Therapeutic potential of targeting MASTL in lung adenocarcinoma

Abstract

Protein kinases play crucial roles in tumor progression and modulation of the immunosuppressive tumor microenvironment. However, the specific function of MASTL (microtubule-associated serine/threonine kinase-like) in lung adenocarcinoma (LUAD) remains poorly understood. In this study, we integrated multi-omics bioinformatic analyses with experimental validation to delineate the clinical significance and biological role of MASTL in LUAD. We found that MASTL is markedly overexpressed in LUAD tissues and exhibits substantial diagnostic value. Elevated MASTL expression served as an independent prognostic indicator of poor overall survival, particularly in early-stage patients. Comprehensive immune profiling revealed a strong association between high MASTL expression and an immunosuppressive tumor microenvironment, characterized by impaired dendritic cell function and altered Th1/Th2 balance. Notably, patients with low MASTL expression showed enhanced sensitivity to immune checkpoint blockade therapy. Phosphoproteomic analysis identified serine 370 (S370) as a novel functional phosphorylation site on MASTL, whose activation correlated with higher tumor grade and dysregulation of key oncogenic pathways, including p53, MYC, mTOR, WNT, and HIPPO signaling. Functionally, pharmacological inhibition of MASTL using MKI-1 suppressed LUAD cell proliferation, induced apoptosis and cell cycle arrest, and reduced cancer stem cell-like properties such as self-renewal and metastatic potential. Importantly, MKI-1 administration significantly inhibited tumor growth in a LUAD xenograft model with good tolerability. Collectively, our findings identify MASTL as a pivotal regulator of tumor progression and immune evasion in LUAD and underscore its potential as both a prognostic biomarker and a promising therapeutic target.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-37735-0.

Introduction

Lung adenocarcinoma (LUAD), the predominant subtype of non-small cell lung cancer (NSCLC), is the leading cause of cancer-related mortality worldwide¹. Despite advances in early detection and treatment, the persistently high recurrence rates and limited efficacy of conventional therapies underscore the urgent need to identify novel biomarkers and therapeutic targets. Achieving this goal requires a comprehensive understanding of the molecular mechanisms driving LUAD progression and therapy resistance^2,3.

In this context, targeting key regulators of the cell cycle represents a potential therapeutic approach for LUAD. Among these regulators, MASTL has garnered attention due to its role in controlling mitotic entry and progression, as well as its emerging association with tumor aggressiveness and poor prognosis in various cancers^4–6. MASTL promotes cell proliferation and survival by inhibiting the tumor-suppressive phosphatase PP2A during the G2/M transition^7,8. Although MASTL dysregulation has been linked to aggressive tumor behavior and poor prognosis in cancers such as breast^9,10, pancreatic¹¹, colorectal¹², and thyroid cancer¹³, its specific role in LUAD remains inadequately characterized.

The tumor microenvironment (TME), particularly the tumor-infiltrating immune cells, plays a critical role in LUAD progression and therapy resistance. Emerging evidence suggests that protein kinases may influence not only tumor cell behavior but also have immunomodulatory functions. For example, EGFR, ERBB2, and Axl have been shown to modulate immune cell infiltration and function, thereby affecting tumor immune evasion and response to therapy in lung cancer^14–18. However, whether MASTL has a regulatory role in the immune microenvironment remains unknown. Investigating the impact of MASTL on the immune microenvironment may provide new insights into its role in LUAD and its potential as a therapeutic target.

In this study, we investigated the potential role of MASTL in driving lung cancer progression by integrating multiple bioinformatic analyses and functional validation, aiming to elucidate the clinical relevance of MASTL in lung cancer and its potential as a therapeutic target. We analyzed the expression patterns and phosphorylation status of MASTL in clinical LUAD and assessed their prognostic significance. Additionally, we examined the association between MASTL expression and immune cell infiltration in LUAD to explore its immunomodulatory role. The functional role of MASTL was further investigated using a specific MASTL inhibitor, which may suggest its potential as a therapeutic target.

Materials and methods

Bioinformatic analysis

MASTL expression analysis

MASTL mRNA expression levels in LUAD tumors versus normal tissues were analyzed using the TIMER2.0 web server (http://timer.cistrome.org/). Both unpaired and paired comparisons of MASTL expression between LUAD tumors and normal tissues were performed. Additionally, Receiver operating characteristic (ROC) analysis was conducted to evaluate the diagnostic value of MASTL in distinguishing LUAD from normal tissues. These analyses (Fig. 1B, C and D) were carried out using the “Gene Expression” and “Diagnostic ROC” modules within the Xiantao Academic cloud database (https://www.xiantao.love).

Fig. 1.

MASTL Expression and Its Diagnostic and Prognostic Significance in LUAD. (A) MASTL mRNA levels were higher in LUAD tumors vs. normal tissues. (B) Unpaired comparison confirmed elevated MASTL in LUAD. (C) Paired comparison confirmed higher MASTL in LUAD vs. adjacent tissues. (D) ROC analysis showed MASTL distinguished LUAD from normal tissues. (E) MASTL predicted 1-, 3-, and 5-year DSS. (F) High MASTL linked to reduced OS in GEPIA data. (G) High MASTL linked to reduced OS in TCGA data. (H) High MASTL linked to reduced survival in stage I patients. (I) High MASTL associated with poorer prognosis in M0 stage. (J) High MASTL linked to reduced survival in N0 stage. * p < 0.05; ** p < 0.01; *** p < 0.001 versus the Normal according to the Student’s t-test.

Survival analysis

The prognostic significance of MASTL was assessed using multiple databases. Overall survival (OS) and disease-specific survival (DSS) analyses were performed using GEPIA2 (http://gepia.cancer-pku.cn/) and TCGA data through the Xiantao Academic cloud database (https://www.xiantao.love). Patients were stratified into high- and low-expression groups based on median MASTL expression levels. Subgroup analyses were conducted for stage I patients, M0 stage, and N0 stage. Survival curves were generated using the R package “survival” (version 3.8.3) with log-rank tests for statistical significance.

Construction and evaluation of prognostic nomogram

Univariate and multivariate Cox regression analyses were performed to identify independent prognostic factors using the TCGA-LUAD dataset. A nomogram was subsequently constructed to predict 1-, 2-, and 3-year overall survival probabilities by integrating MASTL expression with significant clinical parameters. The predictive accuracy of the nomogram was validated using calibration curves. All these analyses, including forest plots and nomogram construction, were conducted using the “Clinical Model” and “Nomogram” modules of the Xiantao Academic cloud database (https://www.xiantao.love).

Association with clinicopathological characteristics

The relationship between MASTL expression and clinicopathological features was analyzed using TCGA-LUAD data. Patients were dichotomized into high and low MASTL expression groups based on median expression levels. Clinical parameters included age (dichotomized at 65 years), gender, pathological T stage, N stage, M stage, TNM staging, and smoking history. Statistical analyses employed chi-square tests for most variables and Fisher’s exact tests for categorical variables with small sample sizes. Results were visualized using stacked bar plots generated with ggplot2 package (version 4.0.0) in R.

Identification of differentially expressed genes and enrichment analysis

Differentially expressed genes (DEGs) associated with MASTL in LUAD were identified from TCGA data. A total of 1,361 DEGs were identified with |logFC| > 1 and p-value < 0.05. Venn analysis was performed to identify overlapping genes between upregulated DEGs and MASTL positively co-expressed genes. Enrichment analyses of Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways were performed^19–21. In particular, KEGG pathway enrichment analysis was performed utilizing the KOBAS tool (http://bioinfo.org/kobas). The resulting data were subsequently visualized through the CNSknowall platform (https://cnsknowall.com), a comprehensive web service for data analysis and visualization.

Protein-protein interaction network analysis

The STRING database (https://string-db.org/) was utilized to construct the protein-protein interaction network of MASTL-associated genes. Venn analysis identified five key overlapping genes from GEPIA2, STRING, and the 156 overlapping DEGs. Correlation analyses between MASTL and these key genes (CCNA2, TTK, KIF11, CCNB1, CDK1) were performed using Pearson correlation and visualized with scatter plots.

Single-cell expression and tumor microenvironment analysis

Tumor Immune Single-cell Hub 2 (TISCH2), which integrated single-cell transcriptional and clinical data from 190 datasets covering six million cells across 50 cancer types, analyzed the gene expression at the single-cell level²². MASTL expression in NSCLC single-cell sequencing data was analyzed using Tumor Immune Single-cell Hub 2 (TISCH2, http://tisch.comp-genomics.org), which integrated single-cell transcriptional data from datasets GSE148071, GSE127465, and EMTAB6149. The ESTIMATE algorithm was employed to calculate ImmuneScore, StromalScore, and ESTIMATEScore from TCGA-LUAD gene expression data using the estimate package (version 1.0.13)²³. Patients were stratified based on MASTL expression and microenvironment scores using median cutoffs. The abundance of twenty-eight immune cell types was estimated using single sample gene set enrichment analysis (ssGSEA) through the GSVA package (version 2.2.1).

Immune infiltration and subgroup survival analysis

Comprehensive immune infiltration analyses were performed combining MASTL expression with various immune parameters. Four-group analyses integrated MASTL expression with immune scores, stromal scores, and Th1/Th2 immune status. Kaplan-Meier survival analysis with log-rank tests was conducted to assess overall survival differences between these groups using survival (version 3.8.3) and survminer (version 0.5.1) packages²⁴. TH1 and TH2 immune cell signatures were calculated using established metagene lists. All visualizations were created using ggplot2 package (version 4.0.0).

Correlation with immune checkpoint genes and immunotherapy response prediction

We downloaded a dataset containing seventy-nine immune checkpoint genes²⁵ and performed Pearson correlation analysis to evaluate the association between MASTL expression and these genes in the TCGA-LUAD dataset. Genes with p-value < 0.05 and absolute correlation coefficient (|r|) > 0.2 were considered statistically significant. The correlation heatmap was generated with the pheatmap package (version 1.0.13). Immunophenoscore (IPS) data for different immunotherapy scenarios were obtained from The Cancer Immunome Atlas (TCIA) database²⁶. Tumor Immune Dysfunction and Exclusion (TIDE) scores were calculated from the TIDE website (http://tide.dfci.harvard.edu/)²⁷Differences in TIDE scores and immunotherapy response rates between MASTL expression groups were assessed using t-tests and chi-square tests. All visualizations were created using ggplot2 package (version 4.0.0).

Drug sensitivity analysis

Drug sensitivity was predicted using the oncoPredict R package (version 1.2) with GDSC2 database(https://www.cancerrxgene.org/) as the training set. The calcPhenotype algorithm was employed to predict values for multiple anti-cancer drugs²⁸. Association analysis between MASTL expression and drug sensitivity was performed using Pearson correlation. Patients were stratified into MASTL-high and MASTL-low groups, and statistical comparisons between groups were conducted using t-tests.

Statistical analysis

All statistical analyses were performed using R software (version 4.2.1)²⁹. Data visualization was primarily conducted using ggplot2 package (version 4.0.0). Statistical significance was set at p < 0.05 unless otherwise specified.

Cell culture and authentication

Established human LUAD cell lines, specifically A549 and H1299, were procured from the American Type Culture Collection (ATCC). The cells were cultured in RPMI 1640 medium (Invitrogen, Carlsbad, CA), supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin. Authentication of all cell lines was conducted utilizing the AmpFlSTR Identifiler PCR Amplification Kit for short tandem repeat (STR) profiling, in accordance with the manufacturer’s guidelines (Thermo Fisher Scientific, Waltham, MA). Routine testing for mycoplasma contamination was performed, and the cells were maintained at 37 °C in a humidified incubator with 5% CO₂. Experimental procedures were carried out using cells that had undergone no more than 10 passages to ensure experimental consistency.

Sphere formation assay

A total of 250 A549 cells and 500 H1299 cells were cultured in 6-well low-attachment plates (Corning, USA) utilizing a cancer stem cell (CSC) medium, which consisted of RPMI 1640 supplemented with 20 ng/mL epidermal growth factor (EGF), 20 ng/mL fibroblast growth factor (FGF), 40 ng/mL insulin-like growth factor (IGF), and 1X B27. The cultures were maintained in a humidified incubator at 37 °C with 5% CO₂. The medium was replenished every three days to ensure sufficient nutrient availability. After a period of 14 days, the quantity and size of the tumor spheres were evaluated microscopically. Tumor spheres exhibiting a diameter greater than 50 μm were classified as positive. The sphere formation efficiency was determined by calculating the ratio of the number of spheres formed to the number of cells initially seeded. Each experimental condition was conducted in triplicate and independently repeated three times.

Cell viability assay

To evaluate the effect of MKI-1 (HY-137552, MCE) on LUAD cell viability, an MTT assay was performed. A549 and H1299 cells were seeded in 96-well plates at densities of 5,000 and 8,000 cells per well, respectively, and allowed to adhere overnight in complete medium. The following day, the culture medium was replaced with fresh medium containing a series of concentrations of MKI-1 (0, 2.5, 5, 10, 20, and 40 µM). After incubating for 24 h, 10 µL of MTT solution (5 mg/mL in PBS) was added to each well, and the plates were further incubated at 37 °C for 4 h. Subsequently, the culture medium containing MTT was carefully aspirated, and 100 µL of dimethyl sulfoxide (DMSO) was added to each well to dissolve the resulting formazan crystals. The plates were gently shaken for 10 min in the dark to ensure complete dissolution. The optical density (OD) at a wavelength of 490 nm was measured using a microplate reader. Cell viability was calculated as a percentage relative to the untreated control group (set as 100% viability).

Migration assay

To assess the effect of MASTL inhibition on cell migration, a Transwell assay was performed. Briefly, A549 and H1299 cells were pre-treated with the indicated concentrations of MKI-1 (0, 2.5, or 5 µM) for 24 h. Subsequently, 30,000 pre-treated A549 cells or 50,000 pre-treated H1299 cells in 200 µL of serum-free RPMI 1640 medium were seeded into the upper chamber of a Transwell insert (8.0 μm pore size). The lower chamber was filled with 800 µL of RPMI 1640 medium supplemented with 10% FBS, which served as a chemoattractant. The plate was incubated at 37 °C in a humidified atmosphere of 5% CO₂ for 24 h. After incubation, non-migratory cells on the upper surface of the membrane were carefully removed with a cotton swab. The migratory cells on the lower surface were fixed with 4% paraformaldehyde for 15 min, stained with 0.1% crystal violet for 30 min, and then imaged under a light microscope. The number of migratory cells was quantified by counting five random fields per well.

Apoptosis assay

Increasing concentrations of MKI-1 (0µM and 5µM) were used to treat LUAD cells (A549 and H1299) for 48 h. Following the treatment period, the cells were collected and subjected to two washes with cold phosphate-buffered saline (PBS). The cells were subsequently resuspended in a 1× binding buffer at a concentration of 1 × 10⁶ cells/mL. A volume of 100 µL of this cell suspension was transferred to a new tube, to which 5 µL of Annexin V-FITC and 5 µL of propidium iodide (PI) were added.The mixture was incubated for 15 min at room temperature in the dark.After incubation, 400 µL of 1× binding buffer was added to each tube, and the cells were analyzed using a flow cytometer. The resulting data were processed using FlowJo™ v10.8.1 Software (BD Life Sciences)³⁰ to determine the percentage of apoptotic cells (Annexin V-positive) and necrotic cells (PI-positive). Each experimental condition was conducted in triplicate and independently repeated three times.

Cell cycle analysis

The progression of the cell cycle was assessed through propidium iodide (PI) staining. LUAD cells were subjected to treatment with MKI-1 at concentrations of 0 µM, 2.5 µM, and 5 µM for a duration of 48 h. Following the treatment, the cells were collected, rinsed with cold PBS, and subsequently fixed in 70% ethanol at −20 °C for a minimum of 24 h. The fixed cells were then washed with PBS and resuspended in 500 µL of a PI staining solution, which contained 50 µg/mL of PI and 0.2 mg/mL of RNase A. The cells were incubated in the dark at 37 °C for 30 min to facilitate DNA staining. The stained cells were analyzed using a flow cytometer, and the distribution of the cell cycle phases (G0/G1, S, and G2/M) was determined utilizing FlowJo™ v10.8.1 Software (BD Life Sciences)³⁰. Each experimental condition was conducted in triplicate and independently repeated three times.

Colony formation

LUAD cells (A549 and H1299) were collected during the logarithmic growth phase and subsequently resuspended in complete RPMI 1640 medium. A total of 500 cells were plated in each well of a 6-well plate and permitted to adhere overnight at 37 °C in a humidified incubator containing 5% CO₂. Following this, the cells were subjected to treatment with varying concentrations of MKI-1 (2.5 µM and 5 µM) or a vehicle control (DMSO) for a duration of two weeks. The culture medium was replenished every three days to ensure optimal growth conditions. After a 14-day incubation period, the colonies were fixed using 4% paraformaldehyde for 15 min and subsequently stained with 0.1% crystal violet for 30 min. The stained colonies were visualized using a light microscope, and the number of colonies containing more than 50 cells was counted manually. The colony formation efficiency (CFE) was determined as the ratio of the number of colonies formed to the number of cells initially seeded, expressed as a percentage. Each experimental condition was conducted in triplicate and independently repeated three times.

Scratch assay

LUAD cells (A549 and H1299) were cultured in 6-well plates at a density of 1.5 × 10⁶ cells per well in complete RPMI 1640 medium. The cells were maintained under standard conditions (37 °C, 5% CO₂) until they reached confluence, thereby forming a uniform monolayer. Upon achieving confluence, a sterile 200 µL pipette tip was employed to create a linear scratch wound across the center of each well. Detached cells were subsequently removed through washing with PBS. Following the scratching procedure, the cells were treated with RPMI 1640 medium supplemented with either 2.5 µM or 5 µM of MKI-1, while a control group was established using DMSO. Wound closure was assessed at 0, 12, and 24 h post-scratch using an inverted light microscope equipped with a 10× objective lens. The migration rates of the cells were quantified by measuring the wound width at multiple locations (n = 5) per scratch, utilizing ImageJ software(version 1. 53a)for analysis. Each experimental condition was conducted in triplicate and independently repeated three times.

Animal experiment

All animal experiments were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee of Sun Yat-sen University (Approval No. SYSU-IACUC-2024-001389). All methods were carried out in accordance with relevant guidelines and regulations, and all methods are reported in accordance with ARRIVE guidelines. Six week old male athymic nude mice were purchased from Zhuhai Bestest Biotechnology Co., Ltd. All mice were housed and experiments were conducted at the Laboratory Animal Center, Shenzhen Campus of Sun Yat-sen University. Athymic nude mice were subcutaneously inoculated with A549 cells (5 × 10⁵ cells per mouse) in a 1:1 mixture of cell suspension and Matrigel (total volume 100 µL). Mice were randomly divided into three groups (n = 8 per group): a control group, an MKI-1 treatment group (25 mg/kg), and another MKI-1 treatment group (50 mg/kg). The injection vehicle comprised a solution of 10% DMSO (Sigma-Aldrich), 1% Tween 20 (Sigma-Aldrich), and 89% saline (Sigma-Aldrich). The control group was administered the vehicle exclusively, whereas the treatment groups received MKI-1 dissolved in the identical vehicle at dosages of 25 mg/kg or 50 mg/kg, delivered via intraperitoneal injection. Drug administration was initiated on day 4 after tumor inoculation and repeated every 4 days for a total of 4 doses. Tumor size was measured regularly using calipers, and volume was calculated as (length × width²)/2. Mice were euthanized when tumors reached a predetermined size via exposure to a rising concentration of carbon dioxide (CO₂) in a controlled chamber, with a flow rate displacing 30% to 70% of the chamber volume per minute, followed by confirmation of death by cervical dislocation. Tumors were collected for further analysis.

Statistical analysis

All statistical analyses were conducted with GraphPad Prism (Version 8.0, CA, USA). A p value less than 0.05 was considered statistically significant (*p < 0.05; **p < 0.01; ***p < 0.001; ns, not significant).

Results

Prognostic significance of MASTL in LUAD

MASTL has been recognized for its role in regulating cell cycle progression and mitotic entry; however, its specific function in LUAD remains inadequately elucidated. To investigate the potential involvement of MASTL in LUAD pathogenesis, we initially examined its expression patterns across normal tissues. Analysis of data from the GTEx project (comprising 17,382 samples from 948 donors) revealed that MASTL expression was markedly elevated in Epstein-Barr Virus (EBV)-transformed lymphocytes and cultured fibroblasts (Fig. S1A). Further validation using the Human Protein Atlas (HPA) dataset demonstrated increased MASTL expression in immune-related tissues, including bone marrow, thymus, lymph nodes, and tonsils, whereas its expression was diminished in 50 other tissue types. This suggests that MASTL expression is predominantly confined to the immune system, with limited expression in other tissues (Fig. S1B).

Building upon these observations in normal tissues, we explored the potential dysregulation of MASTL in LUAD. Utilizing The Cancer Genome Atlas (TCGA) data analyzed via TIMER2.0, we found that MASTL mRNA levels were significantly elevated in LUAD tumor tissues compared to adjacent normal lung tissues (Fig. 1A). This upregulation was further corroborated through both paired and unpaired comparisons between normal lung tissues and LUAD samples (Fig. 1B and C).

To evaluate the diagnostic utility of MASTL, receiver operating characteristic (ROC) curve analysis was conducted, revealing that MASTL expression effectively discriminated LUAD tumor tissues from normal tissues, with an area under the curve (AUC) of 0.764 (95% confidence interval [CI]: 0.718–0.811) (Fig. 1D). Additionally, the prognostic relevance of MASTL expression was assessed in relation to disease-specific survival (DSS) among LUAD patients. ROC analyses indicated moderate predictive performance for 1-year, 3-year, and 5-year DSS, with AUC values of 0.635, 0.604, and 0.616, respectively (Fig. 1E). These findings underscore the potential of MASTL as a clinically relevant biomarker for both diagnosis and prognosis in LUAD.

Further investigation into the association between MASTL expression and patient outcomes was performed using Kaplan-Meier survival analyses based on data from the GEPIA database and TCGA cohort. High MASTL expression was consistently correlated with poorer overall survival (OS) in LUAD patients (hazard ratio [HR] = 1.48, p = 0.007; HR = 1.4, p = 0.024) (Fig. 1F and G). Stratified survival analyses by pathological stage revealed that elevated MASTL expression was particularly predictive of unfavorable prognosis in early-stage patients, including those at TNM stage I (HR = 1.80, p = 0.016), M0 stage (HR = 1.51, p = 0.018), and N0 stage (HR = 1.80, p = 0.006) (Fig. 1H–J). These results suggest that MASTL may serve as a valuable biomarker for early diagnosis and prognosis in LUAD.

Collectively, the data indicate that MASTL holds promise as both a diagnostic and prognostic biomarker in LUAD.

MASTL as an independent prognostic indicator in LUAD

To further investigate whether MASTL expression functions as an independent prognostic factor in LUAD, we conducted both univariate survival analysis and multivariate Cox regression analysis. The univariate analysis revealed that elevated MASTL expression was significantly associated with poorer overall survival (OS) (hazard ratio [HR] = 1.394; 95% confidence interval [CI], 1.144–1.699; p < 0.001). Additionally, advanced pathological TNM (pTNM) stages II (HR = 2.256; 95% CI, 1.547–3.289), III (HR = 3.171; 95% CI, 2.138–4.705), and IV (HR = 3.474; 95% CI, 2.001–6.031) were also significantly predictive of poor OS (p < 0.001) (Fig. 2A). Consistent with these findings, multivariate Cox regression analysis confirmed that high MASTL expression (HR = 1.315; 95% CI, 1.067–1.621; p = 0.01) and advanced pTNM stages II (HR = 2.145; 95% CI, 1.469–3.134), III (HR = 3.078; 95% CI, 2.074–4.567), and IV (HR = 3.230; 95% CI, 1.855–5.623) (p < 0.001) independently predicted unfavorable survival outcomes (Fig. 2B).

Fig. 2.

Forest map of univariate (A) and multivariate (B) Cox analysis of MASTL and other clinical parameters related to the overall survival. (C-D) Nomogram with a calibration curve to predict patients’ overall survival at 1, 2, and 3 years. (E-L) Clinicopathological characteristics associated with MASTL expression: (E) pTNM stage, (F) pT stage, (G) pN stage, (H) pM stage, (I) Smoking status, (J) Gender, (K) Age distribution, and (L) Statistical significance of associations.

Subsequently, a nomogram incorporating these significant variables was developed to assess the prognostic utility of MASTL expression (Fig. 2C). The calibration curve demonstrated strong concordance between predicted and observed 1-, 2-, and 3-year OS rates (Fig. 2D). Analysis of clinicopathological characteristics stratified by MASTL expression levels within The Cancer Genome Atlas (TCGA) cohort revealed that elevated MASTL expression was significantly associated with younger age (< 65 vs. ≥65 years; p = 0.0189), male gender (p = 0.0455), advanced pathological metastasis stage (M0 vs. M1; p = 0.0469), and higher overall pTNM stage (p = 0.0297) (Figs. 2E–L; Supplementary Table 1).

Collectively, these results indicate that the nomogram, which integrates MASTL expression with established clinical parameters, offers a more comprehensive and precise prognostic model for LUAD patient survival than models based on individual factors alone. Importantly, the identification of MASTL as an independent prognostic biomarker underscores its potential involvement in LUAD progression and highlights its clinical relevance for risk stratification.

Identification of MASTL-associated differentially expressed genes and functional enrichment analysis

To further elucidate the molecular mechanisms by which MASTL influences LUAD, we conducted a differential expression analysis utilizing data from TCGA. LUAD samples were stratified into high- and low-MASTL expression groups based on the median expression value, with the low-expression group serving as the reference. This analysis identified 1,361 DEGs using criteria of |log fold change| > 1 and adjusted p-value < 0.05, comprising 974 upregulated and 387 downregulated genes (Fig. 3A).

Fig. 3.

Molecular Mechanisms Underlying MASTL‘s Role in LUAD. (A) Differential expression analysis identified 1,361 DEGs linked to MASTL in LUAD.(B) Venn analysis revealed 156 overlapping genes between upregulated DEGs and MASTL positively co-expressed genes.(C) KEGG secondary classification analysis. (D) GO analysis showed enrichment in BP, CC, MF.(E) Integration of Sankey and Bubble Plots for KEGG Analysis. The Sankey diagram illustrates genes associated with specific KEGG, while the bubble plot displays the GeneRatio values (position), the number of enriched genes (size), and the significance of pathways (color) based on P-values.

Subsequently, to explore the regulatory network associated with MASTL, we performed a co-expression analysis using TCGA data, calculating Pearson correlation coefficients between MASTL and other genes. A total of 591 genes exhibited significant correlation with MASTL expression (|correlation coefficient| > 0.5, p < 0.05), of which 590 were positively correlated and one negatively correlated. Intersection analysis via a Venn diagram revealed 156 overlapping genes between the upregulated DEGs and those positively correlated with MASTL, suggesting their involvement in the MASTL-mediated regulatory network (Fig. 3B, S2).

To comprehensively characterize the functional roles of these 156 genes, we conducted KEGG and GO enrichment analyses. KEGG secondary classification analysis indicated significant enrichment of these genes in higher-order functional categories including replication and repair, signal transduction, cell growth and death, and the endocrine system. At the primary classification level, these genes were also associated with human diseases, underscoring their potential clinical relevance in LUAD (Fig. 3C).

For a more detailed examination, we constructed a KEGG Sankey-bubble plot integrating a Sankey diagram and a bubble plot to visualize specific enriched pathways and their relationships with MASTL-associated genes (Fig. 3E). The Sankey diagram depicted gene-pathway connections, while the bubble plot illustrated enrichment significance, gene counts, and GeneRatio values for each pathway. This analysis revealed significant enrichment of MASTL-associated genes in pathways such as Cell Cycle, Oocyte Meiosis, Progesterone-mediated Oocyte Maturation, Homologous Recombination, Cellular Senescence, p53 Signaling Pathway, and MicroRNAs in LUAD. These pathways are intimately linked to cell cycle regulation, DNA repair, and genomic stability, thereby reinforcing the role of MASTL in LUAD progression.

GO term enrichment analysis further substantiated these findings, demonstrating significant enrichment in Biological Process (BP) terms including regulation of mitotic cell cycle, DNA-templated DNA replication, and DNA replication; Cellular Component (CC) terms such as chromosomal region, spindle, and chromosome centromeric region; and Molecular Function (MF) terms including protein serine/threonine kinase activity, cyclin-dependent protein serine/threonine kinase regulator activity, and tubulin binding (Fig. 3D). Collectively, these results indicate that MASTL and its associated genes are critically involved in LUAD progression through modulation of cell cycle regulation, maintenance of genomic stability, and influence on tumor suppressive pathways. These findings provide a compelling rationale for further investigation into the functional mechanisms of MASTL in LUAD and its potential as a therapeutic target.

Identification of core MASTL-interacting partners in LUAD

To identify downstream interaction partners of MASTL, we first retrieved the top ten MASTL-associated genes from the STRING database (Fig. 4A). In parallel, the top 500 genes exhibiting the highest Pearson correlation coefficients with MASTL expression (PCC > 0.5) in LUAD were obtained from the GEPIA2 database (Table S2). We then performed an intersection analysis of these datasets—namely, the ten interacting genes, the top 500 correlated genes, and the 156 overlapping genes previously identified in Fig. 3B. This integrative approach revealed five genes—CCNA2 (R = 0.704, P < 0.001), TTK (R = 0.727, P < 0.001), KIF11 (R = 0.758, P < 0.001), CCNB1 (R = 0.621, P < 0.001), and CDK1 (R = 0.713, P < 0.001)—that demonstrated strong co-expression and potential interactions with MASTL (Figs. 4B–G). These genes are known to play key roles in mitotic regulation: CCNA2 regulates the G2/M transition and mitotic progression³¹; TTK maintains mitotic fidelity via spindle assembly checkpoint signaling³²; KIF11 facilitates spindle assembly and chromosome segregation³³; CCNB1 governs the G2/M transition and mitotic entry³⁴; and CDK1 drives mitotic entry and progression³⁵. Collectively, these findings highlight the central role of MASTL in coordinating cell cycle regulation and tumor cell survival, suggesting that MASTL and its interacting partners may act as critical drivers of LUAD development and progression.

Fig. 4.

Validation of MASTL- Interacting Partners in LUAD. (A)10 STRING-derived MASTL-associated genes.(B) Venn diagram identified five key genes from GEPIA2, STRING and 156 overlapping genes. (C-G) Scatter plots show strong correlations between MASTL and the CCNA2(C), TTK(D), KIF11(E), CCNB1(F), CDK1(G) in LUAD.

Elevated MASTL expression correlates with an immunosuppressive tumor microenvironment

To investigate the immune microenvironment in lung adenocarcinoma (LUAD), we examined the association between MASTL expression and various immune biomarkers. Our analysis revealed a predominantly negative correlation between MASTL and the majority of immune markers. Notably, nearly all markers associated with dendritic cells exhibited a strong inverse correlation with MASTL, suggesting that MASTL may impede antigen presentation processes to some degree in LUAD (Table 1).Furthermore, Single-cell RNA sequencing datasets (GSE148071, GSE127465, and EMTAB6149) revealed that MASTL expression was notably elevated in malignant cells as well as in monocytes/macrophages, compared with other cell populations (Figs. 5A–C). To further explore the association between MASTL expression and immune cell infiltration, we performed single-sample gene set variation analysis (ssGSVA), which demonstrated distinct infiltration patterns across 28 tumor-infiltrating immune cell (TIIC) subtypes between high and low MASTL expression groups. Notably, the high MASTL expression group exhibited reduced infiltration of most immune cell types, with the exception of activated T helper cells, γδ T cells, and type 2 T helper cells (Fig. 5E). To assess the variations in the tumor microenvironment (TME) between groups exhibiting high and low MASTL expression, the associations between MASTL and stromal as well as immune scores are presented in Fig. 5D. The data demonstrate that samples characterized by elevated MASTL expression levels exhibited significantly reduced immune scores (p < 0.001), stromal scores (p < 0.001), and overall estimation scores (p < 0.001). Collectively, these findings suggest that increased MASTL expression contributes to a broadly immunosuppressive tumor microenvironment, potentially facilitating immune evasion and tumor progression.

Table 1.

Correlation analysis between MASTL and immune cell biomarkers in LUAD.

Immune cell type	Gene biomarker	LUAD
		correlation	p-value
CD8 + T cell	CD8A	0.10	0.02
	CD8B	0.067	0.12
CD4 + T cell	CD4	−0.19	1.34E-05
T cell (general)	CD3D	−0.14	0.001
	CD3E	−0.13	0.002
	CD2	−0.105	0.02
B cell	CD19	−0.20	1.97E-06
	CD79A	−0.27	2.56E-10
Monocyte	CD86	0.03	0.51
M1 Macrophage	NOS2	0.10	0.03
	IRF5	0.02	0.64
	PTGS2	0.14	0.001
M2 Macrophage	CD163	0.08	0.07
	VSIG4	−0.07	0.10
	MS4A4A	−0.03	0.45
Dendritic cell	HLA-DPB1	−0.35	1.17E-16
	HLA-DQB1	−0.30	1.63E-12
	HLA-DRA	−0.29	1.53E-11
	HLA-DPA1	−0.26	9.28E-10
	CD1C	−0.37	8.97E-19
	NRP1	0.12	0.01
	ITGAX	−0.04	0.31
Natural killer cell	KIR2DL3	0.09	0.04
	KIR2DS4	0.08	0.07
	KIR3DL2	0.05	0.25
	KIR3DL3	0.05	0.28
	KIR3DL1	0.05	0.29
	KIR2DL1	0.04	0.33
Neutrophils	CEACAM8	−0.23	1.21E-07
	ITGAM	−0.11	0.01
	CCR7	−0.20	5.09E-06
Th1	TBX21	0.005	0.92
	STAT4	0.04	0.32
	STAT1	0.39	5.79E-21
	IFNG	0.23	4.86E-08
	TNF	−0.04	0.32
Th2	GATA3	0.14	0.001
	STAT6	−0.09	0.05
	STAT5A	−0.08	0.06
	IL13	−0.06	0.15
Tfh	BCL6	0.02	0.64
	IL21	0.13	0.003
	STAT3	0.07	0.09
	IL17A	0.02	0.60
Th17	FOXP3	−0.04	0.41
	CCR8	0.06	0.19
	STAT5B	0.05	0.29
	TGFB1	−0.12	0.005
T cell exhaustion	PDCD1	0.01	0.84
	CTLA4	0.07	0.11
	LAG3	0.06	0.16
	HAVCR2	−0.01	0.82
	GZMB	0.23	7.30E-08
Treg	FOXP3	−0.04	0.41
	STAT5B	0.05	0.29
	TGFB1	−0.12	0.005
	CCR8	0.06	0.19
Effector T cell	CX3CR1	−0.27	9.49E-11
	FGFBP2	−0.12	0.005
	FCGR3A	0.09	0.04
Naive T cell	CCR7	−0.20	5.09E-06
	SELL	−0.11	0.01
Effector memory T cell	DUSP4	0.13	0.002
	GZMK	−0.06	0.14
	GZMA	0.06	0.20
Resident memory T cell	CD69	−0.09	0.03
	CXCR6	0.04	0.33
	MYADM	0.02	0.67
General memory T cell	CCR7	−0.20	5.09E-06
	SELL	−0.11	0.01
	IL7R	−0.04	0.35

Significant values are presented in bold typeface.

Fig. 5.

Immune cell infiltration and TME of LUAD. MASTL expression in the NSCLC single-cell sequencing data GSE148071 (A), GSE127465 (B) and EMTAB6149 (C). (D) Immune scores, stromal scores, and estimate scores between high and low MASTL groups in TCGA cohort. (E) A bundance analysis of twenty-eightimmune cell types between high and low MASTL groups in TCGA cohort *p < 0.05; **p < 0.01; ***p < 0.001.

Survival analysis of immune subgroups

To explore the potential impact of the tumor microenvironment on the prognosis of LUAD, patients were categorized into two subgroups based on the median values of immune and stromal scores. Patients with lower MASTL expression coupled with higher immune and stromal scores exhibited the most favorable prognostic outcomes (Figs. 6A and B). Furthermore, two critical observations guided subsequent subgroup analyses: first, the expression levels of Th1 and Th2 cells were inversely correlated between groups with high and low MASTL expression; second, the Th1/Th2 ratio is recognized as a modulator of the tumor immunosuppressive microenvironment³⁶. Employing the Kaplan-Meier Plotter tool, a subgroup survival analysis was performed. Notably, elevated MASTL expression was significantly correlated with poorer survival in LUAD patients with enriched Th1 cell populations (p = 0.013), whereas no significant associations were detected in patients with reduced Th1 cells (p = 0.15), enriched Th2 cells (p = 0.10), or reduced Th2 cells (p = 0.096) (Fig. 6C–F). Subsequently, Th1/Th2 ratios were calculated for all samples using the single-sample Gene Set Variation Analysis (ssGSVA) algorithm, and patients were dichotomized accordingly. MASTL expression exhibited prognostic significance exclusively within the subgroup characterized by high Th1/Th2 ratios (p = 0.021), indicating that the observed adverse prognosis may be partially attributable to this immunological context (Fig. 6G and H).

Fig. 6.

Immune cell infiltration and prognosis analysis of LUAD. (A) The survival curve according to MASTL expression levels and immune scores. (B) The survival curve according to MASTL expression levels and stromal scores. Groups with high and low ImmuneScore (or StromalScore) were indicated by HIS (or HSS) and LIS (or LSS), respectively. (C-F) K-M survival curves for OS according to MASTL expression in different Type 1 and Type 2 T-helper cells enriched levels. (G) The survival curve of OS according to MASTL expression levels and Th1/Th2 ratios. (H) The survival curve of OS according to MASTL expression levels in high and low Th1/Th2 ratios subgroups. Groups with high and low Th1/Th2 ratios were indicated by Hr and Lr, respectively. *p < 0.05; **p < 0.01; ***p < 0.001.

Enhanced immunotherapy response observed in patients with low MASTL expression

Given the critical role of immune checkpoint molecules in modulating immunotherapy efficacy, we assessed the correlation between MASTL expression and a panel of immune checkpoint genes identified in prior studies (Supplementary Table 3). As illustrated in Fig. 7A, eighteen immune checkpoint genes demonstrated a negative correlation with MASTL expression, whereas PVR and KIR2DL4 showed positive correlations. Subsequently, we evaluated the immunotherapeutic response of LUAD patients utilizing the IPS and TIDE algorithms. Patients exhibiting low MASTL expression presented significantly higher IPS values compared to those with elevated MASTL expression, particularly under conditions of no treatment (p = 0.0036) or anti-CTLA4 therapy (p = 0.0068) (Fig. 7B). Furthermore, the low MASTL cohort displayed markedly reduced TIDE scores (p = 0.031) (Fig. 7C) alongside an increased clinical response rate to immunotherapy (p = 0.0085) (Fig. 7D). Notably, MASTL expression was significantly higher in non-responders relative to responders (p < 0.001) (Fig. 7E). Drug sensitivity analysis revealed nine agents exhibiting stronger correlations with elevated MASTL expression (Fig. 7F) and correspondingly reduced drug sensitivity values in the high MASTL expression group (Fig. 7G), suggesting that these drugs may be more effective for patients with low MASTL levels. Notably, among these nine drugs are Osimertinib, Erlotinib, and Gefitinib, which may serve as potential therapeutic agents for patients exhibiting low MASTL expression. Collectively, these findings suggest that increased MASTL expression in LUAD is predictive of poorer immunotherapy outcomes.

Fig. 7.

Correlation of MASTL with immune checkpoint genes and prediction of immunotherapy response for patients. (A) Heatmap of correlation analysis. Each dot displays Pearson’s correlation coefficient between two different specific genes. (B) Violin diagrams visualized the difference in immunotherapy (PD-1 and CTLA-4 blockers) response to the single and combined use of the two drugs. (C-D) Comparisons of the TIDE score and proportions of non-responders (or responders) to ICIs between high- and low-MASTL expression groups in the TCGA cohort. (E) Differences in MASTL expression levels between the responders and non-responders in TCGA. (F) The correlationbetween drug sensitivity and patients. (G) The difference of drug sensitivity in high- and low- MASTL groups. *p < 0.05; **p < 0.01; ***p < 0.001.

MASTL phosphorylation in LUAD

To elucidate the role of MASTL phosphorylation in LUAD, we conducted an analysis of the PhosphoSitePlus database, identifying serine 370 (S370) as a prominent phosphorylation site on human MASTL. This site has been consistently reported in 24 independent high-throughput studies, predominantly utilizing mass spectrometry-based phosphoproteomics (Fig. 8A). Notably, these datasets encompass samples from non-small cell lung cancer, with S370 phosphorylation recurrently detected across multiple investigations, implying its potential biological significance. Building upon these observations, we performed a clinical correlation analysis of MASTL (S370) phosphorylation employing the UALCAN database. The analysis revealed a significant upregulation of MASTL (S370) phosphorylation in primary LUAD tumors relative to normal lung tissue (Fig. 8B). Moreover, phosphorylation levels at this site exhibited a positive correlation with tumor grade, displaying the highest expression in grade 3 tumors, followed by grades 2 and 1 (Fig. 8C), suggesting an association with tumor aggressiveness. Subsequent pathway analyses demonstrated that increased phosphorylation of MASTL (S370) is significantly linked to the dysregulation of several critical tumor progression pathways, including p53, MYC, mTOR, and receptor tyrosine kinase signaling (Figs. 8D–G). Additionally, this phosphorylation event is associated with chromatin remodeling mechanisms, particularly involving chromatin modifiers and the SWI-SNF complex (Figs. 8H and I). Furthermore, phosphorylated MASTL (S370) correlates with pathways governing stress response and self-renewal, such as Nrf2, Wnt, and HIPPO signaling (Figs. 8J–L). Collectively, these findings suggest that MASTL may possess previously unrecognized biological functions in tumorigenesis by modulating redox homeostasis and cancer stem cell characteristics, highlighting a novel avenue for future research.

Fig. 8.

Phosphorylation of MASTL at Serine 370 in LUAD and Its Clinical Significance. (A) The phosphorylation sites of MASTL were analyzed utilizing the PhosphoSitePlus database, which identified S370 as a frequently phosphorylated site corroborated by numerous high-throughput studies. (B) The expression levels of MASTL (S370) protein were examined across different sample types (tumor versus normal) through the UALCAN database. (C) The expression of MASTL (S370) protein was further analyzed in LUAD tumor grades^1–3 using the UALCAN database. (D-L) The expression of MASTL (S370) protein in LUAD tumors was assessed in relation to alterations in the P53 pathway and other significant pathways, including the MYC/MYCN pathway, mTOR pathway, RTK pathway, chromatin modifiers, SWI-SNF complex, WNT pathway, HIPPO pathway and NRF2 pathway, employing the UALCAN database for analysis. ***P < 0.001, versus the Normal according to one-way ANOVA or the Student’s t-test.

MASTL facilitates proliferation of LUAD cells through modulation of cell cycle and apoptosis

The functional role of MASTL in LUAD has not been fully elucidated. To further investigate the significance of MASTL in LUAD pathophysiology and to assess the therapeutic potential of targeting MASTL, we examined the effects of the selective MASTL inhibitor MKI-1 on the viability of LUAD cell lines. Using MTT assays, the half-maximal inhibitory concentration (IC50) values of MKI-1 were determined to be 10.46 µM for A549 cells and 11.99 µM for H1299 cells (Fig. 9A), demonstrating its potent inhibitory effect on LUAD cell viability. To circumvent cytotoxicity associated with higher doses, sublethal concentrations of MKI-1 were employed in subsequent phenotypic analyses based on these viability data. Inhibition of MASTL was found to significantly influence apoptosis, cell cycle progression, and cellular proliferation. Specifically, treatment with 5 µM MKI-1 markedly increased apoptotic rates in both A549 and H1299 cells, as evidenced by flow cytometric analysis (Figs. 9B-C). Additionally, exposure to MKI-1 at concentrations of 2.5 µM and 5 µM induced cell cycle arrest predominantly at the S phase (Figs. 9D-E), underscoring the critical role of MASTL in regulating cell cycle dynamics. Corroborating these observations, treatment of A549 and H1299 cells with MKI-1 at 2.5 µM and 5 µM resulted in a significant decrease in colony formation capacity (Figs. 9F-G), indicating that MASTL inhibition effectively suppresses tumor cell proliferation.

Fig. 9.

MASTL Inhibition Suppresses LUAD Cell Proliferation and Induces Apoptosis. (A) Cell viability of MKI-1 in A549 and H1299 cells evaluated by MTT assay. (B-C) Apoptosis analysis in A549 and H1299 cells treated with 5 µM MKI-1 by PI and Annexin V staining and flow cytometry analysis. (D-E) Cell cycle analysis revealing S-phase arrest in cells treated with 2.5 µM and 5 µM MKI-1. (F-G) Colony formation assays demonstrating reduced proliferation in A549 and H1299 cells treated with 2.5 µM and 5 µM MKI-1. *P < 0.05, **P < 0.01, ***P < 0.001 versus the 0µM according to Tukey’s test. The data are presented as the means ± SEM of three replicates.

MASTL Inhibition suppresses self-renewal and metastasis of LUAD

Our findings indicate that the activation of MASTL may be linked to multiple CSC-related signaling pathways, suggesting a potential role for MASTL in modulating CSC phenotypes beyond its well-established function in cell cycle regulation. CSCs represent a subpopulation of cells characterized by enhanced aggressiveness relative to their non-CSC counterparts, exhibiting properties such as self-renewal capacity, metastatic potential, and resistance to anticancer therapies. Consequently, we further examined the impact of the MASTL inhibitor MKI-1 on the self-renewal and metastatic abilities of LUAD cell lines using sphere formation assays, transwell migration assays, and wound healing assays. Treatment with MKI-1 at a concentration of 2.5 µM significantly impaired sphere formation and reduced the frequency of CSCs within LUAD cell lines (Figs. 10A and B), indicating its efficacy in targeting stem-like characteristics. Moreover, MKI-1 at concentrations of 2.5 µM and 5 µM substantially inhibited the migratory capacity of A549 and H1299 cells, as demonstrated by migration assays (Figs. 10C and D). The inhibitor also compromised wound healing ability in LUAD cell lines, evidenced by delayed closure of scratch wounds at 12 and 24 h post-treatment in both A549 and H1299 cells (Figs. 10E and F), further corroborating its role in suppressing cell migration. Collectively, these results reinforce the hypothesis that MASTL contributes to the maintenance of CSC phenotypes.

Fig. 10.

MASTL Inhibition Suppresses self-renewal and metastasis of LUAD. (A-B) Sphere formation assays showing reduced CSC frequency in A549 and H1299 cells treated with 2.5 µM MKI-1. (C-D) Migration assays demonstrating suppressed migratory capacities at 2.5 µM and 5 µM MKI-1. (E-F) Scratch wound healing assays revealing delayed wound closure at 12 h and 24 h post-treatment. *P < 0.05, **P < 0.01,***P < 0.001, versus the 0µM according to Tukey’s test. The data are presented as the means ± SEM of at least three replicates.

Inhibition of MASTL attenuates tumor progression in vivo

To assess the therapeutic potential of MASTL inhibition in vivo, a xenograft mouse model was established (Fig. 11A). Tumor-bearing mice received treatment with the MASTL inhibitor MKI-1 at dosages of 25 mg/kg or 50 mg/kg, or a vehicle control. Consistent with prior in vitro data, administration of MKI-1 resulted in a significant, dose-dependent reduction in tumor growth. Representative images of the mice and excised tumors at the study endpoint visually corroborated the decrease in tumor size (Fig. 11B and C). Quantitative evaluation further substantiated these findings: measurements of terminal tumor weight demonstrated a pronounced reduction in the MKI-1–treated cohorts (Fig. 11D), and serial assessments of tumor volume, conducted twice a week, indicated sustained suppression of tumor progression over the treatment period (Fig. 11E). Notably, no significant alterations in body weight were detected across treatment groups (Fig. 11F), suggesting that MKI-1 was well tolerated at the administered doses. Collectively, these in vivo data provide compelling evidence that pharmacological targeting of MASTL via MKI-1 effectively and safely inhibits lung adenocarcinoma growth.

Fig. 11.

MASTL inhibition suppresses tumor growth in a LUAD xenograft model. (A) Schematic of the in vivo experimental timeline. (B) Representative images of mice from each group at the study endpoint. (C) Excised tumors from each treatment group. (D) Weights of tumors collected at the endpoint. (E) Tumor growth curves of xenografts treated with MKI-1 or vehicle. (F) Body weight changes of mice during the treatment period. Statistical significance was determined by one-way ANOVA followed by Tukey’s post hoc test: *p < 0.05; **p < 0.01; ***p < 0.001.

Discussion

Despite notable progress in targeted therapies and immunotherapies for LUAD, patient outcomes remain unsatisfactory, underscoring the critical need to identify novel biomarkers and therapeutic targets^37,38. In this study, we comprehensively investigate the multifaceted functions of MASTL in LUAD, demonstrating its role not only as an independent prognostic biomarker but also as a pivotal contributor to tumor malignancy through regulation of the cell cycle, modulation of the immune microenvironment, and mediation of specific phosphorylation events.

Our findings reveal that MASTL is markedly overexpressed in LUAD tissues, with elevated expression levels significantly correlating with poorer patient prognosis, consistent with observations in breast and colorectal cancers^10,39. Functional enrichment analyses indicate that LUAD samples exhibiting high MASTL expression are enriched in oncogenic pathways associated with organelle division, chromosome segregation, and cell cycle progression, which aligns with MASTL’s established role as a critical mitotic regulator⁴⁰. These results suggest that MASTL may facilitate LUAD tumorigenesis by promoting uncontrolled cellular proliferation and genomic instability, paralleling mechanisms reported in hepatocellular carcinoma^41,42.

Previous studies have highlighted the essential roles of immune cells in LUAD initiation and progression^43–49. Our investigation is the first to systematically elucidate MASTL’s involvement in fostering an immunosuppressive microenvironment within LUAD. Elevated MASTL expression is significantly associated with downregulation of dendritic cell functional genes, diminished immune cell infiltration, and reduced immune scores. Notably, high MASTL expression predicts poorer prognosis specifically in patients exhibiting increased Th1/Th2 ratios, whereas patients with low MASTL expression demonstrate enhanced responses to immunotherapy and a decreased risk of immune exclusion. These findings imply that MASTL may facilitate immune evasion by disrupting T cell homeostasis and antigen presentation, thereby influencing the efficacy of immunotherapeutic interventions.

Through co-expression and protein-protein interaction (PPI) network analyses, we identified several LUAD-associated proteins potentially interacting with MASTL, including CDK1, CCNA2, CCNB1, KIF11, and TTK. Importantly, prior studies have confirmed phosphorylation of MASTL mediated by the CDK1-CCNB1 complex^50,51. Our data further reveal co-expression relationships among CDK1, CCNB1, and MASTL in LUAD, suggesting a novel functional axis involving CDK1-CCNB1-MASTL in this malignancy. Literature reports that reduced expression of CDK1 and CCNB1 correlates with improved survival in LUAD patients⁵², reinforcing the significance of this regulatory pathway. Additionally, proteins such as CCNA2, KIF11, and TTK, which are upregulated in LUAD and associated with poor prognosis^53–55, were also identified as potential MASTL interactors. We recommend subsequent experimental validation of these predicted protein interactions using immunoprecipitation (Co-IP) and GST pull-down assays.

Phosphorylation analyses revealed significant modification of MASTL at serine 370 (S370), with phosphorylation levels closely linked to dysregulation of key signaling pathways including p53, mTOR, and WNT. These observations are consistent with previous reports implicating MASTL in the regulation of mTOR⁹ and WNT¹² pathways in other cancer types^6,11,56,57. To elucidate the functional implications of S370 phosphorylation, we propose future studies employing site-directed mutagenesis and CRISPR-Cas9-mediated gene editing approaches.

Regarding therapeutic interventions, our study confirms the antitumor efficacy of the selective MASTL inhibitor MKI-1^58,59. Although relatively high IC₅₀ values (10–12 µM) were observed in short-term MTT assays, MKI-1 at lower concentrations (2.5–5 µM) exerted significant inhibitory effects across multiple functional assays, including apoptosis induction, cell cycle arrest, colony formation, wound healing, tumor sphere formation, and cell migration. This apparent discrepancy primarily reflects fundamental differences in the biological endpoints assessed by these assays: MTT assays measure acute metabolic inhibition and cell viability within 48–72 h, necessitating higher drug concentrations to achieve significant effects; conversely, long-term assays such as colony and tumor sphere formation evaluate proliferative and self-renewal capacities under sustained drug exposure, while wound healing and migration assays assess cellular motility and repair potential—malignant phenotypes that are highly dependent on intact MASTL signaling. Consequently, sub-IC₅₀ concentrations of MKI-1 can effectively disrupt cell cycle progression and impair long-term survival and motility through continuous inhibition of the MASTL pathway, demonstrating potent functional suppression that extends beyond direct cytotoxicity. These findings align with MKI-1’s documented in vivo antitumor activity at well-tolerated doses, further substantiating the therapeutic potential of sustained MASTL inhibition in tumor growth suppression.

While our study provides comprehensive evidence supporting MASTL as a promising therapeutic target in LUAD, we acknowledge that current insights into MASTL’s protein interaction network, the functional role of S370 phosphorylation, and its immunoregulatory mechanisms are predominantly derived from bioinformatic and correlative analyses. Future investigations should focus on experimentally validating these protein interactions, conducting in-depth functional analyses of phosphorylation sites, and performing in vivo evaluations of MASTL’s immunomodulatory roles. Such investigations are critical for the successful translation of MASTL-targeted interventions into clinical applications. Nevertheless, our integrated multi-omics and experimental results collectively emphasize the central role of MASTL in promoting LUAD progression via cell cycle dysregulation, immune evasion, and specific phosphorylation signaling pathways, thereby underscoring its dual utility as both a prognostic biomarker and a therapeutic target.

Conclusions

Collectively, the results of this study identify MASTL as a pivotal oncogenic driver and a viable therapeutic target in LUAD. Our data indicate that MASTL overexpression is associated with poor clinical outcomes and contributes to tumorigenesis through disruption of cell cycle regulation, promotion of an immunosuppressive tumor microenvironment, and activation of specific phosphorylation signaling pathways. The demonstrated anti-tumor efficacy of the MASTL inhibitor MKI-1 further highlights the therapeutic promise of targeting MASTL. Although further investigation is necessary to fully elucidate the underlying molecular mechanisms, the convergent evidence presented herein provides a strong rationale for the development of MASTL-targeted therapies. Moreover, the prognostic models established in this study, together with mechanistic insights such as the identification of the S370 phosphorylation site, present concrete translational opportunities to enhance both diagnostic precision and targeted treatment strategies in LUAD.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

J. L.: Conceptualization, Methodology, Investigation, Writing – original draft, Project administration. J. L.: Software, Formal analysis, Data Curation, Visualization, Writing – review & editing. J. L.: Investigation, Visualization, Writing – review & editing. Y. J.: Validation, Writing – review & editing. H. H.: Validation, Writing – review & editing. L. Z.: Investigation, Validation. Y. L.: Methodology, Supervision, Project administration. Z. X.: Conceptualization, Resources, Data Curation, Writing – original draft, Supervision, Funding acquisition. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the National Natural Science Foundation of China (Grant No: 82203859), Shenzhen Medical Research Fund (Grant No: D2402019), the National Natural Science Foundation of China (Grant No: 82473130), Guangdong Basic and Applied Basic Research Foundation (Grant No: 2024A1515013092), Shenzhen Science and Technology Program (Grant No. JCYJ20240813150443056), Shenzhen Science and Technology Program (Grant No. JCYJ20220818102011022), and Shenzhen Key laboratory of Bone Tissue Repair and Translational Research (Project No. ZDSYS20230626091402006), Sanming Project of Medicine in Shenzhen (SZSM202311017), Research Start-up Fund of the Seventh Affiliated Hospital, Sun Yat-sen University (Project no. ZSQYBRJH0023).

Data availability

The data presented in this study are available on request from the corresponding author.

Declarations

Competing interests

The authors declare no competing interests.