A Novel Binding Site for the Human Insulin-like Growth Factor-II (IGF-II)/Mannose 6-Phosphate Receptor on IGF-II*

The mammalian insulin-like growth factor (IGF)-II/cation-independent mannose 6-phosphate receptor (IGF2R) binds IGF-II with high affinity. By targeting IGF-II to lysosomal degradation, it plays a role in the maintenance of correct IGF-II levels in the circulation and in target tissues. Loss of IGF2R function is associated with tumor progression; therefore, the IGF2R is often referred to as a tumor suppressor. The interaction between IGF2R and IGF-II involves domains 11 and 13 of the 15 extracellular domains of the receptor. Recently, a hydrophobic binding region was identified on domain 11 of the IGF2R. In contrast, relatively little is known about the residues of IGF-II that are involved in IGF2R binding and the determinants of IGF2R specificity for IGF-II over the structurally related IGF-I. Using a series of novel IGF-II analogues and surface plasmon resonance assays, this study revealed a novel binding surface on IGF-II critical for IGF2R binding. The hydrophobic residues Phe19 and Leu53 are critical for IGF2R binding, as are residues Thr16 and Asp52. Furthermore, Thr16 was identified as playing a major role in determining why IGF-II, but not IGF-I, binds with high affinity to the IGF2R.

The mammalian insulin-like growth factor (IGF)-II/cationindependent mannose 6-phosphate receptor (IGF2R) binds IGF-II with high affinity. By targeting IGF-II to lysosomal degradation, it plays a role in the maintenance of correct IGF-II levels in the circulation and in target tissues. Loss of IGF2R function is associated with tumor progression; therefore, the IGF2R is often referred to as a tumor suppressor. The interaction between IGF2R and IGF-II involves domains 11 and 13 of the 15 extracellular domains of the receptor. Recently, a hydrophobic binding region was identified on domain 11 of the IGF2R. In contrast, relatively little is known about the residues of IGF-II that are involved in IGF2R binding and the determinants of IGF2R specificity for IGF-II over the structurally related IGF-I. Using a series of novel IGF-II analogues and surface plasmon resonance assays, this study revealed a novel binding surface on IGF-II critical for IGF2R binding. The hydrophobic residues Phe 19 and Leu 53 are critical for IGF2R binding, as are residues Thr 16 and Asp 52 . Furthermore, Thr 16 was identified as playing a major role in determining why IGF-II, but not IGF-I, binds with high affinity to the IGF2R.
The mammalian insulin-like growth factor-II (IGF-II) 5 /cation-independent mannose 6-phosphate receptor (IGF2R) has multiple functions due to its ability to interact with a wide variety of ligands. This transmembrane receptor has 15 homologous extracellular repeat domains and a short intracellular domain, which lacks kinase activity (1). The IGF2R plays a major role in targeting mannose 6-phosphate-labeled proteins via the trans-Golgi network and from the cellular membrane to lysosomes for degradation (2). It also binds and internalizes IGF-II, thereby maintaining correct levels of IGF-II locally and in the circulation. These functions are central in the processes of embryonic development and normal growth. Mice lacking the IGF2R have increased levels of circulating IGF-II, are born larger, and die soon after birth due to cardiac hyperplasia. This phenotype is overcome by concomitant deletion of the IGF-II gene (3). Likewise, loss of IGF2R activity in many cancers has been associated with increased tumor cell growth and tumor progression. This can arise from the loss of heterozygosity or mutation of the receptor (4,5), leading to an increase in bioavailable IGF-II, which then acts via the type 1 IGF receptor (IGF-1R) to promote cancer growth and survival.
The extracellular domains of the IGF2R share significant similarity with the single extracellular domain of the cation-dependent mannose 6-phosphate receptor (2). Domains 3 and 9 and more recently domain 5 of the IGF2R were shown to contain mannose 6-phosphate binding sites (6 -8) and act as binding sites for ligands, including the urokinase plasminogen activator receptor, plasminogen, transforming growth factor-␤, and retinoic acid. A distinct IGF-II binding site is located on domain 11, and the additional presence of domain 13 is required for high affinity IGF-II binding equivalent to intact IGF2R (9 -11).
The crystal structure of the IGF2R domain 11 has been solved (12), and specific mapping of the IGF-II binding site on this domain has recently been reported (13). Residues corresponding to the mannose 6-phosphate binding site of the cationdependent mannose 6-phosphate receptor were mutated to Ala, and Tyr 1542 , Thr 1570 , Phe 1567 , and Ile 1572 within this hydrophobic patch were shown to be critical for IGF-II binding. Ile 1572 had previously been identified as essential for IGF-II binding (10). Interestingly, mutation of Glu 1544 (which lies at the edge of this binding site) to Ala or Lys leads to an increase in binding affinity (13). Although the binding site on domain 11 has been mapped, it is not yet clear whether IGF-II contacts domain 13.
IGF-I and IGF-II share Ͼ60% sequence identity, yet the IGF2R only binds IGF-II with high affinity. Little is known about this interaction, since relatively few IGF-II mutants have been made, and a structure of the IGF2R⅐IGF-II complex has not been solved. Residues Phe 48 , Arg 49 , and Ser 50 have been strongly implicated in IGF2R binding (14). However, these residues are conserved between IGF-I and IGF-II and therefore would not drive the specificity of binding. Ala 54 and Leu 55 of IGF-II differ from the corresponding residues in IGF-I (Arg 55 , Arg 56 ) and are important for IGF2R binding (15). In addition, we recently showed that IGF-II residues in the nonconserved C domain are important for IGF2R binding and play a role in conferring the specificity of binding (16). However, the effects of mutation on residue Ala 54 or Leu 55 or the C domain are insufficient to account for the complete lack of binding by IGF-I, suggesting that another determinant provides the IGF-II specificity.
In this study, we have significantly added to our understanding of the IGF2R binding site on IGF-II using a mutagenesis approach and surface plasmon resonance with isolated IGF2R domains. We have revealed a novel hydrophobic patch critical for IGF2R binding, which encompasses residues Phe 19 and Leu 53 and includes Thr 16 and Asp 52 . We have also identified the major determinant on IGF-II responsible for the specificity of IGF2R binding. (Table 1) were purchased from Geneworks Pty. Ltd. (Adelaide, South Australia). Restriction enzymes were from New England Biolabs (Hitchin, UK) or Geneworks Pty. Ltd. Long TM R 3 IGF-I, human IGF-I, and IGF-II were purchased from GroPep Ltd. (Adelaide, South Australia). Greiner Lumitrac 600 96-well plates were from Omega Scientific (Tarzana, CA). Human insulin was purchased from Novo Nordisk (Bagsvaerd, Denmark). The DELFIA europium-labeling kit and DELFIA enhancement solution were purchased from PerkinElmer Life Sciences. Europium-labeled IGF-II was produced as described by Denley et al. (17) according to the manufacturer's instructions. P6 cells (BALB/c3T3 cells overexpressing the human IGF-1R) (18) were a kind gift from Professor R. Baserga (Philadelphia, PA).

Materials-Oligonucleotides
Construction of Expression Plasmids Encoding Human IGF-II Analogues-The IGF expression vector was developed by King et al. (19), and the cDNA encoding IGF-II was introduced into the vector as previously described (20). The QuikChange sitedirected mutagenesis kit (Stratagene) was used to incorporate the 14 different mutations into IGF-II (Table 1). Resultant constructs were transformed into Escherichia coli JM101 (lac Iq) for expression. IGF-II mutants were expressed as fusion proteins with the first 11 amino acids of porcine growth hormone ([Met 1 ]pGH-(1-11)) after isopropyl ␤-Dthiogalactoside induction. Inclusion bodies were isolated as previously described (19).
Refolding and Purification of IGF-II Analogues-Inclusion bodies containing the IGF-II fusion peptides were processed essentially as described (20) with some modifications. Briefly, washed inclusion bodies from a 0.5-liter fermentation were solubilized in 8 M urea containing 40 mM glycine, 0.1 M Tris, and 16 mM dithiothreitol at pH 2.0. Inclusion bodies were immediately desalted on a Superdex 75 column (1 ϫ 30 cm (GE Healthcare) in two 2-ml batches) using the same buffer but with 1.6 mM dithiothreitol. Fractions containing the IGF-II fusion protein were identified by SDS-PAGE and reverse phase chromatogra-phy and pooled prior to folding in 2.5 M urea, 12.5 mM glycine, 0.7 M Tris, 5 mM EDTA, 0.5 mM dithiothreitol, 1 mM 2-hydroxyethyl disulfide, pH 9.1, and dilution to less than 0.1 mg/ml. Cleavage of the fusion partner and a final reverse phase HPLC cleanup were achieved as previously described (21,22). Purified proteins were analyzed by mass spectroscopy and N-terminal sequencing (Dr. Chris Bagley and Chris Cursaro (Adelaide Proteomics Facility)) and were shown to have the correct masses and to be greater than 95% pure. Quantitation of analogues was performed by comparing analytical C4 HPLC profiles with profiles of standard Long TM R 3 IGF-I preparations (23).
Circular Dichroism-CD spectra were recorded on a JASCO J-815 CD spectrometer. Spectra were recorded from 250 to 185 nm with a 1.0-nm step size using a 1.0-s response time and 1.0-nm bandwidth. Three accumulations were recorded for each spectrum. Samples were analyzed in a quartz cuvette with a 0.1-cm path length. Spectra were background-corrected by subtraction of the spectrum of buffer alone. In order to allow direct comparison of spectra without the interference from small differences in quantification of protein concentration, all spectra were normalized to the ellipticity measured at 207 nm (24).
IGF-1R Binding Assay-Receptor binding affinities were measured using an assay similar to that described for analyzing epidermal growth factor binding to the epidermal growth factor receptor (25) and outlined in Ref. 17. Briefly, P6 cells were serum-starved prior to lysis with lysis buffer (20 mM HEPES, 150 mM NaCl, 1.5 mM MgCl 2 , 10% (v/v) glycerol, 1% (v/v) Triton X-100, 1 mM EGTA, and 1 mM phenylmethylsulfonyl fluoride, pH 7.5) for 1 h at 4°C. Lysates were centrifuged for 10 min at 3500 rpm at 4°C, and then 100-l aliquots were added to wells of a white Greiner Lumitrac 600 plate previously coated with anti-IGF-1R antibody 24-31 (2.5 g/ml) (26). Approximately 500,000 fluorescence counts of europiumlabeled IGF-II were added to each well along with increasing concentrations of unlabeled competitor and incubated for 16 h at 4°C. Wells were washed four times with 20 mM Tris, 150 mM NaCl, and 0.1% (v/v) Tween 20 (TBST) and twice with water, and then DELFIA enhancement solution (100 l/well) was added. Time-resolved fluorescence was measured using 340-nm excitation and 612-nm emission filters with a BMG Laboratory Technologies Polarstar fluorimeter (Mornington, Australia).
Kinetic Studies of IGF2R Binding Using Surface Plasmon Resonance-Each IGF2R fragment was coupled to the CM5 biosensor chip using a similar method to that previously described (16). Briefly, the IGF2R fragment to be bound was prepared in 10 mM sodium acetate, pH 4.6, at 18 g/ml and injected over the NHS/EDC activated chip surface in HBS running buffer (10 mM HEPES, 150 mM NaCl, 3.4 mM EDTA, 0.005% surfactant P20, pH 7.4). The surfaces were then deactivated by 1 M ethanolamine. A blank flow cell was used as a reference on all chips. Binding analyses were performed at 25°C using a BIAcore 2000 biosensor (BIAcore, Uppsala, Sweden) within the Flinders Advanced Analytical Laboratories (Flinders University, South Australia).
The IGF2R-d10 -13, IGF2R-d11-13, and IGF2R-d11-14 fragments were coupled to the biosensor surface to give a final resonance value of 2000 response units, whereas IGF2R-d11 and IGF2R-d11-12 were coupled to final values of 500 and 1000 response units, respectively. The proportion of active surface was Ͼ90% for all surfaces except for the IGF2R-d11-14 surface, which was 70% active. IGF-II, IGF-I, and IGF analogues were injected over the chip surfaces at the following concentrations: IGF-II, E12K IGF-II, V14T IGF-II, Q18A IGF-II, F19L IGF-II,  F28L IGF-II, and V43M IGF-II at concentrations ; T16A IGF-II, F19A IGF-II, F19S IGF-II, F19Y  IGF-II, D52K IGF-II, D52E and D52N IGF-II, L53A IGF-II,  IGF-I, and A13T IGF-I at 500, 400, 200, 100, and 50 nM in HBS running buffer for 2.5 min at a flow rate of 40 l/min to minimize mass transfer effects. Dissociation of bound analyte in HBS buffer alone was measured at the same flow rate for 15 min. All flow cells were regenerated by injection of 60 l of 10 mM HCl. Reference flow cell data were subtracted from all runs to account for bulk refractive index due to the buffer.
All kinetic data were analyzed using the BIAevaluation 3.2 software (Biacore AB, Uppsala, Sweden). IGF2R-d11 and IGF2R-d11-12 curves were fitted globally across all concentrations to a steady state affinity model. The steady state model describes the affinity of the interaction at equilibrium. The IGF2R-d10 -13, IGF2R-d11-13, and IGF2R-d11-14 curves were fitted using a two-state conformational change model and were derived from fitting of the 50 nM curves. This model describes a 1:1 binding interaction with a conformational change upon binding, resulting in two association and dissociation rates. This model resulted in the best fitting of the data and has been previously used for analysis of binding to IGF2R fragments (13). Each mutant was analyzed in duplicate in a minimum of two separate experiments. The affinity constant (K A ) derived from eight separate experiments on two separate chips for IGF-II on IGF2R-d10 -13 was 1.5 Ϯ 0.4 ϫ 10 8 M Ϫ1 , and a similar variation between experiments was seen on all surfaces with all mutants analyzed.

Production and Characterization of IGF-II Analogues-
Fourteen novel IGF-II analogues were made to define the IGF2R binding site. Only Val 43 had previously been mutated, and substitutions were designed to probe the properties of each residue. Analogues with Ala substitutions included T16A IGF-II, Q18A IGF-II, F19A IGF-II, and L53A IGF-II. The potential role of hydrophobicity at position 19 was investigated by further substitutions with Ser, Tyr, and Leu. The effect of charge reversal at position 12 (Glu) was probed by substitution with Lys. Val 14 was changed to the equivalent residue in insulin (Thr), and the smaller hydrophobic residue of Leu was used in place of Phe at position 28. Val 43 was substituted with Met. The equivalent mutation in IGF-I (V44M), identified as the first missense mutation in the IGF-I gene, leads to an IGF-I with severely impaired IGF-1R binding (29). The contribution of the Asp 52 side chain was probed by substitution with Asn, Gln, and Glu.
Far UV circular dichroism was used to test the structural integrity of the analogues. Misfolding of IGFs can be identified by a significant change in the far UV CD spectrum (30). The following analogues exhibited spectra indistinguishable from IGF-II: E12K IGF-II, V14T IGF-II, Q18A IGF-II, F19A IGF-II,  F19Y IGF-II, F19S IGF-II, F19L IGF-II, F28L IGF-II, V43M  IGF-II, D52N IGF-II, and L53A IGF-II (Fig. 1). Small deviations in CD signal intensities for T16A IGF-II, D52K IGF-II, and D52E IGF-II were observed, but the introduced changes appear to have little overall effect on secondary structure, and we believe that correct folding was achieved.
Further evidence of structural integrity was provided by analyzing IGF-1R binding. Misfolded IGF-I has a 30-fold lower affinity for the IGF-1R (30), and a similar reduction in IGF-1R affinity is seen for misfolded IGF-II (31). 6 All of the mutants that had CD spectra that differed from IGF-II (T16A, D52K, and D52E) retained the ability to bind the IGF-1R with high affinity (only 2-3-fold lower affinity; Table 2). D52E IGF-II had the greatest change in far UV CD spectrum, yet its IGF-1R binding affinity was only 3-fold lower than IGF-II (Table 2). We therefore believe this mutant is correctly folded, but the difference in CD spectrum represents local structural perturbations.
Although the CD spectra indicated no structural perturbations for V14T IGF-II, F28L IGF-II, and V43M IGF-II, their IGF-1R binding was significantly perturbed. This result was not unexpected, since all three residues correspond to IGF-I residues found in the "classic" binding surface that is essential for IGF-1R binding (32). Mutation of Val 11 in IGF-I (equivalent to Val 14 in IGF-II) to Thr also results in a reduction in IGF-1R binding affinity (3-fold) (33). Mutation of Val 44 IGF-I or Val A3 insulin (Val 43 in IGF-II) to Met or Leu, respectively, disrupts receptor binding without causing gross structural perturbation (34,35).
In contrast, no binding was detectable with F19A IGF-II, F19S IGF-II, or F19Y IGF-II even at a concentration of 500 nM. However, F19L IGF-II had 6.4-and 3.9-fold higher affinities than IGF-II for IGF2R-d11 and IGF2R-d11-12. This would suggest that Phe 19 is involved in a hydrophobic interaction with the IGF2R, and replacing the aromatic ring with a bulky aliphatic side chain can even enhance binding.
Both Leu 53 and Thr 16 are also critical for binding to the low affinity fragments, with no binding detectable even at a concentration of 500 nM. In addition, mutation of Asp 52 to Lys, Glu, or Asn has a significant effect on binding to IGF2R-d11 and IGF2R-d11-12, with the greatest effect seen by substitution with lysine (20-fold lower affinity than IGF-II).

TABLE 4 BIAcore kinetic analysis of IGF-II analogues binding to the high affinity IGF2R fragments
BIAcore curves were fitted using a two-state conformational change model as outlined under "Experimental Procedures." Association rate constants (k a ), dissociation rate constants (k d ), and association equilibrium constants (K A ϭ k a /k d ) are given. Association equilibrium constants of IGF-II analogues relative to IGF-II are also presented (K A of IGF-II analogue/K A of IGF-II). -, no measurable binding at 500 nM. -   ------F19S  ------F19Y  ------F19L 3  detectable with F19A IGF-II, F19S  IGF-II, F19Y IGF-II, L53A IGF-II, and T16A IGF-II to IGF2R-d10 -13, IGF2R-d11-13, and IGF2R-d11-14. F19L IGF-II also showed improved binding over IGF-II for the high affinity frag-ments, predominantly due to slower dissociation rates. However, this effect was only 2-fold on the high affinity fragments versus up to 6.4fold higher affinity with the low affinity fragments. The D52N, D52K, and D52E IGF-II substitutions had a similar decrease in binding to the high affinity IGF2R fragments, as was seen with IGF2R-d11 and IGF2R-d11-12, with the D52K substitution having the greatest effect.
An additional IGF-I mutant (A13T IGF-I) was made using a similar approach as previously described (22). No significant structural changes were detectable by CD analysis following introduction of Thr at this position, and IGF-1R binding affinity was identical to IGF-I (data not shown), suggesting that A13T IGF-I was correctly folded. Interestingly, there is no detectable binding of IGF-I to all IGF2R fragments using surface plasmon resonance. However, the A13T IGF-I did bind IGF2R-d10 -13, albeit with a 25-fold lower affinity than IGF-II (Fig. 5). A similar result was seen for the other high affinity fragments (IGF2R-d11-13 and IGF2R-d11-14) and the low affinity fragments (IGF2R-d11 and IGF2R-d11-12) (data not shown).

DISCUSSION
With the lack of a structure of the IGF-II⅐IGF2R complex and only limited IGF-II mutagenesis data, we undertook a comprehensive study aimed at significantly improving our understanding of the IGF-II⅐IGF2R interaction and the molecular mechanism underlying the specificity of the interaction. A series of 14 novel IGF-II analogues were produced recombinantly and tested for their binding to five different IGF2R fragments. The IGF2R fragments encompassed either domain 11, which is essential for IGF-II binding, or both domains 11 and 13, which together are required to achieve high affinity binding (9 -11). The binding data reveal a completely new IGF-II surface important for interaction with the IGF2R. This surface encompasses residues Thr 16 , Phe 19 , Asp 52 , and Leu 53 of IGF-II, which form a continuous patch (see Fig. 6). This binding surface lies adjacent to residues Ala 54 and Leu 55 , previously identified as important for IGF2R binding (15), but is further away from residues Phe 48 , Arg 49 , and Ser 50 , also reported to be involved in IGF2R binding (14).
The involvement in IGF2R binding of the hydrophobic residues Phe 19 and Leu 53 , along with the adjacent Leu 55 , is consistent with the suggestion that IGF-II binds to the IGF2R using hydrophobic interactions (10,13). Substitution of Phe 19 with

IGF2R Binding Site on IGF-II
polar residues, including the similarly sized Tyr, is detrimental to IGF2R binding. However, an increase in binding affinity to both low and high affinity IGF2R fragments is achieved by substitution with a bulky aliphatic residue (Leu). Consistent with this, a hydrophobic patch on the IGF2R is critical for IGF-II binding (10,13).
Probing the contribution of Asp 52 by substitution with Lys, Glu, or Asn suggests that although charge is important, with the greatest effect seen with the Lys substitution, size also is important at this position for IGF2R binding. A significant reduction in binding was observed for D52E IGF-II.
Mutation of those residues that proved important for IGF-1R binding (Val 14 , Phe 28 , and Val 43 ) had little or no effect on IGF2R binding, and it is noteworthy that they are located on the opposite side of the IGF-II molecule (Fig. 6). This observation is consistent with studies showing that mutation of Phe 26 and Tyr 27 also had little or no effect on IGF2R binding, despite a significant decrease in IGF-1R binding (14,37). Interestingly, all of the mutations that had an effect on IGF2R binding disrupted binding to both the low affinity and high affinity IGF2R fragments. We can therefore deduce that these residues are making contact with domain 11 only, and our observation supports the suggestion that the role of domain 13 is to stabilize the complex, resulting in high affinity binding (9 -11).
Thr 16 of IGF-II is the only residue analyzed in this study that differs from its equivalent residue in IGF-I (Ala 13 ) and therefore can be used to probe the IGF-II binding preference of the IGF2R. Previous studies have identified that Ala 54 , Leu 55 , and the C domain of IGF-II contribute to the specificity of the IGF-II⅐IGF2R interaction. However, mutation of either Ala 54 or Leu 55 to the equivalent residues in IGF-I (Arg and Arg, respectively) only resulted in a 3-4-fold lower binding affinity (15). Similarly, substituting the IGF-II C domain with the IGF-I C domain only reduced binding by 6-fold (16). Therefore, although Ala 54 , Leu 55 , and the IGF-II C domain contain determinants important for IGF2R binding and do play a role in specificity, they are not the major determinants. Since mutation of Thr 16 to Ala resulted in a dramatic loss in binding to all IGF2R fragments tested (with only a 3.5-fold reduction in IGF-1R binding), we believe this residue represents the major determinant for the specificity of the IGF-II/IGF2R interaction. In support of this, we showed that A13T IGF-I does bind to the IGF2R fragments, albeit with lower affinity than IGF-II (Fig. 5),  IGF-II and F19L IGF-II at 50 nM and F19S IGF-II, F19Y IGF-II, D52K IGF-II D52E IGF-II, and D52N IGF-II at 500 nM were passed over the sensor chip as described in the legend to Fig. 2. FIGURE 5. BIAcore analysis of A13T IGF-I binding to IGF2R-d10 -13. IGF-II at 50 nM and A13T IGF-I and IGF-I at 500 nM were passed over a sensor chip surface coupled with IGF2R-d10 -13 as described in the legend to Fig. 2. Association constants (K A ) were determined with a two-state conformational change model for IGF-II and a steady state affinity model for A13T IGF-I. Association constants of A13T IGF-I and IGF-I relative to IGF-II are also presented (K A of IGF-II analogue/K A of IGF-II). Dash, no measurable binding at 500 nM.
whereas there is no detectable binding of IGF-I on either low or high affinity IGF2R fragments.
The ability of the IGF2R to bind IGF-II evolved in mammals after the divergence of monotremes and correlates with the emergence of placental development (38). The cation-independent mannose 6phosphate receptors of nonmammalian species, including chicken and Xenopus, are unable to bind IGF-II due to differences in the sequence of the IGF-II binding site (39). Importantly, Thr 16 is conserved in IGF-II of all mammalian species in which the IGF2R has an IGF-II binding motif.
Thr 16 could play a direct role in IGF2R binding. Thr 16 of IGF-II is found in a deep pocket and is partially surface-exposed, whereas this pocket does not exist in IGF-I, since the corresponding residue (Ala 13 ) is more exposed and forms part of a flatter surface (Fig. 7A). The presence of the bulkier and somewhat polar Thr at position 16 could also be influencing the local structure in this region. There are significant differences between IGF-I and IGF-II in the local structures around the residues Phe 19 , Asp 52 , and Leu 53 , residues identified in this study to be important for  (14,15). IGF-II residues that when mutated inhibit interactions with IGF2R by Յ2and Ն6-fold (orange) or Ն6-fold (red) or have no effect on binding (black) are highlighted. The atomic coordinates (Protein Data Bank code 1IGL) (40) were used to construct this diagram. Molecular graphics images were produced using the UCSF Chimera package from the Resource for Biocomputing, Visualization, and Informatics at the University of California, San Francisco. IGF2R binding (Fig. 7B). These structural perturbations may result from the accommodation of the critical Thr residue in IGF-II.
In summary, we have defined a novel binding surface on IGF-II that is important for binding to domain 11 of the IGF2R and confirmed that high affinity binding is achieved predominantly through hydrophobic interactions. We have also identified Thr 16 as the major determinant of IGF-II binding specificity for the IGF2R. This study provides greater understanding of the IGF-II/IGF2R interaction, which is responsible for regulating the actions of IGF-II in normal development and plays a significant role in tumor suppression.