BIOINFORMATICS ANALYSIS OF HEME-OXYGENASE 1 (HO-1) INTERACTOME IN PROSTATE CANCER

ORTIZ EMILIANO GERMAN; PAEZ ALEJANDRA; GIUDICE JIMENA; VALACCO PIA; LEONARDI DAIANA; COTIGNOLA JAVIER; MARTI MARCELO; VAZQUEZ ELBA; GUERON GERALDINE

Bariloche

Congreso; Third South American Symposium in Signal Transduction and Molecular Medicine; 2015

SISTAM

Prostate Cancer (PCa) is the second leading cause of cancer death in American men. The integration of proteomics with appropriate bioinformatics tools, provide promising findings that may effectively translate into clinical applications. Bioinformatics changes the perspective of proteomics, creating a proper scaffold for data storage, analyses and integration. The aim of this work was to use a proteomic and bioinformatics based approach to identify HO-1 molecular partners. Recombinant GST-HO-1 protein complexes were pulled-down and protein complexes were subject to in-solution or in-gel digestion and then LC-MS/MS. The C-18 reverse phase LC-MS/MS rendered a set of 46 HO-1 candidate partners, which were differentially expressed in PC3 cells transfected with GST-HO-1 respect to control cells (transfection with empty vector GST), thus defining our preliminary HO-1 interactome. For this set of proteins we performed a protein interaction network analysis and an ontology analysis. Results revealed an enrichment of proteins associated to several sub-categories of the cellular component, molecular function and biological processes classifications. This list was further clustered into five functional categories: DNA and transcription, RNA processing, actin and transport-related proteins and other proteins. Our working hypothesis emerges from the detection of nuclear molecular partners of HO-1. This set of nuclear acting proteins comprises the HO-1 interactome that mounts into a transcriptional/ RNA processing regulatory machinery, responsible for modulating the HO-1 target genes. We focused our studies on the RNA processing categories. Interestingly, HO-1 interacts with TRIM28, a modulator of signaling pathways during carcinogenesis, that has been reported to ubiquitinate and destabilize p53 in association with members of the MAGE (melanoma antigen) family of cancer testis antigens. MAGE protein family associates to poor prognosis in cancer. Our results show that MAGE A11 is present in PCa cells and that HO-1 pharmacological induction (hemin 24 h, 80 µM) decreases MAGE A11 protein expression, favoring a less aggressive phenotype. Altogether, this is the first approach in building the HO-1 interactome in prostate cancer, which potentially could define a subset of candidate biomarkers for therapeutic intervention.