volume than the negative control. Arrows indicate the xenograft tumors. F HCCLM3-shSPG21 cells induced a smaller tumor volume than the positive control. Arrows indicate the xenograft tumors. G H&E staining of nude mice tumor samples. H-J TUNEL and PCNA staining showed that the reduction in tumor volume was due to increased apoptosis and decreased proliferation. *P < 0.05, ***P < 0.001

Furthermore, the xenograft mouse model demonstrated that NIH3T3-SPG21 cells induced tumor formation on the flank of nude mice, whereas the negative control nude mice without overexpression of SPG21 did not form tumors within 30 days (Fig. 3E). Knockdown of SPG21 in HCCLM3 cells (HCCLM3-shSPG21 cells) resulted in a significant reduction in tumor volume compared to the positive control (Fig. 3F).

To demonstrate malignant morphological changes, including large nucleoli, nuclear pleomorphism, pathological nuclear division phase, etc., H&E staining of nude mice tumor samples was conducted (Fig. 3G). The results demonstrated that cells in the control group had densely stained nuclei, were highly proliferated and closely packed. In contrast, cells in the SPG21 knockdown group appeared more dispersed with numerous vacuolated structures, and evident signs of nuclear fragmentation and dissolution, indicative of cell necrosis. Additionally, TUNEL and PCNA staining showed that the reduction in tumor volume was attributed to an increase in apoptosis and a decrease in proliferation (Fig. 3H-J). These results suggested that SPG21 might have oncogenic potential in HCC.

HBx increases SPG21 expression and exerts tumorgenicity by interacting with snail to inhibit miR-128-3p

QRT-PCR and western blotting were employed to investigate the effect of HBV on SPG21 expression levels. The results showed that mRNA and protein levels of SPG21 were elevated in HepG2 cells after transfection with HBV1.3 (Supplementary Fig. S5A; Fig. 4A). Moreover, we transfected HBs, HBc, HBe, HBx, and HBp plasmids in HepG2 cells to determine which viral protein of HBV caused elevated SPG21 expression. The results indicated that HBx promoted the mRNA and protein levels of SPG21 (Supplementary Fig. S5B; Fig. 4B).

Fig. 4.

HBx promotes SPG21 levels by interacting with Snail to suppress miR-128-3p. A and B Western blotting showed elevated protein levels of SPG21 in HepG2 cells by HBV1.3 and HBx. C Dual-luciferase reporter assay demonstrated that HBx significantly increased SPG21 promoter activity in HepG2 cells. D Activities of serially 5’ truncated SPG21 promoter cotransfected with HBx in HepG2 cells. E Confocal imaging showed the co-localization of pEGFP-C1-SPG21 and pcDNA3.1-HBx in HEK293T cells (scale bar 10 μm). F Co-IP assay revealed a direct interaction between SPG21-HA and HBx-Flag in HEK293T cells. G Bioinformatics analysis predicted miRNAs that could target SPG21. H TargetScan database predicted two potential binding sites for miR-128-3p. I Dual-luciferase reporter assay in HEK293T cells verified that miR-128-3p targeted SPG21. J and K Protein levels of SPG21 were detected by western blotting in HepG2.2.15 cells with miR-128-3p-mimic and HepG2 cells with miR-128-3p-inhibitor. L Expression of miR-128-3p in HCC tissues and NTs was detected by qRT-PCR. M Expression of SPG21 was significantly negatively related to the level of miR-128-3p using qRT-PCR. N Expression of miR-128-3p in NT and HCC cell lines was detected by qRT-PCR. O and P QRT-PCR showed that the level of miR-128-3p was decreased by HBV1.3 and HBx in HepG2 cells. Q Dual-luciferase reporter assay demonstrated that HBx significantly inhibited miR-128-2 promoter activity in HepG2 cells. R Western blotting showed that HBx elevated the protein level of Snail in HepG2 cells. S Confocal imaging showed the co-localization of pEGFP-C1-Snail and pcDNA3.1-HBx in HEK293T cells (scale bar 7.5 μm). T Co-IP assay revealed a direct interaction between Snail-Flag and HBx-HA in HEK293T cells. U Expression of miR-128-3p was detected by qRT-PCR after knockdown of Snail in HepG2-HBx cells. V MiR-128-2 promoter activity was detected by dual-luciferase reporter assay after knockdown of Snail in HepG2-HBx cells. *P < 0.05, **P < 0.01, ***P < 0.001

The dual-luciferase reporter assay revealed that HBx significantly increased SPG21 promoter activity (Fig. 4C), suggesting that HBx promoted SPG21 expression by elevating SPG21 promoter activity. To determine the shortest sequence required for transcriptional regulation of SPG21, we examined the activities of seven promoters with 5’ truncated sequences (-1300 to + 100, -1100 to + 100, -900 to + 100, -700 to + 100, -500 to + 100, -300 to + 100, -100 to + 100). The results demonstrated that the SPG21 promoter ranging from − 100 to + 100 was the shortest sequence critical for promoter activity (Fig. 4D).

Confocal imaging showed the co-localization of SPG21 and HBx in the cell membrane and cytoplasm (Fig. 4E), and Co-IP assay results further demonstrated a direct interaction between them (Fig. 4F).

MiRNAs play a role by inhibiting the expression of downstream genes [16]. Therefore, we utilized the TargetScan (https://www.targetscan.org/vert_72/), DIANA microT (https://dianalab.e-ce.uth.gr/microt_webserver/#/), and miRDB (https://mirdb.org/) databases to explore the miRNAs that could target SPG21. The intersection of the three databases showed that only one miRNA, miR-128-3p, could target SPG21 (Fig. 4G). The TargetScan database predicted two potential binding sites for miR-128-3p at positions 112–118 and 281–287 of the SPG21 3’ UTR, respectively (Fig. 4H). To further analyze the post-transcriptional regulation of SPG21 by miR-128-3p, we individually or co-mutated the predicted binding targets. The results of the dual-luciferase reporter assay showed that cotransfection of miR-128-3p-mimic and SPG21-3’UTR-WT in HEK293T cells resulted in a significant decrease in fluorescence intensity compared to the control, which was recovered by addition of the 3’UTR mutant (Fig. 4I). Consistently, mRNA and protein levels of SPG21 were decreased in HepG2.2.15 cells after addition of miR-128-3p-mimic, but increased in HepG2 cells after addition of miR-128-3p-inhibitor (Supplementary Fig. S5C-F; Fig. 4J and K). These data suggested that SPG21 was a direct target of miR-128-3p.

The results above demonstrated that SPG21 was highly expressed in HCC tissues compared to corresponding NTs. Therefore, we aimed to explore the expression of miR-128-3p in HCC tissues. QRT-PCR results showed that the miR-128-3p expression level was opposite to SPG21 in 84 pairs of HCC clinical samples (Fig. 4L). There was a significant negative correlation between the expression of SPG21 and miR-128-3p, as determined by Spearman’s correlation analysis (Fig. 4M). Similarly, miR-128-3p was highly expressed in NT compared to HCC cell lines, and its level was lower in HepG2.2.15 cells than in HepG2 cells, indicating that HBV infection might play a role in miR-128-3p expression (Fig. 4N).

After transfection with HBV1.3 in HepG2 cells, the expression level of miR-128-3p was decreased (Fig. 4O), and the viral protein HBx had the most pronounced inhibitory effect on miR-128-3p (Fig. 4P). The results of the dual-luciferase reporter assay showed that the viral protein HBx significantly reduced the activity of the miR-128-2 promoter (Fig. 4Q), indicating that HBx suppressed miR-128-3p expression by reducing the activity of the miR-128-2 promoter.

We have found that miR-128-3p can be regulated by viral proteins, but it remains unclear whether there are transcription factors that can regulate miR-128-3p expression in HCC. Previous studies have shown that the transcription factor Snail can bind to two conserved E-box motifs relative to the transcription start site of the human miR-128-2 stem-loop to reduce the expression level of mature miR-128 in MCF-10 A cells [30]. To further verify whether Snail can suppress the expression level of miR-128-3p in HCC cells, we constructed a 1.4-kb fragment upstream of the human miR-128-2 stem-loop containing two conserved E-box motifs (Supplementary Fig. S5G). The Dual-luciferase reporter assay showed that cotransfection of Snail and miR-128-2-WT in HepG2 cells significantly decreased the fluorescence intensity compared to the control. This inhibitory effect was only absent when both E-box motifs were mutated, suggesting that Snail could bind to either E-box motif to exert an inhibitory effect on the miR-128-2 promoter (Supplementary Fig. S5H). QRT-PCR results showed that knockdown of Snail could inhibit SPG21 expression by elevating miR-128-3p (Supplementary Fig. S5I-K).

In addition, we discovered that HBx could elevate the mRNA and protein levels of Snail (Supplementary Fig. S5L-N; Fig. 4R). Confocal imaging results showed that in HEK293T cells, HBx co-localized with Snail mainly in the cytoplasm, with a small amount in the nucleus (Fig. 4S), and Co-IP assay results further confirmed the direct interaction between them (Fig. 4T).

Knockdown of Snail could reverse the inhibitory effect of HBx on miR-128-3p (Fig. 4U) and similar results were obtained from the dual-luciferase reporter assay (Fig. 4V). In contrast, knockdown of Snail in HepG2 cells transfected with HBx (HepG2-HBx cells) suppressed the expression of SPG21 (Supplementary Fig. S5O). The above results suggested that HBx repressed miR-128-3p and promoted SPG21 expression by interacting with Snail.

As mentioned earlier, HBx can inhibit miR-128-3p expression, thereby promoting SPG21 expression. Consequently, we sought to investigate whether HBx exerts oncogenic properties through this mechanism. RTCA demonstrated that treatment with miR-128-3p-mimic in HepG2-HBx cells resulted in a significant decrease in cell proliferation rate compared to the control (Fig. 5A). Transwell migration assay and matrigel invasion assay revealed that treatment with miR-128-3p-mimic in HepG2-HBx cells inhibited cell migration and invasion compared to the control (Fig. 5B). Furthermore, the foci formation assay also proved that treatment with miR-128-3p-mimic in HepG2-HBx cells reduced the number of colony formations (Fig. 5C).

Fig. 5.

HBx exerts oncogenic properties through miR-128-3p/SPG21. A and D RTCA showed that treatment with miR-128-3p-mimic and knockdown of SPG21 in HepG2-HBx cells decreased the cell proliferation rate. B and E Transwell migration assay and matrigel invasion assay showed that treatment with miR-128-3p-mimic and knockdown of SPG21 in HepG2-HBx cells inhibited cell migration and invasion. C and F Foci formation assay showed that treatment with miR-128-3p-mimic and knockdown of SPG21 in HepG2-HBx cells reduced the number of colony formations. G Xenograft mouse model showed that treatment with miR-128-3p-mimic in Bel-7402-HBx cells reduced tumor volume. Arrows indicate the xenograft tumors. H H&E staining of nude mice tumor samples. I-K TUNEL and PCNA staining showed that the reduction in tumor volume was due to an increase in apoptosis and a decrease in proliferation. **P < 0.01, ***P < 0.001

SPG21 acts as a downstream target of miR-128-3p, and we aimed to further verify whether SPG21 plays a role in HBx exerting tumorigenesis. RTCA demonstrated that knockdown of SPG21 in HepG2-HBx cells significantly reduced the cell proliferation rate compared to the control (Fig. 5D). Transwell migration assay and matrigel invasion assay also showed that knockdown of SPG21 in HepG2-HBx cells significantly inhibited cell migration and invasion (Fig. 5E). Moreover, the foci formation assay showed that knockdown of SPG21 in HepG2-HBx cells significantly decreased the number of colony formations compared to the control (Fig. 5F).

An in vivo tumorigenesis assay in nude mice was conducted to further validate the tumor suppressive effect of miR-128-3p. Treatment with miR-128-3p-mimic in Bel-7402 cells transfected with HBx (Bel-7402-HBx cells) significantly reduced tumor volume compared to the control (Fig. 5G). The results of H&E staining results showed that miR-128-3p-mimic treated Bel-7402-HBx cells exhibited extensive nuclei fragmentation and dissolution, indicative of pronounced cell necrosis (Fig. 5H). Additionally, TUNEL and PCNA staining also confirmed that the above results were due to increased apoptosis and decreased proliferation (Fig. 5I-K). These results indicated that HBx might facilitate tumorigenesis through the miR-128-3p/SPG21 axis.

HBx activates the JNK pathway by elevating TRPM7-mediated calcium influx through miR-128-3p/SPG21

Bioinformatics analysis using the GSE62232 dataset showed that genes associated with HBV-related HCC could activate various cancer pathways including JNK/MAPK. Some studies have reported that the JNK pathway can promote inflammation-mediated hepatocarcinogenesis [31, 32]. Therefore, we aimed to investigate whether SPG21 can be involved in inflammation-mediated hepatocarcinogenesis via the JNK pathway. Western blotting data showed that transfection of SPG21 in HepG2 cells elevated p-JNK, MYC, Bcl-xL, and Cyclin D1 protein levels (Fig. 6A). The levels of inflammatory mediators downstream of the JNK pathway, such as IL-6, TNF-α, VEGF, MMP-2, and MMP-9, were also significantly elevated in the ELISA results (Supplementary Fig. S6A). Knockdown of SPG21 in HepG2.2.15 cells decreased the expression of the downstream proteins (Fig. 6B; Supplementary Fig. S6B).

Fig. 6.

HBx activates the JNK pathway through miR-128-3p/SPG21. A Western blotting showed that overexpression of SPG21 in HepG2 cells increased the levels of p-JNK and its downstream proteins. B Western blotting showed that knockdown of SPG21 in HepG2.2.15 cells decreased the levels of p-JNK and its downstream proteins. C Western blotting showed that treatment with SP600125 in HepG2-SPG21 cells decreased the levels of p-JNK and its downstream proteins. D-F RTCA, transwell migration assay, matrigel invasion assay, and foci formation assay showed that treatment with SP600125 in HepG2-SPG21 cells inhibited cell proliferation, migration, invasion, and clonogenicity. G Xenograft mouse model showed that treatment with SP600125 in Bel-7402 cells transfected with SPG21 (Bel-7402-SPG21 cells) reduced tumor volume. Arrows indicate the xenograft tumors. H H&E staining of tumor samples. I-K TUNEL and PCNA staining showed that the reduction in tumor volume was due to increased apoptosis and decreased proliferation. L and M Western blotting showed that treatment with miR-128-3p-mimic and knockdown of SPG21 in HepG2-HBx cells decreased the levels of p-JNK and its downstream proteins. *P < 0.05, **P < 0.01, ***P < 0.001

To further verify whether SPG21 upregulates these downstream proteins in HepG2 cells by activating the JNK pathway, we used the JNK inhibitor SP600125. The results showed that the addition of SP600125 in HepG2 cells transfected with SPG21 (HepG2-SPG21 cells) significantly reversed the expression of p-JNK and its downstream proteins (Fig. 6C; Supplementary Fig. S6C). RTCA, transwell migration assay, and matrigel invasion assay showed that the ability of SPG21 to promote cell proliferation, migration, and invasion was significantly inhibited by SP600125 (Fig. 6D and E). Moreover, the number of colony formations significantly decreased after SP600125 treatment compared to the control (Fig. 6F). Additionally, the tumor volume was also significantly reduced after SP600125 treatment compared to the control (Fig. 6G). This reduction in tumorigenic potential was further confirmed through H&E, TUNEL, and PCNA staining, which collectively demonstrated the inhibitory effect of SP600125 on the tumorigenic capabilities of SPG21 (Fig. 6H-K). On the other hand, the introduction of the JNK agonist anisomycin following SPG21 knockdown led to a marked increase in the expression levels of p-JNK and its downstream proteins (Supplementary Fig. S7A). This treatment also enhanced cell migration and invasion (Supplementary Fig. S7B).

Aerobic glycolysis is beneficial for tumor proliferation, metastasis, drug resistance, immune escape, etc. It not only provides rapid energy but also supplies raw materials for the synthesis of various biomolecules in tumor cells. Various intermediate metabolites produced during this process, such as pyruvate and lactate, are essential precursors for various other metabolic pathways. Emerging evidence suggests that aerobic glycolysis can promote the malignant progression of HCC [33, 34]. Therefore, we investigated whether SPG21 can promote aerobic glycolysis to drive hepatocarcinogenesis. The results showed that overexpression of SPG21 in HepG2 cells promoted pyruvate kinase activity and lactate production (Supplementary Fig. S8A and B), which was significantly reversed by addition of SP600125 (Supplementary Fig. S8E and F). Additionally, knockdown of SPG21 in HepG2.2.15 cells inhibited pyruvate kinase activity and lactate production (Supplementary Fig. S8C and D).

Given that SPG21 can activate the JNK pathway and SPG21 is a target of miR-128-3p, whether HBx can activate the JNK pathway through miR-128-3p/SPG21 remains to be further validated. Western blotting and ELISA results showed that the addition of miR-128-3p-mimic in HepG2-HBx cells significantly inhibited the expression of p-JNK and its downstream proteins (Fig. 6L; Supplementary Fig. S6D). Similar results were obtained by knocking down SPG21 in HepG2-HBx cells (Fig. 6M; Supplementary Fig. S6E). These results indicated that HBx could activate the JNK pathway through miR-128-3p/SPG21 to promote inflammation-mediated hepatocarcinogenesis.

Studies have shown that HBx can promote the progression of HCC by regulating various types of ion channels [35, 36]. It would be interesting to investigate whether HBx can regulate ion channels through miR-128-3p/SPG21 to promote the progression of HCC.

Several ion channels were screened by qRT-PCR and the results showed that TRPM7, a membrane protein that can promote calcium influx, was highly expressed in HepG2.2.15 cells compared to HepG2 cells (Fig. 7A). Bioinformatics analysis showed that TRPM7 was highly expressed in a variety of tumors, including LIHC (Supplementary Fig. S9A; Fig. 7B). We further validated the bioinformatics results by examining 84 pairs of clinical HCC samples using western blotting and IHC assays, which showed that TRPM7 was highly expressed in HCC tissues (Fig. 7C and D). TIMER2.0 database prediction revealed a significant positive correlation between the expression of TRPM7 and SPG21 (Fig. 7E). Data from clinical samples also confirmed this correlation (Fig. 7F).

Fig. 7.

HBx activates the JNK pathway by promoting TRPM7-mediated calcium influx through miR-128-3p/SPG21. A Ion channels were screened by qRT-PCR. B TRPM7 expression in HCC tissues was significantly higher than in NTs in the UALCAN database. C and D TRPM7 expression was significantly higher in HCC tissues than in NTs, as detected by western blotting, and IHC, respectively. E TIMER2.0 database prediction showed a significant positive correlation between the levels of TRPM7 and SPG21. F Spearman’s correlation analysis showed a significant positive correlation between the protein levels of TRPM7 and SPG21. G Dual-luciferase reporter assay showed that overexpression of SPG21 in HepG2 cells promoted TRPM7 promoter activity. H Activities of serially 5’ truncated TRPM7 promoter cotransfected with SPG21 in HepG2 cells. I Western blotting showed that overexpression of SPG21 in HepG2 cells increased the protein levels of TRPM7 and p-JNK. J Western blotting showed that knockdown of TRPM7 in HepG2.2.15 cells decreased the protein levels of TRPM7 and p-JNK. K TRPM7 and p-JNK protein levels were measured by western blotting in HepG2 cells treated with 200 µmol/L 2-APB and 2 µmol/L Bradykinin. L Flow cytometry showed that overexpression of SPG21 in HepG2 cells elevated [Ca²⁺]_i. M Flow cytometry showed that knockdown of TRPM7 in HepG2.2.15 cells decreased [Ca²⁺]_i. N and O Flow cytometry showed that knockdown of TRPM7 and treatment with 200 µmol/L 2-APB in HepG2-SPG21 cells decreased [Ca²⁺]_i. P and Q Flow cytometry showed that treatment with miR-128-3p-mimic and knockdown of SPG21 in HepG2-HBx cells decreased [Ca²⁺]_i. *P < 0.05, **P < 0.01, ***P < 0.001

QRT-PCR and dual-luciferase reporter assays showed that HBx and SPG21 promoted TRPM7 expression due to elevated TRPM7 promoter activity (Supplementary Fig. S9B and C; Fig. 7G). To determine the shortest sequence essential for transcriptional regulation of TRPM7, we examined the activities of seven promoters with 5’ truncated sequences (-1300 to + 100, -1100 to + 100, -900 to + 100, -700 to + 100, -500 to + 100, -300 to + 100, -100 to + 100). The results indicated that the TRPM7 promoter ranging from − 700 to -500 was the shortest sequence essential for promoter activity (Fig. 7H). Conversely, knockdown of SPG21 significantly inhibited TRPM7 expression (Supplementary Fig. S9D). Overexpression of SPG21 elevated TRPM7 and p-JNK protein levels (Fig. 7I), while knockdown of TRPM7 significantly inhibited p-JNK expression (Supplementary Fig. S9E; Fig. 7J). Bradykinin, a TRPM7 agonist, promoted TRPM7 and p-JNK expression, while 2-APB, acting as a TRPM7 inhibitor, decreased TRPM7 and p-JNK expression (Fig. 7K).

Flow cytometry and calcium imaging showed that overexpression of SPG21 in HepG2 cells elevated [Ca²⁺]_i (Fig. 7L; Supplementary Fig. S10A), which was reversed by knockdown of TRPM7 (Fig. 7N; Supplementary Fig. S10E) and addition of 2-APB (Fig. 7O; Supplementary Fig. S10F). Knockdown of TRPM7 (Fig. 7M; Supplementary Fig. S10B) and SPG21 (Supplementary Fig. S10C and D) in HepG2.2.15 cells decreased [Ca²⁺]_i. Furthermore, we found that overexpression of HBx in HepG2 cells with elevated [Ca²⁺]_i was significantly reversed by the addition of miR-128-3p-mimic (Fig. 7P; Supplementary Fig. S10G), and knockdown of SPG21 showed similar results (Fig. 7Q; Supplementary Fig. S10H).

These results suggested that HBx might enhance the expression of TRPM7 via miR-128-3p/SPG21, consequently facilitating increased calcium influx and activating the JNK pathway.

SPG21 participates in HCC chemoresistance induced by HBx through the JNK pathway via miR-128-3p

Doxorubicin is a chemotherapeutic drug used to treat various cancers by inducing cell death and growth arrest [37]. We used doxorubicin in the concentration range of 0-4.8 µmol/L to detect changes in cell proliferation rate. CCK-8 assay showed that doxorubicin significantly reduced the viability of HepG2 and HepG2.2.15 cells in a concentration-dependent manner, but HepG2 cells exhibited significantly higher sensitivity to doxorubicin than HepG2.2.15 cells further suggested a possible role of HBV in doxorubicin resistance of HCC cells (Supplementary Fig. S11A and B). Flow cytometry also showed that the apoptosis rate was significantly higher in HepG2 cells than in HepG2.2.15 cells at the same doxorubicin concentration, and interestingly overexpression of HBx in HepG2 cells significantly inhibited apoptosis (Fig. 8A). These results suggested that HBx could prevent doxorubicin-induced apoptosis.

Fig. 8.

HBx inhibits doxorubicin-induced apoptosis by activating the JNK pathway through miR-128-3p/SPG21. A Flow cytometry showed the apoptosis rate of HepG2.2.15 cells, HepG2 cells, and HepG2-HBx cells after treatment with 0.6 µmol/L doxorubicin. B Flow cytometry showed that treatment with doxorubicin increased apoptosis in HepG2.2.15 cells, while overexpression of SPG21 inhibited doxorubicin-induced apoptosis. C Western blotting showed that treatment of HepG2.2.15 cells with doxorubicin inhibited the expression of SPG21, p-JNK, and the downstream proteins, which were elevated after overexpression of SPG21. D Flow cytometry showed that the addition of SP600125 increased the apoptosis of HepG2-SPG21 cells treated with doxorubicin. E Western blotting showed that SP600125 inhibited the expression levels of p-JNK and its downstream proteins in HepG2-SPG21 cells treated with doxorubicin. F and G Flow cytometry showed that the addition of miR-128-3p-mimic and knockdown of SPG21 increased the apoptosis of HepG2-HBx cells treated with doxorubicin. H ELISA showed that the addition of SP600125 in HepG2-SPG21 cells treated with doxorubicin suppressed the expression levels of IL-6 and TNF-α. I and J ELISA showed that the addition of miR-128-3p-mimic and knockdown of SPG21 in HepG2-HBx cells treated with doxorubicin suppressed the expression levels of IL-6 and TNF-α. K Xenograft mouse model showed that Bel-7402-HBx cells treated with doxorubicin significantly increased tumor volume compared to the control, whereas the addition of miR-128-3p-mimic reduced tumor volume. Arrows indicate the xenograft tumors. L H&E staining of tumor samples. M-O TUNEL and PCNA staining showed that the reduction in tumor volume was due to increased apoptosis and decreased proliferation. *P < 0.05, **P < 0.01, ***P < 0.001. Dox, doxorubicin

We further investigated whether HBx can inhibit doxorubicin-induced apoptosis by activating the JNK pathway through miR-128-3p/SPG21. QRT-PCR revealed that the expression of SPG21 was decreased after doxorubicin treatment (Supplementary Fig. S11D). Flow cytometry results showed an increase in apoptosis after doxorubicin treatment, which was reversed by the addition of SPG21 (Fig. 8B). Interestingly, the protein levels of SPG21, p-JNK as well as MYC and Bcl-xL, the downstream proteins of the pathway, were also suppressed after doxorubicin treatment, and they could be restored after overexpression of SPG21 (Fig. 8C). The inhibitory effect of SPG21 on apoptosis after treatment with doxorubicin could be reversed by SP600125 (Fig. 8D). Western blotting and ELISA results also showed that SPG21 counteracted the inhibitory effect of doxorubicin on oncoproteins and inflammatory mediators by activating the JNK pathway (Fig. 8E and H). Furthermore, miR-128-3p expression was elevated after doxorubicin treatment (Supplementary Fig. S11C), and the inhibitory effect of HBx on doxorubicin-induced apoptosis could be reversed by miR-128-3p-mimic (Fig. 8F). ELISA results showed that the inhibitory effect of HBx on inhibition of inflammatory mediators by doxorubicin could be reversed by miR-128-3p (Fig. 8I). Flow cytometry and ELISA showed that knockdown of SPG21 in HepG2-HBx cells after doxorubicin treatment had a similar effect to miR-128-3p (Fig. 8G and J). The xenograft mouse model results indicated that miR-128-3p could prevent HBx-induced doxorubicin resistance (Fig. 8K). H&E, TUNEL, and PCNA staining further confirmed the results of the in vivo experiment (Fig. 8L-O). These results revealed that HBx prevented doxorubicin-induced apoptosis by activating the JNK pathway through miR-128-3p/SPG21.

Discussion

HCC is one of the most common malignant tumors worldwide, and various chemotherapeutic drugs have been explored for its treatment, including doxorubicin, sorafenib, and others. However, the frequent emergence of resistance to these drugs, such as doxorubicin and sorafenib, limits their effectiveness [3].

The mechanisms underlying HCC chemoresistance are diverse and include aberrant regulation of alternative splicing, inflammation-activated pathways, and HBV infection, among others. For instance, inhibition of SRSF2 expression promotes splicing of KIAA0101 transcript variant 1, inducing sorafenib resistance by reducing ferroptosis [38]. KIAA0101 transcript variant 1 can also prevent doxorubicin-induced apoptosis by inhibiting the activation of p53 [39]. CKLF1 promotes inflammation-mediated hepatocarcinogenesis and doxorubicin resistance through the IL-6/STAT3 pathway [23], while the human endogenous retrovirus W family envelope facilitates HCC progression and doxorubicin resistance via the inflammation-activated MEK/ERK pathway [40].

HBV promotes sorafenib resistance by decreasing miR-193b expression, elevating the expression of the anti-apoptotic protein Mcl-1 [41]. The viral protein HBx facilitates sorafenib resistance in HCC by increasing the expression of TRERNA1 [42]. Additionally, large hepatitis B surface antigen enhances sorafenib resistance in HCC by decreasing mitophagy [43], and hepatitis B core antigen facilitates HCC cell resistance to doxorubicin through exosomal miR-135a-5p by targeting vesicle-associated membrane protein 2 [18].

In this study, we observed that SPG21 expression was higher in HCC tissues compared to NTs, and the increased SPG21 level was associated with poor overall survival. In contrast, miR-128-3p, which targeted SPG21, functioned as a tumor suppressor in HCC. The miR-128-3p/SPG21 axis played a role in HBx-induced doxorubicin resistance through the activation of the JNK pathway.

Most patients with HCC are already in an advanced stage at the initial diagnosis and have lost the opportunity for surgery, resulting in a high mortality rate. Currently, AFP is the predominant marker utilized in the clinical diagnosis of HCC. However, the sensitivity of AFP ranges from 41 to 65%, with a specificity between 80% and 94%, leading to frequent missed diagnoses, particularly in the early stages of HCC [44]. Hence, it is urgent to explore HCC biomarkers with better sensitivity to improve the survival rate of patients. Our research group has recently identified PCLAF [45] and exosomes [46] as promising biomarkers for HCC diagnosis and prognosis. In the current study, we found a strong association between SPG21 and HCC, suggesting that SPG21 could be a valuable biomarker for diagnosing and prognosticating HCC.

SPG21 is involved in suppressing T cell activation through its non-catalytic alpha/beta hydrolase fold domain, which binds to the hydrophobic C-terminal amino acids of CD4 [14]. A frameshift mutation caused by a single base-pair insertion (601insA) in the SPG21 gene leads to a premature stop codon, resulting in a nonfunctional protein [13]. Targeted disruption of SPG21 in mice impairs hind limb function and increases axonal branching in cerebral cortical neurons [47]. However, the relationship between SPG21 and tumors, including HCC, has not been reported so far. Our findings revealed that SPG21 promoted HCC tumorigenesis and invasion, marking a novel insight into the pathogenic mechanisms of this condition.

HBx acts as a multifunctional trans-activator that can stimulate the expression of diverse proto-oncogenes through direct or indirect protein-protein interactions [48]. Our research indicated that HBx not only elevated the expression of SPG21 but also had a direct interaction with it, underscoring a significant pathway in the modulation of gene expression within the context of HCC.

While miRNAs are known to influence the expression of target genes, there has been a lack of evidence regarding miRNAs that regulate SPG21. Our study identified SPG21 as a direct target of miR-128-3p, marking an important discovery in the understanding of SPG21’s regulatory mechanisms and its role in disease processes.

The miR-128-3p expression status in HCC is controversial. MiR-128-3p is reported to be highly expressed in HCC tissues and promotes HCC progression by targeting CYP2C9 [49]. In contrast, it is mostly documented to be lowly expressed in HCC tissues and acts as a tumor suppressor in HCC progression by targeting PIK3R1 [19] and MICA [20]. Our results showed that the expression of miR-128-3p was downregulated in HCC tissues and significantly negatively correlated with SPG21 levels, suggesting that miR-128-3p may act as a tumor suppressor by targeting SPG21 to inhibit the progression of HCC. The reported high expression of miR-128-3p in HCC tissues may be due to the insufficient number of clinical samples. Only 28 tumor and 27 non-tumor samples are analyzed. In contrast, our data were obtained from 84 pairs of HCC tissues and NTs.

HBx is regarded as an important transcriptional regulator that modulates the expression of a variety of miRNAs [50]. Furthermore, HBx can also interact with transcription factors to regulate the expression of miRNAs and promote HCC progression [51]. Our results demonstrated that HBx interacted with the transcription factor Snail to suppress miR-128-3p expression. This suppression was achieved by inhibiting the promoter activity of miR-128-3p, highlighting a novel mechanism through which HBx contributed to the molecular pathology of HCC.

HBx promotes the malignant progression of HCC by activating the JNK pathway [52]. Existing literature has shown that the JNK pathway promotes inflammation-mediated hepatocarcinogenesis [31, 32]. However, there is controversy about whether the JNK pathway inhibits or promotes the progression of HCC. Activation of the JNK pathway suppresses the progression of HCC by promoting apoptosis and inhibiting proliferation [53, 54]. This may be attributed to the fact that in the initiation phase of HCC, JNK1 promotes apoptosis and necrosis and reduces cell proliferation by decreasing Cyclin D1 and VEGF expression [55]. In contrast, at late stages, JNK1 leads to compensatory proliferation by inhibiting p21 and elevating MYC expression [56]. Apoptosis is known to be anti-inflammatory, and excessive apoptosis leads to secondary necrosis, providing the impetus to promote inflammation [55, 57]. On the other hand, compensatory proliferation is essential for tumor promotion, allowing the initiated hepatocytes to enter the cell cycle [58]. Our study elucidated that HBx activated the JNK pathway through miR-128-3p/SPG21, thereby promoting inflammation-mediated hepatocarcinogenesis. This highlighted a complex interplay between the JNK pathway’s roles across different stages of HCC development and introduced new insights into the mechanisms by which HBx contributed to the disease progression.

Studies have demonstrated that overexpression of TRPM7 activates the JNK pathway [59], whereas silencing of TRPM7 in astrocytes inhibits the JNK pathway [60]. TRPM7 is a non-selective but calcium-conducting cation channel with intrinsic kinase activity that is highly expressed in cancers such as pancreatic, breast, and head/neck cancers [61]. However, there are only three studies on the relationship between TRPM7 and HCC. The first reports that Bradykinin promotes HCC cell migration via TRPM7 [62]. The second reports that blockade of TRPM7 inhibits MRTF/SRF target gene expression and results in the growth arrest of HCC cells and HCC xenografts through oncogene-induced senescence, which is required for TRPM7-mediated magnesium influx and TRPM7 kinase activity [63]. The third reports that carvacrol combined with sorafenib can overcome sorafenib resistance and cardiotoxicity in HCC by inhibiting TRPM7 [64]. Our findings contributed to this emerging field by demonstrating that HBx enhanced TRPM7 expression through miR-128-3p/SPG21. This upregulation stimulated calcium influx, thereby activating the JNK pathway. This insight added a new dimension to our understanding of TRPM7’s role in HCC progression and underscored the complex interplay of molecular mechanisms involved.

Chemotherapy is the main treatment for advanced cancer patients, but chemoresistance occurs frequently. Emerging evidence suggests that HBx promotes chemoresistance of HCC. HBx enhances the resistance of OV6⁺ liver cancer stem-like cells to pirarubicin, oxaliplatin, and hydroxycamptothecin via upregulation of MDM2 [7], causes resistance to 5-fluorouracil by downregulating SHIP2 [8], and increases resistance to 5-fluorouracil, cisplatin, and pharmorubicin through a miR-5188-regulated positive feedback loop [9]. Our findings indicated that HBx promoted resistance of HCC cells to doxorubicin.

The trans-Golgi network plays an important role in resistance to SN-38 in clinical colon cancer [65], and cisplatin resistance in human ovarian carcinoma cells [66]. In our study, SPG21, localized in the trans-Golgi network, promoted doxorubicin resistance of HCC. MiR-128-3p increases the chemosensitivity of temozolomide in glioblastoma [67] and enhances the sensitivity of HCC cells to lenvatinib [68], yet promotes resistance of NSCLC cells to cisplatin, gemcitabine, and paclitaxel [69]. Our data indicated that miR-128-3p enhanced the chemosensitivity of HCC cells to doxorubicin by targeting SPG21. This may be explained by the fact that miR-128-3p regulates different target genes and thus exerts distinct roles in the chemoresistance of tumors.

JNK pathway reduces apoptosis and enhances cisplatin resistance by decreasing p53 expression in NSCLC cells [70]. However, inhibition of the JNK pathway in breast cancer impairs proapoptotic signaling pathway and enhances resistance to anthracyclines [71]. Our data suggested that HBx induced doxorubicin resistance in HCC by activating the JNK pathway through miR-128-3p/SPG21.

In summary, our results showed that high expression of SPG21 in HCC was correlated with advanced TNM stage and poor prognosis. Both in vitro and in vivo results indicated the role of SPG21 as a potential oncogene in HCC. Conversely, miR-128-3p, which targeted SPG21, appeared to function as a tumor suppressor. We also discovered that HBx could amplify TRPM7-mediated calcium influx via miR-128-3p/SPG21, leading to the activation of the JNK pathway and a subsequent reduction in the effectiveness doxorubicin-induced apoptosis (Fig. 9). Furthermore, our findings emphasized the significance of miR-128-3p/SPG21/TRPM7/JNK signaling pathway in the context of HBx-induced hepatocarcinogenesis and chemoresistance. This insight suggested the potential of targeting components within this pathway as a strategy for managing and treating HBV-related HCC.

Fig. 9.

Schema illustrates the mechanism by which SPG21 amplifies HBx-induced carcinogenesis and chemoresistance. HBx interacts with the transcription factor Snail to down-regulate the expression of miR-128-3p and up-regulate the expression of SPG21, the target gene of miR-128-3p. HBx enhances TRPM7-mediated calcium influx through miR-128-3p/SPG21, thereby activating the JNK pathway and increasing the downstream proteins, such as MYC, Bcl-xL, and Cyclin D1. These proteins participate in hepatocarcinogenesis by promoting tumorigenesis and invasion. SPG21 inhibits doxorubicin-induced apoptosis

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

FZ designed the study. PZ performed all the experiments and drafted the manuscript. WY, LJL, QJY, and ZL participated in the analysis of the data. PZ, XBC, XCW, and SD participated in the collection of clinical samples. FZ supervised the study and revised the manuscript. All authors read and approved the final manuscript.

Funding

This work was supported by the National Natural Science Foundation of China (No. 82272321), Fundamental Research Funds for the Central Universities (2042023kf0230), and Translational Medicine and Interdisciplinary Research Joint Fund of Zhongnan Hospital of Wuhan University (Grant No. ZNLH201902).

Data availability

All data supporting the findings of this study are available within the paper and its Supplementary Information.

Declarations

Ethical approval

The animal experiments were approved by the Animal Ethics Committee of Wuhan University, Wuhan University Center for Animal Experiment/A3 Laboratory. This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of Wuhan University, School of Basic Medical Sciences (Wuhan, China). Written informed consent was obtained from each patient.

Competing interests

The authors declare no competing interests.