Identification of putative insulin binding motifs of the insulin receptor

Steve Bottomley¹, Jessica Mitchell², Brian Plewright, Erik Helmerhorst
¹S.Bottomley@curtin.edu.au, Western Australian Biomedical Research Institute, School of Biomedical Sciences, Curtin University of Technology; ²MITCHEJM@ses.curtin.edu.au, Western Australian Biomedical Research Institute, School of Biomedical Sciences, Curtin University of Technology

The aim of this project is to delineate the binding interaction between insulin and its receptor using a combination of experimental and computational approaches.

Overlapping 9 and 15 mer peptides covering the length of the insulin receptor a-subunit sequence were synthesised and measured for their ability to specifically bind ¹²⁵I-insulin. The resulting insulin binding sequences were analysed to:

identify insulin-binding regions of the insulin receptor
identify putative insulin-binding sequence motifs
develop a preliminary insulin-binding scoring matrix that could be used in predictions of insulin binding to protein sequences.

The results indicate at least seven putative insulin-binding regions (R1 to R7) along the length of the insulin receptor and at least six insulin-binding sequence motifs. The amino acid sequence motifs particularly prevalent were (in Prosite format): K-x(0,2)-K; L-x(2,4)-K-x(2,4)-S; C-P-[GT]; and C-x(3,5)-C-x(4,5)-K.

We discuss the possibility that R3 of the insulin receptor (in particular residues 260 to 277) is the putative site of insulin’s principal interaction with its receptor and that this interaction is likely to be electrostatic. We also discuss the potential of sequence motifs, and the insulin-binding scoring matrix, to diagnose insulin binding to other peptides or proteins.