Introduction

Lung cancer has been the top-ranked cause of tumour-related death and a long-standing problem for human beings1,2. Non-small cell lung cancer (NSCLC) and small cell lung cancer are two substantial subtypes of lung cancer. NSCLC mainly consists of lung adenocarcinoma (LUAD), and lung squamous cell carcinoma (LUSC) is predominant and accounts for almost 85% of newly diagnosed lung cancer cases3. Despite the efforts made to study the mechanism and new treatment of NSCLC, the morbidity and mortality of NSCLC remain high. Currently, the 5-year survival rate of NSCLC remains approximately 20%1, which makes it important for us to search for novel biomarkers and clinical therapeutic targets for NSCLC.

Ephrin receptors (Ephs) make up the largest class of receptor tyrosine kinases (RTKs). Among them involved 15 EPH receptors and 9 related ephrins. EPH receptors can be categorized into two groups, EphA and EphB, according to their extracellular domains. EphA receptors mostly combine with GPI-anchored ephrin-A ligands, while EphB receptors primarily engage transmembrane ephrin-B ligands. Dysregulation of EphAs has been revealed in a variety of tumours, such as breast cancer4, gastric cancer5, colorectal cancer6, bladder cancer7, and lung cancer8. Many studies have indicated that EphA receptors are involved in carcinogenesis by promoting cell proliferation, migration, and invasion. For instance, EphA1 cleaved by ADAM12 contributes to lung metastasis9. Overexpression of EphA5 is related to cell migration and invasion in NSCLC10. MiR-448 regulates cell growth and epithelial-to-mesenchymal transition (EMT) by regulating PI3K/AKT signaling by targeting EphA7 in NSCLC11. However, few studies have focused on the synthetic analysis of the expression and prognostic value of EphAs in NSCLC.

Moreover, some members of the EphA family have been regarded as potential therapeutic targets for cancer treatment. Several inhibitors or compounds have been developed to target EphAs. Therapy targeting EphA2, including decreasing EphA2 expression and promoting EphA2 degradation, and EphA2-based immunotherapy are under clinical trials12. EphA5 was also considered a valuable molecular target in lung cancer, and EphA5 mutation was associated with the durable clinical benefit of immune checkpoint inhibitors in lung adenocarcinoma13,14. With immunotherapy proven to be effective in treating advanced lung cancer, it also faces the problems of limited efficiency and relapse15. Therefore, it is of great interest to investigate the possible influence of EphAs on immunotherapy.

In the present study, we investigated the expression and mutations of different EphA family members in NSCLC by bioinformatics to illustrate the expression patterns, prognostic values, and potential functions of EphAs. EphA10 was found to be aberrantly overexpressed and predicted worse clinical prognosis. EphA10 was further analysed and demonstrated a possible correlation with tumour immune infiltrates, which would likely be related to a better immunotherapy response.

Results

Expression of EphAs in Patients with NSCLC

Nine members of the EphA family have been found thus far. The expression levels of EphAs between NSCLC samples and normal samples in the TCGA database were compared (Fig. 1A,B). The results indicated that the expression levels of EphA1, EphA4, EphA5, EphA6, EphA7, EphA8, and EphA10 were significantly increased, whereas EphA3 was significantly downregulated in NSCLC tissues (Fig. 1C). No significant changes were observed in EphA2. The volcano map showed that EphA6, EphA7, EphA8, and EphA10 were significantly upregulated in NSCLC (fold change > 2, p < 0.05). The correlations of EphAs with each other were also analysed by their mRNA expression in TCGA database using Spearman correction. The results showed significant and positive correlations in the following EphAs: EphA1 with EphA2, EphA6 with EphA7, while EphA10 was negatively correlated with EphA7 (Fig. 1D). To further confirm these results, we used the GEPIA (Gene Expression Profiling Interactive Analysis) dataset (http://gepia.cancer-pku.cn/) to compare the mRNA expression of EphA between two different subtypes of NSCLC, lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC), and normal lung tissues. Only EphA10 showed a significantly differential expression in LUAD, while other members showed similar tendencies in both subtypes without prominent differences, and we continued our research in the NSCLC subtype (Fig. 2). Consequently, we concluded that EphA3 was significantly downregulated, while EphA6, EphA7, EphA8, and EphA10 were significantly upregulated in NSCLC tissues.

Figure 1
figure 1

The expression and correlation of EphAs in NSCLC (TCGA). (A) Heatmap of EphA expression in NSCLC and normal samples. The different colours represent the expression trend in different samples. (B) Volcano plot of EphAs expression in NSCLC and normal samples. The differentially expressed genes are labelled. (C) Box plot of EphA expression in NSCLC and normal samples. (D) Correlation between different EphAs in NSCLC. ***p < 0.001, **p < 0.01, *p < 0.05.

Full size image
Figure 2
figure 2

The expression of EphAs in NSCLC (GEPIA). (A) Box plot of EphAs expression in NSCLC. *p < 0.05. (B) Scatter diagram of EphAs expression in NSCLC.

Full size image

Association of EphAs mRNA expression with tumour stage and patient prognosis

The relationship between the expression of EphAs and NSCLC tumour stage was also investigated by GEPIA. Only the EphA1 groups significantly varied, whereas the other groups showed no significant difference (Fig. S1). The impact of EphAs on the survival of NSCLC patients was also characterized by using GEPIA tools. The Kaplan‒Meier curve and log-rank test analyses revealed that the increased EphA5 and EphA8 mRNA levels were significantly associated with better disease-free survival (DFS) (p < 0.05). NSCLC patients with higher levels of EphA2 were predicted to have lower overall survival (OS), whereas those with higher levels of EphA10 were predicted to have poorer DFS (p < 0.05) (Fig. 3).

Figure 3
figure 3

The prognostic value of the mRNA level of EphAs in NSCLC patients (GEPIA).

Full size image

Gene mutations of EphAs in NSCLC

Gene mutations might be a vital form of abnormal EphA expression. We analysed the EphA alterations by using the cBioPortal online tool for LUAD and LUSC (TCGA, PanCancer Atlas; http://www.cbioportal.org). EphAs were altered in 434 samples of 1053 patients with NSCLC (41%) (Fig. 4A). Then, the frequency of EphA mutations in NSCLC samples was also investigated. The results revealed that the mutation frequencies of EphA3, EphA5, and EphA6 were high (12%, 13%, and 12%, respectively), while those of EphA1, EphA2, EphA4, EphA7, EphA8, and EphA10 were relatively lower (less than or equal to 6%) (Fig. 4A). The lollipop graph revealed that the frequently detected mutations in EphAs included synonymous substitutions, missense substitutions, and nonsense substitutions. Among them, the most frequent mutation type in EphAs was missense substitution (Fig. 4B).

Figure 4
figure 4

EphA gene mutation analysis in NSCLC (cBioPortal). (A) EphA gene mutation analysis in NSCLC from the cBioPortal database. (B) The schematic diagrams produced from the cBioPortal database show the mutation frequency of EphA family members.

Full size image

Functional and coexpression analyses of EphAs in NSCLC

The Database for Annotation, Visualization and Integrated Discovery (DAVID) (https://david.ncifcrf.gov/summary.jsp) was used to analyse the gene ontology (GO), including biological processes, cellular components, and molecular functions, of EphA family members. We found that GO:0007169 (transmembrane receptor protein tyrosine kinase signaling pathway), GO:0070372 (regulation of ERK1 and ERK2 cascade), GO:0033628 (regulation of cell adhesion mediated by integrin), and GO:0019838 (growth factor binding) were significantly regulated by EphAs alterations in NSCLC (Fig. 5A). These are well-known entries associated with tumour growth and metastasis. We also performed GO analyses with the WebGestalt tool for further verification. Among the most highly enriched functions in the biological process category, the ephrin receptor signaling pathway, regulation of ERK1 and ERK2 cascade, peptidyl-tyrosine phosphorylation, transmembrane receptor protein tyrosine kinase signaling pathway, and regulation of MAPK cascade were associated with the expression of EphAs. The receptor complex, neuron projection, integral component of plasma membrane, intrinsic component of plasma membrane, and neuron part were enriched in the cellular component category. In the molecular function categories, EphA family genes were primarily enriched in ephrin receptor activity, transmembrane receptor protein kinase activity, and transferase activity. Kyoto Encyclopedia of Genes and Genomes (KEGG)16,17,18 pathway analyses were also used to characterize EphAs and revealed that EphAs were associated with several pathways, including axon guidance, the Rap1 signaling pathway, the Ras signaling pathway, and MAPK signaling (Fig. 5B). We also explored the protein‒protein interaction with EphA family members. The results from GeneMANIA revealed that EphAs were primarily associated with EPHB6, EPHB1, EPHB4, EPHB2, EPHB3, AXL, TEK, ROS1, and so on (Fig. 5C). The NetworkAnalyst database was used to predict the possible protein regulatory network of EphAs (Fig. 5D), which presented valuable leads for further study.

Figure 5
figure 5

Functional and coexpression analyses of EphAs. (A) The functions of EphAs were predicted by Gene Ontology (GO) analysis with DAVID tools. GO enrichment analysis predicted the functional roles of target host genes based on three aspects, including biological processes, cellular components, and molecular functions. (B) WebGestalt tools were also used for GO enrichment and KEGG pathway analysis of EphAs. Bar plots of GO enrichment in cellular component terms, biological process terms, molecular function terms, and KEGG enriched terms for EphAs are shown above. (C) Protein‒protein interaction analysis of the EphA family by GeneMANIA. (D) Regulation network of the EphA family predicted by NetworkAnalyst.

Full size image

Identification of immune-related EphAs

Due to the increasingly widespread use of immunotherapy, we also tried to evaluate the role of EphAs in the tumour immunity of NSCLC. First, we evaluated the correlation between EphAs and the four indicators of ESTIMATE in NSCLC samples (Fig. 6A). Only EphA3 and EphA10 showed a higher ESTIMATE score and decreased tumour purity indicating possible relation with immune infiltration, while other EphAs showed opposite results. Then, the CIBERSORT algorithm was applied to calculate the abundances of 22 immune subtypes of leukocytes in each sample to show immune infiltration in NSCLC. The results are illustrated in the rainbow diagram which showed an abundant immune cell infiltration in NSCLC samples (Fig. 6B). With regard to the positive relationship with immune score, EphA3 and EphA10 were two potential influencing factors for immune infiltration. Since we have found that EphA3 was downregulated in NSCLC and unrelated to prognosis, it is unlikely that EphA3 can be a biomarker for NSCLC. We further confirmed the expression of EphA10 in various cancers in the TCGA database. EphA10 expression was increased in many types of cancer relative to normal control samples (Fig. 6D). Thus, EphA10 was considered for subsequent analyses. We therefore divided the 1006 NSCLC samples from TCGA database into two groups according to EphA10 expression for further research (Fig. 6C).

Figure 6
figure 6

EphA10 is an oncogene associated with immune infiltration. (A) Association between EphAs and ESTIMATE results. (B) Percentage abundance of tumour-infiltrating immune cells in each sample, with different colours and different types of immune cells. (C) NSCLC patients were divided into two groups according to EphA10 expression. (D) The expression of EphA10 in different cancers compared with normal tissues in TCGA database. ***p < 0.001, **p < 0.01, *p < 0.05.

Full size image

EphA10 is associated with immune infiltration and sensitivity to immunotherapy

Subsequently, CIBERSORT, ssGSEA, and ESTIMATE analyses were performed to further investigate the immunological functions related to EphA10 expression. Although CIBERSORT analysis did not demonstrate any significant difference in the proportion of CD8 T cells between the two groups (Fig. 7A), ssGSEA showed that the quantified expression of 20 immune cell subtypes (including activated B cells, activated CD8 T cells, activated dendritic cells, natural killer cells, regulatory T cells, macrophages, and monocytes) was significantly higher in Group 2 (Fig. 7B). These outcomes indicated that patients in Group 2 were inclined to have stronger immune infiltration than patients in Group 1. To further investigate the impact of EphA10 on immunotherapy sensitivity in NSCLC, we compared the expression of several signature targets of immunomodulatory drugs between the two groups. We revealed that most of the immunomodulatory targets (including PD1, CTLA4, CD80, and CD40) were expressed at significantly higher levels in the group with high EphA expression (Fig. 7C and E). Moreover, Group 2 with higher EphA10 expression showed significantly higher stromal, immune, and ESTIMATE scores as well as decreased tumour purity than Group 1 in the ESTIMATE analysis (Fig. 7D). All the above results suggested that higher EphA10 expression may indicate a better response to immunotherapy.

Figure 7
figure 7

Prediction of sensitivity of immune therapy between the two groups. Group 1 with relative low EphA10 expression and Group 2 with higher EphA10 expression. (A) Proportion of immune cells between the two groups. (B) Expression of immune cells between the two groups. (C) Comparison of immunomodulatory drug targets between the two groups. (D) Comparison of functional enrichment, stromal score, immune score, ESTIMATE score, and tumour purity. (E) Comparison of T-cell activation-related immunomodulatory drug targets between the two groups. ***p < 0.001, **p < 0.01, *p < 0.05.

Full size image

Validation of the oncogenic role of EphA10 in NSCLC

We also validated the oncogenic role of EphA10 in NSCLC by molecular experiments. The EphA10 mRNA levels in NSCLC and adjacent noncancerous lung tissues were measured by qRT‒PCR, and EphA10 expression was higher in NSCLC samples than in noncancerous tissues (Fig. 8A,B). The Immunohistochemistry staining from The Human Protein Atlas website (https://www.proteinatlas.org/) also showed a high level of EphA10 expression in NSCLC tissue (Fig. 8C). The mRNA and protein expression level of EphA10 was significantly reduced after transfection with two siRNAs against EphA10 (Fig. 8D–F). CCK-8 assays and clonogenic assays were performed to confirm that cell growth was significantly inhibited in EphA10 knockdown cells compared with control cells after transfection (Fig. 8G–J). Transwell assays further demonstrated that loss of EphA10 considerably suppressed the migration and invasion abilities of NSCLC cells (Fig. 8K,L). These data confirmed the crucial role of EphA10 in promoting NSCLC progression.

Figure 8
figure 8

Validation of the oncogenic role of EphA10 in NSCLC. (A) and (B) Expression of EphA10 in tissue samples by qRT-PCR. (C) Expression of EphA10 in normal lung tissue and NSCLC tissue samples by IHC staining. (D) and (E) The mRNA level of EphA10 was downregulated in EphA10-knockdown cells. (F) The protein level of EphA10 was downregulated in EphA10-knockdown cells. (G) and (H) CCK-8 assays were performed to analyse cell proliferation. (I) and (J) Clonogenic assays were performed to assess cell proliferation. (K) and (L) Transwell assays were performed to analyse cell motility. ***p < 0.001, **p < 0.01, *p < 0.05.

Full size image

Discussion

Although the abnormal expression of EphAs has been reported in different types of cancers, integrated bioinformatics analysis of EphAs in NSCLC has not been performed before. We investigated the expression profile, mutation status, and prognostic values of EphAs in NSCLC for the first time. We found that dysregulated EphA10 might be a new biomarker for NSCLC prognosis and immunotherapy response. Hopefully, this research provides a comprehensive point of view to interpret the roles of EphA family members in NSCLC.

EphA1 is the family member most homologous to EphA2, and they colocalize in lung vascular endothelial cells19. This is consistent with our findings that EphA1 and EphA2 were both highly coexpressed in NSCLC. Although little has been studied about EphA1 in lung cancer, it has been reported to mediate cell adhesion and thus promote tumour lung metastasis9, while another study claimed that EphA1 can play an anticancer role20. According to our findings, EphA1 was upregulated in NSCLC, and its expression was significantly related to tumour stage, suggesting that EphA1 could be a biomarker for advanced NSCLC. However, the precise function of EphA1 in NSCLC needs further validation.

EphA2 is the most studied member of the EphA family and is a key regulator of tumorigenesis. While EphA2 can play complicated roles in multiple human malignancies, it mainly acts as an oncogene in lung cancer4,12. It was reported that EphA2 promotes tumour cell proliferation, migration and invasion, as well as cancer stem cell-like properties in NSCLC21. The expression of EphA2 in NSCLC is associated with K-RAS mutations, epidermal growth factor receptor (EGFR) activation, smoking history, and poor prognosis22. Moreover, inhibition of EphA2 promotes cell apoptosis and reverses acquired resistance to EGFR tyrosine kinase inhibitors (TKIs)23,24, making EphA2 a promising therapeutic target for NSCLC. Although we did not find significant differential expression of EphA2 in the current study, we confirmed that a higher level of EphA2 was related to poor survival.

EphA3 and EphA4 are considered tumour suppressors in lung cancer25,26. EphA3 was also one of the most commonly mutated genes in lung cancer. Interestingly, somatic EphA3 mutations can reverse the tumour-suppressive effects of wild-type EphA327. Consistent with a previous study, we found low expression and a high mutation frequency of EphA3 in NSCLC. We also revealed that EphA3 might be associated with immune infiltration, which might need further research for clarification. On the other hand, EphA4 inhibited cell migration and invasion and predicted favourable prognosis, while it was overexpressed in LUAD at the same time28. In our study, EphA4 was significantly more highly expressed in NSCLC than in normal samples. Since EphA4 can also play an oncogenic role in other types of cancer29, the function and molecular system of EphA4 might be complicated and remain to be discovered.

EphA5 is another important factor of the EphA family. EphA5 acted as an oncogene and enhanced malignant phenotypes in lung cancer30. EphA5 is also related to many biological processes, such as cytotoxicity, the cell cycle, and DNA damage repair10,13. It can promote cell stemness by activating the Wnt signaling pathway31. Moreover, mutations of EphA5 were found to predict better survival for LUAD patients, implying the potential of EphA5 to serve as a prognostic marker for LUAD patients after immune checkpoint blockade therapy14,32. All the above factors make EphA5 a candidate therapeutic target for lung cancer13. Our data were in accordance with previous studies. We found that EphA5 was upregulated in NSCLC tissues and exhibited a high mutation rate, while high expression of EphA5 indicated longer DFS.

Although little is known about EphA6 in lung cancer, it showed a strong correlation with EphA7 in our results. EphA7 was found to be highly expressed in different cancers, including lung cancer. EphA7 is negatively regulated by miRNAs such as miR-944 and miR-44811,33. Downregulation of EphA7 can suppress cell proliferation and motility by inhibiting EMT and PI3K/AKT signaling pathways11,34. Something more intriguing is that the secreted form of EphA7 was predominantly detected in any subtype of lung cancer other than LUAD, which made it a possible biomarker to distinguish different subtypes of lung cancer35. Consistently, we found that EphA7 was expressed at higher levels in NSCLC, especially in LUSC tissues. EphA8 is also a proven oncogene in glioma, breast cancer, gastric cancer, etc. and is related to poor prognosis in breast cancer and oral tongue squamous cell carcinoma36,37,38,39. In contrast, our findings showed the opposite: high EphA8 predicted longer DFS in NSCLC.

Last but not least, the study further focused on EphA10 because our results regarding EphA10 revealed something meaningful. EphA10, as one of the two pseudokinases of the Eph receptor family, is a catalytically defective RTK poorly characterized in NSCLC40,41,42. Previous studies have recognized the oncogenic role of EphA10 in multiple tumours, especially breast cancer41,43,44. EphA10 is highly expressed in breast cancer and promotes cell proliferation and migration45. It was also associated with clinicopathological characteristics, lymph node metastasis, stage progression, and poor prognosis in breast cancer patients46. Given its lack of kinase activity, its biological function was suggested to be realized by interacting with other catalytically active components, such as EphA745,47. Consistently, our findings also revealed that EphA10 was significantly upregulated in NSCLC and related to poor prognosis, indicating that EphA10 might be of great importance in the tumorigenesis and progression of NSCLC. However, the correlation analysis did not show a connection between EphA7 and EphA10. These results indicated that the biochemical mechanisms underlying EphA10 activation were different in NSCLC and warrant more thorough research. According to recent research, pseudokinases such as EphA10 could also serve as mediators in recruiting other proteins for downstream signaling48, and this may be a new lead in exploring the functional role of EphA10 in NSCLC.

More importantly, we found that EphA10 was significantly related to the immune therapy response. In the ESTIMATE analysis, EphA10 was significantly related to higher stromal, immune, and ESTIMATE scores. Tumour-infiltrating immune cells are closely related to the occurrence, development, and outcomes of tumours49,50. In ssGSEA, most of the immune cell subtypes showed significantly higher expression in the high-EphA10 expression group, and patients with a high degree of immune cell filtration were more likely to have a better immune response51. Subsequently, immune checkpoint therapy targets, including PD-L1, were also highly expressed in Group 2, implying that patients with high EphA10 levels may benefit more from immunotherapy. In accordance with our results, the immunomodulatory role of EphA10 has been addressed in breast cancer. It was reported that EphA10 expression was positively related to PD-L1 and that targeting EphA10 suppressed tumour regression, therapeutic response rate, and T-cell-mediated antitumor immunity in breast cancer52. Since our previous studies indicated that RTKs can be involved in the regulation of PD-L153,54, and EphA10 is a member of RTKs, EphA10 may regulate immune infiltration by regulating immune checkpoint targets. Although these results indicated that patients with high EphA10 levels might benefit from immune therapy, current data still showed shorter DFS in high-EphA10 patients, because some of the patients might have been diagnosed at a late stage or did not receive standard treatment. Our molecular experiments proved that EphA10 promoted cell proliferation and motility in NSCLC cells which could also lead to poor prognosis. Taken together, these findings suggested that EphA10 plays a critical role in the tumorigenesis of NSCLC, and targeting EphA10 can be a promising new strategy for strengthening immunotherapy in EphA10-positive patients.

Materials and methods

The cancer genome atlas (TCGA)

The Cancer Genome Atlas databases contain sequencing and pathological data of over 30 different types of cancers55. R software with TCGAbiolinks package56 was used to download transcriptome profiling data, including HTSeq-Counts and HTSeq-FPKM, as well as clinical information from TCGA databases. All the included cases had complete clinical information (including age, sex, TNM stage, and prognostic information). HTSeq-FPKM of 1113 NSCLC and normal tissue samples were used for further analyses after log2(FPKM + 1) transformation, while HTSeq-Counts were used for differential expression analysis.

Differentially expressed genes

HTSeq-Counts, the expression profiling data of TCGA cases, were analysed to screen out the differentially expressed genes by the DESeq2 package57. The differential expression threshold was |log2FoldChange|> 1, and a p value < 0.05 was considered statistically significant.

Gene expression profiling interactive analysis (GEPIA)

GEPIA (http://gepia.cancer-pku.cn/) is an innovative internet platform providing customized analyses of gene expression in multiple cancers, including differential expression analysis, correlation analysis, and survival analysis, using the RNA sequencing data of 9736 tumours and 8587 normal samples from TCGA databases and Genotype-issue Expression (GTEx) projects58.

cBioPortal database analysis

cBioPortal (http://www.cbioportal.org/) is a developed interactive website that provides comprehensive analyses of complex tumour genomics and clinical profile data from the TCGA database belonging to the Memorial Sloan Kettering Cancer Center for Cancer Genomics59,60. The frequency and forms of EphA gene mutations were investigated using lung adenocarcinoma (TCGA, Nature 2014) and lung squamous cell carcinoma (TCGA, Nature 2012) datasets. The gene mutations, putative copy number alterations (CNAs) from genomic identification of significant targets in cancer (GISTIC), mRNA expression Z scores (RNA-seq v.2 RSEM), and protein expression Z scores (reversed-phase protein array [RPPA]) were all included in the genomic profiles.

Functional analyses

The Database for Annotation, Visualization, and Integrated Discovery (DAVID) was used for Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG)16,17,18 analyses for the functions of the EphA family61,62. Gene ontology analyses mainly contain three aspects: biological processes (BP), cellular components (CC), and molecular functions (MF).

WebGestalt (https://www.webgestalt.org/)63 is also a functional enrichment analysis website applicable for GO and KEGG analyses of EphAs. The FDR method is BH. Terms with a p value of < 0.05 were considered significant.

Protein‒protein interaction analyses

GeneMANIA (http://genemania.org/)64 was applied to provide information on protein and genetic interactions, pathways, coexpression, colocalization, and protein domain similarity of submitted genes.

NetworkAnalyst (https://dev.networkanalyst.ca/NetworkAnalyst)65 is an analytics platform for visualizing comprehensive gene expression profiling and meta-analysis. It was applied to investigate the protein regulatory network of EphAs.

Immune infiltration analyses

ESTIMATE, a method that determines the fractions of stromal and immune cells based on gene expression signatures in tumour samples, was used to investigate the tumour microenvironment (TME) of NSCLC patients from the TCGA database. The R package estimate66 was applied to investigate the stromal score (stromal content), immune score (extent of immune cell infiltration), ESTIMATE score (synthetic mark of stroma and immune), and tumour purity of each cluster. CIBERSORT is a method that calculates cell composition based on the expression profiles. This deconvolution algorithm was applied to calculate the proportion of 22 immune cells in each patient67. The combination of 22 immune cell type proportions in each sample was considered 100%. Moreover, single-sample gene set enrichment analysis (ssGSEA) was applied to investigate the infiltration extent of 28 immune cell types according to the expression levels of genes in 28 published immune cell gene sets68 by the R package GSVA69.

Specimen collection and quantitative real-time PCR assays

11 NSCLC tissues and 11 noncancerous tissues were gathered from the First Affiliated Hospital of Soochow University with informed consent and approval from the Ethics Committee of the First Affiliated Hospital of Soochow University. Total RNA was extracted with TRIzol reagent (Invitrogen, USA). The specific primers for target genes were EphA10 forward 5’-TCCTCCTGGATTCCAAAGCCTC-3’, reverse 5’-CGGGGACACCTTGTAAAACC-3’; β-actin forward 5’-CACAGAGCCTCGCCTTTGCC-3’, reverse 5’-ACCCATGCCCACCATCACG-3’. Gene expression levels were quantified according to the comparative ΔΔCt method, and β-actin was used as the internal control.

Ethical approval

The study was approved by the Ethics Committee of the First Affiliated Hospital of Soochow University (2022-276) in accordance with the ethical standards of the Declaration of Helsinki. Informed written consent was obtained from all participants. Confidentiality of patient data was guaranteed.

RNA interference

Predesigned short interfering RNA (siRNA) sequences that target different coding regions of EphA10 were directly synthesized by GenePharma (Suzhou, China). The target sequences were si-EphA10-1 sense 5’-GCAAUGUGCUGGAGCCCAATT-3’, anti-sense 5’-UUGGGCUCCAGCACAUUGCTT-3’; si-EphA10-2 sense 5’-GGAAGUUAUCCUCCUGGAUTT-3’, anti-sense 5’-AUCCAGGAGGAUAACUUCCTT-3’. Scrambled siRNA was used as a negative control. Cells were transiently transfected with 100 pmol of siRNA sequences using Lipofectamine 2000 (Invitrogen, USA). After 48 h of transfection, the cells were harvested for further experiments.

Western blot analysis

Western blot analysis was performed as previously described3. The primary antibodies used in this study were anti-EPHA10 (DF4044) (Affinity, China), and anti-β-actin (#3700) (Cell Signaling Technology, USA). The bands were developed by an electrochemiluminescence reagent, imaged with a ChemiDoc XRS + (Bio-Rad, USA), and finally quantified with ImageJ software (National Institutes of Health, Bethesda, MD, USA). We exhibited the individual bands instead of the whole membrane because they were cut before hybridization to be hybridized with different antibodies.

Cell viability assay

Cell proliferation ability was examined according to the manufacturer’s instructions by a Cell Counting Kit-8 (CCK-8) (Beyotime, Shanghai, China). Clonogenic assays were performed as previously described3. The experiment was performed in triplicate.

Migration and invasion assays

Transwell assays were performed as described previously3. The experiment was performed in triplicate.

Statistical analysis

R (4.1.3) software was applied for statistical analyses. The Wilcoxon rank-sum test was applied for box plot analyses. Pearson’s coefficient was used for correlation analysis. Survival curves were formed by the Kaplan–Meier method (log-rank test). Statistical comparisons were calculated using Student’s t tests, and a p value < 0.05 was considered significant.

Conclusions

In conclusion, we systemically elucidated the expression patterns and prognostic values of EphAs in NSCLC. Among them, EphA10 was significantly differentially expressed and related to the prognosis of NSCLC. Furthermore, EphA10 expression was positively related to immune infiltration and levels of immune checkpoint genes. Our findings suggested that EphA10 can be a promising biomarker for NSCLC prognosis and the status of tumour immunity as well as the great potential in further exploring the biological mechanism and clinical value of EphA10.