Deciphering the oncogenic role of Rac family small GTPase 3 in hepatocellular carcinoma through multiomics integration

Abstract

BACKGROUND

Hepatocellular carcinoma (HCC) remains a lethal malignancy due to its molecular complexity and chemoresistance. Rac family small GTPase 3 (RAC3), a tumorigenic GTPase understudied in HCC, drives recurrence via E2F transcription factor 1 (E2F1)-mediated transcriptional activation. This study integrates multiomics and clustered regularly interspaced short palindromic repeats (CRISPR) screening to delineate RAC3’s roles. RAC3 overexpression correlates with advanced HCC and patient age, while its knockout suppresses proliferation. Mechanistically, RAC3 dysregulates cell-cycle checkpoints through E2F1 binding. Pharmacological RAC3 inhibition disrupts tumor growth and synergizes with chemotherapy to overcome resistance.

AIM

To explore RAC3’s expression, clinical links, and HCC mechanisms via multiomics and functional genomics.

METHODS

Multiomic integration of The Cancer Genome Atlas (TCGA), Gene Expression Omnibus, and Genotype-Tissue Expression datasets was performed to analyze RAC3 mRNA expression. Immunohistochemistry quantified RAC3 protein in 108 HCC/adjacent tissue pairs. Kaplan–Meier/Cox regression assessed prognostic significance using TCGA data. CRISPR screening validated RAC3’s necessity for HCC proliferation. Functional enrichment identified associated pathways; hTFtarget/JASPAR predicted transcription factors, validated via chromatin immunoprecipitation sequencing (ChIP-seq).

RESULTS

RAC3 exhibited significant mRNA and protein overexpression in HCC tissues, which was correlated with advanced tumor stages and reduced overall survival rates (hazard ratio = 1.82, 95%CI: 1.31–2.53). Genetic ablation of RAC3 suppressed HCC cell proliferation across 16 cell lines. Pathway analysis revealed RAC3’s predominant involvement in cell-cycle regulation, DNA replication, and nucleocytoplasmic transport. Mechanistic investigations identified E2F1 as a pivotal upstream transcriptional regulator, and ChIP-seq analysis validated its direct binding to the RAC3 promoter region. These findings suggest that RAC3 drives HCC progression through E2F1-mediated cell-cycle dysregulation.

CONCLUSION

This study identified RAC3 as a key HCC oncogenic driver; its overexpression links to poor prognosis/resistance. Targeting the RAC3/E2F1 axis offers a new therapy, which highlights RAC3 as a biomarker/target.

Key Words: Rac family small GT Phase 3; Hepatocellular carcinoma; Immunohistochemistry; High-throughput sequencing; Clustered regularly interspaced short palindromic repeats screening

Core Tip: This study identifies Rac family small GTPase 3 (RAC3) as a critical oncogenic driver in hepatocellular carcinoma (HCC) through multi functional validation. RAC3 overexpression correlates with advanced tumor stage and poor prognosis (hazard ratio = 1.82), while clustered regularly interspaced short palindromic repeats screening confirms its necessity for HCC proliferation. Mechanistically, E2F transcription factor 1 transcriptionally activates RAC3, which drives cell-cycle dysregulation. These findings position RAC3 as a promising therapeutic target for combating chemotherapy resistance in HCC.

INTRODUCTION

Hepatocellular carcinoma (HCC) is a major global health burden, ranking as the fourth-most common malignancy and the second leading cause of cancer-related mortality in China[1]. Its pathogenesis is closely linked to smoking, alcohol abuse, dietary aflatoxin exposure, hepatitis B/C viral infections, and metabolic disorders (obesity, type 2 diabetes)[2-4]. Current therapeutic strategies, which include surgical resection, liver transplantation, chemotherapy, immunotherapy, and targeted therapies, have limited success. While curative surgery remains the gold standard for early-stage HCC, over 70% of patients present with advanced disease at diagnosis due to the absence of early symptoms, which renders them ineligible for surgical intervention[5]. Furthermore, postoperative recurrence rates exceeding 50% and the emergence of therapy resistance significantly compromise patients’ long-term survival[6]. These challenges underscore the urgent need to elucidate novel molecular drivers of HCC progression and to identify actionable therapeutic targets.

As central modulators that govern cytoskeletal organization and fundamental cellular activities, including cell adhesion, migratory capacity, and proliferative processes, Rho GTPases have emerged as significant contributors to carcinogenesis[7]. Among these molecular switches are those in the Rac subfamily, which play central roles in driving oncogenic progression by modulating neoplastic proliferation, apoptotic evasion, metastatic dissemination, and treatment refractoriness[8]. Rac family small GTPase 3 (RAC3), also termed RAS-related C3 botulinum toxin substrate 3, is a Rac subfamily member that governs cellular growth, cytoskeletal remodeling, and kinase signaling[9,10]. Although it exhibits low baseline expression in normal tissues[11,12], RAC3 is abnormally overexpressed across multiple malignancies, including breast cancer[13], bladder cancer[14], endometrial carcinoma[15], and lung adenocarcinoma[16], where it drives aggressive phenotypes. For instance, RAC3 promotes cisplatin resistance in bladder cancer via P21-activated kinase 1–extracellular regulated kinase 1/2 pathway activation[17,18], enhances endometrial cancer progression through fatty acid synthase-mediated proliferation and immunosuppression[14,19], and facilitates breast cancer invasion by suppressing apoptosis[15,20]. Despite its established oncogenic roles in these cancers, the involvement of RAC3 in HCC pathogenesis remains entirely unexplored.

Its unresolved questions underscore the need to systematically characterize RAC3’s expression patterns, pathological significance, and mechanistic underpinnings in HCC pathogenesis. By deciphering the RAC3-centric signaling architecture, this study seeks to identify innovative therapeutic targets that are capable of simultaneously tackling the clinical hurdles of neoplastic relapse and treatment refractoriness in HCC therapeutic interventions.

MATERIALS AND METHODS

Data acquisition and preprocessing

Microarray and RNA sequencing (RNA-seq) datasets for HCC and adjacent normal liver tissues were systematically retrieved from Oncomine, ArrayExpress, Gene Expression Omnibus (GEO), Genotype-Tissue Expression (GTEx), and The Cancer Genome Atlas (TCGA). The inclusion criteria were as follows: (1) Human HCC tissue samples; (2) A dataset size of ≥ 3 samples; and (3) The availability of RAC3 mRNA expression data. Duplicate datasets and those with incomplete metadata were excluded. Raw RAC3 mRNA expression values were log2(x + 1) transformed to stabilize variance, and batch effects were corrected using the SVA[21] package in R. Normalization and standardization of the merged expression matrixes were performed using limma (microarray) and edgeR (RNA-seq) pipelines.

The RAC3 expression profiles were stratified into HCC (n = 3414) and non-HCC (n = 3036) cohorts. Standardized mean difference (SMD) was calculated in STATA v15.1 to quantify expression disparities. Heterogeneity was assessed via the I² statistic, with a random-effects model applied when I² ≥ 50%. Statistical significance was defined as P < 0.05 (two-tailed). Egger’s regression test evaluated publication bias, and sensitivity forest plots identified potential heterogeneity sources. The diagnostic accuracy of RAC3 was evaluated via summary receiver operating characteristic (SROC) curves, with area under curve (AUC) values computed to estimate sensitivity and specificity.

Tissue specimens and immunohistochemistry

Formalin-fixed, paraffin-embedded HCC specimens and paired adjacent nontumor tissues (n = 108 pairs) were obtained from treatment-naïve patients undergoing radical resection at The First Affiliated Hospital of Guangxi Medical University (June 2023–June 2024). The inclusion criteria were as follows: (1) Histopathologically confirmed HCC; and (2) No prior radiotherapy, chemotherapy, or targeted therapy. Recurrent HCC or incomplete clinical records were excluded. Written informed consent was obtained under institutional ethics approval (No. 2022-KT-NSFC-127). Immunohistochemistry (IHC) staining was performed using validated anti-RAC3 antibodies (1:100 dilution, Catalog No. 125299, Chengdu Zen-Bioscience Co. Ltd.). Three blinded pathologists independently scored staining intensity (0: None; 1: Weak; 2: Moderate; 3: Strong) and percentage of positive cells (0: < 5%; 1: 5%–25%; 2: 26%–50%; 3: > 50%). The overall IHC score is calculated as the product of staining intensity and the proportion of positive cells, with a maximum score of 12. Scores of 7 or below are considered low expression, while scores of 8–12 are regarded as high expression[22]. Clinicopathological correlations were analyzed using independent sample t-tests (continuous variables) and Pearson’s correlation (categorical variables) using Statistical Package for the Social Sciences (SPSS) v23.0.

Prognostic analysis

Prognostic data from the TCGA-LIHC, GSE10143, GSE27150, and GSE76427 cohorts were analyzed. Patients were stratified into high-RAC3 and low-RAC3 expression groups using maximally selected rank statistics (maxstat package for R v4.3.2). Overall survival (OS) differences were assessed via Kaplan–Meier analysis (survival package for R), with log-rank tests for significance. Multivariate Cox proportional hazards models that incorporated age, sex, tumor stage, and histological grade identified independent prognostic factors.

Clustered regularly interspaced short palindromic repeats screening for RAC3 dependency

A genome-wide clustered regularly interspaced short palindromic repeats (CRISPR)-Cas9 knockout screen was conducted across 16 HCC cell lines. Gene essentiality scores were computed using the CERES algorithm, where negative scores (< 0) indicated RAC3-dependent growth suppression and positive scores (> 0) suggested oncogenic dependence[23].

Differentially expressed and co-expressed gene identification

Differentially expressed genes (DEGs) between HCC and non-HCC tissues were identified using the limma-voom pipeline (R v4.3.2), with thresholds of |log2FC| > 1 and an adjusted P value of < 0.05 (Benjamini-Hochberg correction). RAC3 co-expressed genes (CEGs) were defined by Pearson’s correlation (r ≥ 0.3, P < 0.05) across ≥ 3 independent datasets.

Functional enrichment analysis

Intersected upregulated DEGs and RAC3 CEGs underwent Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment using ClusterProfiler (R package). Biological processes, cellular components, molecular functions, and pathways with a false discovery rate of < 0.05 were retained. The results were visualized via ggplot2 (R package).

Transcriptional regulation of RAC3

The hTFtarget database (http://bioinfo.life.hust.edu.cn/hTFtarget) was queried to predict RAC3-associated transcription factors (TFs) with regulatory potential (RP) scores of > 0.6. TF binding motifs in the RAC3 promoter (2 kb upstream of transcription start site) were identified using JASPAR (http://jaspar.genereg.net/). Chromatin immunoprecipitation sequencing peaks at the RAC3 transcription start site were validated via Cistrome DB (http://cistrome.org/db/#/) and visualized in Integrative Genomics Viewer (IGV v3.0.1). Survival associations of candidate TFs were analyzed using the Kaplan–Meier plotter (log-rank P < 0.05).

Statistical analysis

Analyses were performed in SPSS v25.0 (descriptive statistics, t-tests), STATA v15.1 (meta-analysis), R v4.3.2 (bioinformatics pipelines), and GraphPad Prism v8.0.0 (visualization). The statistical methods employed in this study were reviewed by Mei-Meng Huang of The Department of Statistics at Guangxi Medical University.

RESULTS

Upregulation of RAC3 in HCC

Our integrative analytical framework processed 75 Gene Expression Omnibus Series (GSE) datasets through platform-specific harmonization and generated 36 standardized expression matrixes. These were cross-referenced with curated datasets from TCGA and GTEx repositories, which culminated in a composite cohort comprising 3414 histologically confirmed HCC specimens and 3036 non-neoplastic controls (comprehensive metadata annotations available in Supplementary Table 1). As depicted in Figure 1A, RAC3 mRNA expression was significantly upregulated in HCC tissues compared to non-HCC tissues (SMD = 0.59, 95%CI: 0.54–0.65, P < 0.001). Given the substantial heterogeneity observed (I² = 83%, P < 0.001), a random-effects model was employed. Egger’s test (Figure 1B) indicated no significant publication bias (P = 0.082). Sensitivity analysis (Figure 1C) revealed that none of the included studies identified a source of heterogeneity. The SROC curve (Figure 2A) demonstrated that RAC3 mRNA exhibited strong diagnostic accuracy for HCC tissues (AUC = 0.76, 95%CI: 0.72–0.79), with corresponding sensitivity and specificity (Figure 2B) of 0.63 (95%CI: 0.55–0.70) and 0.77 (95%CI: 0.70–0.82), respectively. The positive and negative diagnostic likelihood ratios (Figure 2C) were 2.71 (95%CI: 2.16–3.41) and 0.49 (95%CI: 0.41–0.58), respectively.

Figure 1 Meta-analysis of Rac family small GTPase 3 dysregulation in hepatocellular carcinoma using standardized mean difference. A: Forest plot visualizing pooled standardized mean difference estimates across studies; B: Egger’s regression test for publication bias assessment; C: Leave-one-out sensitivity analysis of effect size stability.

Figure 2 Diagnostic potential of Rac family small GTPase 3 for discriminating hepatocellular carcinoma from non-malignant liver conditions. A: Summary receiver operating characteristic curve with area under curve estimation; B: Pooled sensitivity and specificity with 95%CI; C: Diagnostic likelihood ratios [positive likelihood ratio (LR) and negative LR] across included studies. AUC: Area under curve; SROC: Summary receiver operating characteristic.

As demonstrated in Table 1, high RAC3 protein expression was observed in 69 of 108 HCC tissues (63.9%), whereas 39 cases showed low expression (36.1%). In contrast, of 108 adjacent liver tissues, only 30 exhibited high RAC3 protein expression (27.8%), with most (72.2%) showing low expression. IHC profiling demonstrated the predominant subcellular localization of RAC3 in the cytoplasmic and membranous compartments of neoplastic hepatocytes (Figure 3A). Quantitative histomorphometric analysis revealed marked overexpression in malignant specimens relative to matched peritumoral tissue (Figure 3B). Clinicopathological correlation analysis established statistically significant associations between RAC3 immunoreactivity scores and elevated serum alpha-fetoprotein concentrations, as well as confirmed satellite lesions (Table 2). The receiver operating characteristic curve, constructed based on IHC scores, yielded an AUC value of 0.773 (Figure 3C), which further underscores the diagnostic potential of RAC3 in HCC.

Figure 3 Clinical relevance of Rac family small GTPase 3 dysregulation in hepatocellular carcinoma. A: Representative immunohistochemical (IHC) staining demonstrating differential Rac family small GTPase 3 (RAC3) expression between hepatocellular carcinoma and adjacent non-tumorous liver tissues; B: Quantitative comparison of IHC H-scores in matched tumor-normal pairs (P < 0.001, paired t-test); C: Diagnostic performance evaluation using receiver operating characteristic curve analysis (area under curve = 0.82, 95%CI: 0.76–0.89); D: Kaplan–Meier survival curves stratified by RAC3 expression levels with clinicopathological correlations. AUC: Area under curve; HCC: Hepatocellular carcinoma; RAC3: Rac family small GTPase 3.

Table 1 Comparative analysis of Rac family small GTPase 3 protein expression in hepatocellular carcinoma tissues and paired adjacent non-tumorous liver tissues, n (%).

Group	Number	Rac family small GTPase 3 expression		χ2	P value
		Low	High
HCC tissues	108	39 (36.1)	69 (63.9)	28.4	< 0.001
Non-HCC tissues	108	78 (72.2)	30 (27.8)

HCC: Hepatocellular carcinoma.

Table 2 Association between Rac family small GTPase 3 protein immunoreactivity and clinicopathological characteristics of hepatocellular carcinoma using an in-house developed immunohistochemistry assay.

Clinicopathological parameters	Number	Rac family small GTPase 3 expression		χ2	P value
		High	Low
Age (years)
≥ 60	69	17	52	0.645	0.422
< 60	39	7	32
Gender
Male	93	62	31	2.239	0.135
Female	15	7	8
Pathologic grade
I-II	27	21	6	3.010	0.083
III-IV	81	48	33
Clinical stage
I-II	102	64	38	1.041	0.308
III-IV	6	5	1
Capsular invasion
Yes	76	53	23	3.802	0.051
No	32	16	16
Microvascular invasion
Yes	41	25	16	0.243	0.622
No	67	44	23
Satellite nodule
Yes	11	4	7	4.022	0.045
No	97	65	32
Liver cirrhosis
Yes	65	39	26	1.070	0.301
No	43	30	13
Alpha-fetoprotein
≥ 8.78 ng/mL	69	39	30	4.495	0.034
< 8.78 ng/mL	39	30	9
Hepatitis B virus
Yes	95	59	36	1.088	0.297
No	13	10	3

Prognostic association of RAC3 transcriptional activation in HCC

The cross-platform bioinformatics integration of TCGA and GEO repositories demonstrated a robust correlation between RAC3 transcriptional upregulation and diminished median OS duration in HCC cohorts, which indicates its potential role as an independent risk factor linked to unfavorable clinical outcomes (Figure 3D). Univariate Cox regression analysis of the TCGA-LIHC cohort identified age and RAC3 overexpression as statistically significant predictors of poor prognoses, thereby validating the clinical relevance of RAC3. Subsequent multivariate analysis confirmed RAC3 as an independent prognostic marker for HCC [hazard ratio (HR) = 2.220, 95%CI: 1.428-3.451, P < 0.001] (Table 3), which underscores its clinical utility in risk stratification.

Table 3 Univariate and multivariate Cox proportional hazards regression analyses of overall survival-associated factors in The Cancer Genome Atlas liver hepatocellular carcinoma cohort.

Variables	Univariate analysis			Multivariate analysis
	HR	95%CI	P value	HR	95%CI	P value
Age	1.609	1.040-2.491	0.033	1.564	0.997-2.453	0.051
Gender	0.758	0.496-1.158	0.200	0.855	0.549-1.332	0.488
Pathologic stage	1.134	0.749-1.717	0.553	1.037	0.686-1.568	0.862
Clinical stage	0.925	0.606-1.411	0.717	0.952	0.629-1.441	0.815
New tumor event after initial treatment	1.058	0.686-1.630	0.799	1.097	0.707-1.704	0.679
Rac family small GTPase 3	2.254	1.460-3.480	< 0.001	2.220	1.428-3.451	< 0.001

HR: Hazard ratio.

RAC3 promotes HCC cell proliferation

CRISPR knockout screening technology was employed to assess the dependency score of RAC3 and to evaluate its impact on HCC cell proliferation. CRISPR/Cas9-mediated knockout of RAC3 in a panel of 16 established HCC cellular models exerted profound growth-suppressive effects (Figure 4). These functional genomic findings mechanistically support RAC3’s essential oncogenic driver function in hepatocarcinogenesis and malignant progression.

Figure 4  Functional genomic profiling identifies Rac family small GTPase 3 as a pro-growth dependency in hepatocellular carcinoma.

Enrichment analysis of RAC3-related genes

Initially, 65535 CEGs and 8676 upregulated DEGs were identified. Intersection analysis was conducted and yielded 6773 overlapping genes (all appearing in ≥ 3 datasets). GO and KEGG enrichment analyses were performed on these overlapping genes. The GO analysis indicated that biological processes were predominantly associated with chromosome segregation, DNA replication, and ribosome biogenesis (Figure 5A). Cellular components were significantly enriched in chromosomal regions, spindles, and ribosomal subunits (Figure 5B). Molecular functions were primarily linked to ribosomal structural constituents, catalytic activity, DNA interaction, and adenosine triphosphate hydrolase activity (Figure 5C). The KEGG pathway analysis highlighted enrichment in cell-cycle regulation, nucleocytoplasmic transport, and DNA replication pathways (Figure 5D).

Figure 5 Functional convergence of Rac family small GTPase 3-coexpressed genes with hepatocellular carcinoma-upregulated transcripts: A multi-layer enrichment landscape. A: Gene Ontology enrichment: Biological processes; B: Cellular component localization; C: Molecular function annotation; D: Kyoto Encyclopedia of Genes and Genomes pathway topology; E: Gene set enrichment analysis of hepatocellular carcinoma-related pathway activation.

Additionally, gene set enrichment analysis demonstrated that KEGG pathways were predominantly associated with basal cell carcinoma, cell-cycle regulation, and extracellular matrix receptor interactions (Figure 5E).

E2F transcription factor 1 activates RAC3 transcription

Overexpression of RAC3 is linked to unfavorable prognoses in HCC patients, which points to a need to explore the molecular control pathways of RAC3 in HCC. From the hTFtarget database, 89 TFs potentially related to RAC3 (RP score > 0.6) were identified. These TFs intersected with 4964 genes positively correlated with RAC3 (r ≥ 0.7), resulting in 11 TFs closely associated with RAC3, namely KDM2B, E2F transcription factor 1 (E2F1), GATA1, NANOG, SUPT5H, MBD3, TFAP2A, SMARCA4, NCAPH2, ESR1, and USF2 (Figure 6A and B)[24]. Based on an extensive literature review and enrichment analysis results, E2F1 was selected as a potential upstream regulator of RAC3. To mechanistically dissect the E2F1-mediated transcriptional activation of RAC3, we performed in silico prediction of canonical E2F1-binding motifs (TTTSSCGC consensus sequence) in the 2000 bp promoter-proximal region of the RAC3 Locus using the JASPAR 2024 database (matrix MA0024.2). The three most strongly correlated motifs were identified as TGGGCGGGAGG, GTGGCGGCAGG, and CAAGCGGGAGC (Figure 6C and D)[25]. The results demonstrated the presence of E2F1-binding peaks at RAC3’s transcription start site (Figure 6E)[26,27]. SMD analysis was used to calculate the integrated E2F1 gene expression in HCC and revealed a significant upregulation in HCC tissues compared to non-HCC tissues (Figure 7A). Additionally, survival modeling showed that high E2F1 expression in HCC correlated with adverse prognosis (HR = 2.11, 95%CI: 1.45–3.07) (Figure 7B-D)[28].

Figure 6 E2F transcription factor 1 activates Rac family small GTPase 3 transcription by binding to the Rac family small GTPase 3 promoter region. A: Intersection Venn diagram of Rac family small GTPase 3 (RAC3)-associated upregulated genes and related transcription factors (TFs); B: Target gene RAC3 and its related TFs (it was sourced from the hTFtarget database); C and D: The binding site of E2F transcription factor 1 (E2F1) in the promoter region of RAC3 and the ten predominant binding sequences (they were obtained from JASPAR); E: E2F1 binding peak in the translation initiation region of RAC3 [it was derived from integrated analyses using Cistrome DB and Integrative Genomics Viewer (IGV v3.0.1)]. CEGs: Co-expressed genes; RAC3: Rac family small GTPase 3; RP: Regulatory potential.

Figure 7 Clinical significance of E2F transcription factor 1 dysregulation in hepatocellular carcinoma: Expression patterns and survival implications. A: Transcriptional overexpression of E2F transcription factor 1 (E2F1) mRNA in hepatocellular carcinoma vs non-tumorous liver tissues; B–D: Prognostic stratification using Kaplan–Meier Plotter survival analytics including overall survival, progression-free survival, recurrence-free survival (log-rank P < 0.01). E2F1: E2F transcription factor 1; OS: Overall survival; PFS: Progression-free survival; RFS: Recurrence-free survival.

DISCUSSION

HCC represents one of the most aggressive malignancies worldwide and has significant implications for human health. Surgical resection is the primary curative treatment for HCC, but its effectiveness is limited primarily to early-stage cases, which leaves advanced-stage patients with few treatment options. HCC’s high recurrence and metastasis rates highlight an urgent need to understand the pathophysiological mechanisms behind HCC progression that can be targeted in new therapies.

Evidence is emerging on the oncogenic role of RAC3 in multiple malignancies, including renal cell carcinoma[24], breast cancer[25-27], lung adenocarcinoma[28,29], and bladder cancer[8,18,30,31]. However, its involvement in HCC pathogenesis has remained unexplored until now. By integrating multiomics data from the GEO, TCGA, and GTEx databases, complemented by immunohistochemical validation, our study demonstrated notable RAC3 upregulation in HCC tissues. Moreover, survival analysis revealed that RAC3 overexpression correlates with a poor prognosis of HCC patients. Importantly, CRISPR-mediated RAC3 knockout inhibited proliferation across 16 HCC cell lines, which implicates RAC3 as a significant pro-tumorigenic driver and a promising prognostic biomarker of HCC.

Functional enrichment analysis further revealed that RAC3-associated genes are predominantly enriched in cell-cycle-related pathways. Cell-cycle dysregulation, a hallmark of cancer, arises from mutations or aberrant expression of key regulatory proteins[32]. DNA impairment checkpoints and repair processes play dual roles in sustaining genomic integrity: They either coordinate repair pathways or induce proliferative checkpoint activation coupled with intrinsic apoptosis execution in response to telomeric crises[32,33]. The mechanistic insights delineate actionable therapeutic vulnerabilities within cell-cycle regulatory networks and DNA repair fidelity maintenance systems and thus inform precise intervention strategies against tumorigenic hyperproliferation. Our findings position RAC3 as a promising therapeutic target for HCC treatment.

In integrated computational prediction, E2F1 emerged as the predominant transcriptional regulator of RAC3, with binding motifs localized to the RAC3 promoter. The E2F family (E2F1–E2F8) governs diverse processes, including cell-cycle progression, apoptosis, and chemoresistance[34]. Of them, E2F1 is a well-documented oncogene that is overexpressed in multiple cancers and linked to malignant transformation and poor prognoses[35-39]. Notably, E2F1 contributes to chemoresistance in colon cancer (5-fluorouracil/oxaliplatin)[39], lung cancer (cisplatin/gefitinib)[40,41], breast cancer (tamoxifen/palbociclib)[42], and HCC (oxaliplatin)[43]. Significantly, RAC3 has emerged as a key player in chemotherapy resistance. For instance, RAC3 silencing reverses paclitaxel resistance in lung adenocarcinoma[16], enhances cisplatin sensitivity in endometrial carcinoma[14], and inhibits clonogenic potential in platinum-refractory urothelial carcinoma cells[18]. In non-small cell lung cancer, the DDX1/ADAR1/RAC3 axis modulates cisplatin chemosensitivity[44], while RAC1/RAC3 dual inhibition overcomes cetuximab resistance in head and neck squamous cell carcinoma[45]. These findings collectively underscore the therapeutic relevance of the E2F1/RAC3 axis in cancer.

Despite these insights, our study has limitations. First, the prognostic value of RAC3 protein expression in HCC requires validation by clinical follow-up studies. Second, the mechanistic interplay between E2F1 and RAC3 in HCC requires experimental confirmation. Future work should adopt in vitro and in vivo models to validate the E2F1/RAC3 axis, dissect its impact on cell-cycle dynamics, and explore its translational potential in overcoming therapy resistance.

CONCLUSION

This study delineates a novel E2F1-RAC3 regulatory axis in HCC wherein E2F1 drives RAC3 transcription to promote uncontrolled proliferation. The dual functions of E2F1 and RAC3 in both tumor progression and chemoresistance highlight their promise as therapeutic candidates. Targeting this axis may disrupt cell-cycle dysregulation and mitigate therapeutic resistance, offering a promising strategy for HCC management.

ACKNOWLEDGEMENTS

The authors would like to thank Guangxi Zhuang Autonomous Region Clinical Medicine Research Center for molecular pathology and intelligent pathology precision diagnosis and Guangxi Key Laboratory of Enhanced Recovery after Surgery for Gastrointestinal Cancer for technical support as well as all the public databases used in this study.