Hierarchical protein folding: asymmetric unfolding of an insulin analogue lacking the A7-B7 interchain disulfide bridge.

The landscape paradigm of protein folding can enable preferred pathways on a funnel-like energy surface. Hierarchical preferences may be manifest as a nonrandom pathway of disulfide pairing. Stepwise stabilization of structural subdomains among on-pathway intermediates is proposed to underlie the disulfide pathway of proinsulin and related molecules. Here, effects of pairwise serine substitution of insulin's exposed interchain disulfide bridge (Cys(A7)-Cys(B7)) are characterized as a model of a late intermediate. Untethering cystine A7-B7 in an engineered monomer causes significantly more marked decreases in the thermodynamic stability and extent of folding than occur on pairwise substitution of internal cystine A6-A11 [Weiss, M. A., Hua, Q. X., Jia, W., Chu, Y. C., Wang, R. Y., and Katsoyannis, P. G. (2000) Biochemistry 39, 15429-15440]. Although substantially disordered and without significant biological activity, the untethered analogue contains a molten subdomain comprising cystine A20-B19 and a native-like cluster of hydrophobic side chains. Remarkably, A and B chains make unequal contributions to this folded moiety; the B chain retains native-like supersecondary structure, whereas the A chain is largely disordered. These observations suggest that the B subdomain provides a template to guide folding of the A chain. Stepwise organization of insulin-like molecules supports a hierarchic view of protein folding.

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.

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.

Mini-proinsulin and mini-IGF-I: homologous protein sequences encoding non-homologous structures.

Protein minimization highlights essential determinants of structure and function. Minimal models of proinsulin and insulin-like growth factor I contain homologous A and B domains as single-chain analogues. Such models (designated mini-proinsulin and mini-IGF-I) have attracted wide interest due to their native foldability but complete absence of biological activity. The crystal structure of mini-proinsulin, determined as a T3R3 hexamer, is similar to that of the native insulin hexamer. Here, we describe the solution structure of a monomeric mini-proinsulin under physiologic conditions and compare this structure to that of the corresponding two-chain analogue. The two proteins each contain substitutions in the B-chain (HisB10-->Asp and ProB28-->Asp) designed to destabilize self-association by electrostatic repulsion; the proteins differ by the presence or absence of a peptide bond between LysB29 and GlyA1. The structures are essentially identical, resembling in each case the T-state crystallographic protomer. Differences are observed near the site of cross-linking: the adjoining A1-A8 alpha-helix (variable among crystal structures) is less well-ordered in mini-proinsulin than in the two-chain variant. The single-chain analogue is not completely inactive: its affinity for the insulin receptor is 1500-fold lower than that of the two-chain analogue. Moreover, at saturating concentrations mini-proinsulin retains the ability to stimulate lipogenesis in adipocytes (native biological potency). These results suggest that a change in the conformation of insulin, as tethered by the B29-A1 peptide bond, optimizes affinity but is not integral to the mechanism of transmembrane signaling. Surprisingly, the tertiary structure of mini-proinsulin differs from that of mini-IGF-I (main-chain rms deviation 4.5 A) despite strict conservation of non-polar residues in their respective hydrophobic cores (side-chain rms deviation 4.9 A). Three-dimensional profile scores suggest that the two structures each provide acceptable templates for threading of insulin-like sequences. Mini-proinsulin and mini-IGF-I thus provide examples of homologous protein sequences encoding non-homologous structures.

Mapping the functional surface of insulin by design: structure and function of a novel A-chain analogue.

Functional surfaces of a protein are often mapped by combination of X-ray crystallography and mutagenesis. Such studies of insulin have yielded paradoxical results, suggesting that the native state is inactive and reorganizes on receptor binding. Of particular interest is the N-terminal alpha-helix of the A-chain. Does this segment function as an alpha-helix or reorganize as recently proposed in a prohormone-convertase complex? To correlate structure and function, we describe a mapping strategy based on protein design. The solution structure of an engineered monomer ([AspB10, LysB28, ProB29]-human insulin) is determined at neutral pH as a template for synthesis of a novel A-chain analogue. Designed by analogy to a protein-folding intermediate, the analogue lacks the A6-A11 disulphide bridge; the cysteine residues are replaced by serine. Its solution structure is remarkable for segmental unfolding of the N-terminal A-chain alpha-helix (A1 to A8) in an otherwise native subdomain. The structure demonstrates that the overall orientation of the A and B chains is consistent with reorganization of the A-chain's N-terminal segment. Nevertheless, the analogue's low biological activity suggests that this segment, a site of clinical mutation causing diabetes mellitus, functions as a preformed recognition alpha-helix.

5-Hydroxytryptophan: an absorption and fluorescence probe which is a conservative replacement for [A14 tyrosine] in insulin.

Use of insulin's intrinsic tyrosine absorption and fluorescence to monitor its interaction with the insulin receptor is limited because the spectral properties of the receptor tryptophan residues mask the spectral properties of the hormone tyrosine residues. We describe the synthesis of an insulin analog where A14 tyrosine is replaced by a tryptophan analog, 5-hydroxytryptophan. This insulin is spectrally enhanced since 5-hydroxytryptophan has an absorption band above 300 nm which is at lower energies than the absorption of tryptophan. Steady-state and time-resolved fluorescence parameters indicate that 5-hydroxytryptophan reports the same information about the environment of the A14 side chain as does the corresponding tryptophan-containing insulin. The synthetic hormone is a full agonist compared to porcine insulin, but has slightly reduced specific activity. Consequently, this spectrally enhanced insulin analog will be useful for hormone-receptor interaction studies since it can be observed by both absorption and fluorescence even in the presence of the tryptophan-containing receptor.

High-potency hybrid compounds related to insulin and amphioxus insulin-like peptide.

We describe the synthesis and biological evaluation of five two-chain, insulin-like compounds structurally related both to insulin and to a putative insulin like peptide (ILP) whose sequence was deduced from a cDNA cloned from Branchiostoma californiensis (amphioxus), a primitive vertebrate [Chan, S. J., Cao, Q.-P., & Steiner, D. F. (1990) Proc. Natl. Acad. Sci. U.S.A. 87, 9319-9323]. The present compounds feature an A-chain corresponding to the A-domain of the putative amphioxus ILP A-domain, except that amino acid substitutions have been made at positions A2, A3, A5, and/or A8, linked via disulfide bonds to the B-chain of bovine insulin. Amphioxus ILP [2 Ile] A/insulin B, amphioxus ILP [2 Ile, 8 His] A/insulin B, amphioxus ILP [2 Ile, 5 Gln, 8 His] A/insulin B, and amphioxus ILP [2 Ile, 3 Ile, 5 Gln, 8 His] A/insulin B all display insulin-like metabolic activity and growth-promoting activity (mitogenesis) equal to or greater than that of natural insulin. Amphioxus ILP [8 His] A/insulin B shows activity in these assays greater than that of its parent compound, but not as high as compounds featuring Ile rather than Leu at position A2. In contrast, the parent compound of the present analogues, i.e., amphioxus ILP A/insulin B, displays potencies ranging from 4.0 to 9.8% relative to insulin in insulin receptor binding and lipogenesis assays, respectively. This parent compound displayed activity in growth factor assays too low for exact quantitation [Chu, Y.-C., Hu, S. Q., Zong, L., Burke, G. T., Gammeltoft, S., Chan, S. J., Steiner, D. F., & Katsoyannis, P. G. (1994) Biochemistry (in press)].(ABSTRACT TRUNCATED AT 250 WORDS)

Cross-linking of a B25 azidophenylalanine insulin derivative to the carboxyl-terminal region of the alpha-subunit of the insulin receptor. Identification of a new insulin-binding domain in the insulin receptor.

To identify a site within the insulin receptor ectodomain which forms a binding pocket for B25 Phe and is responsible for initiating conformational changes required for high affinity binding of insulin we have used a novel photoreactive insulin, despentapeptide-(B26-B30) [B25 p-azidophenylalanine-alpha-carboxamide] insulin (APC insulin). This derivative has a highly photoreactive azido group incorporated into the aromatic ring of the B25 phenylalanine amide. APC insulin bound to human insulin receptors overexpressed on a transfected Chinese hamster ovary cell line (P3-A) with an apparent potency of 9-fold relative to that of native insulin and stimulated lipogenesis in rat adipocytes with an average potency equal to porcine insulin. Addition of biotin to the B1 Phe amino group to form despentapeptide-(B26-B30) [B1 (6-biotinylamidocaproyl)phenylalanine B25 p-azidophenylalanine-alpha-carboxamide] insulin derivative (Bio-APC insulin) did not adversely affect receptor-binding affinity and provided a convenient ligand for purification of cross-linked complexes. The efficiency of receptor cross-linking with these reagents was high (70%). To identify the site(s) of cross-linking, the insulin receptor in P3-A cells was first metabolically labeled with various individual 3H-labeled amino acids and then photoaffinity labeled with 125I-Bio-APC insulin, isolated, and digested with Lys-C endoproteinase. The resulting cross-linked peptide fragments were separated by streptavidin-affinity chromatography and sequenced. The smallest identified fragment comprised residues 704-718 of the COOH terminus of the alpha-subunit of the insulin receptor. This B25 Phe cross-linked region of the alpha-subunit lies just upstream of the Exon 11-encoded 12-amino acid COOH-terminal region. Aromatic residues in this predicted alpha-helical region may form a binding pocket for B25 Phe to initiate conformational changes required for stabilizing the high affinity binding state.

Insulin-like compounds related to the amphioxus insulin-like peptide.

Three insulin-like compounds consisting of two disulfide-linked polypeptide chains have been synthesized. The A-chains of these compounds correspond either to the A- or to the A + D-domain of the putative amphioxus insulin-like peptide (amphioxus ILP), and their B-chains correspond either to the B-chain of insulin or to a slightly modified (i.e., [1-Thr]) B-domain of amphioxus ILP. The biological potency of these compounds was evaluated in mammalian cells or cell fractions containing either human or rat insulin receptors or human or mouse insulin-like growth factor I (IGF-I) receptors, with respect to binding affinity, insulin-like metabolic activity (lipogenesis), and growth factor activity (mitogenesis). Amphioxus ILP A/bovine insulin B and amphioxus ILP A + D/bovine insulin B exhibited potencies ranging from 2.0 to 9.8% relative to natural insulin, and both compounds were full agonists in lipogenesis assays, stimulating lipogenesis to the same maximal extent as seen with natural insulin. Amphioxus ILP A/amphioxus ILP [1-Thr]B stimulated lipogenesis with a potency of 0.01% relative to natural insulin. We consider this compound also likely to be a full agonist. In assays measuring binding to IGF-I receptors and stimulation of mitogenesis, these compounds displayed some activity although the activity was too low for exact quantification. These results suggest that amphioxus ILP has retained an overall structural similarity to mammalian insulin and IGF-I but has also accumulated substantial mutations which markedly reduce its ability to bind and activate their cognate receptors.(ABSTRACT TRUNCATED AT 250 WORDS)

Contribution of the B16 and B26 tyrosine residues to the biological activity of insulin.

We report the synthesis and biological evaluation of five insulin analogues in which one or both of the B-chain tyrosine residues have been substituted. [B16 Phe]insulin and [B16 Trp]insulin display a very modest reduction in potency (c. 65%) relative to porcine insulin; [B26 Phe]insulin is less active (30-50%), and the doubly substituted [B16 Phe, B26 Phe]insulin displays still lower potency (c. 35%). The further substitution of Asp for B10 His in [B16 Phe, B26 Phe]insulin raises its activity to approximately twofold greater than natural insulin, an increase of approximately fivefold over the parent compound. We conclude that the bulk and/or aromaticity of the amino acid residue at position B16, but not its hydrogen-bonding capacity, contributes to the biological activity of the hormone. We further conclude that hydrogen-bonding capacity or special side-chain packing characteristics are required at the B26 position for insulin to display high biological activity.

The effect of placement of tryptophan residues in selected A-chain positions on the biological profile of insulin.

In continuation of our efforts to study the solution structure and conformational dynamics of insulin by time-resolved fluorescence spectroscopy, we have synthesized and examined the biological activity of five insulin analogues in which selected naturally occurring residues in the A-chain have been replaced with the strongly fluorescent tryptophan residue. The potency of these analogues was evaluated in lipogenesis assays in isolated rat adipocytes, in receptor binding assays using rat liver plasma membranes, and in two cases, in receptor binding assays using adipocytes. [A3 Trp]insulin displays a potency of 3% in receptor binding assays in both liver membranes and in adipocytes, but only 0.06% in lipogenesis assays as compared to porcine insulin. [A10 Trp]insulin displays a potency of ca. 40% and ca. 25% in rat liver receptor binding and lipogenesis assays, respectively. [A13 Trp]insulin displays a potency of ca. 39% in rat liver receptor binding assays, but only ca. 9% in receptor binding in adipocytes; in lipogenesis assays, [A13 Trp]insulin displays a potency of ca. 12%, comparable to its potency in adipocyte receptor binding assays. [A15 Trp]insulin exhibits a potency of 18% and 9% in rat liver receptor binding and lipogenesis assays, respectively. The doubly substituted analogue, [A14 Trp, A19 Trp] insulin, displays a potency of ca. 0.7% in both rat liver receptor binding assays and lipogenesis assays.(ABSTRACT TRUNCATED AT 250 WORDS)

Steric requirements at position B12 for high biological activity in insulin.

The alpha-helix formed by the amino acid residues 9-19 of the B-chain of insulin is involved in the stabilization of its three-dimensional structure. We have shown that modification at positions B9, B10, B12, and B16 results in analogues possessing biological activities ranging from ca. 0.2% to ca. 500% relative to that of natural insulin. The lowest potency was displayed by [B12 Asn]insulin, in which the hydrophobic B12 Val residue was replaced by the hydrophilic Asn residue. We now report the synthesis of four insulin analogues in which hydrophobicity is retained, and only the spatial arrangement of atoms in the B12 region is altered. Substitution of B12 Val with alpha-aminoisobutyric acid (Aib), D-Ala, and Phe led to analogues possessing biological activities, in lipogenesis assays, of 8.5%, 2%, and 0.2%, respectively, relative to that of natural insulin. Inversion of the B11-B12 sequence, -Leu-Val-, led to an analogue displaying 3.3% activity. A synthetic B-chain in which the B11 Leu-B12 Val sequence was replaced by B11 Ala-B12 Ile was incapable of combining with the natural A-chain. We conclude that the Val residue in the B12 position in insulin fulfills special side-chain packing requirements involved in the stability of the structure of insulin. Even slight steric alteration at position B12 results in a distortion of the overall conformation of the B-chain which affects its ability to combine with the natural A-chain. This distortion is retained in the corresponding analogue, which is reflected in diminished biological potency.

The A14 position of insulin tolerates considerable structural alterations with modest effects on the biological behavior of the hormone.

As part of our aim to investigate the contribution of the tyrosine residue found in the 14 position of the A-chain to the biological activity of insulin, we have synthesized six insulin analogues in which the A14 Tyr has been substituted by a variety of amino acid residues. We have selected three hydrophilic and charged residues--glutamic acid, histidine, and lysine--as well as three hydrophobic residues--cycloleucine, cyclohexylalanine, and naphthyl-(1)-alanine--to replace the A14 Tyr. All six analogues exhibit full agonist activity, reaching the same maximum stimulation of lipogenesis as is achieved with porcine insulin. The potency for five of the six analogues, [A14 Glu]-, [A14 His]-, [A14 Lys]-, [A14 cycloleucine]-, and [A14 naphthyl-(1)-alanine]-insulins in receptor binding assays ranges from 40-71% and in stimulation of lipogenesis ranges from 35-120% relative to porcine insulin. In contrast, the potency of the sixth analogue, [A14 cyclohexylalanine]insulin, in both types of assays is less than 1% of the natural hormone. The retention time on reversed-phase high-performance liquid chromatography for the first five analogues is similar to that of bovine insulin, whereas for the sixth analogue, [A14 cyclohexylalanine]insulin, it is approximately 11 min longer than that of the natural hormone. This suggests a profound change in conformation of the latter analogue. Apparently, the A14 position of insulin can tolerate a wide latitude of structural alterations without substantial decrease in potency. This suggests that the A14 position does not participate directly in insulin receptor interaction. Only when a substitution which has the potential to disrupt the conformation of the molecule is made at this position, is the affinity for the receptor, and hence the biological potency, greatly reduced.

Insulin analogues with modifications in the beta-turn of the B-chain.

The beta-turn formed by the amino acid residues 20-23 of the B-chain of insulin has been implicated as an important structural feature of the molecule. In other biologically active peptides, stabilization of beta-turns has resulted in increases in activity. We have synthesized three insulin analogues containing modifications which would be expected to increase the stability of the beta-turn. In two analogues, we have substituted alpha-aminoisobutyric acid (Aib) for the Glu residue normally present in position B21 or for the Arg residue normally present in position B22; in a third compound, we have replaced the Glu residue with its D-isomer. Biological evaluation of these compounds showed that [B21 Aib]insulin displays a potency ca. one-fourth that of natural insulin, while [B22 Aib]insulin is less than 10% as potent. In contrast, [B21 D-Glu]insulin is equipotent with natural insulin. We conclude that the beta-turn region of the insulin molecule normally possesses considerable flexibility, which may be necessary for it to assume a conformation commensurate with high biological activity. If this is the case, [B21 D-Glu]insulin may exhibit a stabilized geometry similar to that of natural insulin when bound to the insulin receptor.

Superactive insulins.

The substitution of aspartic acid for the naturally-occurring histidine residue in position B10 in human insulin results in an insulin analogue which displays an in vitro potency 4- to 5-fold greater than the parent compound. This substitution has been introduced into six insulin analogues which, before modification, display potencies ranging from less than 0.01-fold to 3-fold relative to natural insulin. In each case, the resulting aspartic acid-substituted analogue is substantially more potent than the parent compound. Thus, it is now possible to prepare "tailor-made" insulins with enhanced potency.

An insulin-like hybrid consisting of a modified A-domain of human insulin-like growth factor I and the B-chain of insulin.

We have synthesized an insulin-like compound, consisting of the B-chain of bovine insulin and an A-chain corresponding to the A-domain of human insulin-like growth factor-I (IGF-I), in which the isoleucine residue normally present in position 2 of the A-domain of IGF-I has been replaced with glycine. Biological evaluation of the compound indicated that its insulin-like activity (insulin receptor-binding and stimulation of lipogenesis) was 0.2%, and its growth-factor activity (stimulation of thymidine incorporation) was less than 1%, both relative to natural insulin. We conclude that interactions between IleA2 and TyrA19, which are crucial to high biological activity in insulin, are also present in IGF-I, and are equally critical for its biological activity.

An insulin-like compound consisting of the B-chain of bovine insulin and an A-chain corresponding to a modified A- and the D-domains of human insulin-like growth factor I.

We report the synthesis and biological evaluation of a two-chain, disulfide-linked, insulin-like compound consisting of the B-chain of bovine insulin and an A-chain corresponding to the A- and D- domains of human insulin-like growth factor-I (IGF-I) in which the A-domain amino-acid residues -Phe49-Arg50-Ser51-found in IGF-I have been replaced by -Ala-Gly-Val-, the homologous region of sheep insulin. The compound is indistinguishable from a previously reported compound whose A-chain corresponds to the A- and D-domains of IGF-I without the substitution, in assays for insulin-like activity as well as in assays for growth-promoting activity. We conclude that these A-domain residues do not contribute significantly to the interaction of IGF-I with either insulin or IGF-I receptors.

A biologically active insulin analogue with modification in the A2 position. [2-Valine-A] sheep insulin.

The synthesis and biological evaluation of [2-Valine-A] insulin ([Val2-A]insulin) is reported. In this insulin, the isoleucine residue in position A2, invariant in the majority of mammalian insulins, is substituted by valine. The same substitution, along with four others, occurs naturally in the insulin produced by the owl monkey. Owl monkey insulin exhibits ca. 20% of the activity of porcine insulin in in vitro insulin assays using human cells in culture. [Val2-A]insulin displays 20-22% of the activity of bovine insulin in in vitro insulin assays using rat liver plasma membranes or isolated rat adipocytes. We suggest that the substitution of valine for isoleucine at position A2 is responsible for all or most of the diminution in potency of owl monkey insulin relative to porcine insulin. The data are discussed with regard to previous findings with insulin analogues in which isoleucine A2 was replaced with norleucine, glycine and alanine.

A highly potent insulin: des-(B26-B30)-[AspB10,TyrB25-NH2]insulin(human).

An insulin analogue that embodies two distinct structural modifications, each of which independently increases insulin activity, has been synthesized and evaluated for biological activity. The analogue, des-(B26-B30)-[AspB10,TyrB25-NH2]insulin is the most potent insulin analogue yet described; it displays an 11- to 13-fold higher activity than natural insulin. The findings are discussed with regard to the receptor-binding domains of insulin.