Introduction

Head and neck cancer is the 7 th most prevalent malignancy worldwide, with approximately 650,000 new cases and 320,000 deaths annually1. As the most widespread type, head and neck squamous cell carcinomas (HNSCCs) which mainly arise from mucosal epithelium in the larynx, pharynx and oral cavity account for approximately 90% of head and neck cancer1. Tobacco use, alcohol consumption, and human papillomavirus (HPV) infection all play a crucial carcinogenic role in HNSCC2. Generally, HNSCCs are characterized by local recurrence, lymph node and distant metastasis, which lead to poor prognosis and unsatisfactory quality life3,4,5. The 5-year overall survival (OS) of patients in advanced stage with HNSCC is only nearly 50%3. In addition, Chemo- and immunotherapy resistance are also the main reasons for low OS rate6,7,8. Hence, it is vital to reveal the mechanism of HNSCC tumorigenesis and progression for improving the prognosis of patients with HNSCC.

RAS related 2 (RRAS2), a transforming GTPase, shares downstream effectors with Ras subfamily proteins, and associates with the plasma membrane and may function as a signal transducer9. RASS2 plays a crucial role in activating signal transduction pathways that control cell proliferation. Mutations of RRAS2 are related to the growth of various tumors10,11. Recently, increasing studies suggested the role of RRAS2 in tumorigenesis. Romain M Larive et al. demonstrated that RAAS2 was associated with primary breast tumorigenesis and the advanced lung metastatic steps of cancer cells. RRAS2 facilitated tumorigenesis of breast cancer in a phosphatidylinostiol-3 kinase (PI3 K)-dependent and signaling autonomous manner12. In oral cancer, RRAS2 was identified to be overexpressed in tumor tissues, and associated with unsatisfactory prognosis of oral squamous cell carcinomas13. Meanwhile, Muzafar A. Macha et al. showed an interaction between RRAS2 and ERK2, PI3 K13. However, fewer studies have investigated the carcinogenic role and therapy resistance of RRAS2 in HNSCCs.

In the present study, HNSCC RNA expression files were obtained from TCGA (https://portal.gdc.cancer.gov/) to investigate the carcinogenic role and prognosis of RRAS2. Meanwhile, data from GEO (https://www.ncbi.nlm.nih.gov/geo) and HPA databases (https://www.proteinatlas.org/) were used to identify RRAS2 expression in HNSCC. Furthermore, we revealed the crucial role of RRAS2 in immune microenvironment and drug therapy. On the other hand, we identified the upstream signal and various biological experiments were conducted to explore the carcinogenic role of RRAS2 in HNSCC. Finally, gene set enrichment analysis (GSEA) was used to predict the candidate KEGG pathways influenced by RRAS2.

Tumor microenvironment analysis

Immune cell and function scores of HNSCC patients were obtained via single sample gene set enrichment analysis (ssGSEA) in R software (Version 4.2.2, https://www.r-project.org/). Subsequently, the immune scores difference analysis between RRAS high- and low-expression groups was conducted. In addition, bioinformatics analysis was performed to investigate the correlation between RRAS expression and immune cell levels. On the other hand, Tumor mutation burden (TMB) is visualized as an indicator for immunotherapeutic response. Therefore, we obtained the TMB(TMB) data from UCSC Xena database (http://xena.ucsc.edu/), following exploration the correlation between TMB level and RRAS2 expression.

Methods and materials

Data acquisition and bioinformatics analysis

The RRAS2 expression of pan-cancer was obtained from GEPIA2 database (http://gepia2.cancer-pku.cn/), and the prognostic value of RRAS2 in various cancers was further explored. The HNSC gene expression profiles and corresponding clinical data were downloaded from TCGA database (https://portal.gdc.cancer.gov), following extraction of RRAS2 expression level. Furthermore, we investigated the role of RRAS2 in HNSCC prognosis through survival and Cox regression analysis. On the other hand, the RNA and protein expression profile were further validated in GEO (https://www.ncbi.nlm.nih.gov/geo) and HPA database (https://www.proteinatlas.org/) respectively.

Drug susceptibility analysis

Drug susceptibility data were downloaded from the CellMiner database (https://discover.nci.nih.gov/cellminer/home.do). Furthermore, the RRAS2 expression level and corresponding drug susceptibility were extracted, following investigation of the correlation between RRAS2 expression and drug therapy resistance.

Cell culture and transfection

HNSCC cell lines SCC9 and SCC25 were obtained from ATCC, and all cell lines were seeded in Dulbecco’s modified Eagle’s medium (DMEM) with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin at 37 °C with 5% CO2. Small interference RNA (siRNA) targeting RRAS2 and its corresponding negative control were designed and synthesized by TINGKE Biotech Co., Ltd. (Chengdu, China). Cells were cultured in 6/12-well plate, following transfection of RRAS2 siNRA and its corresponding negative control into cells according to the instruction of the lipofectamine 3000 regents when the density reached approximately 80%. Subsequently, transfected cells were collected for further analysis. The siRNA sequence was listed (Table 1).

Table 1 The SiRNA and primers sequence used in this study.
Full size table

CCK8 assay

Briefly, approximately 2 × 103 cells/well were seeded in 96-well plate. About 12 h later, the cell number baseline was determined. Subsequently, the cell proliferation viability was evaluated by Cell Counting Kit (Vazyme) at 24, 48 and 72 h, respectively.

Cell migration and invasion assay

Approximately 5 × 104 cells per well were cultured in the upper chamber of 24-well plate with 200 µl serum-free medium, and 600 µl completed medium was placed in the bottom well. After 24–48 h, the positive migrated cells were counted by staining with crystal violet. The cell invasion ability assay was done with 100 µl 1:8 diluted Matrigel in the bottom of the upper chamber, and the positive invasive cells were analyzed after 24–48 h.

Cell wound healing assay

Transfected cells were seeded in 6-well. When cell density reached approximately 100% to form monolayer cells, a linear wound was generated across the cell by a sterile 200 µl pipette tip, following which the medium was exchanged the completed medium to serum-free medium. The wound healing images were recorded at 0–24 h, respectively.

RNA isolation and qRT-PCR

The total RNA was extracted according to the TRIZOL manufacturer’s instructions, and the same amounts of RNAs were reversed transcribed to cDNA. Furthermore, the relative expression of targeted genes to GAPDH was evaluated by quantitative real-time polymerase chain reaction (qRT-PCR). The primer sequences were listed (Table 1).

Western blotting assay

Proteins were isolated via RIPA lysis buffer, and degenerated at 95 ℃ for 15 min. Subsequently, samples were separated by PAGE gels, and further transferred into PVDF membranes, following incubation in 5% skim milk at room temperature for 1 h. Then, PVDF membranes were washed by PBS for 3 times, and incubated in primary antibodies at 4 ℃ overnight. Finally, the membranes were incubated in secondary antibodies at room temperature for 1 h and imaged by ECL reagents. The RRAS2 (Cat# 12530-1-AP) and Actin (Cat# 66009-1-Ig) antibodies were obtained from Proteintech official website.

Gene set enrichment analysis (GSEA)

GSEA was used to evaluate the KEGG pathways mediated by RRAS2 according to Kanehisa M’s studies14,15,16. In briefly, HNSCC cases in TCGA were divided into “high (H)” and “low (L)” groups in line with median expression of RRAS2. Furthermore, group files and gene expression profiles were imported into the GSEA_4.1.0 software (Version 4.1.0, https://www.gsea-msigdb.org/gsea/index.jsp). The gene-set database chose KEGG with the L group as the control group.

Statistical analysis

SPSS software (IBM, Version 26.0, https://www.ibm.com/spss) was used to conducted difference analysis. Student’s t test, one/two-way ANOVA, and long rank test were used. p < 0.05 *, p < 0.01 **, p < 0.001***, p < 0.0001****.

Results

RASS2 was overexpressed in various cancers and correlated with HNSCC prognosis

To investigate the role of RRAS2 in HNSCC, we firstly identified its expression in various cancers. We suggested overexpression of RRAS2 in DLBC (Lymphoid Neoplasm Diffuse Large B-cell Lymphoma), GBM (Glioblastoma multiforme), HNSC (Head and Neck squamous cell carcinoma), LGG (Brain Lower Grade Glioma), PAAD (Pancreatic adenocarcinoma), and THYM (Thymoma) (Fig. 1A). In addition, high expression of RRAS2 was correlated with ACC (Adrenocortical carcinoma), LIHC (Liver hepatocellular carcinoma), LUAD (Lung adenocarcinoma), PAAD, and SARC (Sarcomav) poor prognosis (Figure S1). We further identified that RRAS2 mRNA level was upregulated in HNSCC tumor cases in GEO database (Fig. 1B). Meanwhile, RRAS2 protein level was significantly elevated in tumor tissues, and correlated with lymph node metastasis (Fig. 1C). Moreover, high RRAS2 expression showed unsatisfactory prognosis of HNSCC patients (Fig. 1D). On the other hand, RRAS2 was identified as a candidate independent prognostic biomarker for HNSCC through univariate and multivariate Cox regression analysis (Fig. 1E, F). These findings suggested that RRSA2 was overexpressed and might be a potential biomarker for HNSCC.

Fig. 1
figure 1

RASS2 was overexpressed in various cancers and correlated with HNSCC prognosis. (A) RRAS2 expression in various cancer types, suggesting that upregulation of RRAS2 could be seen in DLBC, GBM, HNSC, LGG, PAAD, THYM. (B) RRAS2 expression in GEO database, suggesting that RRAS2 was overexpressed in HNSCC tumor cases (The one-way ANOVA test was used). (C) RRAS2 protein level in HSCC tissues, suggesting that high RRAS2 expression might be associated with HNSCC metastasis. (D) Survival analysis indicated the lower overall survival rate of HNSCC patients in RRAS2 high expression group. E; F. Univariate and multivariate Cox regression analysis according to RRAS2 expression and clinical information including Age, Grade, Stage, Tumor size and Lymph node metastasis, showing that RRAS2 might an independent prognostic biomarker for HNSCC.

Full size image

RRAS2 was associated with immune infiltration of HNSCC

To further investigate the role of RRAS2 in HNSCC, we obtained the immune cell and function scores of HNSCC patients, and correlation analysis between RRAS2 expression and immune scores. RRAS2 overexpression might inhibit the function of various immune cells, especially TILs, NK cell, B cell, C8+ T cell, etc. (Fig. 2A). In addition, RRAS2 might impair the immune process via suppressing the function of antigen presenting cells, such as aDCs and DCs (Fig. 2A). Moreover, RRSA2 contributed to the inactivation of Cytolytic activity and Type II IFN response (Fig. 2B). On the other hand, high expression of RRAS2 was negatively related to naïve B cell, C8+ T cell and T cells follicular helper level (Fig. 2C, D, E). Meanwhile, RRAS2 was positively associated with TMB level of HNSCC patients (Fig. 2F), which is an indicator for immunotherapy response. These observations suggested that RRAS2 might induce the immune escape of tumor cell, and lead to HNSCC tumorigenesis, progression and immunotherapy resistance.

Fig. 2
figure 2

RRAS2 was associated with immune infiltration of HNSCC. (A) RRAS2 expression was negatively related to various cell scores, especially B cells, CD8 + T cells, NK cells and etc. (B) High expression of RRAS2 was associated with unsatisfactory immune response, especially Type-II IFN response. C; D; E. The correlation between RRAS2 expression and B naive cells, CD8 + T cells and T cells follicular helper. F. High expression of RRAS2 meant increased tumor mutation burden (TMB), suggesting that RRAS2 might induced the immunotherapy resistance of HNSCC.

Full size image

Upregulated RRAS2 contributed to drug resistance

To investigate the role of RRAS2 in drug resistance, we obtained drug susceptibility data from Cellminer database and then conducted the correlation between RRAS2 expression and response to drug therapy. The results showed that overexpression of RRAS2 was related to unsatisfied response to main chemotherapy drug, such as fluorouracil and paclitaxel (Fig. 3A, B). RRAS2 also showed inhibitory role in vincristine and isotretinoin, which are prevalent drugs for various cancers (Fig. 3C, D). On the other hand, RRAS2 was correlated with the remarkably poor response of tamoxifen and raloxifene (Fig. 3E, F). These findings suggested that RRAS2 might induce the chemotherapy resistance of HNSCC cases.

Fig. 3
figure 3

Upregulated RRAS2 contributed to drug resistance. A; B. Overexpression of RRAS2 showed inhibitory fluorouracil and paclitaxel susceptibility. C; D; E; F. RRAS2 expression was negatively associated with various chemotherapy drugs.

Full size image

Inhibition of RRAS2 weakened HNSCC cell migration and invasion abilities

To further determine the biological function of RRAS2 in HNSCC, we transfected siRNA targeting to RRAS2 in HNSCC cell lines (Fig. 4A, B; Figure S2), and then detected the transformation of cancer cell phenotype. Inhibition of RRAS2 expression showed little effect on cell proliferation of OSCC cells (Fig. 4C). Therefore, we further explored the role of RRAS2 in OSCC invasion. We suggested that Knockdown of RRAS2 inhibited migrated ability of cancer cells (Fig. 4D, E). In addition, Knockdown of RRAS2 suppressed HNSCC invasion in vitro (Fig. 4F, G). On the other hand, scratch healing assay also showed inhibited invasive ability in RRAS2 knockdown cells (Fig. 4H, I). Genetically, inhibition of RRAS2 decreased N-cad, Vimentin, and elevated E-cad expression in HNSCC cells (Fig. 4J, K). These observations showed that RRAS2 might induce the epithelial-mesenchymal transition (EMT) of HNSCC.

Fig. 4
figure 4

Inhibition of RRAS2 weakened HNSCC cell migration and invasion abilities. A; B. The efficiency of siRNA targeting to RRAS2 was identified by Q-PCR and western blotting. The WB images were cropped, and the original images were presented. C. Inhibition of RRAS2 had no effect on SCC25 and HN12 cell proliferation. D; E. Knockdown of RRAS2 inhibited HN12 and SCC15 migration. F; G. Knockdown of RRAS2 inhibited HN12 and SCC15 invasion. H; I. Scratch healing assay suggested that inhibition of RRAS2 weakened HNSCC cells invasion. J; K. Knockdown of RRAS2 downregulated N-cad and Vimentin, while upregulated E-cad expression.

Full size image

RRAS2 expression was induced by H3K18la

We further investigated the regulation of RRAS2 expression in HNSCC. According to the previous study, RRAS2 expression was mediated by H3 lysine 18 lactylation (H3K18la)17. However, whether H3K18la induced RRAS2 expression in HNSCC remains unclear. Hence, we conducted correlation analysis between histone lactylation “writer” EP300 and RRAS2 expression, which suggested that EP300 was positively related to RRAS2 expression (Fig. 5A). Furthermore, we showed that lactate-accumulation-associated genes including HIF1 A, LDH1 and LDHB might induce RRAS2 expression (Fig. 5B). On the other hand, we demonstrated that lactate indeed induced the expression of RRAS2 in HNSCC cell lines (Fig. 5C). Meanwhile, inhibition of H3K18la suppressed RRAS2 expression (Fig. 5D). In addition, activation of H3K18la by extraneous lactate promoted RRAS2 protein expression, and inactivation of H3K18la by 2-DG, an inhibitor of aerobic glycolysis, inhibited RRAS2 expression (Fig. 5E; Figure S2). Moreover, we suggested that lactate could reversed the inhibited role of 2-DG in RRAS2 expression (Fig. 5F). These results showed that RRAS2 was mediated by H3K18la in HNSCC.

Fig. 5
figure 5

RRAS2 expression was induced by H3K18la. (A) RRAS2 expression was positively associated with lactylation “writer” EP300. (B) High expression of lactate-production genes might induce the expression of RRAS2. (C) Sodium lactate induced the mRNA expression of RRAS2. (D) 2-DG, an inhibitor of H3K18la, inhibited RRAS2 mRNA expression of HNSCC cells. (E) Sodium lactate and 2-DG increased and decreased the RRAS2 protein expression respectively. The WB images were cropped, and the original images were presented. (F) 2-DG inhibited the expression of RRAS2, while lactate could rescue the inhibited role of 2-DG.

Full size image

Knockdown of RRAS2 impaired the H3K18la activation induced migration and invasion

We further identify the H3K18la-induced RRAS2 carcinogenic role. We suggested that activation of H3K18la facilitated HNSCC migration and invasion, while knockdown of RRAS2 reversed the promoted role of lactate (Fig. 6A-D). Scratch healing assay also showed that the carcinogenic role of H3K18la was in a RRAS2 dependent manner (Fig. 6E, F). In addition, High H3K18la level induced the expression of N-cad, Vimentin, while knockdown of RRAS2 reversed the role of H3K18la agonist (Fig. 6G, H). These observations suggested that RRAS2 might be a crucial target for the carcinogenic role of histone lactylation in HNSCC progression.

Fig. 6
figure 6

Knockdown of RRAS2 impaired the H3K18la activation induced migration and invasion. A; B. Activation of H3K18la by sodium lactate enhanced HNSCC migration and invasion, while inhibition of RRAS2 impaired the facilitated role of H3K18la. C; D. Statistical analysis of migration and invasion assay. E. Scratch healing assay indicated that the promoted role of H3K18la in HNSCC progression was retarded by silence of RRAS2. F. Statistical analysis of scratch healing assay. G; H. Sodium lactate stimulation increased the expression of N-cad, Vimentin, and decreased E-cad. However, knockdown of RRAS2 reversed the role of Sodium lactate.

Full size image

Overexpression of RRAS2 was related to HNSCC-associated pathways

To predict the potential mechanism, we carried out GSEA to select KEGG pathways mediated by RRAS2. As shown, RRAS2 contributed to the activation of the KEGG-focal adhesion and KEGG-WNT-signaling-pathway, which plays a crucial role in HNSCC (Fig. 7A, B). In addition, RRAS2 might decreased the activation of drug metabolism and primary immunodeficiency, suggesting that RRAS2 plays a crucial role in drug and immunotherapy resistance(Fig. 7C, D), and targeting histone Kla might impair the activation of these pathways induced by RRAS2, thereby inhibiting HCC development and elevating the therapy response.

Fig. 7
figure 7

RRAS2 contributed to the activation of cancer-related KEGG pathways and undesirable therapy response. A; B. RRAS2 played a crucial role in focal adhesion and WNT signaling pathway, which were associated with HNSCC initiation and progression. C; D. High expression of RRAS2 indicated poor primary immunodeficiency and drug metabolism. The KEGG pathways were obtained from GSEA software and official website (www.kegg.jp/kegg/kegg1.html).

Full size image

Discussion

The pathogenesis of HNSCC is remarkably complex. The patients with HNSCC develop rapidly, exhibit high aggressiveness, and respond poorly to existing treatments including immunotherapy, leading to unsatisfactory survival rates and prognosis. Hence, it is vital to investigate the effective biomarkers and therapeutic targets to improve the overall survival of HNSCC. RRAS2 was related to the plasma membrane and might function as a signal transducer, which plays an important role in activating signal transduction pathways that regulate cell proliferation. Mutations of RRAS2 are relevant to the growth of various tumors9. For instance, Claudia Cifuentes et al. suggested that overexpression of RRAS2 in mice is sufficient to develop triple negative breast cancer (TNBC). In human breast cancer, RRAS2 is overexpressed in 68% of tumors across grade, location, and molecular type, surpassing the prevalence of any previously implicated alteration. RRAS2 upregulation remarkably higher and more frequent in TNBC and young parous patients18. In addition, upregulation of RRAS2 could be seen in squamous cell carcinoma. Chengxi Sun found that knockdown of RRAS2 impaired the angiogenesis and colony and tumorsphere formation in cutaneous squamous cell carcinoma19. These findings clarified the crucial role of RRAS2 in oncogenesis. However, its role in HNSCC progression remains unclear.

In our study, we suggested that H3K18la-mediated RRAS2 was elevated in various cancers, including HNSCC. Overexpression of RRAS2 was associated with unsatisfied overall survival rate as well as metastasis, and might be served as an independent biomarker for HNSCC. The majority of studies elucidated the crucial role of RRAS2 in various cancers. In esophageal cancer cells, Raghibul Hasan et al. demonstrated that downregulation of RRAS2 could sensitize the chemotherapy to cis-platin20. We suggested that RRAS2 expression was positively related to the sensibility to Fluorouracil, one of the most common chemotherapy drugs for HNSCC21, suggesting that RRAS2 might a crucial target for chemotherapy of squamous cell carcinoma. We suggested that high expression of RRAS2 was negatively related to various immune cell sores and immune function scores, especially B cell, CD8 + T cell, NK cell and Type-II IFN response, respectively. Mendoza P et al. demonstrated that RRAS2 is required for germinal center formation to aid B cells during energetically demanding processes22, suggesting that RRAS2 might play a crucial role in immune system. However, there are fewer studies to explore the role of RRAS2 in immunotherapy.

On the other hand, we showed that RRAS2 expression was induced by histone lactylation. H3K18la-induced RRAS2 mediated the EMT of HNSCC. H3K18la is a newly identified histone posttranslation, and is sensitive to lactate levels in the tumor microenvironment (TME)17. Owing to the aerobic glycolysis, HNSCC is characterized by an amount of production of lactate23. The role of lactate not only plays a crucial role in tumor cells, but also influences surrounding cells in the TME, which promotes tumor cells to acclimate stromal cells and immune cells24. Tumor-derived lactate induces the histone lysine lactylation, and then stimulates gene expression and enhances or inhibits protein function25, thereby facilitating progression of cancers. This finding establishes a link between the epigenetic function of nuclear histones and glycolytic metabolism in the tumor cytoplasm. Recently, numerous studies have elucidated the crucial role of histone lactylation. For instance, Yu et al. suggested that lactate induced the lactylation of histone, and then drove oncogenesis of ocular melanoma by enhancing m6 A reader protein YTHDF2 expression26. Xie et al. demonstrated that inhibition of histone lactylation suppressed the proliferation and migration of tumor cells27. Huang et al. suggested that both overall lactylation level and H3K18la in OSCC tissues were increased. RRAS2 was a potential downstream target of the H3K18la, and RRAS2 was identified to be a potential prognostic biomarker for OSCC28. Huang et al. also showed that Histone Kla-induced BCAM enhanced oral squamous cell carcinoma (OSCC) progression and cis-platinum resistance28. Therefore, understanding of the H3K18la-RRAS2 axis might be helpful to provide promising chemotherapy strategies for clinical intervention of HNSCC. More studies should be conducted to investigate the mechanism of RRAS2 in HNSCC.

Conclusion

In conclusion, we demonstrated that RRAS2 mediated by H3K18la was upregulated in HNSCC, and associated with metastasis and poor prognosis. RRAS2 was identified to be associated with chemotherapy resistance and immune infiltration. In addition, we suggested that H3K18la-induced RRAS2 facilitated HNSCC migration, invasion and EMT. These observations contribute to a better understanding of the H3K18la-RRAS2 axis in HNSCC, providing promising therapy strategies for clinical intervention of HNSCC.

Limitations

Although this study may have some clinical importance in HNSCC, we also should consider several limitations. First, there is lack of detailed mechanistic data to support the RRAS2 in HNSCC. Second, RRAS2 was associated with immune infiltration in HNSCC. However, we failed to explore the role of RRAS2 in immune microenvironment. In addition, HPV infection was known to affect TMB and immune infiltration. However, we failed to explore the association between RRAS2 and HPV positive/negative status. On the other hand, the role of RRAS2 in vivo should be further explored. Therefore, more information needs to be harvested to investigate the detailed mechanism of RRAS2 in HNSCC in the further study.