ISSN 1470-3947 (print) | ISSN 1479-6848 (online)

Endocrine Abstracts (2017) 50 OC1.4 | DOI: 10.1530/endoabs.50.OC1.4

An investigation into sodium-iodide symporter (NIS) dimerization and its impact on radioiodide uptake in thyroid cancer

Rebecca J. Thompson1,2, Alice Fletcher2, Hannah Nieto1,2, Mohammed Alshahrani1,2, Katie Baker1,2, Jonathan W. Mueller1,2, Nicholas H.F. Fine1,2, David J. Hodson1,2, Martin L. Read1,2, Kristien Boelaert1,2, Vicki E. Smith1,2 & Christopher J. McCabe1,2

1Institute of Metabolism and Systems Research, University of Birmingham, Birmingham, UK; 2Centre for Endocrinology, Diabetes and Metabolism, Birmingham Health Partners, Birmingham, UK.

The ability of the thyroid to accumulate iodide via the sodium-iodide symporter (NIS) can be utilised to successfully treat the majority of thyroid cancers with radioiodide. However, approximately 25% of thyroid cancers lose this functional NIS activity and become unresponsive to radioiodide therapy, resulting in a poorer prognosis. Our knowledge of NIS regulation is limited, but as dimerisation of NIS has been proposed, we sought to investigate NIS dimerisation and its impact on radioiodide uptake. Dimerisation of wild-type NIS was confirmed using proximity ligation assays (PLA) in both a thyroid (SW1736) and non-thyroid (HeLa) cell line. To quantitatively assess NIS dimerisation using Förster resonance energy transfer (FRET), novel constructs conjugating NIS to one of the fluorescent proteins citrine (YFP) or cerulean (CFP) were created. YFP/CFP ratio of the NIS-fluorophore constructs increased compared to fluorophores alone (1.71±0.10 vs 1.09±0.16, P<0.05 in SW1736 cells, n=3 and 1.73±0.10 vs 1.13±0.04, P<0.01 in HeLa cells, n=3), further validating NIS dimerisation. To identify residues potentially involved in dimerisation, a homology model of NIS structure was built based on the dimeric crystal structure of the bacterial protein vSGLT using the modelling platform Phyre2. Using site-directed mutagenesis, we then mutated five residues identified from our homology model (D237A, Y242A, T243A, Q471A and A525F), and two putative dimerisation motifs identified in the literature (a glycine zipper motif in transmembrane domain (TMD) 12, with key glycine residues mutated to valine, and a leucine zipper motif in TMD6, with key leucine residues mutated to alanine). PLA suggested that all mutants still retained the ability to dimerise, indicating that dimerisation involves multiple, or as yet undiscovered, residues. In summary, NIS dimerisation has been conclusively demonstrated using two discreet methodologies. Further work is ongoing to determine the critical residues, cellular localisation and regulation of NIS dimerisation and its impact on radioiodide uptake.