Immunological Role of TP53 Somatic Mutation Classification in Human Cancers

Background TP53 is a very common tumor suppressor gene and has implicated in various cancers. A systematic immunological analysis of TP53 somatic mutation classification in multiple cancers is still lacking. Methods To assess the immunological value of TP53 somatic mutation classification in various cancers, we integrated a series of bioinformatics methods to analyze the role of TP53 gene across the public databases, such as UCSC Xena, Cancer Cell Line Encyclopedia (CCLE), Ensembl, and Genotype−Tissue Expression (GTEx). Results The results revealed that the TP53 expression level had significant difference in tumor tissues and normal tissues, and it had a high expression level in most malignant tumors. Moreover, the missense mutation is the most common type of TP53 mutation in most cancers. In addition, the Cox proportional hazards model analysis and Kaplan−Meier (KM) survival analysis demonstrated that the TP53 expression is a high-risk factor in brain lower-grade glioma (LGG), prostate adenocarcinoma (PRAD), and uterine carcinosarcoma (UCS), which is opposite in uterine corpus endometrial carcinoma (UCEC). Besides, compared to the TP53 nontruncating mutation classification samples, we found that TP53 truncating mutation samples had lower TP53 expression levels in certain types of cancer. Notably, TP53 was associated with the mismatch repair (MMR) gene in some cancers which contained truncating or nontruncating mutation. Based on the classification of truncating or nontruncating mutation, we also discovered that TP53 expression was positively or negatively correlated with the immune score, stromal score, and the levels of immune cell infiltration in different cancers. Conclusions Our research reveals an overarching landscape of immunological value on TP53 status in various malignant tumors. According to our results, we demonstrate that TP53 also plays an immunological role in various cancers.


Introduction
Cancer is a malignant disease with a high mortality rate. Up till now, there is no efective treatment to absolutely cure cancer patients which most likely predicts poor quality of life [1]. However, immunotherapy has been reported to achieve great improvement and regarded as one of the major breakthroughs in cancer treatments [2][3][4]. It is feasible to seek out potential immunotherapy biomarker through the utility of databases which contained sequencing data.
TP53 is a well-studied tumor suppressor gene and is always the hotspot of tumor research. Nowadays, studies on TP53 function have concentrated on the correlation between variation pattern of TP53 and prognosis [5][6][7]. TP53 has a prominent role in preventing tumor development and maintaining genomic stability (GS) [8,9]. Te cancer-associated function of the p53 protein depends on its fve function domains, including transactivation domain, proline rich domain, DNA-binding domain, oligomerization domain, and carboxy-terminal regulatory domain [10,11]. In various cancers, TP53 was explicitly linked to cancer development and prognosis [12][13][14], but these conclusions were still controversial [15]. Te most widely accepted concept is that the nontruncated mutation commonly induces high expression of TP53 [13]. Previous studies have demonstrated that TP53 mutations are associated with their expression, cancer prognosis [13], and immune − related research [16]. Integrated studies combined with these features with classifcation of TP53 somatic mutation are few.
Our work makes it possible to explore the association between TP53 mutation classifcation and immunological function by multiple databases, such as UCSC Xena, CCLE, Genotype -Tissue Expression (GTEx), and Ensembl. Tis study may provides new perspective on the relationship between gene variation classifcation and microenvironment for researchers. We present the following article in accordance with the MDAR reporting checklist.

TP53 Gene Expression Data.
Te expression data across 33 types of cancers which contains 11057 samples were downloaded from UCSC Xena. Te expression data across 31 health tissues which contains 7858 samples and 28 tumor cell lines were downloaded from GTEx and CCLE. Expression data were extracted using in-house Perl scripts and further analysis using R software (version 4.1.2, R Foundation for Statistical Computing, Vienna, Austria).

TP53 Variation and Classifcation
Criteria. In this part, the varscan2 variation fles with TP53 gene form UCSC Xena (accessed 18 February 2022) were used for analyzing base alterations. According to the mutation classifcation method in previous studies [13,15], we distinguished between truncating (including nonsense mutation, frameshift mutation, and splice site mutation) and nontruncating (including missense mutation, in-frame deletion mutation, and in-frame insertion mutation) TP53 mutation.

Correlation between TP53 and Prognosis.
We obtained relevant clinical data from the UCSC Xena. Overall survival (OS) data of patients were used to estimate the prognosis status of diferent TP53 classifcation. Te patients were divided into two sections by the mean value of TP53 expression as a cutof value [17]. Te KM method and log-rank test were used for survival analysis. Moreover, Cox proportional hazards model analysis was performed to analyze the hazard ratio among diferent types of cancer.

Correlation between TP53 and Mismatch Repair Genes.
We calculated the Pearson correlation coefcient to estimate the relationship between TP53 and mismatch repair (MMR) genes, including MLH1, MSH2, MSH6, and PMS2. P < 0.05 was considered for the statistical signifcance.

Correlation between TP53 and Immunity.
We used the ESTIMATE algorithm to infer the degree of tumor infltration [18,19]. Moreover, the CIBERSORT method was further used for exploring the correlation between TP53 and immunity [20]. Te annotation fle with Ensembl gene converted to gene symbol was downloaded by Ensembl database.

Te Landscapes of TP53 Expression across Multiple
Datasets. In various normal samples, we found signifcant diference of TP53 expression, as shown in Figure 1(a) (P < 2.2e − 16). Te skin tissue expresses TP53 on the highest level and the brain tissue expressed the lowest. We further extracted TP53 expression values from cancer cell lines using CCLE datasets, and the consequence was presented in Figure 1(b), displaying the analogous diference in cell lines (P � 1.9e − 07).
In addition, we discovered the similar signifcance of TP53 expression in 33 cancer types (P < 2.2e − 16). As shown in Figure 2(a), pheochromocytoma and paraganglioma (PCPG) and kidney chromophobe (KICH) are on the lowest level while UCEC is on the highest level.

TP53 Mutation in Pan-Cancer Cohorts.
After eliminating those cancers with the small number samples (for instance, cancer types with less than thirty samples were excluded), 19 cancer types were screened for depicting the TP53 mutation types in pan-cancer, and the landscape was exhibited in Figure 3. Consistent with the previous studies [13,21], we found that missense mutations accounted for the most of TP53 variants in 19 types of cancer. Next, nonsense mutation ranked second.

Prognostic Impact of TP53 across TP53-Mut Cancers.
In Figure 4, forest plots showed that TP53 was high-risk gene in LGG (hazard ratio � 1.614), PRAD (hazard ratio � 11.97), and UCS (hazard ratio � 2.748), while it was a low-risk gene in UCEC (hazard ratio � 0.5717). Furthermore, we carried out prognostic impact of TP53 expression using mean value as criteria and revealed the signifcant diference between high and low expression groups among LGG (P � 0.0084), PRAD (P � 0.0035), UCS (P � 0.0098), and UCEC (P � 0.0068).  with the earlier conclusion, the results (see Figure 5) showed that TP53 expression levels were upregulated in truncating mutation relative to nontruncating patients [15]. Furthermore, we performed survival analysis for the sake of the evaluation of TP53 mutation's prognostic value. Unfortunately, no signifcance diference was detected between TP53 mutation types and overall survival time in any type of cancer ( Figure 5). However, KM (Kaplan−Meier) results indicated a clear trend that individuals with truncating mutation had longer survival time in specifc cancers, such as READ, BLCA, and PAAD.
In nontruncating TP53-mut cancer types ( Figure 6(b)), we also explored the positive correlation across multiple cancers. Interestingly, compared with diferent TP53 nontruncating classifcation, there was opposite correlations in UCEC (MSH6 correlation coefcient/P-value � 0.24/0.005), SKCM (MSH6 correlation coefcient/P-value � 0.37/0.03),  3.6. Correlations between TP53 and Immunity. Tumor microenvironment (TME) played an important role during neoplasm occurrence and progression [16,25,26]. In this  Journal of Oncology study, we calculate the stromal score, immune score, and estimate score in 19 TP53-mut cancers using the ESTIMATE method. As shown in Table 1, in TP53 truncating mutation cancers, such as BRCA, HNSC, LIHC, LUAD, LUSC, SKCM, PAAD, SARC, and STAD, TP53 was signifcantly positively correlated with the immune score, as well as stromal score and estimate score, while it was negatively correlated with UCS. Conversely, in nontruncating mutation cancers (for instance, BRCA, GBM, OV, and PRAD), TP53 was significantly negatively associated with the estimated TME scores, while LIHC and PAAD had the opposite consequences. Subsequently, we investigated the immune cell infltration levels among TP53-mut caners. Te results implied the diverse signifcant correlation between levels of immune cell infltration and TP53 expression among TP53 truncating or nontruncating mutant cancers. Signifcant correlation is screened and presented in Table 2.

Correlation between TP53 and Biological Function.
We carried out a thorough inspection of Gene Set Enrichment Analysis (GSEA) to investigate the relationship between TP53 and biological function inTP53-mut tumor tissues, and the results are shown in Supplementary Figures  (Available here).

Discussion
In this article, the expression level of TP53 gene was diverse in tumor or normal tissues. TP53 expression was higher in most cancers than in normal tissues except KICH.
Besides, we analyzed the variation of TP53 in 19 cancers and discovered that missense mutation was the dominant subtype, which is consistent with the previous conclusion [13]. At the same time, the evidences showed that patients with the higher expression of TP53 had a worse survival in UCS, LGG, and PRAD. On the contrary, in UCEC, the higher expression of TP53 gene was linked to better survival.
In previous studies, the prognosis roles of TP53 mutation were controversial [15]. Meanwhile, the expression level of TP53 is often related to the mutation types [13]. In order to examine the impact of TP53 mutations on prognosis, TP53 mutations were divided into truncating and nontruncating  Figure 6: Association between TP53 expression and MMR genes in 19 TP53-mut cancers. (a) Te relationship between TP53 expression and MMR genes in truncating TP53-mut cancers. (b) Te relationship between TP53 expression and MMR genes in nontruncating TP53-mut cancers. For each cell, the top left triangle represents the P value and the bottom right triangle represents the correlation coefcient. * indicates P < 0.05, * * indicates P < 0.01, * * * indicates P < 0.001. 8 Journal of Oncology mutation groups referring to the published reliable classifcation method [13]. Our results demonstrated that the patients with truncating mutations presented lower TP53 expression. Besides, Kaplan−Meier analysis showed a clear trend that individuals with truncating TP53 mutation had longer survival time in BLCA (P-value � 0.12), PAAD (P -value � 0.24), and READ (P-value � 0.12), consistent with the results of previous published literature [13]. Dan et al. [22] clarifed that the abnormal expression of mismatch repaired genes induced the increased frequency of somatic mutation. In colorectal cancer, Perez et al. [31] indicated that mismatch-repaired defciency can induce TP53 mutation. Fang et al. [32] reported that TP53 defection and mismatch-repaired defciency commonly occurred in early carcinosarcoma. Te correlation between TP53 mutation classifcation and the expression of the MMR genes was analyzed. In the truncating mutation group, TP53 expression is positively correlated with MMR genes expression in LGG, LIHC, and HNSC, while negative correlation was found in other four cancers, including COAD, SKCM, UCEC, and STAD. In contrast, in the nontruncating mutation group, we detected the positive correlation in most cancers. Compared to TP53 nontruncating classifcation, there were opposite correlations in UCEC, SKCM, and STAD.
We further explored the relationship between TP53 mutation classifcation and tumor microenvironment.

Journal of Oncology 9
According to the ESTIMATE algorithm [18,19], we calculated the stromal score, immune score, and estimate score. In the TP53 truncating mutation group, TP53 is signifcantly positively correlated with immune score, as well as stromal score and estimate score in specifc cancers, such as BRCA, HNSC, LIHC, LUAD, LUSC, SKCM, PAAD, SARC, and STAD. Whereas, it is negatively correlated in UCS. Conversely, in the nontruncating mutation group, TP53 is signifcantly negatively correlated with stromal/estimate/immune score in BRCA, GBM, OV, and PRAD, while LIHC and PAAD had opposite results. Correlation between the degree of immune cell infltration and TP53 expression was estimated further. In truncating or nontruncating mutation samples, obvious associations between the previous two factors were shown in most cancers. Finally, GSEA results indicated that TP53 (truncating or nontruncating) was involved in some immune−related function and pathways. Some limitations of this paper include the limited databases and scarce experimental verifcation. In the future, we will adopt more valuable databases and experimental results to confrm and improve our work.

Conclusions
In conclusion, this might be the frst comprehensive and systematic research to evaluate the immune − related mechanisms of TP53 mutation classifcation in diferent cancer. According to our results, TP53 is related to immunological function based on diferent mutation classifcation in various cancers. It is worth mentioning that these fndings might extend better understanding of TP53 gene underlying the mechanism in the immune system.

Ethical Approval
Te authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Conflicts of Interest
Te authors declare that they have no conficts of interest.