Endoplasmic reticulum stress-related genes contribute to lung cancer risk: a multiomics data integration study

Introduction

Lung cancer remains a significant public health concern, ranking as the second most prevalent cancer in terms of both incidence and mortality in the United States and as the leading cause of cancer-related mortality in China (1,2). Genetic factors significantly contribute to the development of lung cancer, with heritability estimated at approximately 18% to 24% (3,4). Genome-wide association studies (GWAS) that investigate single-nucleotide polymorphisms (SNPs) in relation to disease provide significant opportunities to elucidate the role of genetic variations in lung cancer (5,6), with recent investigations identifying several susceptibility loci associated with lung cancer risk. In addition to loci identified through GWAS, genetic variants in detoxification-related genes such as glutathione S-transferases (GSTs) and cytochrome P450 1A1 (CYP1A1) have also been linked to increased lung cancer susceptibility, particularly among smokers, and have shown population-specific associations in multiple cohorts (7,8). Additionally, specific gene mutations have been successfully translated into therapeutic targets, showing promising results in clinical trials (9,10). However, most mutations are noncoding and exert their effects by modulating gene expression (11,12), although the biological function of the vast number of mutations remains unclear (13). Quantitative trait loci (QTL) serve as essential tools for investigating the regulatory mechanisms underlying genetic variations. In particular, cis-acting QTL (cis-QTL), which are located close to genes (within <1 megabase), regulate the expression levels of adjacent genes (14). Numerous studies have identified several QTL potentially associated with gene expression (14-16). Therefore, integrating GWASs and QTL data can further elucidate the biological mechanisms underlying genetic variations in lung cancer development and progression, thus providing new insights for diagnosis and therapeutic strategies.

The endoplasmic reticulum (ER) is a crucial organelle in eukaryotic cells, involved in numerous vital biological processes, including protein targeting, secretion, and vesicle trafficking (17). When proteins within the ER become unfolded, misfolded, or excessively accumulated, cells activate adaptive pathways to restore ER dysfunction—a phenomenon known as endoplasmic reticulum stress (ERS) (18-20). This protective mechanism enhances the ability of the ER to manage protein folding and restore cellular homeostasis. However, prolonged or severe ERS may lead to apoptosis (21). ERS plays a critical role in various diseases, including diabetes, neurodegenerative disorders, and cancer (22). ERS has been shown to promote tumor progression and drug resistance across multiple cancer types, including breast, colorectal, glioma, and hepatocellular carcinomas, primarily by enabling malignant cells to adapt to adverse microenvironmental stresses such as hypoxia and oxidative stress (22-24). Conversely, in certain contexts, excessive ERS can trigger immunogenic cell death and enhance antitumor immunity—for instance, via ERO1A-related pathways in lung cancer (25,26). These seemingly contradictory findings may reflect the context-dependent roles of ERS and the diverse regulatory mechanisms involved across different cancer types. Despite its clinical relevance, the role of ERS in lung cancer has been insufficiently evaluated, and its precise function remains unclear, warranting further investigation (25,27).

We thus conducted a study using large-scale GWAS and multiomics cis-QTL data, combined with summary data-based Mendelian randomization (SMR) and colocalization analysis, to clarify the causal link between ERS-related gene expression and lung cancer risk, along with the underlying regulatory mechanisms. Large-sample-size data and SMR analysis can help to eliminate confounding factors (28), and were thus used to clarify the role of ERS genes in lung cancer development and provide new insights into diagnosis and treatment. We present this article in accordance with the STREGA reporting checklist (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-474/rc).

Methods

Study design

This study leveraged publicly available cis-QTL and GWAS datasets to clarify the causal association between ERS-related genes and lung cancer. The discovery cohort was based on GCST004748, and SMR analysis was employed to investigate the causal relationships between ERS-related genes and lung cancer across three dimensions: DNA methylation, gene expression, and protein expression. The study results were validated using the C3 BRONCHUS LUNG EXALLC cohort. Furthermore, colocalization analysis was conducted to validate the key results obtained from the SMR analysis, thereby enhancing the reliability of the conclusions. By integrating findings from multiple omics layers, the study ultimately identified two ERS-related genes potentially associated with lung cancer risk. Finally, the expression levels and prognostic value of these two genes in lung cancer tissues were comprehensively analyzed. Figure 1 illustrates the overall study design and analytical workflow.

Figure 1 Study design and analytical workflow. **, P<0.01; ***, P<0.001. eQTL, expression quantitative trait loci; GWAS, genome-wide association study; HEIDI, heterogeneity in dependent instruments; LUAD, lung adenocarcinoma; LUSC, lung squamous cell carcinoma; mQTL, methylation quantitative trait loci; NSCLC, non-small cell lung cancer; pQTL, protein quantitative trait loci; SMR, summary data-based Mendelian randomization; SNP, single-nucleotide polymorphism; TPM, transcripts per million.

Data sources of GWAS and QTL

GWAS data

The discovery cohort, GCST004748 (6), was downloaded from the GWAS Catalog (https://www.ebi.ac.uk/gwas/) and includes 29,266 cases, 56,450 controls, and a total of 7,884,149 single-nucleotide polymorphisms (SNPs). Participants were of European ancestry, with 88% aged over 50 years, 62% of cases being male, and 91% having a history of smoking. Histologic subtypes included adenocarcinoma (39%), squamous cell carcinoma (25%), and small cell carcinoma (9%). The validation cohort, C3 BRONCHUS LUNG EXALLC (version R10), was obtained from the FinnGen database (29) and comprises 6,340 cases, 314,193 controls, and a total of 20,191,463 SNPs. The median age at diagnosis was 71.62 years, and all participants were of European ancestry, with diagnoses based on the international classification of diseases codes.

Cis-QTL data

The summary data for blood expression QTL (eQTL) were obtained from the eQTLGen Consortium, which includes genetic data on blood gene expression from 31,684 individuals (14). The summary data for blood methylation QTL (mQTL) were derived from a meta-analysis of two European cohorts: the Brisbane Systems Genetics Study (n=614) and the Lothian Birth Cohorts (n=1,366) (16). The summary data for blood protein QTL (pQTL) were obtained from the work of Pietzner et al. (15), which included 10,708 European individuals.

ERS-related genes

ERS-related genes were retrieved from the Gene Cards database (https://www.genecards.org) using the keyword “Endoplasmic Reticulum Stress”. The search was limited to the “Protein Coding” category, resulting in the identification of 2,350 protein-coding ERS-related genes.

SMR analysis

SMR software (version 1.3.1) was employed for SMR, multi-SMR, and heterogeneity in dependent instruments (HEIDI) tests to evaluate the relationships between methylation, gene expression, and protein abundance of ERS-related genes and lung cancer (16,28). The reference sample for the SMR analysis was sourced from the European population data in the Complex Trait Genetics Lab database (30). The analysis employed a window centered around the corresponding gene ±1,000 kb and a P value threshold of 5.0×10⁻⁸ to select cis-QTL, excluding any SNPs with allele frequency differences greater than the predefined threshold (set at 0.2) between any two datasets (including the reference sample, cis-QTL summary data, and outcome summary data). For mQTL, eQTL, and pQTL, the maximum allowable proportion of SNPs exhibiting allele frequency differences exceeding 0.2 was set to 0.05. Only results meeting the criteria of P_SMR <0.05, P_SMR-multi <0.05, and P_HEIDI >0.05 were included in subsequent colocalization and integrative analyses of eQTL, mQTL, and pQTL.

Colocalization analysis

Colocalization analysis was performed using the “coloc” package in R (The R Foundation for Statistical Computing) to identify common causal variants between cis-QTL (mQTL, eQTL, and pQTL) associated with ERS genes and lung cancer. The colocalization region windows for mQTL-GWAS (31), eQTL-GWAS (32), and pQTL-GWAS (33) analyses were set at ±500, ±1,000, and ±1,000 kb, respectively. P12 was calculated by conducting a sensitivity analysis of the posterior probability of colocalization (PP.H4) with respect to the prior probability of colocalization. Colocalization was considered successful if the cis-QTL satisfied the following criteria: (I) P12 =5×10⁻⁵ with PP.H4 >0.5; or (II) P12 =1×10⁻⁵ with PP.H3 <0.5 (34). The colocalization results for key genes are shown in Figure S1.

Multiomics results integration

To further investigate the regulatory relationships among methylation sites, gene expression, and protein abundance, SMR analyses were performed using mQTL as exposures and eQTL as outcomes. Similarly, eQTL served as the exposure and pQTL as the outcome in subsequent SMR analyses. The intersection of SMR results from cis-QTL-GWAS, colocalization results, and those obtained from this step was selected. Only outcomes meeting the criteria of P_SMR <0.05, P_SMR-multi <0.05, and P_HEIDI >0.05 were selected.

Expression and prognosis analysis

Data from the Gene Expression Profiling Interactive Analysis (GEPIA) (35) database were used to analyze differences in KLC1 and BCL2L1 gene expression between lung cancer samples and normal tissues. Based on the median expression levels of KLC1 and BCL2L1, patients with lung cancer in The Cancer Genome Atlas (TCGA; https://portal.gdc.cancer.gov/) database were categorized into high- and low-expression groups. The CIBERSORT algorithm was employed to calculate immune cell infiltration scores, enabling the evaluation of associations between gene expression levels and the relative abundance of different immune cell types in the tumor samples. Data from the Kaplan-Meier Plotter website (36) were used to assess the relationship between the expression levels of KLC1 and BCL2L1 and overall survival (OS) in lung cancer patients [univariate Cox regression was performed to estimate the hazard ratio (HR)].

Immunohistochemical (IHC) assays and analysis

Lung cancer tissues and matched normal tissues (n=18) were collected from the First Affiliated Hospital of Kunming Medical University. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Human tissue samples were obtained with informed consent and approval from the Ethics Committee of First Affiliated Hospital of Kunming Medical University (No. 2020-L42). The tissue sections were subjected to microwave heating in 10 mM of sodium citrate solution and subsequently treated with 3% H₂O₂ for 10 minutes to inactivate endogenous peroxidases. Following this, the sections were blocked with goat serum and incubated overnight at 4 ℃ with primary antibodies, followed by incubation with a secondary antibody for 1 hour at 37 ℃. Signals were visualized using a diaminobenzidine chromogen, and images were captured using a light microscope. The antibodies employed in this IHC analysis were BCL2L1 (bs-1336R; Bioss Antibodies, Woburn, MA, USA) and KLC1 (bs-11212R; Bioss Antibodies). Cell staining intensity and the percentage of positive cells were quantified using ImageJ software version 1.53t (US National Institutes of Health, Bethesda, MD, USA). The final IHC score was determined by multiplying the cell staining intensity score by the percentage of positive cells.

Statistical analysis

All statistical analyses, with the exception of SMR, were conducted using R software (version 4.1.1). The R packages “ggplot2” and “ggrepel” were employed to generate Manhattan plots, while “forestplot” was used to construct forest plots. The code for generating the “SMR Locus Plot” and “SMR Effect Plot” was adapted from Zhu et al. (28). Differentially expressed genes (DEGs) were identified using the “DESeq2” package with thresholds set at |log₂ fold change| >1 and an adjusted P value <0.01 (Benjamini-Hochberg method). Gene enrichment analysis was performed using the Metascape database (37) and gene set enrichment analysis (GSEA) database (38). Potential drug targets were identified through queries to the Drug-Gene Interaction Database (DGIdb) (https://dgidb.org/).

Results

ERS-related gene methylation and lung cancer

Through SMR analysis, a total of 509 methylation sites in proximity to 243 ESR-related genes were identified as potentially causally associated with lung cancer risk. In the colocalization analysis, 168 of these sites (corresponding to 97 genes) demonstrated strong evidence of colocalization (PP.H4 >0.5). For instance, the methylation levels of CpG sites cg08257293, cg12873919, cg13989999 (BCL2L1), cg11963233 (CDH2), and cg23090046 (KLC1) were negatively correlated with lung cancer risk [odds ratio (OR) <1], whereas methylation at cg26027576 (CLU) was positively correlated with lung cancer risk (OR >1). Notably, different CpG sites within the same gene exhibited inconsistent effects, as illustrated by two CpG sites in TP73: cg19692322 [OR =1.17; 95% confidence interval (CI): 1.01–1.35] and cg21418707 (OR =0.86; 95% CI: 0.76–0.98). Due to the volume of data, Figure 2 presents only a selected subset of CpG sites with strong colocalization evidence and potential relevance to lung cancer risk. The complete results are provided in website: https://cdn.amegroups.cn/static/public/tlcr-2025-474-1.xlsx. Key CpG site results, including cg10869879, cg00907267 (ERBB3), cg08257293, cg12873919, and cg13989999 (BCL2L1), were further validated using the FinnGen R10 lung cancer cohort (Table 1). The complete mQTL-GWAS results are available in website: https://cdn.amegroups.cn/static/public/tlcr-2025-474-2.xlsx.

Figure 2 SMR results for the association between the CpG site methylation of ERS-related genes and lung cancer risk. CI, confidence interval; ERS, endoplasmic reticulum stress; HEIDI, heterogeneity in dependent instruments; ID, identification; OR, odds ratio; PP.H4, posterior probability for hypothesis 4; PP.H3, posterior probability for hypothesis 3; SMR, summary data-based Mendelian randomization.

ERS-related gene expression and lung cancer

SMR analysis identified 64 ERS-related genes associated with lung cancer risk, of which, 35 genes demonstrated negative correlations and 29 positive correlations. The complete results of the eQTL-GWAS analysis are presented in website: https://cdn.amegroups.cn/static/public/tlcr-2025-474-3.xlsx. Colocalization analysis further revealed that 20 of these identified genes exhibited strong evidence of colocalization (PP.H4 >0.5) within the defined colocalization region windows; the corresponding SMR results are shown in Figure 3. Among these genes, CCNA2, STING1, MAP1A, and BCL2L1 were validated in the FinnGen R10 lung cancer cohort (Table 1), with the complete eQTL-GWAS validation results provided in website: https://cdn.amegroups.cn/static/public/tlcr-2025-474-4.xlsx.

Figure 3 SMR results for the association between the gene expression of ERS-related genes and lung cancer risk. CI, confidence interval; ERS, endoplasmic reticulum stress; HEIDI, heterogeneity in dependent instruments; ID, identification; OR, odds ratio; PP.H4, posterior probability for hypothesis 4; PP.H3, posterior probability for hypothesis 3; SMR, summary data-based Mendelian randomization.

ERS-related proteins and lung cancer

SMR analysis identified 19 ERS-related proteins associated with lung cancer risk. Among them, the abundance of 10 proteins, including ERBB3, GSR, KLC1, and KRT5, was positively correlated with lung cancer risk, whereas the abundance of 9 proteins, such as DLG3, DNAJB12, and FKBP4, showed negative correlations. Colocalization analysis revealed that 9 of these proteins exhibited strong evidence of colocalization (PP.H4 >0.5) within the defined colocalization region windows. Figure 4 presents the SMR and colocalization results for these proteins. However, these findings were not validated in the FinnGen R10 lung cancer cohort. The complete pQTL-GWAS results for the discovery and validation cohorts are provided in website: https://cdn.amegroups.cn/static/public/tlcr-2025-474-5.xlsx, https://cdn.amegroups.cn/static/public/tlcr-2025-474-6.xlsx, respectively.

Figure 4 SMR results for the association between the protein abundance of ERS-related genes and lung cancer risk. CI, confidence interval; ERS, endoplasmic reticulum stress; HEIDI, heterogeneity in dependent instruments; ID, identification; OR, odds ratio; PP.H4, posterior probability for hypothesis 4; PP.H3, posterior probability for hypothesis 3; SMR, summary data-based Mendelian randomization.

Multiomics results integration

By intersecting the results of mQTL-GWAS and eQTL-GWAS analyses, 5 genes, including RETSAT, IRF4, KLC1, CDH2, and BCL2L1, were identified as being potentially associated with lung cancer risk at both the methylation and gene expression levels (Figures 2,3). To investigate the regulatory relationship between methylation and gene expression, SMR analysis was conducted using mQTL as exposures and eQTL as outcomes. The analysis revealed that cg23090046 (KLC1), cg12873919 (BCL2L1), and cg13989999 (BCL2L1) may negatively regulate the expression of KLC1 and BCL2L1, respectively, as indicated by P_SMR <0.05, P_SMR-multi <0.05, and P_HEIDI >0.05 (Table 2). Similarly, to further characterize the regulatory relationship between gene expression and protein abundance, SMR analysis was conducted using eQTL as exposures and pQTL as outcomes. The results demonstrated that the expression of KLC1 was positively correlated with its protein abundance (P_SMR <0.05, P_SMR-multi <0.05, and P_HEIDI >0.05; Table 2). The complete results of the mQTL-eQTL and eQTL-pQTL SMR analyses are provided in website: https://cdn.amegroups.cn/static/public/tlcr-2025-474-7.xlsx; https://cdn.amegroups.cn/static/public/tlcr-2025-474-8.xlsx, respectively.

The integration of multiomics results revealed that the methylation level of the cg23090046 site (KLC1) was negatively correlated with lung cancer risk, while both the gene and protein expression levels of KLC1 were positively correlated with lung cancer risk. Specifically, cg23090046 methylation was negatively correlated with KLC1 gene expression, and the gene expression was positively correlated with KLC1 protein abundance. This suggests that the methylation site cg23090046 may regulate KLC1 gene expression, thereby influencing KLC1 protein synthesis and subsequently increasing the risk of lung cancer (Figure 5). Similarly, the integration of multiomics evidence suggests that the methylation levels of the cg12873919 and cg13989999 sites (BCL2L1) are negatively correlated with lung cancer risk, whereas BCL2L1 gene expression is positively correlated with lung cancer risk. The methylation levels of cg12873919 and cg13989999 are negatively correlated with BCL2L1 gene expression, indicating that these methylation sites may regulate BCL2L1 gene expression, thereby further increasing the risk of lung cancer (Figure 6). However, the relationships between BCL2L1 protein abundance and lung cancer risk, as well as between BCL2L1 gene expression and BCL2L1 protein abundance, were not statistically significant.

Figure 5 Integrated multiomics analysis of KLC1. (A) SMR analysis results from the integration of lung cancer GWAS and mQTL data at the KLC1 locus. (B) SMR analysis results of lung cancer GWAS and eQTL at the KLC1 locus. (C) SMR analysis results of lung cancer GWAS and pQTL at the KLC1 locus. (D) Correlation of SNP effect sizes for KLC1 between lung cancer GWAS and QTL data (mQTL, eQTL, pQTL). cis-QTL, cis-acting quantitative trait loci; eQTL, expression quantitative trait loci; GWAS, genome-wide association study; mQTL, methylation quantitative trait loci; pQTL, protein quantitative trait loci; SMR, summary data-based Mendelian randomization; SNP, single-nucleotide polymorphism.

Figure 6 Integrated multiomics analysis of BCL2L1. (A) SMR analysis results from the integration of lung cancer GWAS and mQTL data at the BCL2L1 locus. (B) SMR analysis results of lung cancer GWAS and eQTL data at the BCL2L1 locus. (C) Correlation of SNP effect sizes for BCL2L1 between lung cancer GWAS and QTL data (mQTL, eQTL). (D) Manhattan plots displaying the SMR results for KLC1 and BCL2L1 at the multiomics level. CHR on the X-axis indicates chromosome numbers. cis-QTL, cis-acting quantitative trait loci; eQTL, expression quantitative trait loci; GWAS, genome-wide association study; mQTL, methylation quantitative trait loci; pQTL, protein quantitative trait loci; SNP, single-nucleotide polymorphism; SMR, summary data-based Mendelian randomization.

Expression and prognostic analysis

We further examined the gene expression of KLC1 and BCL2L1 in patients with lung cancer. The results demonstrated that KLC1 expression in lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) was lower than that in normal lung tissue samples. Similarly, BCL2L1 expression in LUAD and LUSC was lower than that in normal lung tissue samples; however, the differences were not statistically significant (Figure 7A). Patients with lung cancer were stratified into high-expression and low-expression groups based on the median gene expression levels. Differences in types of immune cell infiltration and OS rates between the two groups were compared. Immune cell infiltration analysis revealed that patients with high BCL2L1 expression in LUAD and LUSC had a higher abundance of M2 macrophages compared to those with a low expression. Additionally, patients with LUSC and low KLC1 expression exhibited a higher proportion of M1 macrophages, whereas patients with LUAD and low KLC1 expression demonstrated a higher proportion of neutrophils as compared to those with expression (the detailed results are presented in Figure 7B). Prognostic analysis suggested that patients with high KLC1 and BCL2L1 expression had worse OS compared to those with low expression (KLC1: HR =1.29; BCL2L1: HR =1.17; Figure 7C). IHC revealed that in LUAD, 7 (46.7%) samples exhibited moderate density, and 4 (26.7%) exhibited high density of KLC1. In LUSC, 7 (50%) samples exhibited moderate density, and 3 (21.4%) exhibited high density of KLC1 (Figure 7D). The IHC results showed that the expression levels of BCL2L1 and KLC1 in lung cancer tissues were significantly higher than those in normal tissues, demonstrating a statistically significant difference (P<0.05; Figure 7E).

Figure 7 Expression and prognostic analysis of BCL2L1 and KCL1. (A) Differential expression analysis of KLC1 and BCL2L1 between lung cancer and normal samples. (B) Comparison of immune cell infiltration levels in the lung cancer microenvironment between high- and low-expression groups for KLC1 and BCL2L1. (C) Overall survival analysis comparing high- and low-expression groups of KLC1 and BCL2L1. (D) Immunohistochemical analyses of KLC1 in lung cancer tissue from the HPA database. (https://www.proteinatlas.org/ENSG00000126214-KLC1/cancer/lung+cancer). (E) Immunohistochemical staining of BCL2L1 and KLC1 in lung cancer and normal lung tissues (scale bar: 50 µm). *, P<0.05; **, P<0.01; ***, P<0.001. CI, confidence interval; HPA, Human Protein Atlas; HR, hazard ratio; CI, confident interval; LUAD, lung adenocarcinoma; LUSC, lung squamous cell carcinoma; N, normal lung tissues; T, lung cancer tissues; TPM, transcripts per million.

To further clarify the function of KCL1, we divided patients with NSCLC TCGA into two groups based on the median expression level of KCL1 and used DESeq2 to identify DEGs. A total of 918 DEGs were identified. Enrichment analysis revealed that these genes may be involved in the inhibition of B-cell immune protein secretion and the progression of NSCLC. Furthermore, we explored the potential drug interaction with KCL1 using DGIdb, and the results suggested that irinotecan, fluorouracil, and leucovorin calcium could serve as potential therapeutic agents targeting KCL1 (Figure S2 and Table S1).

Discussion

In this study, we integrated large-scale lung cancer GWAS and cis-QTL data, along with SMR and colocalization analyses, to determine the causal link between gene methylation, gene expression, and protein abundance of ERS-related genes and lung cancer risk. We identified 20 ERS-related genes and 9 proteins with strong evidence of having a causal link to lung cancer risk. By integrating multi-omics datasets, our study not only identified ERS-related genes whose methylation, transcription, and protein expression levels are all associated with lung cancer risk, thereby enhancing the robustness and credibility of the findings, but also revealed potential regulatory cascades linking specific CpG sites to gene and protein expression, and ultimately to disease susceptibility. This comprehensive framework provides a valuable roadmap for future experimental studies aimed at validating these mechanistic pathways. We found that low methylation at cg23090046 increased KLC1 expression and kinesin light chain 1 abundance, thereby contributing to an elevated risk of lung cancer. Similarly, low methylation at cg12873919 and cg13989999 promoted BCL2L1 expression, further increasing the risk of lung cancer. DNA methylation is thought to play a critical role in gene silencing, potentially by blocking promoters where activating transcription factors are supposed to bind, which supports the plausibility of our findings (39). Prognostic analysis confirmed that lung cancer patients with high expression levels of KLC1 and BCL2L1 exhibited poorer OS as compared to those with low expression levels. By leveraging large-scale GWAS and cis-QTL data, this study examined how genetic variations influence gene expression and protein abundance through methylation, revealing the mechanisms by which ERS-related genes may impact lung cancer risk. Given that few studies have investigated the role of ERS in lung cancer and the existing data remain contradictory (22,25), the findings of this study provide valuable insights into the biological role of ERS in lung cancer and suggest potential gene targets for diagnosis and treatment. Figure 8 illustrates the regulatory mechanisms and potential functions of the ERS-related genes KLC1 and BCL2L1 in lung cancer.

Figure 8 Diagram illustrating the regulatory mechanisms and functions of the ERS-related genes KLC1 and BCL2L1 in lung cancer. ERS, endoplasmic reticulum stress.

The kinesin superfamily is responsible for the intracellular transport of proteins, chromosomes, vesicles, and organelles (40). Kinesins are heterotetramers composed of two heavy chains and two light chains (41). The heavy chains act as molecular motors, facilitating movement along microtubules, while the light chains connect the cargo to the heavy chains (42,43). To date, fifteen kinesin families have been identified, each with distinct functions (43,44). A study has found that KLC1 mediates inflammatory responses by participating in transendothelial migration (42). Additionally, in neurodegenerative diseases such as Alzheimer’s disease and Down syndrome, reduced KLC1 levels may be associated with impaired anterograde axonal transport, potentially contributing to disease development (45). However, the role of KLC1 in cancer, particularly in lung cancer, remains underexplored. Moamer et al. found that KLC1 acts as a regulator of epithelial-mesenchymal plasticity in breast cancer, inducing an epithelial/luminal phenotype and inhibiting epithelial-mesenchymal transition, invasion, metastasis, and the expression of stem cell markers (46). Fujii et al. discovered that the KLC1-ROS1 fusion protein promotes glioma cell proliferation and invasion under serum-deprived conditions in a JAK-STAT-dependent manner (47). Meril et al. used multiplex immunofluorescence to show that low KLC1 expression is associated with poorer OS and progression-free survival (PFS) in patients with certain types of endometrial cancer (48). Furthermore, germline variants in tumors may influence the tumor mutation burden (TMB) in stomach adenocarcinoma by regulating KLC1 expression levels (49). In patients with lung cancer, KLC-ALK fusion genes and protein expression have been detected, and these patients may benefit from ALK-targeted therapy (42,50,51). In summary, our findings support the association between KLC1 and malignant tumor phenotypes, suggesting that it may serve as a potential therapeutic target in cancer. We found that methylation at the cg23090046 site may regulate KLC1 gene and protein expression, potentially increasing lung cancer risk through ERS. Additionally, high KLC1 expression is associated with poor prognosis in patients with NSCLC. Enrichment analysis further revealed that KCL1 might be involved in the inhibition of B-cell immune protein secretion and the progression of NSCLC. Given its role in intracellular transport, KLC1 may regulate immune responses by modulating the trafficking of cytokines or immune receptors, such as those involved in B cell activation and antigen presentation. This may impair B cell-mediated immunity and contribute to immune suppression in the tumor microenvironment (52). Additionally, KLC1 was identified in the DGIdb database as a potential interactor with irinotecan and leucovorin calcium, suggesting a possible link between KLC1 and chemotherapy efficacy in NSCLC. Although this interaction is based on curated and inferred sources, it warrants further experimental validation. These findings raise the possibility that KLC1 could serve as a novel therapeutic target, particularly in the context of immune modulation and chemosensitization.

The BCL2 protein family, which consists of key regulators of apoptosis, has been extensively studied in cancer (53). Research suggests that BCL2 exerts antiapoptotic effects in lung cancer, thereby contributing to tumor formation and chemotherapy resistance (54,55). Interestingly, BCL2 can promote tumor-associated neutrophils (TANs) to increase BCL-xL expression by secreting granulocyte-macrophage colony-stimulating factor (GM-CSF), thereby enhancing the tumor-supporting function of TANs. Targeted blockade of BCL-xL inhibits this effect, thereby reducing tumor growth (56). BCL2 is considered a potential therapeutic target in cancer, and its inhibitors may induce tumor apoptosis or reduce the number of cancer stem cells (57,58). This study identified a positive causal relationship between the expression of the ERS-related gene BCL2L1 and lung cancer risk, in which low methylation levels at the cg12873919 and cg13989999 loci increase lung cancer risk by promoting BCL2L1 expression. Kim et al. reported that ERS-induced death signals activate BAK through BH3-only molecules (tBID, BIM, and PUMA), thereby initiating a BAX/BAK-dependent mitochondrial death program and promoting apoptosis (59). BCL2 and BCL-xL prevent BAK activation by binding to activated BH3 proteins (60,61). Our findings and other evidence suggest that BCL2L1 may inhibit ERS-induced apoptotic signals in lung cancer. We hypothesize that lung cancer may evade ERS-induced apoptosis via BCL2L1, thereby promoting tumor development. BCL2L1 can be targeted by selective BCL-xL inhibitors, such as navitoclax (ABT-263), which have shown antitumor effects in preclinical models (62). In addition, we observed that high BCL2L1 expression correlates with increased M2 macrophage infiltration, a phenotype commonly linked to immunosuppression and tumor promotion. It is possible that BCL2L1 influences macrophage polarization through inhibition of ERS-mediated apoptosis or by altering cytokine signaling within the tumor microenvironment. These findings highlight the potential of BCL2L1 not only as a prognostic biomarker but also as a candidate for targeted immunomodulatory therapy in lung cancer. However, this hypothesis needs to be validated by further experiments.

This study has several limitations. First, all primary GWAS and cis-QTL datasets used in this study were derived from individuals of European ancestry. Given the known variability in genetic architecture, risk allele frequencies, and linkage disequilibrium patterns across populations, especially between European and East Asian groups, our findings may not be directly generalizable to other ethnic populations. For example, EGFR mutations are much more frequent in East Asian lung cancer patients than in Europeans, and variants in TP63 have shown population-specific associations with lung cancer risk (63,64). Further validation in non-European cohorts is necessary to assess the broader applicability of our results. Second, although key associations at the methylation and transcript levels were validated, the lack of protein QTL data for several ERS-related genes, especially BCL2L1, hindered our ability to confirm protein-level effects and complete the causal chain. Third, this study is primarily computational and lacks experimental validation. Although robust methods like SMR and colocalization were used to support causal inference, the roles of KLC1 and BCL2L1 in ERS and immune regulation remain hypothetical. Further functional studies are needed to confirm their mechanistic involvement in lung cancer. Finally, although the prognostic associations of KLC1 and BCL2L1 are statistically significant, they are derived from retrospective cohort data, which limits causal interpretation. Survival analyses alone cannot establish a direct mechanistic link, and prospective studies are needed to validate these findings.

Conclusions

In summary, the present study identified a causal relationship between ERS-related genes and lung cancer risk using a multiomics approach. Additionally, two potential ERS-related genes were identified as potentially playing crucial roles in ERS regulation in lung cancer and may serve as promising therapeutic targets.

Acknowledgments

We would like to thank the participants and investigators of the FinnGen study. We also extend our gratitude to the authors of the publicly available datasets used in this study, as well as to the participants and researchers involved in those original studies.

Footnote

Reporting Checklist: The authors have completed the STREGA reporting checklist. Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-474/rc

Data Sharing Statement: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-474/dss

Peer Review File: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-474/prf

Funding: This study was funded by the Natural Science Foundation of China (No. 82160696), the Yunnan Provincial Department of Science and Technology - Kunming Medical University Joint Special Project (No. 202401AY070001-033), the Yunnan Province Medical Elite Talent Project (No. L-2024022), and the Clinical Examination Industry-Academia-Research Talent Workstation Project.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-474/coif). M.S. received payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events from BMS, Pierre Fabre, Novartis, AstraZeneca. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All data used in this study were obtained from publicly available research projects, each of which had received prior ethical approval. The hematoxylin and eosin-stained tissue slides analyzed in this study were collected with appropriate consent and approved by the Ethics Committee of the First Affiliated Hospital of Kunming Medical University (approval No. 2020-L42). This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Han B, Zheng R, Zeng H, et al. Cancer incidence and mortality in China, 2022. J Natl Cancer Cent 2024;4:47-53. [Crossref] [PubMed]
2. Siegel RL, Giaquinto AN, Jemal A. Cancer statistics, 2024. CA Cancer J Clin 2024;74:12-49. [Crossref] [PubMed]
3. Mucci LA, Hjelmborg JB, Harris JR, et al. Familial Risk and Heritability of Cancer Among Twins in Nordic Countries. JAMA 2016;315:68-76. [Crossref] [PubMed]
4. Sampson JN, Wheeler WA, Yeager M, et al. Analysis of Heritability and Shared Heritability Based on Genome-Wide Association Studies for Thirteen Cancer Types. J Natl Cancer Inst 2015;107:djv279. [Crossref] [PubMed]
5. Imielinski M, Berger AH, Hammerman PS, et al. Mapping the hallmarks of lung adenocarcinoma with massively parallel sequencing. Cell 2012;150:1107-20. [Crossref] [PubMed]
6. McKay JD, Hung RJ, Han Y, et al. Large-scale association analysis identifies new lung cancer susceptibility loci and heterogeneity in genetic susceptibility across histological subtypes. Nat Genet 2017;49:1126-32. [Crossref] [PubMed]
7. Kudhair BK, Abdulridha FM, Hussain GM, et al. The association of combined GSTM1, GSTT1, and GSTP1 genetic polymorphisms with lung cancer risk in male Iraqi Waterpipe Tobacco (Nargila) smokers. Cancer Epidemiol 2024;93:102689. [Crossref] [PubMed]
8. Kudhair BK, Lafta IJ, Alabid NN. Waterpipe tobacco smoking and gene variants of CYP1A1-Ile462Val and-MspI polymorphisms are possibly associated with the risk of lung cancer in the Iraqi population. Meta Gene 2020;23:100623.
9. Kwak EL, Bang YJ, Camidge DR, et al. Anaplastic lymphoma kinase inhibition in non-small-cell lung cancer. N Engl J Med 2010;363:1693-703. [Crossref] [PubMed]
10. Wu YL, Tsuboi M, He J, et al. Osimertinib in Resected EGFR-Mutated Non-Small-Cell Lung Cancer. N Engl J Med 2020;383:1711-23. [Crossref] [PubMed]
11. Maurano MT, Humbert R, Rynes E, et al. Systematic localization of common disease-associated variation in regulatory DNA. Science 2012;337:1190-5. [Crossref] [PubMed]
12. Ward LD, Kellis M. Interpreting noncoding genetic variation in complex traits and human disease. Nat Biotechnol 2012;30:1095-106. [Crossref] [PubMed]
13. Human genomics. The Genotype-Tissue Expression (GTEx) pilot analysis: multitissue gene regulation in humans. Science 2015;348:648-60. [Crossref] [PubMed]
14. Võsa U, Claringbould A, Westra HJ, et al. Large-scale cis- and trans-eQTL analyses identify thousands of genetic loci and polygenic scores that regulate blood gene expression. Nat Genet 2021;53:1300-10. [Crossref] [PubMed]
15. Pietzner M, Wheeler E, Carrasco-Zanini J, et al. Mapping the proteo-genomic convergence of human diseases. Science 2021;374:eabj1541.
16. Wu Y, Zeng J, Zhang F, et al. Integrative analysis of omics summary data reveals putative mechanisms underlying complex traits. Nat Commun 2018;9:918. [Crossref] [PubMed]
17. Boyce M, Yuan J. Cellular response to endoplasmic reticulum stress: a matter of life or death. Cell Death Differ 2006;13:363-73. [Crossref] [PubMed]
18. Kaufman RJ. Stress signaling from the lumen of the endoplasmic reticulum: coordination of gene transcriptional and translational controls. Genes Dev 1999;13:1211-33. [Crossref] [PubMed]
19. Mori K. Tripartite management of unfolded proteins in the endoplasmic reticulum. Cell 2000;101:451-4. [Crossref] [PubMed]
20. Schröder M, Kaufman RJ. The mammalian unfolded protein response. Annu Rev Biochem 2005;74:739-89. [Crossref] [PubMed]
21. Breckenridge DG, Germain M, Mathai JP, et al. Regulation of apoptosis by endoplasmic reticulum pathways. Oncogene 2003;22:8608-18. [Crossref] [PubMed]
22. Yuan S, She D, Jiang S, et al. Endoplasmic reticulum stress and therapeutic strategies in metabolic, neurodegenerative diseases and cancer. Mol Med 2024;30:40. [Crossref] [PubMed]
23. Tameire F, Verginadis II, Koumenis C. Cell intrinsic and extrinsic activators of the unfolded protein response in cancer: Mechanisms and targets for therapy. Semin Cancer Biol 2015;33:3-15. [Crossref] [PubMed]
24. Rouschop KM, van den Beucken T, Dubois L, et al. The unfolded protein response protects human tumor cells during hypoxia through regulation of the autophagy genes MAP1LC3B and ATG5. J Clin Invest 2010;120:127-41. [Crossref] [PubMed]
25. Kim H, Bhattacharya A, Qi L. Endoplasmic reticulum quality control in cancer: Friend or foe. Semin Cancer Biol 2015;33:25-33. [Crossref] [PubMed]
26. Liu L, Li S, Qu Y, et al. Ablation of ERO1A induces lethal endoplasmic reticulum stress responses and immunogenic cell death to activate anti-tumor immunity. Cell Rep Med 2023;4:101206. [Crossref] [PubMed]
27. Marciniak SJ. Endoplasmic reticulum stress in lung disease. Eur Respir Rev 2017;26:170018. [Crossref] [PubMed]
28. Zhu Z, Zhang F, Hu H, et al. Integration of summary data from GWAS and eQTL studies predicts complex trait gene targets. Nat Genet 2016;48:481-7. [Crossref] [PubMed]
29. Kurki MI, Karjalainen J, Palta P, et al. FinnGen provides genetic insights from a well-phenotyped isolated population. Nature 2023;613:508-18. [Crossref] [PubMed]
30. de Leeuw CA, Neale BM, Heskes T, et al. The statistical properties of gene-set analysis. Nat Rev Genet 2016;17:353-64. [Crossref] [PubMed]
31. Morrow JD, Glass K, Cho MH, et al. Human Lung DNA Methylation Quantitative Trait Loci Colocalize with Chronic Obstructive Pulmonary Disease Genome-Wide Association Loci. Am J Respir Crit Care Med 2018;197:1275-84. [Crossref] [PubMed]
32. GTEx Consortium. Genetic effects on gene expression across human tissues. Nature 2017;550:204-13. Erratum in: Nature 2018;553:530. [Crossref] [PubMed]
33. Yoshiji S, Butler-Laporte G, Lu T, et al. Proteome-wide Mendelian randomization implicates nephronectin as an actionable mediator of the effect of obesity on COVID-19 severity. Nat Metab 2023;5:248-64. [Crossref] [PubMed]
34. Kousathanas A, Pairo-Castineira E, Rawlik K, et al. Whole-genome sequencing reveals host factors underlying critical COVID-19. Nature 2022;607:97-103. [Crossref] [PubMed]
35. Tang Z, Li C, Kang B, et al. GEPIA: a web server for cancer and normal gene expression profiling and interactive analyses. Nucleic Acids Res 2017;45:W98-W102. [Crossref] [PubMed]
36. Győrffy B. Integrated analysis of public datasets for the discovery and validation of survival-associated genes in solid tumors. Innovation (Camb) 2024;5:100625. [Crossref] [PubMed]
37. Zhou Y, Zhou B, Pache L, et al. Metascape provides a biologist-oriented resource for the analysis of systems-level datasets. Nat Commun 2019;10:1523. [Crossref] [PubMed]
38. Subramanian A, Tamayo P, Mootha VK, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A 2005;102:15545-50. [Crossref] [PubMed]
39. Moore LD, Le T, Fan G. DNA methylation and its basic function. Neuropsychopharmacology 2013;38:23-38. [Crossref] [PubMed]
40. Hirokawa N. Kinesin and dynein superfamily proteins and the mechanism of organelle transport. Science 1998;279:519-26. [Crossref] [PubMed]
41. Gindhart JG Jr, Desai CJ, Beushausen S, et al. Kinesin light chains are essential for axonal transport in Drosophila. J Cell Biol 1998;141:443-54. [Crossref] [PubMed]
42. Cyrus BF, Muller WA. A Unique Role for Endothelial Cell Kinesin Light Chain 1, Variant 1 in Leukocyte Transendothelial Migration. Am J Pathol 2016;186:1375-86. [Crossref] [PubMed]
43. Verhey KJ, Hammond JW. Traffic control: regulation of kinesin motors. Nat Rev Mol Cell Biol 2009;10:765-77. [Crossref] [PubMed]
44. Hirokawa N, Noda Y, Tanaka Y, et al. Kinesin superfamily motor proteins and intracellular transport. Nat Rev Mol Cell Biol 2009;10:682-96. [Crossref] [PubMed]
45. Chen XQ, Das U, Park G, et al. Normal levels of KIF5 but reduced KLC1 levels in both Alzheimer disease and Alzheimer disease in Down syndrome: evidence suggesting defects in anterograde transport. Alzheimers Res Ther 2021;13:59. [Crossref] [PubMed]
46. Moamer A, Hachim IY, Binothman N, et al. A role for kinesin-1 subunits KIF5B/KLC1 in regulating epithelial mesenchymal plasticity in breast tumorigenesis. EBioMedicine 2019;45:92-107. [Crossref] [PubMed]
47. Fujii T, Nakano Y, Hagita D, et al. KLC1-ROS1 Fusion Exerts Oncogenic Properties of Glioma Cells via Specific Activation of JAK-STAT Pathway. Cancers (Basel) 2023;16:9. [Crossref] [PubMed]
48. Meril S, Muhlbauer Avni M, Lior C, et al. Loss of EIF4G2 mediates aggressiveness in distinct human endometrial cancer subpopulations with poor survival outcome in patients. Oncogene 2024;43:1098-112. [Crossref] [PubMed]
49. Sun X, Xue A, Qi T, et al. Tumor Mutational Burden Is Polygenic and Genetically Associated with Complex Traits and Diseases. Cancer Res 2021;81:1230-9.
50. Choi YL, Lira ME, Hong M, et al. A novel fusion of TPR and ALK in lung adenocarcinoma. J Thorac Oncol 2014;9:563-6. [Crossref] [PubMed]
51. McCoach CE, Blakely CM, Banks KC, et al. Clinical Utility of Cell-Free DNA for the Detection of ALK Fusions and Genomic Mechanisms of ALK Inhibitor Resistance in Non-Small Cell Lung Cancer. Clin Cancer Res 2018;24:2758-70. [Crossref] [PubMed]
52. Lacy P, Stow JL. Cytokine release from innate immune cells: association with diverse membrane trafficking pathways. Blood 2011;118:9-18. [Crossref] [PubMed]
53. García-Sáez AJ. The secrets of the Bcl-2 family. Cell Death Differ 2012;19:1733-40. [Crossref] [PubMed]
54. Inoue-Yamauchi A, Jeng PS, Kim K, et al. Targeting the differential addiction to anti-apoptotic BCL-2 family for cancer therapy. Nat Commun 2017;8:16078. [Crossref] [PubMed]
55. Barbie DA, Tamayo P, Boehm JS, et al. Systematic RNA interference reveals that oncogenic KRAS-driven cancers require TBK1. Nature 2009;462:108-12. [Crossref] [PubMed]
56. Bodac A, Mayet A, Rana S, et al. Bcl-xL targeting eliminates ageing tumor-promoting neutrophils and inhibits lung tumor growth. EMBO Mol Med 2024;16:158-84. [Crossref] [PubMed]
57. Zeuner A, Francescangeli F, Contavalli P, et al. Elimination of quiescent/slow-proliferating cancer stem cells by Bcl-XL inhibition in non-small cell lung cancer. Cell Death Differ 2014;21:1877-88. [Crossref] [PubMed]
58. Aird D, Teng T, Huang CL, et al. Sensitivity to splicing modulation of BCL2 family genes defines cancer therapeutic strategies for splicing modulators. Nat Commun 2019;10:137. [Crossref] [PubMed]
59. Kim H, Tu HC, Ren D, et al. Stepwise activation of BAX and BAK by tBID, BIM, and PUMA initiates mitochondrial apoptosis. Mol Cell 2009;36:487-99. [Crossref] [PubMed]
60. Cheng EH, Wei MC, Weiler S, et al. BCL-2, BCL-X(L) sequester BH3 domain-only molecules preventing BAX- and BAK-mediated mitochondrial apoptosis. Mol Cell 2001;8:705-11. [Crossref] [PubMed]
61. Kim H, Rafiuddin-Shah M, Tu HC, et al. Hierarchical regulation of mitochondrion-dependent apoptosis by BCL-2 subfamilies. Nat Cell Biol 2006;8:1348-58. [Crossref] [PubMed]
62. Leverson JD, Phillips DC, Mitten MJ, et al. Exploiting selective BCL-2 family inhibitors to dissect cell survival dependencies and define improved strategies for cancer therapy. Sci Transl Med 2015;7:279ra40. [Crossref] [PubMed]
63. Midha A, Dearden S, McCormack R. EGFR mutation incidence in non-small-cell lung cancer of adenocarcinoma histology: a systematic review and global map by ethnicity (mutMapII). Am J Cancer Res 2015;5:2892-911.
64. Miki D, Kubo M, Takahashi A, et al. Variation in TP63 is associated with lung adenocarcinoma suscep’tibility in Japanese and Korean populations. Nat Genet 2010;42:893-6. [Crossref] [PubMed]