Identification of genes regulated by the different HIF-3alpha isoforms: tool development for tumor data analysis

Author(s)

Fábio Malta de Sá Patroni

• Document type: Master's Dissertation
• Press: Campinas, SP.
• Institution: Universidade Estadual de Campinas (UNICAMP). Instituto de Biologia
• Defense date: 2019-02-11

Examining board members

Sandra Martha Gomes Dias; Marcelo Mendes Brandão; Mariana Maschietto

• Advisor: Sandra Martha Gomes Dias; Marcelo Falsarella Carazzolle

Abstract

Because of its importance in metabolic adaptation and cancer progression, obtaining functional information about HIFs that will respond to its mechanisms of action is of paramount importance for understanding the disease. Of special interest is the less studied isoform, HIF-3alpha, and its several variants by alternative splicing. Three of the variants (-3alpha1, -3alpha3 and -3alpha9) present DNA interaction domain bHLH and NTAD domain of gene transcription regulation in a way that potentially directly regulate the gene transcription of target genes. In fact, HIF-3alpha9 has already been shown to regulate a set of genes themselves. Other 3 isoforms (-3alpha4, -3alpha5 and 3alpha7) do not present the bHLH domain or the NTAD, being able to promote regulation not as active elements, but as suppressors, as already proven for the HIF-3alpha4 gene. In this work, we developed a platform for the analysis of tumor tissue data available by The Cancer Genome Atlas using the R programming language. The program was called GDCtools and is already available on GitHub (https://github.com/Facottons/GDCtools). It allows analyses based on the separation of tumor groups according to continuous characteristics (genetic expression, level of isoforms, methylation, protein level and survival) or discrete (clinical characteristics), followed by differential gene expression analysis using the EBSeq, DESeq2 and edgeR, accompanied by the possibilities of Venn diagram intersection and pathway enrichment analysis. In addition, it is possible to use co-expression analysis, which correlates the expression variations of different genes in a given tumor and clusters them in color-labeled modules. With this package and extra analyses, we studied the different isoforms of the HIF3A gene. In this study we searched for differentially expressed isoforms between normal and tumor related tissues, followed by the evaluation of the impact of tumor expression on the tumors‘ prognosis. Our studies pointed to two isoforms, HIF-3alpha2 in colon adenocarcinoma and HIF-3alpha3 in prostatic adenocarcinoma, whose alteration in expression level indicated worsening or improvement (respectively) in relapse-free survival of the respective tumors and led to the enrichment of pathways that may explain such impacts. Thus, these isoforms have the potential to be involved in the progression of these tumor types and deserve further laboratory studies to prove their involvement in these diseases (AU)

• FAPESP's process: 15/26059-7 - Identification of genes regulated by the several HIF-3a isoforms: development of tumor patients transcriptomics data analysis tool
• Grantee: Fábio Malta de Sá Patroni
• Support Opportunities: Scholarships in Brazil - Master