Background
Lung cancer, one of the most common malignant tumors worldwide, remains the first leading cause of cancer deaths [
1]. Lung adenocarcinoma (LUAD) is the most frequently diagnosed histological subtype of non-small cell lung cancer (NSCLC), followed by squamous cell carcinoma [
2]. Various factors such as cigarette smoking, second-hand or passive smoking, air pollution, genetic alteration, asbestos, and radon put individuals under the risk of LUAD [
3]. Although alternative treatments such as targeted therapy and immune therapy for LUAD patients have been progressed rapidly over the past decades, the average 5-year survival rate of patients with LUAD remains less than 20% [
4,
5]. A better understanding of the biological processes and molecular mechanisms underlying lung cancer pathogenesis will be of great significance to clinicians and improve patients’ outcomes. Thus it is of vital importance to keep searching for new tumor biomarkers and other potential genetic targets [
6].
SEC61G, also known as Sec61 gamma subunit, is a central member of the SEC61 complex, a heterotrimeric protein channel formed by three subunits, SEC61 α, β, and γ [
7]. Sec61 combined Sec62 and Sec63 serves as core component of the protein translocation machinery in the endoplasmic reticulum (ER) membrane and participates in protein folding, post-translational modification, translocation and unfolded protein response (UPR), especially under conditions of ER stress such as hypoxia and glucose deprivation in the tumor microenvironment [
8,
9]. High frequency of mutation and overexpression of SEC62 and SEC63 have been observed in kinds of cancers, suggesting the potential role of ER protein in tumor development [
10,
11]. Several studies have demonstrated that SEC61G was overexpressed in Glioblastoma [
12], gastric cancer [
13], hepatocellular carcinoma [
14,
15] and breast carcinomas [
16]. SEC61G gene was also found to coamplify with epidermal growth factor receptor in patients with glioblastoma and served as a potential prognostic marker [
12,
17]. However, the potential correlation between SEC61G and lung adenocarcinoma has not been characterized.
In this present study, we comprehensively investigated the prognostic impact of SEC61G expression in LUAD patients through the gene expression profile and the matching clinical information of LUAD patients from the Cancer Genome Atlas (TCGA) and the Gene Expression Omnibus (GEO) database. A nomogram based on several independent risk factors was constructed. Functional enrichment and Gene set enrichment analysis (GSEA) were performed to explore the underlying mechanism of SEC61G involved in LUAD pathogenesis. We also identified the association between SEC61G expression and genetic alteration and methylation. Additionally, single-sample Gene Set Enrichment Analysis (ssGSEA) and TIMER were used to explore the correlation between the infiltration of immune cells and the expression of SEC61G. Finally, cell functional experiments were performed to verify the biological behavior of SEC61G in lung adenocarcinoma cells (LAC). We hope our study will help others get further insight into understanding the potential role of SEC61G in tumor pathogenesis and contribute to the improvement of molecular targeted therapy and prognosis for LUAD patients.
Methods
Data acquisition
The gene expression data, phenotype data, and the corresponding clinicopathological information of the TCGA-LUAD project and other tumor were acquired from the UCSC Xena browser (version: 2019-07-20,
http://xenabrowser.net/datapages/). Clinical parameters, including age, gender, TNM stage and pathological stage, were evaluated. Transcriptome profiling data of patients with LUAD in the GSE11969 dataset from the Gene Expression Omnibus (https://www/
ncbi.nlm.nih.gov/geo/) database were used for external validation of survival analyses. The two databases’ exclusion criteria were as follows: cases without complete gene expression data and survival information. Finally, 497 cases were extracted for further analysis. Patients with LUAD were classified into low- and high-expression groups according to the optimal cut-off value of SEC61G.
The expression levels of SEC61G in various cancers including LUAD were acquired from the TIMER database [
18]. (
https://cistrome.shinyapps.io/timer/). The Broad Institute Cancer Cell Line Encyclopedia Database (CCLE;
http://www.broadinstitute.org/ccle) was used to validate the expression levels of SEC61G in different types of cancer cell lines [
19]. The datasets of IMvigor 210 Clinical Trial was also download via the IMvigor210CoreBiologies R package and used for the survival analyses [
20].
Co-expressed and functional enrichment analysis
We used R (version 3.6.1) to identify the genes co-expressed with SEC61G in the TCGA-LUAD cohort. Spearman’s correlation coefficient (
r) was calculated to evaluate the correlation between SEC61G and co-expressed genes, of which |
r| > 0.35 and
P < 0.001 were selected. Meanwhile, co-expressed analysis was also performed to explore the correlation between SEC61G and ER-stress related genes which was obtained from the Molecular Signatures Database (
https://www.gsea-msigdb.org/gsea/msigdb/genesets.jsp). To explore the underlying biological mechanism of SEC61G in LUAD pathogenesis, Metascape [
21](
https://metascape.org) was used conducted gene ontology (GO) analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis for genes co-expressed with SEC61G as previously selected.
Gene set enrichment analysis
Gene Set Enrichment Analysis (GSEA) is a computational method that determines whether a set of a priori defined genes show statistically significant and consistent differences between biological states [
22]. In this study, We conducted GSEA using the “clusterProfiler” R package (3.8.0, [
23]) to elucidate the statistically significant function and pathway difference between high and low SEC61G expression groups of LUAD. H.all.v7.0.symbols.gmt in the MSigDB Collections was used as the reference gene collection. The expression level of SEC61G was regarded as a phenotype label. Adjusted
P-value < 0.001, FDR q-value < 0.001 and |NES| > 1.5 was considered as statistically significant.
Analysis between SEC61G expression and its correlation with immune infiltration
We conducted the ssGSEA (single-sample Gene Set Enrichment Analysis) method from the GSVA package [
24] in R software to investigate the correlation between SEC61G expression and the immune cell infiltration level based on the published signature gene lists [
25]. Spearman correlation was carried out to evaluate the correlation between SEC61G and immune cell infiltration. TIMER online tool was applied to validate the association between the SEC61G expression and the level of immune cell infiltration.
SEC61G methylation level and its prognostic analysis
The copy number variation (CNV) and methylation level data of SEC61G were acquired from the cBioPortal web platform (
https://www.cbioportal.org/) and a comparison of the varying SEC61G gene expressions in SEC61G copy number variation groups (Kruskal-Wallis test) and the correlation between SEC61G methylation level and SEC61G gene expression (Spearman correlation) was conducted. The SMART online platform (
http://www.bioinfo-zs.com/smartapp/) was used to visualize the methylation levels of SEC61G in pan-cancer and normal samples from the TCGA database. The UALCAN online tool (
http://ualcan.path.uab.edu/) was used to identify the differences in the promoter methylation level of SEC61G between LUAD and normal tissues from TCGA data. MethSurv online tool (
https://biit.cs.ut.ee/methsurv/) was used to explore the prognostic value of the SEC61G methylation level in the TCGA-LUAD cohort.
Prognostic model generation and prediction
Univariate analysis and multivariate cox regression analysis were used to determine the optimal prognostic model. Then, a nomogram was constructed to predict the prognosis by R packages rms. The patients were stratified into high and low risk groups based on the optimal cut-off value. The difference in OS between the high-risk group and low-risk group was analyzed by the Kaplan-Meier method with a two-sided log-rank test. The concordance index(C-index) and calibration curves were used to evaluate nomogram models’ quality. The C-index is between 0.5 and 1.0, where 1.0 indicates the model has a perfect capacity to distinguish outcomes with the model correctly, and 0.5 indicates random probability. The calibration curve is evaluated graphically by plotting the nomogram’s predicted probability against the observed rates. Overlap with the reference line indicates that the consistency of the model is perfect.
Cell lines and cell culture
Human lung carcinoma cell lines A549 and H1299 were obtained from ATCC. A549 cells were cultured in F12K medium supplemented with 10% foetal bovine serum (FBS, AusGeneX, Australia), 100 U/ml penicillin, and 100 mg/ml streptomycin. H1299 cells were cultured in RPMI 1640 medium supplemented with 10% FBS, 100 U/ml penicillin, and 100 mg/ml streptomycin. Cell cultures were maintained in a humidified incubator consisting 5% CO2 at 37 °C.
Small interfering RNA (siRNA) transfection
Knockdown of the expression of SEC61G in lung cancer cells was accomplished by small interfering RNAs (siRNAs) transfection using the Lipofectamine™ 3000 transfection reagent, according to the manufacturer’s instructions. Three small interfering RNA (siRNAs) targeting SEC61G (si-1, si-2 and si-3) and siRNA negative control (NC) were obtained from GenePharma (D010003; Shanghai, China). siRNAs were designed and synthesized by GenePhama (Shanghai, China). The siRNA sequences were as follows:
-
SEC61G si-1 sense: 5′-CAGCAAUAGGAUUUGCUAUAATT-3′;
-
SEC61G si-1 antisense: 5′-UUAUAGCAAAUCCUAUUGCUGTT-3′;
-
SEC61G si-2 sense: 5′-AUCUUAGAGAUUGGUGAACAATT-3′;
-
SEC61G si-2 antisense: 5′-UUGUUCACCAAUCUCUAAGAUTT-3′;
-
SEC61G si-3 sense: 5′-AGCCAAGUCGGCAGUUUGUAATT-3′;
-
SEC61G si-3 antisense: 5′-UUACAAACUGCCGACUUGGCUTT-3′.
Quantitative real-time polymerase chain reaction (qRT-PCR)
The Trizol RNA extraction kit was used to extracted total RNA from transfected cells. Then the reverse transcription kit was employed to reversely transcribe RNA into cDNA. Subsequently, qRT-PCR was conducted to measure the expression of SEC61G. The calculation of inhibitory efficacy of the genes was performed with 2 − ΔΔCt method. GAPDH acted as the internal references. The primer sequences are as follows:
-
SEC61G-forward: 5′-ACGTGTCCCTGGCATTTTAG-3′;
-
SEC61G-reverse: 5′-TCAGCCACCAACAATGATGT-3′;
-
GAPDH-forward: 5′-GAAGGTGAAGGTCGGAGTC-3′;
-
GAPDH-reverse: 5′-GAAGATGGTGATGGGATTTC-3′.
Western blotting
RIPA extraction reagent (Beyotime, Shanghai, China) was used for lysing cells to extract the total protein, and BCA Protein Assay (Thermo-Fisher Scientific, Waltham, MA, USA) was utilized to determine the protein concentration. Subsequently, samples were subjected to SDS-PAGE. Following the electrophoresis, proteins were transferred to polyvinylidene difluoride (PVDF) membranes (Millipore, Billerica, MA, USA). Next, the membranes and primary antibody (anti-SEC61G: 1:1000, 111,472–2-AP, ProteinTech, USA; anti-β-actin: 1:100000, AC026, ABclonal, USA) were incubated overnight at 4 °C and then were incubated with secondary antibody (goat anti-rabbit IgG-HRP, 1:5000, AS014; ABclonal, USA) for 1 h at room temperature. Sequentially, protein bands were visualized using chemiluminescence.
Cell counting kit-8 (CCK8) assay
After transfection, cells were seeded into a 96-well plate according to the standard of 5000 cells per well. The proliferation of A549 or H1299 cells in the si-SEC61G and NC group was detected respectively. Cell viability was measured every 12 h. Briefly, 10 μl CCK8 reagent (MCE, Shanghai, China) was added daily to each hole in the 96-well plates and incubated at 37 °C for 1.5 h. The microplate reader (Infinite® M1000 PRO, TECAN, Switzerland) was used to measure the OD value at 450 nm.
Briefly, 1000 cells were seeded into a 6-well plate and incubated in culture medium with 10% foetal bovine serum at 37 °C. The plates were incubated at 37 °C, 5% CO2 incubator for 2 weeks with frequent observation. Then the supernatant was removed and cells were washed carefully twice with PBS. Cells were fixed with 4% paraformaldehyde for 0.5 h before dyeing with 0.1% crystal violet. The number of clones was counted directly by the naked eye. The size and number of clones were compared finally.
Cell apoptosis
The culture medium was replaced with a serum-free medium after 48 h of transfection., Cells were obtained by centrifugation after another 24 h of starvation and then resuspended by cold PBS (4 °C) and then centrifuged again. The supernatant was carefully removed. Cells were spread around in the binding buffer and complied with the Annexin V-FITC staining manufacturer’s instructions (US Everbright® Inc., Suzhou, China). FACS Calibur flow cytometer (BD Biosciences, Mountain View, CA) was used to determine the effects of SEC61G on lung cancer cell apoptosis.
Cell invasion assay
Cell invasion capability was conducted using the 24-well Transwell chambers (8 μm pore size; Corning, NY, USA). The transwell chambers were coated with 100 μl matrigel (5× dilution; 100 μL/well; BD Biosciences, Bedford, MA) in a 24-well plate at 37 °C for 4 h. Then the upper compartment containing 100 μL of serum-free medium was added with 2 × 104 cells. Meanwhile, the lower compartment was added with 600 μL of medium with 10% FBS. After being cultured for 48 h in a 37 °C incubator, cotton sticks were used to wipe off the cells remaining in the upper chamber, and the cells penetrating the filter were fixed with 4% paraformaldehyde and mixed with 0.4% crystal violet solution. Subsequently, the stained cells were observed and photographed under a microscope at 100× magnification (Olympus, Tokyo, Japan). At least four fields in each chamber were observed under the microscope. The average number of invasion cells in each microscopic fields was counted.
Statistical analysis
All statistical analyses were performed with the statistical package IBM SPSS Statistics software (SPSS26.0), GraphPad Prism 8.0 and R (version 3.6.1). The Wilcoxon signed-rank test and Wilcoxon rank-sum test were used to analyze the expression of SEC61G in paired and non-paired samples. X-tile software (Version 3.6.1) was conducted to identify the optimal cut-off value. The relationships between clinicopathological features and SEC61G expression were evaluated using Wilcoxon signed-rank test and logistic regression. Univariate analysis and multivariate analysis were done by applying Cox logistic regression model to identify independent variables, including age, gender, T stage, N stage, M stage, pathological stage, and SEC61G expression. The 95% confidence interval (Cl) of HR was calculated to assess individual factors hazard risk. The “survival” R package (version:0.1.3) and the “survminer” R package (version 0.4.8) were used to draw survival curves. The DESeq2 package (version: 3.26.5) was used to identify differential expression analysis between low and high SEC61G expression groups. The student’s t-test was carried out for comparisons between each group in vitro experiments. All tests were two-sided, and a P-value < 0.05 was considered as statistically significant.
Discussion
Lung adenocarcinoma, which constitutes almost 50% of NSCLCs, is a severe global public health problem with high mortality and morbidity in cancer patients [
1,
2]. Although newly molecular targeted therapy and immunotherapy have shed lights on NSCLC treatment, the therapeutic efficacy is still limited due to the high heterogeneity of lung cancer and lung cancer patients’ outcome is still far from satisfactory [
26]. This study aimed to identify a novel prognostic biomarker for LUAD patients and help to guide individualized treatments and optimize the therapeutic strategy.
Several studies have demonstrated that SEC61G was overexpressed in various malignancies. However, the potential correlation between SEC61G and lung adenocarcinoma remains unclear. Our study showed that SEC61G was upregulated in LUAD tissues compared with adjacent normal tissues in the TCGA-LUAD cohort, which was consistent with previous studies founded in glioblastoma and hepatocellular carcinoma [
14,
17]. Our analysis also found that SEC61G is significantly overexpressed in multiple malignancies in the TCGA database, such as gastric, liver cancer and breast carcinomas. These findings indicated that SEC61G could be a potential diagnostic marker in some type of cancers. In our study, the SEC61G gene was demonstrated as a potential diagnostic marker in LUAD, and its AUC exceeded 0.85. In the meanwhile, we also found that SEC61G correlated with the T stage, N stage and pathological grade of LUAD, further corroborating that the expression of SEC61G might be associated with the degree of malignancy of LUAD.
Then we demonstrated that patients with high SEC61G expression were significantly related to shorter overall survival (OS). Multivariate Cox regression analyses revealed that SEC61G expression served as an independent prognostic factor of OS. This result was also validated in LUAD patients from GEO database. Considering that SEC61G is a strong prognostic factor, we constructed a nomogram based on age, pathological stage and the SEC61G expression to predict the 1-, 3-and 5-year survival probability in LUAD patients. Calibration curves suggested that the actual prognosis was closely corresponded to the 1-and 3-year predicted prognosis, indicating good prediction performance of the nomogram. The nomogram could help to identify the high-risk patients and choose the more aggressive therapeutic strategy, which had better predictability than individual prognostic factors.
Although various mechanisms can contribute to elevated gene expression levels, DNA methylation and CNVs are two of the most common situations. DNA methylation can not only regulate gene expression but also plays a key role in tumorigenesis [
27]. Previous studies demonstrated that several highly-expressed genes due to hypomethylation were associated with poor prognosis in NSCLC [
28,
29]. Our results showed that SEC61G expression was associated with SEC61G hypomethylation (R = − 0.290,
p < 0.001) and the promoter methylation level of SEC61G in LUAD is lower than that in normal tissue. More importantly, SEC61G methylation was related to the prognosis of LUAD, and hypomethylated patients have worse OS, which is in line with the prognostic value of the mRNA expression of SEC61G. Hence, SEC61G hypomethylation might partly contribute to SEC61G overexpression and was associated with poor prognosis in LUAD.
SEC61G encodes the core subunit of SEC61 complex in the endoplasmic reticulum membrane and plays a critical role in protein translocation and cellular calcium homeostasis [
8]. Currently, the biological mechanism of SEC61G in tumors is still under exploration. In this study, GO and KEGG analysis indicated that SEC61G was significantly associated with the proliferation-associated biological process such as DNA replication, cell cycle and cell division, which is consistent with previous research in hepatocellular carcinoma [
14]. To further verify the biological mechanism of SEC61G, we conducted a series of In vitro experiments in A427 and H1299 cells. Our experiments here demonstrated that knockdown of SEC61G inhibits LAC cell proliferation, invasion, and favorited apoptosis, which is in accordance with our bioinformatic prediction. It is worth noting that GSEA analysis found SEC61G was significantly related to the E2F targets pathway. E2Fs are a complex family of transcriptional regulators and play a key role in protecting cells from cell cycle-generated genomic errors and abnormal proliferation [
30]. Li et al. found that overexpressed E2F genes were associated with poor prognosis in lung cancer patients [
31]. Therefore, SEC61G might participate in the E2F-related pathway to regulate the cell cycle of lung cancer cells. Based on the above findings, overexpression of SEC61G might take an active part in cell cycle, cell division and E2F-related pathway in LUAD and leading to the occurrence and progression of lung cancer.
Another important finding was that SEC61G was significantly related to the unfolded protein response (UPR), which is an adaptive mechanism to reinstate ER proteostasis under the ER stress. Potent ER stress responses have been reported in the majority of human cancer, including brain, lung, breast, colon, gastric, pancreatic, prostate skin. Productive, non-lethal ER stress or activation of UPR-mediated cytoprotective functions, could promote tumorigenesis, mediate resistance to treatment and orchestrate various immune-evasive mechanisms [
32]. UPR modulators such as IRE1α kinase inhibitors, PERK inhibitors and eIF2α inhibitors have shown notable anti-tumor efficacy in preclinical cancer models and hence provide a new insight of targets therapies [
33‐
35]. It’s worth noting that SEC61G might confers a selective growth advantage by facilitating a cytoprotective response to ER stress [
17]. Given that Our study showed significant association between SEC61G expression and ER stress response, we suspected that high expression of SEC61G play a critical role in malignant tumor behavior of LUAD via adaptive ER stress responses and SEC61G-related inhibitors might encouraging for the potential treatment of LUAD. Further studies are needed to explore the detailed molecular mechanisms through which ER stress response pathways stimulates the expression of SEC61G and induce its prosurvival effects. With the advent of immunotherapy, the tumor microenvironment has received more and more attention. Studies have shown that immune cells account for a large proportion of the tumor microenvironment and play a critical role in tumor development [
25]. Our study demonstrated that the SEC61G expression was negatively correlated with CD4+ T cells, CD8+ T cells, NK cells, DCs and B cells, indicating that SEC61G might play an inhibitory role in both innate immunity and adaptive immunity [
36]. There is accumulating evidence indicating that high levels of CD4+ T cells, CD8+ T cells, NK cells, DCs, and B cells infiltration are correlated with better prognosis in LUAD patients [
37‐
39]. Additionally, patients with higher levels of CD8+ T cells and CD4+ cells infiltration were more likely to benefit from immunotherapy [
40,
41]. Interestingly, We found that metastatic urothelial cancer patients with higher SEC61G expression level had a worse prognosis than those patients with low SEC61G expression who both received the atezolizumab immunotherapy in The IMvigor 210 study. Hence, it is reasonable to speculate that the SEC61G-mediatedimmunosuppression was one of the underlying causes of poor outcomes in LUAD patients and could provide a reference for the efficacy of immunotherapy in LUAD patients.
Although this current study enhanced a better understanding of the relationship between SEC61G and LUAD, some limitations of our study needed to be noted. First, our investigations into the role of SEC61G in tumors were based on the TIMER, TCGA and GEO databases, which lacks verification from our own clinical samples. Second, given that our study design is limited, additional key signaling pathways associated with SEC61G might be missed, and specific details on these relevant pathways and SEC61G-mediated immunosuppression are still unclear. Traditional in-house experimental studies and prospective studies are needed for further validation.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.