Clinical and molecular data for LUAD were acquired from The Cancer Genome Atlas (TCGA) database, and accessed through the GDC Data Portal (https://portal.gdc.cancer.gov/). Data preprocessing and analysis were conducted using Perl (v5.30.0) and R (v4.3.3). Differential expression analysis and clinical correlation assessments were conducted using the limma package in R. Survival analysis and Cox regression modeling (univariate and multivariate) were performed using the survival package. Functional enrichment analyses (KEGG and GSEA) were performed with thresholds of p < 0.05 and |log2FC| > 1. Tumor microenvironment (TME) profiles were analyzed using the ESTIMATE package, while immune cell infiltration levels were quantified via CIBERSORT to deconvolute specific immune cell subsets. Finally, the analysis of pan-cancer differences was done based on the TIMER database (http://timer.comp-genomics.org).

The clinicopathological data and tissue specimens (histologically verified tumor and NAT) from 80 patients with LUAD were analyzed in this study. All participants underwent curative surgical resection at the Affiliated Hospital of North Sichuan Medical College between January 2019 and December 2020. Ethical approval for this retrospective cohort study (approval number: 2024ER270 - 1) including a waiver of informed consent granted by the Ethics Committee.

IHC for all formalin formalin-fixed and paraffin-embedded (FFPE) tissue sections was conducted following established protocols (26). Briefly, FFPE tissue sections underwent standard processing, including incubation with primary antibodies followed by biotin-conjugated secondary antibodies, and were finally mounted using neutral balsam. Two independent associate chief pathologists evaluated IHC results. IHC results were quantified using the Immunoreactive Score (IRS) system, which combines staining intensity and percentage of positive cells into a composite score (27). For animal tissues, we analyzed staining patterns by calculating the ratio of integrated optical density (IOD) to positive tissue area.

The human LUAD cell lines (H1299 and A549) were obtained from Pricella (Wuhan, China). The H1299 was cultured in RPMI 1640 medium containing 10% fetal bovine serum (FBS), while the A549 cell line was maintained in Ham’s F - 12K medium with 10% FBS. The normal bronchial epithelial cell line BEAS - 2B was obtained from Biocode (Zhejiang, China), and was grown in high-glucose DMEM containing 10% FBS. Each cell line was maintained in a 37 °C incubator with 5% carbon dioxide (CO₂) and high moisture levels.

Lentiviral vectors for CCDC58 knockdown (shCCDC58) and negative control vectors (shNC), both containing puromycin resistance markers and the green fluorescent protein (GFP) gene, were acquired from GENE (Shanghai, China). The target sequences were:

Following the manufacturer’s protocol, we performed lentiviral transduction and selected transfected cells using puromycin (4 μg/mL for 1 week, then 2 μg/mL for maintenance). Three stable CCDC58 knockdown groups (shCCDC58-1, shCCDC58-2, and shCCDC58-3) and one negative control group (shNC) were employed in this study.

According to the manufacturer’s guidelines, we conducted total RNA extraction, cDNA synthesis, and qPCR using the following kits (Accurate Biotechnology, Hunan, China): the SteadyPure Universal RNA Extraction Kit for RNA isolation, the Evo M-MLV RT Premix Kit for cDNA synthesis, and the SYBR Green Pro Taq HS Premixed qPCR Kit for amplification. Using the 2^(-ΔΔCT) method, we quantified relative gene expression (28). The oligonucleotide primer sequences designed for amplification of the target gene’s coding DNA sequence (CDS) are listed below:

WB was conducted following established protocols (29). Briefly, primary antibody incubation was performed overnight on the prepared membranes, using the following specific antibodies: CCDC58 (OmnimAbs, 1:1000); β-ACTIN and GAPDH (Affinity, 1:5000 each); BAX, p-AKT, BCL - 2, AKT, N-cadherin, E-cadherin, and Vimentin (all HuaBio; 1:10000, 1:1000, 1:4000, 1:4,000, 1:5000, 1:5000, and 1:20,000, respectively); and PI3K and p-PI3K (Zenbio, 1:1500 each). Subsequently, the membranes were incubated to a horseradish-peroxidase-conjugated goat anti-rabbit IgG secondary antibody (dilution, 1:5000) from FineTest (Wuhan, China). Protein bands were visualized using an ECL substrate kit (Epizyme, Shanghai, China) on a ChemiDoc™ XRS+ system (Bio-Rad, USA) and quantified with ImageJ software.

Cell proliferation was assessed via the CCK - 8 kit (APExBIO, USA) in 96-well plates containing 5,000 cells per well. Cell viability assessments were conducted at 0, 24, 48, and 72 hours. Following the manufacturer’s protocol, 10 μL of CCK - 8 solution was combined with 100 μL of medium in each well. Plates were incubated for 2 hours at 37 °C under 5% CO₂ with controlled humidity, after which optical density was quantified at 450 nm.

In colony formation experiments, cells (1,000 per well) were seeded in 6-well plates. The cultures were incubated at 37 °C in a 5% CO₂-enriched humidified environment for 7–14 days. Visible colonies were washed with ice-cold PBS, fixed with 4% paraformaldehyde (PFA) for 30 minutes, and stained for 20 minutes using 0.1% crystal violet. Colonies containing 50 or more cells were counted with ImageJ software.

Transwell chambers with 8 μm pores (Corning, USA) were used to analyze cell migration. In short, 50,000 cells in a serum-free medium were placed in the upper chambers, with the lower chambers filled with complete medium containing 10% FBS to act as a chemoattractant. After being incubated for 24 hours at 37 °C in a humidified atmosphere containing 5% CO₂, migrated cells were fixed with 4% PFA and stained with 0.1% crystal violet. Microscopic images were acquired and analyzed using ImageJ to quantify cell migration.

In wound healing experiments, 5 × 10⁵ cells per well were plated into 6-well plates and grown to 90% confluency before initiating the assay. Uniform wounds were then introduced into the monolayer using a sterile 200 μL pipette tip. After replacing the medium with serum-free conditions, we monitored wound closure by acquiring images at 0 and 48 hours post-wounding. Using ImageJ, we quantified cell migration by measuring the wound distance over time.

The cell apoptosis assay was performed when the cell confluence reached 70%. After harvesting, the cells were washed two times with ice-cold PBS and resuspended in Binding Buffer. Samples were analyzed with a flow cytometer (ACEA Bioscience, USA) after being incubated in the dark for 10 minutes at room temperature with Annexin V-APC/7-AAD staining solution (KeyGEN BioTECH, China).

For cell cycle profiling, harvested cells were washed with PBS, fixed in 70% ethanol, and treated with RNase and propidium iodide (KeyGEN BioTECH, China) for 60 minutes. Samples were analyzed using a flow cytometer (ACEA Bioscience, USA).

Four-week-old male BALB/c nude mice were obtained from SPF Biotechnology (Beijing, China) and randomly divided into two groups (n = 5). A549 cells transfected with negative control shRNA (shNC) or CCDC58-targeting shRNA (shCCDC58-3) were subcutaneously injected into the right flank at 3 × 10⁶ cells per mouse. The tumor’s growth was tracked every three days using caliper measurements, and its volume was determined by the formula (length × width²)/2. After four weeks, mice were euthanized humanely, and xenografts were excised for downstream analysis. Prior to commencement, the study received full ethical clearance (NSMC Ethical Animal Review [2024] 127) from North Sichuan Medical College Animal Ethics Committee.

R statistical software (version 4.3.3) and GraphPad Prism (version 9.5) were used for statistical analyses. All experiments included three independent biological replicates, with quantitative data presented as mean ± standard deviation. Intergroup differences were statistically assessed using Student’s t-test, whereas multigroup comparisons required application of one-way ANOVA. To examine potential correlations between CCDC58 expression levels and clinicopathological characteristics, Pearson’s χ² test was implemented. A p-value less than 0.05 was considered statistically significant, with significance levels indicated as: *p < 0.05, **p < 0.01, ***p < 0.001.

Pan-cancer analysis identified CCDC58 overexpression in LUAD ( Figure 1A ), which was validated in TCGA datasets showing significantly higher CCDC58 levels in LUAD versus NAT ( Figure 1B ). IHC of 80 paired LUAD/NAT confirmed this overexpression pattern ( Figure 1C ). CCDC58 expression was significantly upregulated in LUAD cell lines (H1299, A549) compared to normal bronchial epithelial BEAS - 2B cell line, with consistent increases observed at both mRNA and protein levels ( Figures 1D, E ). These consistent findings across bioinformatics, clinical samples, and cells suggest that CCDC58 plays a critical role in LUAD pathogenesis.

Analysis of the TCGA dataset using bioinformatics revealed that higher CCDC58 expression was linked to poor overall survival ( Figure 2A , Supplementary Figure S1 ). CCDC58 emerged as an independent predictor of prognosis in both univariate and multivariate Cox regression analyses, demonstrating clinical relevance comparable to tumor stage ( Figures 2B, C ). CCDC58 expression patterns demonstrated significant clinicopathological correlations, with particularly elevated levels observed in metastatic (M1) compared to non-metastatic (M0) cases ( Figure 2D ), in N2 versus N0 cases ( Figure 2E ), and in T2/T4 versus T1 cases ( Figure 2F ). The protein was upregulated in advanced-stage (III/IV) compared to early-stage (I) disease ( Figure 2G ) and exhibited higher expression in male patients ( Figure 2H ). No significant age-dependent differences were observed ( Figure 2I ). In our institutional LUAD cohort, CCDC58 expression levels showed significant associations with T stage and overall stage, but not with N stage, gender, age, or smoking history ( Table 1 ). The partial discordance with TCGA data may reflect our cohort’s N0 predominance and limited sample size. In a word, CCDC58 may be a key clinical marker for tumor growth and metastasis, and can independently predict a poor prognosis for patients in LUAD.

KEGG pathway analysis demonstrated significant enrichment of CCDC58-associated genes in cell cycle regulation pathways ( Figure 3A ). Enrichment of cell cycle-related genes was demonstrated in LUAD specimens with elevated CCDC58 expression through GSEA ( Figure 3B ), consistent with prior observations. Furthermore, CCDC58 expression levels significantly influenced TME composition, as evidenced by alterations in Stromal, Immune, and ESTIMATE scores ( Figure 3C ). Immune cell infiltration analysis of 10 cell subtypes showed distinct correlation patterns with CCDC58 expression levels ( Figure 3D ). Specifically, higher CCDC58 expression positively correlated with increased infiltration of activated Macrophages M1, T cells CD4 memory activated, T cells follicular helper, T cells CD8 and Macrophages M0, while showing negative correlations with B cells memory, Monocytes, Dendritic cells resting, Mast cells resting and T cells CD4 memory resting ( Figure 3E ). These findings suggest CCDC58 promotes LUAD progression through dual mechanisms, including influence of cell cycle and immunomodulation of TME.

We generated three lentiviral shRNA constructs targeting CCDC58 (shCCDC58-1, -2, -3) in LUAD cell lines. Based on validation by qPCR and WB, shCCDC58–3 showed optimal knockdown efficiency ( Figures 4A–D ) and was selected for functional studies. CCDC58 knockdown produced significant anti-tumor effects, reducing colony formation capacity ( Figure 4E ) and decreasing cell proliferation ( Figures 4F, G ) compared to controls. These results establish CCDC58 as a critical regulator of LUAD cell growth and clonogenicity, highlighting its therapeutic potential.

Functional assays demonstrated that CCDC58 knockdown significantly impaired cell motility, reducing wound closure ( Figures 5A, B ) and decreasing migration capability ( Figures 5C, D ) compared to controls. Considering the pivotal regulatory role of EMT in tumor aggressive growth and distant metastasis in malignancies (30), we analyzed EMT marker expression by WB. CCDC58 knockdown influenced EMT, characterized by E-cadherin upregulation concurrent with N-cadherin and vimentin downregulation ( Figure 5E ). These findings demonstrate that CCDC58 promotes LUAD cell motility through EMT regulation, suggesting its potential role in metastatic progression.

Cell cycle analysis by flow cytometry demonstrated that CCDC58 knockdown induced G1-phase arrest in both H1299 and A549 cell lines ( Figures 6A, B ). Concomitantly, apoptotic rates were significantly increased in these cell lines ( Figures 6C, D ). To understand how CCDC58 influences apoptosis at the molecular level, we used WB to study the expression of proteins associated with apoptosis. In results, knocking down CCDC58 suppressed the expression of anti-apoptotic protein BCL - 2 and amplified the expression of pro-apoptotic protein BAX ( Figure 6E ). These results indicate that CCDC58 not only affects the cell cycle but also plays a significant role in apoptosis.

To investigate how CCDC58 influences LUAD progression, we examined its effect on tumor-related signaling pathways. Based on previous studies linking CCDC58 to the PI3K/AKT pathway (25), we analyzed key pathway components by WB. CCDC58 knockdown significantly reduced p-PI3K and p-AKT ( Figures 7A, B ), suggesting CCDC58 promotes LUAD progression through influencing PI3K/AKT pathway activity.

To investigate CCDC58’s role in LUAD tumorigenesis, we developed a xenograft model by subcutaneously injecting stably transfected A549 cells into nude mice which were randomized into control (shNC) and CCDC58-knockdown (shCCDC58-3) groups ( Figures 8A, B ). CCDC58 suppression significantly inhibited tumor growth, reducing tumor volume ( Figures 8C, D ) and decreasing final tumor weight ( Figure 8E ) compared to controls. IHC analysis demonstrated that CCDC58 knockdown reduced proliferation (decreased Ki-67 protein expression) while promoting apoptosis (increased BAX and decreased BCL - 2 protein expression) ( Figure 8F ). These results establish CCDC58’s dual oncogenic function in LUAD by simultaneously promoting proliferation and suppressing apoptosis.