Architecture and topologies of gene regulatory networks associated with breast cancer, adjacent normal, and normal tissues

Abstract

Most cancer studies employ adjacent normal tissues to tumors (ANTs) as controls, which are not completely normal and represent a pre-cancerous state. However, the regulatory landscape of ANTs and how it differs from tumor and non-tumor-bearing normal tissues is largely unexplored. Among cancers, breast cancer is the most commonly diagnosed cancer and a leading cause of death in women worldwide, with a lack of sufficient treatment regimens due to various reasons. Hence, we aimed to gain deeper insights into normal, pre-cancerous, and cancerous regulatory systems of the breast tissues towards the identification of ANT and subtype-specific candidate genes. For this, we constructed and analyzed eight gene regulatory networks (GRNs), including five different subtypes (viz. Basal, Her2, LuminalA, LuminalB, and Normal-Like), one ANT, and two normal tissue networks. Whereas several topological properties of these GRNs enabled us to identify tumor-related features of ANT; escape velocity centrality (EVC+) identified 24 functionally significant common genes, including well-known genes such as E2F1, FOXA1, JUN, BRCA1, GATA3, ERBB2, and ERBB3 across different subtypes and ANT. Similarly, the EVC+ also helped us to identify tissue-specific key genes (Basal: 18, Her2: 6, LuminalA: 5, LuminalB: 5, Normal-Like: 2, and ANT: 7). Additionally, differential correlation along with functional, pathway, and disease annotations highlighted the cancer-associated role of these genes. In a nutshell, the present study revealed ANT and subtype-specific regulatory features and key candidate genes which can be explored further using in vitro and in vivo experiments for better and effective disease management at an early stage.