Targeting ABCA3 impedes tumor progression and EGFR-TKI resistance in EGFR-mutant LUAD

Abstract

Lung adenocarcinoma (LUAD) with epidermal growth factor receptor (EGFR) mutations is prevalent in East Asian NSCLC patients and responds initially to EGFR-tyrosine kinase inhibitors (EGFR-TKIs), but resistance inevitably develops. This study identifies ATP-binding cassette subfamily A member 3 (ABCA3) as a key protein involved in tumor progression and TKI resistance in EGFR-mutant cancers. ABCA3 was significantly upregulated in EGFR-mutant LUAD cells compared to WT, promoting cell viability, proliferation, migration, and clonogenicity, while suppressing apoptosis. Mechanistic analyses revealed that ABCA3 enhanced cholesterol uptake and activated the PI3K/AKT/mTOR pathway, contributing to tumor growth. Moreover, the transcription factor SP1 was found to induce ABCA3 expression, especially following EGFR-TKI treatment. ABCA3 was markedly elevated in EGFR-TKI-resistant cell line, and its inhibition restored drug sensitivity. These findings suggest that ABCA3 plays a central role in mediating both tumor progression and resistance in EGFR-mutant LUAD. Targeting ABCA3 may represent a promising strategy to suppress tumor growth and overcome EGFR-TKI resistance, providing a novel therapeutic avenue for patients with EGFR-mutated non-small cell lung cancer.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12935-026-04221-6.

Introduction

Lung cancer ranks first in terms of cancer incidence and mortality in the world [1]. Non-small cell lung cancer (NSCLC) counts for 85% of total lung cancer diagnoses. Within NSCLC classification, lung adenocarcinoma (LUAD) is the most prevalent histologic subtype [2]. One prevalent genetic mutation in LUAD is the epidermal growth factor receptor (EGFR) mutation, which is especially prominent in Asians, constituting up to 50% of cases [3]. Although third-generation EGFR tyrosine kinase inhibitors (TKIs), Osimertinib, has been accepted as a first-line therapy for advanced EGFR-mutant NSCLC in recent years [4, 5], acquired resistance to Osimertinib still inevitably develops in the late period of treatment [6]. The mechanism for resistance includes driver gene mutation, histological transformation, bypass pathway activation and downstream signaling pathways activation [7].

ATP-binding cassette (ABC) family is one of the largest transporter gene families [8] which transports substrates across biological membranes. The ABC transporters bind ATP and utilize the energy to drive various molecules such as sugars, amino acids, lipids and proteins across the lipid membranes against the concentration gradient [8]. Cholesterol metabolic dysregulation is associated with tumor development, and lipid transport also affects tumor progression [9, 10]. In a variety of cancers, the ABCA subfamily affects biological processes such as proliferation, viability and migration of cancer cells by influencing cholesterol and lipid transport processes [11].

ABCA3 gene was originally cloned in 1996 from a cDNA library. The gene contains a 5112-nucleotidetrapped-long open reading frame which encodes 1704 amino acids, and its molecular mass is 191 kDa. It has homology to ABCA1, ABCA2, including the HH1 hydrophobic residues [12, 13]. Although the ABCA3 transporter has been detected in many organs such as stomach, brain, kidney and liver, it was found highly expressed in alveola type II cells (AT2) in lung [13, 14]. In alveola type II cells, the ABCA3 transporter is localized at the outer membrane of lamellar bodies (LBs) and can help transport the phosphatidylcholine as well as other phospholipid species [14, 15]. Specially, ABCA3 reduces intracellular free cholesterol and phosphatidylcholine levels while inducing the expression of SREBP-regulated genes [16]. Besides, ABCA3 interferes with the multidrug resistance pathway in some cancers and leads to poor prognosis [17–20].

Although ABCA3 takes effects on types of cancers [21, 22], the comprehensive understanding of its abnormal expression in terms of clinical relevance, oncogenesis contribution, drug resistance and biological function in LUAD progression is currently limited, necessary for further exploration.

Herein, the present research endeavors to systematically elucidate how ABCA3 affects biological function in EGFR-Mutant LUAD and its potential role in drug resistance through a synergistic application of bioinformatics, transcriptomic analysis and experimental research. In summary, targeting ABCA3 suppresses the growth and migration of EGFR-mutant lung adenocarcinoma cells by downregulating the PI3K/AKT/mTOR signaling pathway and enhances their sensitivity to EGFR-TKIs.

Materials and methods

Data source

The transcriptome of LUAD patients were downloaded from The Cancer Genome Atlas (TCGA) database (https://portal.gdc.cancer.gov/). And the RNA expression profiles of normal lung tissues were downloaded from Genotype Tissue Expression Project (GTEx) database (https://commonfund.nih.gov/GTEx).

Cell culture and transfection

Human EGFR-mutant lung adenocarcinoma cell lines NCI-H1975, HCC827, NCI-H1650, PC9 and EGFR-wild type lung carcinoma cell lines NCI-H2122, NCI-H1299 and A549 were ordered from the American Type Culture Collection (ATCC).

These cells were authenticated by short tandem repeat (STR) analysis.

The third-generation TKIs, Osimertinib (HCC827OR and H1975OR), were established, maintained, and authenticated as previously reported by our research group. Cells were treated with increasing dosages of Osimertinib, after several months, the cells developed an acquired resistance and then drug pressure was removed during the subsequent experiments.

Cells were tested for mycoplasma contamination annually using mycoBlue mycoplasma detector (D101, Vazyme) according to the manufacturer’s instructions.

Cells were propagated in DMEM or RPMI-1640 supplemented with 10% FBS, 1% alanine glutamine, 1% penicillin and streptomycin. All cells were cultured at 37 °C in an atmosphere of 5% CO2.

For transient overexpression of SP1, cells were transfected with 2.5 µg of pCMV-SP1 plasmid or an empty pCMV vector (control) using Lipofectamine™3000 (Invitrogen, USA) according to the manufacturer’s instructions. Briefly, plasmid DNA and Lipofectamine 3000 and P3000 reagent were diluted separately in Opti-MEM medium and then mixed at room temperature for 10 min before being added to cells. After 4 h of incubation, the transfection medium was replaced with fresh complete culture medium, and cells were further incubated for 24–48 h before subsequent assays.

SiRNAs and gene knockdown

Small interfering RNA targeting ABCA3 and SP1 (sense, antisense) were purchased from GenePharma (GenePharma Corporation, Shanghai, China). Transfection was carried out using the Lipofectamine and siRNA oligonucleotides (final mixed in Opti-MEM) according to the manufacturer’s instructions. The H1975, HCC827, NCI-H1650, PC9, NCI-H2122, NCI-H1299 and A549 cells were seeded in six-well plates (2 × 105 cells/well) and cultured overnight until 40% confluent. Knockdown of all cells was performed after transfecting. The sense strand sequences of the siRNAs used are as follows:

siABCA3 #1: 5’-GCCCACUACUGCAAGAAAUTT-3’,

siABCA3 #2: 5’-CGGACACUGACAGAAUUAUTT-3’,

siSP1 #1: 5’-CCUCACAGCCACACAACUUTT-3’,

siSP1 #2: 5’- GCAGACCUUUACAACUCAATT-3’.

Cell growth curve and cell viability assays

For cell growth curve analysis, cells were seeded in 96-well plates with three replicates per group, the density of which was controlled at approximately 4000 cells per well. Cell growth was monitored using the IncuCyte ZOOM live cell analysis system. For cell viability assay, the medium was replaced with fresh culture medium containing 10% Cell Counting Kit-8 (CCK8) reagent (APExBIO) according to the manufacturer’s instructions.

RNA extraction and quantitative polymerase chain reaction

RNA was extracted using an RNA extraction kit (Takara) according to the manufacturer’s instructions. The concentrations were measured with NanoDrop (Thermo Fisher Scientific), and reverse transcription was carried out using the RevertAid First Strand cDNA Synthesis Kit (Qiagen). Real-time PCR was run in a LightCycler 480 II system (Roche) using default reaction settings.

Colony formation assay

Cells were seeded in six-well plates, exposed to treatments, and cultured in complete medium for about 2 weeks. Cells were then fixed with 4% paraformaldehyde for 15 min and stained with 0.01% crystal violet. Colonies of > 50 cells were counted using the ECLIPSE TI Live-Cell Imaging System.

Cell cycle and apoptosis assays

Cell cycle phases and apoptosis were determined by flow cytometry analyses based on propidium iodide (PI) staining of cellular DNA content and annexin V–fluorescein isothiocyanate (FITC)/PI double staining of cell death, respectively, according to the reagent kit manufacturer’s instructions (MultiSciences, CCS012 and MultiSciences, AT101). All samples were analyzed on a Flow Cytometer (Thermo Scientific, Attune NxT). FlowJo-V10 software was used to quantify populations.

RNA sequencing

Total mRNA samples were extracted with an RNA extraction kit (Vazyme), then sequenced on an DNBSEQ platform. Differentially expressed gene (DEG)-normalized read counts [fragments per kilobase of exon per million (FPKM)] were calculating using RSEM (v.1.3.1). DEG analysis was performed using the DESeq2(v1.34.0). Kyoto Encyclopedia of Genes and Genomes (KEGG), and gene set enrichment (GSEA) were annotated by KEGG pathway database (https://www.genome.jp/kegg/) and GSEA database (https://www.gsea-msigdb.org/gsea/index.jsp) respectively.

Western blot for pathway detection

Cells and tumor samples were lysed by radioimmunoprecipitation assay (RIPA) buffer with 1 mM phenylmethylsulfonyl fluoride (PMSF) protease inhibitors to extract the total cell protein. Protein concentrations were determined by bicinchoninic acid (BCA) Protein Assay Kit. Proteins were subjected to SDS–polyacrylamide gel electrophoresis (SDS-PAGE) gel separation and transferred to polyvinylidene difluoride (PVDF) membranes. Membranes were blocked with 5% nonfat milk in 1× tris-buffered saline–Tween 20 for one hour at room temperature and blotted with diluted primary antibodies at 4 °C with gentle shaking overnight. After incubation with horseradish peroxidase–conjugated anti-rabbit immunoglobulin G (IgG) or anti-mouse IgG antibodies, the immunoblots were subjected to electrochemiluminescence and scanned using an Odyssey FC imaging system.

Crisper-Cas9

LentiCRISPRv2 vector from Addgene (#52961) was digested with BsmBI and linked with annealed oligonucleotide (sgABCA3: 5’-TCTGCGATGAACGGCGGGTA-3’ (sense)). Then human lung cancer cell line HCC827 was transfected using Polyethyleneimine with LentiCRISPRv2 and packaging vectors pMD2G and psPAX2. After 48 h transfection, viral supernatants were collected and stored at −80 °C. HCC827 cells were inoculated into the 6-well plate with 5 × 10⁴cells/well and transducted with the viral supernatant containing 8 µg/mL Polybrene, then 50 µg/ml puromycin was used to generate the stable cell lines.

In vivo mouse xenograft study and ethics approval

The 5-week-old female BALB/c nude mice were purchased and raised in a pathogen-free environment. In total, HCC827 cells (1 × 10⁷) stably transfected with sgABCA3, or packaging vector were subcutaneously inoculated into the right flanks of the BALB/c nude mice (n = 4 mice in each group) to establish a lung adenocarcinoma model to investigate the effect of ABCA3 on cell proliferation in vivo. Every 3 days, mice were weighed, tumor sizes were assessed with a digital caliper. Twenty-five days later, the mice were killed, tumors were weighed immediately. All procedures and experiments involving animal studies were evaluated and approved by the Institutional Animal Care and Use Committee (IACUC) and carried out in accordance with the Animal Care and Use Rules of Shanghai Jiao Tong University School of Medicine (JUMC2023-132). The tumor volume was calculated according to the formula: V = (a × b²)/2, (a and b are the maximal and minimal diameters in millimeters respectively).

Results

1. ABCA3 is highly expressed in EGFR-mutant NSCLC, and its knockdown reduces the viability and proliferation of EGFR-mutant cells

Given that ATP-binding cassette (ABC) transporters influence key biological processes such as cancer cell proliferation, survival, and migration through their roles in drug efflux and metabolic reprogramming [8], we investigated the relationship between the ABC transporter family and tumorigenesis and progression. In particular, we focused on the role of ABCA3 in lung adenocarcinoma (LUAD), as this protein is closely associated with critical pathways of cholesterol metabolism and is rarely studied in the tumorigenesis and progression of lung cancer [11]. Lung adenocarcinoma is characterized by diverse driver gene alterations, each associated with distinct mutational profiles. Keys among these are KRAS, EGFR, BRAF, NTRK3, ALK, PI3KCA, ROS1, MET, and ERBB2, which are crucial for pathogenesis and serve as targets for therapy. Therefore, we investigated the correlation between ABCA3 expression and the mutational status of these genes.

Notably, TCGA database analysis revealed that ABCA3 expression was significantly elevated in LUAD samples harboring EGFR or KRAS mutations compared with their wild-type counterparts. (Fig. 1A). Conversely, ABCA3 expression decreased in samples with NTRK3, PI3KCA, or ROS1 mutations. No significant difference in ABCA3 expression was observed in BRAF, ALK, MET, or ERBB2 mutated samples (Fig. 1A). This strong correlation between elevated ABCA3 expression and major oncogenic drivers like EGFR and KRAS in LUAD suggests a potential role for ABCA3 in tumor progression, specifically in the context of these two genetic alterations.

Fig. 1.

ABCA3 is highly expressed in EGFR-mutant cell lines, and its knockdown reduces the viability and proliferation of EGFR-mutant cells. (A) Correlation between ABCA3 expression and genetic abnormalities of driver genes in LUAD. ABCA3 expression level in wild type and mutated group of KRAS, EGFR, BRAF, NTRK3, ALK, PI3KCA, ROS1, MET and ERBB2. (B)Results of cell counting kit-8 assay of EGFR mutant cells (H1975, H1650 and HCC827) and EGFR WT cells (H1299, A549 and H2122) with ABCA3 knockdown by siABCA3 #1 and #2, indicating cell viability. (C) and (D) The IncuCyte method was used to plot the growth curves of the EGFR-mutant cells and EGFR WT cells. All quantitative data are derived from three independent biological experiments. Data are presented as mean ± SD. Statistical significance was determined using Students’ t test. *P < 0.05, **P < 0.01, ***P < 0.001; NS, not significant

To test this assumption, we utilized EGFR-mutant cell lines (H1975, H1650, and HCC827), KRAS-mutant cell lines (A549 and H2122), and driver gene-negative cells (H1299). We first assessed the protein expression level of ABCA3 (Fig. S1A), the Western blot analysis showed that the expression of ABCA3 was higher in the EGFR-mutant cell lines (HCC827, H1975, PC9) compared to the EGFR wild-type (WT) cell lines (A549, H1299, H2122). ABCA3 was then knocked down in these cell lines using two independent siRNAs (siABCA3 #1 and #2). We assessed cell growth and viability using a Cell Counting Kit-8 (CCK-8) assay. In EGFR-mutant cells, ABCA3 knockdown significantly inhibited cell viability at 48 h compared to control groups. In contrast, extremely slight decrease in cell viability was observed in EGFR WT cell lines at the 48-hour time point (Fig. 1B). These findings consistently demonstrate that ABCA3 knockdown inhibits the growth of EGFR-mutant cells (Fig. 1C), while EGFR WT cells remain unaffected (Fig. 1D).

• 2.ABCA3 knockdown suppresses cell migration, proliferation and promotes apoptosis

To further investigate the biological characteristics after knocking down ABCA3 in EGFR-mutant cells, we tested cell migration, proliferation and apoptosis. As shown by the wound healing assay, the migration index of EGFR-mutant cells (H1975 and HCC827) markedly decreased at 24 h following siABCA3 transfection, whereas EGFR WT cell (A549) showed no significant change compared with the control group (Fig. 2A and B). Colony formation assays revealed that cell proliferation was almost completely suppressed in H1975 and HCC827 cells upon ABCA3 knockdown (Fig. 2C and D). Moreover, flow cytometric analysis demonstrated that the proportion of apoptotic HCC827 and H1975 cells increased by approximately 20% relative to controls (Fig. 2E and F). Collectively, these findings indicate that targeting ABCA3 exerts potent anti-tumor effects in EGFR-mutant LUAD cells.

Fig. 2.

ABCA3 knockdown inhibits cell migration and proliferation, and promotes apoptosis. (A) and (B) Migration ability of EGFR-mutant cells (H1975 and HCC827) and EGFR WT cells (A549) in wound healing assay. (C) and (D) H1975 and HCC827 cell viability with ABCA3 knockdown by siRNA #1 and #2 showed in clone formation assay. (E) and (F) Apoptosis result of EGFR-mutant cells (H1975 and HCC827) showed by flow cytometric analysis. All quantitative data are derived from three independent biological experiments. Data are presented as mean ± SD. Statistical significance was determined using Students’ t test. *P < 0.05, **P < 0.01, ***P < 0.001; NS, not significant

• 3. SP1-regulated ABCA3 drives EGFR-mutant tumor cell growth by modulating the PI3K/AKT/mTOR signaling pathway

To unravel how ABCA3 mediates tumor progression specifically in EGFR-mutant LUAD, we performed RNA-seq on ABCA3 knockdown group and control group in H1975 (EGFR-mutant) and A549 (EGFR-WT) cell lines. Our goal was to pinpoint signaling pathways significantly altered by ABCA3 knockdown in H1975 cells, while showing minimal or no changes in A549 cells. This comparative approach aimed to reveal ABCA3-mediated pathways that EGFR-mutant LUAD cells might depend more heavily on, offering insights into mutation-specific vulnerabilities linked to ABCA3 function.

Gene Set Enrichment Analysis (GSEA) revealed thirteen signaling pathways that were significantly affected by ABCA3 knockdown specially in H1975 cells (Fig. 3A). Among the upregulated pathways, we observed activation of the G2M checkpoint, E2F targets, and DNA repair pathways. Conversely, several pathways were markedly suppressed, including the critical PI3K/AKT/mTOR signaling and cholesterol homeostasis (Fig. 3B). We also detected a context-dependent alteration in KRAS signaling, which was downregulated in H1975 cells but upregulated in A549 cells, highlighting the influence of genetic background on cellular responses.

Fig. 3.

SP1-regulated ABCA3 drives EGFR-mutant tumor cells growth by modulating the PI3K/AKT/mTOR signaling pathway. (A) Venn plot of GSEA activated (NES > 1) and suppressed pathway (NES < 1) in ABCA3-Knockdown A549 and H1975 groups. NES, normalized enrichment score. (B) GSEA Hallmark pathways specifically regulated in H1975(EGFR-Mutant) compared with A549 (EGFR–wild type) cells following ABCA3 knockdown. (C) PI3K/AKT/mTOR pathway is identified by GSEA analysis in H1975 ABCA3 knockdown vs. control group. P values are calculated by permutation testing. (D) KEGG pathway enrichment analysis indicated that PI3K-Akt pathway served as one of the major enriched signaling in ABCA3 KD group in H1975. (E) Relative mRNA levels of the indicated genes of PI3K/AKT/mTOR pathway in A549 and H1975 cells ABCA3 knockdown vs. control group. (F) Western blot analysis of expression of proteins related to PI3K/AKT/mTOR pathway in H1975 and HCC827 cells treated with siABCA3 or negative control (siNC) for 24 h. (G) Venn diagrams of four gene lists: transcription factors predicted by the hTFtarget, ENCODE, FIMO_JASPAR and KnockTF databases. (H) Spearman correlation analysis between ABCA3 and SP1 or MYC in the TCGA datasets. Spearman correlation coefficients (ρ) and corresponding P values are shown; P < 0.05 indicates statistical significance. (I and J) SP1 was over-expressed in HCC827 and H1975 cell lines and RT-qPCR and Western blot were used to test the mRNA and protein level of ABCA3. SP1 was knocked down in HCC827 and H1975 and Western blot demonstrated the mRNA and protein level of ABCA3. All quantitative data are derived from three independent biological experiments. Data are presented as mean ± SD. Statistical significance was determined using the methods described in the Materials and Methods section. P < 0.05, *P < 0.01, **P < 0.001; NS, not significant

Focusing on pathways associated with tumorigenesis and progression, we found that ABCA3 knockdown in H1975 cells enhanced DNA repair while inhibiting cholesterol homeostasis, angiogenesis, and the PI3K/AKT/mTOR pathway (Fig. 3B). Consistently, GSEA confirmed a significant downregulation of the PI3K/AKT/mTOR pathway following ABCA3 depletion (Fig. 3C). Moreover, KEGG pathway enrichment analysis of differentially expressed genes further validated the prominent involvement of the PI3K/AKT signaling pathway in ABCA3-knockdown H1975 cells (Fig. 3D).

Consistent with these observations, ABCA3 depletion led to reduced expression of key genes involved in the PI3K/AKT/mTOR signaling pathway in H1975 cells, while no significant changes were detected in A549 cells (Fig. 3E). To further validate the sequencing results, western blot analysis demonstrated that ABCA3 knockdown markedly reduced the phosphorylation levels of AKT and mTOR in both H1975 and HCC827 cells (Fig. 3F; Fig S2A). Collectively, these results strongly suggest that ABCA3 knockdown specifically suppresses the PI3K/AKT/mTOR pathway in EGFR-mutant LUAD cells.

To further investigate the mechanism behind ABCA3’s role in tumor progression, we predicted upstream transcription factors (TFs) by integrating data from hTFtarget, ENCODE, FIMO_JASPAR, and KnockTF databases. This analysis identified SP1 and MYC as potential upstream transcription factors for ABCA3 (Fig. 3G). Upon analyzing the TCGA database, we found a positive correlation between ABCA3 and SP1 expressions across various cancer types, whereas the correlation between ABCA3 and MYC was considerably weaker (Fig. 3H). We next verified the regulatory effect of SP1 on ABCA3 expression using RT-qPCR and western blot analysis in HCC827 and H1975. RT-qPCR analysis revealed that SP1 overexpression increased the mRNA level of ABCA3, whereas SP1 knockdown led to its reduction (Fig. 3I). Consistently, western blot analysis confirmed that ABCA3 protein expression was regulated by SP1 (Fig. 3J; FigS2B and C).

In summary, our findings indicate that ABCA3 promotes tumor progression in EGFR-mutant LUAD by activating the PI3K/AKT/mTOR pathway, and this process is transcriptionally regulated by SP1.

• 4.ABCA3 knockdown suppresses tumor growth in vivo

To further evaluate the in vivo effects of ABCA3, we established a murine LUAD patient graft xenograft (PDX) model in BALB/c nude mice. Mice were subcutaneously inoculated with either sgABCA3-HCC827 cells or sg-control HCC827 cells (Fig. 4 A). Over a 25-day period, tumor volume and body weight were meticulously monitored. As demonstrated in Fig. 4 B-D, tumors from the sgABCA3-treated group exhibited a significant inhibition of both tumor volume and tumor weight compared to the control group. Importantly, no significant difference in body weight was observed between the two groups of mice throughout the study period (Fig. 4 E), indicating minimal systemic toxicity. To further validate the mechanism, western blot analysis confirmed reduced expression of ABCA3 and decreased phosphorylation levels of AKT and mTOR in tumor tissues from the sgABCA3 group (Fig. 4 F). These in vivo findings are consistent with our in vitro results (Figs. 2 and 3), collectively demonstrating that ABCA3 knockdown suppresses the growth of EGFR-mutant LUAD by inhibiting the PI3K/AKT/mTOR signaling pathway.

Fig. 4.

ABCA3 knockdown suppresses tumor growth in vivo. (A) Schematic illustration of the in vivo experimental design. BALB/c nude mice (n = 4) were subcutaneously inoculated with HCC827 sgABCA3 or sgNC cells. (B) Photos of the formed tumors at the end of the experiment. (C) Tumor growth curve during the experiment. (D)Tumor weight at the end of the experiment. (E) Body weight curve during the experiment. (F) Western blot analysis of expression of proteins related to PI3K/AKT/mTOR pathway in HCC827-PDX cells in vivo. Bars represent mean ± SD of replicate. Statistical significance was determined using two-way Anova. *P < 0.05, **P < 0.01, ***P < 0.001; NS, not significant

• 5. ABCA3 knockdown enhances osimertinib sensitivity in EGFR-mutant NSCLC cell lines

Finally, to define ABCA3’s precise role in Osimertinib resistance within EGFR-mutant NSCLC, we aimed to investigate whether targeting its regulation could not only partially reverse this resistance but also uncover a crucial vulnerability for therapeutic intervention. This work is critical for expanding treatment options and improving outcomes for patients who develop resistance to standard therapies.

Studies indicate that ABCA3 contributes to multidrug resistance in cancer by effusing chemotherapeutic drugs [23]. Furthermore, it’s verified the activation of the PI3K/AKT/mTOR signaling pathway fosters resistance to antineoplastic drugs [24]. Accordingly, we assumed ABCA3 might mediate resistance to targeted therapies. To test our hypothesis, we established osimertinib-resistant NSCLC cell lines to explore whether the regulation of ABCA3 can partly or completely reverse that resistance [25].

First, our previous RNA-seq analysis [25] of the ABCA family revealed a significant upregulation of ABCA3 mRNA expression in osimertinib-resistant (OR) cells (Fig. 5 A). We validated this finding using RT-qPCR, which confirmed that ABCA3 was significantly overexpressed in H1975OR and HCC827OR cells compared to their parental H1975 and HCC827 counterparts (Fig. 5 B, C, and D).

Fig. 5.

ABCA3 knockdown enhances osimertinib sensitivity in EGFR-mutant NSCLC cell lines. (A) RNA-seq analysis of mRNA expression of ABCA family in parental and OR cell lines were shown in violin charts. n = 5 per group. (B) and (C) Results of ABCA mRNA expression in violin charts and heatmap through RT-qPCR analysis. n = 3 per group. (D) ABCA3 expression in parental and OR cell lines through RT-qPCR analysis. (E) Western blot results of SP1 and ABCA3 expression in parental and OR cell lines. (F) Cell viability was assessed to validate resistance to Osimertinib in H1975OR and HCC827OR cells and demonstrate the cell viability variation of H1975, H1975OR, HCC827OR under treatment with Osimertinib and ABCA3 knockdown. A non-linear regression model is used to draw sigmoidal dose-response curves. n = 3 per group. (G) H1975, H1975OR, HCC827 and HCC827OR cells were treated with osimertinib (800 nM) with and without ABCA3 knockdown. Cells viability was assessed using a CCK-8 assay. (H) Western blot analysis of expression of proteins related to PI3K/AKT/mTOR pathway in HCC827OR and H1975OR with or without ABCA3 knockdown for 24 h. (I)Western blot analysis of ABCA3 and PI3K/AKT/mTOR pathway related protein after ABCA3 knockdown and/or Osimertinib treatment for 24 h. (J) Schematic representation of the proposed mechanisms underlying the effect of ABCA3 to PI3K/AKT/mTOR pathway in EGFR-mutant NSCLC cells. All quantitative data are derived from three independent biological experiments. Bars represent mean ± SD of replicate. Statistical significance was determined using Students’ t test. *P < 0.05, **P < 0.01, ***P < 0.001; NS, not significant

To determine ABCA3’s role in drug resistance, we treated H1975, H1975OR, HCC827, and HCC827OR cells with graded concentrations of Osimertinib with or without ABCA3 transfection, followed by assessment of cell viability. We subsequently performed Western blot analysis, and the results demonstrated elevated expression of both SP1 and ABCA3 in the Osimertinib-Resistant (HCC827-OR and H1975-OR) cell lines compared to their corresponding parental sensitive controls (Fig. 5 E; Fig S2D). This finding confirmed the elevated expression of the SP1-ABCA3 axis at the protein level in resistant cells, prompting us to further investigate whether targeting ABCA3 could reverse TKI resistance in OR cells. Furthermore, the half-maximal inhibitory concentration (IC₅₀) values for Osimertinib were notably lower in both H1975OR and HCC827OR cells after ABCA3 knockdown. Meanwhile, ABCA3 knockdown also enhanced Osimertinib sensitivity in the parental cell lines (Fig. 5 F). Western blotting results further demonstrated that ABCA3 knockdown inhibited the p-AKT and p-mTOR in HCC827OR and H1975OR cells (Fig. 5 H; Fig S2E). The combined therapeutic approach of ABCA3 knockdown and Osimertinib treatment further intensified the inhibition of p-mTOR, pointing to a promising clinical strategy (Fig. 5 I; Fig S2F and G). It has been reported that PI3K/AKT/mTOR pathway activation contributed to the development of drug resistance against osimertinib, thereby bypassing the inhibitory effects of the drug [26, 27]. Collectively, data suggested that ABCA3 knockdown can partially restore Osimertinib sensitivity in EGFR-mutant NSCLC through inhibition of the PI3K/AKT/mTOR pathway, a mechanism consistent with our observations in EGFR-mutant cell lines.

Together, our findings reveal that ABCA3 is overexpressed in osimertinib-resistant EGFR-mutant NSCLC cells. Notably, ABCA3 knockdown partially restored sensitivity to osimertinib, primarily through inhibition of the PI3K/AKT/mTOR signaling pathway. We therefore propose a mechanistic model in which EGFR-TKI pressure induces SP1-mediated transcriptional activation of ABCA3, leading to altered cholesterol metabolism and subsequent activation of the PI3K/AKT/mTOR pathway, ultimately promoting cell proliferation and drug resistance (Fig. 5 J).

Discussion

In this study, we conducted bioinformatics, transcriptomic analysis and experimental research to gain a more comprehensive understanding of the potential functions and regulatory mechanisms of ABCA3 in lung adenocarcinoma (LUAD). Initially, by bioinformatic analysis, we discovered that ABCA3 mRNA levels modulate in a range of malignant tumors. Higher ABCA3 expression level was found in the mutated group of major driver genes including EGFR and KRAS in LUAD. Mutation in the EGFR is a predominant oncogenic driver in LUAD [28]. Interestingly, our finding contrasted with a previous report suggesting that ABCA3 was a tumor-suppressive factor in the development of LUAD [29]. This irreconciliation of these findings may likely lie in different genetic backgrounds. In our study, we mainly discussed EGFR-mutant and wild type cell lines. This suggested the importance of gene function within specific molecular subtypes for tumor.

To further investigate the functional implications and heterogeneity of ABCA3 in LUAD, we executed cellular phenotype assays in EGFR-mutant and wild type cell lines. We demonstrated that the depletion of ABCA3 in EGFR-mutant cells led to diminished cell viability, proliferation, and migration, and an increase in apoptotic rates. This outcome is likely due to the deprivation of ABCA3, leading to deregulation of cholesterol and lipid homeostasis. However, similar biological effects were not observed in wild type LUAD cells. In vivo studies further confirmed that ABCA3 knockdown curtailed cell proliferation of EGFR-mutant cells. Furthermore, our study indicates that blocking ABCA3 increases the sensitivity of EGFR-mutant NSCLC cells to EGFR TKIs. Current targeted therapeutic approaches for LUAD patients with EGFR mutation like Osimertinib, which targets the EGFR T790M mutation, showing promising results in clinical settings [30]. A growing amount of evidence shows that the ABCA3 gene is associated with cancer, including breast cancer, leukemia and lung cancer. The changes in expression level of the ABCA3 will affect prognosis of patients with breast cancer [21, 31]. The fact that ABCA3 can cause cellular drug resistance in leukemia from chronic to acute has also been widely described and is thought to be related with the overexpression of the ABCA3 transporters that function as drug efflux pumps [22, 32–34]. In lung cancer, several studies have described that ABCA3 protects cancer cells from drug toxicity in ways of material transporting and drug extrusion [35–38].

To explore the mechanism how ABCA3 knockdown only inhibited EGFR-mutant cells viability, we performed transcriptome sequencing on both mutant and wild type cells with siABCA3 intervention. Through GSEA analysis we uncovered that ABCA3 was associated with various functions including angiogenesis, DNA repair, cholesterol homeostasis and PI3K/AKT/mTOR signaling pathway, which is dependent to EGFR-mutant cells. PI3K/AKT/mTOR pathway is one of the intracellular signaling cascades initiated by the dimerization of EGFR [39]. Activating this pathway fosters both tumor survival and resistance to anticancer drugs [24, 40]. The pathway’s influence extends to the regulation of various cellular physiological processes, with its activation of downstream effectors being essential for tumor proliferation, invasion, and metastasis [40]. In light of this, the strategic targeting of the PI3K/AKT/mTOR pathway to modulate tumor metabolism has become a recognized therapeutic strategy in cancer. In our study, PI3K/AKT/mTOR signaling pathway is verified to be suppressed after silencing ABCA3 [41].

Our research findings endorse the hypothesis that ABCA3 contributes to the neoplastic process in EGFR-mutant NSCLC by triggering PI3K/AKT/mTOR pathway. By inhibiting ABCA3, the PI3K/AKT/mTOR pathway was suppressed, and both in vitro and in vivo experiments had shown that the cell’s growth and survival capabilities, as well as tumor growth, were inhibited, and the responsiveness to EGFR TKI is enhanced, indicating that ABCA3 could potentially become a therapeutic target in EGFR mutation and EGFR-TKIs resistance.

According to the results of RNA-seq, we believe that cholesterol homeostasis is a potential mechanism by which ABCA3 affects the PI3K/AKT/mTOR pathway. ABCA3 is also involved in LDL processing and intracellular cholesterol trafficking [11]. However, we didn’t further explore Relationship between ABCA3 and cholesterol metabolism reprogramming. Overall, our investigations have delineated ABCA3 as a potential therapeutic target and predictive biomarker of EGFR-TKIs sensitivity. These results contribute the field of precision oncology, enabling more accurate and targeted therapeutic interventions.

Conclusion

Our research has demonstrated that knocking down ABCA3 inhibits growth and migration in EGFR-mutant LUAD cells, via downregulating PI3K/AKT/mTOR signaling axis, and increases the drug sensitivity to EGFR-TKIs. ABCA3 plays a vital role in cancer progression and may be a potential therapeutic and drug resistance target for patients with EGFR-mutant LUAD.

Statistical analysis

For statistical analysis, we used GraphPad Prism Software 10 (GraphPad Software, Inc.) and SPSS 22.0 (IBM Corp., Armonk, NY, USA) and the measurement data were shown as mean ± standard deviation in the bar charts. Independent samples t-test was used to compared two-group comparison and one-way ANOVA was used for different-group multiple comparisons. p-values of < 0.05 were considered statistically significant (NS: not significant, *p < 0.05, **p < 0.01, ***p < 0.001).

Author contributions

(I) Conception and design: YQL and CZ (II) Administrative support: YQL and CZ and LZ (III) Provision of study materials or patients: YQL and CZ (IV) Collection and assembly of data: ZXC, GTL, JYY, and ZMC (V) Data analysis and interpretation: ZXC, GTL, JYY, and ZMC (VI) Manuscript writing: All authors (VII) Final approval of manuscript: All authors.

Funding

This work was supported by grants from the National Natural Science Foundation of China (No.82273950) and Collaborative Innovation Center for Clinical and Translational Science by Chinese Ministry of Education & Shanghai.

Data availability

The RNA-seq data are available in the NCBI’s Gene Expression Omnibus (GEO) database under accession number PRJNA1301097.

Declarations

Ethics approval and consent to participate

declaration: Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.