Preparation and activity of nitrated insulin dimer

Nitration of insulin using tetranitromethane causes polymerisation involving cross-linked tyrosyl residues. By performing this reaction with insulin crystals, in which it is known that B16 tyrosine of one monomer is closely associated with B26 of the neighbouring monomer within the dimer, it has been possible to isolate a covalent dimer of insulin cross-linked between these two tyrosines. It was, however, first necessary to block the reactive A14 tyrosine. Both rhombohedral (hexameric) and cubic (dimeric) pig insulin crystals were used, the latter proving successful in yielding a pure dimeric product as shown by oxidative sulphitolysis and HPLC. The purified nitrated dimer was biologically active (ca. 10% potency compared to monomeric insulin in a lipogenesis assay) suggesting that the residues responsible for insulin's action are present on the surface of the dimer and not buried in the interface.     

Ionization of tyrosine residues in horse-heart ferricytochrome c and its guanidinated and acetylated-guanidinated derivatives

Spectrophotometric titration curves were obtained at 242 nm for native and fully guanidinated horse-heart ferricytochrome c. The cytochrome c data were fit over the pH range 9-12 (I = 0.35) by a theoretical curve with pK' values of 10.35 and 11.70. The slope of the experimental data increases sharply above pH 12.5 suggesting that two tyrosine residues with pK' values greater than 12.5 are exposed by conformation change. The guanidinated cytochrome c data after correction for the alkaline spin-state transition were fit over the entire pH range 9-13.6 (I = 0.35) by a theoretical curve with pK' values 10.37, 10.78, 11.50, and 13.60. These results along with viscosity measurements indicate that the unfolding transition occurs at higher pH in the guanidinated derivative. N-Acetylimidazole was used to acetylate specific tyrosyl groups of guanidinated cytochrome c. Assignments of acetylated tyrosine residues were confirmed by peptide mapping of 14C-labelled derivatives. Spectrophotometric titrations with rapid data acquisition of two monoacetylated derivatives allowed assignments of pK'1 (10.37) to Tyr-67 and pK'4 (13.60) to Tyr-97. The basis for the large differences in acidity and chemical reactivity of these two residues is not obvious from the crystallographic structure and may arise from differences in solvent access due to motions of the polypeptide chain.     

Ionization behavior of native and mutant insulins: pK perturbation of B13-Glu in aggregated species

Upscale titration from pH 2.5 to 11.2 is used as a means for probing solvent accessibility of ionizing groups in zinc-free preparations of native and mutant insulins. Stoichiometry and pK alpha values of ionizing groups in the titration curves are determined by iterative curve fitting. Under denaturing conditions, the titration curve of human insulin is in good agreement with that predicted from the sum of unperturbed titrations of the constituent ionizing groups and yields an apparent isoionic point of 5.3. Under nondenaturing conditions where aggregation and precipitation occur, titrations show that only five out of six carboxylate residues of human insulin ionize in the expected region. Consequently, one carboxylate ionization is masked and the apparent isoionic point located at pH 6.4. Correlation between ionization behavior and patterns of aggregation and solubility is established by titrations of mutant insulins and of dilute native insulin. Titration of an unusually soluble species, B25-Phe----His, shows that precipitation is not responsible for the masked carboxylate ionization of native insulin. Titrations of mutants B13-Glu----Gln and B9-Ser----Asp show that the masked ionization probably originates from monomer-monomer interactions in the insulin dimer. We conclude that the B13-Glu side chain is responsible for the masked carboxylate ionization in aggregated forms of human insulin.     

The effect of iodination of particular tyrosine residues on the hormonal activity of insulin

Insulin dissolved in aqueous or methanolic buffer was iodinated to give preparations containing an average of between one and five iodine atoms per insulin monomer. The resultant preparations were fragmented in various ways and the ratio of tyrosine to monoiodotyrosine and di-iodotyrosine was determined in each fragment. This has allowed the distribution of iodine between the combined A-chain tyrosine residues and the individual B-chain tyrosine residues to be determined. The hormonal activity of each of these iodinated insulin preparations was measured from their effect on the production of (14)CO(2) from [1-(14)C]glucose by isolated adipose cells. The results were interpreted as meaning that the iodination of tyrosine residue A19 or B16 leads to the inactivation of insulin. Speculations are made about the nature of an interaction between insulin and a receptor site on the target tissue.     

1H n.m.r. studies of insulin. Reversible transformation of 2-zinc to 4-zinc insulin hexamer

1H n.m.r. studies at 270 MHz were made of the transformation of 2 Zn insulin hexamer to 4 Zn hexamer produced by the addition of anions (thiocyanate ion). Four separate H2 histidine resonances were observed for the B5 and B10 histidines in 2 Zn hexamer at pH 7 and 9 and four separate resonances also occurred in the 4 Zn hexamer. The observation of these resonances and others from phenylalanine, tyrosine and leucine residues showed that the 2 Zn to 4 Zn transformation probably occurred in solution in a similar manner to that observed in the crystal. Furthermore as occurred in the crystal, it was found that in solution the transformation was reversible (on removal of thiocyanate) and that 2 Cd insulin was unable to undergo the transformation. Des-Phe-Bl-insulin did not undergo the transformation. Addition of SCN- to Zn-free insulin (mainly dimer) produced only a small transformation, consistent with the idea that Zn2+ promotes formation of hexamer from dimer but probably does not otherwise affect the transformation.     

High resolution 1H-NMR studies of Des-(B26-B30)-insulin; assignment of resonances and properties of aromatic residues

The assignments of 1H resonances of the eight aromatic residues of Des-(B26-B30)-insulin are reported, based on pH titration, selective spin decoupling and its 500 MHz 1H two-dimensional (2D)-COSY spectrum. The pK values of the three tyrosines A14, A19 and B16 are 10.84, 11.27 and 10.40, respectively. Tyrosine A19 is buried in a hydrophobic environment, while Tyrosine B16 is exposed in a relatively hydrophilic state. Among the three phenylalanines, the ring proton resonances of Phe-B25 undergo abnormal upfield shifts, probably due to the ring currents of the nearby Phe-B24 and Tyr-B16. From this study of the low-field region of 1H-NMR spectrum of Des-(B26-B30)-insulin, we conclude that this molecule probably maintains the major structural features of insulin in aqueous solution, but there are some readjustments of the peptide conformation.     

Four new monomeric insulins obtained by alanine scanning the dimer-forming surface of the insulin molecule

The residues A21Asn, B12Val, B16Tyr, B24Phe, B25Phe, B26Tyr and B27Thr, buried in the dimer of insulin, were identified by means of alanine-scanning mutagenesis. The receptor binding activity, in vivo biological potency and self-association properties of the seven single alanine human insulin mutants were determined. Four of the seven single alanine mutants, [B12Ala]human insulin, [B16Ala]human insulin, [B24Ala]human insulin and [B26Ala]human insulin, are monomeric insulin, which indicates that B12Val, B16Tyr, B24Phe and B26Tyr are crucial for the formation of insulin dimer. The monomeric [B16Ala]human insulin and [B26Ala]human insulin retain 27 and 54% receptor binding activity, respectively, and nearly the same in vivo biological potency compared with native insulin, so they could be developed as the fast-acting insulin.     

Proton nuclear magnetic resonance study of the B9(Asp) mutant of human insulin. Sequential assignment and secondary structure

The sequence-specific 1H nuclear magnetic resonance (n.m.r.) assignment of 49 of the 51 amino acid residues of human B9(Asp) insulin in water at low pH is reported. Spin systems were identified using a series of two-dimensional n.m.r. techniques. For the majority of the amino acid residues with unique spin systems, particularly Ala, Thr, Val, Leu, Ile and Lys, the complete spin systems were identified. Sequence-specific assignments were obtained from sequential nuclear Overhauser enhancement (NOE) connectivities. The results indicate that the solution structure of the mutant closely resembles the crystal structure of native insulin. Thus, the NOE data reveal three helical domains all consistent with the secondary structure of the native human 2Zn insulin in the crystal phase. Numerous slowly exchanging amide protons support these structural elements, and indicate a relatively stable structure of the protein. A corresponding resemblance of the tertiary structures in the two phases is also suggested by slowly exchanging amide protons, and by the extreme chemical shift values observed for the beta-protons of B15(Leu) that agree with a close contact between this residue and the aromatic rings of B24(Phe) and B26(Tyr), as found in the crystal structure of the 2Zn insulin. Finally, there are clear indications that the B9(Asp) insulin mutant exists primarily as a dimer under the given conditions.     

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.     

Differential binding of monoiodinated insulins to muscle and liver derived receptors and activation of the receptor kinase

The binding affinity of monoiodoinsulin analogues to receptors purified from rat skeletal muscle and liver were compared. Insulin iodinated at tyrosine B26 bound to both muscle and liver derived insulin receptors with higher affinity than the A14-iodoisomer or native insulin. The affinity of the B26-iodoanalogue was greater for muscle than for liver derived receptors; by Scatchard analysis the affinity ratio B26/A14 was 2.8 for muscle and 1.3 for liver. The affinity of muscle and liver derived receptors for A14-iodoinsulin was not different. Dose response curves of autophosphorylation and exogenous tyrosine kinase activation showed significantly increased sensitivity to the B26-iodoanalogue (compared to the A14-iodoisomer or native insulin) in muscle derived receptors, but not in liver. The difference in affinity between muscle and liver derived insulin receptors towards B26-monoiodotyrosyl-insulin likely reflects the observed structural difference between the insulin receptor alpha-subunits from muscle and liver.     

High resolution1H-NMR studies of Des-(B26–B30)-insulin; assignment of resonances and properties of aromatic residues

The assignments of¹H resonances of the eight aromatic residues of Des-(B26–B30)-insulin are reported, based on pH titration, selective spin decoupling and its 500 MHz¹H two-dimensional (2D)-COSY spectrum. The pK values of the three tyrosines A14, A19 and B16 are 10.84, 11.27 and 10.40, respectively. Tyrosine A19 is buried in a hydrophobic environment, while Tyrosine B16 is exposed in a relatively hydrophilic state. Among the three phenylalanines, the ring proton resonances of Phe-B25 undergo abnormal upfield shifts, probably due to the ring currents of the nearby Phe-B24 and Tyr-B16. From this study of the low-field region of¹H-NMR spectrum of Des-(B26–B30)-insulin, we conclude that this molecule probably maintains the major structural features of insulin in aqueous solution, but there are some readjustments of the peptide conformation.     

Correlation proton magnetic resonance studies at 250 MHz of bovine pancreatic ribonuclease. I. Reinvestigation of the histidine peak assignments.

The deuterium exchange kinetics of the C(2) protons of the four histidine residues of native bovine pancreatic ribonuclease A have been followed at pH 6.5 and 8.0 by proton magnetic resonance spectroscopy (1H NMR). Comparison of the order of exchange of the histidine peaks with tritium exchange rates into individual histidine residues [Ohe, M., Matsuo, H., Sakiyama, F., and Narita, K. (1974), J. Biochem. (Tokyo) 75, 1197] supports the previous assignment of histidine NMR peaks H(1) and H(4) to histidine-105 and histidine-48 but requires reassignment of peaks H(2) and H(3) to histidine-119 and histidine-12, respectively. Ribonuclease A samples having differentially deuterated histidines have been used to verify the existence of crossover points in the histidine proton magnetic resonance titration curves and to observe the discontinuous titration curve of histidine-48. Proton magnetic resonance peaks have been assigned to the C(4) protons of the four histidine residues of ribonuclease A on the basis of their unit proton areas and by matching their titration shifts with the more readily visible C(2)-H peaks of the histidines. The pK' values derived from the C(4)-H data agree, within experimental limits, with those derived from C(2)-H data. The C(4)-H peaks were assigned to histidine-12, -48, -105, and -119 of ribonuclease A on the basis of their pH dependence, pK' values, shifts of their pK' values in the presence of inhibitor cytidine 3'-phosphate, and by comparison with the assignments of the histidine C(2)-H peaks above.     

Zymogen activation in serine proteinases. Proton magnetic resonance pH titration studies of the two histidines of bovine chymotrypsinogen A and chymotrypsin Aalpha.

Reversible unfolding of bovine chymotrypsinogen A in 2H2O either by heating at low pH or by exposure to 6 M guanidinium chloride results in the exchange of virtually all the nitrogen-bound hydrogens that give rise to low-field 1H NMR peaks, without significant exchange of the histidyl ring Cepsilon1 hydrogens. These preexchange procedures have enabled the resolution of two peaks, using 250-MHz correlation 1H NMR spectroscopy, that are attributed to the two histidyl residues of chymotrypsinogen A. Assignments of the Cepsilon1 hydrogen peaks to histidine-40 and -57 were based on comparison of the NMR titration curves of the native zymogen with those of the diisopropylphosphoryl derivative. Two histidyl Cepsilon1 H peaks were also resolved with solutions of preexchanged chymotrypsin Aalpha. The histidyl peaks of chymotrypsin Aalpha were assigned by comparison of NMR titration curves of the free enzyme with those of its complex with bovine pancreatic trypsin inhibitor (Kunitz). The NMR titration curves of histidine-57 in the zymogen and enzyme and histidine-40 in the zymogen exhibit two inflections; the additional inflections were assigned to interactions with neighboring carboxyl groups: aspartate-102 in the case of histidine-57 and aspartate-194 in the case of histidine-40 of the zymogen. In bovine chymotrypsinogen A in 2H2O at 31 degrees C, histidine-57 has a pK' of 7.3 and aspartate-102 a pK' of 1.4, and the histidine-40-aspartate-194 system exhibits inflections at pH 4.6 and 2.3. In bovine chymotrypsin Aalpha under the same conditions, the histidine-57-aspartate-102 system has pK' values of 6.1 and 2.8, and histidine-40 has a pK' of 7.2. The results suggest that the pK' of histidine-57 is higher than the pK' of aspartate-102 in both zymogen and enzyme. A significant difference exists in the structure and properties of the catalytic center between the zymogen and activated enzyme. In addition to the difference in pK' values, the chemical shift of histidine-57, which is highly abnormal in the zymogen (deshielded by 0.6 ppm), becomes normalized upon activation. These changes may explain part of the increase in the catalytic activity upon activation. The 1H NMR chemical shift of the Cepsilon1 H of histidine-57 in the chymotrypsin Aalpha-pancreatic trypsin inhibitor (Kunitz) complex is constant between pH 3 and 9 at a value similar to that of histidine-57 in the porcine trypsin-pancreatic trypsin inhibitor complex [Markley, J.L., and Porubcan, M. A. (1976), J. Mol. Biol. 102, 487--509], suggesting that the mechanisms of interaction are similar in the two complexes.     

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.     

Degradation of the four monoiodoinsulin isomers by the insulin protease of rat liver

The cleavage of insulin by the partially purified insulin protease was studied using the four [¹²⁵I]tyrosine-monoiodoinsulins (tyrosine A-14 and A-19 of the A-chain; tyrosine B-16 and B-26 of the B-chain). The rates of conversion of the four isomers to trichloroacetic acid-soluble form was in the order B-26 > A-14 > A-19 > B-16. The following was observed in experiments which gave 19/14/5/3 percent conversion to trichloroacetic acid-soluble products: the loss of ability to bind to IM-9 lymphocytes was approx. 55% for all four isomers. About 70% of the radioactivity was in the ‘insulin’ peak, and about 30% was in peptides smaller than insulin as judged by gel filtration on Sephadex G-50. The descending limb of the ‘insulin’ peak contained significant amounts of radioactive material not binding to IM-9 lymphocytes. This material showed multiple peaks when applied to high performance liquid chromatography. Other experiments were designed to cause an almost complete degradation of the isomers. Under these conditions, the radioactivity eluted on Sephadex G-50 largely as iodotyrosine (and some small peptides) using the A-14, B-16 and B-26 isomers, whereas iodotyrosine was absent using the A-19 isomer. Thus, the insulin protease appears to first degrade insulin to multiple products with molecular sizes slightly smaller than insulin and subsequently to small peptides (e.g. containing tyrosine A-19) and amino acids (e.g. tyrosine A-14, B-16 and B-26).     

Identification of the nitration site of insulin by peroxynitrite.

Our previous investigation indicated that insulin can be nitrated by peroxynitrite in vitro. In this study, the preferential nitration site of the four tyrosine residues in insulin molecule was confirmed. Mononitrated and dinitrated insulins were purified by RP-HPLC. Following reduction of insulin disulfide bridges, Native-PAGE indicated that A-chain was preferentially nitrated. Combination of enzymatic digestion of mononitrated insulin with endoproteinase Glu-C, mass spectrometry confirmed that Tyr-A14 was the preferential nitration site when insulin was treated with peroxynitrite. Tyr-A19, maybe, was the next preferential nitration site. According to the crystal structure, Tyr-B26 between the two tyrosine residues in insulin B-chain was likely easier to be nitrated by peroxynitrite.     

A novel view of pH titration in biomolecules

When individual titratable sites in a molecule interact with each other, their pH titration can be considerably more complex than that of an independent site described by the classical Henderson-Hasselbalch equation. We propose a novel framework that decomposes any complex titration behavior into simple standard components. The approach maps the set of N interacting sites in the molecule onto a set of N independent, noninteracting quasi-sites, each characterized by a pK'(a) value. The titration curve of an individual site in the molecule is a weighted sum of Henderson-Hasselbalch curves corresponding to the quasi-sites. The total protonation curve is the unweighted sum of these Henderson-Hasselbalch curves. We show that pK'(a) values correspond to deprotonation constants available from methods that can be used to assess total proton uptake or release, and establish their connection to protonation curves of individual residues obtained by NMR or infrared spectroscopy. The new framework is tested on a small molecule diethylenetriaminepentaacetate (DTPA) exhibiting nonmonotonic titration curves, where it gives an excellent fit to experimental data. We demonstrate that the titration curve of a site in a group of interacting sites can be accurately reconstructed, if titration curves of the other sites are known. The application of the new framework to the protein rubredoxin demonstrates its usefulness in calculating and interpreting complicated titration curves.     

Photochemically induced nuclear polarization study of the accessibility of tyrosines in insulin

The accessible tyrosines of bovine insulin were studied by the photochemically induced dynamic nuclear polarization (photo-CIDNP) method. Tyrosine 1H nuclear polarization is observed in acidic, neutral, and basic solutions at all concentrations studied, in the absence of added salts as well as in the presence of 0.05-0.1 M chloride or phosphate. At pH 2.1 in the presence of chloride, at concentrations of 640 microM and above, most of the nuclear polarization at delta 6.82 originates from one group of tyrosines. On the basis of the crystallographic model, these are assumed to be the A14 tyrosines. We explored the possibility of a genuine concentration dependence of the photo-CIDNP intensity of insulin due to aggregation. In order to discern between such effects and trivial kinetic effects traceable to the optical irradiation method, the effects of concentration changes on polarization were examined in three apparently nonassociating trypsin inhibitor proteins. In insulin, the intensity of Tyr-A 14 polarization changes slowly at concentrations above 1 mM, suggesting that these residues are similarly accessible in all association states. At insulin concentrations below 320 microM, additional tyrosine emission signals were observed. These signals are probably due to B16 and B26 tyrosines of monomers. Polarization transfer effects from Tyr-A14 are evident in the tetramer and hexamer. Enhanced absorption effects in the two histidines (B5 and B10) of the insulin monomer were observed at pH 10 in the presence of 0.1 M phosphate.     

Design and synthesis of a novel biotinylated photoreactive insulin for receptor analysis.

B1-(4-Azido-salicyloyl)-[B1-biocytin,B2-lysine]insulin was synthesized by double Edman degradation of A1,B29-Msc2-insulin and stepwise acylation at the N-terminus of the B-chain. This derivative is homogeneous in RP-HPLC and has a biological in vitro activity of 20% and receptor binding of 15%, relative to insulin. Radioiodination and HPLC gave the B1-labelled 125I-derivative (I) as well as the 4 isomers with 125I-labelled tyrosine (A14, A19, B16, B26). UV-induced crosslinking of I with insulin receptors led to specific labelling of the alpha-subunit (Mr 130,000). The peptide bond LysB2-AspB3 is completely cleavable by trypsin (EC 3.4.21.4). I is thus a new tool for the analysis of the hormone-binding region by making possible the isolation of tryptic, biotinylated receptor fragments labelled by the dipeptide 125I-4-azidosalicyloyl-biocytinyl-Lys.