The relationship between insulin bioactivity and structure in the NH2-terminal A-chain helix

High precision protein functional site detection using 3D convolutional neural networks

Motivation

Accurate annotation of protein functions is fundamental for understanding molecular and cellular physiology. Data-driven methods hold promise for systematically deriving rules underlying the relationship between protein structure and function. However, the choice of protein structural representation is critical. Pre-defined biochemical features emphasize certain aspects of protein properties while ignoring others, and therefore may fail to capture critical information in complex protein sites.

Results

In this paper, we present a general framework that applies 3D convolutional neural networks (3DCNNs) to structure-based protein functional site detection. The framework can extract task-dependent features automatically from the raw atom distributions. We benchmarked our method against other methods and demonstrate better or comparable performance for site detection. Our deep 3DCNNs achieved an average recall of 0.955 at a precision threshold of 0.99 on PROSITE families, detected 98.89 and 92.88% of nitric oxide synthase and TRYPSIN-like enzyme sites in Catalytic Site Atlas, and showed good performance on challenging cases where sequence motifs are absent but a function is known to exist. Finally, we inspected the individual contributions of each atom to the classification decisions and show that our models successfully recapitulate known 3D features within protein functional sites.

Availability and implementation

The 3DCNN models described in this paper are available at https://simtk.org/projects/fscnn.

Supplementary information

Supplementary data are available at Bioinformatics online.

Insulin in motion: The A6-A11 disulfide bond allosterically modulates structural transitions required for insulin activity

The structural transitions required for insulin to activate its receptor and initiate regulation of glucose homeostasis are only partly understood. Here, using ring-closing metathesis, we substitute the A6-A11 disulfide bond of insulin with a rigid, non-reducible dicarba linkage, yielding two distinct stereo-isomers (cis and trans). Remarkably, only the cis isomer displays full insulin potency, rapidly lowering blood glucose in mice (even under insulin-resistant conditions). It also posseses reduced mitogenic activity in vitro. Further biophysical, crystallographic and molecular-dynamics analyses reveal that the A6-A11 bond configuration directly affects the conformational flexibility of insulin A-chain N-terminal helix, dictating insulin’s ability to engage its receptor. We reveal that in native insulin, contraction of the C_α-C_α distance of the flexible A6-A11 cystine allows the A-chain N-terminal helix to unwind to a conformation that allows receptor engagement. This motion is also permitted in the cis isomer, with its shorter C_α-C_α distance, but prevented in the extended trans analogue. These findings thus illuminate for the first time the allosteric role of the A6-A11 bond in mediating the transition of the hormone to an active conformation, significantly advancing our understanding of insulin action and opening up new avenues for the design of improved therapeutic analogues.

Effects of localized interactions and surface properties on stability of protein-based therapeutics

OBJECTIVES

Protein-based therapeutics garner significant attention because of exquisite specificity and limited side effects and are now being used to accomplish targeted delivery of small-molecule drugs. This review identifies and highlights individual chemical attributes and categorizes how site-specific changes affect protein stability based on published high-resolution molecular analyses.

KEY FINDINGS

Because it is challenging to determine the mechanisms by which the stability of large, complex molecules is altered and data are sparse, smaller, therapeutic proteins (insulin, erythropoietin, interferons) are examined alongside antibody data. Integrating this large pool of information with the limited available studies on antibodies reveals common mechanisms by which specific alterations affect protein structure and stability.

SUMMARY

Physical and chemical stability of therapeutic proteins and antibody drug conjugates (ADCs) is of critical importance because insufficient stability prevents molecules from making it to market. Individual moieties on/near the surface of proteins have substantial influence on structure and stability. Seemingly small, superficial modification may have far-reaching consequences on structure, conformational dynamics, and solubility of the protein, and hence physical stability of the molecule. Chemical modifications, whether spontaneous (e.g. oxidation, deamidation) or intentional, as with ADCs, may adversely impact stability by disrupting local surface properties or higher order protein structure.

Inhibition of Insulin Amyloid Fibrillation by a Novel Amphipathic Heptapeptide: MECHANISTIC DETAILS STUDIED BY SPECTROSCOPY IN COMBINATION WITH MICROSCOPY

The aggregation of insulin into amyloid fibers has been a limiting factor in the development of fast acting insulin analogues, creating a demand for excipients that limit aggregation. Despite the potential demand, inhibitors specifically targeting insulin have been few in number. Here we report a non-toxic and serum stable-designed heptapeptide, KR7 (KPWWPRR-NH2), that differs significantly from the primarily hydrophobic sequences that have been previously used to interfere with insulin amyloid fibrillation. Thioflavin T fluorescence assays, circular dichroism spectroscopy, and one-dimensional proton NMR experiments suggest KR7 primarily targets the fiber elongation step with little effect on the early oligomerization steps in the lag time period. From confocal fluorescence and atomic force microscopy experiments, the net result appears to be the arrest of aggregation in an early, non-fibrillar aggregation stage. This mechanism is noticeably different from previous peptide-based inhibitors, which have primarily shifted the lag time with little effect on later stages of aggregation. As insulin is an important model system for understanding protein aggregation, the new peptide may be an important tool for understanding peptide-based inhibition of amyloid formation.

All-atom structural models of insulin binding to the insulin receptor in the presence of a tandem hormone-binding element

Insulin regulates blood glucose levels in higher organisms by binding to and activating insulin receptor (IR), a constitutively homodimeric glycoprotein of the receptor tyrosine kinase (RTK) superfamily. Therapeutic efforts in treating diabetes have been significantly impeded by the absence of structural information on the activated form of the insulin/IR complex. Mutagenesis and photo-crosslinking experiments and structural information on insulin and apo-IR strongly suggest that the dual-chain insulin molecule, unlike the related single-chain insulin-like growth factors, binds to IR in a very different conformation than what is displayed in storage forms of the hormone. In particular, hydrophobic residues buried in the core of the folded insulin molecule engage the receptor. There is also the possibility of plasticity in the receptor structure based on these data, which may in part be due to rearrangement of the so-called CT-peptide, a tandem hormone-binding element of IR. These possibilities provide opportunity for large-scale molecular modeling to contribute to our understanding of this system. Using various atomistic simulation approaches, we have constructed all-atom structural models of hormone/receptor complexes in the presence of CT in its crystallographic position and a thermodynamically favorable displaced position. In the "displaced-CT" complex, many more insulin-receptor contacts suggested by experiments are satisfied, and our simulations also suggest that R-insulin potentially represents the receptor-bound form of hormone. The results presented in this work have further implications for the design of receptor-specific agonists/antagonists.

The A-chain of Insulin Contacts the Insert Domain of the Insulin Receptor

The contribution of the insulin A-chain to receptor binding is investigated by photo-cross-linking and nonstandard mutagenesis. Studies focus on the role of Val(A3), which projects within a crevice between the A- and B-chains. Engineered receptor alpha-subunits containing specific protease sites ("midi-receptors") are employed to map the site of photo-cross-linking by an analog containing a photoactivable A3 side chain (para-azido-Phe (Pap)). The probe cross-links to a C-terminal peptide (residues 703-719 of the receptor A isoform, KTFEDYLHNVVFVPRPS) containing side chains critical for hormone binding (underlined); the corresponding segment of the holoreceptor was shown previously to cross-link to a Pap(B25)-insulin analog. Because Pap is larger than Val and so may protrude beyond the A3-associated crevice, we investigated analogs containing A3 substitutions comparable in size to Val as follows: Thr, allo-Thr, and alpha-aminobutyric acid (Aba). Substitutions were introduced within an engineered monomer. Whereas previous studies of smaller substitutions (Gly(A3) and Ser(A3)) encountered nonlocal conformational perturbations, NMR structures of the present analogs are similar to wild-type insulin; the variant side chains are accommodated within a native-like crevice with minimal distortion. Receptor binding activities of Aba(A3) and allo-Thr(A3) analogs are reduced at least 10-fold; the activity of Thr(A3)-DKP-insulin is reduced 5-fold. The hormone-receptor interface is presumably destabilized either by a packing defect (Aba(A3)) or by altered polarity (allo-Thr(A3) and Thr(A3)). Our results provide evidence that Val(A3), a site of mutation causing diabetes mellitus, contacts the insert domain-derived tail of the alpha-subunit in a hormone-receptor complex.

Structural interpretation of reduced insulin activity as seen in the crystal structure of human Arg-insulin

The N-terminal glycine of the A-chain in insulin is reported to be one of the residues that binds to the insulin receptor. Modifications near this region lead to variations in the biological activity of insulin. One such modification viz., an addition of an arginine at the N-terminal A-chain, was reported to possess two-thirds the activity of native insulin. The crystal structure of 2 zinc human arg (A0) insulin has been elucidated to 2A resolution to understand the mechanism of reduction in insulin activity. A conformational transition from T6 to T3R3(f) and a decrease in the surface accessibility of residues in the so called receptor binding region have been observed. The presence of arginine has also induced distortions in the A chain N-terminal helix. The subtle conformational alterations like decrease in surface accessibility, alterations in the charge surface and changes in the relative orientation of the two helices in the A chain may be responsible for the reduction in activity.

Solution Structure and Characterization of the LGR8 Receptor Binding Surface of Insulin-like Peptide 3

Insulin-like peptide 3 (INSL3), a member of the relaxin peptide family, is produced in testicular Leydig cells and ovarian thecal cells. Gene knock-out experiments have identified a key biological role in initiating testes descent during fetal development. Additionally, INSL3 has an important function in mediating male and female germ cell function. These actions are elicited via its recently identified receptor, LGR8, a member of the leucine-rich repeat-containing G-protein-coupled receptor family. To identify the structural features that are responsible for the interaction of INSL3 with its receptor, its solution structure was determined by NMR spectroscopy together with in vitro assays of a series of B-chain alanine-substituted analogs. Synthetic human INSL3 was found to adopt a characteristic relaxin/insulin-like fold in solution but is a highly dynamic molecule. The four termini of this two-chain peptide are disordered, and additional conformational exchange is evident in the molecular core. Alanine-substituted analogs were used to identify the key residues of INSL3 that are responsible for the interaction with the ectodomain of LGR8. These include Arg(B16) and Val(B19), with His(B12) and Arg(B20) playing a secondary role, as evident from the synergistic effect on the activity in double and triple mutants involving these residues. Together, these amino acids combine with the previously identified critical residue, Trp(B27), to form the receptor binding surface. The current results provide clear direction for the design of novel specific agonists and antagonists of this receptor.

Study of the insulin dimerization: binding free energy calculations and per-residue free energy decomposition

A calculation of the binding free energy for the dimerization of insulin has been performed using the molecular mechanics-generalized Born surface area approach. The calculated absolute binding free energy is -11.9 kcal/mol, in approximate agreement with the experimental value of -7.2 kcal/mol. The results show that the dimerization is mainly due to nonpolar interactions. The role of the hydrogen bonds between the 2 monomers appears to give the direction of the interactions. A per-atom decomposition of the binding free energy has been performed to identify the residues contributing most to the self association free energy. Residues B24-B26 are found to make the largest favorable contributions to the dimerization. Other residues situated at the interface between the 2 monomers were found to make favorable but smaller contributions to the dimerization: Tyr B16, Val B12, and Pro B28, and to an even lesser extent, Gly B23. The energy decomposition on a per-residue basis is in agreement with experimental alanine scanning data. The results obtained from a single trajectory (i.e., the dimer trajectory is also used for the monomer analysis) and 2 trajectories (i.e., separate trajectories are used for the monomer and dimer) are similar.

Chiral mutagenesis of insulin. Foldability and function are inversely regulated by a stereospecific switch in the B chain

How insulin binds to its receptor is unknown despite decades of investigation. Here, we employ chiral mutagenesis-comparison of corresponding d and l amino acid substitutions in the hormone-to define a structural switch between folding-competent and active conformations. Our strategy is motivated by the T --> R transition, an allosteric feature of zinc-hexamer assembly in which an invariant glycine in the B chain changes conformations. In the classical T state, Gly(B8) lies within a beta-turn and exhibits a positive phi angle (like a d amino acid); in the alternative R state, Gly(B8) is part of an alpha-helix and exhibits a negative phi angle (like an l amino acid). Respective B chain libraries containing mixtures of d or l substitutions at B8 exhibit a stereospecific perturbation of insulin chain combination: l amino acids impede native disulfide pairing, whereas diverse d substitutions are well-tolerated. Strikingly, d substitutions at B8 enhance both synthetic yield and thermodynamic stability but markedly impair biological activity. The NMR structure of such an inactive analogue (as an engineered T-like monomer) is essentially identical to that of native insulin. By contrast, l analogues exhibit impaired folding and stability. Although synthetic yields are very low, such analogues can be highly active. Despite the profound differences between the foldabilities of d and l analogues, crystallization trials suggest that on protein assembly substitutions of either class can be accommodated within classical T or R states. Comparison between such diastereomeric analogues thus implies that the T state represents an inactive but folding-competent conformation. We propose that within folding intermediates the sign of the B8 phi angle exerts kinetic control in a rugged landscape to distinguish between trajectories associated with productive disulfide pairing (positive T-like values) or off-pathway events (negative R-like values). We further propose that the crystallographic T -->R transition in part recapitulates how the conformation of an insulin monomer changes on receptor binding. At the very least the ostensibly unrelated processes of disulfide pairing, allosteric assembly, and receptor binding appear to utilize the same residue as a structural switch; an "ambidextrous" glycine unhindered by the chiral restrictions of the Ramachandran plane. We speculate that this switch operates to protect insulin-and the beta-cell-from protein misfolding.

Diabetes-Associated Mutations in Human Insulin: Crystal Structure and Photo-Cross-Linking Studies of A-Chain Variant InsulinWakayama†,‡

Naturally occurring mutations in insulin associated with diabetes mellitus identify critical determinants of its biological activity. Here, we describe the crystal structure of insulin Wakayama, a clinical variant in which a conserved valine in the A chain (residue A3) is substituted by leucine. The substitution occurs within a crevice adjoining the classical receptor-binding surface and impairs receptor binding by 500-fold, an unusually severe decrement among mutant insulins. To resolve whether such decreased activity is directly or indirectly mediated by the variant side chain, we have determined the crystal structure of Leu(A3)-insulin and investigated the photo-cross-linking properties of an A3 analogue containing p-azidophenylalanine. The structure, characterized in a novel crystal form as an R(6) zinc hexamer at 2.3 A resolution, is essentially identical to that of the wild-type R(6) hexamer. The variant side chain remains buried in a nativelike crevice with small adjustments in surrounding side chains. The corresponding photoactivatable analogue, although of low affinity, exhibits efficient cross-linking to the insulin receptor. The site of photo-cross-linking lies within a 14 kDa C-terminal domain of the alpha-subunit. This domain, unrelated in sequence to the major insulin-binding region in the N-terminal L1 beta-helix, is also contacted by photoactivatable probes at positions A8 and B25. Packing of Val(A3) at this interface may require a conformational change in the B chain to expose the A3-related crevice. The structure of insulin Wakayama thus evokes the reasoning of Sherlock Holmes in "the curious incident of the dog in the night": the apparent absence of structural perturbations (like the dog that did not bark) provides a critical clue to the function of a hidden receptor-binding surface.

Structural and functional characteristics of the Val44Met insulin-like growth factor I missense mutation: correlation with effects on growth and development

We have previously described the phenotype resulting from a missense mutation in the IGF-I gene, which leads to expression of IGF-I with a methionine instead of a valine at position 44 (Val44Met IGF-I). This mutation caused severe growth and mental retardation as well as deafness evident at birth and growth retardation in childhood, but is relatively well tolerated in adulthood. We have conducted a biochemical and structural analysis of Val44Met IGF-I to provide a molecular basis for the phenotype observed. Val44Met IGF-I exhibits a 90-fold decrease in type 1 IGF receptor (IGF-1R) binding compared with wild-type human IGF-I and only poorly stimulates autophosphorylation of the IGF-1R. The ability of Val44Met IGF-I to signal via the extracellular signal-regulated kinase 1/2 and Akt/protein kinase B pathways and to stimulate DNA synthesis is correspondingly poorer. Binding or activation of both insulin receptor isoforms is not detectable even at micromolar concentrations. However, Val44Met IGF-I binds IGF-binding protein-2 (IGFBP-2), IGFBP-3, and IGFBP-6 with equal affinity to IGF-I, suggesting the maintenance of overall structure, particularly in the IGFBP binding domain. Structural analysis by nuclear magnetic resonance confirms retention of near-native structure with only local side-chain disruptions despite the significant loss of function. To our knowledge, our results provide the first structural study of a naturally occurring mutant human IGF-I associated with growth and developmental abnormalities and identifies Val44 as an essential residue involved in the IGF-IGF-1R interaction.

Enhancing the Activity of Insulin at the Receptor Interface: Crystal Structure and Photo-Cross-Linking of A8 Analogues

The receptor-binding surface of insulin is broadly conserved, reflecting its evolutionary optimization. Neighboring positions nevertheless offer an opportunity to enhance activity, through either transmitted structural changes or introduction of novel contacts. Nonconserved residue A8 is of particular interest as Thr(A8) --> His substitution (a species variant in birds and fish) augments the potency of human insulin. Diverse A8 substitutions are well tolerated, suggesting that the hormone-receptor interface is not tightly packed at this site. To resolve whether enhanced activity is directly or indirectly mediated by the variant A8 side chain, we have determined the crystal structure of His(A8)-insulin and investigated the photo-cross-linking properties of an A8 analogue containing p-azidophenylalanine. The structure, characterized as a T(3)R(3)(f) zinc hexamer at 1.8 A resolution, is essentially identical to that of native insulin. The photoactivatable analogue exhibits efficient cross-linking to the insulin receptor. The site of cross-linking lies within a 14 kDa C-terminal domain of the alpha-subunit. This contact, to our knowledge the first to be demonstrated from the A chain, is inconsistent with a recent model of the hormone-receptor complex derived from electron microscopy. Optimizing the binding interaction of a nonconserved side chain on the surface of insulin may thus enhance its activity.

Diabetes-associated mutations in insulin: consecutive residues in the B chain contact distinct domains of the insulin receptor

How insulin binds to and activates the insulin receptor has long been the subject of speculation. Of particular interest are invariant phenylalanine residues at consecutive positions in the B chain (residues B24 and B25). Sites of mutation causing diabetes mellitus, these residues occupy opposite structural environments: Phe(B25) projects from the surface of insulin, whereas Phe(B24) packs against the core. Despite these differences, site-specific cross-linking suggests that each contacts the insulin receptor. Photoactivatable derivatives of insulin containing respective p-azidophenylalanine substitutions at positions B24 and B25 were synthesized in an engineered monomer (DKP-insulin). On ultraviolet irradiation each derivative cross-links efficiently to the receptor. Packing of Phe(B24) at the receptor interface (rather than against the core of the hormone) may require a conformational change in the B chain. Sites of cross-linking in the receptor were mapped to domains by Western blot. Remarkably, whereas B25 cross-links to the C-terminal domain of the alpha subunit in accord with previous studies (Kurose, T., et al. (1994) J. Biol. Chem. 269, 29190-29197), the probe at B24 cross-links to its N-terminal domain (the L1 beta-helix). Our results demonstrate that consecutive residues in insulin contact widely separated sequences in the receptor and in turn suggest a revised interpretation of electron-microscopic images of the complex. By tethering the N- and C-terminal domains of the extracellular alpha subunit, insulin is proposed to stabilize an active conformation of the disulfide-linked transmembrane tyrosine kinase.

A Comparison of the Dynamic Behavior of Monomeric and Dimeric Insulin Shows Structural Rearrangements in the Active Monomer

Molecular dynamics (MD) simulations (5-10ns in length) and normal mode analyses were performed for the monomer and dimer of native porcine insulin in aqueous solution; both starting structures were obtained from an insulin hexamer. Several simulations were done to confirm that the results obtained are meaningful. The insulin dimer is very stable during the simulation and remains very close to the starting X-ray structure; the RMS fluctuations calculated from the MD simulation agree with the experimental B-factors. Correlated motions were found within each of the two monomers; they can be explained by persistent non-bonded interactions and disulfide bridges. The correlated motions between residues B24 and B26 of the two monomers are due to non-bonded interactions between the side-chains and backbone atoms. For the isolated monomer in solution, the A chain and the helix of the B chain are found to be stable during 5ns and 10ns MD simulations. However, the N-terminal and the C-terminal parts of the B chain are very flexible. The C-terminal part of the B chain moves away from the X-ray conformation after 0.5-2.5ns and exposes the N-terminal residues of the A chain that are thought to be important for the binding of insulin to its receptor. Our results thus support the hypothesis that, when monomeric insulin is released from the hexamer (or the dimer in our study), the C-terminal end of the monomer (residues B25-B30) is rearranged to allow binding to the insulin receptor. The greater flexibility of the C-terminal part of the beta chain in the B24 (Phe-->Gly) mutant is in accord with the NMR results. The details of the backbone and side-chain motions are presented. The transition between the starting conformation and the more dynamic structure of the monomers is characterized by displacements of the backbone of Phe B25 and Tyr B26; of these, Phe B25 has been implicated in insulin activation.

A new cell secreting insulin

The pancreatic beta-cell is the only cell in animals that expresses the insulin gene and secretes insulin protein. We have found copious release of immunoreactive and bioactive insulin into the medium from the primary culture of carp adipocytes. Glucose augmented this release to more than 2-fold, and glucose transporter, Glut2, was detected in these cells. These all reflect characteristics of a pancreatic beta-cell. The expression of the adipocyte-specific flotillin gene, the presence of peroxisomal proliferator-activated receptor gamma and Glut4, and the colocalization of insulin and leptin confirmed the identity of these cells as adipocytes. Purified carp adipocyte insulin (AdpInsl) comigrated with porcine and bovine insulin in SDS-PAGE, indicating the similarity of their molecular sizes (5.5 kDa). AdpInsl strongly reduced hyperglycemia in streptozotocin-induced diabetic rats. It also stimulated significantly higher glucose uptake in carp and hamster adipocytes than porcine insulin. Adipocyte RNA hybridized with rat and zebrafish insulin cDNA showing the expression of the insulin gene in this cell. Using oligonucleotide primers designed on the basis of conserved insulin domain, AdpInsl cDNA was reverse transcribed, cloned, and sequenced. The deduced amino acid sequence of AdpInsl A and B chain exhibited 98% homology with zebrafish and more than 70% homology with human, porcine, and murine insulin. To understand the structure-function relationship between AdpInsl and mammalian beta-cell insulin, we have analyzed the amino acid sequences and three-dimensional structure of AdpInsl. In the critical determinant segment for receptor binding, AdpInsl has His at the A8 position instead of Thr in human and porcine insulin, and this attributed greater biological activity to AdpInsl. Our results show that carp adipocyte is a unique cell. As an insulin target cell it can express the insulin gene and secrete highly active insulin protein; thus, it may serve as a natural alternative to pancreatic beta-cell insulin.

Mechanism of insulin chain combination. Asymmetric roles of A-chain alpha-helices in disulfide pairing

The A and B chains of insulin combine to form native disulfide bridges without detectable isomers. The fidelity of chain combination thus recapitulates the folding of proinsulin, a precursor protein in which the two chains are tethered by a disordered connecting peptide. We have recently shown that chain combination is blocked by seemingly conservative substitutions in the C-terminal alpha-helix of the A chain. Such analogs, once formed, nevertheless retain high biological activity. By contrast, we demonstrate here that chain combination is robust to non-conservative substitutions in the N-terminal alpha-helix. Introduction of multiple glycine substitutions into the N-terminal segment of the A chain (residues A1-A5) yields analogs that are less stable than native insulin and essentially without biological activity. (1)H NMR studies of a representative analog lacking invariant side chains Ile(A2) and Val(A3) (A chain sequence GGGEQCCTSICSLYQLENYCN; substitutions are italicized and cysteines are underlined) demonstrate local unfolding of the A1-A5 segment in an otherwise native-like structure. That this and related partial folds retain efficient disulfide pairing suggests that the native N-terminal alpha-helix does not participate in the transition state of the reaction. Implications for the hierarchical folding mechanisms of proinsulin and insulin-like growth factors are discussed.

Non-standard insulin design: structure-activity relationships at the periphery of the insulin receptor

The design of insulin analogues has emphasized stabilization or destabilization of structural elements according to established principles of protein folding. To this end, solvent-exposed side-chains extrinsic to the receptor-binding surface provide convenient sites of modification. An example is provided by an unfavorable helical C-cap (Thr(A8)) whose substitution by favorable amino acids (His(A8) or Arg(A8)) has yielded analogues of improved stability. Remarkably, these analogues also exhibit enhanced activity, suggesting that activity may correlate with stability. Here, we test this hypothesis by substitution of diaminobutyric acid (Dab(A8)), like threonine an amino acid of low helical propensity. The crystal structure of Dab(A8)-insulin is similar to those of native insulin and the related analogue Lys(A8)-insulin. Although no more stable than native insulin, the non-standard analogue is twice as active. Stability and affinity can therefore be uncoupled. To investigate alternative mechanisms by which A8 substitutions enhance activity, multiple substitutions were introduced. Surprisingly, diverse aliphatic, aromatic and polar side-chains enhance receptor binding and biological activity. Because no relationship is observed between activity and helical propensity, we propose that local interactions between the A8 side-chain and an edge of the hormone-receptor interface modulate affinity. Dab(A8)-insulin illustrates the utility of non-standard amino acids in hypothesis-driven protein design.

Properties of small molecules affecting insulin receptor function

Small molecules with insulin mimetic effects and oral availability are of interest for potential substitution of insulin injections in the treatment of diabetes. We have searched databases for compounds capable of mimicking one epitope of the insulin molecule known to be involved in binding to the insulin receptor (IR). This approach identifies thymolphthalein, which is an apparent weak agonist that displaces insulin from its receptor, stimulates auto- and substrate phosphorylation of IR, and potentiates lipogenesis in adipocytes in the presence of submaximal concentrations of insulin. The various effects are observed in the 10(-5)-10(-3) M range of ligand concentration and result in partial insulin activity. Furthermore, analogues of the related phenol red and fluorescein molecules fully displace insulin from the IR ectodomain, however, without insulin agonistic effects. The interactions are further characterized by NMR, UV-vis, and fluorescence spectroscopies. It is shown that both fluorescence and UV-vis changes in the ligand spectra induced by IR fragments occur with Kd values similar to those obtained in the displacement assay. Nevertheless, insulin itself cannot completely abolish binding of the small molecules. Determination of the binding stoichiometry reveals multiple binding sites for ligands of which one overlaps with the insulin binding site on the receptor.

Activities of monomeric insulin analogs at position A8 are uncorrelated with their thermodynamic stabilities

Previous studies have demonstrated that the potency and thermodynamic stability of human insulin are enhanced in concert by substitution of Thr(A8) by arginine or histidine. These surface substitutions stabilize the N-terminal alpha-helix of the A chain, a key element of hormone-receptor recognition. Does enhanced stability necessarily imply enhanced activity? Here, we test by structure-based mutagenesis the relationship between the stability and activity of the hormone. To circumvent confounding effects of insulin self-association, A chain analogs were combined with a variant B chain (Asp(B10), Lys(B28), and Pro(B29) (DKP)) to create a monomeric template. Five analogs were obtained by chain combination; disulfide pairing proceeded in each case with native yield. CD and (1)H NMR spectra of the DKP analogs are essentially identical to those of DKP-insulin, indicating a correspondence of structures. Receptor binding affinities were determined by competitive displacement of (125)I-insulin from human placental membranes. Thermodynamic stabilities were measured by CD titration; unfolding was monitored as a function of guanidine concentration. In this broader collection of analogs receptor binding affinities are uncorrelated with stability. We suggest that receptor binding affinities of A8 analogs reflect local features of the hormone-receptor interface rather than the stability of the free hormone or the intrinsic C-capping propensity of the A8 side chain.

Flexibility and bioactivity of insulin: an NMR investigation of the solution structure and folding of an unusually flexible human insulin mutant with increased biological activity

The structure and folding of a novel human insulin mutant, [Thr(B27) --> Pro, Pro(B28) --> Thr]insulin (PT insulin), in aqueous solution and in mixtures of water and 2,2,2-trifluoroethanol (TFE) have been studied by NMR spectroscopy. It was found that PT insulin has a highly flexible structure in pure water and is present in at least two different conformations, although with an overall tertiary structure similar to that of native insulin. Furthermore, the native helical structures are poorly defined. Surprisingly, the mutant has a biological activity about 50% higher than native insulin. In contrast, in TFE/water solution the mutant reveals a propensity of forming a well-defined structure at the secondary structure level, similar to monomeric native insulin. Thus, as shown by a detailed determination of the structure from 208 distance restraints and 52 torsion angle restraints by distance geometry, simulated annealing, and restrained energy minimization, the native insulin helices (A2-A7, A13-A19, and B10-B19) as well as the beta-turn (B20-B23) are formed in 35% TFE. However, the amount of tertiary structure is decreased significantly in TFE/water solution. The obtained results suggest that only an overall tertiary fold, as observed for PT insulin in pure water, is necessary for expressing the biological activity of insulin, as long as the molecule is flexible and retains the propensity to form the secondary structure required for its receptor binding. In contrast, a compact secondary structure, as found for native insulin in solution, is unnecessary for the biological activity. A model for the receptor binding of insulin is suggested that relates the increased bioactivity to the enhanced flexibility of the mutant.

Hierarchical protein "un-design": insulin's intrachain disulfide bridge tethers a recognition alpha-helix

A hierarchical pathway of protein folding can enable segmental unfolding by design. A monomeric insulin analogue containing pairwise substitution of internal A6-A11 cystine with serine [[Ser(A6),Ser(A11),Asp(B10),Lys(B28),Pro(B29)]insulin (DKP[A6-A11](Ser))] was previously investigated as a model of an oxidative protein-folding intermediate [Hua, Q. X., et al. (1996) J. Mol. Biol. 264, 390-403]. Its structure exhibits local unfolding of an adjoining amphipathic alpha-helix (residues A1-A8), leading to a 2000-fold reduction in activity. Such severe loss of function, unusual among mutant insulins, is proposed to reflect the cost of induced fit: receptor-directed restoration of the alpha-helix and its engagement in the hormone's hydrophobic core. To test this hypothesis, we have synthesized and characterized the corresponding alanine analogue [[Ala(A6),Ala(A11),Asp(B10),Lys(B28), Pro(B29)]insulin (DKP[A6-A11](Ala))]. Untethering the A6-A11 disulfide bridge by either amino acid causes similar perturbations in structure and dynamics as probed by circular dichroism and (1)H NMR spectroscopy. The analogues also exhibit similar decrements in thermodynamic stability relative to that of the parent monomer as probed by equilibrium denaturation studies (Delta Delta G(u) = 3.0 +/- 0.5 kcal/mol). Despite such similarities, the alanine analogue is 50 times more active than the serine analogue. Enhanced receptor binding (Delta Delta G = 2.2 kcal/mol) is in accord with alanine's greater helical propensity and more favorable hydrophobic-transfer free energy. The success of an induced-fit model highlights the applicability of general folding principles to a complex binding process. Comparison of DKP[A6-A11](Ser) and DKP[A6-A11](Ala) supports the hypothesis that the native A1-A8 alpha-helix functions as a preformed recognition element tethered by insulin's intrachain disulfide bridge. Segmental unfolding by design provides a novel approach to dissecting structure-activity relationships.

Ligand-induced conformational change in the minimized insulin receptor

Within the class of insulin and insulin-like growth factor receptors, detailed information about the molecular recognition event at the hormone-receptor interface is limited by the absence of suitable co-crystals. We describe the use of a biologically active insulin derivative labeled with the NBD fluorophore (B29NBD-insulin) to characterize the mechanism of reversible 1:1 complex formation with a fragment of the insulin receptor ectodomain. The accompanying 40 % increase in the fluorescence quantum yield of the label provides the basis for a dynamic study of the hormone-receptor binding event. Stopped-flow fluorescence experiments show that the kinetics of complex formation are biphasic comprising a bimolecular binding event followed by a conformational change. Displacement with excess unlabeled insulin gave monophasic kinetics of dissociation. The rate data are rationalized in terms of available experiments on mutant receptors and the X-ray structure of a non-binding fragment of the receptor of the homologous insulin-like growth factor (IGF-1).

Conformational correlation and coupled motion between residue A21 and B25 side chain observed in crystal structures of insulin mutants at position A21

The C-terminal residue of the insulin A chain is invariant and kept as asparagine in all known insulin molecules from hagfish through birds to mammals. To get information on the role of this conserved residue, which is still unclear, the three-dimensional structures of four human insulin mutants, A21 Asn-->Gly, A21 Asn-->Asp, A21 Asn-->Ala, and A21 Asn-->Gln DesB30, were determined by X-ray crystallography. The four mutants crystallize separately into two kinds (rhombohedral and cubic) of crystals. In the refined structures, conformational correlation and coupled motion between the A chain C-terminal residue A21 and the B25 side chain was observed, in contrast to the nearly unchanged general structures as compared with the native insulin structures in their respective crystals. A detailed analysis suggests that residue A21 can affect insulin receptor binding by interaction with the B25 side chain and the B chain C-terminal segment to assist the B25 side chain rearranging into the 'active' conformation.