A Transient Protein-Protein Interaction Paves the Way for Human Steroid Sulfation

Program: Abstracts - Orals, Poster Previews, and Posters
Session: SUN 203-235-Steroid Hormone Actions, Biosynthesis and Metabolism (posters)
Bench to Bedside
Sunday, April 3, 2016: 1:15 PM-3:15 PM
Exhibit/Poster Hall (BCEC)

Poster Board SUN 227
Jonathan W Mueller*1, Jan Idkowiak1, Cecilia Vallet2, Rebecca Hardman1, Tarsis Gesteira3, Joanne Christine McNelis1, Ian T Rose1, Vivek Dhir1, Edina Rosta3, Shirley Knauer2 and Wiebke Arlt1
1University of Birmingham, Birmingham, United Kingdom, 2University of Duisburg-Essen, Essen, Germany, 3King's College London, London, United Kingdom
Human sulfation pathways depend on provision of the universal sulfate donor PAPS by the two PAPS synthase isoforms PAPSS1 and PAPSS2. Mutations in PAPSS2 have been identified as a monogenic cause of androgen excess presenting with premature adrenarche and polycystic ovary syndrome, due to decreased sulfation of the androgen precursor DHEA by DHEA sulfotransferase (SULT2A1) and hence increased conversion of DHEA to active androgens [NEJM 2009, 360(22):2310-8; JCEM 2015, 100(4):E672-80]. Here we examined why ubiquitously expressed PAPSS1 cannot compensate for the impact of PAPSS2 deficiency on DHEA sulfation.  First, we carried out siRNA-mediated knockdown of PAPSS1/2 in the adrenal cell line NCI-H295R1. Realtime-PCR confirmed >90% knockdown and the impact at protein level was assessed by Western blot and SULT2A1 activity assays. Efficient knockdown of PAPSS2 reduced DHEA sulfation to 30±5%. Strikingly, PAPSS1 knockdown did not impact on DHEA sulfation, providing in vitro evidence for non-overlapping functionality of the two PAPSS isoforms. To test whether subcellular localisation impacts on isoform-specific capacity to support DHEA sulfation, we used HEK293 cells to co-express SULT2A1 with either wild-type PAPSS1/2 or exclusively nuclear or cytoplasmic PAPSS1/2 variants.  Wild-type and nuclear PAPSS1/S2 equally supported DHEA sulfation; however, SULT2A1 activity with exclusively cytoplasmic expression of PAPSS2 was 60% higher than with cytoplasmic PAPSS1; suggesting an isoform-specific protein-protein interaction of SULT2A1 with PAPSS2, but not PAPSS1. Such interaction seems to be transient as it was not amenable to a GFP-trap pulldown. Using a cysteine-reactive crosslinker on NCIh295 cell lysate, we could detect a band of about 100 kDa that was immunoreactive with a SULT2A1 and a PAPSS2 antibody. Overexpressed PAPSS-EGFP and SULT2A1-mRFP fusion constructs resulted in a clear FRET signal. Finally, homology models of full-length PAPSS2 were built and used for docking studies with different structures of human SULT2A1 (1efh, 3f3y and 4ifb). The sulfotransferase was exclusively found at the APS kinase domain near the putative PAPS exit site. Probing this interface by mutagenesis is currently underway. Assuming this novel interaction to be specific to PAPSS2, this may explains the observed different functionality of the two PAPS synthases in facilitating steroid sulfation.

Nothing to Disclose: JWM, JI, CV, RH, TG, JCM, ITR, VD, ER, SK, WA

*Please take note of The Endocrine Society's News Embargo Policy at https://www.endocrine.org/news-room/endo-annual-meeting/pr-resources-for-endo