ODC1 Polyamine Metabolism Drives Prostate Cancer via AKT and Splicing

TL;DR

This study shows that ODC1, a key enzyme in polyamine metabolism, promotes prostate cancer by affecting gene expression and splicing, especially through the AKT pathway.

Contribution

The study reveals a novel mechanistic link between ODC1 activity, splicing regulation, and prostate cancer progression via the AKT pathway.

Findings

Reducing ODC1 expression slowed prostate cancer cell growth, movement, and increased cell death.

Over 1,000 differentially expressed genes and 2,000 alternative splicing events were linked to cancer-related pathways.

Genes like CAV1 and ITGB1 connected ODC1 activity to the AKT signaling pathway in prostate cancer progression.

Abstract

Prostate cancer is an aggressive disease with limited quantifiable biomarkers. One gene of interest is ODC1, which encodes ornithine decarboxylase, the rate‐limiting enzyme converting ornithine to putrescine in polyamine metabolism. Although ODC1 is known to be involved in prostate cancer development, exactly how it drives the disease mechanistically is not fully understood. To explore this, we created a prostate cancer cell model with reduced ODC1 expression and examined its effects on tumour behaviours. Knocking down ODC1 significantly slowed cell growth and movement while increasing cell death. Using RNA sequencing, we identified over one thousand differentially expressed genes, with 565 upregulated and 497 downregulated, primarily linked to angiogenesis and cell adhesion. We also found more than two thousand alternative splicing events connected to cell cycle regulation and protein modification. Notably, genes including CAV1, ITGB1, BNIP3, and YTHDF2 were associated with the AKT signalling pathway, suggesting a functional link between ODC1 activity and cancer progression. These results indicate that ODC1 influences prostate cancer cell behaviour by regulating both gene expression and splicing, particularly affecting pathways involved in angiogenesis, adhesion, and the cell cycle. This points to the AKT pathway and polyamine metabolism as potentially valuable targets for future prostate cancer therapies.

Figures

ODC1 Knockdown Promotes Malignant Phenotype in PC3 Cells. (A) RT‐qPCR detection of ODC1 knockdown efficiency in PC3 cells. (B) WB detection of ODC1 knockdown efficiency in PC3 cells. (C) CCK8 assay demonstrates reduced proliferation capacity in PC3 cells following ODC1 knockdown. (D, E) ODC1 knockdown diminishes PC3 cell migration capacity but has negligible effect on invasion capacity. (F) Flow cytometry reveals increased apoptosis in PC3 cells following ODC1 knockdown. (G) TNMplot database indicates high ODC1 expression in prostate cancer tissues.

ODC1 regulates gene expression in PC‐3 cells. (A) Principal component analysis (PCA) based on the FPKM value of all detected genes. The ellipse for each group is the confidence ellipse. (B) Volcano plot showing all differentially expressed genes (DEGs) between the treatment and control groups with DEseq2. FDR < 0.05 and fold change (FC) ≥ 2 or ≤ 0.5. (C). Pie chart showing the types of DEGs. (D) Hierarchical clustering heat map showing the expression of all DEGs. FPKM values are log2‐transformed and then median‐centered by each gene.

ODC1 regulates gene expression in PC‐3 cells. (A) Bubble diagram showing the most enriched GO biological processes for the upregulated and downregulated DEGs. (B) Boxplot showing the expression pattern and statistical difference of DEGs (error bars represent the mean ± SEM. *p < 0.001). (C) Western blot validation of five key gene expression levels (PERP, CAV1, ITGB1, BNIP3, CDKN3) in PC3 cells after ODC1 knockdown, showing PERP upregulation and downregulation of the remaining four genes.

ODC1 influences the alternative splicing of genes in PC‐3 cells. (A) Bar plot showing the number of significant regulated alternative splicing events (RASEs). X‐axis: RASE number. Y‐axis: types of AS events. (B) Bubble Diagram showing the most enriched GO biological processes obtained for RASGs. (C) Boxplot showing the ratio and statistical differences in RASEs (error bars represent the mean ± SEM. p < 0.01, *p < 0.001).

Full text

Introduction

1

Prostate cancer (PCa), a prevalent urological malignancy, demonstrates substantial disease burden with elevated incidence and mortality. Epidemiological statistics from the American Cancer Society indicate PCa comprised 21% of all cancer diagnoses in 2020, corresponding to 191,930 new cases and 33,330 fatalities annually in the US [1]. Androgen deprivation therapy (ADT) remains the first‐line treatment for patients with metastatic PCa. However, almost all patients with PCa eventually become unresponsive to ADT; this is referred to as castration‐resistant prostate cancer (CRPC) [2]. CRPC is primarily mediated through dysregulated androgen receptor signalling (ARS). CRPC progression is characterised by constitutive AR activation, wherein aberrant ARS emerges as the principal pathogenic determinant driving therapeutic resistance and metastatic dissemination [3]. CRPC remains challenged by limited durable treatment efficacy, with disease progression significantly correlating with elevated morbidity and impaired patient quality of life [4]. The identification of clinically actionable, quantifiable biomarkers remains an unmet need in prostate cancer management, with current biomarker panels demonstrating limited prognostic and predictive accuracy across heterogeneous patient cohorts [5]. While ornithine decarboxylase 1 (ODC1) has been validated as a prostate adenocarcinoma biomarker with elevated expression profiles, its molecular mechanisms driving oncogenic progression remain systematically uncharacterised [5]. To address this gap, this study elucidated how ornithine decarboxylase 1 (ODC1), a novel biomarker for prostate cancer (PCa), influences PCa progression and identified potential therapeutic targets.

In cancer, increased ODC1 expression is related to cell invasion and angiogenesis [6]. In addition, the small‐molecule irreversible inhibitor difluoromethylornithine (DFMO) of ODC1 has been used in the treatment of various adult cancers [7]. ODC1 encodes a rate‐limiting enzyme, ODC, which produces putrescine, spermidine, and spermine via spermidine synthase (SPDSY) and spermine synthase (SPMSY). The intracellular concentrations of polyamines are controlled by the ODC antizyme 1 (AZ), which has a high affinity for ODC monomers. Polyamines play important roles in regulating gene expression, maintaining DNA structure, RNA processing and translation, and cell proliferation and differentiation [8, 9]. The prostate gland exhibits physiologically elevated polyamine concentrations, with oncogenic transformation driven by polyamine metabolic dysregulation through tumour progression pathways [10], and polyamine metabolic enzyme inhibitors or polyamine analogs can arrest cell growth or induce cell death in PCa [11].

Genetic risk factors predispose individuals to the development of PCa. Studies have demonstrated that candidate genes for PCa include those that participate in the androgen pathway and testosterone metabolism [12]. Although studies have concluded that ODC1 is a biomarker for PCa, the function of ODC1 and the molecular mechanisms involved in the occurrence and development of PCa remain unclear. Although the treatment of PCa has progressed in recent years, current treatments remain unsatisfactory and have limited effects on patient survival time. Recently, there has been renewed interest in the molecular pathogenesis of PCa. Common genetic pathways include the AR, PI3K‐PTEN, WNT, TGF‐β/SMAD4, DNA repair, and cell cycle pathways [13]. The aim of our study was to evaluate the role of ODC1 in the development of PCa through regulating polyamine synthesis via its effects on gene expression and alternative splicing. Furthermore, we aimed to identify therapeutic targets based on polyamine synthesis in ODC1‐expressing CRPC.

Materials and Methods

2

RNA sequencing was employed to examine differentially expressed genes (DEGs) and allele‐specific expression (ASEs) in ODC1‐knockdown PC‐3 cells. Commercially synthesised siRNA duplexes, including non‐targeting control (siNegative: 5′’‐UUCUCCGAACGUGUCACGUTT‐3′, sense strand) and ODC1‐specific sequences (siODC1: 5′‐GCUGUGACCUGCCUGAAAUTT‐3′, sense strand), were obtained from GenePharma (Suzhou, Jiangsu, China).

Culture of PC‐3 Cells and Plasmid Transfections

2.1

The PC‐3 human prostate carcinoma cell line and DU145 cells were acquired and propagated under established culture protocols as previously detailed [14]. Transient gene silencing was achieved by employing Lipofectamine RNAiMAX Transfection Reagent per the supplier's technical manual. Cellular lysates were prepared at 48 h post‐transfection for concurrent reverse transcription quantitative PCR (RT‐qPCR) and immunoblotting investigations.

Assessment of Gene Expression

2.2

GAPDH served as the normalisation control to determine the efficacy of ODC1 knockdown.

Western Blot

2.3

This investigation employed an optimised immunoblotting protocol adapted from Ye's methodology [15]. Key procedural steps included: (1) Protein transfer onto polyvinylidene difluoride (PVDF) membranes; (2) Blocking with 5% (w/v) non‐fat dried milk in TBST buffer; (3) Primary antibody incubation overnight at 4°C; (4) HRP‐conjugated secondary antibody treatment for 1 h at ambient temperature. Protein‐antibody complexes were visualised using enhanced chemiluminescence detection system with 30‐s exposure intervals.

CCK‐8 Assay

2.4

Cell proliferation was quantified using CCK‐8. PC‐3 cells (1 × 10^4^ cells/well) in 96‐well plates were cultured for 48 h. After adding 10 μL CCK‐8 reagent per well, plates were incubated for 3 h before measuring absorbance at 450 nm using an ELX800 microplate reader (Biotek, USA).

Annexin V Apoptosis Assay

2.5

Apoptosis analysis employed Annexin V‐Fluor647/PI kit. PC‐3 cells (5 × 10^5^/well) in 6‐well plates were maintained 24 h prior to 48‐h plasmid transfection. Cells were stained with 5 μL Annexin V‐Fluor647 and 10 μL PI in 100 μL binding buffer (15‐min dark incubation, 25°C), then analyzed via BD FACSCanto flow cytometer (USA) with 488/670 nm wavelength settings.

Cell Invasion and Migration Assay

2.6

Cell invasion and migration were evaluated using Corning Transwell chambers. For invasion assays, inserts were coated with Matrigel and polymerised at 37°C for 1 h. Serum‐starved PC‐3/DU145 cells (5 × 10^4^ cells/insert) in 0.2 mL serum‐free medium were seeded into upper chambers, with 10% FBS‐containing medium as chemoattractant in lower chambers. After 24‐h (migration) or 48‐h (invasion) incubation, non‐migrated cells were removed using cotton swabs. Membranes were fixed with 4% paraformaldehyde, stained with 0.1% crystal violet, and quantified under an MF52‐N inverted microscope at 200× magnification.

RNA‐Seq Preparation and Data Analysis

2.7

RNA Extraction

2.7.1

DNase‐treated total RNA (RQ1 DNase, Promega) was assessed for purity via A260/A280 ratio (1.8–2.2) using a SmartSpec Plus spectrophotometer (Bio‐Rad), with RNA integrity confirmed by 1.5% agarose gel electrophoresis. RNA‐seq libraries were constructed from 1 μg DNase‐treated RNA using VAHTS mRNA capture beads and the VAHTS Universal V8 Kit. Following mRNA fragmentation and double‐stranded cDNA synthesis, Illumina‐compatible adapters were ligated after end repair/A‐tailing. Libraries (300–500 bp) were PCR‐amplified, quantified, and cryopreserved for strand‐specific sequencing via dUTP strand marking.

High‐Throughput Sequencing

2.7.2

Illumina NovaSeq 6000 (150 bp PE) sequencing was performed on strand‐specific libraries constructed with VAHTS V8 kits.

RNA Sequencing Data Preprocessing and Read Alignment

2.7.3

Sequencing reads containing over two ambiguous bases (N) were excluded. Adapter sequences and low‐quality regions were trimmed from raw reads using FASTX Toolkit, with subsequent removal of reads shorter than 16 nucleotides. Processed reads were aligned to the GRCh38 genome via HISAT2 [16] permitting four mismatches. Uniquely aligned reads were quantified as FPKM (fragments per kilobase per million mapped reads) [17].

DEG Analysis

2.7.4

To screen the DEGs, the package DESeq2 [18] was used with the following identification conditions: p‐value threshold < 0.05; expression ratio exceeding 2‐fold upregulation or declining below 0.5‐fold.

Alternative Splicing Analysis

2.7.5

Alternative splicing events (ASEs) were analysed using the established ABLas computational pipeline [19], which identifies 10 distinct ASE types through splice junction read analysis: exon skipping (ES), alternative 5′ splice sites (A5SS), alternative 3′ splice sites (A3SS), mutually exclusive 5′UTRs (5pMXE), mutually exclusive 3′UTRs (3pMXE), mutually exclusive exons (MXEs), cassette exon A3SS&ES, and A5SS&ES. Statistical significance of ODC1‐mediated ASE alterations was determined by Student's t‐test, with a 5% probability threshold defining regulatory significance.

Validation of Gene Expression

2.7.6

RT‐qPCR validation (QuantStudio5, SYBR Green Master Mix) confirmed RNA‐seq DEGs/ASEs with GAPDH‐normalised 2^−ΔΔCT^ quantification [20]. Detailed information on the primers used to detect pre‐mRNA splicing is provided in Additional file 1. PCR amplifications were quantified using the 2^−ΔΔCT^ method and the statistical significance was calculated using two‐way ANOVA.

Results

3

ODC1 Knockdown Inhibits Cellular Proliferation and Promotes Apoptosis in PCa Cells

3.1

ODC1 knockdown was successfully achieved in both PC‐3 and DU145 cells using two distinct siRNA duplexes. Reverse transcription quantitative PCR (RT‐qPCR) analysis demonstrated substantial reductions in ODC1 mRNA expression across both cell lines (Figures 1A and S1A). Concordantly, immunoblotting revealed commensurate protein downregulation, confirming effective ODC1 silencing at the translational level (Figures 1B and S1B). These findings validate the successful establishment of ODC1‐knockdown cell models in both PC‐3 and DU145 prostate cancer cell lines for subsequent mechanistic investigations and RNA‐seq. ODC1 knockdown markedly suppressed cell proliferation and migration (p < 0.05; Figures 1C,E and S1C,E), promoted apoptosis (p < 0.05, Figures 1F and S1F), and inhibited cell invasion; however, the results were not statistically significant (Figures 1D and S1D). In addition, we analysed the expression of ODC1 in PCa and para‐cancer tissues using the TNMplot database and found that ODC1 was expressed at higher levels in PCa tissues than those in the para‐cancer tissues (p < 0.001, Mann–Whitney U test. Figure 1G). These results indicated that ODC1 inhibition decreased the proliferation of PCa cells in vitro.

ODC1 Knockdown Promotes Malignant Phenotype in PC3 Cells. (A) RT‐qPCR detection of ODC1 knockdown efficiency in PC3 cells. (B) WB detection of ODC1 knockdown efficiency in PC3 cells. (C) CCK8 assay demonstrates reduced proliferation capacity in PC3 cells following ODC1 knockdown. (D, E) ODC1 knockdown diminishes PC3 cell migration capacity but has negligible effect on invasion capacity. (F) Flow cytometry reveals increased apoptosis in PC3 cells following ODC1 knockdown. (G) TNMplot database indicates high ODC1 expression in prostate cancer tissues.

ODC1 Regulates Gene Expression in PC‐3 Cells

3.2

Consistent with prior studies documenting ODC1 upregulation in malignancies and its enzymatic role in polyamine biosynthesis, we performed RNA sequencing (Illumina NovaSeq 6000) on ODC1‐knockdown PC3 cells and control groups. Principal component analysis (PCA) of genome‐wide expression profiles identified ODC1 expression status as the main factor influencing the pattern of gene expression (Figure 2A). Next, we investigated ODC1‐regulated transcription by predicting DEGs. Differential expression analysis employing the edgeR package [18] identified ODC1‐modulated transcripts under stringent criteria (FDR‐adjusted p < 0.05; |log2(fold change)| ≥ 1). A total of 1062 genes were differentially expressed, among which we identified 565 upregulated and 497 downregulated DEGs (Figure 2B). Among the 1062 DEGs, 748 (70.4%) were mRNAs, 247 (23.3%) were lncRNAs, and 67 (6.3%) were others (Figure 2C). The hierarchical clustering heat map of all DEGs revealed that the ODC1‐knockdown groups showed the same regulatory consistency in DEG expression levels as the control group (Figure 2D). These results demonstrated that ODC1 can regulate gene expression in PC‐3 cells.

ODC1 regulates gene expression in PC‐3 cells. (A) Principal component analysis (PCA) based on the FPKM value of all detected genes. The ellipse for each group is the confidence ellipse. (B) Volcano plot showing all differentially expressed genes (DEGs) between the treatment and control groups with DEseq2. FDR < 0.05 and fold change (FC) ≥ 2 or ≤ 0.5. (C). Pie chart showing the types of DEGs. (D) Hierarchical clustering heat map showing the expression of all DEGs. FPKM values are log2‐transformed and then median‐centered by each gene.

ODC1 Knockdown Resulted in Some DEGs Associated With Multiple Biological Pathways in PC‐3 Cells

3.3

Gene ontology (GO) enrichment analysis of the 1062 differentially expressed genes (DEGs) was conducted across both expression polarities (upregulated/downregulated) to elucidate their putative biological roles. We identified 565 DEGs enriched in ‘transcribed regulation of RNA polymerase II’; ‘transcribed regulation of DNA templates; ‘extracellular matrix tissue’; ‘collagen fiber tissue’; ‘angiogenesis‘; ‘skeletal system development’; ‘cell adhesion’; ‘response to toxic substances'; ‘multicellular biological development’; ‘positive regulation of MAPK cascade pathway;’ and other biological processes by GO pathway enrichment analysis (Figure 3A). The 497 downregulated DEGs were mainly associated with ‘cell matrix adhesion’; ‘apoptosis process'; ‘negative regulation of extrinsic apoptosis signalling pathway in absence of ligand’; ‘response to hypoxia‘; ‘positive regulation of cell population proliferation’; ‘establishment of mitotic spindle orientation’; ‘angiogenesis'; ‘positive regulation of apoptosis process‘; and ‘cytokine‐mediated signalling pathway’ (Figure 3A). These results suggested that the regulation of transcription by ODC1 is complex and involves multiple biological pathways in PC‐3 cells.

*ODC1 regulates gene expression in PC‐3 cells. (A) Bubble diagram showing the most enriched GO biological processes for the upregulated and downregulated DEGs. (B) Boxplot showing the expression pattern and statistical difference of DEGs (error bars represent the mean ± SEM. **p < 0.001). (C) Western blot validation of five key gene expression levels (PERP, CAV1, ITGB1, BNIP3, CDKN3) in PC3 cells after ODC1 knockdown, showing PERP upregulation and downregulation of the remaining four genes.

Validation of DEGs in PCa Cells

3.4

To determine whether these DEGs are involved in important biological processes, six DEGs (PERP, BNIP3, CDC20, CAV1, CDKN3, and ITGB1) listed in the top 20 up/down regulated differential genes (FC ≥ 2 or ≤ 0.5, p < 0.05) and related to PCa in our previous studies were selected, and RT‐qPCR was performed in PC‐3 cells and DU145 cells. The results demonstrated that five out of six selected genes (PERP, BNIP3, CAV1, CDKN3, and ITGB1) were significantly different (error bars represent the mean ± SEM. ***p < 0.001, Figures 3B and S2A). In addition, we also used Western blot to verify the expression of these five genes in PC‐3 cells and DU145 cells. The results were consistent with those of RT‐qPCR (Figures 3C and S2B). PERP was the only upregulated DEG selected and was associated with cell adhesion; the remaining four genes were down regulated DEGs (BNIP3, CAV1, CDKN3, and ITGB1) associated with ‘apoptotic process’; ‘angiogenesis’; and ‘positive regulation of angiogenesis’ respectively. CDKN3 was not enriched in any pathway. The results were highly consistent with the RNA‐seq data. Together, these data suggest that ODC1 regulates DEGs associated with angiogenesis, cell apoptosis, and cell adhesion pathways in PCa.

ODC1 Influences the Alternative Splicing (AS) of Genes in PC‐3 Cells

3.5

Given that ODC1 is the rate‐limiting enzyme of polyamine biosynthesis and polyamines have been shown to directly regulate spliceosomal function, we further investigated whether ODC1 knockdown affects alternative splicing patterns. To investigate the role of ODC1 knockdown in AS regulation, RNA‐seq of PC‐3 cells was performed. Overall, 2117 ASEs were found, including: cassette Exon, MXE, IntronR, ES, A5SS&ES, A5SS, A3S&ES, A3SS, 5pMXE, and 3pMXE (Figure 4A). These genes were significantly enriched in biological processes such as ‘phosphorylation’; ‘cellular protein modification process’; ‘protein phosphorylation’; ‘protein sumoylation’; ‘cell cycle’; ‘cell division’; ‘lipid phosphorylation’; ‘synaptic structure or activity regulation’; ‘membrane fusion’; ‘protein stability regulation’; and other biological processes by GO analysis (Figure 4B).

*ODC1 influences the alternative splicing of genes in PC‐3 cells. (A) Bar plot showing the number of significant regulated alternative splicing events (RASEs). X‐axis: RASE number. Y‐axis: types of AS events. (B) Bubble Diagram showing the most enriched GO biological processes obtained for RASGs. (C) Boxplot showing the ratio and statistical differences in RASEs (error bars represent the mean ± SEM. **p < 0.01, **p < 0.001).

Validation of ASs in PC‐3 Cells

3.6

To determine the role of RASGs in biological processes, three genes (YTHDF2, SIRT2, and TRIOBP) were selected from PCa‐related ASEs identified in our previous study and used to perform RT‐qPCR. The results revealed significant differences in two genes (YTHDF2 and SIRT2; error bars represent the mean ± SEM. **p < 0.01, ***p < 0.001), and the two genes were all related to the biological process of cell cycle (Figure 4C). We found that YTHDF2 and SIRT2 showed AS ratio changes consistent with the RNA‐seq results, demonstrating AS regulation of ODC1 in PCa.

Discussion

4

Ornithine decarboxylase 1 (ODC1) serves as the catalytic gatekeeper of polyamine biogenesis, orchestrating pleiotropic physiological functions through its rate‐limiting enzymatic activity. We demonstrated that ODC1 knockdown markedly inhibited cellular proliferation and migration, promoted apoptosis, and showed a trend toward inhibiting cell invasion in PCa cells. These findings are consistent with previous reports that elevated ODC1 expression is associated with aggressive tumour behaviour. Our RNA sequencing analysis revealed a comprehensive transcriptomic profile of ODC1‐regulated genes, revealing marked transcriptional reprogramming and spliceosomal remodelling through differential gene expression and alternative splicing dynamics. Notably, differentially expressed genes (DEGs) were predominantly enriched in pathways related to angiogenesis and cell adhesion, while alternative splicing events (ASEs) were enriched in cell cycle and protein modification pathways. These observations suggest that ODC1 may influence prostate cancer progression through multiple molecular mechanisms, potentially mediated by the AKT signalling pathway. Further exploration of these pathways could provide valuable insights into the role of ODC1 in prostate cancer and identify novel therapeutic targets.

To delineate ODC1 and polyamine regulatory mechanisms, RNA sequencing analysis was conducted in ODC1‐knockdown PC‐3 prostate cancer cells. Among the DEGs identified, five (PERP, BNIP3, CDKN3, CAV1, and ITGB1) were identified by RT to have significant differences (***p < 0.001). In prostate cancer (PCa), CAV1 upregulation under androgen deprivation phosphorylates AR (Ser81), activating androgen‐independent signalling while concurrently promoting lipid synthesis via ACC1/FASN induction. This metabolic shift potentially sustains polyamine anabolism through acetyl‐CoA availability. Enhanced polyamine levels may drive therapeutic resistance [21]. While CAV1 exhibits tumour‐suppressive roles in colorectal cancer, its expression is dysregulated in canine PCa [22, 23]. A previous study reported that VEGF‐mediated pathological angiogenesis is markedly reduced in CAV1 knock‐out mice [24]. Kucharzewska et al. reported polyamine‐dependent regulation via AKT signalling in hypoxia, and that polyamines are involved in angiogenesis [25]. Caveolae are abundant in vascular endothelia and are the main endothelial plasmalemmal vesicles; thus, CAV1 dysregulation affects tumour‐associated permeability and angiogenesis [26]. During cancer development, CAV1 may act through lipid synthesis and angiogenesis, which may be associated with polyamine metabolism.

As a Beta‐integrin family member, ITGB1 is ubiquitously expressed in normal and neoplastic cells. Its overexpression in human malignancies, particularly during prostate cancer metastasis, facilitates integrin‐mediated signalling essential for cellular proliferation, migration, and survival [27]. Pellinen and others showed that ITGB1 can upregulate CAV1, and induces oncogenic TGFβ signalling in PCa [28]. This was consistent with our results; both ITGB1 and CAV1 were downregulated in ODC1‐knockdown cells in our study, and interestingly, CAV1 and ITGB1 were enriched and positively regulated angiogenesis, respectively. This suggests that there may be a strong association between ODC1/ITGB1/CAV1 expression and Pca progression (i.e., downregulation of VEGF, E‐cadherin, and Rho family members) via lipid synthesis and angiogenesis.

BNIP3, elevated in CRPC and linked to poor prognosis, induces apoptosis via mitochondrial Bcl‐xL/beclin interactions [29, 30]. Its blockade restores PI3K inhibitor sensitivity in PTEN‐null CRPC, where PI3K overexpression is prevalent [29, 31]. Therefore, ODC1 may regulate gene expression through regulation of polyamine metabolism (i.e., PI3K/AKT/mTOR signalling pathway). CDKN3 emerged as an additional downregulated DEG in our analyses, functioning as a key modulator of cell cycle progression and genomic replication fidelity in prostate carcinogenesis [32]; however, no pathway enrichment was identified by GO analysis.

PERP, the only upregulated differentially expressed gene (DEG) identified in our study, is a tumour suppressor protein composed of four transmembrane regions, two extracellular loop domains, and three cytoplasmic domains. Multiple studies have shown that PERP plays an important role in cell–cell adhesion, and knocking out PERP leads to cell adhesion defects [33, 34, 35]. In our study, PERP was significantly augmented in ODC1‐knockdown groups compared to control groups and was enriched in cell adhesion through GO analysis, consistent with the previous study. PERP may also participate in apoptosis and cell cycle progression. Flamigni et al. showed that putrecine, spermidine, and spermine participate in programmed cell death, supporting a role for polyamines in the regulation of p53 gene expression [36]. However, this could not be confirmed in our study, possibly due to insufficient samples or crosstalk between cell adhesion and apoptosis, and PERP being mainly involved in cell adhesion.

ODC1‐mediated polyamine depletion may influence alternative splicing through direct regulatory mechanisms. Recent evidence demonstrates that polyamines interact with acidic phosphorylatable motifs in SF3 spliceosomal complex members, preventing kinase‐mediated phosphorylation and thereby modulating splice site selection [37]. Upon ODC1 knockdown and subsequent polyamine depletion, reduced SF3 protection from phosphorylation likely alters the splicing landscape, explaining the 2117 alternative splicing events we observed enriched in cell cycle regulation. Transcriptomic analysis of alternative splicing events in ODC1‐depleted cells identified significant differential splicing of YTHDF2 and SIRT2 pre‐mRNAs, both demonstrating enrichment in cell cycle regulation and post‐translational modification pathways. YTHDF2 and its upstream co‐factors promote prostate cancer proliferation/migration through m6A‐dependent degradation of LHPP and AKX‐1 to regulate AKT phosphorylation (S473/T308) [38], while concurrently delaying G2 phase entry via m6A‐YTHDF2‐mediated cell cycle regulation to suppress adipogenesis [38, 39]. As noted, fatty acids supply acetyl‐CoA for polyamine synthesis, regulated by AKT under hypoxia. Analyses revealed YTHDF1 ASEs demonstrated enrichment in cell cycle regulation and protein modification signalling pathways, suggesting ODC1 potentially modulates YTHDF2 expression through adipogenic mechanisms via the PI3K/AKT axis. SIRT2 can regulate lipid synthesis and fatty acid oxidation via deacetylation of the nuclear transcription factor FOXO1, and bind to peroxisome proliferators‐activated receptor‐γ (PPAR‐γ), thus inhibiting transcriptional activity and affecting fat hydrolysis [40]. Fat hydrolysis can provide acetyl coenzyme A, and acetylated polyamines can return to polyamine metabolism with PAOX and acetyl coenzyme A, which may explain the elevated levels of polyamines in tumour cells. In the cell cycle, SIRT2 can activate the anaphase‐promoting complex/cyclosome (APC/C) and regulate the progression of mitosis, thus inhibiting cancer progression [41].

Core oncogenic pathways implicated in polyamine metabolic reprogramming encompass the PTEN‐PI3K‐mTORC1 signalling axis, alongside Wnt/β‐catenin, AKT, and RAS signalling cascades [6]. In our study, CAV1, BNIP3, and YTHDF2 were all more or less correlated with the AKT pathway. ITGB1, as a positive regulator of CAV1, may also participate in the AKT pathway. Thus, we speculate that AKT may be a key factor regulating polyamines and PCa progression.

This study has several limitations. It was conducted solely in vitro using prostate cancer cell lines (PC‐3 and DU145), which may not reflect the heterogeneity of clinical tumours. In vivo validation is lacking, limiting the translational relevance. Moreover, although numerous DEGs and ASEs were identified, only a few were functionally validated. The mechanistic link between ODC1 and the AKT signalling pathway remains correlative. An important limitation is that while we documented both differential gene expression and alternative splicing changes following ODC1 knockdown, direct mechanistic validation of polyamine‐mediated splicing regulation (such as SF3 complex phosphorylation analysis or polyamine rescue experiments) was not performed. Future studies should measure cellular polyamine levels post‐knockdown and determine whether exogenous polyamine supplementation can rescue the observed splicing alterations. Further multidisciplinary investigation remains imperative to validate these observations and delineate ODC1‐targeted therapeutic paradigms in PCa.

Conclusions

5

ODC1 knockdown in PCa cells markedly suppressed proliferative and migratory capacities while inducing apoptotic activity and reducing invasive potential. However, the effect was not significant, suggesting that ODC1 may play a key role in cellular function. The RNA‐seq data revealed that DEGs were mainly enriched in angiogenesis and cell adhesion, whereas ASs were mainly enriched in the cell cycle and protein modification pathways. Our results indicate that ODC1 may regulate polyamine biosynthesis, thereby influencing gene expression and alternative splicing followed by the progression of PCa, which may be regulated by the AKT pathway. Thus, Akt and polyamines may represent promising therapeutic targets. Thus, further research on ODC1‐regulated DEGs and ASs would be helpful to elucidate the molecular mechanism underlying PCa and to identify therapeutic targets.

Author Contributions

Jian Ma: conceived and designed the research framework, led the data collection process, conducted the majority of data analysis, and was primarily responsible for writing the initial draft of the manuscript. Peng Chen: supervised the overall research project, provided critical guidance throughout the study, participated in the data interpretation, and revised the manuscript for important intellectual content. Ting Pan: assisted in the experimental design, contributed to the data collection, and participated in the initial data analysis. Shengli Sun: helped with data processing, offered valuable suggestions during the data analysis phase, and participated in the review and editing of the manuscript. Musitapa Mutalifu: participated in the fieldwork, contributed to the collection of primary data, and provided insights relevant to the research topic. Wei Yang: conducted literature reviews, provided theoretical support for the research, and participated in the manuscript revision. Yue Niu: assisted in the experimental operations, contributed to the quality control of data collection, and participated in the preparation of figures and tables. All authors have read and approved the final version of the manuscript, and agree to be accountable for all aspects of the work.

Funding

This work was supported by Provincial Natural Science Foundation General Project (Project Number: 2022D01C285).

Disclosure

Cell line authentication and mycoplasma detection: All cell lines employed in this study were expert‐certified prior to experimentation. Each cell line underwent routine testing and was confirmed free from mycoplasma contamination.

Ethics Statement

The authors have nothing to report.

Consent

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: ODC1 knockdown promotes malignant phenotype in DU145 cells. (A) RT‐qPCR detection of ODC1 knockdown efficiency in DU145 cells. (B) WB detection of ODC1 knockdown efficiency in DU145 cells. (C) CCK8 assay after ODC1 knockdown demonstrates reduced proliferation capacity in DU145 cells. (D, E) ODC1 knockdown diminishes DU145 cell migration capacity, with negligible impact on invasion. (G) Flow cytometry reveals increased apoptosis in DU145 cells following ODC1 knockdown.

Figure S2: (A, B) RT‐qPCR and WB validation of expression alterations in five key genes after ODC1 knockdown in DU145 cells, showing PERP upregulation and downregulation in the remaining genes.