Investigation on Potential Correlation Between the RNA-Binding Protein of Evolutionarily Conserved MEX3 Family and Non–Small-Cell Lung Cancer

As mRNA binding proteins, MEX3 (muscle excess 3) family highlights its unique characteristics and plays an emerging role in post-transcriptionally regulating programmed of biological processes, including tumor cell death and immunological relevance. These have been shown to be involved in various diseases, however, the role of MEX3 in non-small-cell lung cancer (NSCLC) has not been fully elucidated. In this study, we found that the sequence or copy number of MEX3 gene did not change signicantly, which can explain the stability of malignant tumor development through the COSMIC database. Further, gene expression in NSCLC was examined using the Oncomine™ database, and the prognostic value of each gene was analyzed by Kaplan-Meier analysis. The results showed that overexpressed of MEX3A, MEX3B, MEX3C and MEX3D were associated with signicantly lower OS in patients with NSCLC and LUAD, while overexpressed of MEX3D was associated with signicantly poorer OS in patients with LUSC. We also applied the Tumor Immune Estimation Resource (TIMER) tool to assess the correlations between distinct MEX3 and the inltrating immune cell landscape.

New Gene) domain 4,6 . This particular indicates that the MEX3 proteins play vital role in the balance between self-renewal and differentiation by mediating self-ubiquitination or ubiquitination of target protein and promoting RING-dependent degradation of HLA-A2 (human leukocyte antigen serotype A2) mRNA 6,7 . The ability of MEX3 proteins can not only interact with different RNA sequences, but also diversify the mechanisms enhanced by the RING domain, despite the increasing complexity of regulation, so far, little evidence support these redundant 8 . Consistent with the multi-pathway disease presented by cancer and the multiple roles of MEX3 in regulating gene expression, MEX3 is involved in multiple biological processes in the occurrence and development of cancer 9 . MEX3 mediates cancer cell proliferation, migration, tumor immune escape mechanism, and the transcription level changes in different cancer types [10][11][12] ; according to the tumor type and family members, the expression of MEX3 is related to the increase or decrease of patient survival rate.
However, there are few studies on the correlation between MEX3 family and lung cancer, and the conclusions are limited. In agreement, here, we draw attention to investigating on large sample of databases to explore MEX3 family expression, prognostic value, and immune-related effects in nonsmall-cell lung cancer(NSCLC),thereby providing further insights into tumor heterogeneity and targeted therapy.

Materials And Methods
Somatic Mutations of MEX3 in lung cancer.
We applied COSMIC database 13

MEX3 prognostic analysis
Kaplan-Meier plotter 15,16 were used to assess the prognostic relevance of MEX3 A-D speci cally expressed in NSCLC samples. In our study, Affymetrix Identity (Jetset best Probe 17 , as shown in Table I) identi ed available genes, used median gene expression values to divide patient samples into high and low expression group, 95% con dence interval (CI) calculated log-rank P value and hazard ratio (HR), "Array quality control" and "Exclude biased array " were selected to obtain numerical results through univariate Cox regression analysis. Finally, the valuable Kaplan-Meier survival curve (OS, Overall survival) was generated according to these parameters.

Immunological correlation of MEX3
The immune correlation of MEX3 was completed through the Tumor Immune Estimation Resource (TIMER 2.0, timer.comp-genomics.org/), a web server for comprehensive analysis of tumor in ltrating immune cells 18 , including The Cancer Genome Atlas (TCGA) of cancer genome maps from 32 cancer types, a total of 10,897 samples, six subsets of TIIC including B cells, CD4 + T cells, CD8 + T cells, macrophages, neutrophils and dendritic cells were involved 19 . Here, we investigate the signi cance of MEX3A-D mRNA expression data and invasion of six immune cell types in lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC).

Results
Four MEX3 genes were assessed for mutations by COSMIC database. Prior to February 20, 2021, the characteristics of MEX3 in Table II, and the genetic alteration affecting MEX3 in lung cancer samples in Table III. We found that the regulatory mechanism types in tumors, including lung cancer, are mainly missense mutations. Further, in the regulatory mechanism types study of lung cancer, three regulatory mechanism types were found in Fig. 1, Point Mutations, Copy Number Variation (CNV), Gene Expression. While in these regulatory mechanism types, point mutations mainly occurred in MEX3B, with the highest mutation frequency being 1.06%; CNV was con rmed on MEX3A, with the highest mutation frequency 1.99%; Gene Expression (Overexpressed) was determined on MEX3A, with the highest frequency 14.03%. No translocations, insertions, deletions, or loss of heterozygosity were identi ed. In summary, the results indicate that the sequence or copy number of the MEX3 gene has not altered signi cantly, except for MEX3A overexpression, which revealed that it is stable and not easy to mutate, leading to malignant proliferation of tumor cells. It may explain the development of malignant tumors.
Oncomine™ database analysis revealed that MEX3 expression in tissues of NSCLC patients compared to normal. The column bar graph in Fig. 2  Secondly, the effect of MEX3B mRNA expression on prognosis was examined, once again, the survival curves of patients with NSCLC, LUAD, and LUSC were described (Fig. 3b). Over-expressed of MEX3B was associated with decreased OS in all NSCLC (HR = 1.37; CI, 1.  Table S1. MEX3A and MEX3B were signi cantly associated with classi cation 2,while MEX3A was associated with classi cation 1 and MEX3B was associated with classi cation 3 shown in Table S2. MEX3A was correlated with 0 lymph node status (Table S3), meanwhile, four MEX3s were correlated with gender (Table S4) in NSCLC patients. For MEX3A, it was signi cantly associated to smoking, but MEX3C and MEX3D no signi cant (Table S5). In contrast, MEX3 was not associated to prognosis with or without chemotherapy treatment in NSCLC patients (Table S6).
The relationship between MEX3 and the immune microenvironment of LUAD and LUSC, as well as information of tumor purity were obtained by TIMER 2.0 database. In LUAD, MEX3A was positively correlated with tumor purity and negatively correlated with the level of dendritic cell in ltration. MEX3B was positively correlated with tumor purity, CD4 + T cell and macrophage in ltration level. MEX3C, not only correlated with tumor purity, but also with CD8 + T cells, CD4 + T cells, macrophages, and neutrophils in ltration levels were positively correlated. Finally, we found that MEX3D has statistically related with tumor purity, but positively in CD4 + T cells, macrophages and neutrophils (Fig. 4a). In LUSC, similarly, MEX3A was shown to be positively correlated with tumor purity, B cells, negatively with CD4 + T cells.
MEX3B and MEX3C were positively correlated in tumor purity, CD8 + T cells and CD4 + T cells. In addition, MEX3D has no signi cation on tumor purity, but positively correlates with CD4 + T cells and macrophages (Fig. 4b).
Here, further, we assessed that overexpressed MEX3 had no statistically signi cant difference in the survival rate of NSCLC patients, regardless of whether it was 1 year, 3 years or 5 years (P > 0.05).

Discussion
The human MEX3 family is differentially expressed in healthy tissues of different origins 9 , so in the same way, we are interested in understanding how it is expressed in abnormal tissues, particularly in cancer. In the human protein atlas, members of the MEX3 family are expressed in heterogeneously types of tumors 22,23 . Many evidences indicated that MEX3A promotes cell proliferation and inhibits cell apoptosis bladder cancer 24,25 , gastric cancer 26 , colorectal cancer 27 . Furthermore, increased MEX3A levels are also reported in liver cancer, which were signi cantly associated with a poor patients' survival 28 . This study revealed that over-expression of MEX3A had signi cantly poor prognosis of NSCLC and LUAD, but not in LUSC. In addition, increased expression of MEX3A in NSCLC patients is also associated with stage I, classi cation tumor, lymph node status, and male, whether chemotherapy has little effect on the prognosis. For MEX3B,there are few studies on tumors. It may be a ubiquitination of Runx3(runt-related transcription factor 3) and can increase invasion of gastric cancer cells 29 . Our study revealed that high expression of MEX3B mRNA had decreased OS in all NSCLC and LUAD cases. Over-expressed MEX3B was also associated with classi cation T2 and T3, and gender of NSCLC patients.
Recently, MEX3A and MEX3C proteins have been reported to be negatively correlated posttranslational regulators of several target genes 30 . In colorectal cancer, MEX3C has been identi ed as an unstable gene that is frequently lost in CIN+ (cervical intraepithelial neoplasia+) tumors 31,32 ,regulating lipid metabolism through JNK(c-Jun N-terminal kinases) pathway in bladder cancer 33 and breast cancer 34 . Based on these observations, we hypothesized that MEX3C is an important part of in uencing metastasis and prognosis in NSCLC. Consistent with our ndings, high expression of MEX3C mRNA was associated with poor prognostic OS in NSCLC and LUAD patients. Several study 35,36 demonstrated that MEX3D reverses apoptosis by interacting with au-rich elements (AREs) and enhances the degradation of BCL2 (B-cell lymphoma2) mRNA. Moreover, MEX3D is frequently deleted in various human cancers 37 , can also participate in the modulation by chemotherapy in AML (acute myeloid leukemia) 38 and overexpression in androgen-independent prostate cancer 39 . In this study, according to Kaplan-Meier analysis, high expression of MEX3D was observed, it was a good prognostic indicator, not only in LUAD and LUSC, but also in all NSCLC.
In immune response, MEX3B can be used as a co-receptor in the innate antiviral response of toll-like receptor 40 . In melanoma 41 , under-expression of MEX3B is associated with antibodies against the programmed cell death 1 (PD-1) receptor, while over-expression can inhibit T cell-mediated tumor elimination. MEX3C is involved in the regulation of proteins, degradation and ubiquitination, and it has been identi ed as a new type of RNA-binding E3 ubiquitin ligase, which is responsible for posttranscriptional regulation 42 . In present study, our results showed that MEX3B and MEX3C expression was positively related to tumor purity, and CD8 + T cell and CD4 + T cell in ltrating levels. While the function of MEX3A and MEX3D in immune responses has not yet been su cient research evidence. Abundant evidence 8,9 that, MEX3 proteins have the ability to regulate gene expression in tumor suppression, negative correlation. In the follow-up, we can explore the research of the targeted inhibitor of MEX3 and it could be identi ed as a marker of immunotherapy detail.
In summary, from the research results, in the COSMIC database, the sequence or copy number of MEX has no major alterations, which can explain its more copy number in malignant tumors, that is, malignant proliferation. After that, we used the Oncomine™ database to detect the expression of MEX3 in NSCLC, and Kaplan-Meier analyzed the prognostic value of genes. The results indicated that in NSCLC and LUAD, the lower prognosis of overexpressed MEX3A, MEX3B, MEX3C, and MEX3D was statistically signi cant, besides, in LUSC, MEX3D is also highly expressed with poor OS. In addition, MEX3B and MEX3C were positively related to tumor purity, of CD8 + T cells and CD4 + T cells in ltrating levels in TIMER.
These data re ect the potential association of MEX3 in non-small cell lung cancer. We found that most of MEX3 is highly expressed in NSCLC. High expression indicates a poor prognosis and has a certain immune correlation. Therefore, these conclusions can lay a framework for the prognosis of NSCLC patients and the development of treatment strategies in the future. But of course, our research needs to be improved and deepened. In the future, we need to further analyze its post-transcriptional regulatory mechanism and immunomodulatory effects, which is currently the focus of tumor immunotherapy direction.
Declarations Acknowledgements Not applicable.

Availability of data and materials
The datasets analyzed in the present study are available in the COSMIC database

Competing interests
The authors declare that they have no competing interests.
Authors' contributions MZ, HL Z, JJD were responsible for the design of the study and interpretation of the data. LF C, HX H, ZX F and XD L examined the archives and identi ed the databases included in the study, and they also have revised critically the manuscript for important intellectual content. All authors contributed to the writing of the manuscript. All authors read and approved the nal manuscript.    Abbreviation: HR, hazard ratio; CI, con dence interval. b The prognostic value of MEX3B expression. Survival curves were plotted for all NSCLC patients (n=1924). Survival curves were plotted for LUAD patients (n=672). Survival curves were plotted for LUSC patients (n=271). Data was analyzed using Kaplan-Meier Plotter. Patients with expression above the median are indicated in red line, and patients with expressions below the median in black line. Abbreviation: HR, hazard ratio; CI, con dence interval. c The prognostic value of MEX3C expression. Survival curves were plotted for all NSCLC patients (n=1925). Survival curves were plotted for LUAD patients (n=719). Survival curves were plotted for LUSC patients (n=524). Data was analyzed using Kaplan-Meier Plotter. Patients with expression above the median are indicated in red line, and patients with expressions below the median in black line. Abbreviation: HR, hazard ratio; CI, con dence interval. d The prognostic value of MEX3D expression. Survival curves were plotted for all NSCLC patients (n=1925). Survival curves were plotted for LUAD patients (n=719). Survival curves were plotted for LUSC patients (n=524). Data was analyzed using Kaplan-Meier Plotter. Patients with expression above the median are indicated in red line, and patients with expressions below the median in black line. Abbreviation: HR, hazard ratio; CI, con dence interval.