A Pan-cancer analysis of GPR183 correlated with tumor immunity

Background G-protein-coupled receptors(GPRs) are the largest family of membrane proteins in the human genome, positively involving in human physiological and metabolic activities. Drugs targeted to GPRs have developed fast. However, there are still few GPR-targeted drugs for cancer treatment. GPR183, an orphan seven-transmembrane receptor, have been widely studied in immune regulation. Yet, its role in tumor immunity remains mysterious. Results In our study, we performed a pan-cancer analysis of GPR183. Though statistically signicantly differed between tumor and normal samples in various cancers, the expression of GPR183 showed a dual trend. Besides, we applied integrated network analysis to predict GPR183 related genes and analyzed their function with GO and Pathway analysis. GPR183 was strongly associated with immune responses. Moreover, we evaluated the prognostic value of GPR183 in various cancers and found it related to the prognosis of BRCA, SKCM, LUSC, LGG and UVM. We further explore its immunoregulatory function based on differentially expressed genes. Finally, we estimated the tumor immune inltration associated with GPR183 expression using “TIMER”. The GPR183 expression was highly correlation with macrophages and dendritic cells inltration. Conclusions a multiple immunoregulatory role tumor and UVM). We also found GPR183 had intimate connection to PD-1 signals. GPR183 might become a potential therapeutic target for cancer treatment.

Background G-protein-coupled receptors(GPRs) are the largest family of membrane proteins in the human genome, positively involving in human physiological and metabolic activities [1]. GRPs can be activated by a series of extracellular signals including small peptides, metabolites, neurotransmitters, ions and proteins. After ligand binding, GPRs change their conformation and mediate the complex cytosolic siganllings, leading to a cellular response [2]. Aberrant regulation or mutations of GPRs, G proteins and downstream siganllings are implicated with various diseases such as heart failure [3], asthma [4] and bipolar disorder [5]. In addition, increasing evidences have pointed to GPRs and their downstream signallings as relevant players in cancer initiation and progression [6][7][8]. GPRs have been proved to play multiple roles in cell differentiation, cell proliferation, immune responses, invasion, angiogenesis and metastasis [7,9,10]. Thus, drugs targeted to GPRs for cancer prevention and treatment become promising. Yet, the development of such drugs is limited by our knowledge of regulatory role in cancer of many GPR members in cancer.
G protein-coupled receptor 183(GPR183, also known as EBI2), an orphan seven-transmembrane receptor, have been widely studied in immune regulation [11]. Previous researches have demonstrated that GPR183 positively participated in immune responses of B cell and Dendric cells [12][13][14][15]. GPR183 played vital role in cell communication, migration and development of lymphocytes [16,17]. Recently, some studies have found GPR183 mediated intestinal immunity and in ammation by inducing migration and development of Group 3 innate lymphoid cells(ILC3s) [18][19][20]. Though GPR183 has intimate connection to body immunity, its role in tumor immunity remains mysterious. Given the broad prospects of GPRs drugs in cancer treatment, unveiling the regulatory mechanisms especially immune-related functions of GPR183 in tumor progression might provide a new potential treating target.
In this study, we rst performed a pan-cancer analysis of GPR183 to explore the mutational and expression landscapes in various cancers. In order to study the regulatory role of GPR183 in cancers, we applied integrated network analysis of GPR183 to predict related genes. Then we performed functional analysis including GO and pathway analysis to further explore the regulatory mechanism of GPR183. Besides, we applied cox survival regression and Kaplan-Meier(KM) analysis to estimate the prognostic value of GPR183 in different cancer types from The Cancer Genome Atlas(TCGA) (https://portal.gdc.cancer.gov/). We found that GPR183 associated with prognosis in ve cancer types(BRCA, SKCM, LUSC, LGG and UVM). To gure out the role of GPR183 in these cancer types, we identi ed differentially expressed genes(DEGs) associated between high and low GRP183 expression group. In addition, we separately analyzed the function of commonly upregulated differentially expressed genes and prognosis-related differentially expressed genes. Finally, we estimated correlation between GPR183 expression and the tumor immune-in ltration levels with online tool "TIMER" [21].

Pan-Cancer Expression Landscape of GPR183
Among all of the 33 cancer types from the TCGA database, 12 cancer types showed differential GPR183 expression between tumor and normal tissues( Fig. 1C and Figure S1). In BLCA, COAD, LUSC, PRAD, READ and UCEC, GPR183 expression was statistically signi cant higher in normal tissues( Fig. 1C and Figure S1). While in CHOL, GBM, HNSC, KIRC, KIRP and THCA, GPR183 had an inverse distribution( Fig. 1C and Figure S1). However, after compared to paired normal tissue data from GTEx, cancers with statistically signi cant high expression in normal sites were BLCA, COAD, LUSC and READ (Fig. 1B). In cancer types including ESCA, GBM, KIRC, KIRP, LAML, LGG and STAD, GPR183 expression was higher in tumor site (Fig. 1B).

Pan-Cancer Mutational Landscape of GPR183
In the analysis of GPR183 somatic single-base mutations(SSM) in different cancer types, the top ve cancers were UCEC, DLBC, ACC, COAD and SKCM (Fig. 1A). Among the top ve cancer types with highest GPR 183 mutations, the overall mutation landscapes were similar (Fig. 2). The most common variant classi cation, variant type and SNV class were separately missense mutation, SNP and C > T (Fig. 2).

Integrated Network Analysis of GPR183
In order to explore the function of GPR183, we applied integrated network analysis of GPR183. 20 genes were closely related to GPR183 in different ways including co-expression, predicted genes and colocalization (Fig. 3A). According to bioprocess analysis, GPR183 involved in various responses to stimulus including peptide, various organic compounds, lipids and bacterium. Besides, GPR183 expression was associated with metabolic and immune process (Fig. 3B). Moreover, GPR183 expression was linked to regulation of programmed cell death (Fig. 3B). The pathway analysis showed that a strong correlation between GPR183 and immune regulation (Fig. 3C). GPR183 participated in regulation of multiple signals such as INF-γ, SMADs family, AKT, EGFR, MECP2 and IGF-2 (Fig. 3C). In addition, GPR183 showed association with TP53 regulation (Fig. 3C).

Evaluation of prognostic association with GPR183 expression in Pan-cancer analysis
Using cox regression survival analysis, GPR183 expression was associated with prognosis in LGG and SKCM (Table 1, Table S1). After Kaplan-Meier(KM) analysis, we found high GPR183 expression was positively associated with prognosis in BRCA and SKCM( Fig. 4A and 4D), while negatively in LGG, LUSC and UVM (Fig. 4B, 4C and 4E).

Overall function analysis of GPR183 in prognosis related cancers
In order to explore the mechanisms of GPR183 in tumor progress and prognosis, we analyzed differentially expressed genes in high and low GPR183 groups in BRCA, SKCM, LUSC, LGG and UVM(|log2 fold-change| > 1, adjusted p-value < 0.01, Table S2). There were 824, 472, 1215, 1077 and 1110 upregulated genes in BRCA, SKCM, LUSC, LGG and UVM. Inversely, there were 284, 52, 180, 366 and 99 downregulated genes in BRCA, SKCM, LUSC, LGG and UVM. We found 18 commonly upregulated expressed genes in high GPR183 groups among these cancer types( Fig. 5A and 5B). The GO analysis demonstrated that GPR183 was related to regulation of immune response and ionic homeostasis like calcium (Fig. 5C). The molecular function of GPR183 were signaling receptor, molecular transducer, immune receptor and CPR activities (Fig. 5D). Besides, GPR183 expressed mainly in the membrane (Fig. 5E). Moreover, according to pathway analysis, GPR183 was involved in immune responses such as innate immune responses, signaling by interleukins and neutrophil degranulation (Fig. 5F).

Differential GPR183-related expression patterns in cancers with opposite prognosis
Though associated with prognosis in ve cancer types(BRCA, SKCM, LUSC, LGG and UVM), GPR183 seemed to exert a dual function. In order to explore the potential mechanisms causing opposite clinical outcomes in the ve cancers, we divided the ve cancers into two groups(Prognosis-Positive and Prognosis-Negative) determined by the GPR183-related prognosis. 258 genes were statistically signi cantly unregulated in Prognosis-Positive group (Fig. 6, Table S3). 222 genes were upregulated in Prognosis-Negative group (Fig. 6, Table S3). Using pathway analysis based on group-speci c genes(144 genes in Prognosis-Positive group, 108 genes in Prognosis-Negative group), still we found many immune-related pathways enriched both in the two groups (Table S4). Yet, the detailed regulatory function differed between them. The speci c pathways in Prognosis-Positive group were related to signalings of TNFs, Interleukin-2 and CLRs (Table 2). While among the top ve speci c pathways in Prognosis-Negative group, the function of GPR183 was strongly associated with PD-1 signaling (Table 3).

Association between GPR183 expression and tumor immune in ltration
Considering close correlation between GPR183 and immunity, we estimated the correlation between GPR183 expression and immune in ltration levels in the ve cancers (Fig. 7). Yet, except for UVM, GPR183 was strongly associated with immune in ltration in other four cancer types. Among all the six lymphocytes, macrophages and dendritic cells showed a quite high correlation. Discussion G-protein-coupled receptors(GPRs) are essential for the regulation of body physiological activities and homeostasis [1]. The extensive functions of GPRs cover from cell development, communication, differentiation, proliferation to apoptosis. Aberrant mutations or expression of GPRs could cause various disorders. Bene t from increasing knowledge on the regulation mechanisms of physiology and pathology, GPRs-targeted drugs have been constantly developing. Recently, GPRs have been found positively involved in tumor progression such as invasion and metastasis [9,22]. G Protein-Coupled Receptor Kinases(GRKs) were reported associated with tumorigenesis and cancer progression [23]. GPR87, GPR64 and GPR1 were reported separately associated with pancreatic ductal adenocarcinoma (PDA), endometrioid endometrial carcinoma and choriocarcinoma, which might become candidates of targeting by cancer therapeutics. [24][25][26]. Yet, the knowledge of many members of GPRs especially orphan receptors in cancer initiation and progression are still lacking, which limited the development of GPR-related drugs for cancer treatment.
In previous studies, G protein-coupled receptor 183(GPR183, also known as EBI2) has been reported to positively participate in immune responses including development and migration of lymphocytes [13,17]. It played vital role in intestinal immunity and in ammation through regulating ILC3s [18]. Though GPR183 has been proved strongly associated with immune responses, there are few studies on its regulation of tumor immunity. Thus, we applied a pan-cancer analysis method to explore the role of GPR183 in tumor immunity.
Though statistically signi cantly differed between tumor and normal samples in various cancers, the expression of GPR183 showed a dual trend. Besides, the mutational landscapes of cancers related to GPR183 mutations were similar. We identi ed 20 predicted genes related to GPR183 according to the integrated network analysis. The GO and pathway analysis showed strong correlation between GPR183 and immune responses. We found GPR183 not only positively regulated immunity through signals such as INF-γ, SMADs family, AKT, EGFR, MECP2 and IGF-2, but also might exert inhibitory function by activating programmed cell death. Therefore, the immunoregulatory role of GPR183 could be diverse.
In previous studies, GPR183 was reported to participate in progression of chronic lymphocytic leukemia-like Bcell malignancies, colitis and neuroin ammation [19,27,28]. We found that GPR183 was associated with prognosis of ve cancers(BRCA, SKCM, LUSC, LGG and UVM). However, the trend of prognostic correlation of GPR183 expression was polarized(High GPR183 expression was positive prognostic factor for BRCA and SKCM, but negative for LUSC, LGG and UVM). We assumed the difference originated from dual immunoregulatory role of GPR183.
To gure out the regulatory function of GPR183 in tumor progression, we rst identi ed differentially expressed genes between GPR183 high and low expression groups and found 18 commonly upregulated genes in high GPR183 group among all ve cancer types. A study demonstrated the pro-in ammatory instinct of GPR183 [29]. Function analysis of these 18 genes also showed GPR183 activated and promoted immune responses. Moreover, GPR183 promoted innate immunity, which might related to ILC3s [18]. Yet, as we found in Prognosis-Positive and Prognosis-Negative groups, the speci c regulatory mechanisms of GPR183 differed, contributing to opposite prognosis in these cancers. Among the top ve speci c pathways in the Prognosis-Positive group(BRCA and SKCM), we found TNFs and CLRs(C-type lectin receptor) related immune-activating pathways, with already proved proin ammatory performance, enriched [30,31]. We considered that GPR183 promoted anti-tumor immune responses via TNFs or CLRs in BRCA and SKCM. Among the top ve speci c pathways in the Prognosis-Negative group(LUSC, LGG, and UVM), PD-1 has been proved helping tumor immune escape and thus become one vital target for cancer immunotherapies [32]. We estimated GPR183 might reduce immune responses and aggravate tumor progression via PD-1 signals in the LUSC, LGG, and UVM. Yet, the exact mechanisms of immunoregulatory function of GPR183 need further experimental validation.
Previous studies have showed GPR183 was close linked to immunity [14,19,27]. We also found GPR183 was associated with tumor immune in ltration in BRCA, SKCM, LUSC and LGG. In addition, GPR183 showed a high correlation with dendritic cell in ltration. As reported, GPR183 was positively interacted with dendritic cells [15], which could also happen in tumor progression.

Conclusions
In sum, we rst performed a pan-cancer analysis of GPR183 and further evaluate its prognostic value and possible regulatory mechanisms. GRP183 showed a multiple immunoregulatory role in tumor progression. GPR183 was associated with prognosis of BRCA, SKCM, LUSC, LGG and UVM. Our study was limited by the lack of experimental validation. However, we evaluated the overall distribution and function of GPR183 in various cancers. We also found GPR183 had intimate connection to PD-1 signals. Considering different prognostic roles in BRCA, SKCM, LUSC, LGG and UVM, GPR183 might become a potential therapeutic target for cancer treatment.

Integrated Network Analysis of GPR183
To explore the function of GPR183, we applied GeneMANIA(http://genemania.org/) [37] to perform integrated network analysis of GPR183. The settings were all default.

Survival Analysis of GPR183 Expression
We used the "OncoLnc"(http://www.oncolnc.org/) [42] to perform cox survival regression analysis of GPR183 in various cancers. We also applied GEPIA to perform Kaplan-Meier(KM) analysis of GPR183 high/low expression groups determined by the median expression.
6. Analysis of differentially expressed genes Using R package "TCGAbiolinks" [43], we downloaded RNA-seq data of BRCA, SKCM, LUSC, LGG and UVM from TCGA(https://portal.gdc.cancer.gov/). The format of RNA-seq data was "Counts". The tumor samples were divided into two groups by the median expression of GPR183. We applied R package "DeSeq2" to identify differentially expressed genes(DEGs) associated with GRP183 expression in tumor samples. The threshold of DEGs were |log2 fold-change| > 1 and adjusted p-value < 0.01.

Commonly upregulated or downregulated genes related to GPR183
We used R package "VennDiagram" [44] to visualize the commonly upregulated genes in BRCA, SKCM, LUSC, LGG and UVM. We also analyze the DEGs distribution in two cancer groups with opposite GPR183-related prognosis with this package.

Statistical analyses
The Wilcoxon test was used to estimate the statistical signi cance for GPR183 expression difference between tumor and normal groups. All statistical analyses were conducted using R software(version 3.6.1). A two-tailed P values < 0.05 were considered statistically signi cant. Availability of data and materials

Abbreviations
The datasets analyzed during the current study are available in the TCGA(https://portal.gdc.cancer.gov/) and the Genotype-Tissue Expression (GTEx) project [35].

Competing interests
The authors declare that they have no competing interests.

Funding
The study was sponsored by the Interdisciplinary Program of Shanghai Jiao Tong University (project number YG2019QNA07, YG2019QNA51).
Authors' contributions WJ C, the corresponding author, was involved in design of the work, analysis of data, and approval of the nal version to be published. LS contributed to statistical analysis of the data and drafting the manuscript. YJ W, HZ and TW contributed to acquisition and analysis of data. YY and LX L were responsible for drafting the manuscript, analysis and interpretation of the data. All authors read and approved the nal manuscript.    Expression and mutational landscape of GPR183 in TCGA cohorts. (A). "Y" axis shows mutational rates of GPR183 (simple somatic mutation) in all TCGA cohorts. (B). Pan-cancer expression landscape of GPR183. "T" stands for tumor tissue and "N" stands for paired normal tissue. The red or green color of the cancer type means that GPR183 expression is higher or lower in tumor tissue compared to paired normal samples. (C). Bar plot shows the gene expression pro le across all tumor samples and paired normal tissues. The height of bar represents the median expression of certain tumor type or normal tissue.  Integrated network functional analysis of GPR183. (A). The integrated network of GPR183. The colors of pink, purple, blue, yellow separately means physical interactions, co-expression, co-localization and predicted interaction. Node size stands for its weight in the network. (B). GO analysis(BP) of predicted genes. The size of bubbles means numbers of input genes. The color stands for p-values. "X" axis represents fold enrichment.
(C). Pathway analysis of predicted genes. The color stands for p-values. "X" axis represents input numbers of genes.

Figure 5
Overall function analysis of GPR183 in prognosis related cancers. (A). The Venn diagram of commonly upregulated genes in high GPR183 expression group in BRCA, LGG, LUSC, SKCM and UVM. (B). The integrated network of commonly upregulated genes. The colors of pink, purple, blue, yellow separately means physical interactions, co-expression, co-localization and predicted interaction. Node size stands for its weight in the network. (C-E). GO analysis(BP, MF, CC) of commonly upregulated genes. The size of bubbles means numbers of input genes. The color stands for p-values. "X" axis represents fold enrichment. (E). Pathway analysis of commonly upregulated genes. The color stands for p-values. "X" axis represents input numbers of genes.

Figure 6
The Venn diagram of commonly upregulated genes in high GPR183 expression group in Prognosis-Positive(BRCA and SKCM) and Prognosis-Negative(LUSC, LGG and UVM) groups.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.