This site complies with the HONcode standard for trustworthy health information:
verify here.

Ligand Specificity Between Insulin and IGFr

« Back to Volume 23, Issue 1, March 2007 - Table of Contents

Although the structure of the insulin receptor (IR) has been generally described, high-resolution data remain unknown. A member of the tyrosine kinase receptor family that also includes the type 1 insulin-like growth factor receptor (IGF1R), the IR is a transmembrane glycoprotein with 2 extracellular ligand-binding a- subunits. Each ectodomain consists of 2 leucine-rich repeat domains (L1 and L2) separated by a cys-rich region (CR) in its N terminal half and 3 fibronectin type III domains in its C terminal half. Both L1 and CR have been identified as important for determining IR ligand specificity, based on evidence from IR/IGF1R receptor chimera experiments and mutational analyses. Much of these interpretations were based on inferences from the three-dimensional structure of IGF1R’s L1-CR-L2 domains, that has been determined to 2.6 A resolution.¹

Lou et al determined the crystal structure of the corresponding L1-CR-L2 fragment of IR (IR485) at 2.3 A resolution. Like the L1-CR-L2 fragment of IGF1R (IGF1R462), IR485 did not bind ligand, adopted a bilobal structure, and had N-linked carbohydrate moieties. Global movement of the L2 domain relative to the L1-CR domains was enabled by the structures of both IR485 and IGF1R462. However, the relative orientation of the L2 domains showed a 32 o angle of rotation in both IR molecules, in contrast to a 17 o angle (molecule 1) and 32 o angle (molecule 2) relative to L1 of IGF1R462. The greatest differences between IR and IGF1R individual domains occurred in the regions that determine ligand specificity. The first of these was the left side of the ligand-binding surface of the L1 domain, despite 70% sequence identity between the types of receptors. In IR, the side chain of F39 formed part of the ligand-binding surface involving the second b -sheet; its counterpart in IGF1R, S35, was located differently and not involved in ligand binding. The second major difference involved the CR domain, which had 47% sequence identity between the 2 receptors, and more specifically, its sixth module (0% sequence identity). The IR contained a four-residue insertion and an extra disulfide bond in its sixth module. Although this insertion was pretty small, it formed an a -helix that extended deeper into the putative binding pocket and was kept there by the extra disulfide bond. Furthermore, the abundance of positively charged residues in IR’s CR domain created opposite electrostatic potential to that from the acidic residues in the corresponding IGF1R region. IGF1R’s CR electronegativity complements the electropositive nature of the IGF-I C and D domains, and contributes to IGF binding. The L2 domain of IR was found to closely resemble that of IGF1R.

In addition to delineating the structural differences between the ligand-binding domains of IR and IGF1R, the authors described a possible model for insulin-IR interaction.

Lou M, Garret TP, McKern NM, et al. The first three domains of the insulin receptor differ structurally from the insulin-like growth factor 1 receptor in the regions governing ligand specificity. Proc Natl Acad Sci USA. 2006;103:12429-34.

Editor’s Comment

Not only are insulin (and more specifically, proinsulin) and the insulin-like growth factors (IGFs) structurally related, so too are their receptors. Both the IR and the type 1 IGF receptor (IGF1R) are a 2b 2 transmembrane tyrosine kinases that signal through common pathways. As a further complication, hybrid receptors that mix IR and IGF1R subunits have been identified and are shown to be functionally active.^2,3 Thus, one of the outstanding questions remains: how do these 2 systems achieve specificity? This paper provides one of the answers.

Adda Grimberg , MD

References - (linked to )

« Back to Volume 23, Issue 1, March 2007 - Table of Contents

Last Updated: 6/12/2009