Distinct co-expression networks using multi-omic data reveal novel interventional targets in HPV-positive and negative head-and-neck squamous cell cancer

Abstract

The human papillomavirus (HPV) is present in a significant fraction of head-and-neck squamous cell cancer (HNSCC). However, a comprehensive understanding of disease progression profiles comparing HPV+ and HPV-HNSCC cases is still lacking. The main goal of this study was to identify distinct co-expression patterns between HPV+ and HPV-HNSCC and to provide insights into potential regulatory mechanisms/effects (such as methylation and mutation) within the analyzed networks. For conducting this, we selected 276 samples from The Cancer Genome Atlas database comprising data of gene expression, methylation profiles and mutational patterns, in addition to clinical information (HPV status and tumor staging). We further added external information such as the identification of transcription factors to the networks. Genes were selected as differentially expressed and differentially methylated based on HPV status, of which 12 genes were doubly selected, including SYCP2, GJB6, FLRT3, PITX2 and CCNA1. Weight correlation network analysis was used to identify co-expression modules and a systematic approach was applied to refine them and identify key regulatory elements integrating results from the other omics. Three main modules were associated with distinct co-expression patterns in HPV+ versus HPV-HNSCC. The molecular signatures found were mainly related to cell fate specification, keratinocyte differentiation, focal adhesion and regulation of protein oligomerization. This study provides comprehensive insights into complex genetic and epigenetic particularities in the development and progression of HNSCC in patients according to HPV status, identifying unseen gene interactions, and may contribute to unveiling specific genes/pathways as novel therapeutic targets for HNSCC.

Human papillomavirus elevated genetic biomarker signature by statistical algorithm

“…This gene is downregulated in HPV induced tumors. SPRR2G shows downregulation in HPV-positive cases as compared with HPVnegative cases (Costa et al, 2018).…”

“…This gene shows a downregulated pathway and is connected to HPV-positive tumors. This gene is significantly downregulated and shows more methylation in HPV-negative HNSCC (Costa, Boroni, & Soares, 2018).…”

Integration of epigenetic, transcriptomic and proteomic data

“…However, a low overlap is observed between rebound CpGs and genes. Although this is not an uncommon occurrence, and indeed multiple examples can be found in the literature (Costa et al, 2018;Guo et al, 2020;Imran et al, 2021;Yang and Wang, 2015), this observation poses the question on how to integrate DNA methylation and RNA-seq data in case-control studies. The approach used in this study is to analyze DMGs and DEGs from a systemic perspective, meaning looking at the pathways associated with these sets of genes.…”

“…An intuitive approach to the problem is to analyze the omics separately, and to select shared molecular events (i.e., changes associated to the same gene across omics) in a second step. Indeed, examples of this approach can be found in analyses of paired epigenetics/transcriptomics data (Costa et al, 2018;Imran et al, 2021) and transcriptomics/proteomics data Dumitriu et al (2016). However, a limitation of this strategy is that the overlap between the omics may be poor (Guo et al, 2020;Yang and Wang, 2015), as the analysis treats the multi-omics data as independent.…”

Gene regulatory network reconstruction incorporating 3D chromosomal architecture reveals key transcription factors and DNA elements driving neural lineage commitment

“…To date, GRN analysis has been used successfully by ourselves and others to pinpoint factors and pathways underpinning normal development (Viiri et al 2019;M.-A. Mendoza-Parra et al 2016), cancer progression Costa, Boroni, and Soares 2018;Verfaillie et al 2015) and transdifferentiation (M.-A. Mendoza-Parra et al 2016;Ieda et al 2010).…”