Premium
Bayesian Analysis of E3 Ubiquitin Ligase/AQP2 Interactions in the Renal Collecting Duct
Author(s) -
Medvar Barbara L,
Sarkar Abhijit,
Knepper Mark
Publication year - 2016
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.30.1_supplement.969.30
Subject(s) - ubiquitin ligase , ubiquitin , nedd4 , ubiquitin protein ligases , biology , computational biology , aquaporin 2 , microbiology and biotechnology , genetics , gene , mechanical engineering , water channel , engineering , inlet
Aquaporin‐2 (AQP2) is regulated in part via vasopressin‐mediated changes in protein half‐life that are in turn dependent on AQP2 ubiquitination. Here we addressed the question, “What E3 ubiquitin ligase is responsible for AQP2 ubiquitination?” using large‐scale data integration based on Bayes’ Rule. The first step was to bioinformatically identify all E3 ligase genes coded by the human genome. This analysis used the software ABE (Automated Bioinformatics Extractor) to identify all proteins with HECT, RING‐Finger or U‐box domains and combined the resulting list with data from existing databases. The 377 E3 ubiquitin ligases identified in the human genome have been used to create an online database. Bayes’ analysis ranked these 377 E3 ligases regarding probability of interaction with AQP2 using (1) expression data (the mouse mpkCCD transcriptome and the mouse mpkCCD proteome), (2) co‐localization data of E3 ubiquitin ligases with AQP2 in mpkCCD subcellular fractions (quantitative proteomic data), (3) presence of E3 ligases in AQP2‐expressing tubule segments (rat single‐tubule RNA‐seq data), (4) annotation from UniProt records regarding cytoplasmic versus nuclear localization and (5) likelihood of membrane association based on presence of membrane‐spanning domains or phospholipid interaction domains. Application of Bayes’ Rule to these data identified the top ranked E3 ligases from the analysis (in order of probability): NEDD4=NEDD4L>AMFR>STUB1>ITCH>ZFPL1 . Three of these proteins (Nedd4, Nedd4‐2 and Itchy) are HECT domain proteins. Prior phosphoproteomic analysis has revealed that Nedd4‐2 undergoes a marked increase in phosphorylation in response to vasopressin in the region that mediates protein‐protein interactions (multi‐WW domain region), suggesting a mechanism whereby AQP2 ubiquitination could be regulated.