A Comprehensive and Systematic Analysis Revealed the Role of ADAR1 in Pan-Cancer Prognosis and Immune Implications

Adenosine deaminase RNA specific 1 (ADAR1) has been identified as an enzyme that deaminates adenosine within the dsRNA region to produce inosine, whose amplification reinforced the exhaustion of the immune system. Although there were currently cellular and animal assays supporting the relationship between ADAR1 and specific cancers, there was no correlation analysis that has been performed at the pan-cancer level. Therefore, we first analyzed the expression of ADAR1 in 33 cancers based on the TCGA (The Cancer Genome Atlas) database. ADAR1 was highly expressed in most cancers, and there was a closely association between ADAR1 expression and prognosis of patients. Furthermore, pathway enrichment analysis revealed that ADAR1 was involved in multiple antigens presenting and processing inflammatory and interferon pathways. Moreover, ADAR1 expression was positively correlated with CD8+ T cell infiltration levels in renal papillary cell carcinoma, prostate cancer, and endometrial cancer and negatively correlated with Treg cell infiltration. In addition, we further found that ADAR1 expression was closely associated with various immune checkpoints and chemokines. Meanwhile, we observed that ADAR1 may be involved in the regulation of pan-cancer stemness. In conclusion, we provided a comprehensive understanding of the oncogenic role of ADAR1 in pan-cancer, and ADAR1 might serve as a new potential target for antitumor therapy.


Introduction
Despite advances in surgery, chemotherapy, and radiotherapy, the global morbidity and mortality of malignant tumors are on the rise [1]. Tumors remodel the tumor immune microenvironment (TIM) through various factors, such as dysfunction of immune checkpoints and secretion of chemokines [2]. Although there are advances in immunotherapy drugs targeting PD-1, PD-L1, and CTLA4, only a minority of patients benefit from immunotherapy [3]. The prognostic model constructed by combining multiple genes has been used to evaluate the efficacy of tumor immunotherapy, but there is no rigorous clinical proof [4]. Therefore, the exploration of appropriate immunotherapeutic targets is particularly urgent. ADAR1, adenosine deaminase RNA specific 1, is the main RNA-editing enzyme responsible for the deamination of adenosine to produce inosine (A-to-I) [5]. ADAR1 also participates in a variety of biological processes in a non-RNA-editing manner, the most important of which is the creation of protein-protein interactions through the double-stranded RNA-binding domain (dsRBD) of ADAR1, which directly and systematically regulates protein-based immunity response [6]. As an enzyme, ADAR1 edits endogenous double-stranded RNA (dsRNA), making it unstable and unable to be recognized by nucleic acid sensors (NAS), thereby closing interferon-stimulated gene (ISG) expression and downregulating interferon (IFN) expression [7]. Previous research demonstrates that ISG-positive tumor cells are uniquely susceptible to ADAR1 deficiency, which sensitises tumor cells to immunotherapy and overcomes resistance to checkpoint blockade [8]. The depletion of ADAR1 in cancer cells was susceptible to death by inflammation [9]. ADAR1 prevented the pathway to immune and translational catastrophe by blocking dsRNA activation.
In this study, we evaluated the expression and variation of ADAR1 and analyzed the relationship with the prognosis of patients. We investigated the relationship between ADAR1 and immune cell infiltration, immune suppressor genes, and chemokines. Our findings provided a new insight into the functional role of ADAR1 in pan-cancer and highlighted that ADAR1 may serve as a new potential target for cancer immunotherapy.

Genomic Modifications of ADAR1 in Cancer Patients.
Alterations in ADAR1 in cancer patients were obtained from the online cBioPortal database (http://www.cbioportal.org/). Cancer genomic alterations in ADAR1 include copy number amplification, profound deletions, missense mutations of unknown significance, and mRNA upregulation [10].

Survival
Analysis of the Prognostic Value of ADAR1. The Kaplan-Meier (KM) survival analysis was performed to identify different survival outcomes between the two group differences. The univariate Cox regression model was applied to determine the favourable or unfavourable prognosis of ADAR1. The KM analysis was performed by the R packages "survminer" and "survival" and by the R packages "survival" and "forestplot." 2.5. GSEA. GSEA is a method to analyze the function or pathway of target genes affecting tumor genetics. The Kyoto Encyclopedia of Genes and Genomes (KEGG) database and HALLMARK database were adopted in the R package "clus-terProfiler" for GSEA [13,14]. The significant enrichment results were demonstrated on the basis of net enrichment score (NES), gene ratio and p value. Gene sets with NES > 1, NOM p < 0:05, and FDR q < 0:25 were considered to be significantly enriched.
2.6. Correlation Analysis between ADAR1 Expression and Immune Function. TISIDB (http://cis.hku.hk/TISIDB/) is a web server that integrates multiple heterogeneous data types for tumor and immune system interactions [15]. We used the "Lymphocyte," "Immunomodulator," and "Chemokine" modules in TISIDB to analyze the correlation between ADAR1 expression and the level of immune infiltration, immune checkpoints, and chemokines for multiple cancer types in TCGA database. 2.8. Statistical Analysis. Gene expression data from TCGA database were analyzed by Student's t-test. The expression of ADAR1 was correlated with the abundance scores of immune cells assessed using Spearman's correlation analysis. All analyses were performed with R software (version 4.1.1, http://www.r-project.org) loaded with R packages ("ggplot2," "ggpubr," "limma," "survival," "survminer," "clusterProfiler," "ESTIMATE," "enrichplot," and "forestplot"). p < 0:05 was considered statistically significant to provide confidence in the data analysis.

The mRNA Expression Level and Copy Number
Variation of ADAR1 in Pan-Cancer. We first evaluated the expression of ADAR1 in 33 cancers. Figure 1(a) shows the ranking of ADAR1 expression in cancer tissues from top to bottom. Subsequently, we observed the expression level of ADAR1 in different cancers. The results showed that ADAR1 was significantly upregulated in 14 cancers, including BRCA, LUAD, ESCA, LUSC, STAD, CHOL, CESC, HNSC, UCEC, PCPG, BLCA, COAD, THCA, and LIHC ( Figure 1(b)). It is widely recognized that gene copy number affects the level of gene expression [16]. Therefore, we 2 Disease Markers further analyzed ADAR1 copy number variation (CNV) by the cBioPortal database. We observed that most cancers with CNV exhibited copy number amplification (Figure 1(c)).

Protein Level and PPI Network Analysis of ADAR1 in
Pan-Cancer. We previously analyzed mRNA expression level and the copy number of ADAR1 in pan-cancer. Next,   Disease Markers we further investigated the protein level of ADAR1 in pancancer. The result showed that the protein levels of ADAR1 were highest in breast cancer and lung cancer by HPA database. (Figure 2(a) and Supplementary Figure 1). As a potential immunomodulatory gene, a location on the cell membrane is essential. To explore the potential role of ADAR1 in pan-cancer, we constructed a PPI network of ADAR1 via the GeneMANIA and STRING databases. Results from two databases showed that ADAR1 may interact with STAT2, IFIT1, and IFIT3 (Figures 2(b) and 2(c)). Given that STAT2, IFIT1, and IFIT3 play a role in regulating immune cell function, we firmly believed that ADAR1 may be involved in the regulation of the TIM [17,18].

The Prognostic Significance of ADAR1 in Pan-Cancer.
We further evaluated the prognostic value of ADAR1 in pan-cancer. We first plotted the KM curves by the median value of ADAR1 expression; the results showed that high expression of ADAR1 in ACC, LGG, LUAD, PAAD, and UCEC had poor overall survival (OS) (Figures 3(a)-3(e)). Interestingly, patients with high ADAR1 expression in ACC, KICH, and KIRP had poor disease-specific survival (DSS) (Figures 3(f)-3(h)). Likewise, UCEC and ACC patients with high ADAR1 expression had poor progression-free survival (PFS), and KIRP and ACC patients with high ADAR1 expression had poor disease-free survival (DFS) (Figures 3(i)-3(l)). In addition, we further analyzed the expression of ADAR1 in the four aspects of OS, DSS, DFS, and PFS by the univariate Cox analysis. The results showed that ADAR1was a risk factor for patients with LGG, KRIP, LIHC, ACC, and UCEC in terms of OS (Figure 4(a)). Meanwhile, DSS analysis revealed that ADAR1 served as a risk factor for patients with KIRP, LGG, and ACC (Figure 4(b)). The DFS analysis revealed that ADAR1 served as a risk factor for patients with UCEC, PRAD, ACC, and KIRP but may be a protective factor for OV (Figure 4(c)). The PFS analysis revealed that ADAR1 served as a risk factor for patients with ACC, KIRP, LGG, UCEC,     (Figure 4(d)). These data strongly suggested that ADAR1 plays an important role in tumor prognosis.

Enrichment Analysis of ARDR1 Expression in Pan-
Cancer. To preliminarily explore the mechanism of ADAR1 in pan-cancer, we performed GSEA analysis using KEGG, HALLMARKER, and immunologic signature gene sets, respectively. The results indicated that ADAR1 was involved in a variety of signaling pathways. In KEGG, ADAR1 is closely related to pathways of cell adhesion, antigen processing, and chemokine expression (Figures 5(a)-5(f)). In HALLMARK sets, ADAR1was involved in the inflammatory and interferon pathways (Figures 5(g)-5(l)). These results suggested that ADAR1 may play an important role in regulating the tumor immune microenvironment.

Association of ADAR1 Expression with Immune Cell
Infiltration, Immune Checkpoints, and Chemokines in Pan-Cancer. Tumor-infiltrating immune cells are an important part of the complex microenvironment that regulates the development and progression of cancers [3]. Tumorinfiltrating immune cells are the main performers of tumor immune responses [19]. To explore whether ADAR1   Given that deficiencies in immune surveillance are an important cause of poor prognosis in various cancers. Tumors evade immune cell attack by exploiting multiple pathways, such as regulation of immune checkpoints and secretion of leukocyte chemokines. Therefore, we further investigated the relationship between ADAR1 and the above functions. The results showed that ADAR1 was positively correlated with the expression levels of immunosuppressive genes such as CD274, PDCD1, LAG3, CTLA4, TIGIT, CD96, and IDO1 in most tumors (Figure 7(a)). ADAR1 positively correlated with chemokines, such as CCL14 and CCL28 in most tumors (Figure 7(b)). These data suggested that ADAR1 may regulate the expression of immune checkpoints and chemokines to modulate the tumor immune microenvironment.