This study utilized the Xiantao Academic platform (https://www.xiantaozi.com/) (21) to conduct differential expression and clinical correlation analyses of PPP2R1A in LUAD. Data associated with 598 mRNA LUAD samples, along with 59 matched adjacent normal tissue samples, all acquired from The Cancer Genome Atlas (TCGA database https://tcga-data.nci.nih.gov/). The mRNA expression data from TCGA were normalized as Transcripts Per Million (TPM). The parameters for differential expression analysis were set as follows: 1) tumor type selected as ‘LUAD’, transcriptomic data format selected as ‘TPM’, and other parameters set to default values. Clinical and pathological variables encompassed Gender, T stage, N stage, M stage, and overall stage classification. As all data utilized in this study were sourced exclusively from the publicly available TCGA database, ethics committee or institutional review board approval was not required. The analysis strictly adhered to TCGA publication guidelines.

In this study, utilizing the cBioPortal platform (https://www.cbioportal.org/) (22), we interrogated alteration frequency, copy number variations, and mutation types of PPP2R1A across malignancies within the TCGA pan-cancer atlas, defining its comprehensive genetic alteration landscape.

Using the pan-cancer analysis platform GEPIA2 (http://gepia2.cancer-pku.cn/) (23), we systematically retrieved and downloaded the top 100 gene lists with significant co-expression relationships with the PPP2R1A in lung adenocarcinoma. Using the built-in ‘correlation analysis’ module in GEPIA2, the top three most highly correlated genes in the co-expression network were validated using scatter plots based on TCGA tumor data.

Systemic immune microenvironment profiling was performed by using the TIMER 2.0 tool (http://timer.cistrome.org/) (24, 25). The platform integrates TCGA pan-cancer data encompassing transcriptomes of 16,283 tumor samples (representing 11,300 patients) with immune cell infiltration estimates across 33 malignancies. We leveraged two analytical modules: (1) the ‘timer-Gene’ module was used to assess PPP2R1A expression correlations with immune infiltration levels specifically in lung adenocarcinoma (LUAD), and (2) the ‘Immune-gene’ module was used to evaluate PPP2R1A-immune infiltration associations in the TCGA cohort.

In this study, we constructed the protein interaction network of PPP2R1A using the STRING database (https://string-db.org/) (26). The specific workflow is as follows: first, gene name input: PPP2R1A, Species selection: Homo sapiens. The interaction sources were selected as ‘text mining and experiments’, interaction thresholds were defined as follows: (1) minimum interaction score = 0.150 (low confidence cutoff), and (2) maximum interactor displays limit = 50 proteins.’ The remaining parameters were set to the system default configurations.

Prognostic significance of PPP2R1A in lung adenocarcinoma was interrogated via the Kaplan-Meier Plotter (https://kmplot.com/) (27). Patients were stratified into high- and low-expression groups by median expression of PPP2R1A, the relationship between PPP2R1A expression and Failure-free survival (FP), Overall survival (OS), and Post-progression survival (PPS) was subsequently analyzed to explore the prognostic value of PPP2R1A in lung adenocarcinoma. Statistical significance was assessed using the log-rank test, and hazard ratios (HR) with 95% confidence intervals (CI) were calculated.

Utilize CPTAC (https://cptac-data-portal.georgetown.edu/) The “Protein Expression Analysis with cProstite” module of the database systematically analyzes the expression characteristics of PPP2R1A protein in LUAD. By integrating high-depth mass spectrometry proteomics and matching transcriptome data, the relative protein abundance and mRNA expression levels of PPP2R1A in tumor tissue and adjacent non tumor tissue were quantitatively compared, and its clinical significance was further evaluated. To more clearly illustrate the relationship between PPP2R1A protein abundance and mRNA expression levels, we have included a scatter plot depicting the protein abundance and mRNA expression levels in tumor and adjacent non-tumor tissues. We have also calculated the correlation between the two. The proteomic data were normalized as log₂ (TMT ratio), with median centering across samples, and the accompanying mRNA data were normalized as Counts Per Million (CPM), followed by log₂ transformation.

The A549 lung adenocarcinoma cell line used in this study (28, 29) was obtained from the Shanghai Cell Bank, Chinese Academy of Sciences. Prior to subsequent experiments, a subpopulation of A549 cells was subjected to CRISPR/Cas9-mediated PPP2R1A gene knockdown to establish a gene-modified cell model, which was used for comparative analysis with the parental wild-type A549 cells. To ensure the reliability of the gene-modified model, the knockdown efficiency was verified by quantitative real-time polymerase chain reaction (qRT-PCR) and Western blotting. Importantly, this PPP2R1A knockdown was confirmed to be stable; the modified cells were continuously passaged and cultured, and the persistent downregulation of PPP2R1A expression was monitored and validated for at least 10 consecutive passages to exclude the possibility of gene expression recovery during long-term culture.

Cells (both wild-type and stable PPP2R1A-knockdown A549 cells) were cultured in a humidified incubator maintained at 37 °C with 5% CO₂. The culture medium consisted of high-glucose Dulbecco’s Modified Eagle Medium (DMEM) (Gibco, C11995500BT), supplemented with 10% fetal bovine serum (Oricell, FBSSR-01021) and 1% penicillin-streptomycin (Gibco, 15140122). All cell-based assays were performed with triplicate replicates.

Log-phase cells were harvested for total RNA extraction using TRIzol reagent (SparkZol Reagent, AC0101-B). cDNA synthesis was then performed using reverse transcription reagents (Tolobio, 22107). After the addition of the corresponding primers, cDNA, and fluorescent dye (Tolobio, 22204-1), the reaction mixture was transferred to an eight-well plate and analyzed using the Roche LightCycler 96 system to quantify the expression levels of target genes. Gene expression fold changes for each triplicate sample set were calculated using the 2^−ΔΔCT method, with normalization against GAPDH. The primer sequences used were: PPP2R1A-F: CGGAGACCCCAACTACCTGC; PPP2R1A-R: GCCACATTGAAGCGGACATT.

Cell proliferative potential in lung adenocarcinoma was assessed using two complementary approaches: CCK-8 viability assays and clonogenic formation assays. Log-phase cells were trypsinized (25200056, Gibco) and plated at 300 cells per well in six-well plates (140675, Thermo). For the clonogenic assay, cells were cultured for 14 days, allowing colonies to form into morphologically distinguishable clusters. The six-well plates were washed twice with PBS (Gibco, C10010500BT) to remove suspended cells and residual substances. Cells were then fixed with 4% paraformaldehyde (Biosharp, BL539A) for 30 minutes, followed by staining with crystal violet (Solarbio, G1063) for 10 minutes. After drying, images were captured, and the stained colonies were counted and statistically analyzed.

For the CCK-8 assay, 2,000 cells in the exponential growth phase were carefully seeded into each well of a 96-well plate (Thermo, 167008) to ensure uniformity and optimal conditions for subsequent analysis. Cells were cultured for 24, 48, 72, and 96 hours. Following incubation, 10 μL of CCK-8 reagent (HY-K0301-500T, MCE) was added to each well. Absorbance was measured at 450 nm using an INFINITE 200 PRO microplate reader (3005030301), and growth kinetic curves were generated.

The day before the experiment, logarithmic-phase A549 cells were trypsinized and resuspended as single-cell suspension, then seeded into 6-well plates at 3×10⁵ cells per well and cultured overnight until 90–100% confluence was reached. The cells covering the plate were removed, and 200 μL of autoclaved wall head solution was applied to draw a straight line perpendicular to the six-well plate. After rinsing with PBS, serum-free medium was added. Images were captured under a microscope (Axio Vert A1, 07060200), and the initial scratch area (A_0h) was recorded. After 48 hours of culture, images were taken at the same location. The migration kinetics were assessed by calculating the wound closure percentage using the following formula: Closure rate = [(A_0h - A_48h)/A_0h] × 100, where A_x represents the wound area at X hours post-scratching.

To assess migratory and invasive capacities, Transwell assays were conducted. Log-phase cells were harvested and resuspended in serum-free medium at a concentration of 1 × 10⁵ cells/mL. The Transwell system (TCS003024, JET) was set up with 200 μL of serum-free cell suspension in the upper compartment and 600 μL of complete medium (10% serum) in the lower compartment. The system was then incubated at 37 °C for 24 hours to allow migration and invasion. After incubation, the chamber was removed and fixed with paraformaldehyde for 30 minutes, followed by staining with 0.5% crystal violet for 5 minutes. The chamber was rinsed with PBS, and excess cells were removed using a cotton swab. Multiple fields of view were captured under a microscope, and statistical analysis was performed. For the invasion assay, 100 μL of matrix gel (Corning, 356234) was added to the Transwell insert before the experiment, allowed to solidify in a 37 °C incubator, and then the subsequent steps were performed.

All statistical analyses in this study were performed using SPSS statistical software (version 26.0; IBM, NY, USA). Data were presented as mean ± standard deviation (Mean ± SD) or median (Median) and interquartile range (IQR). Intergroup comparisons were statistically evaluated using either the Student’s t-test or the Mann-Whitney U test, with the selection contingent upon the underlying data distribution, while for comparing multiple groups, one-way analysis of variance (ANOVA) or Kruskal Wallis test is used. To evaluate the prognostic significance of PPP2R1A in lung adenocarcinoma (LUAD), Kaplan-Meier survival curves were generated with subsequent intergroup comparisons conducted via log-rank testing. Hazard ratios (HR) and 95% confidence intervals (CI) were derived from the Cox model. To systematically explore the potential association between PPP2R1A expression levels, clinical characteristics, and patient survival prognosis, used Cox proportional hazards regression model to perform multivariate survival analysis on survival outcome data. Differential expression analysis of PPP2R1A was performed via the Xiantao platform using TCGA-derived LUAD and matched normal cohorts, with expression disparities quantified by distribution-appropriate statistical tests (Student’s t-test or Mann-Whitney U test). Correlation analysis was performed to calculate the correlation between PPP2R1A and other genes using the Spearman correlation coefficient, employing either the Pearson or Spearman method depending on the distribution characteristics of the data. For multiple comparisons in subgroup analyses, False Discovery Rate (FDR) correction was applied where appropriate. Statistical significance was established at P < 0.05, with a more stringent threshold of P < 0.001.

Using the Xiantao online tool to assess the expression levels of PPP2R1A mRNA, we detected a significant increase in PPP2R1A expression in 26 cancers (Figure 1A). Unpaired analysis results showed a significant upregulation of PPP2R1A in lung adenocarcinoma. (P < 0.05) (Figure 1B). Paired analysis results were consistent with unpaired results, further confirming that PPP2R1A expression levels were significantly increased in lung adenocarcinoma samples (P < 0.05) (Figure 1C). Subsequently, Further assessment of PPP2R1A’s diagnostic potency in lung adenocarcinoma yielded an AUC of 0.593 (95% CI: 0.54-0.64), indicating modest discriminative capacity for malignant identification (Figure 1D). Notably, analysis of the independent CPTAC dataset revealed a contrasting pattern, with both PPP2R1A protein (TMT log₂ ratio) and mRNA (gene expression count) levels being significantly lower in LUAD tumor tissues compared to adjacent normal tissues (P < 0.001; Supplementary Figures 1A, B). Furthermore, the scatter plot in (Supplementary Figure 1C) illustrates the correlation between PPP2R1A mRNA and protein levels across tumor and adjacent normal tissues. The total correlation coefficient was -0.5090, indicating a moderate negative correlation between mRNA and protein levels. This discrepancy between TCGA and CPTAC datasets may stem from differences in sample cohorts, normalization methodologies, or post-transcriptional regulatory mechanisms.

We divided patients into high and low expression groups based on PPP2R1A expression levels. Kaplan-Meier (KM) survival analysis showed that PPP2R1A overexpression in patients with lung adenocarcinoma was significantly associated with poor FP and OS outcomes. Subsequently, we analyzed the relationship between PPP2R1A expression and adverse FP and OS outcomes in different clinical subgroups (Figures 2A-C). The results showed that patients with stage I LUAD exhibited the strongest PPP2R1A-related survival disadvantage, suggesting a higher carcinogenic impact in early disease (Figures 2D, E).

Subgroup analysis based on clinical pathological parameters further revealed the relationship between PPP2R1A and the clinicopathological features of lung adenocarcinoma. No statistically significant correlation was observed between PPP2R1A expression and either gender or T stage among lung adenocarcinoma patients (Figures 3A, B). The expression level of PPP2R1A was closely associated with N stage in lung adenocarcinoma patients, compared with N0 and N1 stage lung adenocarcinoma patients, PPP2R1A expression levels were significantly higher in N2 stage lung adenocarcinoma patients (Figure 3C). While PPP2R1A expression levels were higher in M1 stage lung adenocarcinoma patients, this difference did not reach statistical significance (Figure 3D). PPP2R1A expression was significantly associated with lung adenocarcinoma stage; advanced-stage (III-IV) LUAD patients exhibited significantly higher PPP2R1A expression compared to early-stage (I-II) cases (Figure 3E).

A protein-protein interaction (PPI) network for PPP2R1A was constructed through the STRING database, revealing close interactions with members of the protein phosphatase complex (e.g., PPP2CA, PPP2R2A) and INTS family proteins (e.g., INTS6, INTS1) (Figure 4A). In the TCGA lung adenocarcinoma cohort, PPP2R1A expression demonstrated significant positively correlated with PRPF31 (R = 0.72), SCAF1 (R = 0.72), and PTOV1 (R = 0.71) (Figures 4B-D). A Venn diagram shows that there is one common gene, STRN4, shared between the interacting genes and the expression-related genes (Figure 4E). In different tumors, PPP2R1A was significantly positively correlated with PRPF31, PTOV1, and SCAF1 (Figure 4F). To predict the functions of PPP2R1A-related genes, we performed Gene Ontology (GO) functional enrichment analysis on 150 PPP2R1A-related genes obtained from screening. Gene Ontology (GO) enrichment analysis revealed that the three most enriched GO terms in the three processes (BP, CC, MF) were ‘peptide serine dephosphorylation,’ ‘chromosome centromere region,’ and ‘chemokine receptor binding,’ indicating that PPP2R1A regulates protein dephosphorylation, cell cycle, and metabolic pathways, and synergistically drives lung adenocarcinoma progression with key genes such as PRPF31 and SCAF1 (Figures 5A-C). KEGG pathway enrichment analysis revealed that key pathways involving PPP2R1A-associated genes include ‘cell cycle,’ ‘cell senescence,’ and ‘PPAR signaling pathway,’ suggesting that PPP2R1A promotes tumor progression by regulating proliferation and metabolism (Figure 5D). Finally, we used the cBioPortal platform to investigate the genetic mutation characteristics of PPP2R1A in lung adenocarcinoma. Based on TCGA-LUAD data, the mutation frequency of PPP2R1A in lung adenocarcinoma was only 2.3%, primarily consisting of missense mutations in the phosphatase domain (e.g., Q372L) (Figures 6A-C).

To investigate PPP2R1A’s immunomodulatory potential, we assessed its co-expression patterns with tumor-infiltrating immune cell signatures across the LUAD. Immunoinfiltration analysis revealed PP2R1A showed significant positive correlations with macrophage and CD4 T cells infiltration (P < 0.05), but not significantly correlated with the infiltration level of myeloid dendritic cells, neutrophils, B lymphocytes and CD8 T cells (Figure 7A). To further characterize the macrophage polarization state associated with PPP2R1A, we analyzed its correlation with canonical M1 (NOS2, IL6, TNF, CD86, CD80) and M2 (TGFB1, CD163, MRC1, ARG1, IL10) macrophage markers. Contrary to our initial hypothesis, PPP2R1A expression demonstrated a significant negative correlation with the majority of M1 markers (e.g., CD80: R = -0.194, P < 0.001; CD86: R = -0.149, P < 0.001). Notably, it also showed significant negative correlations with several key M2 markers, including IL10 (R = -0.181, P < 0.001) and ARG1 (R = -0.129, P < 0.01). A notable exception was a significant positive correlation with the M2-related cytokine TGFB1 (R = 0.216, P < 0.001) (Supplementary Figures 2A, B). This complex correlation pattern suggests that PPP2R1A is linked to a specific immunosuppressive context, potentially defined more by TGFB1 signaling than a classical M2 polarization.

Furthermore, we investigated the relationship between PPP2R1A and regulatory T cell (Treg) infiltration by examining key Treg signature genes (FOXP3, IL2RA, CCR8, CTLA4). Strikingly, PPP2R1A expression was significantly negatively correlated with the functional Treg markers CTLA4 (R = -0.230, P < 0.001) and CCR8 (R = -0.148, P < 0.001), as well as with IL2RA (CD25) (R = -0.104, P < 0.05), while showing no correlation with the master transcription factor FOXP3 (R = 0.001, P > 0.05) (Supplementary Figure 2C). This inverse relationship with critical Treg effector molecules suggests that high PPP2R1A expression may be associated with an altered, or potentially dysfunctional, Treg compartment within the LUAD tumor microenvironment.

Additionally, we investigated the relationship between PPP2R1A expression levels and the tumor immune microenvironment. First, we divided 598 patients into high and low expression groups according to the expression level of the PPP2R1A in the TCGA dataset. Subsequently, we performed comparative analysis of immune cell infiltration patterns between PPP2R1A-high and -low expression groups across 24 immune cell subtypes. In the high PPP2R1A expression group, sixteen cell types, including B cells, activated dendritic cells (ADC), cytotoxic cells, and T cells, were significantly elevated (Figure 7B). Additionally, sixteen of the 24 immune cell subtypes were negatively correlated with PPP2R1A expression (Figure 7C). Next, we employed TIMER2 to analyze the correlation between PPP2R1A gene expression patterns and immune infiltration levels across various tumor types in the TCGA. This analysis revealed consistent and statistically significant positive correlations between PPP2R1A expression and cancer-associated fibroblast (CAF) infiltration in multiple tumor types, including: head and neck squamous cell carcinoma, hepatocellular carcinoma, lung adenocarcinoma, malignant mesothelioma, and other tumor types (Figure 8A). Subsequently, a scatter plot generated using the most relevant algorithm presented the correlation between PPP2R1A gene expression and cancer-associated fibroblasts across diverse malignancies (Figure 8B).

Hierarchical Cox regression analysis revealed the association between different immune cell subpopulations and prognostic features in lung adenocarcinoma patients (Figures 9A, B). The results showed that reduced mesenchymal stem cell levels and decreased natural killer T cells both demonstrating significant prognostic increased risk of death in lung adenocarcinoma patients. Conversely, the enrichment of regulatory T cells (Tregs) exhibited the strongest adverse prognostic impact, followed by CD8+ T cell infiltration. The prognostic dichotomy was particularly pronounced in T helper cell subpopulations: the depletion of type 1 T helper cells and the enrichment of type 2 T helper cells both demonstrated significant clinical implications.

Based on bioinformatics predictions, PPP2R1A is closely linked to the progression of lung adenocarcinoma. To further explore its role in LUAD, this study performed cellular experiments for validation. qPCR analysis revealed a significant reduction in mRNA expression levels in the PPP2R1A knockdown group (P < 0.01), confirming the efficiency of gene knockdown (Figure 10A). CCK-8 assays revealed that the relative OD450 values in the PPP2R1A knockdown (KD) group were significantly lower than those in the wild-type (WT) group at 48, 72, and 96 hours. The fold changes (KD/WT) at 0, 24, 48, 72, and 96 hours were 1.01, 1.17, 1.62, 1.51, and 1.67, respectively (all P < 0.05 after 24 hours) (Figure 10B). Clonogenic assays demonstrated that PPP2R1A knockdown severely impaired colony formation, with the KD group forming significantly fewer visible colonies than the WT control group (fold change = 2.62, P < 0.001) (Figure 10C). All quantitative data are now presented as mean ± SD from n=3 independent experiments.

Cell scratch assays revealed a notable impairment in wound closure in the KD group compared to the WT group at the 48-hour time point (fold change = 1.99, P < 0.001) (Figure 11A). Transwell migration assays showed a significant decrease in matrix-penetrating cells in the PPP2R1A-KD group, indicating reduced migratory capacity (fold change = 1.25, P < 0.01) (Figure 11B). Additionally, the matrix gel invasion assay confirmed that PPP2R1A knockdown led to a significant reduction in the number of invasive cells (fold change = 1.26, P < 0.01) (Figure 11C). All representative images have been replaced with higher-resolution versions, and quantitative data (fold changes, exact p-values) are now explicitly stated in the figure legends and main text.