The insulin and IGF-I receptors are members of the superfamily of receptor tyrosine kinases (RTKs). Unlike most RTKs that are single-chain monomeric transmembrane polypeptides, the insulin and IGF-I receptors are covalent dimers composed of two extracellular α subunits and two transmembrane β subunits containing the tyrosine kinase domains. The α subunits contain the ligand binding sites, of which at least three subdomains have been defined by photoaffinity crosslinking, alanine-scanning mutagenesis or minimized receptor constructs. All RTKs are dimeric or oligomeric in the ligand-activated form. The residues of insulin involved in receptor binding have been mapped by alanine-scanning mutagenesis. They form at least two major epitopes that partially overlap with the dimer- and hexamer-forming surfaces of the insulin molecule, and we propose that insulin is using those surfaces to asymmetrically cross-link the two receptor α subunits. This mechanism provides a structural basis for high affinity binding and negative cooperativity, and probably also operates in the IGF-receptor interaction. It also provides a structural basis for the approximation and transphosphorylation of the kinase domains and triggering of the signalling cascade.

Evolutionary Biology of the Insulin Peptide Family and Their Receptors

Insulin and the insulin-like growth factors (IGF)-I and -II belong to a phylogenetically ancient family of peptide hormones and growth factors^1-3 that play a fundamental role in the control of essential cellular and physiological processes such as the cell cycle, survival or apoptosis, cell migration, proliferation and differentiation, and body growth, metabolism, reproduction, and longevity.

Proteins of the insulin superfamily are synthesized as prepro-proteins consisting of 4 domains (pre, B, C, A). These are then processed by proteolytic removal of the pre domain and in some cases the C- domain and fold to form mature proteins, in which the A and B domains are covalently linked by two disulfide bonds. The IGF precursors have additional C-terminal D and E peptides; the latter is removed proteolytically post-translationally. The basic fold is shared for all molecules in the superfamily whose structure is known although the lengths of the secondary structural elements vary and the connecting loops can exhibit several conformations; the B domain contains a single α helix which lies across the 2 helices of the A domain (Fig. 1).

Figure 1

The insulin peptide family protein folds. A) Insulin. The A-chain is red, the B-chain yellow. Adapted from Baker E et al. Phil Trans R Soc Lond 1988; B19:369-456. B) IGI. The A domain is red, the B domain yellow, the C domain blue and the D domain green. (more...)

These properties have lead to the development of criteria for the identification of superfamily members from data obtained from genomic and cDNA cloning studies.

Candidate open reading frames should have at least a putative signal peptide, a B domain and an A domain defined as follows:

1. A domain predicted to contain two helices joined by a loop.
2. B domain with an extended N-terminal coil followed by a tight turn and a central helix.
3. A hydrophobic core forming the interface between the A and B domains.
4. At least three disulfide bonds formed by conserved cysteine residues.

In addition in most family members the B and A domains are encoded by individual exons separated by an intron at the 3' end of the coding region for the B domain. These criteria encompass the plethora of vertebrate and non vertebrate insulin family members identified during the last decade.

In humans, the insulin-like peptide family comprises ten members, the closely related insulin and insulin-like growth factors (IGF)-I and II, and the seven peptides related to relaxin (INSL/RLFs).^4,5 While insulin and the IGFs bind to receptors belonging to the superfamily of receptor tyrosine kinases (RTKs), the relaxin-like peptides were recently shown unexpectedly to bind to leucine-rich repeat-containing G protein-coupled receptors (LGRs),⁴ involved in the development and physiology of the reproductive tract.^6,7

The sequence of the zebrafish and fugu genomes has revealed the existence of two insulin genes; evidence suggests that a genome duplication event occurred on the lineage leading to teleost fish.⁸

Several insulin-like family members have been identified in invertebrates as well, including bombyxins in the silkworm (38 putative genes), Bombyx mori⁹ and related peptides in the sweet potato hornworm Agrius convolvuli,¹⁰ a locust insulin-related peptide (LIRP) in the grasshopper Locusta migratoria,^11,12 seven different molluscan insulin-like peptides (MIP) in the pond snail, Lymnea stagnalis, an insulin-like gene in the sea slug Aplysia californica,¹³ and seven insulin-like genes in the fruit fly Drosophila melanogaster.¹⁴ In addition, three new gene families of insulin-like peptides with atypical disulphide bond pattern (38 putative peptides in all) have been recently identified in the nematode C. elegans that may be ligands (both agonists and antagonists) for the daf-2 insulin-like receptor, involved in reproduction, growth and adult life span.¹⁵ Surprisingly, one of the C. elegans peptides (INS-6) was reported to bind and activate the human insulin receptor, despite absolute lack of conservation of any of the amino acids involved in mammalian receptor binding (see below),¹⁶ a finding which warrants further investigation. In contrast to the abovementioned insulin-related peptides, the reported sequence of an insulin-like peptide in the sponge Geodia cydonium¹⁷ is suspiciously too close to human insulin (60-80%) to represent a true phylogenetically ancient ancestor1 and probably represents contamination by rodent material, as later acknowledged by the authors (W.E.G., Muller, personal communication quoted in ref. 18).

Evidence for the existence of receptors related to the insulin receptor has also been reported in various invertebrate species such as D. melanogaster,¹⁹ B. mori,²⁰ the mosquitos Aedes aegypti²¹ and Anopheles gambiae,²² the molluscs, L. stagnalis,²³ Biomphalaria glabrata,²⁴ Aplysia californica²⁵ and the Pacific oyster Crassostrea gigas,²⁶ C. elegans,²⁷ the human blood flukes Schistosoma mansoni²⁸ and Schistosoma japonicum,²⁹ the tapeworm Echinococcus multilocaris,³⁰ and the cnidarian Hydra vulgaris³¹ as well as sponges, although in this last case identification was based only on homologies to the insulin receptor kinase domain.^32,33

Interestingly, the protochordate amphioxus (Branchiostoma californiensis) contains a single gene for an insulin-like peptide and a related receptor, both of which have hybrid characterictics of the insulin and IGF systems and may therefore represent the ancestral genes,^34,35 both have undergone duplications in the transition from protochordates to vertebrates. This has been corroborated by the sequencing of the genome of the sea squirt Ciona intestinalis, another protochordate.³⁶

The concept that the Amphioxus insulin-like peptide represents the common ancestral gene of insulin and IGFs was challenged by the discovery of both an insulin and an IGF molecule in the urochordate tunicate Chelyosoma productum;³⁷ however, the predicted aminoacid sequences for the tunicate insulin and IGF peptides share a remarkably high sequence identity including in the evolutionarily highly variable C domain, and Chan and Steiner therefore suggested that the tunicate sequences are the product of a recent gene duplication event.¹

While a detailed discussion of the evolutionary biology of the insulin signaling system is beyond the scope of the review and the evolutionary record is of necessity very incomplete, several points deserve further comment. Firstly prior to the emergence of the vertebrates, the full spectrum of the biological effects of the signaling system (regulation of metabolism, growth, cell survival, feeding, longevity and reproductive function and developmental timing) appear to be mediated by a single receptor. Secondly there has been a high degree of conservation of function from the insects, as manifested in Drosophila, to mammals. However the nematode C. elegans seems to be an exception to this general rule. Firstly the existence of the very large number of genes encoding insulin-like peptides (38) in this species is atypical, although a similar number exists in Bombyx. The genes for many of these are organized into clusters of 3-7. This clustering, the tandem arrangement and sequence similarity between genes in a given cluster suggest that these clusters may have arisen relatively recently by gene duplication. However sequence diversity within the clusters suggests that they are continuing to rapidly evolve. The completion of the sequencing of the C. briggsae genome³⁸ should provide further insights into this issue. Recent database mining suggests the presence of multiple insulin receptor-like proteins as well, with a very short C-rich region between the two L domains.³⁹ Secondly C. elegans seems to be the only species where both insulin-like peptides function as both agonists and antagonists. Finally signaling by the insulin receptor homolog in this species appears to inhibit growth and anabolism in constrast to the stimulation universally seen in other species.^40,41 However, inhibiting the insulin/IGF system appears to prolong lifespan in C.elegans as well as in Drosophila,⁴² rodents⁴³ and possibly humans.⁴⁴

Within vertebrate species, multiple types of receptor are observed. In fish there are genes encoding multiple receptor types with high homology to either mammalian insulin or IGF-I receptors but not mammalian IRR (insulin receptor-related receptor).¹ The Fugu rubripes genome⁴⁵ has four genes encoding receptors from this family, two with high homology to the mammalian insulin receptors and two with high homology to mammalian IGF-I receptors. The absence of IRR type receptors in fish is corroborated by the failure to identify DNAs encoding such proteins by molecular cloning studies in multiple species of fish, where insulin and IGFI-like receptor have been identified by such techniques. In reptiles and amphibian only insulin and IGF-I receptors have been identified.^46-51 However this cannot be considered as definitive until genomes of examples of each have been sequenced. In mammals there are insulin, IGF-I and IRR receptors and all the proteins appear to be highly conserved. The same three types of receptor have also been identified in birds.

Throughout evolution, the prototypical domain structure of these proteins (see below) appears to have been well conserved. However certain homologs have either additional N terminal or C terminal domains or both. These can vary considerably in size from 30 to 40 amino acids (N- and C-terminal extensions of Bombyx IR) to 200 to 400 amino acids (N- and C-terminal extensions of Drosophila IR and C terminal extension of Daf-2). The primary structures of these domains vary considerably. Of particular interest are the C-terminal domains of the Drosophila and C. elegans homologs. These contain several YXXM motifs, potential docking sites for PI3 kinase when phosphorylated, suggesting that they may play a role in signal transduction by these receptors.

The signal transduction pathways used by these receptors also appear to be well conserved throughout evolution.^{14,40,44,52-54} While detailed discussion of this topic is beyond the scope of this review, it should be mentioned that there is extensive genetic evidence that the insulin receptor homologs of C. elegans and Drosophila both utilize signal transduction pathways closely related to the mammalian signal transduction pathway mediated by PI3 kinase activation in response to receptor autophosphorylation. Genetic evidence also suggests that the biological effects mediated by these receptors in mammals, nematodes and fruit flies are highly conserved. Insulin receptor and IGF receptor knockout studies in mice indicate that in addition to regulation of intermediary metabolism and growth, signaling via these receptors is involved in the regulation of energy homeostasis, regulation of reproductive function and of longevity.⁵⁵ Similar biological roles for signaling by the appropriate receptor homologs have been demonstrated in mutational studies in Drosophila⁴⁰ and C. elegans.⁵⁶

It has been hypothesized that the ancestral insulin-like gene functioned primarily as a “mitogenic” growth factor, and that its duplication allowed insulin to develop as a metabolic regulator while the mitogenic activity was retained by the IGF-I gene.^1,57

Structure of the Insulin and IGF-I Receptor Genes and Predicted Protein Tertiary Structure

As mentioned above, the insulin, IGF-I and IRR receptors are members of the RTK family. In humans, the RTK superfamily comprises ˜60 members distributed into ˜20 subfamilies depending on the modular architecture of their extracellular domains and the degree of identity in their intracellular tyrosine kinase domains (Fig.2). The importance of these receptors is demonstrated by the discovery of a growing number of congenital genetic syndromes linked to gain-of-function (constitutive activation) or loss-of-function (inactive or dominant-negative receptor) mutations.⁵⁸ Some of the RTKs are oncogenes⁵⁹ and at least a dozen of the RTK families, including the IGF-I and insulin receptors^59,60 have been implicated in human cancers due to amplification, overexpression, loss of parental imprinting or somatic gene mutations. Consequently, RTKs are major targets for anticancer therapy.^61-64 The principles for designing structure-based agonists and antagonists of RTKs have been recently reviewed.⁶⁵

Figure 2

Modular structure of the receptor tyrosine kinases. This figure is adapted from reference 113, updated with information compiled from references 58, 59, 223-225 and various databases such as Pfam. We found that all published figures on the domain organisation (more...)

The receptor for insulin was first characterised biochemically by direct binding studies using radiolabelled insulin in the early seventies.^66-69 Evidence for its subunit structure was provided in the early eighties,^70,71 as was the demonstration that it is a receptor tyrosine kinase,^72-74 which catalyzes the transfer of the γ phosphate of ATP to tyrosine residues on protein substrates, the first being the receptor itself. Cloning of insulin receptor complementary DNA (cDNA) was achieved in the mid-eighties,^75-77 soon followed by that of the IGF-I receptor⁷⁸ and by the sequencing of both genes,^79,80 giving valuable insight into the receptor structure and organisation.

The gene for a mammalian related receptor (insulin receptor-related receptor or IRR) was identified in humans and guinea pigs in 1989,⁸¹ with expression in heart, skeletal muscle, kidney, liver, pancreas and sympathetic ganglia.⁸² No ligand for this receptor has been so far identified. The mouse knockout has no phenotype, so the physiological relevance of this receptor has been questioned. However, it has recently been shown that it is required for testis determination in mice since XY mice that are mutant for all three insulin related receptors (Ir, Igf1r and Irr) develop ovaries and show a completely female phenotype.⁸³ Also, it has been reported that coexpression of IRR and IGF-IR correlates with enhanced apoptosis and differentiation in human neuroblastomas.⁸⁴ Thus, it is quite possible that the IRR plays a role in modulating the activity of the insulin and IGF-I receptors, possibly through the formation of hybrid receptors, similar to the role of the ligand-less ErbB2 receptor in the EGF receptor family.

The insulin and IGF-I receptors differ from the majority of RTKs in that they are covalent dimeric structures, although all the RTKs dimerize or oligomerize upon ligand binding,⁸⁵ resulting in activation of the kinase by transphosphorylation. A claim that monomeric insulin receptor mutants could autophosphorylate⁸⁶ was recently found to be artifactual.⁸⁷

The insulin receptor has a modular structure encoded by a gene with 22 exons and 21 introns.⁸⁸ The receptor sequence modules encoded by the various exons is shown in Fig. 3. Exon 11 is alternatively spliced, resulting in two isoforms (A and B) of the insulin receptor differing by the absence or presence of a 12 residue-sequence (717-729). The physiological significance of this alternative splicing is still unclear; it is absent in the IGF-I receptor which has no equivalent to exon 11. The two isoforms differ slightly in affinity for insulin,^89-91 but the A isoform has significantly higher affinity for IGF-I (40 nM vs 350 nM)⁹⁰ and IGF-II (close to that of insulin).⁹² The IGF-I receptor binds IGF-II with a lower affinity than IGF-I and insulin with a 500-fold lower affinity.⁹³

Figure 3

Modular structure of the insulin receptor. Left: cartoon of the α2β2 structure of the insulin receptor, drawn to scale. On the left half of the receptor, spans of the 22 exon-encoded sequences. On the right half, spans of predicted protein (more...)

The deduced amino acid sequences of the receptors indicate that they are synthesized as single chain preproreceptors with a 30-residue signal peptide, which is cleaved cotranslationally; the single chain precursor is glycosylated cotranslationally, and folded and dimerized under the guidance of the chaperones calnexin and calreticulin⁹⁴ prior to transport to the Golgi apparatus where it is processed at a tetrabasic RKRR furin protease cleavage site (732-735) to yield the mature α₂β₂ receptor. Interestingly, the B-isoform of the proreceptor binds insulin with high affinity but not the A-isoform.⁹⁵

In cells expressing both insulin and IGF-I receptors, heterodimeric hybrid receptors are formed consisting of one half of each. They have been reported to bind IGF-I with high affinity and insulin with low affinity;⁹⁶ the relative affinities are dependent of the insulin isoform involved.⁹⁷ Their physiological role is unknown.

Comparative sequence analysis of the insulin/IGF-I receptors and the related EGF receptor families had generated a hypothetical tertiary structure of the extracellular part of the receptor, suggesting that the N-terminal half consists of two homologous globular domains (L1, residues 1-120 and L2, residues 211-428), separated by a cysteine-rich region.⁹⁸ The Cys-rich region was predicted to consist of a series of disulfide-linked modules similar to those found in the tumor-necrosis factor (TNF) receptor and later in laminin.⁹⁹ All these predictions were confirmed by the determination of the crystal structure of this domain (see below). The C-terminal half of the receptors was predicted to consist of three fibronectin type III (FnIII) domains, each with a seven-stranded β-sandwich structure.^100-102 The second FnIII domain comprizes the C-terminal part of the α-subunit and the N-terminal part of the β-subunit and contains a large insert domain of ˜120-130 residues of unknown structure containing the site of cleavage between a- and β-subunits. The intracellular portion of the β-subunit contains the kinase catalytic domain (980-1255) flanked by two regulatory regions, a juxtamembrane region involved in docking insulin receptor substrates (IRS) 1-4 and Shc as well as in receptor internalization, and a C-terminal tail containing two phosphotyrosine binding sites. The detailed organization of the modular domains of the insulin receptor is shown in Fig. 3. The IGF-I receptor has a very similar organization with sequence homology varying from 41% to 84% depending on the domain, being maximal in the kinase domain (see ref. 103 for details).

Modular Receptor Structures Elucidated by X-Ray Crystallography

The 3D structure of the L1/Cys-rich/L2 (L1CL2) domain fragment (amino acids 1-460) of the IGF-I receptor has been solved at 2.6 Å resolution^103,110 (Fig. 4). An extended bilobed structure (40 x 48 x 105 Å) comprizes the two globular L domains with a new type of right-handed β-helix fold flanking the Cys-rich domain. They appear to be part of the leucine-rich repeat superfamily.¹¹¹ While L1 (1-150) contacts the Cys-rich domain along its length, there is minimal contact with L2 (300-460). The Cys-rich domain comprizes an array of disulfide-linked modules resembling as predicted those in TNF and laminin. The different orientations of L1 and L2 relative to the Cys-rich domain are probably an artifact of crystal packing and the position of L2 is probably more parallel to L1 in the native structure, as shown in the recently determined structures of the structurally related EGF receptor complex with EGF or TGFa (for review see ref. 112). The flexibility between the Cys-rich domain and L2 may be important for ligand binding.¹¹³ A cavity of ˜30 Å diameter occupies the centre of the molecule and represents a potential binding pocket, although this construct does not bind IGF-I (see below for further discussion). Fig. 4 shows also the equivalent domains (3-468) of the insulin receptor modelled on the IGF-I structure.

Figure 4

Structures of the insulin and IGF-I receptors and their ligands, and mapping of binding domains. Left: 3-D structure of the L1CL2 domain of the IGF-I receptor determined by X-ray crystallography. The amino acids determined by alanine scanning mutagenesis (more...)

The 3D structures of the tyrosine kinase domain of the insulin receptor, both in the inactive state (unphosphorylated)¹¹⁴ and in the active state (tris-phosphorylated in combination with an ATP analogue and a peptide substrate)¹¹⁵ have been determined. The structure of the IGF-I receptor kinase in complex with an ATP analogue and a specific peptide substrate has also been solved.¹¹⁶ The structures have revealed the determinants of substrate preference for tyrosine rather than serine or threonine and a novel autoinhibition mechanism whereby Tyr 1162, one of the three tyrosines that are autophosphorylated in the activation loop in response to insulin (1158, 1162, 1163) is bound in the active site, hydrogen bonded to a conserved Asp 1132 in the catalytic loop. Tyr 1162 in effect competes with protein substrates before autophosphorylation. In the activated state, Tyr 1163 becomes hydrogen-bonded to a conserved Arg 1155 in the beginning of the activation loop, which stabilizes the repositioned tris-phosphorylated loop. Mutation of the Asp 1161 (which participates in several hydrogen bonds that stabilize the closed conformation of the loop) to Ala substantially increases the ability of the unphosphorylated kinase to bind ATP. The structure of this mutant kinase has also been solved.¹¹⁷ Covalent dimerization of the soluble insulin and IGF-I receptor kinase domains fused to the homodimeric glutathion-S-transferase (GST) resulted in 10-100-fold increases in the phosphotransferase activity in both the auto- and substrate phosphorylation reactions and rendered the autophosphorylation reaction concentration-independent.¹¹⁸

Ligand Binding Properties

The binding of insulin to its receptor, whether studied on whole cells, purified membranes or purified receptor protein in solution, is complex (Fig. 5), as has been appreciated since the very first binding studies.^57,119-121 Scatchard plots are curvilinear, indicating the coexistence of high and low affinity binding sites. Dissociation studies have shown that the dissociation rate of prebound ligand is accelerated by the presence of unlabelled ligand, consistent with the existence of negative cooperativity between binding sites. Dose-response curves for the accelerated dissociation are bell-shaped (self-antagonism) with a loss of the acceleration of dissociation at concentrations of insulin over 100 nM. See reference 57 for a detailed discussion. IGF-I binding to its receptor shows a similar phenomenology, except that the dose-response curves for accelerated dissociation are sigmoid with no loss of response at high concentration.¹²²

Figure 5

Ligand binding properties of the insulin and IGF-I receptors. A) Curvilinear Scatchard plot of insulin binding (IGF-I's is similar). B) Accelerated dissociation of labelled insulin, the hallmark of negative cooperativity (IGF-I gives similar data). C) (more...)

Alterations in the various components of this complex kinetic behaviour have been observed with certain amino acid substitutions or deletions in the insulin or receptor molecule, or with various minimized receptor constructs, and have provided essential clues as to the nature of the ligand binding mechanism and established the reality and specificity of the negatively cooperative behaviour (see below).

Receptor Crosslinking with Bifunctional and Photoreactive Ligands

Early crosslinking experiments using radiolabelled insulin and bifunctional N-hydroxysuccinimide esters demonstrated that insulin's B-chain was crosslinked to a 55 kDa chymotryptic N-terminal fragment of the receptor α-subunit.¹²³

Photoaffinity labelling of affinity-purified insulin receptor with a radioactive insulin photoprobe derivatized at Lys B29 labelled a 23-kDa fragment suggested to comprize residues 205-316, i.e., most of the Cys-rich region, on the basis of its immunoreactivity.¹²⁴ In contrast, insulin with a different photoaffinity reagent at B29 labelled a 14-kDa tryptic fragment shown by N-terminal sequencing to start at Leu 20 and assumed to extend to approximately Asn 120,¹²⁵ but due to glycosylation more probably to Arg 86 or Lys 102.¹⁰³ Subsequent studies with chimeric or alanine-mutated insulin receptors have not supported a role for the Cys-rich domain in insulin binding (unlike IGF-I binding) but confirmed the importance of the N-terminal sequence (see below). Insulin derivatized at Phe B1 labelled a tryptic fragment of the soluble receptor ectodomain starting at Gly 390 and estimated to extend to Arg 488, covering the domains encoded by exon 6 and 7 (second half of L2 and first FnIII domain).¹²⁶ The importance of this region in insulin binding is supported by other experimental evidence (see below). Two groups showed that insulin derivatized at Phe B25 crosslinks covalently to the insulin receptor,^127,128 the contact site was mapped to the 15-residue sequence Thr 704 to Lys 718 (nomenclature for B isoform) near the end of the A-chain by sequencing of a photo-labeled peptide generated by Lys-C-endoproteinase digestion, in the insert within the second FnIII domain.¹²⁸ The fact that photoaffinity labels only 4 residues apart at the C-terminus of the insulin B-chain label segments of the receptor located respectively at the very N-terminus and at the very C-terminus of the a subunit more than 550 residues apart, strongly suggests that these receptor regions must be located close together in space despite their wide separation in the primary sequence.

All the available evidence suggests that all the insulin contact sites are located in the α subunit.

Definition of Ligand Binding Specificity Using Chimeric Insulin/IGF-I Receptors

Extensive studies using a variety of chimeric receptor constructs with swapped domains of the insulin and IGF-I receptors have unequivocally demonstrated that the determinants of ligand binding specificity reside in different segments of the two receptors.^129-131 Both whole receptor constructs and soluble ectodomains have been used with relatively consistent results. These data have been recently exhaustively reviewed.^103,132 In brief, the Cys-rich region plus flanking regions from L1 and L2 are prime requirements for IGF-I binding, especially the residues 253-266 in the variable loop of module 6 in the Cys-rich region. Alanine scanning studies suggest that this loop does not directly participate in ligand binding implying that its role in modulating affinity must be indirect. The loop in the insulin receptor exhibits significant charge differences from that of the IGFR; two lysine and two arginine residues in the insulin receptor loop compared to two glutamates and one aspartate in the IGFR loop, suggesting it might produce an unfavourable charge environment in the putative binding cavity. In addition the IR loop is significantly longer than that of the IGFR, possibly limiting steric access of the bulkier IGF molecule to the binding site. This mechanism is supported by the finding that a mini-IGF-I with a shortened C-domain and a deleted D domain exhibits a 5 fold increase in its affinity for the insulin receptor.¹³³ Further support is provided the finding that shortening the insulin receptor loop increases its affinity for IGF-I (Whittaker and Groth, unpublished observations). In contrast the N-terminal residues 1-137 in the L1 domain of the insulin receptor (in agreement with the photoaffinity labelling data of Wedekind et al¹²⁵) and residues 325-524, comprising most of the L2 domain and part of the first FnIII domain, are important determinants of insulin binding. Studies with chimeric receptors have localized the N-terminal element contributing to insulin specificity to residues 1-68 and Kjeldsen et al have further suggested that F39 in the insulin receptor may play a significant role in this.¹³⁴

Natural Receptor Mutations That Affect Insulin Binding in Syndromes of Extreme Insulin Resistance

Naturally occuring mutations in the insulin receptor have been characterized in several dozen patients with genetic syndromes associated with extreme insulin resistance: acanthosis nigricans type A, leprechaunism (Donohue syndrome), Rabson-Mendenhall syndrome¹³⁵ and congenital fiber-type dysproportion myopathy.^136-138 Most of the patients carry compound heterozygous mutations, exceptionally homozygous. The functional consequences of the mutations fall into five classes: impaired biosynthesis of full-length receptor due to premature termination mutations, impaired transport to the cell surface due to perturbed protein folding, impaired tyrosine kinase activity, alterations in binding affinity and/or negative cooperativity, and accelerated receptor degradation.¹³⁵ One mutation can affect several of these parameters. A large number of these mutations alter the tyrosine kinase domain (reviewed in ref. 114).

The mutations that affect binding affinity or other ligand binding properties provide clues to the localization of the binding domains that complement the in vitro studies. The majority of mutations that impair ligand binding have been found in the exon 2-encoded L1 domain. A natural mutation at Asn 15 (to Lys) decreased insulin binding 5-fold,¹³⁹ and the importance of this residue was confirmed later by alanine scanning mutagenesis (see below). A natural mutation of Asp 59 to Gly decreased binding affinity 4-fold.¹⁴⁰ A natural mutation of Arg 86 to Pro results in lack of transport to the plasma membrane, loss of binding affinity and a constitutively activated kinase.^141,142 A natural mutation of Leu 87 to Pro reduced the affinity 6-fold,¹⁴³ also confirmed by alanine mutagenesis. A natural deletion of Lys 121 created a temperature-sensitive alteration in insulin binding and negative cooperativity.¹⁴⁴ A mutation of Val 140 to Leu has been reported, but no assessment of the binding affinity was done.¹⁴⁵

A mutant receptor with Arg 252 mutated to Cys in the Cys-rich domain also showed reduced affinity (as well as endocytosis) but showed normal signalling; the abnormal Cys 252 is close enough to the first Cys at position 8 in the L1 domain to create an abnormal disulfide that may lock the L1 domain in a conformation unfavorable for high affinity binding.¹⁴⁶ Two patients with a mutation to His at the same position have been described;^147,148 both cell surface expression and binding affinity were decreased.

Studies of a receptor with Ser 323 mutated to Leu in the L2 domain (found in two patients with Rabson-Mendenhall syndrome^149,150 and one leprechaun patient¹⁴⁷) have generated somewhat conflicting results, Roach et al finding that it decreases binding affinity of the full length receptor without perturbing intracellular transport and membrane insertion while Whittaker and Mynarcik found that both binding affinity and secretion of the recombinant extracellular domain were impaired when that mutation was introduced, suggesting a disturbed structure. In the latter study these effects were found to be more profound with mutation of the A isoform and furthermore mutation to Ala had no significant effect on binding.¹⁵¹

Another leprechaun patient had a homozygous Asn 431 to Asp mutation, in the L2 domain; this decreased processing and receptor signalling but retained substantial binding affinity, indicating that no major disruption of the binding site was caused by this mutation.¹⁴⁷

Finally, an extensively studied mutation of Lys 460 to Glu which retarded ligand dissociation at acid pH and enhanced negative cooperativity, and accelerated receptor degradation, pointed to a possible key role of the C-terminal region of the L2 domain in negative cooperativity, leading to further exploration of this domain (see below).¹⁵²

Mapping of Ligand Binding Sites on the Insulin and IGF-I Receptors by Site-Directed and Alanine-Scanning Mutagenesis

Alanine scanning studies of the secreted recombinant insulin receptor ectodomain have provided the most definitive demonstration that the domains implicated in ligand binding by affinity labeling studies and chimeric receptor experiments are in fact components of a ligand binding site and have also provided a more detailed functional characterization of this binding site. Alanine scanning studies of the L1 domain of the insulin receptor¹⁵³ (Fig.4) indicate that side chains of residues in the first five turns of the β helix form two functional epitopes with discrete footprints on the base of the domain. The first of these is composed of Asp 12, Arg 14 and Asn 15; Gln 34, Leu 36 and Leu 37; Phe 64; Val 94, Glu 97; Glu 120 and Lys 121 which are located in the second β strands of turns 1-5 of the β helix, respectively. Those amino acids, providing the majority of the free energy of binding (Arg 14, Asn 15 and Phe 64, whose mutation to alanine inactivates insulin binding), are located centrally within the footprint and those making more minor contributions are more peripheral, as has been observed for interactions of growth hormone with its receptor.¹⁵⁴ The second functional epitope is located in the bulge between the first and second β sheets of the β helix and consists of Leu 87, Phe 89, Asn 90 and Tyr 91. This epitope had been earlier identified by site-directed mutagenesis of Phe 89 to Leu.¹⁵⁵

Alanine scanning mutagenesis of the secreted receptor also confirms that the C-terminus of the α subunit, amino acids 704-715, is also an essential component of this ligand binding site.^156,157 In this region, alanine mutations of Phe 705, Glu 706, Tyr 708, His 710 and Asn 711 inactivate insulin binding and mutations of Leu 709 and Phe 714 produce a 30-100 fold reduction in affinity, while mutations of Asp 707, Val 713 and Val 715 were without effect. A conflicting result has been reported regarding mutation of Asp 707 to Ala, found in a leprechaun patient, with abolition of binding.¹⁵⁸ Analogue binding studies reveal that the free energy contribution of this region to insulin binding is considerably greater than that of the L1 domain, consistent with the observation that both the L1 and L1/CRD/L2 domains are incapable of binding insulin when expressed on their own. In the insulin receptor, the cysteine rich domain does not appear to be involved in insulin binding. Alanine scanning of ligand accessible residues just adjacent to the L1 domain fail to demonstrate a functional epitope for insulin binding¹⁵⁹ and more distal residues in the Cys-rich domain are too far from the L1 hotspot residues for simultaneous contacts with the putative insulin binding site to occur, even with bridging by water molecules.

Alanine scanning of the A and B isoforms extracellular domains¹⁵⁹ showed that a number of mutations produced differential effects on the two isoforms. Mutation of Asn 15 in the L1 domain and Phe 714 at the C terminus of the alpha subunit inactivated the A isoform but only reduced the affinity of the B isoform 40- to 60-fold. At the C terminus of the alpha subunit, mutations of Asp 707, Val 713 and Val 715 produced 7- to 16-fold reductions in affinity of the A isoform but were without effect on the B isoform. In contrast, alanine mutations of Tyr 708 and Asn 711 inactivated the B isoform but only reduced the affinities of the A isoform 11- and 6-fold, respectively. This indicates that the energetic contributions of certain side chains differ in each isoform, suggesting that different molecular mechanisms are used to obtain the same affinity.

Alanine scanning studies of the secreted recombinant IGF-I receptor¹⁶⁰ (Fig. 4) indicates that the L1 and Cys-rich domains and the C-terminal peptide of the a subunit 692-702) corresponding to 705-715 of the insulin receptor are components of the IGF-I binding site. In the L1 the side chains of residues in the first 4 turns of the β helix (Asp 8 and Asn 11; Tyr 28, His 30 and Leu 33; Leu 56 and Arg 59; Phe 90 respectively) appear to form a discrete functional epitope on the base of the domain. Trp 79 which is situated in a bulge in the fourth turn of the L1 β helix and Arg 240, Phe 241, Glu 242 and Phe 251 in the cysteine rich domain form a second functional epitope. None of these mutations in L1 or the cysteine rich domain caused a greater than 25 fold decrease in affinity for IGF-I. Alanine mutations of all residues in the C-terminal peptide of the α subunit with the exception of Phe 695, Ser 699 and Val 702 compromised IGF-I binding. However in contrast to the corresponding insulin receptor mutants, only mutation of Phe 701 (corresponding to Phe 714 of the insulin receptor) to alanine appeared to inactivate ligand binding.

Reconstitution of Modular Minimized Receptor Constructs with Low and High Affinity

It should first be mentioned that discussion of the affinity of receptor constructs should carefully consider the conditions of the experiment since the affinity of the wild type receptor varies widely depending on the assay conditions. In physiological buffer on whole cells, the high affinity component of insulin binding is usually in the sub-nM range (e.g., 0.3 nM on IM-9 cells) while low affinity binding is typically ˜10 nM. However, when detergent solubilized holoreceptor or some other dimeric constructs are studied in solution, the high affinity can be optimized to the low pM range.¹⁶¹ In that context, nM is “low affinity”.

Studies with purified receptors have established that the holoreceptor binds only one insulin molecule with high affinity, and at least one additional molecule with low affinity.¹⁶¹ It has the same negatively cooperative binding as the cell-bound receptor. In contrast, a purified secreted dimeric ectodomain comprising the two a-subunits and the extracellular portions of the β-subunits binds two molecules of insulin, but with low affinity; it shows a linear Scatchard plot and a fast dissociation rate that is not accelerated by unlabeled insulin.¹⁶² The bivalency of the insulin receptor was also demonstrated by photoaffinity labelling.¹⁶³ Monomeric αβ receptor halves prepared by disulfide bond reduction also bind insulin but only with low affinity.^164,165 When the soluble ectodomain is reduced to αβ form, no significant change in affinity is observed.¹⁶¹ These experiments have established that each αβ receptor half contains partial binding sites for insulin but that cooperation between the two receptor halves is required to create one site with high affinity. High affinity and negative cooperativity could be restored in the soluble ectodomain by fusing it to dimeric constant domains from immunoglobulin Fc and λ subunits¹⁶⁶ or leucine zipper,¹⁶⁷ or by extending the construct to include the transmembrane domains.^168,169 These data show that it is the improper orientation of the β-subunits that disturbs the structure of the soluble ectodomain, although the beta subunits do not contain binding determinants, and that this is corrected by constraints imposed by membrane insertion or TM domain interactions.

A variety of attempts at deleting various modules from the insulin receptor extracellular domain have been made, with the aims of mapping domains essential for ligand binding as well as making minimized receptor structures more amenable to crystallization (some of which are illustrated in Fig. 6.

Figure 6

Progressive reconstitution of a minimal dimeric receptor structure with high affinity and negative cooperativity: ligand binding properties of minimized receptor constructs. See text for explanation. Reprinted with permission from De Meyts P and Whittaker (more...)

Kadowaki et al (1990) deleted the sequence 486-569 (i.e., 67% of what was later identified as the first FnIII domain including Cys 524 involved in one of the α-α disulfide bonds) and found that the construct had decreased binding affinity, three times faster dissociation rate (measured only at 5 minutes) and altered negative cooperativity as shown by near linearization of the Scatchard plot.¹⁵² This was not the main emphasis of the paper and the properties of this construct were not extensively quantitated. In contrast, Sung et al made a somewhat more extensive deletion of this FnIII domain (485-599) and reported that it did not affect insulin binding, but abolished the interaction of various anti-receptor antibodies,¹⁷⁰ in agreement with the finding that region 450-601 is a major immunogenic determinant. The results of the two studies regarding the role of FnIII₀ in insulin binding affinity are thus conflicting, but recent studies on purified receptor constructs have established that the first FnIII domain is unquestionably a major determinant of high affinity binding and negative cooperativity. Kristensen et al deleted the region 487-599 from a high affinity soluble ectodomain fused to the Fc region of IgG.¹⁷¹ This led to complete loss of high affinity binding (from pM to 3.9 nM, i.e., more than a 1,000-fold loss of affinity). However, in Table 1 of that paper, only the low affinity of the wild-type receptor is tabulated for comparison with the various low affinity deletion mutants. The statement in the discussion that deletion of 234 aminoacids in the central part of the IR α-subunit does not compromise ligand binding implicitly refers to the most extensive deletion yielding a monomeric fragment (due to loss of all the α-α disulfides)—referred to as “minimized insulin receptor” or mIR, see below—as compared to the low affinity ectodomain. The statement does not apply to the dimeric FnIII₀ deletion mutant compared to the Fc-fused high affinity wild type, which is the appropriate control. This has been apparently misunderstood by some authors who have reviewed the data by Kristensen et al as being in agreement with Sung et al.¹⁷² Adding to the confusion, another paper with some of the same authors quote Sung et al and Kadowaki et al as both showing decreased affinity and linear Scatchard plots in the FnIII₀ deletion mutants.¹⁷³ The issue has been clarified in two later studies^174,175 which have shown unequivocally that the presence of both exon 10 (which contains the cysteines 682, 683 and 685 involved in α-α disulfide bonds, as well as the CT sequence 704-717 essential for insulin binding) and the FnIII₀ domain (which contains the 524 cysteine) are required for high affinity binding. Introducing only FnIII₀ into the minireceptor does not improve the nM affinity, and the ligand dissociation rate is fast without acceleration by unlabeled insulin, despite the fact that the construct forms dimers. When both domains are introduced in the mini-receptor, the affinity of the dimeric construct increases 1,000 fold to the pM range, dissociation of tracer is slow, and the negative cooperativity is reintroduced as shown by accelerated dissociation.

Compared to the full ectodomain, this high affinity construct lacks only the 47 amino acid sequence 602-649 encoded by exon 9 (first half of the second FnIII domain just before the insert) and the alternatively spliced exon 11 plus 3 C-terminal residues from the a subunit, and of course the β-subunit extracellular domain; this again suggests that the low affinity of the soluble ectodomain is due to structure disruption by misoriented β-subunits. It also shows that the exon 9-encoded domain has no other structural function than to provide the bond with the β-subunit, and that ˜90% of the a-subunit native structure is required to maintain the integral binding properties of the native receptor, without contributions from the β-subunit.

An interesting finding of Surinyia et al investigating similar constructs,¹⁷⁵ is that while the FnIII₀ module introduced by Brandt et al¹⁷⁴ is extended by 9 amino acids from the N-terminal sequence of the second FnIII domain, the introduction of a FnIII₀ module terminating at the exact boundary of FnIII₀ (amino acid 593) also restores the high affinity even in the absence of exon 10, and also slow dissociation, but not the accelerated dissociation (meaning that the receptor is locked in the high affinity state). This 9-amino acid sequence contains three prolines, thus possibly with some degree of secondary structure; this linker may thus be critical in providing an orientation of the CT domain that is essential for negative cooperativity. If the amino acids encoded by exon 10 are present, the presence or absence of the 9-amino acid sequence does not make a difference and negative cooperativity is observed in both cases (Siddle, K., Brandt, J., personal communication).

The binding properties of the minireceptor construct of Kristensen et al¹⁷¹ were quite unexpected. As mentioned above, the L1CL2 fragment (1-468) equivalent to the IGF-IR fragment crystallized by Garret et al¹¹⁰ does not bind the ligand. Fusion of a 16-amino acid peptide (amino acids 704-719 from the C-terminus of the a-subunit to its C-terminus resulted in a protein with low (nM) affinity for insulin; these amino acids have been shown to be an insulin contact site by photoaffinity labelling and alanine scanning mutagenesis. Similar findings were obtained by Molina et al.¹⁷³ Molina et al mentioned that this construct binds with only 10-fold lower affinity than the native receptor, but their assay conditions were not optimized to detect very high affinity binding (low tracer, long incubation time). The reconstitution of nM affinity with the corresponding CT peptide was true also for the IGF-I receptor. Further a construct with only the L1C sequence (1-308) and the CT peptide binds insulin. In fact, the peptide does not need to be covalently attached to the receptor domain, coincubation of L1CL2 or L1C with a synthetic CT peptide reconstitutes the nM affinity, even if the N-terminal domain is further shortened to 1-255.¹⁷⁶ This, together with the alanine scanning and photoaffinity data, suggest that one binding site on the receptor comprises residues of the L1 domain and the CT domain, which are probably close to each other in the native structure as discussed above.

The minireceptor paradigm was also used to investigate the role of the L1CL2 and CT domains in the specificity of ligand binding in the insulin/IGF-I receptor family.¹⁷⁷ It was shown that the CT domain of the insulin-receptor related receptor (IRR), an orphan receptor related to the above that binds neither insulin nor IGF-I, abolishes binding to both insulin and IGF-I minireceptors, surprisingly since there are only 4 aminoacid differences. It was shown that mutating Phe 714 to Ala in the insulin receptor CT peptide abolishes the ability of the CT peptide to confer binding to L1CL2, confirming the alanine scanning of the full receptor. The constructs with L1CL2 of the insulin receptor with either IR or IGF-IR CT sequences show low selectivity for ligands, binding both insulin and IGF-I with affinities in the nM range. In contrast, the IGF-IR L1CL2 with either the IGF-IR or IR CT domains discriminates more than 1000 fold against insulin.

Attempts at Insulin Receptor Structure Definition by Electron Microscopy

While the crystal structure of the N-terminal fragment of the IGF-I receptor and the homology modelling studies of the FnIII repeats provide reasonably accurate predictions of the structures of the modules forming the extracellular domain of the receptors, they fail to provide any indication of their overall organization in the receptor dimer. Some clues to the answer to this question have been provided by single molecule electron microscopic imaging of either purified recombinant receptor extracellular domain^178,179 or of purified full length receptor using a variety of different techniques.^180-183 The results of such studies have been somewhat variable, but suggested T or Y shapes for the receptor (see ref. 65 for review).

Tulloch et al have reported the structure of EM images of the insulin receptor ectodomain decorated with Fab' fragments of conformation-specific monoclonal anti-receptor antibodies directed towards defined epitopes of the receptor, using negative staining techniques.¹⁷⁹ This approach has the advantages of providing both controls for the structural integrity of the molecules imaged and also, within the limits of the resolution of the technique, clues to the localization of subdomains. In contrast to the Y shaped structure deduced from the studies discussed above, the consensus image of the molecular envelope of the receptor ectodomain, in this study, was that of a U shaped prism of overall dimensions 90 x 80 x 120 Å; despite the differences in the overall image, the dimensions are similar to those found in the other studies. The prism has a central cleft 30-40 Å wide, adequate to accommodate a single molecule of insulin. Fab fragments directed towards epitopes in the first FnIII domain or in the insert region of the second FinIII domain appear to bind to the base of the U. A third Fab fragment directed towards an epitope in the distal part of the cysteine rich domain binds half way up the uprights of the U on different sides at each end. This suggests an arrangement of the receptor dimer in which the 2 L1CL2 fragments are arranged in an antiparallel configuration and form the uprights and the base of the cavity of the U and the FnIII domains form the membrane proximal base, implying that the L2 domain occupies a similar position to the L1 domain relative to the cysteine rich domain, in contrast to what is observed in the crystal structure. The dimensions of the crystallized fragment can be accommodated within such a structural model.

More recently Luo and coworkers have reported the quarternary structure of the insulin-insulin receptor complex based on 3D reconstructions from low-dose, low temperature, dark-field scanning electron microscopic studies of the complex.¹⁸³ The complexes were found to be approximately spherical with a diameter of approximately 150 Å. Covalent labeling of the receptor molecules with nanogold labelled insulin indicated that the majority of the receptors only bound a single molecule of insulin although binding of two insulin molecules was occasionally observed. This procedure also identified the location of the L1CL2 fragment within the reconstruction. Using the published coordinates of the IGF-I receptor N-terminal fragment, the insulin receptor tyrosine kinase catalytic domain, and of Type III repeats from fibronectin, they produced a model of the domain structure within the protein envelope of the receptor dimer obtained from the EM images. In this model the L1CL2 structure occupies the central core of the extracellular domain dimer with the FnIII domains surrounding them just proximal to the membrane.

A number of assumptions made in this reconstruction and the details of the proposed insulin-receptor interface inferred in this and subsequent publications^184,185 have been questioned.⁶⁵ Nevertheless, these studies make it likely that the insulin-receptor complex is a globular molecule with a diameter of 150Å. In the extracellular component the L1CL2 domains form the membrane distal core and the FnIII domains form the membrane proximal part.

Mapping of Receptor-Binding Sites on the Insulin and IGF-I Molecules

Insulin aggregates as a function of concentration, first into dimers, and in the presence of zinc, three dimers form a hexamer (the storage form in the granules of the β-cell). The insulin molecule has one obvious flat surface studded with aromatic and aliphatic residues, which aggregates primarily through nonpolar forces. This dimer-forming surface is made of mostly B-chain residues: Gly B8, Ser B9, Val B12, Tyr B16, Gly B23, Phe B24, Phe B25, Tyr B26, Thr B27, Pro B28, Asn A21. The surfaces of the dimer buried in the hexamer comprize both polar and nonpolar residues from the A and B chains: Leu A13, Tyr A14, Glu A17, Phe B1, Val B2, Gln B4, Gln B13, Ala B14, Leu B17, Val B18, Cys B19, Gly B20.^186,187

There has been a consensus for many years that a number of surface residues that have been widely conserved during the evolution of vertebrates are probably involved in receptor binding: Gly A1, Gln A5, Tyr A19, Asn 21, Val B12, Tyr B16, Gly B23, Phe B24, Phe B25, Tyr B26 (“classical binding surface”).^186-188 De Meyts et al showed that a subset of residues from this surface (A21, B23-26) was essential for the negative cooperativity in binding.¹²¹ In addition, Ile A2 and Val A3 (a residue mutated in a diabetic patient: insulin Wakayama¹⁸⁹), which are not on the surface of the molecule, probably become exposed and interact with the receptor upon displacement of the B-chain C-terminus during the receptor binding process,¹⁹⁰ although the extent of this rearrangement has been debated.¹⁹¹

The concept that the classical binding surface was the only one involved in receptor binding was challenged by studies of the properties of phylogenetically ancient insulins from the hagfish (reviewed in ref. 57 and more recently from the lamprey¹⁹² (Holst, P.A, Conlon, J.M, Whittaker, J., De Meyts, P., unpublished data)). Despite absolute conservation of the classical binding surface, these insulins display different binding behaviour from most mammalian insulins: low affinity for the holoreceptor, but better relative affinity for the soluble low affinity ectodomain, low metabolic potency, slow association kinetics, decreased negative cooperativity with a sigmoid (not bell-shaped) dose-response curve. This strongly suggested that certain residues outside the classical binding surface must contact the receptor and their mutation in hagfish and lamprey insulins explain the abnormal behaviour. The evolutionarily divergent hystricomorph insulins exhibit similar binding behaviour but the interpretation of its underlying mechanism was less clearcut since they also have a few substitutions in the classical binding surface.⁵⁷ This concept was validated when we found that insulin analogues with mutations at two residues in the hexamer-forming surface, Leu A13 to Ser and Leu B17 to Gln, bound to the holoreceptor with similar characteristics to hagfish insulin.⁵⁷

These findings prompted a reexamination of the structure-function relationship of insulin by alanine scanning mutagenesis. Kristensen et al produced 21 new insulin analogue constructs with single alanine substitutions and tested their binding to the soluble ectodomain fused to the IgG fusion protein.¹⁹³ Since it is difficult to accurately determine the high affinity (pM) binding component of such constructs in competition assays due to ligand depletion, requirement for minimal tracer amounts and long incubation time, the authors focused on the low affinity component of binding. They showed that mutation of residues A2, A19, B23 and B24 from the classical binding surface and of B13 were most disruptive of binding. Mutations of B6 and B8 were also disruptive, probably due to conformational alterations.

Since high affinity binding is more amenable to precise determination in the context of the cell-bound receptor, we tested the binding affinity of the alanine-scanned analogues using IM-9 human lymphocytes, a “gold standard” for insulin receptor studies for nearly three decades. We reconfirmed the importance of the residues from the classical binding surface A2, A19, B16, B26. Others have shown that mutations of A3¹⁹⁴ (Aladdin, H. and De Meyts, P., unpublished data) and B12¹⁹⁵ to Ala are also very disruptive. In addition, we found that a number of residues outside the classical surface disrupted binding, validating the concept of a second receptor binding surface: Ser A12, Leu A13, Glu A17, His B10, Glu B13, Leu B17¹⁹⁶ (Theede, A-M., Aladdin, H., De Meyts, P., manuscript in preparation). Interestingly, all of these are involved in the formation of the hexamer. Mutation of Thr A8 to Ala (as in ox insulin) impacted binding moderately; previous studies of analogues mutated in this position had already suggested that it may extend the classical binding surface¹⁹⁷ and its mutation to His explains the enhanced affinity of chicken and turkey insulins (Piron, M.A. and De Meyts, P., unpublished data).

Moreover, we recently showed by introducing substitutions from hystricomorph¹⁹⁸ or hagfish (Sajid, W., Andersen, A.S. and and De Meyts, P., unpublished data) insulins into human insulin, alone or in combination, that the aberrant behaviour of these species is largely explained by a small number of deleterious mutations in the novel binding surface.

To date, alanine scanning studies of insulin and its receptor have only determined the effect of mutations on equilibrium binding affinity. This may underestimate the impact of the mutations. We have performed direct binding kinetic studies of several of the analogues and found that some mutations affect association and dissociation rates in compensatory ways; thus, A13 Ala and B17 Ala had a nearly 20-fold decrease in association rates, but a slower dissociation rate as well (similar to hagfish and hystricomorph insulins). This may apply to insulin receptor mutations as well but has not been tested. This contrasts with the growth hormone-receptor system where the majority of alanine mutations primarily affected dissociation rates with proportional effects on affinity.¹⁹⁹ However in this system, mutation of several charged residues and a mutation resulting in a localized structural perturbation decreased association rates but not to the extent observed with A13 and A17 Ala mutations of insulin.

From the above, it therefore appears, perhaps unsurprisingly for a small compact molecule, that insulin uses more or less the same functional surfaces for receptor binding as for self-aggregation.

The structure-function relationships of the IGF-I and II molecules have not been as extensively mapped as insulin's (see refs. 103, 200 for detailed review). Unlike insulin, the IGFs conserve a permanent C-peptide and feature a D-peptide extension at the C-terminus (Fig. 1). 3D structures have been determined by solution NMR or crystallography in the presence of detergents.^201-203 Mutagenesis data of IGF-I have shown Ala 8, Asp 12, Phe 23 and Tyr 24 in the B domain; Tyr 31, Arg 36, Arg 37 in the C-peptide and Met 59, Tyr 60 and Ala 62 in the A domain to be important for high affinity binding to the IGF-I receptor (see refs. 103, 200 for review). Chimeric receptors show that the determinants that favor binding of Arg 36 and Arg 37 lie between residues 217 and 286 in the Cys-rich region.²⁰⁴ Zhang et al showed that Ala replacement of both Lys 65 and Lys 68 in the D domain caused a 10-fold drop in affinity for the IGF-I receptor;²⁰⁴ however, Bayne et al showed that removal of the entire D domain had an insignificant effect on IGF-I receptor binding.¹³³ Mutation of the highly conserved Val 43 in IGF-II to Leu, equivalent to the key conserved insulin-binding residue Val A3, resulted in a 16-fold reduction in IGF-I receptor binding. Moreover, a patient with a dwarfism phenotype due to a homozygous missense mutation at the equivalent Val 44 (to Met) in IGF-I has recently been found (Karperien, M. and Walenkamp, M.J.E., personal communication). This position is homologous to insulin's Val A3 mutated in the diabetic patient with insulin Wakayama,¹⁸⁹ suggesting that this aminoacid is critical for receptor binding of both ligands (the affinity of both mutants for their respective receptor is reduced to less than 1%). The phenotype of the IGF-I mutant patient is comparable to that of a patient with a partial IGF-I gene deletion.²⁰⁵

The C domain of IGF-I plays a major role in IGF-I receptor binding since its deletion or replacement by short linkers alter the binding affinity dramatically.^133,206,207 The most recent X-ray structure determination of IGF-I has provided more details of the structure of the C-peptide and shows that when compared with the structure of insulin it produces a rearrangement of the B22-29 segment corresponding to the C-terminal insulin sequence important for receptor binding.²⁰¹ Interestingly, a chimeric molecule with the A and B chains of insulin fused to the IGF-I C-peptide (ICP) binds with high affinity to both insulin and IGF-I receptors, while proinsulin's C-peptide causes a 20-fold drop in affinity for the insulin receptor.²⁰⁸

Thus, while it is likely that the low affinity of insulin for the IGF-I receptor largely results from the lack of the C-peptide, more studies are needed to fully understand the low affinity of IGF-I for the insulin receptor (see Note Added in Proof).

Mechanism of Ligand Binding and Receptor Activation

We have now laid all the pieces of the puzzle on the table, how do we assemble them? In other words, how do the bits and pieces of structural information discussed above help us match the putative binding sites on insulin with those on the receptor? An important clue to the possible topography of the receptor complex was the discovery that human growth hormone (hGH) forms a 1:2 complex with the extracellular domain of its receptor, i.e., that a single GH molecule is a bivalent ligand with two opposite surfaces binding to basically the same site on the receptor.²⁰⁹ Such a mechanism could also explain how the insulin receptor generates one high affinity binding site from low affinity parts.

Twenty five years ago, Martin Raff was the first to speculate that the complex insulin binding kinetics may be the result of insulin having more than one binding site for the receptor and its consequent ability to crosslink the receptor (see discussion p. 116 in ref. 210). Later, Lee et al²¹¹ and Yip²¹² also proposed that insulin may contact both halves of the receptor. Schäffer was the first to put forward a model that attempted to integrate all the available kinetic information, proposing that it may be two different receptor sites (labelled site 1 and site 2) that contact an insulin molecule, in contrast to the GH-GH receptor interaction.¹⁶¹ Schäffer found that insulin analogues with mutations in the hexamer-forming surface e.g., at A13 and B17 had a biological activity lower than was expected from affinity, slow binding kinetics and a higher relative affinity for the soluble receptor ectodomain than for the holoreceptor, and concluded that these residues are probably involved in receptor binding. De Meyts had independently found that the same analogues had very slow association and dissociation rates in IM-9 cells, and reduced negative cooperativity with a sigmoid rather than bell-shaped dose-response curve (properties reminiscent of hagfish and hystricomorph insulins), and came to the same conclusion.⁵⁷ Schäffer proposed that the insulin molecule would have two binding surfaces labelled site 1 (the classical binding surface overlapping the dimer-forming surface) and site 2 (the new binding site overlapping the hexamer-forming surface), and that a single insulin molecule would crosslink the insulin receptor's site 1 and site 2. This would explain the 1:2 stoichiometry and the coexistence of high and low affinity sites (curved Scatchard plot). While the affinity of the crosslinked insulin molecule is in the pM range, Schäffer determined the affinity of site 1, using the high affinity analogue X92, to be 1 nM.

However, a model with a single crosslink does not predict the rapid acceleration of tracer dissociation induced by cold ligand²¹³ (Shymko and De Meyts, unpublished modelling). This is also demonstrated experimentally by the absence of negative cooperativity in the hGH receptor system. To explain accelerated dissociation, Schäffer proposed a cis-induced conformational change disturbing the crosslink upon binding of a second molecule to the unoccupied site 1. Also, the single crosslink per se does not explain the bell-shaped curve for negative cooperativity. Schäffer postulates an additional competing mechanism such as the binding of a molecule to the unoccupied binding site 2, “putting a lid” on the complex, sterically impairing dissociation.

While not questioning the plausibility of these postulates, De Meyts proposed that all three parameters of the negative cooperativity (high and low affinity sites, accelerated dissociation and bell-shaped dose-response curve) could be explained by the crosslinking mechanism per se without need for additional mechanisms, if the receptor binding sites 1 and 2 were disposed in an antiparallel symmetry⁵⁷ (Fig. 7). A second insulin molecule bridging both leftover sites 1 and 2 upon partial dissociation of the first crosslink would prevent its rebinding and accelerate dissociation of the first bound insulin (Fig.7). A bell-shaped curve is the hallmark of any effect that requires crosslinking (as demonstrated by the dose-response curve for biological effects of GH.²¹⁴ The concept of antiparallel symmetry was supported by the EM studies of Tulloch et al.¹⁷⁹

Figure 7

Symmetrical model of bivalent crosslinking insulin binding mechanism. Both α-subunit N-terminal pairs of binding sites (1 and 2) are represented in a symmetrical antiparallel arrangement as also suggested by Tulloch et al. The first insulin molecule (more...)

Whichever modality of the crosslinking model proves to be correct, the general concept proposed by Schäffer and De Meyts is widely supported by the body of evidence discussed above and it also provides a molecular basis for activation of the receptor tyrosine kinase and signalling pathways. The motion involved in simultaneous interaction of the bivalent ligand with the two half-receptors may approximate the two kinase domains and permit transphosphorylation. This is analogous to the mechanism demonstrated for the dimeric bacterial aspartate receptor, which also exhibits negative cooperativity.²¹⁵ In addition, in the cytokine family, previously thought to be activated by ligand-induced dimerization, structural and functional evidence has been presented that demonstrates that the erythropoietin receptor exists as a preformed dimer in the absence of ligands, in which the stems of the extracellular domain and thus presumably the two janus kinase domains are too far apart (73 Å) for transactivation.^216,217 The binding of an EPO-like peptide ligand changes the structure and brings the stems closer together (39 Å).

It remains to identify sites 1 and 2 on the insulin receptor. We should clarify here that the nomenclature we use for site 1 and 2 is the opposite of that used by Schäffer,¹⁶¹ who labelled site 1 (classical binding surface) and site 2 (new surface) according to their order of discovery. If one postulates an ordered, sequential mechanism as is the case for hGH, then it is mutations in the site that bind last that create antagonism.²¹⁸ Since mutations at the C-terminus of the B-chain create antagonists for negative cooperativity,⁵⁷ we have postulated that this site binds last (and is therefore site 2), and that site 1 is the site containing Leu A13 and Leu B17, in agreement with the fact that mutations of these two residues slow down the initial association 20-fold, suggesting that they normally make the initial contact.

Photoaffinity labelling data suggest that insulin's site 2 binds to a receptor site composed of the L1 domain and the CT domain,^125,128 whose functional epitopes have been identified by alanine scanning.^153,156 This is consistent with the L1C-CT construct binding insulin with nM affinity.¹⁷¹

The insulin receptor binding site 1 has not been mapped by alanine scanning yet, but several lines of evidence suggest that is is located within the structures encoded by exons 6-8, i.e., the C-terminal part of L2 and the first FnIII domain (FnIII₀):

1. The B1-photoaffinity analogue covalently labels a peptide mapping to 390-488 generated by tryptic digestion of the receptor.¹²⁶
2. All autoimmune anti-receptor antibodies from patients with extreme insulin resistance recognize an epitope contained within residues 450-601.²¹⁹ Such antibodies lock the patient's insulin receptors into a low affinity state with fast dissociation rate and lack of negative cooperativity; plasmapheresis restores high affinity, slow dissociation and negative cooperativity.²²⁰ Monoclonal anti-receptor antibodies that inhibit insulin binding bind to the same epitope.^103,219 Chimeric insulin receptors with residues 450-601 replaced by the corresponding residues from the IGF-I receptor had decreased ability to bind insulin.²¹⁹ Monoclonal antibodies that activate the receptor bind to an epitope within residues 469-592.¹⁰³
3. A Mutation of Lys 460 to Glu found in a leprechaun patient enhances negative cooperativity and alters pH dependence of binding; mutation of Lys 460 to Arg suppresses negative cooperativity and linearizes the Scatchard plot¹⁵² (Wallach and De Meyts, unpublished). Interestingly, the equivalent residue in the IGF-I receptor is Arg and its mutation does not affect binding kinetics (Wallach and De Meyts, unpublished).
4. Substitution of the insulin receptor domain encoded by exons 6 and 7 into the IGF-I receptor induces a bell-shaped instead of sigmoid dose-response curve for negative cooperativity.

In summary, the studies described above support the concept that insulin binds asymmetrically to two discrete sites in the receptor dimer, crosslinking the constituent monomers. One of these sites, the binding site of the secreted recombinant receptor, has been characterized in detail and is partly located in the L1 domain and also contains a peptide from the C-terminus of the α-subunit. The second site, in contrast, is considerably less well studied but is located in the L2-FnIII₀ region. These findings are consistent with the topology of the extracellular portion of the receptor that has been proposed on the basis of crystallographic and EM studies.

Conclusions and a Word of Caution

In the absence of a definitive three-dimensional structure of a high affinity ligand-receptor complex, circumstantial evidence generated by a multitude of biochemical and molecular biological approaches have generated a plausible working model of the insulin and IGF-I receptor interaction that provides a rational explanation for a variety of complex equilibrium and kinetic properties, including a solution to the three decades-old riddle of the aberrant behaviour of hagfish and hystricomorph insulins. This model has also provided a rational basis for the design of agonist and antagonist mimetic peptides,^221,222 (reviewed in ref. 65).

However, the concept described above of a single bivalent ligand sitting in the middle of a head-to-head dimeric receptor complex (which appears to be the rule in the RTK or cytokine ligand-receptor structures that have been solved so far), is called into question by the recently determined structure of the EGF receptor (which has significant structural homology to the insulin and IGF-I receptors) complexed to EGF or TGFα (reviewed in ref. 112). In those structures, two ligand molecules bind to the external surfaces of a back-to-back receptor dimer with a 2:2 stoichiometry (Fig. 8) rather than the 1:2 we have advocated for the insulin and IGF-I receptor. This means that we have to remain open-minded to the fact that insulin and IGF-I may link the two a-subunit binding epitopes in a cis-fashion within the same subunit rather than a trans-fashion between two subunits as we have proposed. However, the majority of the free energy for the formation of the EGF receptor back-to-back dimer is contributed primarily by contacts between two protruding β-hairpin arms in the CR domains that are entirely absent from the insulin and IGF-I receptors. Therefore it is still plausible that the insulin/IGF-I receptor dimer structure will rather resemble the head-to-head structures also present in the EGF receptor crystal structures.

Figure 8

Crystal structure of the complex of human epidermal growth factor and receptor extracellular domains. The two dimerized L1-CR-L2 domains are shown as tube models respectively in red and green. The two bound EGF molecules are shown as CPK models in yellow. (more...)

It is clear therefore that the hypotheses and models generated over the last three decades of insulin and IGF-I ligand-receptor binding studies still need to be confirmed or invalidated by three-dimensional structural data of the high affinity ligand-receptor complexes.

Note Added in Proof

Since this chapter was submitted, two new reviews have discussed the insulin receptor binding mechanism.^227,228

Sørensen et al²²⁹ have examined the molecular mechanisms underlying the lower affinity of IGF-II as compared to IGF-I for the IGF-I receptor by evaluating the effect of alanine mutations of residues forming the ligand binding site¹⁶⁰ on the affinity of the secreted recombinant receptor for IGF-II. In contrast to the functional epitope for IGF-I, which is composed of residues in the L1, cysteine rich and the Fn1 insert domain, there was no evidence for involvement of cysteine rich domain residues in binding IGF-II. Furthermore while identical residues in the L1 and insert domains were used to bind both ligands, there were small but significant differences in the free energy contributions of many of these residues to the binding of the two ligands.

Whittaker et al have now characterized the functional binding epitope of the native insulin receptor²³⁰ (previous studies have focused on a secreted ectodomain). Alanine mutations Arg14, Phe 64 in the L1 domain and Phe 705, Glu 706, Tyr 708, His 710, and Asn 711 in the insert domain inactivate the insulin binding function of both forms of the receptor. Surprisingly, alanine mutation of Val 715, which has no impact on the affinity of the secreted receptor, inactivates the full-length receptor. Significant differences in effect on affinity of the holo-receptor for insulin were also observed for alanine mutations of Asp 12, Leu 37, Phe 89, Glu 97, and Glu 120 in the L1 domain and of Leu 709 and Phe 714 in the insert domain. It is possible that these differences are a consequence of the proposed sequential binding mechanism.

Chakravarty et al²³¹ have characterized the binding properties of a hybrid receptor dimer composed of a wild type IGF-I receptor monomer and an inactivating mutant monomer. The mutation of Phe 701 to alanine abolished the ability of both the secreted and full length receptors to bind IGF-I while the hybrid receptor bound IGF-I with an affinity indistinguishable from that of the wild type receptor. However, in dissociation experiments, no ligand-induced acceleration of dissociation was observed. These results are consistent with the proposed bivalent cross-linking model of ligand binding.⁵⁷

Additional photo-affinity cross-linking experiments^232-235 have shown that Val A3 and Thr A8 in insulin's A-chain N-terminal α-helix crosslink to the CT domain in the insulin receptor's insert (like insulin's Phe B25) while Tyr B16 in the central a-helix of insulin's B-chain and Phe B24 in its B chain C-terminus crosslink to L1 domain of the receptor. These results confirm the role of the receptor's L1CT domain (our site 2) in binding the “classical” insulin binding surface.

Two studies have addressed the mechanisms for the higher affinity of IGF-II than IGF-I for the A isoform of the insulin receptor. Denley et al²³⁶ have used chimeras where the C and D domains were exchanged between IGF-I and IGF-II either singly or together, and concluded that the C domain and to a lesser extent the D domain represent the principal determinants of the binding difference between IGF-I and IGF-II to IR-A. This is somewhat surprising since a single chain fusion protein containing insulin fused to C-peptide of IGF-I has high affinity for both the A-isoform of the insulin receptor's and the IGF-I receptor.²⁰⁸ In contrast, Gauguin and De Meyts (manuscript in preparation²³⁷) found that the substitutions at the IGF-I and IGF-II positions equivalent to insulin's A8, A10, B5 and B16 explain the low affinities of the IGFs for the A-isoform of the insulin receptor, while two substitutions in IGF-II at positions equivalent to insulin's A18 and B14 have a positive effect on affinity that explains why IGF-II binds better than IGF-I.

Finally, the articles describing the Val 44 IGF-I mutation in a dwarf patient have been published.^238,239

Acknowledgements

The Receptor Systems Biology Laboratory at the Hagedorn Research Institute is an independent basic research component of Novo Nordisk A/S. The laboratory is also supported by grants from the Danish Medical Research Council through the Center for Growth and Regeneration, Medicon Valley Academy and Øresund IT Academy, and grants from the National Institutes of Health (5R01 DK065890) and the Juvenile Diabetes Research Foundation (1-2000-198) to J.W. Extensive discussions on the ligand binding mechanism with Lauge Schäffer and Jacob Brandt, and on insulin structure with Michael A. Weiss, are gratefully acknowledged, as well as the collaboration with Claus Kristensen on alanine scanning insulin mutants and Asser S. Andersen in making other recombinant mutated insulins.

References

1.

Chan SJ, Steiner DF. Insulin through the ages: Phylogeny of a growth promoting and metabolic regulatory hormone. Am Zool. 2000;40:213–222.

2.

Conlon JM. Evolution of the insulin molecule: Insights into structure-activity and phylogenetic relationships. Peptides. 2000;22:1183–1193. [PubMed: 11445250]

3.

Steiner DF, Chan SJ, Welsh JM. et al. Structure and evolution of the insulin gene. Annu Rev Genet. 1985;19:463–484. [PubMed: 3909947]

4.

Hsu SY. New insights into the evolution of the relaxin-LGR signaling system. Trends Endocrinol Metab. 2003;14:303–309. [PubMed: 12946872]

5.

Hsu SY, Nakabayashi K, Nishi S. et al. Relaxin signaling in reproductive tissues. Mol Cell Endocrinol. 2003;202:165–170. [PubMed: 12770746]

6.

Kumagai J, Hsu SY, Matsumi H. et al. INSL3/Leydig insulin-like peptide activates the LGR8 receptor important in testis descent. J Biol Chem. 2002;277:31283–31286. [PubMed: 12114498]

7.

Ferlin A, Simonato M, Bartoloni L. et al. The INSL3-LGR8/GREAT ligand-receptor pair in human cryptorchidism. J Clin Endo Metab. 2003;88:4273–4279. [PubMed: 12970298]

8.

Irwin DM. A second insulin gene in fish genomes. Gen Comp Endo. 2003;135:150–158. [PubMed: 14644655]

9.

Kondo H, Ino M, Suzuki A. et al. Multiple gene copies for Bombyxin, an insulin-related peptide of the sikmoth Bombyx mori: Structural signs for gene rearrangement and duplication responsible for generation of multiple molecular forms of bombyxin. J Mol Biol. 1996;259:926–937. [PubMed: 8683595]

10.

Iwami M, Furuya I, Kataoka H. Bombyxin-related peptides: cDNA structure and expression in the brain of the hornworm Agrius convolvuli. Insect Biochem Mol Biol. 1996;26:25–32. [PubMed: 8673077]

11.

Lagueux M, Lwoff L, Meister M. et al. cDNAs from neurosecretory cells of brains of Locusta migratoria encoding a novel member of the superfamily of insulins. Eur J Biochem. 1990;187:249–254. [PubMed: 1688797]

12.

Smit AB, Geraerts PM, Meester I. et al. Characterization of a cDNA clone encoding molluscan insulin-related peptide II of Lymnaea stagnalis. Eur J Biochem. 1991;199:699–703. [PubMed: 1868853]

13.

Floyd PD, Li L, Rubakhin SS. et al. Insulin prohormone processing, distribution, and relation to metabolism in Aplysia californica. J Neurosci. 1999;19:7732–7741. [PMC free article: PMC6782465] [PubMed: 10479677]

14.

Brogiolo W, Stocker H, Ikeya T. et al. An evolutionarily conserved function of the Drosophila insulin receptor and insulin-like peptides in growth control. Curr Biol. 2001;11:213–221. [PubMed: 11250149]

15.

Pierce SB, Costa M, Wisotzkey R. et al. Regulation of DAF-2 receptor signaling by human insulin and ins-1, a member of the unusually large and diverse C. elegans insulin gene family. Genes Dev. 2001;15:672–686. [PMC free article: PMC312654] [PubMed: 11274053]

16.

Hua QX, Nakagawa SH, Wilken J. et al. A divergent INS protein in Caenorhabditis elegans structurally resembles human insulin and activates the human insulin receptor. Genes Dev. 2003;17:826–831. [PMC free article: PMC196032] [PubMed: 12654724]

17.

Robitzki A, Schroder HC, Ugarkovic D. Demonstration of an endocrine signaling circuit for insulin in the sponge Geodia cydonium. EMBO J. 1989;8:2905–2909. [PMC free article: PMC401354] [PubMed: 2531072]

18.

Duret L, Guex N, Peitsch MC. et al. New insulin-like proteins with atypical disulfide bond pattern characterized in Caenorhabditis elegans by comparative sequence analysis and homology modeling. Genome Res. 1998;8:348–353. [PubMed: 9548970]

19.

Fernandez R, Tabarini D, Azpiazu N. et al. The Drosophila insulin receptor homolog: A gene essential for embryonic development encodes two receptor isoforms with different signaling potential. EMBO J. 1995;14:3373–3384. [PMC free article: PMC394404] [PubMed: 7628438]

20.

Fullbright G, Lacy ER, Büllesbach EE. The prothoracicotropic hormone bombyxin has specific receptors on insect ovarian cells. Eur J Biochem. 1997;245:774–780. [PubMed: 9183018]

21.

Graf R, Neuenschwander S, Brown MR. et al. Insulin-mediated secretion of ecdysteroids from mosquito ovaries and molecular cloning of the insulin receptor homologue from ovaries of bloodfed Aedes aegypti. Insect Mol Biol. 1997;6:151–163. [PubMed: 9099579]

22.

Holt RA, Subramanian GM, Halpern A. et al. The genome sequence of the malaria mosquito Anopheles gambiae. Science. 2002;298:129–149. [PubMed: 12364791]

23.

Roovers E, Vincent ME, van Kesteren E. et al. Characterization of a putative molluscan insulin-related peptide receptor. Gene. 1995;162:181–188. [PubMed: 7557427]

24.

Lardans V, Coppin JF, Vicogne J. et al. Characterization of an insulin receptor-related receptor in Biomphalaria glabrata embryonic cells. Bioch Biophys Acta. 2001;1510:321–329. [PubMed: 11342169]

25.

Jonas EA, Knox RJ, Kaczmarek LK. et al. Insulin receptor in Aplysia neurons: Characterization, molecular cloning, and modulation of ion currents. J Neurosci. 1996;16:1645–1658. [PMC free article: PMC6578688] [PubMed: 8774433]

26.

Gricourt L, Bonnec G, Boujard D. et al. Insulin-like system and growth regulation in the Pacific oyster Crassostrea gigas: hrIGF-1 effect on protein synthesis of mantle edge cells and expression of an homologous insulin receptor-related receptor. Gen Comp Endocrinol. 2003;134:44–56. [PubMed: 13129502]

27.

Kimura KD, Tissenbaum HA, Liu Y. et al. Daf-2, an insulin receptor-like gene that regulates longevity and diapause in Caenorhabditis elegans. Science. 1997;277:942–946. [PubMed: 9252323]

28.

Verjovski-Almeida S, DeMarco R. et al. Transcriptome analysis of the acoelomate human parasite Schistosoma mansoni. Nat Genetics. 2003;35:148–157. [PubMed: 12973350]

29.

Hu W, Yan Q, Shen DK. et al. Evolutionary and biomedical implications of a Schistosoma japonicum complementary DNA resource. Nat Genetics. 2003;35:139–147. [PubMed: 12973349]

30.

Konrad C, Kroner A, Spiliotis M. et al. Identification and molecular characterisation of a gene encoding a member of the insulin receptor family in Echinococcus multilocularis. Int J Parasitol. 2003;33:301–312. [PubMed: 12670515]

31.

Steele RE, Lieu P, Mai NH. Response to insulin and the expression pattern of a gene encoding an insulin receptor homolog suggest a role for an insulin-like molecule in regulating growth and patterning in Hydra. Dev Genes Evol. 1996;206:247–259. [PubMed: 24173564]

32.

Schäder H, Schröder HC, Gamulin V. et al. Molecular cloning of a tyrosine kinase gene from the marine sponge Geodia cydonium: A new member belonging to the receptor tyrosine kinase class II family. Mol Membr Biol. 1994;11:101–107. [PubMed: 7920862]

33.

Skorokhod A, Gamulin V, Gundacker D. et al. Origin of insulin receptor-like tyrosine kinases in marine sponges. Biol Bull. 1999;197:198–206. [PubMed: 10573839]

34.

Chan SJ, Cao QP, Steiner DF. Evolution of the insulin superfamily: Cloning of a hybrid insulin/ insulin-like growth factor cDNA from amphioxus. Proc Natl Acad Sci USA. 1990;87:9319–9323. [PMC free article: PMC55156] [PubMed: 1701257]

35.

Pashmforoush M, Chan SJ, Steiner DF. Structure and expression of the insulin-like peptide receptor from amphioxus. Mol Endo. 1996;10:857–866. [PubMed: 8813726]

36.

Dehal P, Satou Y, Campbell RK. et al. The draft genome of Ciona intestinalis: Insights into chordate and vertebrate origins. Science. 2002;298:2157–2167. [PubMed: 12481130]

37.

McRory JE, Sherwood NM. Ancient divergence of insulin and insulin-like growth factor. DNA Cell Biol. 1997;16:939–949. [PubMed: 9303436]

38.

Stein LD, Bao Z, Blasiar D. et al. The genome sequence of Caenorhabditis briggsae: A platform for comarative genomics. PLoS Biology. 2003;1:166–192. [PMC free article: PMC261899] [PubMed: 14624247]

39.

Dlakic M. A new family of putative insulin receptor-like proteins in C. elegans. Curr Biol. 2002;12:R155–R157. [PubMed: 11882301]

40.

Garofalo RS. Genetic analysis of insulin signaling in Drosophila. Trends Endocrinol Metab. 2002;13:156–162. [PubMed: 11943559]

41.

Efstratiadis A. Genetics of mouse growth. Int J Dev Biol. 1998;42:955–976. [PubMed: 9853827]

42.

Tatar M, Kopelman A, Epstein D. et al. A mutant Drosophila insulin receptor homolog that extends life-span and impairs neuroendocrine function. Science. 2001;292:107–110. [PubMed: 11292875]

43.

Holzenberger M, Dupont J, Ducos B. et al. IGF-1 receptor regulates lifespan and resistance to oxidative stress in mice. Nature. 2003;421:182–187. [PubMed: 12483226]

44.

Barbieri M, Bonafe M, Franceschi C. et al. Insulin/IGF-I-signaling pathway: An evolutionarily conserved mechanism of longevity from yeast to humans. Am J Physiol Endocrinol Metab. 2003;285:E1064–E1071. [PubMed: 14534077]

45.

Aparicio S, Chapman J, Stupka E. et al. Whole-genome shotgun assembly and analysis of the genome of Fugu rubripes. Science. 2002;297:1301–1310. [PubMed: 12142439]

46.

Scavo L, Shuldiner AR, Serrano J. et al. Genes encoding receptors for insulin and insulin-like growth factor I are expressed in Xenopus oocytes and embryos. Proc Natl Acad Sci USA. 1991;88:6214–6218. [PMC free article: PMC52053] [PubMed: 1648732]

47.

Hainaut P, Kowalski A, Giorgetti S. et al. Insulin and insulin-like-growth-factor-I (IGF-I) receptors in Xenopus laevis oocytes. Comparison with insulin receptors from liver and muscle. Biochem J. 1991;273:673–678. [PMC free article: PMC1149816] [PubMed: 1847619]

48.

Foty RA, Liversage RA. Detection of insulin receptors in newt liver and forelimb regenerates and the effects of local insulin deprivation on epimorphic regeneration. J Exp Zool. 1993;266:299–311. [PubMed: 8354993]

49.

Shemer J, Penhos JC, LeRoith D. Insulin receptors in lizard brain and liver: Structural and functional studies of alpha and beta subunits demonstrate evolutionary conservation. Diabetologia. 1986;29:321–329. [PubMed: 3522330]

50.

Shemer J, Raizada M, LeRoith D. Structural and functional studies on insulin receptors from alligator brain and liver. Comp Biochem Physiol. 1987;86:55–61. [PubMed: 3829632]

51.

Hart C, Shemer J, Penhos JC. et al. Frog brain and liver show evolutionary conservation of tissue-specific differences among insulin receptors. Gen Comp Endocrinol. 1987;68:170–178. [PubMed: 3322930]

52.

Kozma SC, Thomas G. Regulation of cell size in growth, development and human disease: PI3K, PKB and S6K. Bio Essays. 2002;24:65–71. [PubMed: 11782951]

53.

Leevers SJ. Growth control: Invertebrate insulin surprises! Curr Biol. 2001;11:R209–R212. [PubMed: 11301264]

54.

Puig O, Marr MT, Ruhf ML. et al. Control of cell number by Drosophila FOXO: Downstream and feedback regulation of the insulin receptor pathway. Genes Dev. 2003;17:2006–2020. [PMC free article: PMC196255] [PubMed: 12893776]

55.

Kitamura T, Kahn CR, Accili D. Insulin receptor knockout mice. Annu Rev Physiol. 2003;65:313–332. [PubMed: 12471165]

56.

Kenyon C, Chang J, Gensch E. et al. C. elegans mutant that lives twice as long as wild type. Nature. 1993;366:461–464. [PubMed: 8247153]

57.

De MeytsP. The structural basis of insulin and insulin-like growth factor-I (IGF-I) receptor binding and negative cooperativity, and its relevance to mitogenic versus metabolic signaling. Diabetologia. 1994;37(suppl 2):S135–S148. [PubMed: 7821729]

58.

Robertson SC, Tynan JA, Donoghue DJ. RTK mutations and human syndromes. When good receptors turn bad. Trends in Genetics. 2000;16:265–271. [PubMed: 10904266]

59.

Blume-Jensen P, Hunter T. Oncogenic kinase signalling. Nature. 2001;411:355–365. [PubMed: 11357143]

60.

Gray A, Stenfeldt MathiasenI, De MeytsP. The insulin-like growth factors and insulin-signalling systems: An appealing target for breast cancer therapy? Hormone Metab Res. 2003;35:857–871. [PubMed: 14710369]

61.

Drevs J, Medinger M, Schmidt-Gersbach C. et al. Receptor tyrosine kinases: The main targets for new anticancer therapy. Curr Drug Targets. 2003;4:113–121. [PubMed: 12558064]

62.

Zwick E, Bange J, Ullrich A. Receptor tyrosine kinases as targets for anticancer drugs. Trends Mol Med. 2002;8:17–23. [PubMed: 11796262]

63.

Levitzki A. Tyrosine kinases as targets for cancer therapy. Eur J Cancer. 2002;38(Suppl 5):S11–S18. [PubMed: 12528768]

64.

Brunelleschi S, Penengo L, Santoro MM. et al. Receptor tyrosine kinases as target for anti-cancer therapy. Curr Pharm Des. 2002;8:1959–1972. [PubMed: 12171522]

65.

De MeytsP, Whittaker J. Structural biology of insulin and IGF1 receptors: Implications for drug design. Nat Rev Drug Discov. 2002;1:769–783. [PubMed: 12360255]

66.

House PD, Weidemann MJ. Characterization of an [125 I]-insulin binding plasma membrane fraction from rat liver. Biochem Biophys Res Comm. 1970;41:541–548. [PubMed: 4320281]

67.

Freychet P, Roth J, Neville DM. Insulin receptors in the liver: Specific binding of I-125 insulin to the plasma membrane and its relation to insulin bioactivity. Proc Natl Acad Sci USA. 1971;68:1833–1837. [PMC free article: PMC389303] [PubMed: 4331561]

68.

Cuatrecasas P. Insulin-receptor interactions in adipose tissue cells: Direct measurement and properties. Proc Natl Acad Sci USA. 1971;68:1264–1268. [PMC free article: PMC389168] [PubMed: 5288373]

69.

Gammeltoft S, Gliemann J. Binding and degradation of 125I-labelled insulin by isolated rat fat cells. Biochim Biophys Acta. 1973;320:16–32. [PubMed: 4748361]

70.

Yip CC, Yeung CW, Moule ML. Photoaffinity labeling of insulin receptor proteins of liver plasma membrane preparations. Biochemistry. 1980;19:70–76. [PubMed: 6986166]

71.

Massague J, Czech MP. The subunit structures of two distinct receptors for insulin-like growth factors I and II and their relationship to the insulin receptor. J Biol Chem. 1982;257:5038–5045. [PubMed: 6279656]

72.

Shia MA, Pilch PF. The beta subunit of the insulin receptor is an insulin-activated protein kinase. Biochemistry. 1983;22:717–721. [PubMed: 6340724]

73.

Roth RA, Cassell DJ. Insulin receptor: Evidence that it is a protein kinase. Science. 1983;219:299–301. [PubMed: 6849137]

74.

Kasuga M, Fujita-Yamaguchi Y, Blithe DL. et al. Tyrosine-specific protein kinase activity is associated with the purified insulin receptor. Proc Natl Acad Sci USA. 1983;80:2137–2141. [PMC free article: PMC393772] [PubMed: 6188161]

75.

Ebina Y, Ellis L, Jarnagin K. et al. The human insulin receptor cDNA: The structural basis for hormone-activated transmembrane signalling. Cell. 1985;40:747–758. [PubMed: 2859121]

76.

Ullrich A, Bell JR, Chen EY. et al. Human insulin receptor and its relationship to the tyrosine kinase family of oncogenes. Nature. 1985;313:756–761. [PubMed: 2983222]

77.

Whittaker J, Okamoto AK, Thys R. et al. High-level expression of human insulin receptor cDNA in mouse NIH 3T3 cells. Proc Natl Acad Sci USA. 1987;84:5237–5241. [PMC free article: PMC298830] [PubMed: 3299382]

78.

Ullrich A, Gray A, Tam AW. et al. Insulin-like growth factor-I receptor primary structure: Comparison with insulin receptor suggests structural determinants that define functional specificity. EMBO J. 1986;5:2503–2512. [PMC free article: PMC1167146] [PubMed: 2877871]

79.

Seino S, Seino M, Nishi S. et al. Structure of the human insulin receptor gene and characterization of its promoter. Proc Natl Acad Sci USA. 1989;86:114–118. [PMC free article: PMC286414] [PubMed: 2911561]

80.

Abbott AM, Bueno R, Pedrini MT. et al. Insulin-like growth factor I receptor gene structure. J Biol Chem. 1992;267:10759–10763. [PubMed: 1316909]

81.

Shier P, Watt V. Primary structure of a putative receptor for a ligand of the insulin family. J Biol Chem. 1989;264:14605–14608. [PubMed: 2768234]

82.

Zhang B, Roth RA. The insulin receptor-related receptor: Tissue expression, ligand binding specificity, and signaling capabilities. J Biol Chem. 1992;267:18320–18328. [PubMed: 1326521]

83.

Nef S, Verma-Kurvari S, Merenmles J. et al. Testis determination requires insulin receptor family function in mice. Nature. 2003;426:291–295. [PubMed: 14628051]

84.

Weber A, Huesken C, Bergmann E. et al. Coexpression of insulin receptor-related receptor and insulin-like growth factor 1 receptor correlates with enhanced apoptosis and dedifferentiation in human neuroblastomas. Clin Cancer Res. 2003;9:5683–5692. [PubMed: 14654552]

85.

Heldin CH, Ostman A. Ligand-induced dimerization of growth factor receptors: Variations on the theme. Cytokine Growth Factor Rev. 1996;7:3–10. [PubMed: 8864349]

86.

Lu K, Guidotti G. Identification of the cysteine residues involved in the class I disulfide bonds of the human insulin receptor: Properties of insulin receptor monomers. Mol Biol Cell. 1996;7:679–691. [PMC free article: PMC275922] [PubMed: 8744943]

87.

Jianping WuJ, Guidotti G. Construction and characterization of a monomeric insulin receptor. J Biol Chem. 2002;31:27809–27817. [PubMed: 12023954]

88.

Seino S, Seino M, Bell GI. Human insulin receptor gene. Diabetes. 1990;39:129–133. [PubMed: 2227119]

89.

McClain DA. Different ligand affinities of the two human insulin receptor splice variants are reflected in parallel changes in sensitivity for insulin action. Mol Endo. 1991;5:734–739. [PubMed: 2072930]

90.

Yamaguchi Y, Flier JS, Yokota A. et al. Functional properties of two naturally occuring isoforms of the human insulin receptor in chinese hamster ovary cells. Endocrinology. 1991;129:2058–2066. [PubMed: 1655392]

91.

Gu JL, De MeytsP. The structure and function of the insulin receptor. In: Zai-Ping L, Zi-Xian L, You-Shang-Z, eds. Retrospect and Prospect of Protein Research. Singapore: World Scientific Publishing Co. 1991:120–125.

92.

Frasca F, Pandini G, Scalia P. et al. Insulin receptor isoform A, a newly recognized, high-affinity insulin-like growth factor II receptor in fetal and cancer cells. Mol Cell Biol. 1999;19:3278–3288. [PMC free article: PMC84122] [PubMed: 10207053]

93.

De MeytsP, Wallach B, Christoffersen CT. et al. The insulin-like growth factor-I receptor. Structure, ligand binding mechanism and signal transduction. Horm Res. 1994;42:152–169. [PubMed: 7868068]

94.

Bass J, Chiu G, Argon Y. et al. Folding of insulin receptor monomers is facilitated by the molecular chaperones calnexin and calreticulin and impaired by rapid dimerization. J Cell Biol. 1998;141:637–646. [PMC free article: PMC2132748] [PubMed: 9566965]

95.

Pashmforoush M, Yoshimasa Y, Steiner DF. Exon 11 enhances insulin binding affinity and tyrosine kinase activity of the human insulin proreceptor. J Biol Chem. 1994;269:32639–32648. [PubMed: 7798270]

96.

Soos M, Field CE, Siddle K. Purified hybrid insulin/insulin-like growth factor-i receptors bind insulin-like growth factor-I, but not insulin, with high affinity. Biochem J. 1993;290:419–426. [PMC free article: PMC1132290] [PubMed: 8452530]

97.

Pandini G, Frasca F, Mineo R. et al. Insulin/IGF-I hybrid receptors have different biological characteristics depending on the insulin receptor isoform involved. J Biol Chem. 2002;277:39684–39695. [PubMed: 12138094]

98.

Bajaj M, Waterfield MD, Schlessinger J. et al. On the tertiary structure of the extracellular domains of the epidermal growth factor and insulin receptors. Biochim Biophys Acta. 1987;916:220–226. [PubMed: 3676333]

99.

Ward CW, Hoyne PA, Flegg RH. Insulin and epidermal growth factor receptors contain the cysteine repeat motif found in the tumor necrosis factor receptor. Proteins. 1995;22:141–153. [PubMed: 7567962]

100.

Marino-Buslje C, Mizuguchi K, Siddle K. et al. A third fibronectin type III domain in the extracellular region of the insulin receptor family. FEBS Lett. 1998;441:331–336. [PubMed: 9883910]

101.

Mulhern TD, Booker GW, Cosgrove L. A third fibronectin-type-III domain in the insulin-family receptors. TIBS. 1998;23:465–466. [PubMed: 9868364]

102.

Ward CW. Members of the insulin receptor family contain three fibronectin type III domains. Growth factors. 1999;16:315–322. [PubMed: 10427505]

103.

Adams TE, Epa VC, Garrett TP. et al. Structure and function of the type 1 insulin-like growth factor receptor. Cell Mol Life Sci. 2000;57:1050–1093. [PubMed: 10961344]

104.

Schaffer L, Ljungqvist L. Identification of a disulfide bridge connecting the alpha-subunits of the extracellular domain of the insulin receptor. Biochem Biophys Res Commun. 1992;189:650–653. [PubMed: 1472036]

105.

Sparrow LG, McKern NM, Gorman JJ. et al. The disulfide bonds in the C-terminal domains of the human insulin receptor ectodomain. J Biol Chem. 1997;272:29460–29467. [PubMed: 9368005]

106.

Cheatham B, Kahn CR. Cysteine 647 in the insulin receptor is required for normal covalent interaction between alpha- and beta-subunits and signal transduction. J Biol Chem. 1992;267:7108–7115. [PubMed: 1551916]

107.

Hedo JA, Kasuga M, Van ObberghenE. Direct demonstration of glycosylation of insulin receptor subunits by biosynthetic and external labeling: Evidence for heterogeneity. Proc Natl Acad Sci USA. 1981;78:4791–4795. [PMC free article: PMC320249] [PubMed: 6946427]

108.

Elleman TC, Frenkel MJ, Hoyne PA. et al. Mutational analysis of the N-linked glycosylation sites of the human insulin receptor. Biochem J. 2000;347:771–779. [PMC free article: PMC1221015] [PubMed: 10769182]

109.

Leconte I, Auzan C, Debant A. et al. N-linked oligosaccharide chains of the insulin receptor beta subunit are essential for transmembrane signaling. J Biol Chem. 1992;267:17415–17423. [PubMed: 1324936]

110.

Garrett TP, McKern NM, Lou M. et al. Crystal structure of the first three domains of the type-1 insulin-like growth factor receptor. Nature. 1998;394:395–399. [PubMed: 9690478]

111.

Ward CW, Garrett TP. The relationship between the L1 and L2 domains of the insulin and epidermal growth factor receptors and leucine-rich repeat modules. BMC Bioinformatics. 2001;2:4. [PMC free article: PMC37351] [PubMed: 11504559]

112.

Jorissen RN, Walker F, Pouliot N. et al. Epidermal growth factor receptor: Mechanisms of activation and signalling. Exp Cell Res. 2003;284:31–53. [PubMed: 12648464]

113.

Hubbard SR, Till JH. Protein tyrosine kinase structure and function. Annu Rev Biochem. 2000;69:373–398. [PubMed: 10966463]

114.

Hubbard SR, Wei L, Ellis L. et al. Crystal structure of the tyrosine kinase domain of the human insulin receptor. Nature. 1994;372:746–754. [PubMed: 7997262]

115.

Hubbard SR. Crystal structure of the activated insulin receptor tyrosine kinase in complex with peptide substrate and ATP analog. EMBO J. 1997;16:5572–5581. [PMC free article: PMC1170189] [PubMed: 9312016]

116.

Favelyukis S, Till JH, Hubbard SR. et al. Structure and autoregulation of the insulin-like growth factor 1 receptor kinase. Nat Struct Biol. 2001;8:1058–1063. [PubMed: 11694888]

117.

Till JH, Ablooglu AJ, Frankel M. et al. Crystallographic and solution studies of an activation loop mutant of the insulin receptor tyrosine kinase: Insights into kinase mechanism. J Biol Chem. 2001;276:10049–10055. [PubMed: 11124964]

118.

Baer K, Al HasaniH, Parvaresch S. et al. Dimerization-induced activation of soluble insulin/IGF-1 receptor kinases: An alternative mechanism of activation. Biochemistry. 2001;40:14268–14278. [PubMed: 11714281]

119.

De MeytsP, Roth J, Neville DM. et al. Insulin interactions with its receptors: Experimental evidence for negative cooperativity. Biochem Biophys Res Comm. 1973;55:154–161. [PubMed: 4361269]

120.

Kahn CR, Freychet P, Roth J. et al. Quantitative aspects of the insulin-receptor interaction in liver plasma membranes. J Biol Chem. 1974;249:2249–2257. [PubMed: 4362069]

121.

De MeytsP, Van ObberghenE, Roth J. et al. Mapping of the residues of the receptor binding region of insulin responsible for the negative cooperativity. Nature. 1978;273:504–509. [PubMed: 661960]

122.

Christoffersen CT, Bornfeldt KE, Rotella CM. et al. Negative cooperativity in the insulin-like growth factor-I (IGF-I) receptor and a chimeric IGF-I/insulin receptor. Endocrinology. 1994;135:472–475. [PubMed: 8013387]

123.

Waugh SM, DiBella EE, Pilch PF. Isolation of a proteolitically derived domain of the insulin receptor containing the major site of crosslinking/binding. Biochemistry. 1989;28:3448–3455. [PubMed: 2742847]

124.

Yip CC, Hsu H, Patel RG. et al. Localization of the insulin-binding site to the cysteine-rich region of the insulin receptor alpha-subunit. Biochem Biophys Res Comm. 1988;157:321–329. [PubMed: 3058126]

125.

Wedekind F, Baer PontzenK, Bala Mohan S. et al. Hormone binding site of the insulin receptor: An analysis using photoaffinity-mediated avidin complexing. Biol Chem Hoppe Seyler. 1989;370:251–258. [PubMed: 2540769]

126.

Fabry M, Schaefer E, Ellis L. et al. Detection of a new hormone contact site within the insulin receptor ectodomain by the use of a novel photoreactive insulin. J Biol Chem. 1992;267:8950–8956. [PubMed: 1577732]

127.

Shoelson SE, Lee J, Lynch CS. et al. BpaB25 insulins. Photoactivatable analogues that quantitatively cross-link, radiolabel, and activate the insulin receptor. J Biol Chem. 1993;268:4085–4091. [PubMed: 8382690]

128.

Kurose T, Pashmforoush M, Yshimasa Y. et al. Cross-linking of a B25 azidophenylalanine insulin derivative to the carboxyl-terminal region of the a-subunit of the insulin receptor. J Biol Chem. 1994;269:29190–29197. [PubMed: 7961885]

129.

Kjeldsen T, Andersen AS, Wiberg FC. et al. The ligand specificities of the insulin receptor and the insulin-like growth factor-I receptor reside in different regions of a common binding site. Proc Natl Acad Sci USA. 1991;88:4404–4408. [PMC free article: PMC51668] [PubMed: 1852007]

130.

Andersen AS, Kjeldsen T, Wiberg FC. et al. Identification of determinants that confer ligand specificity on the insulin receptor. J Biol Chem. 1992;267:13681–13686. [PubMed: 1320025]

131.

Schumacher R, Mosthaf L, Schlessinger J. et al. Insulin and insulin-like growth factor-I binding specificity is determined by distinct regions of their cognate receptors. J Biol Chem. 1991;266:19288–19295. [PubMed: 1655782]

132.

Tavare JM, Siddle K. Mutational analysis of insulin receptor function: Consensus and controversy. Biochim Biophys Acta. 1993;1178:21–39. [PubMed: 8329456]

133.

Bayne ML, Applebaum J, Underwood D. et al. The C region of human insulin-like growth factor (IGF) I is required for high affinity binding to the type I IGF receptor. J Biol Chem. 1988;264:11004–11008. [PubMed: 2472386]

134.

Kjeldsen T, Wiberg FC, Andersen AS. Chimeric receptors indicate that phenylalanine 39 is a major contributor in insulin specificity of the insulin receptor. J Biol Chem. 1994;269:32942–32946. [PubMed: 7806523]

135.

Taylor SI, Cama A, Accili D. et al. Mutations in the insulin receptor gene. Endocrine Rev. 1992;13:566–595. [PubMed: 1330507]

136.

Vorwerk P, Christoffersen CT, Muller J. et al. Alternative splicing of exon 17 and a missense mutation in exon 20 of the insulin receptor gene in two brothers with a novel syndrome of insulin resistance (congenital fiber-type disproportion myopathy). Horm Res. 1999;52:211–220. [PubMed: 10844410]

137.

Vestergaard H, Klein HH, Hansen T. et al. Severe insulin-resistant diabetes mellitus in patients with congenital muscle fiber type disproportion myopathy. J Clin Invest. 1995;95:1925–1932. [PMC free article: PMC295737] [PubMed: 7706500]

138.

Klein HH, Muller R, Vestergaard H. et al. Implications of compound heterozygous insulin receptor mutations in congenital muscle fibre type disproportion myopathy for the receptor kinase activation. Diabetologia. 1999;42:245–249. [PubMed: 10064106]

139.

Kadowaki T, Kadowaki H, Accili D. et al. Substitution of lysine for asparagine at position 15 in the alpha-subunit of the human insulin receptor. A mutation that impairs transport of receptors to the cell surface and decreases the affinity of insulin binding. J Biol Chem. 1990;265:19143–19150. [PubMed: 2121734]

140.

Rouard M, Bass J, Grigorescu F. et al. Congenital insulin resistance associated with a conformational alteration in a conserved beta-sheet in the insulin receptor L1 domain. J Biol Chem. 1999;274:18487–18491. [PubMed: 10373457]

141.

Longo N, Langley SD, Griffin LD. et al. Activation of glucose transport by a natural mutation in the human insulin receptor. Proc Natl Acad Sci USA. 1993;90:60–64. [PMC free article: PMC45599] [PubMed: 8419945]

142.

Grønskov K, Vissing H, Shymko RM. et al. Mutation of arginine 86 to proline in the insulin receptor alpha subunit causes lack of transport of the receptor to the plasma membrane, loss of binding affinity and a constitutively activated tyrosine kinase in transfected cells. Biochem Biophys Res Comm. 1993;192:905–911. [PubMed: 8484793]

143.

Nakae J, Morioka H, Ohtsuka E. et al. Replacements of leucine 87 in human insulin receptor alter affinity for insulin. J Biol Chem. 1995;270:22017–22022. [PubMed: 7665623]

144.

Liu R, Zhu J, Jospe N. et al. Deletion of lysine 121 creates a temperaturesensitive alteration in insulin binding by the insulin receptor. J Biol Chem. 1995;270:476–482. [PubMed: 7814414]

145.

Rique S, Nogues C, Ibanez L. et al. Identification of three novel mutations in the insulin recepor gene in type A insulin resistant patients. Clinical Genetics. 2000;57:67–69. [PubMed: 10733238]

146.

Hamer I, Foti M, Emkey R. et al. An arginine to cysteine 252 mutation in insulin receptors from a patient with severe insulin resistance inhibits receptor internalisation but preserves signalling events. Diabetologia. 2002;45:657–667. [PubMed: 12107746]

147.

Maassen JA, Tobias ES, Kayserilli H. et al. Identification and functional assessment of novel and known insulin receptor mutations in five patients with syndromes of severe insulin resistance. J Clinical Endo Metab. 2003;88:4251–4257. [PubMed: 12970295]

148.

Nakashima N, Umeda F, Yanase T. et al. Insulin resistance associated with substitution of histidine for arginine 252 in the alpha-subunit of the human insulin receptor: Trial of insulin-like growth factor I injection therapy to enhance insulin sensitivity. J Clin Endo Metab. 1995;80:3662–3667. [PubMed: 8530617]

149.

Roach P, Zick Y, Formisano P. et al. A novel human insulin receptor gene mutation uniquely inhibits insulin binding without impairing posttranslational processing. Diabetes. 1994;43:1096–1102. [PubMed: 8070609]

150.

Krook A, Soos MA, Kumar S. et al. Functional activation of mutant human insulin receptor by monoclonal antibody. Lancet. 1996;347:1586–1590. [PubMed: 8667867]

151.

Whittaker J, Mynarcik DC. Phenotype of the Ser-323 to Leu mutation of the insulin receptor is isoform dependent. Diabetes. 1998;47(Suppl 1):264 (abstract).

152.

Kadowaki H, Kadowaki T, Cama A. et al. Mutagenesis of lysine 460 in the human insulin receptor. Effects upon receptor recycling and cooperative interactions among binding sites. J Biol Chem. 1990;265:21285–21296. [PubMed: 2123490]

153.

Williams PF, Mynarcik DC, Qin YG. et al. Mapping of an NH2-terminal ligand binding site of the insulin receptor by alanine scanning mutagenesis. J Biol Chem. 1995;270:1–5. [PubMed: 7852381]

154.

Clackson T, Wells JA. A hot spot of binding energy in a hormone-receptor interface. Science. 1995;267:383–386. [PubMed: 7529940]

155.

De MeytsP, Gu JL, Shymko RM. et al. Identification of a ligand-binding region of the human insulin receptor encoded by the second exon of the gene. Mol Endo. 1990;4:409–416. [PubMed: 2188117]

156.

Mynarcik DC, Yu GQ, Whittaker J. Alanine-scanning mutagenesis of a C-terminal ligand binding domain of the insulin receptor a subunit. J Biol Chem. 1996;271:2439–2442. [PubMed: 8576204]

157.

Mynarcik DC, Williams PF, Schäffer L. et al. Analog binding properties of insulin receptor mutants. Identification of amino acids interacting with the COOH terminus of the B chain of the insulin molecule. J Biol Chem. 1997;272:2077–2081. [PubMed: 8999905]

158.

Hart LM, Lindhout D, Van der Zon GC. et al. An insulin receptor mutant (Asp707 --> Ala), involved in leprechaunism, is processed and transported to the cell surface but unable to bind insulin. J Biol Chem. 1996;271:18719–18724. [PubMed: 8702527]

159.

Whittaker J, Sørensen H, Gadsboll VL. et al. Comparison of the functional insulin binding epitopes of the A and B isoforms of the insulin receptor. J Biol Chem. 2002;277:47380–47384. [PubMed: 12270939]

160.

Whittaker J, Groth AV, Mynarcik DC. et al. Alanine scanning mutagenesis of a type 1 insulin-like growth factor receptor ligand binding site. J Biol Chem. 2001;276:43980–43986. [PubMed: 11500492]

161.

Schäffer L. A model for insulin binding to the insulin receptor. Eur J Biochem. 1994;221:1127–1132. [PubMed: 8181471]

162.

Markussen JM, Halstrom J, Wiberg F. et al. Immobilized insulin for high capacity affinity chromatography of insulin receptors. J Biol Chem. 1991;266:18814–18818. [PubMed: 1918001]

163.

Yip CC, Jack E. Insulin receptors are bivalent as demonstrated by photoaffinity labeling. J Biol Chem. 1992;267:13131–13134. [PubMed: 1618814]

164.

Sweet LJ, Morrison BD, Pessin J. Isolation of functional a-β heterodimers from the purified human placental a2-β2 heterotetrameric insulin receptor concept. Structural basis for high affinity ligand binding. J Biol Chem. 1987;262:6939–6942. [PubMed: 3294822]

165.

Boni-Schnetzler M, Scott W, Waugh SM. et al. The insulin receptor. Structural basis for high affinity ligand binding. J Biol Chem. 1987;262:8395–8401. [PubMed: 3597378]

166.

Bass J, Kurose T, Pashmforoush M. et al. Fusion of insulin receptor ectodomains to immunoglobulin constant domains reproduces high-affinity insulin binding in vitro. J Biol Chem. 1996;271:19367–19375. [PubMed: 8702623]

167.

Hoyne PA, Cosgrove LJ, McKern NM. et al. High affinity insulin binding by soluble insulin receptor extracellular domain fused to a leucine zipper. FEBS Lett. 2000;479:15–18. [PubMed: 10940380]

168.

Whittaker J, Garcia P, Yu GQ. et al. Transmembrane domain interactions are necessary for negative cooperativity of the insulin receptor. Mol Endo. 1994;8:1521–1527. [PubMed: 7877620]

169.

Florke RR, Schnaith K, Passlack W. et al. Hormone-triggered conformational changes within the insulin-receptor ectodomain: Requirement for transmembrane anchors. Biochem J. 2001;360:189–198. [PMC free article: PMC1222217] [PubMed: 11696007]

170.

Sung CK, Wong KY, Yip CC. et al. Deletion of residues 485-599 from the human insulin receptor abolishes antireceptor antibody binding and influences tyrosine kinase activation. Mol Endo. 1994;8:315–324. [PubMed: 8015549]

171.

Kristensen C, Wiberg FC, Schäffer L. et al. Expression and characterization of a 70-kDa fragment of the insulin receptor that binds insulin. Minimizing ligand binding domain of the insulin receptor. J Biol Chem. 1998;273:17780–17786. [PubMed: 9651379]

172.

Marino-Buslje C, Martin-Martinez M, Mizuguchi K. et al. The insulin receptor: From protein sequence to structure. Biochem Soc Trans. 1999;27:715–726. [PubMed: 10917671]

173.

Molina L, Marino-Buslje C, Quinn DR. et al. Structural domains of the insulin receptor and IGF receptor required for dimerisation and ligand binding. FEBS Lett. 2000;467:226–230. [PubMed: 10675543]

174.

Brandt J, Andersen AS, Kristensen C. Dimeric fragment of the insulin receptor alpha-subunit binds insulin with full holoreceptor affinity. J Biol Chem. 2001;276:12378–12384. [PubMed: 11278498]

175.

Surinya KH, Molina L, Soos MA. et al. Role of insulin receptor dimerization domains in ligand binding, cooperativity, and modulation by anti-receptor antibodies. J Biol Chem. 2002;277:16718–16725. [PubMed: 11875066]

176.

Kristensen C, Andersen AS, Ostergaard S. et al. Functional reconstitution of insulin receptor binding site from nonbinding receptor fragments. J Biol Chem. 2002;277:18340–18345. [PubMed: 11901156]

177.

Kristensen C, Wiberg FC, Andersen AS. Specificity of insulin and insulin-like growth factor I receptors investigated using chimeric mini-receptors. Role of C-terminal of receptor alpha subunit. J Biol Chem. 1999;274:37351–37356. [PubMed: 10601304]

178.

Schaefer EM, Erickson HP, Federwisch M. et al. Structural organization of the human insulin receptor ectodomain. J Biol Chem. 1992;267:23393–23402. [PubMed: 1385419]

179.

Tulloch PA, Lawrence LJ, McKern NM. et al. Single-molecule imaging of human insulin receptor ectodomain and its Fab complexes. J Struct Biol. 1999;125:11–18. [PubMed: 10196112]

180.

Woldin CN, Hing FS, Lee J. et al. Structural studies of the detergent-solubilized and vesicle-reconstituted insulin receptor. J Biol Chem. 1999;274:34981–34992. [PubMed: 10574975]

181.

Christiansen K, Tranum-Jensen J, Carlsen J. et al. A model for the quaternary structure of human placental insulin receptor deduced from electron microscopy. Proc Natl Acad Sci USA. 1991;88:249–252. [PMC free article: PMC50787] [PubMed: 1986371]

182.

Tranum-Jensen J, Christiansen K, Carlsen J. et al. Membrane topology of insulin receptors reconstituted into lipid vesicles. J Membr Biol. 1994;140:215–223. [PubMed: 7932656]

183.

Luo RZ, Beniac DR, Fernandes A. et al. Quaternary structure of the insulin-insulin receptor complex. Science. 1999;285:1077–1080. [PubMed: 10446056]

184.

Ottensmeyer FP, Beniac DR, Luo RZ. et al. Mechanism of transmembrane signaling: Insulin binding and the insulin receptor. Biochemistry. 2000;39:12103–12112. [PubMed: 11015187]

185.

Yip CC, Ottensmeyer P. Three-dimensional structural interactions of insulin and its receptor. J Biol Chem. 2003;278:27329–27332. [PubMed: 12764141]

186.

Blundell TL, Dodson GG, Hodgkin DC. et al. Insulin: The structure in the crystal and its reflection in chemistry and biology. Adv Prot Chem. 1972;26:279–402.

187.

Baker E, Blundell TL, Cutfield JF. et al. The structure of 2Zn pig insulin at 1.5 Å resolution. Phil Trans R Soc Lond. 1988;B19:369–456. [PubMed: 2905485]

188.

Pullen RA, Lindsay DG, Wood SP. et al. Receptor-binding region of insulin. Nature. 1976;259:369–373. [PubMed: 175286]

189.

Nanjo K, Sanke T, Miyano M. et al. Diabetes due to secretion of a structurally abnormal insulin (insulin Wakayama). Clinical and functional characteristics of [LeuA3] insulin. J Clin Invest. 1986;77:514–519. [PMC free article: PMC423373] [PubMed: 3511099]

190.

Hua XH, Shoelson SE, Kochoyan M. et al. Receptor binding redefined by a structural switch in a mutant human insulin. Nature. 1991;354:238–241. [PubMed: 1961250]

191.

Ludvigsen S, Olsen HB, Kaarsholm NC. A structural switch in a mutant insulin exposes key residues for receptor binding. J Mol Biol. 1998;279:1–7. [PubMed: 9636695]

192.

Conlon JM, Bondareva V, Rusakov Y. et al. Characterization of insulin, glucagon, and somatostatin from the river lamprey, Lampetra fluviatilis. Gen Comp Endocrinol. 1995;100:96–105. [PubMed: 8575665]

193.

Kristensen C, Kjeldsen T, Wiberg FC. et al. Alanine scanning mutagenesis of insulin. J Biol Chem. 1997;272:12978–12983. [PubMed: 9148904]

194.

Nakagawa SH, Tager HS. Importance of aliphatic side-chain structure at positions 2 and 3 of the insulin A chain in insulin-receptor interactions. Biochemistry. 1992;31:3204–3214. [PubMed: 1554705]

195.

Nakagawa SH, Tager HS, Steiner DF. Mutational analysis of invariant valine B12 in insulin: Implications for receptor binding. Biochemistry. 2000;39:15826–15835. [PubMed: 11123908]

196.

Jensen A-M. Analysis of structure-activity relationships of the insulin molecule by alanine-scanning mutagenesis. Master's thesis Copenhagen University. 2000

197.

Weiss MA, Hua Q-X, Jia W. et al. Activities of insulin analogues at position A8 are uncorrelated with thermodynamic stability. In: Federwisch M, Dieken ML, De Meyts P, eds. Insulin and Related Proteins - Structure to Function and Pharmacology. Dordrecht, Boston, London: Kluwer Academic Publishers. 2002:103–119.

198.

Palsgaard J. Receptor-binding properties of 25 different hystricomorph-like human insulin analogs. Copenhagen, Denmark: Master's thesis Copenhagen University. 2003

199.

Wells JA. Binding in the growth hormone receptor complex. Proc Natl Acad Sci USA. 1996;93:1–6. [PMC free article: PMC40168] [PubMed: 8552582]

200.

Van den Brande JL. Structure of the human insulin-like growth factors: Relationship to function. In: Schofield PN, ed. The Insulin-Like Growth Factors. Structure and Biological Functions. Oxford, New York, Tokyo: Oxford University Press. 1992:12–44.

201.

Brzozowski AM, Dodson EJ, Dodson GG. et al. Structural origins of the functional divergence of human insulin-like growth factor-I and insulin. Biochemistry. 2002;41:9389–9397. [PubMed: 12135360]

202.

Vajdos FF, Ultsch M, Schaffer ML. et al. Crystal structure of human insulin-like growth factor-1: Detergent binding inhibits binding protein interactions. Biochemistry. 2001;40:11022–11029. [PubMed: 11551198]

203.

Schaffer ML, Deshayes K, Nakamura G. et al. Complex with a phage display-derived peptide provides insight into the function of insulin-like growth factor I. Biochemistry. 2003;42:9324–9334. [PubMed: 12899619]

204.

Zhang W, Gustafson TA, Rutter WJ. et al. Positively charged side chains in the insulin-like growth factor-I C- and D- regions determine receptor binding specificity. J Biol Chem. 1994;269:10609–10613. [PubMed: 8144650]

205.

Woods KA, Camacho-Hubner C, Bergman RN. et al. Effects of insulin-like growth factor I (IGF-I) therapy on body composition and insulin resistance in IGF-I gene deletion. J Clin Endo Metab. 2000;85:1407–1411. [PubMed: 10770174]

206.

Gill R, Wallach B, Verma C. et al. Engineering the C-region of human insulin-like growth factor- 1: Implications for receptor binding. Protein Eng. 1996;9:1011–1019. [PubMed: 8961354]

207.

Geddes S, Holst P, Grotzinger J. et al. Structure-function studies of an IGF-I analogue that can be chemically cleaved to a two-chain mini-IGF-I. Protein Eng. 2001;14:61–65. [PubMed: 11287679]

208.

Kristensen C, Andersen AS, Hach M. et al. A single-chain insulin-like growth factor I/insulin hybrid binds with high affinity to the insulin receptor. Biochem J. 1995;305:981–986. [PMC free article: PMC1136354] [PubMed: 7848300]

209.

de VosAM, Ultsch M, Kossiakoff AA. Human growth hormone and extracellular domain of its receptor: Crystal structure of the complex. Science. 1992;255:306–312. [PubMed: 1549776]

210.

DeLisi C, Blumenthal R. Physical chemical aspects of cell surface events in cellular regulation. New York, Amsterdam, Oxford: Elsevier North Holland. 1979;116

211.

Lee J, O'Hare T, Pilch P. et al. Insulin receptor autophosphorylation occurs asymmetrically. J Biol Chem. 1993;268:4092–4098. [PubMed: 8440700]

212.

Yip CC. The insulin-binding domain of the insulin receptor is encoded by exons 2 and 3. J Cell Biochem. 1992;48:19–25. [PubMed: 1316357]

213.

Hammond BJ, Tikerpae J, Smith GD. An evaluation of the cross-linking model for the interaction of insulin with its receptor. Am J Physiol. 1997;272:E1136–E1144. [PubMed: 9227462]

214.

Ilondo MM, Damholt A, Cunningham B. et al. Receptor dimerization determines the effects of growth hormone in primary rat adipocytes and cultured IM-9 lymphocytes. Endocrinology. 1994;134:2397–2403. [PubMed: 8194466]

215.

Yeh JI, Biemann HP, Prive GG. et al. High-resolution structures of the ligand binding domain of the wild-type bacterial aspartate receptor. J Mol Biol. 1996;262:186–201. [PubMed: 8831788]

216.

Remy I, Wilson IA, Michnick SW. Erythropoietin receptor activation by a ligand-induced conformation change. Science. 1999;283:990–993. [PubMed: 9974393]

217.

Livnah O, Stura EA, Middleton SA. et al. Crystallographic evidence for preformed dimers of erythropoietin receptor before ligand activation. Science. 1999;283:987–990. [PubMed: 9974392]

218.

Fuh G, Cunningham BC, Fukunaga R. et al. Rational design of potent antagonists to the human growth hormone receptor. Science. 1992;256:1677–1680. [PubMed: 1535167]

219.

Zhang B, Roth RA. A region of the insulin receptor important for ligand binding (residues 450-601) is recognized by patients autoimmune antibodies and inhibitory monoclonal antibodies. Proc Natl Acad Sci USA. 1991;88:9858–9862. [PMC free article: PMC52820] [PubMed: 1719540]

220.

Muggeo M, Flier JS, Abrams RA. et al. Treatment by plasma exchange of a patient with autoantibodies to the insulin receptor. N Engl J Med. 1979;300:477–480. [PubMed: 759936]

221.

Pillutla RC, Hsiao KC, Beasley JR. et al. Peptides identify the critical hotspots involved in the biological activation of the insulin receptor. J Biol Chem. 2002;277:22590–22594. [PubMed: 11964401]

222.

Schäffer L, Brissette RE, Spetzler JC. et al. Assembly of high affinity insulin receptor agonists and antagonists from peptide building blocks. Proc Natl Acad Sci. 2003;100:4435–4439. [PMC free article: PMC153573] [PubMed: 12684539]

223.

Avruch J. Receptor tyrosine kinases. In: De Groot L, Jameson LJ, eds. Endocrinology. 4th ed. Philadelphia: WB Saunders Co. 2001:25–47.

224.

Loren CE, Englund C, Grabbe C. et al. A crucial role for the Anaplastic lymphoma kinase receptor tyrosine kinase in gut development in Drosophila melanogaster. EMBO Rep. 2003;4:781–786. [PMC free article: PMC1326337] [PubMed: 12855999]

225.

Robinson DR, Wu YM, Lin SF. The protein tyrosine kinase family of the human genome. Oncogene. 2000;19:5548–5557. [PubMed: 11114734]

226.

Ogiso H, Ishitani R, Nureki O. et al. Crystal structure of the complex of human epidermal growth factor and receptor extracellular domains. Cell. 2002;110:775–787. [PubMed: 12297050]

227.

De MeytsP. My favorite molecule. Insulin and its receptor: Structure, function and evolution. BioEssays. 2004;26:1351–1362. [PubMed: 15551269]

228.

Siddle K. The insulin receptor and downstream signalling. In: Kumar S, O'Rahilly S, eds. Insulin Resistance: Insulin Action and Its Disturbances in Disease. John Wiley and Sons, Ltd. 2005:1–62.

229.

Sørensen H, Whittaker L, Hinrichsen J. et al. Mapping of the insulin-like growth factor II binding site of the Type 1 insulin-like growth factor receptor by alanine scanning mutagenesis. FEBS Letters. 2004;565:19–22. [PubMed: 15135045]

230.

Whittaker J, Whittaker L. Characterization of the functional insulin binding epitopes of the full length insulin receptor. J Biol Chem. 2005;280:20932–20936. [PubMed: 15799978]

231.

Chakravarty A, Hinrichsen J, Whittaker L. et al. Rescue of ligand binding of a mutant IGF-I receptor by complementation. Biochem Biophys Res Commun. 2005;331:74–77. [PubMed: 15845360]

232.

Wan Z, Xu B, Chu YC. et al. Enhancing the activity of insulin at the receptor interface: Crystal structures and photo-cross-linking of A8 analogues. Biochemistry. 2004;43:16119–16133. [PubMed: 15610006]

233.

Wan ZL, Huang K, Xu B. et al. Diabetes associated mutations in human insulin: Crystal structure and photo-crosslinking studies of A-chain variant insulin Wakayama. Biochemistry. 2005;44:5000–5016. [PubMed: 15794638]

234.

Huang K, Xu B, Chu YC. et al. How insulin binds: The B-chain a-helix contacts the L1 β-helix of the insulin receptor. J Mol Biol. 2004;341:529–550. [PubMed: 15276842]

235.

Xu B, Hu SQ, Chu YC. et al. Diabetes-associated mutations in insulin: Consecutive residues in the B-chain contact distinct domains of the insulin receptor. Biochemistry. 2004;43:8356–8372. [PubMed: 15222748]

236.

Denley A, Bonython ER, Booker GW. et al. Structural determinants for high affinity binding of insulin-like growth factor II to insulin receptor (IR)-A, the exon 11 minus isoform of the IR. Mol Endo. 2004;18:2502–2512. [PubMed: 15205474]

237.

Gauguin L. Structural basis for the lower affinity of the insulin-like growth factors for the insulin receptor. Master's thesis, University of Copenhagen. 2005 [PubMed: 18048361]

238.

Walenkamp MJ, Karperien M, Pereira AM. et al. Homozygous and heterozygous expression of a novel insulin-like growth factor mutation. J Clin Endo Metab. 2005;90:2855–2864. [PubMed: 15769976]

239.

Denley A, Wang CC, McNeil KA. et al. Structural and functional characteristics of the Val44Met insulin-like growth factor-1 missense mutation: Correlation with effects on growth and development. Mol Endo. 2005;19:711–721. [PubMed: 15576456]