Therapeutic and prognostic potential of GPCRs in prostate cancer from multi-omics landscape

Abstract

Prostate cancer (PRAD) is a common and fatal malignancy. It is difficult to manage clinically due to drug resistance and poor prognosis, thus creating an urgent need for novel therapeutic targets and prognostic biomarkers. Although G protein-coupled receptors (GPCRs) have been most attractive for drug development, there have been lack of an exhaustive assessment on GPCRs in PRAD like their molecular features, prognostic and therapeutic values. To close this gap, we herein systematically investigate multi-omics profiling for GPCRs in the primary PRAD by analyzing somatic mutations, somatic copy-number alterations (SCNAs), DNA methylation and mRNA expression. GPCRs exhibit low expression levels and mutation frequencies while SCNAs are more prevalent. 46 and 255 disease-related GPCRs are identified by the mRNA expression and DNA methylation analysis, respectively, complementing information lack in the genome analysis. In addition, the genomic alterations do not exhibit an observable correlation with the GPCR expression, reflecting the complex regulatory processes from DNA to RNA. Conversely, a tight association is observed between the DNA methylation and mRNA expression. The virtual screening and molecular dynamics simulation further identify four potential drugs in repositioning to PRAD. The combination of 3 clinical characteristics and 26 GPCR molecular features revealed by the transcriptome and genome exhibit good performance in predicting progression-free survival in patients with the primary PRAD, providing candidates as new biomarkers. These observations from the multi-omics analysis on GPCRs provide new insights into the underlying mechanism of primary PRAD and potential of GPCRs in developing therapeutic strategies on PRAD.

Keywords: G protein-coupled receptors; multi-omics; prognostic model; prostate cancer; virtual screening.

FIGURE 1

Differentially expressed GPCRs (DEGpcrs) between 495 primary prostate tumor samples and 151 normal samples. (A) Volcano plot showing the DEGpcrs. Red and blue dots represent the significantly up- and downregulated genes, respectively. (B) Barplot showing subfamily distribution of the DEGpcrs. (C) G protein linkage of all the 767 GPCRs (right), the 293 GPCRs expressed in both prostate tumor and normal samples (middle), and the 46 DEGpcrs (left).

FIGURE 2

Statistics of GPCR Mutations. (A) A landscape of the GPCRs mutated in 276 primary prostate tumor samples generated by Maftools visualization module. (B) Boxplots of the expression levels of SSTR1 (left) between the SSTR1 mutated and unmutated groups, and OR51D1 (right), an example of a gene that is dysregulated between the mutated and unmutated tumor samples of the other GPCR.

FIGURE 3

SCNAs of GPCRs in PRAD. (A) The number of heterozygous/homozygous deletions, and low-/high-level amplifications for all GPCRs. (B) Scatter plot of recurrent amplifications and deletions in 500 primary prostate tumor samples. (C) Box plots showing the expression of the significantly deleted and amplified GPCRs between the tumor and normal samples. (D) Distribution of Pearson’s correlation coefficients between the GPCR expression and its linear SCNA values.

FIGURE 4

Differential methylation of GPCRs in the 491 PRAD patients. (A) Bar plot and Upset showing the number of DMGpcrs. (B) Bar plots and scatter plots showing the number and identity of DMEGs, respectively.

FIGURE 5

The predicted binding modes of the ligands with (A) GPR160 and (B) CRHR2 at the ligand binding pocket (dashed box). Different ligands are represented by different colored sticks, salmon: Irinotecan, yellow: Cinacalcet, orange: Simeprevir, and cyan: Glecaprevir.

FIGURE 6

The diagnostic performance of the prognostic model. (A,B) Kaplan–Meier survival analysis showing PFI differences between the high-risk (red) and low-risk (blue) groups in the training set (A) and test set (B). (C,D) The ROC curve showing the AUC value of the risk model in the training set (C) and testing set (D).