Bioinformatic characterization of SLC25A39 across cancers with validation in hepatocellular carcinoma

Abstract

Introduction

Solute carrier family 25 member 39 (SLC25A39) is crucial for regulating mitochondrial oxidative metabolism and homeostasis, linking to various physiological and pathological processes. Its expression landscape and mechanistic roles across cancer types remain largely unexplored. This study systematically profiled SLC25A39 in pan-cancer and assessed its prognostic relevance, molecular associations, and therapeutic potential.

Methods

Multi-omics database resources were utilized to perform an integrated analysis of SLC25A39, including its expression profiles, prognostic value, molecular mechanisms, and associations with the immune microenvironment across diverse tumors. The oncogenic role of SLC25A39 was confirmed through in vitro gene knockdown in hepatocellular carcinoma.

Results

SLC25A39 was predominantly over expressed in nearly all types of cancer. SLC25A39 serves as an unfavorable prognosis marker. Its expression was primarily linked to immune cell infiltration, the cancer immunity cycle, major histocompatibility complex, immune checkpoints, tumor mutation burden, microsatellite instability, and RNA modifications like m1A, m5C, m6A, as well as DNA methylation sites. Functional enrichment analysis reveals that SLC25A39 is commonly associated with immune-related pathways and cell cycle-related pathways in the pan-cancer dataset. Knockdown of SLC25A39 in Huh-7 and HepG2 cells significantly inhibited hepatocellular carcinoma cell proliferation.

Conclusions

SLC25A39 induces pro-tumorigenic effects across various cancer types. It serves as a potential biomarker.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12672-026-05621-w.

Introduction

Malignant tumors are among the leading contributors to the global disease burden, representing an urgent public health challenge [1]. The pathogenesis of these diseases is highly complex, arising from the interplay between various genetic susceptibilities and environmental exposures [2]. Although traditional methods like surgery, radiotherapy, and chemotherapy are improving. However, its clinical application remains limited. Drug resistance, disease recurrence, and adverse reactions to new drugs are some of the challenges [3, 4]. We need to identify new drug targets and develop reliable biomarkers for diagnosis and prognosis as soon as possible.

Breakthroughs in high-throughput sequencing technologies have fundamentally reshaped the landscape of cancer research. Unlike traditional approaches that focus on a single cancer type, the pan-cancer analysis strategy integrates multi-cancer datasets to systematically investigate shared molecular mechanisms and cancer type-specific differences. This approach not only enhances the understanding of fundamental biological processes in tumors but also provides critical scientific evidence for the development of broad-spectrum therapeutic interventions [5, 6].

Mitochondria, as the central hub of energy metabolism and a regulator of redox homeostasis in eukaryotic cells, have been widely implicated in tumorigenesis and other pathological conditions. The SLC25A39 gene, located on chromosome 17, encodes an inner mitochondrial membrane transporter belonging to the SLC25 solute carrier family. Recent studies have shown that SLC25A39 is involved in cell survival by regulating mitochondrial oxidative metabolism and maintaining mitochondrial homeostasis [7]. Glutathione (GSH) serves as the primary intracellular antioxidant, and upon entering the mitochondrial matrix, it transforms into mitochondrial GSH (mGSH), directly participating in mitochondrial oxidation [8]. The loss of SLC25A39 leads to a significant decrease in mGSH, inducing mitochondrial stress and metabolic disruptions, and impacting the cell death mechanism [9, 10].

During tumor progression, mitochondria act as a central hub influencing oxidative stress responses, energy supply, signal transduction, and cell fate decisions, thereby driving malignant phenotypes. Imbalance in redox homeostasis and remodeling of reactive oxygen species (ROS) signaling represent core mechanisms underlying the pro-tumorigenic effects of mitochondria [11]. The phenomenon of metabolic reprogramming of mitochondria that endows cancer cells with proliferative and metastatic advantages has been widely reported [12]. Although SLC25A39 has been identified as a critical regulator of mGSH transport and mitochondrial oxidative metabolism, its functional role in cancer has not yet been systematically characterized. Current evidence indicates that this gene is abnormally overexpressed in colorectal cancer, where it promotes tumor cell proliferation, migration, and inhibits apoptosis [13]. however, its comprehensive functional mechanisms in a pan-cancer context remain to be elucidated.

Materials and methods

Data collection

Multi-omics datasets were integrated for the pan-cancer analysis. Specifically, pan-cancer transcriptomic expression profiles and the corresponding clinical annotations were obtained from The Cancer Genome Atlas (TCGA) database. Molecular feature data, including DNA methylation profiles, copy number variations, transcript isoform expression levels, microsatellite instability (MSI) scores, and tumor mutation burden (TMB) scores, as well as survival outcome information, were retrieved from the UCSC Xena platform.

To provide normal tissue controls, gene expression data from supplementary normal samples were obtained from the Genotype-Tissue Expression (GTEx) project. In addition, bulk RNA-seq and single-cell RNA sequencing (scRNA-seq) datasets from the Gene Expression Omnibus (GEO) database were incorporated to further expand the coverage of the pan-cancer cohorts. Detailed information for all datasets used in this study is summarized in Supplementary Table 1. Clinical trial number: not applicable.

Prognostic and diagnostic analyses

The expression levels of SLC25A39 in pan-cancer tissues and corresponding normal tissues were evaluated using R software. Univariate Cox regression analysis and Kaplan-Meier (KM) survival analysis were conducted with the “survival” package to assess the association between SLC25A39 expression and prognostic outcomes, including overall survival (OS), disease-specific survival (DSS), disease-free interval (DFI), and progression-free interval (PFI). Subsequently, SLC25A39 expression levels were analyzed across different clinical conditions, including patient age, T stage, M stage, N stage, and overall clinical stage. Receiver operating characteristic (ROC) curves were generated using the “timeROC” package to evaluate the sensitivity and accuracy of SLC25A39 expression for cancer diagnosis in a pan-cancer context.

Single-cell expression distribution

To characterize the cellular expression heterogeneity of SLC25A39 across cancers, a systematic analysis was performed on 14 single-cell RNA sequencing (scRNA-seq) datasets. Data preprocessing was conducted using the Seurat package (v5.2.1) with strict quality control criteria: cells with more than 300 detected genes, a total unique molecular identifier (UMI) count below 30,000, and a mitochondrial gene proportion of less than 20% were retained. The standard preprocessing workflow included normalization of expression data (NormalizeData), identification of highly variable genes (FindVariableFeatures), data scaling (ScaleData), and principal component analysis (RunPCA). Integration of samples was performed using the IntegrateLayers function, and batch effects were corrected with the Harmony algorithm. Nonlinear dimensionality reduction for visualization was achieved via Uniform Manifold Approximation and Projection (UMAP). Cell community identification was performed by constructing a neighbor graph (FindNeighbors) followed by Louvain clustering (FindClusters). Finally, clusters were annotated manually according to canonical cell type marker expression profiles.

Gene set enrichment analysis

To elucidate the biological functions of SLC25A39 in pan-cancer, each cancer type cohort was divided into high- and low-expression groups based on the median expression level of the gene. Following differential mRNA expression analysis between the two groups, gene set enrichment analysis (GSEA) was performed using the clusterProfiler package. Both the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway database and the Hallmark gene sets were employed as reference backgrounds to evaluate pathway enrichment. The normalized enrichment score (NES) and the false discovery rate (FDR) were calculated for each enrichment result.

Tumor immune microenvironment and genetic alteration analyses

To investigate the associations between SLC25A39 expression and tumor immune characteristics, multi-dimensional immunogenomic data were integrated. Pan-cancer immune status data from the TIP database were used to compare anti-tumor immune activity between the high- and low-expression groups of SLC25A39. The abundance of immune cell infiltration in each sample was quantified using the single-sample gene set enrichment analysis (ssGSEA) and ESTIMATE algorithms. Tumor mutation burden (TMB), microsatellite instability (MSI) scores, and whole-genome DNA methylation data for all pan-cancer cohorts were obtained from the UCSC Xena platform and standardized.In addition, transcriptomic data for key immune-related molecules—including major histocompatibility complex (MHC) genes, immune checkpoint molecules, and RNA modification-related regulators (covering m1A, m5C, and m6A pathways)—were systematically collected. Spearman’s rank correlation analysis was conducted to comprehensively examine the associations between SLC25A39 expression, tumor immune microenvironment characteristics, and indicators of genomic stability.

Potential therapeutic drug prediction

For each cancer type, the top 100 most significantly upregulated and downregulated mRNA genes between the high- and low-expression SLC25A39 groups were identified. These differential genes and their corresponding fold changes were uploaded to the SPIED3 platform (http://92.205.225.222/HGNC-SPIED3-QF.py), using compound expression profiles from 1,309 small molecules in the Connectivity Map (CMAP) database as the reference dataset. Compounds showing a negative correlation with the expression trend of the differential genes were considered potential therapeutic agents capable of reversing the observed expression patterns. Furthermore, drug activity scores and RNA-seq expression data for the NCI-60 cancer cell lines were downloaded from the CellMiner database. Only U.S. Food and Drug Administration (FDA)-approved and clinically validated drugs were included. The correlations between SLC25A39 expression levels and drug activity scores were evaluated, with higher scores indicating greater drug sensitivity.

Cell culture and establishment of SLC25A39 knockdown cell lines

Human hepatocellular carcinoma cell lines (Huh-7 and HepG2) and human embryonic kidney cells (HEK293T) were obtained from the Cell Bank of the Chinese Academy of Sciences. All cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin at 37 °C in a humidified incubator containing 5% CO₂. shRNA plasmids targeting SLC25A39 were purchased from Thermo Fisher Scientific and transfected into HEK293T cells to produce lentiviral particles using FuGENE 6 (Promega, Madison, WI, USA) and jetPRIME (Polyplus-transfection, Illkirch, France). Lentiviruses were used to infect Huh-7 and HepG2 cells to establish stable SLC25A39 knockdown cell lines. The shRNA sequences were as follows: sh-SLC25A39-1, GUGGUUACCUCUCUCUUCATT; sh-SLC25A39-2, UCUACUUCACUGCCUAUGATT.

Western blotting and quantitative real-time PCR

Total proteins were extracted from cells using RIPA lysis buffer (Solarbio, China), and protein concentrations were determined with a BCA protein assay kit (Solarbio, China). SLC25A39 protein levels were detected by western blotting using a primary antibody against SLC25A39 (1:1000, A0993, ABclonal, China).

Total RNA was extracted with TRIzol reagent (Tiangen, China) and quantified. Complementary DNA (cDNA) was synthesized using the PrimeScript™ RT reagent kit (Takara, Japan). Quantitative PCR reactions were performed with SYBR Green reagent (Takara, Japan) on a real-time PCR thermocycler (Agilent, USA). β-actin was used as the internal control, and the relative expression level of SLC25A39 was calculated using the 2⁻^ΔΔCt method. The primer sequences were as follows: SLC25A39: forward AGGGACTCTTTGCAGGCTTC, reverse GCCGAACTCATAGGTGCTGA; β-actin: forward CATGTACGTTGCTATCCAGGC, reverse CTCCTTAATGTCACGCACGAT.

Cell proliferation assay

Cell viability was assessed using a Cell Counting Kit-8 (CCK-8) assay (Beyotime, China). Cells were seeded into 96-well plates and incubated for 24 h, 48 h, 72 h, and 96 h, followed by the addition of CCK-8 reagent. Absorbance was measured at 450 nm with a microplate reader.

Colony formation assay

Approximately 1,000 cells were seeded in 6-well plates and cultured for 14 days. Colonies were fixed with methanol and stained with crystal violet solution.

Wound healing assay

Cells were seeded in 6-well plates until they reached approximately 80% confluence. A scratch was created on the cell monolayer using a pipette tip. Cells were washed twice with PBS and cultured in serum-free medium. Images of wound closure were captured at 0 h, 24 h, and 48 h under a microscope.

Apoptosis assay

Cells were seeded in 6-well plates, incubated for 48 h, and harvested. Apoptotic cells were stained with an Annexin V-FITC apoptosis detection kit (Beyotime, China) according to the manufacturer’s instructions and analyzed by flow cytometry.

Statistical analysis

Spearman rank correlation was used to evaluate the correlations between two variables. The Wilcoxon rank-sum test was used to compare two groups, while the Kruskal-Wallis test was applied for comparisons among more than two groups. Kaplan-Meier survival curves and Cox proportional hazards regression models were used for survival analyses. Statistical analyses and figure generation were performed using R software (version 4.3.2). A p-value < 0.05 was considered statistically significant. *P < 0.05; **P < 0.01; ***P < 0.001.

Results

Aberrant expression of SLC25A39 in pan-cancer

The expression differences of SLC25A39 between tumor tissues and normal tissues were comprehensively analyzed using transcriptomic data from TCGA and GTEx projects. Overall, SLC25A39 was found to be dysregulated in the majority of tumor types, with a predominant trend of upregulation. Significant overexpression was observed in bladder urothelial carcinoma (BLCA), breast invasive carcinoma (BRCA), cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC), cholangiocarcinoma (CHOL), colon adenocarcinoma (COAD), esophageal carcinoma (ESCA), glioblastoma multiforme (GBM), head and neck squamous cell carcinoma (HNSC), kidney chromophobe (KICH), kidney renal papillary cell carcinoma (KIRP), liver hepatocellular carcinoma (LIHC), lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LUSC), prostate adenocarcinoma (PRAD), rectum adenocarcinoma (READ), stomach adenocarcinoma (STAD), and uterine corpus endometrial carcinoma (UCEC). In contrast, SLC25A39 was significantly downregulated in kidney renal clear cell carcinoma (KIRC) (Fig. 1A-B). Validation using 10 cancer datasets from the GEO yielded consistent results, demonstrating upregulation of SLC25A39 in 9 cancer types and downregulation specifically in KIRC (Fig. 1C).

Fig. 1.

Aberrant expression of SLC25A39 in pan-cancer. A Expression analysis of SLC25A39 across pan-cancer cohorts based on TCGA data. B Expression analysis integrating TCGA and GTEx pan-cancer datasets. C Expression analysis across pan-cancer cohorts from the GEO database. D Structural diagram of SLC25A39 transcript isoforms and their expression profiles in pan-cancer analysis

Given the widespread dysregulation of SLC25A39 in pan-cancer contexts, the structural characteristics of SLC25A39 transcript isoforms and their expression differences between normal and tumor tissues were further investigated. The results revealed considerable heterogeneity in transcript expression across multiple cancers. Notably, non-coding transcript isoforms of SLC25A39 were markedly upregulated in several cancer types, including LIHC, CHOL, LUSC, and LUAD, suggesting that SLC25A39 may also exert biological functions in the form of non-coding RNAs (Fig. 1D).

Expression distribution of SLC25A39 at the single-cell level

Following the preliminary analysis of SLC25A39 expression in pan-cancer, it was observed that the gene was dysregulated in most tumor types, predominantly exhibiting upregulation. To further elucidate the cell type-specific expression patterns of SLC25A39 in various cancers, single-cell transcriptomic datasets were analyzed. The results demonstrated that SLC25A39 was broadly distributed among diverse cellular populations across multiple cancer types; however, its expression patterns exhibited cell type-specificity (Supplementary Fig. 1). For instance, in BRCA, CHOL, KIRC, and LUAD, SLC25A39 expression was primarily enriched in immune cell clusters. In contrast, in CRC, ESCA, OV, and PRAD, the gene was mainly expressed in epithelial and endothelial cell populations. Such differential expression patterns suggest that SLC25A39 may be functionally associated with distinct immune-related processes within the tumor microenvironment.

Prognostic and diagnostic value of SLC25A39 in pan-cancer

To investigate the prognostic significance of SLC25A39 across different cancer types, univariate Cox proportional hazards regression analysis and KM survival analysis were performed to assess its associations with four prognostic outcomes: OS, DSS, DFI, and PFI. The results indicated that SLC25A39 expression was significantly associated with poor prognosis in ACC, CESC, hHNSC, KIRC, KIRP, LGG, LIHC, LUAD, MESO, PRAD, READ, SARC, SKCM, TGCT, UCS, and UVM. Notably, in ACC, LGG, and LIHC, SLC25A39 expression demonstrated significant correlations with all four prognostic outcomes (Fig. 2A). In addition, variables that were significant in the univariate Cox analysis were further included in the multivariate Cox analysis to assess whether SLC25A39 is an independent prognostic factor (Supplementary Table 2).

Fig. 2.

Prognostic and diagnostic value of SLC25A39 in pan-cancer. A Univariate Cox regression and KM survival analyses evaluating correlations between SLC25A39 expression and four prognostic outcomes-OS, DSS, DFI, and PFI-across multiple cancer types. B Association of SLC25A39 expression with clinicopathological characteristics including age, TNM stage, and overall tumor stage. C Diagnostic ROC curves for SLC25A39 in 17 TCGA cancer types exhibiting significant expression differences

To further elucidate the relationship between SLC25A39 expression and clinicopathological features, expression levels were compared across different age groups, TNM classifications, and tumor stages. The results revealed that tumor stage exhibited broader correlations with SLC25A39 in multiple cancer types. Significant upregulation of SLC25A39 was observed in higher-stage tumors in ACC, BLCA, HNSC, KIRP, and LUSC, whereas in THCA and UCS the opposite trend was observed (Fig. 2B).

We plotted diagnostic ROC curves for 17 TCGA cancer types with differential expression. The results showed that SLC25A39 had high diagnostic performance in some cancer types (Fig. 2C).

Gene set enrichment analysis

To further investigate the potential biological functions through which SLC25A39 may be broadly involved in tumorigenesis and cancer progression, samples from each cancer type were stratified into high- and low-expression groups based on SLC25A39 expression levels. Differentially expressed genes were then identified, followed by GSEA. In the KEGG pathway analysis, multiple cancer types were enriched in tumor-related signaling pathways, including cytokine-cytokine receptor interaction, calcium signaling pathway, cell adhesion molecules (CAMs), and cAMP signaling pathway. In addition, several immune-related pathways were commonly downregulated in multiple cancers, such as primary immunodeficiency, antigen processing and presentation, Th17 cell differentiation, natural killer cell-mediated cytotoxicity, and chemokine signaling pathway, suggesting that SLC25A39 may be involved in tumor immune evasion (Fig. 3A). In the Hallmark pathway analysis, cell cycle-related pathways, including G2M checkpoint and E2F targets, were significantly activated in multiple cancer types. Pathways associated with genomic stability, such as DNA repair and mitotic spindle, were also upregulated, potentially contributing to uncontrolled proliferation of tumor cells. Furthermore, several cancer-related pathways, including MYC targets, inflammatory response, and epithelial-mesenchymal transition (EMT), were significantly enriched (Fig. 3B). It is noteworthy that certain cancer types exhibited distinct heterogeneity. for example, in LGG, enrichment patterns in several pathways were opposite to those observed in most other cancers.

Fig. 3.

Gene set enrichment analysis (GSEA) of SLC25A39 in pan-cancer. KEGG pathway enrichment (A) and hallmark gene set enrichment (B) results for high-versus low-expressing SLC25A39 groups across multiple cancer types

Immune characteristics of SLC25A39 in the tumor microenvironment across pan-cancer

The progression of a malignant tumor relies on the environment where a tumor cell resides. To further investigate SLC25A39’s influence on immune status in pan-cancer, it is essential to assess the activity scores of the cancer-immunity cycle and analyze the correlation between SLC25A39 expression and immune cell infiltration using the ESTIMATE and ssGSEA algorithms. In most cancer types, the activity of the cancer-immunity cycle showed a negative correlation with SLC25A39 expression, suggesting that high SLC25A39 expression may impair immune cell recognition and killing of tumor cells, thereby promoting immune evasion. In contrast, in LGG and LAML, high SLC25A39 expression was associated with more active anti-cancer immune responses (Fig. 4).

Fig. 4.

Relationship between SLC25A39 expression and cancer immunity cycle activity in various cancer types

Immune-related modulators are crucial in shaping the tumor environment and determining the effectiveness of cancer immunotherapy. Therefore, we also examine the relationship between SCL25A39, MHC molecules, and immune checkpoints. The results indicated significant correlations, with heterogeneous patterns across cancer types and molecular subtypes. For MHC molecules, negative correlations with SLC25A39 were observed in the majority of cancers, whereas the transporter associated with antigen processing binding protein showed the opposite trend (Fig. 5A). Similarly, most immune checkpoint genes displayed negative correlations with SLC25A39, whereas immune checkpoints such as CD276, TNFRSF18, and TNFRSF4 exhibited positive correlations (Fig. 5B).

Fig. 5.

Associations between SLC25A39 expression and immune regulatory molecules in pan-cancer. A Correlation between SLC25A39 expression and MHC molecule expression. B Correlation between SLC25A39 expression and immune checkpoint gene expression

SCL25A39 expression showed a negative association with immune and stromal scores in various cancers, according to ESTIMATE, but a positive correlation in LGG. These findings suggest that SLC25A39 may be involved in reshaping the tumor microenvironment by altering immune cell infiltration or stromal components (Supplementary Fig. 2A). Additionally, analysis of immune cell infiltration levels reveals that, apart from natural killer CD56 bright cells and T-helper 2 (Th2) cells, other immune cell populations exhibit a negative correlation between SLC25A39 and pan-cancer (Supplementary Fig. 2B).

Genetic alterations of SLC25A39 in pan-cancer

Genomic stability impacts gene expression and tumor development, and is associated with immunotherapy response. Analysis of TMB and MSI revealed that SLC25A39 expression was positively correlated with TMM and MSI in STAD, UCEC, LUSC, PRAD, BLCA, HNSC, and LIHC (Fig. 6A, B).

Fig. 6.

Genetic and epigenetic alterations associated with SLC25A39 in pan-cancer. A Association of SLC25A39 with TMB across various cancers. B Link between SLC25A39 and MSI in different cancers. C CNV of SLC25A39 in diverse cancers. D Expression of SLC25A39 related to DNA promoter methylation levels. E Expression of SLC25A39 associated with m1A, m5C, and m6A modification enzymes

To investigate potential mechanisms underlying the dysregulated expression of SLC25A39 in pan-cancer, CNV and DNA promoter methylation analyses were performed. In the CNV analysis, marked copy number amplification events were observed in multiple cancer types, which may contribute to the upregulation of SLC25A39 expression (Fig. 6C). However, copy number deletions were also identified in several cancers, such as KICH and OV, indicating that additional regulatory mechanisms, such as DNA methylation, may also affect its expression. Significant correlations between SLC25A39 expression and methylation status at promoter-associated CpG sites were observed in multiple cancer types, suggesting that DNA methylation may participate in the regulation of SLC25A39 (Fig. 6D).

Post-transcriptional regulation via RNA modification also impacts expression; abnormal expression of the enzyme responsible for RNA modification is linked to tumorigenesis and tumor progression. The correlation analysis reveals that SLC25A39 expression is significantly linked to the expression of genes associated with enzymes involved in N1-methyladenosine (m1A), 5-methylcytosine (m5C), and N6-methyladenosine (m6A) RNA modification (Fig. 6E).

Potential therapeutic drug analysis

To identify potential therapeutic agents targeting the pro-tumorigenic pathways associated with SLC25A39, the top 100 DEGs from each cancer type identified in the pan-cancer differential expression analysis were collected. These DEGs were compared with perturbation-induced gene expression profiles from the CMAP database. The top 30 negatively correlated compounds were chosen, among which canadine, quinethazone, and sulfamethoxypyridazine could potentially reverse cancer-related DEGs (Fig. 7A).

Fig. 7.

Potential therapeutic agents associated with SLC25A39 in pan-cancer. A Top 30 negatively correlated compounds identified via CMAP analysis using the top 100 DEGs from each cancer type. B Correlation between SLC25A39 expression and drug sensitivity for FDA-approved and clinically validated anti-tumor agents based on CellMiner dataset analyses

In addition, we analyzed FDA-approved and clinically validated drugs from the CellMiner database. The results showed that the sensitivity to four antitumor drugs was correlated with SLC25A39. Sensitivity to 6-thioguanine, 8-chloro-adenosine, floxuridine, and raltitrexed was positively correlated with SLC25A39, suggesting that patients with high SLC25A39 expression may be more sensitive to these drugs (Fig. 7B). Molecular docking analysis showed that the four drugs above formed a stable binding pose with SLC25A39, with binding energies of 5.2 kcal/mol, 7.0 kcal/mol, 9.4 kcal/mol, and 6.4 kcal/mol, respectively (Supplementary Fig. 3A). We further selected floxuridine to test its effects in HCC. The CCK-8 assay showed that the 24-h IC50 of floxuridine in HCC was 11.26 nM (Supplementary Fig. 3B), and floxuridine reduced the protein expression level of SLC25A39 (Supplementary Fig. 3C).

Knockdown of SLC25A39 inhibits proliferation of hepatocellular carcinoma cells

Because hepatocellular carcinoma shows a relatively high level of abnormal expression among different cancer types, and to improve the reliability of our study, we performed experimental validation in HCC cell. Initially, upregulated SLC25A39 expression in HCC was confirmed across ten independent cohorts (Fig. 8A). Stable SLC25A39 knockdown was then established in Huh-7 and HepG2 cell lines using short hairpin RNA-mediated silencing, and knockdown efficiency was verified by WB and qPCR analyses (Fig. 8B). Consistent results were obtained from CCK-8 assays, which showed decreased proliferation rates upon SLC25A39 silencing (Fig. 8C). Wound healing assays further demonstrate that the wound closure rate significantly decreased at specific time points, suggesting impaired migration ability of HCC cells (Fig. 9A). The reduction in colony formation of HCC cells was also observed following the knockdown of SLC25A39, with both the number and size of colonies being smaller compared to the control group in the colony formation assay (Fig. 9B). Finally, the knockdown of SLC25A39 resulted in significant apoptosis of HCC cells, as observed through the AV/PI staining assay (Fig. 9C).

Fig. 8.

Functional validation of SLC25A39 knockdown in HCC cells. A Validation of SLC25A39 overexpression in ten independent HCC cohorts. B Verification of SLC25A39 knockdown efficiency in Huh-7 and HepG2 cells by Western blotting and PCR. C CCK-8 assay results indicating decreased cell proliferation after SLC25A39 silencing

Fig. 9.

Functional validation of SLC25A39 knockdown in HCC cells. A Wound healing assay demonstrating impaired migratory ability in knockdown cells. B Colony formation assay results showing reduced colony number and size following SLC25A39 knockdown. C AV/PI staining showing increased apoptotic rates after SLC25A39 knockdown. D Protein expression levels of the related molecules were examined by Western blotting

To determine whether mitophagy mediates the pro-apoptotic effect caused by SLC25A39 loss, we measured the protein levels of apoptosis- and mitophagy-related markers in HCC control cells, SLC25A39 knockdown cells, and cells treated with a mitophagy inhibitor. The results showed that, compared with the HCC control group, SLC25A39 knockdown increased mitophagy and pro-apoptotic levels, while treatment with the mitophagy inhibitor reversed these changes (Fig. 9D).

Discussion

The complexity and high diversity of cancer pose a significant challenge to public health systems worldwide. The limitations of conventional therapy highlight the urgent need to identify new therapeutic targets and diagnostic markers. In recent years, the emergence of pan-cancer studies has provided transformative insights into common oncogenic mechanisms and the development of broad-spectrum therapeutic approaches. Within this framework, mitochondria, as the central hub coordinating cellular energy metabolism and redox homeostasis, have attracted considerable attention due to their functional impairments being closely associated with malignant tumor progression. Evidence indicates that mitochondrial metabolic reprogramming, by reshaping energy supply networks and oxidative stress signaling pathways, drives tumor cell microenvironmental adaptation and resistance to apoptosis, making it one of the focal topics in cancer biology research [14, 15]. Notably, the mitochondrial glutathione transporter SLC25A39, a critical regulator of oxidative metabolism, influences cell fate determination by maintaining mitochondrial redox balance, and its dysfunction can lead to metabolic disorders [9, 10]. However, the expression dynamics, mechanistic roles, and clinical relevance of this gene across the pan-cancer spectrum have not yet been systematically elucidated.

TIME exerts a decisive influence on cancer progression and profoundly shapes the ultimate efficacy of immunotherapy [16]. Substantial evidence has demonstrated that various immune cell subsets within the TME can exert dual effects while exhibiting substantial heterogeneity in functional states and spatial distribution [17]. Therefore, the identification of novel predictive biomarkers and the systematic characterization of immune infiltration profiles in cancer patients are crucial for achieving individualized precision immunotherapy. In this research, SLC25A39 expression levels inversely correlate with the infiltration of anti-tumor immune cells like CD8 + T cells and NK cells. Conversely, a positive correlation exists between SLC25A39 expression and Th2 cells, which exhibit pro-tumor potential [18]. Initially, the most striking fact was that LGG exhibited a distinct course where immune cell infiltration levels correlated with SLC25A39 expression. This inconsistency might stem from varying tumors and a complex tumor microenvironment, wherein certain parts may recruit specific immune contexts, active stages, and recruitment sites with differing functions. Some more recent studies have suggested that Th2 might be exerting pro-tumor effects in certain ways [19]. SLC25A39 was associated with three immune regulatory genes, MHC and immune checkpoint. The immune checkpoint acts as an immunosuppressive regulator, which cancer cells exploit to survive and evade immune system destruction, making it a key target for immunotherapy drugs [20]. MHC molecules encode a variety of proteins crucial for initiating cellular immunity. These proteins play a central role in antigen presentation and kickstarting the entire process, potentially enhancing the effectiveness of immune checkpoint blockers [21].

Genomic alterations represent fundamental mechanisms that drive cancer initiation, progression, and therapeutic responses. Mutations, rearrangements, copy number amplifications, or deletions affecting key genes can disrupt normal biological regulatory pathways, resulting in uncontrolled cell proliferation, abnormal growth, and tumorigenesis [22, 23]. Among the metrics used to evaluate genomic instability and mutational burden in tumors, TMB and MSI have been widely established as key biomarkers for predicting the efficacy of immunotherapies, particularly immune checkpoint inhibitors. In general, higher TMB or MSI scores correlate with increased tumor immunogenicity, indicating a greater likelihood of clinical benefit from such treatments [24, 25]. Our correlational findings suggest that TMB and MSI are associated with certain immune-related features, which have been previously linked to immunotherapy outcomes in selected cancer types. However, direct evidence of predictive value for immunotherapy response is not provided by this study. Future studies in patient cohorts receiving immune checkpoint inhibitors are necessary to evaluate whether TMB or MSI indeed predict clinical response.

Epigenetic dysregulation is pervasive in cancer, with RNA modifications representing one of the critical regulatory mechanisms that modulate gene expression and cellular functions during tumor development. These modifications have been regarded as potential therapeutic targets for cancer intervention [26, 27]. The present analysis revealed that SLC25A39 expression was significantly correlated with key regulatory genes within multiple RNA modification pathways, including m6A, m5C, and m1A, across various cancer types. Taking m5C modification as an example, a general pattern was observed in most cancers: genes functioning as “writers” (e.g., NSUN5, DNMT3B) and “readers” (e.g., ALYREF, YBX1) were positively correlated with SLC25A39 expression, whereas “erasers” (e.g., TET1, TET2, TET3) exhibited negative correlations. This pattern suggests that SLC25A39 may act in concert with m5C-associated effector proteins-either writers or readers-to influence mRNA stability, translational efficiency, or degradation rates, thereby modulating the proliferative and metastatic capacity of cancer cells. In addition, aberrant DNA methylation represents another common form of epigenetic alteration in cancer [28]. Hypermethylation of promoter regions frequently leads to the epigenetic silencing of gene expression [29]. The current findings revealed correlations between SLC25A39 expression and DNA methylation levels, with distinct regulatory relationships observed among different cancer types. Such regulation may influence the transcriptional activity of SLC25A39, consequently affecting its oncogenic functions in various tumor contexts.

While this study has elucidated the multifaceted roles of SLC25A39 in pan-cancer settings across multiple datasets and analytical dimensions, certain limitations must be acknowledged. First, the research was primarily based on data acquired from public databases and subsequent statistical analyses, which may be inherently influenced by inter-cohort heterogeneity. Hence, the results derived from bioinformatic approaches should be interpreted with caution. Second, the absence of validation using clinical specimens constitutes another limitation. In addition, we validated our findings only in HCC cell lines, which may limit their generalizability. The inhibitory effect of fluorouracil on SLC25A39 protein levels still requires further experimental and clinical validation. however, the findings provide a foundation and clear direction for future investigations.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Liufeng Qin and Yuanqian Yao performed the data analysis and drafted the manuscript. Jianlin Lv designed the research study. All authors read and approved the final manuscript.

Funding

This work was supported by The National Natural Science Foundation of Guangxi (Grant No. 2022BS032).

Data availability

The datasets analysed during the current study are available in the TCGA and GEO repository (TCGA, https://portal.gdc.cancer.gov/, TCGA-LIHC) and the Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo/).

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.