GATA5 impedes NSCLC progression by activating SDPR and repressing CNR1 expression

Abstract

Background

Non-small cell lung cancer (NSCLC) remains a leading cause of cancer-related mortality, necessitating the identification of novel molecular mechanisms and therapeutic targets. GATA5, a member of the GATA family of transcription factors, has been implicated as a potential tumor suppressor in various malignancies; however, its role and regulatory mechanisms in NSCLC pathogenesis remain largely unexplored.

Methods

GATA5 expression and promoter methylation status were assessed in NSCLC cell lines, patient tissues, and public databases using RT-PCR, western blot, methylation-specific PCR, and immunohistochemistry. Functional studies were performed using in vitro assays (proliferation, colony formation, apoptosis, migration, invasion) and in vivo xenograft models. Mechanistic insights were obtained through RNA sequencing, chromatin immunoprecipitation, dual-luciferase reporter assays, and rescue experiments. The therapeutic potential of CNR1 inhibition was evaluated using the selective antagonist AM251 alone and in combination with cisplatin.

Results

GATA5 was frequently silenced by promoter hypermethylation in NSCLC, and low GATA5 expression correlated with poor patient prognosis. Functionally, GATA5 acted as a potent tumor suppressor, inhibiting proliferation, migration, invasion, and tumor growth while promoting apoptosis. Mechanistically, GATA5 directly bound to the SDPR promoter to activate its transcription, and SDPR was identified as a critical mediator of GATA5’s tumor-suppressive effects. The GATA5-SDPR axis negatively regulated CNR1, an oncogenic effector. Pharmacological inhibition of CNR1 with AM251 reversed the pro-tumorigenic effects of GATA5 loss and synergized with cisplatin to enhance anti-tumor efficacy both in vitro and in vivo.

Conclusions

Our findings establish the GATA5-SDPR-CNR1 axis as a critical tumor-suppressive pathway in NSCLC. Targeting CNR1 represents a promising therapeutic strategy to overcome cisplatin resistance and improve outcomes in patients with GATA5-silenced tumors.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12967-026-08727-5.

Background

Lung cancer remains the leading cause of cancer-related mortality worldwide, with non-small cell lung cancer (NSCLC) constituting the majority of cases [1]. Despite advances in targeted therapies and immunotherapy, a substantial proportion of patients experience primary or acquired resistance, underscoring the urgent need to elucidate novel molecular drivers and therapeutic vulnerabilities [2, 3]. Transcription factors and their epigenetic regulation play pivotal roles in tumorigenesis, offering promising avenues for intervention [4, 5]. The GATA family of zinc-finger transcription factors governs lineage-specific differentiation and tissue homeostasis across development [6, 7]. While GATA6 have been extensively characterized as an oncogenic or context-dependent regulator in lung adenocarcinoma [8, 9], the function and regulatory mechanism of GATA5 in NSCLC pathogenesis remain largely unexplored.

Emerging evidence has implicated GATA5 as a potential tumor suppressor in various malignancies, where it regulates key processes including cell proliferation, apoptosis, and differentiation [10–12]. Its expression is frequently modulated by promoter hypermethylation in colorectal and other cancers [13–15]. However, its role in lung carcinogenesis, particularly its transcriptional targets, downstream signaling pathways, and potential clinical utility in NSCLC, has not been systematically investigated. Moreover, the potential link between GATA5 and therapeutic resistance in NSCLC remains entirely unexplored. A comprehensive understanding of GATA5’s function and regulation in NSCLC could provide crucial insights into disease pathogenesis and reveal new therapeutic opportunities.

Importantly, the promoter hypermethylation that silences GATA5 not only explains its frequent downregulation in NSCLC but also represents a potential therapeutic entry point. DNA methyltransferase inhibitors (DNMTis) have shown promise in reactivating epigenetically silenced tumor suppressors in various cancers. Thus, targeting the methylation-mediated silencing of GATA5 could provide a novel strategy to restore its tumor-suppressive function and potentially overcome therapy resistance. This perspective bridges the gap between clinical observations (poor prognosis associated with GATA5 loss) and molecular mechanisms (epigenetic silencing), strengthening the translational narrative from discovery to application.

In this study, we systematically investigated the expression, function, and mechanism of GATA5 in NSCLC. We discovered that GATA5 is frequently silenced by promoter hypermethylation in NSCLC, and its loss correlates with poor patient prognosis. Through comprehensive in vitro and in vivo functional analyses, we established GATA5 as a potent tumor suppressor. Mechanistically, we identified Serum Deprivation Response (SDPR) as a direct transcriptional target mediating GATA5’s effects and further delineated a downstream oncogenic effector, Cannabinoid Receptor 1 (CNR1), which is repressed by the GATA5-SDPR axis. As a core receptor of the endocannabinoid system, CNR1 is predominantly expressed in the central nervous system [16, 17]. In alveolar rhabdomyosarcoma, its inhibition markedly reduces pulmonary metastasis formation [18]. Additionally, CNR1 promotes head and neck squamous cell carcinoma growth via the p38/MAPK pathway [19]. Collectively, these findings underscore a pro-tumorigenic role for CNR1 across multiple cancer types, although its function in lung cancer remains poorly understood. Importantly, we demonstrated the translational relevance of this axis by showing that pharmacological inhibition of CNR1 not only reverses the tumor-promoting effects of GATA5 loss but also synergizes with standard cisplatin chemotherapy to suppress tumor growth. Collectively, our work defines a previously unrecognized GATA5- SDPR-CNR1 tumor-suppressive axis, establishes CNR1 as a functionally relevant and druggable node in NSCLC, and provides a compelling rationale for evaluating CNR1 inhibition, alone or in combination, as a strategy to overcome chemoresistance.

Materials and methods

Cell lines

Eight lung cancer cell lines (A549, H2122, PC9, H1975, CAL-12T, H1651, H460, H1299) and a normal bronchial epithelial cell line (BEAS-2B) were obtained from the American Type Culture Collection (ATCC). All non-small cell lung cancer (NSCLC) cell lines were cultured in RPMI-1640 medium (Gibco, Thermo Fisher Scientific) supplemented with 10% fetal bovine serum (Vivacell) and 1% penicillin-streptomycin (Beyotime) and maintained at 37 °C in a humidified atmosphere containing 5% CO₂. To investigate the effects of cisplatin (CDDP) and the CNR1 inhibitor AM251, cells were treated with varying concentrations of CDDP (AbMole, M2223) or AM251 (AbMole, M2379) prior to subsequent experimental procedures.

Tumor samples

Human tumor tissues, paired adjacent non-tumor tissues, and normal lung tissues used in this study were obtained from Chongqing University Cancer Hospital and stored at -80 °C. All tissue samples were evaluated and confirmed by pathologists as suitable for research. Written informed consent was obtained from all participants.

RNA extraction, PCR and qPCR

Total RNA was isolated from cells or tissues using the Super FastPure Cell RNA Isolation Kit (NuoViz, Nanjing, China). For reverse transcription, 1 µg of total RNA was converted to cDNA in a 20 µL reaction. Semi-quantitative PCR was performed using Go-Taq DNA Polymerase (Promega) with GAPDH or β-actin as an internal control. The 10 µL reaction mixture contained 2 µL of cDNA template. Amplified products were separated on a 2% agarose gel for visualization. Quantitative real-time PCR (qPCR) was performed using SYBR Green (Promega) on an ABI 7500 Real-Time PCR System (Applied Biosystems) following the manufacturer’s protocol, with GAPDH or β-actin serving as the reference gene. Primer sequences are listed in the supplementary Table 1.

Plasmid construction and transfection

The following overexpression recombinant plasmids were constructed: pLV3-CMV-3×FLAG-Puro-GATA5, pLV3-CMV-3×FLAG-Puro-SDPR, and pLV3-CMV-3×FLAG-G418-CNR1. The following knockdown recombinant plasmids were constructed: PLKO.1-U6-shRNA-Puro-GATA5 and PLKO.1-U6-shRNA-G418-SDPR. According to the manufacturer’s instructions, the expression or control vectors were transfected into A549, H1299, H460, and H1651 cell lines using polyethylenimine (PEI, Thermo Fisher Scientific) as the transfection reagent. Stable cell lines were subsequently established by selection with puromycin (STEMCELL) or G418 (Biosharp). Puromycin selection used 1–2 µg/mL for 7–10 days, while G418 selection employed 400–800 µg/mL with a 2- to 4-week duration.

Methylation-specific PCR (MSP), 5-Aza-2’-deoxycytidine (Aza) and trichostatin A (TSA) treatment

Genomic DNA was extracted from tissues using a DNA extraction kit (TIANGEN, China). To assess the methylation status of GATA5, DNA was treated with bisulfite. Methylation-specific primers are provided in the Appendix. Methylation-specific PCR (MSP) was performed using Taq-Gold polymerase. PCR products were analyzed by agarose gel electrophoresis and visualized using a gel imaging system (BIO-RAD). To verify epigenetic regulation, A549 and H1299 cells were treated with 10 µM of the DNA methyltransferase inhibitor 5-Aza-2′-deoxycytidine (Aza, Merck, Germany) for 72 h, followed by treatment with 100 nM of the histone deacetylase inhibitor trichostatin A (TSA, Sigma-Aldrich, USA) for an additional 24 h [20]. RNA was then extracted for RT-PCR analysis. Primer sequences are listed in the supplementary Table 1.

Cell proliferation and colony formation assays

For the cell proliferation assay, cells were seeded at a density of 2000 cells per well in 96-well plates. Cell viability was measured at 0, 24, 48, and 72 h using a CCK-8 kit (bgbiotech) by reading the absorbance at 450 nm. For the colony formation assay, cells were plated at 600 cells per well in 6-well plates. After 14 days of culture, when visible colonies had formed, cells were fixed with 4% paraformaldehyde for 30 min at room temperature and stained with 0.1% crystal violet solution for 30 min. After washing, colonies were photographed and counted using ImageJ software.

Flow cytometry analysis of cell apoptosis

To analyze apoptosis, stably transfected cells were double-stained with Annexin V-FITC and PI, incubated in the dark, and then analyzed by flow cytometry.

Transwell assays

Cell migration and invasion abilities were assessed using Transwell chambers (8.0 μm pore size, Corning). For the invasion assay, chambers were pre-coated with Matrigel (Biosharp). Cells were seeded into the upper chamber and incubated for 24 (migration) or 48 (invasion) hours. Cells that migrated or invaded to the lower surface were fixed with 4% paraformaldehyde, stained with crystal violet, photographed under a microscope, and counted using ImageJ software.

H&E staining and immunohistochemistry

Human and mouse tumor tissue specimens were fixed in paraffin, dehydrated, embedded, and sectioned. Subsequently, H&E staining and immunohistochemistry were performed. The primary antibodies used were: anti-GATA5 (#55433-1-AP, Proteintech) and anti-Ki-67 (#9449S, Cell Signaling Technology).

Western blot

Protein samples were separated by 10% SDS-PAGE (Epizyme, China) and transferred onto PVDF membranes (Merck, Germany). The membranes were blocked with 5% non-fat milk for 2 h at room temperature and washed three times with TBST. They were then incubated overnight at 4 °C with primary antibodies on a rocking platform. Primary antibodies included: anti-GATA5 (#55433-1-AP, Proteintech), anti-SDPR (#12339-1-AP, Proteintech), anti-CNR1 (#17978-1-AP, Proteintech), and anti-GAPDH (#sc-47724, Santa Cruz). After washing, membranes were incubated with appropriate horseradish peroxidase-conjugated secondary antibodies for 1 h at room temperature. Following additional washes, protein bands were visualized using an ECL chemiluminescence detection system.

Dual luciferase assay

Potential GATA5 transcription factor binding sites in the SDPR promoter region were predicted using the JASPAR database. The wild-type SDPR promoter sequence containing the predicted binding site (SDPR-wt) and a mutant version (SDPR-mut) were cloned into a Fluc-SV40-hRluc reporter vector by Miaoling Biology (Wuhan, China). Twenty-four hours prior to transfection, vector control and GATA5-overexpressing A549 and H1299 cells were seeded into 24-well plates. The SDPR-wt or SDPR-mut reporter plasmids were then transfected into the cells. After 48 h, luciferase activity was measured using a dual-luciferase reporter assay kit (Promega) according to the manufacturer’s instructions.

Chromatin immunoprecipitation (ChIP) assay

A549 and H1299 cells were cross-linked with 1% formaldehyde for 10 min. Chromatin was then sonicated to shear DNA to fragments of 200–500 bp. The samples were incubated overnight at 4 °C with anti-FLAG antibody (#66008-4-Ig, Proteintech) or control IgG (#2729, Cell Signaling Technology). Immune complexes were captured using Protein A/G magnetic beads for 2 h. After washing and reverse cross-linking, the purified DNA was analyzed by PCR using primers specific for different regions of the SDPR promoter (sequences provided in the supplementary figure). Primer sequences are listed in the supplementary Table 1.

Subcutaneous tumor formation experiment in nude mice

Four- to six-week-old BALB/c nude mice (equal numbers of males and females) were used for the xenograft experiment. Cells in the logarithmic growth phase were resuspended in 100 µL PBS (5 × 10⁶ cells) and injected subcutaneously into the dorsal flank near the thigh of each mouse. Starting on day 4 post-injection, tumor length and width were measured every three days using a caliper, and tumor volume was calculated. Mice were euthanized when tumors reached an appropriate size. Tumors were excised, weighed, and photographed. Tumor tissues were collected for paraffin embedding, followed by H&E staining and immunohistochemistry.

RNA sequencing analysis

RNA sequencing was performed by Applied Protein Technology (Shanghai, China). RNA integrity was assessed using an Agilent Bioanalyzer 4150 system (Agilent Technologies, USA). Qualified RNA samples were used to construct sequencing libraries, which were also quality-checked on the Bioanalyzer. Sequencing was performed on the Illumina NovaSeq 6000 / MGISEQ-T7 platform. Differential gene expression analysis between groups was performed using the DESeq2 package in R. Genes with an absolute log2 fold change > 1 and an adjusted P-value (padj) < 0.05 were considered differentially expressed.

Statistical analysis

Statistical analyses were performed using GraphPad Prism 10 software. All data are presented as mean ± standard deviation (SD). Differences in mRNA expression between paired tumor and adjacent non-tumor tissues from lung cancer patients were analyzed using a paired two-tailed Student’s t-test. Differences between two independent groups were compared using an unpaired two-tailed Student’s t-test. Comparisons among multiple groups were performed using two-way analysis of variance (ANOVA). A P-value of less than 0.05 was considered statistically significant.

Result

GATA5 is hypermethylated and downregulated in NSCLC cells and tissues

We first investigated the GATA transcription factor family in NSCLC. A Venn analysis of GATA family genes and relevant datasets (GSE19804, GSE75037, GSE27262) identified four differentially expressed members: GATA2, GATA5, MTA3, and MTA1 (Fig. 1A). As the role of GATA5 remained unclear, we assessed its expression profile via TCGA and GTEx data. We confirmed markedly lower GATA5 mRNA levels in primary NSCLC tumors than in normal samples (Fig. 1B-D). Consistently, IHC and western blot revealed substantially decreased GATA5 protein expression in NSCLC tissues (Fig. 1E-F).

Fig. 1.

The levels of expression and the methylation status of the promoter region of GATA5 in tissues and cell lines derived from NSCLC. (A) The Venn diagram illustrates the identification of four differentially expressed genes derived from the GATA5 gene family, utilizing data from the GSE19804, GSE75037, and GSE27262 datasets. (B) Comparative analysis of GATA5 expression levels between healthy human lung tissue and NSCLC cancer tissue using the TCGA database. (C) Comparative analysis of GATA5 expression levels between healthy human lung tissue and NSCLC cancer tissue using the combined TCGA+GTEx databases. (D) Analysis of GATA5 expression differences between adjacent non-tumor tissue and tumor tissue in lung cancer patients using the TCGA database. (E) Immunohistochemical images of GATA5 in NSCLC tissue and paired normal tissue. (F) The detection of GATA5 protein expression was conducted using western blot in four pairs of NSCLC tissues alongside their corresponding normal tissue samples. (G) The expression levels of GATA5 were assessed in NSCLC cell lines (n = 8) and a single sample of normal lung epithelial cells (n = 1) using RT-PCR and western blot analysis. (H) The Kaplan-Meier survival curves derived from the Kaplan-Meier Plotter database illustrate the relationship between GATA5 expression levels and the survival outcomes of patients diagnosed with NSCLC. (I) An examination of the methylation status of the GATA5 promoter in LUAD was conducted utilizing data from The Cancer Genome Atlas (TCGA) database. (J) An examination of the methylation status of the GATA5 promoter in LUSC utilizing data from TCGA. (K) Methylation status of GATA5 sites in blood and tissue. (L) MSP detection of GATA5 methylation in NSCLC cell lines. (M) Detection of gene expression in NSCLC cell lines following administration of demethylating agents. The data were presented as the mean ± SD. ***p < 0.001

We next examined GATA5 expression in a panel of NSCLC cell lines. Both RT-PCR and western blot analysis confirmed notably low GATA5 levels in A549, H1975, CAL-12T, and H1299 cells (Fig. 1G). Clinically, Kaplan-Meier analysis indicated that higher GATA5 expression correlated with significantly better overall survival in NSCLC patients (Fig. 1H).

Analysis of TCGA methylation data revealed significantly higher GATA5 promoter methylation in both LUAD and LUSC tumors compared to normal tissues (Fig. 1I-J). In patient-derived samples, methylation levels at four specific GATA5 loci (179, 198, 222, 669) were substantially elevated in circulating tumor DNA (ctDNA) and tumor tissue DNA relative to cell-free DNA (cfDNA) (Fig. 1K). Methylation-specific PCR (MSP) further confirmed prominent GATA5 promoter methylation in NSCLC cell lines (Fig. 1L). Treatment with the DNA methyltransferase inhibitor Aza and the histone deacetylase inhibitor TSA restored GATA5 expression in A549 and H1299 cells, confirming that its silencing is mediated by epigenetic mechanisms (Fig. 1M). Collectively, these findings demonstrate that GATA5 is frequently downregulated in NSCLC due to promoter hypermethylation, and its low expression is associated with poor patient prognosis.

Overexpression of GATA5 inhibits NSCLC progression in vitro and in vivo

To assess the functional role of GATA5, we overexpressed it in low-expressing A549 and H1299 cells (Fig. 2A). This overexpression significantly inhibited cell proliferation (CCK-8 assay) and colony formation (Fig. 2B-C), while promoting apoptosis (flow cytometry) (Fig. 2D). Furthermore, Transwell assays demonstrated that GATA5 overexpression markedly suppressed the migratory and invasive capacities of NSCLC cells (Fig. 2E-F).

Fig. 2.

The heightened expression of GATA5 was found to suppress the progression of NSCLC cells both in vivo and in vitro. (A) RT-PCR and western blot validation of GATA5 overexpression transfection efficiency. (B) The CCK8 assay was employed to assess the impact of GATA5 overexpression on the proliferation of NSCLC cells. (C) Colony formation assays were conducted to evaluate the effects of GATA5 overexpression on the proliferation of NSCLC cells. (D) Flow cytometry to examine the influence of GATA5 overexpression on NSCLC cell apoptosis. (E-F) Transwell assay to determine the effect of GATA5 overexpression on NSCLC cell migration and invasion capabilities. Subcutaneous tumor formation in nude mice to evaluate the effect of GATA5 overexpression on H1299 cell proliferation in vivo: (G) Images of subcutaneous tumors in nude mice, (H) Tumor growth curve, (I) Statistical analysis of excised tumor weights. ***p < 0.001

In a subcutaneous xenograft model, mice injected with GATA5-overexpressing H1299 cells exhibited significantly slower tumor growth and lower final tumor weights compared to controls (Fig. 2G-I). Histological analysis (H&E staining) confirmed reduced tumor cell density, and IHC revealed a significantly lower Ki67 proliferation index in GATA5-overexpressing tumors (Fig. S1A). Together, these data establish GATA5 as a potent tumor suppressor that constrains NSCLC cell growth, survival, migration and invasion both in vitro and in vivo.

Knockdown of GATA5 promotes NSCLC progression in vitro and in vivo

Conversely, we knocked down GATA5 in high-expressing H460 and H1651 cells (Fig. 3A). GATA5 knockdown enhanced cell proliferation and colony formation (Fig. 3B-C), reduced apoptosis (Fig. 3D), and increased migration and invasion (Fig. 3E). In vivo, knockdown of GATA5 accelerated tumor growth and increased tumor weight (Fig. 3F-H), resulting in denser tumor tissue and a higher Ki67 index (Fig. S2A).

Fig. 3.

The silencing of GATA5 facilitated the progression of NSCLC cells both in vivo and in vitro. (A) qRT-PCR and western blot validation of GATA5 knockdown efficiency. (B) The CCK-8 assay was utilized to evaluate the impact of GATA5 silencing on the proliferation of NSCLC cells. (C) The colony formation assay demonstrated the influence of GATA5 silencing on the proliferation of NSCLC cells. (D) Flow cytometric analysis demonstrated the impact of GATA5 silencing on apoptosis in NSCLC cells. (E) The Transwell assay was utilized to assess the impact of GATA5 knockdown on the migratory and invasive abilities of NSCLC cells. The development of subcutaneous tumors in nude mice demonstrated the influence of GATA5 silencing on the proliferation of H1651 cells in vivo: (F) Photographs of subcutaneous tumors in nude mice, (G) Tumor growth curves, (H) Statistical analysis of excised tumor weights. **p < 0.01, ***p < 0.001

GATA5 directly binds to the SDPR promoter to activate transcription

To delineate the transcriptional targets of GATA5, we performed RNA sequencing in H1299 cells stably overexpressing GATA5 or empty vector (Supplementary Table 2). Analysis with thresholds of |log2FC|≥1 and p adj < 0.05 identified 1,875 differentially expressed genes (610 upregulated; 1,025 downregulated) (Fig. 4A), with a heatmap visualizing the top 30 from each category (Fig. 4B, Fig. S3A). To pinpoint direct targets, we integrated our RNA-seq data within silico predictions from FIMO_JASPAR and PWMEnrich_JASPAR databases, as well as GATA5-associated genes from TCGA-NSCLC, via a Venn analysis. This approach identified 27 high-confidence candidate target genes (13 upregulated, 14 downregulated) (Fig. 4C, Fig. S3B). Among these, SDPR (CAVIN2), a caveolae-associated gene previously reported to be downregulated by promoter methylation and to inhibit migration and invasion while promoting apoptosis in lung cancer, exhibited the most significant upregulation and was selected for further validation [21, 22].

Fig. 4.

GATA5 promoted SDPR transcription in NSCLC cells. (A) The volcano plot illustrates the comprehensive distribution of genes that exhibit differential expression between H1299 cells with GATA5 overexpression and those containing an empty vector. (B) A heatmap illustrating the results of hierarchical clustering analysis was generated to compare the upregulated genes between H1299 cells overexpressing GATA5 and those containing an empty vector. (C) Combining the FIMO-JASPAR, PWMEnrich_JASPAR datasets, upregulated genes in RNA sequencing, GATA5 related genes in TCGA database, and 13 target genes were identified through Venn analysis. (D) A Spearman correlation analysis was conducted to evaluate the association between the expression levels of SDPR and GATA5. (E) qRT-PCR revealed mRNA expression of SDPR in cells stably transfected with GATA5. (F) Dual luciferase assays measured luciferase activity from the SDPR promoter in A549 and H1299 cells expressing empty vector versus GATA5 overexpression. (G) ChIP-qPCR detected interactions between GATA5 and the SDPR promoter region. (H) The TCGA database was utilized to analyze and contrast the expression levels of SDPR in healthy human lung tissue with those found in NSCLC tissue. (I) Comparative analysis of SDPR expression levels between healthy human lung tissue and NSCLC cancer tissue using the combined TCGA+GTEx databases. (J) Analysis of SDPR expression differences between adjacent non-tumor tissue and tumor tissue in lung cancer patients using the TCGA database. (K) The Kaplan-Meier survival analysis demonstrated the correlation between the expression levels of SDPR and the survival outcomes in patients diagnosed with NSCLC. The data were presented as the mean ± SD. ***p < 0.001

A positive correlation between GATA5 and SDPR expression was confirmed in NSCLC tissues (Fig. 4D). Ectopic expression of GATA5 in A549 and H1299 cells significantly increased both SDPR mRNA (qRT-PCR) and protein (western blot) levels (Fig. 4E, Fig. S3C). A dual-luciferase reporter assay demonstrated that GATA5 overexpression activated the wild-type SDPR promoter but not a mutant version in which these sites were altered (Fig. 4F). To determine if this regulation was direct, we used the JASPAR database to predict two potential GATA5 binding sites on the SDPR promoter. Chromatin immunoprecipitation (ChIP)-qPCR analysis further confirmed the direct enrichment of GATA5 at both predicted binding sites on the endogenous SDPR promoter (Fig. 4G).

Consistent with its role as a tumor suppressor, analysis of TCGA and GTEx data revealed that SDPR mRNA expression was significantly lower in primary NSCLC tumors compared to normal tissues (Fig. 4H-I). This downregulation was corroborated in our cohort of TCGA-NSCLC patient-matched tumor and adjacent non-tumor tissues (Fig. 4J). Clinically, Kaplan-Meier survival analysis indicated that higher SDPR expression was associated with significantly better overall survival in NSCLC patients (Fig. 4K). Moreover, IHC and western blot revealed substantially decreased SDPR protein expression in NSCLC tissues (Fig. S3D-E). In summary, we identify SDPR as a direct transcriptional target of GATA5, whose expression is frequently lost in NSCLC and is associated with improved patient survival.

GATA5 suppresses NSCLC malignancy by promoting SDPR expression

We next performed functional rescue experiments to determine if SDPR is a critical mediator of GATA5’s tumor-suppressive effects. A549 and H1299 cells were co-transfected with GATA5 overexpression and SDPR knockdown plasmids (transfection efficiency confirmed by western blot, Fig. 5A). In vitro functional assays demonstrated that silencing SDPR effectively rescued the suppression of cell proliferation (CCK-8, Fig. 5B), colony formation (Fig. 5C), and the induction of apoptosis (Fig. 5D) mediated by GATA5 overexpression. Furthermore, SDPR knockdown reversed the inhibitory effects of GATA5 on cell migration and invasion (Transwell assays, Fig. 5E). These findings were substantiated in vivo, where SDPR knockdown significantly attenuated the ability of GATA5 overexpression to inhibit subcutaneous tumor growth in nude mice (Fig. 5F-H). Thus, the tumor-suppressive effects of GATA5 are largely dependent on its ability to transcriptionally activate SDPR expression.

Fig. 5.

The inhibition of SDPR has the potential to negate the tumor-inhibitory influence of GATA5 in NSCLC cells. (A) Western blot validation of co-transfection efficiency for GATA5 overexpression and SDPR knockdown plasmids. (B-C) CCK8 and colony formation assays demonstrated the effect of SDPR knockdown on GATA5-induced proliferation in NSCLC cells. (D) Flow cytometry analysis of the effect of SDPR knockdown on GATA5-induced apoptosis in NSCLC cells. (E) Transwell assay demonstrated the impact of SDPR knockdown on GATA5-induced migration and invasion in NSCLC cells. Subcutaneous tumor formation in nude mice revealed the effect of SDPR knockdown on GATA5-induced in vivo suppression of NSCLC cell growth: (F) Photographs of subcutaneous tumors in nude mice, (G) Tumor growth curves, (H) Statistical analysis of excised tumor weights. The data were presented as the mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001

Upregulation of CNR1 reverses the tumor-suppressive effect of SDPR in NSCLC

To elucidate the downstream mechanism of the GATA5-SDPR axis, we integrated our GATA5 RNA-seq data with a published SDPR-related gene set [23]. Venn diagram analysis identified eight overlapping candidate genes, including five upregulated and three downregulated targets (Fig. 6A, Fig. S4A). We focused on Cannabinoid Receptor 1 (CNR1), a downregulated gene, is most regulated by GATA5 and SDPR (Fig. S4B), as Kaplan-Meier analysis indicated that high CNR1 expression correlated with poor survival in NSCLC patients (Fig. S4C), and it has been implicated in promoting growth in other cancers, though its role in lung cancer remains unclear [18, 24, 25].

Fig. 6.

The overexpression of CNR1 has the potential to counteract the tumor-suppressive influence of SDPR in NSCLC cells. (A) Venn diagram analysis of genes downregulated by GATA5 and SDPR in transcriptomic sequencing. (B) qRT-PCR detection of CNR1 mRNA changes following co-transfection of SDPR knockdown and GATA5 overexpression plasmids. (C) Western blot validation of co-transfection efficiency for SDPR and CNR1 overexpression plasmids. (D-E) CCK8 and colony formation assays demonstrated the effect of CNR1 overexpression on SDPR-induced proliferation in NSCLC cells. (F) Flow cytometry analysis demonstrated the effect of CNR1 overexpression on SDPR-induced apoptosis in NSCLC cells. (G-H) Transwell assays demonstrated the effect of CNR1 overexpression on SDPR-induced migration and invasion capabilities in NSCLC cells. Subcutaneous tumor formation experiments in nude mice revealed the impact of CNR1 overexpression on SDPR-mediated in vivo suppression of NSCLC cell growth: (I) Photographs of subcutaneous tumors in nude mice, (J) Tumor growth curves, (K) Statistical analysis of excised tumor weights. The data were presented as the mean ± SD. **p < 0.01, ***p < 0.001

IHC and western blot analysis confirmed elevated CNR1 protein levels in NSCLC tissues compared to adjacent normal tissues (Fig. S4D-E). To validate the proposed GATA5-SDPR-CNR1 axis, qRT-PCR showed that SDPR knockdown reversed the downregulation of CNR1 mRNA induced by GATA5 overexpression (Fig. 6B). IHC revealed that GATA5 overexpression in subcutaneous xenografts upregulated SDPR but downregulated CNR1, whereas GATA5 knockdown produced opposite effects (Fig. S1A, Fig. S2A). Subsequent rescue experiments were conducted by co-transfecting A549 and H1299 cells with SDPR and CNR1 overexpression plasmids (efficiency confirmed by WB, Fig. 6C). Ectopic expression of CNR1 effectively rescued the SDPR-mediated suppression of proliferation (Fig. 6D), colony formation (Fig. 6E), and apoptosis (Fig. 6F), as well as the inhibition of migration and invasion (Fig. 6G-H). In vivo, CNR1 overexpression markedly reversed the tumor growth inhibition driven by SDPR in the subcutaneous xenograft model (Fig. 6I-K). These results establish CNR1 as a key downstream effector that functionally opposes the tumor-suppressive actions of SDPR.

Upregulation of CNR1 counteracts the tumor-suppressive role of GATA5

We conducted a final set of experiments to directly link CNR1 to GATA5 function. Immunofluorescence analysis indicated that GATA5 overexpression led to reduced CNR1 protein levels. We then co-transfected A549 and H1299 cells with GATA5 and CNR1 overexpression plasmids (Fig. 7A). Overexpression of CNR1 successfully rescued the GATA5-induced suppression of proliferation (CCK-8, Fig. 7B), colony formation (Fig. 7C), and apoptosis (Fig. 7D), and it abrogated the inhibitory effects of GATA5 on cell migration and invasion (Fig. 7E-F). The in vivo subcutaneous tumorigenesis assay confirmed that CNR1 overexpression could counteract the potent tumor-suppressive effect of GATA5 (Fig. 7G-I). Collectively, these data demonstrate that CNR1 acts downstream of the GATA5-SDPR axis and its overexpression is sufficient to antagonize the tumor-suppressive function of GATA5 in NSCLC.

Fig. 7.

The heightened expression of CNR1 has the potential to counteract the tumor-inhibiting influence of GATA5 within NSCLC cells. (A) Western blot validation of co-transfection efficiency for GATA5 and CNR1 overexpression plasmids. (B-C) CCK8 and colony formation assays demonstrated the effect of CNR1 overexpression on GATA5-induced proliferation in NSCLC cells. (D) Flow cytometry revealed the impact of CNR1 overexpression on GATA5-mediated apoptosis promotion in NSCLC cells. (E-F) Transwell assays demonstrated the effect of CNR1 overexpression on GATA5-induced migration and invasion capabilities in NSCLC cells. Subcutaneous tumor formation experiments in nude mice revealed the impact of CNR1 overexpression on the in vivo inhibitory effect of GATA5 on NSCLC cells: (G) Photographs of subcutaneous tumors in nude mice, (H) Tumor growth curves, (I) Statistical analysis of excised tumor weights. The data were presented as the mean ± SD. **p < 0.01, ***p < 0.001

The CNR1 inhibitor AM251 modulates the GATA5-SDPR-CNR1 axis in NSCLC

To further validate the functional significance of the GATA5-SDPR-CNR1 axis in NSCLC progression, we employed AM251, a potent and selective CNR1 antagonist, to pharmacologically inhibit CNR1 downstream signaling. After transfection of H460 and H1651 cells with GATA5 knockdown plasmids, cells were treated with AM251 (transfection and inhibition efficiency confirmed by western blot, Fig. 8A). Functional assays revealed that AM251 treatment effectively rescued the pro-proliferative phenotype induced by GATA5 knockdown, as evidenced by CCK-8 and colony formation assays (Fig. 8B-C). Flow cytometric analysis demonstrated that AM251 counteracted the anti-apoptotic effect resulting from GATA5 silencing (Fig. 8D). Similarly, Transwell assays showed that AM251 attenuated the enhanced migratory and invasive capacities conferred by GATA5 knockdown (Fig. 8E-F). These in vitro findings were corroborated in vivo, where AM251 treatment significantly reversed the promotive effect of GATA5 knockdown on subcutaneous tumor growth (Fig. 8G-I). These results demonstrate that pharmacological inhibition of CNR1 by AM251 potently reverses the oncogenic effects of GATA5 loss, providing direct evidence that the tumor-suppressive function of the GATA5-SDPR axis is mediated through suppression of CNR1 signaling.

Fig. 8.

The CNR1 inhibitor AM251 could modulate the GATA5-SDPR-CNR1 axis in NSCLC cells. (A) Western blot validation of protein expression changes following GATA5 knockdown and addition of the CNR1 inhibitor AM251. (B-C) CCK8 and colony formation assays demonstrated the effect of AM251 addition on GATA5 knockdown-induced proliferation in NSCLC cells. (D) Flow cytometry revealed the impact of AM251 addition on GATA5 knockdown-induced apoptosis promotion in NSCLC cells. (E-F) Transwell assays demonstrated the effect of AM251 on GATA5-mediated promotion of NSCLC cell migration and invasion. Subcutaneous tumor formation experiments in nude mice revealed the impact of AM251 on GATA5 knockdown-induced promotion of NSCLC cell growth in vivo: (G) Photographs of subcutaneous tumors in nude mice, (H) Tumor growth curves, (I) Statistical analysis of excised tumor weights. The data were presented as the mean ± SD. **p < 0.01, ***p < 0.001

GATA5 enhances the sensitivity of NSCLC cells to cisplatin

Based on our previous finding that GATA5 overexpression promotes apoptosis, we investigated whether GATA5 could modulate the sensitivity of NSCLC cells to cisplatin, a first-line chemotherapeutic agent that induces apoptotic cell death [26–28]. CCK-8 assays demonstrated that GATA5 overexpression synergized with cisplatin to inhibit NSCLC cell proliferation more effectively than cisplatin treatment alone (Fig. 9A). Apoptosis assays further confirmed that GATA5 overexpression significantly enhanced cisplatin-induced cell death (Fig. 9B). The synergistic effect was validated in vivo, where GATA5 overexpression combined with cisplatin treatment resulted in the most potent inhibition of subcutaneous tumor growth compared to all other groups (Fig. 9C-E). These findings indicate that GATA5 overexpression sensitizes NSCLC cells to cisplatin-induced apoptosis and growth inhibition, both in vitro and in vivo, suggesting a role for GATA5 in modulating chemotherapeutic response.

Fig. 9.

GATA5 enhanced the sensitivity of NSCLC cells to cisplatin, while the CNR1 inhibitor AM251 enhanced cisplatin-induced sensitivity in NSCLC cells. (A) CCK8 assay demonstrated proliferation of Vector and GATA5-overexpressing NSCLC cells with and without cisplatin treatment. (B) Flow cytometry revealed apoptosis in Vector and GATA5-overexpressing NSCLC cells with and without cisplatin treatment. The effect of Vector and GATA5-overexpressing NSCLC cells with and without cisplatin treatment on subcutaneous tumors in nude mice: (C) Photographs of subcutaneous tumors in nude mice, (D) Tumor growth curves, (E) Statistical analysis of excised tumor weights. (F) CCK8 assay resulted for A549 and H1299 cell proliferation following treatment with AM251, CDDP, or combined AM251 and CDDP. The effect of AM251, CDDP or AM251 and CDDP combination therapy on subcutaneous tumors in nude mice: (G) Photographs of subcutaneous tumors in nude mice, (H) Tumor growth curves, (I) Statistical analysis of excised tumor weights. The data were presented as the mean ± SD. **p < 0.01, ***p < 0.001

The CNR1 antagonist AM251 enhances cisplatin sensitivity in NSCLC

We next evaluated whether inhibiting the downstream effector CNR1 could similarly enhance chemosensitivity. CCK-8 assays revealed that the combination of AM251 and cisplatin resulted in the most potent suppression of NSCLC cell proliferation compared to either agent alone, and this difference was statistically significant (Fig. 9F). This strong synergistic effect was confirmed in vivo, where the co-administration of AM251 and cisplatin exhibited the most robust inhibition of subcutaneous tumor growth compared to all monotherapy groups (Fig. 9G-I). Therefore, pharmacological inhibition of CNR1 synergizes with standard cisplatin chemotherapy to achieve superior anti-tumor efficacy, highlighting a promising therapeutic strategy of combining CNR1 antagonists with conventional chemotherapeutics for NSCLC treatment.

Discussion

Lung cancer, particularly NSCLC, represents a highly aggressive malignancy characterized by pronounced invasiveness, rapid progression, and frequent acquisition of therapy resistance. These attributes contribute to poor patient prognosis and low 5-year survival rates, presenting major challenges for current anticancer strategies [29, 30]. There is thus a pressing need to identify reliable biomarkers for early detection and improved therapeutic intervention. Given its frequent and tumor-specific promoter hypermethylation in NSCLC, GATA5 also holds promise as a non-invasive diagnostic biomarker for early detection. Plasma-based assays have shown that GATA5 methylation, when incorporated into multi-gene panels, can detect early-stage NSCLC with high sensitivity and specificity [31, 32]. Moreover, sputum assessment of GATA5 methylation has been associated with significantly increased lung cancer risk [33]. These findings suggest that GATA5 methylation could complement low-dose CT screening, improve risk stratification of indeterminate nodules, and facilitate early diagnosis—an area warranting prospective validation.

The GATA family of transcription factors, which share a conserved zinc-finger DNA-binding domain, play diverse roles in cancer biology [34]. Accumulating evidence has implicated several GATA members in NSCLC pathogenesis. For instance, GATA1 upregulates HSD17B6 to suppress cisplatin resistance [35]; GATA2 overexpression inhibits lung cancer progression [36]; GATA3 modulates radiotherapy sensitivity in NSCLC [37]; GATA4 induces cellular senescence by downregulating WNT7B [38]; and GATA6 impairs proliferation and migration via metabolic reprogramming [8]. Among these, GATA5 has been associated with tumorigenesis in various cancers: it acts synergistically with ARDDC3 and ELP3 methylation to drive malignant phenotypes in breast cancer [39], and its reduced expression correlates with poor recurrence-free survival in renal clear cell carcinoma, suggesting potential prognostic utility [40]. However, the specific role of GATA5 in NSCLC has remained unclear.

Here, we demonstrate that GATA5 acts as a novel tumor suppressor in NSCLC. Integrated bioinformatic analyses reveals that GATA5 is frequently downregulated and hypermethylated in NSCLC tissues, and its high expression is associated with improved patient survival. We validated these findings in clinical samples and cell lines, confirming that promoter hypermethylation is a primary mechanism silencing GATA5 expression. Clinically, promoter hypermethylation-driven silencing of GATA5 is a tumor-specific event. Its methylation status, detectable in easily accessible specimens like plasma and sputum, holds promise as a non-invasive biomarker for early detection and lung cancer risk assessment [31, 33]. Functionally, GATA5 overexpression suppressed proliferation, migration, and invasion while promoting apoptosis in NSCLC cells. Consistent with these in vitro results, GATA5 inhibited subcutaneous tumor growth in vivo, whereas its knockdown exerted opposing effects.

Mechanistically, we uncover a previously uncharacterized transcriptional cascade: GATA5 directly binds the SDPR promoter and activates its transcription, leading to increased SDPR protein levels. Rescue experiments confirm that SDPR knockdown fully abrogates the tumor-suppressive effects of GATA5, establishing SDPR as an indispensable mediator in this axis. Similarly, SDPR is epigenetically silenced in breast cancer, and its loss promotes growth and metastasis [41]. In KRAS-mutant NSCLC, reduced SDPR expression is linked to an immunosuppressive microenvironment and poor prognosis [42]. Mechanistically, SDPR collaborates with CAVIN1 to facilitate EGFR endocytosis and degradation, thereby inhibiting AKT/STAT3 signaling [43].

Further investigation into SDPR-mediated mechanisms led to the identification of CNR1 as a key downstream effector. CNR1 expression was negatively regulated by the GATA5-SDPR axis, and CNR1 overexpression reversed the inhibitory effects of both SDPR and GATA5 on NSCLC malignancy. Although CNR1 is reportedly oncogenic in Hodgkin lymphoma [24] and endometrial cancer [25], its role in lung cancer has been poorly characterized. Notably, high CNR1 expression predicted unfavorable survival in NSCLC patients. We further showed that AM251, a selective CNR1 inhibitor, counteracted the oncogenic effects of GATA5 knockdown, confirming functional involvement of CNR1 in the proposed axis. Although AM251 validated CNR1 as a functional target, its off-target effects, neuropsychiatric toxicity, and unfavorable pharmacokinetics preclude clinical use. To translate our findings, peripherally restricted CB1 antagonists or localized delivery strategies should be prioritized, as they may retain anti-tumor efficacy without central adverse effects. Importantly, this GATA5-SDPR-CNR1 axis represents a distinct molecular mechanism, underscoring functional divergence within the GATA family despite structural homology in DNA binding.

Given the continued clinical challenge of cisplatin resistance in NSCLC [44, 45], and considering GATA5’s potent pro-apoptotic activity [46], we next evaluated its impact on chemosensitivity. Strikingly, GATA5 overexpression synergizes with cisplatin to suppress tumor growth both in vitro and in vivo. Moreover, co-administration of AM251 and cisplatin yields significantly enhanced antitumor efficacy compared to either agent alone, providing compelling preclinical evidence that CNR1 inhibition can overcome cisplatin resistance.

The upstream drivers of GATA5 promoter hypermethylation in NSCLC remain incompletely understood. Tobacco smoke exposure is a well-established environmental trigger [32, 33]. GATA5 promoter methylation has been detected in sputum of cancer-free heavy smokers and predicts subsequent lung cancer risk [33]. In NSCLC patients, GATA5 methylation varies significantly with smoking history [32]. Environmental and occupational exposures are also implicated: radiation exposure at the MAYAK nuclear facility increased GATA5 methylation prevalence in lung adenocarcinomas [47]. Mechanistically, tobacco carcinogens such as 4-(methylnitro- samino)-1-(3-pyridyl)-1-butanone (NNK) induce DNMT1 accumulation [48], and both DNMT1 and DNMT3B overexpression correlate with promoter hypermethylation in smoking-related lung cancer [49].Therefore, we speculate that DNMT1 and/or DNMT3B may be the principal enzymes mediating GATA5 silencing. However, whether these isoforms are directly recruited to the GATA5 promoter, or whether other DNMT family members (e.g., DNMT3A) also contribute, requires systematic investigation through chromatin immunoprecipitation and CRISPR-based knockout approaches in future studies. The molecular mechanism by which SDPR negatively regulates CNR1 expression remains elusive. Given that SDPR is a caveolae-associated protein involved in EGFR endocytosis and downstream signaling modulation [43], it is plausible that SDPR may regulate CNR1 indirectly through affecting transcription factors or signaling cascades (e.g., AKT/STAT3) that control CNR1 transcription, rather than through direct promoter binding. Alternatively, SDPR might influence CNR1 mRNA stability or protein turnover via post-transcriptional or post-translational mechanisms. Elucidating these connections represents a compelling direction for future research and will further solidify the mechanistic framework of the GATA5-SDPR-CNR1 axis.

Conclusion

In summary, our study identifies GATA5 as a tumor suppressor gene frequently silenced by promoter hypermethylation in NSCLC. We delineate a novel GATA5-SDPR-CNR1 regulatory axis that significantly influences NSCLC pathogenesis and chemoresistance. These findings nominate GATA5 methylation as a potential prognostic biomarker, and support the rational development of CNR1-targeted combination therapies as a viable strategy for GATA5-deficient NSCLC patients.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

Aza

5-Aza-2’-deoxycytidine

CDDP

Cisplatin

CHIP

Chromatin immunoprecipitation

H&E

Hematoxylin and Eosin

IHC

Immunohistochemistry

LUAD

Lung adenocarcinoma

LUSC

Lung squamous cell carcinoma

MSP

Methylation-Specific

NSCLC

Non-small cell lung cancer

PCR

Polymerase Chain Reaction

PBS

Phosphate-Buffered Saline

qRT-PCR

Quantitative Reverse Transcription PCR

TSA

Trichostatin A

Author contributions

S.Y. designed and performed the experiments, analyzed the data, and drafted the manuscript. J.H., F.D. and C.X. performed Bioinformatics analysis. L.R., W.C. and N.L. contributed to animal studies. Y.H., W.S. and X.X. performed statistical analysis. Y.L., Z.Z., Y.T. and T.W. revised the manuscript. T.X. and Y.W. conceived and supervised the study and interpreted the results for important intellectual content. All authors reviewed and approved the final version of the manuscript.

Funding

This study was supported by the Chongqing Science and Health Joint Medical Research Project (No. 2023GGXM002).

Data availability

All data can be provided as needed.

Declarations

Ethics approval and consent to participate

All animal experiments were approved by the Animal Welfare Ethics Committee of Chongqing University Affiliated Cancer Hospital (No. CQCH-LAE-A0000202078) and conducted to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health.

Consent for publication

Not applicable.

Conflict of interest

The author declares no conflict of interest.