Formation of the collagen-like triple-helical structure in oligopeptides during elongation of the molecular chain

Oligotripeptides Z-(Gly-Pro-Pro)_n-OMe (n = 1,2,…,8), Z-Gly-Pro_Pro-Gly-Pro-Gly-OMe, Z-Gly-Pro-Pro-Gly-Pro-Gly-Gly-Pro-Pro-OMe, Z-Gly-Pro-Pro-(Gly-Pro-Gly)₂-Gly-Pro-Pro-OMe, and Z-(Gly-Pro-Ala)_n-OMe (n = 1,2,…,4) were synthesized step-by-step and then studied by means of x-ray diffraction, ir spectroscopy, the kinetics of hydrogen-deuterium exchange of peptide groups, and circular dichroism,. Different stages in the formation of a triple helix in Z-(Gly-Pro-Pro)_n-OMe were revealed during the chain elongation. In the solid state, at the first stage a conformation of the polyproline II-type is formed in the tripeptide and in the second stage a triple helical complex appears in the hexapeptide. Interpeptide hydrogen bonds in this complex are still of low order. At further stages an ordered set of interpeptide hydrogen bonds is gradually formed. It is shown that the degree of order of interpeptide H bonds depends on the length of the molecular chain, the amino acid composition, and residue sequence in the triplets.     

Bound water in the collagen-like triple-helical structure.

The ir amide bands of the triple-helical polytripeptides and collagens upon hydration of films are investigated. On the basis of our assignment of the amide I components, the formation of hydrogen bonds between the peptide backbone and structural water is studied. The C1O1--HOH hydrogen bonds are found more ordered than the C3O3--HOH hydrogen bonds. The specific incorporation of water in the triple helix is followed by multistep conformational changes and by increasing of the interpeptide hydrogen-bond strength. The formation of the polypeptide hydrate structure depending on the amino acid composition and the chain length is examined.     

A triple-helical model for (Gly-Pro-Hyp)n with cis peptide units

Synthetic regular polytripeptides of the type (Gly-R₂-R₃) where R₂, R₃, or both, are imino acids have been widely studied as model compounds for collagen. One such polytripeptide is poly(Gly-Pro-Hyp), since triplets with this sequence constitute about 10% of collagen. Recently, a new model has been proposed for this polytripeptide in which one of the three peptide bonds in the tripeptide unit is in the cis conformation, and the γ-hydroxyl group of hydroxyproline forms a direct interchain hydrogen bond within the triple helix. We have confirmed this structure by model building using computer techniques, and the helical parameters obtained by us are close to the experimentally observed values. The model is also found to be comparable in stability with other models from energy considerations.     

Amide I band of IR spectrum and structure of collagen and related polypeptides

The infrared amide I band of collagens (rat and cod skin) and related compounds (polyproline, polyglycine, and polytripeptides) was studied. Assignment of amide I-band components for polyproline II and polytripeptides (Gly-Pro-Pro)_n and (Gly-Pro-Gly)_n in the solid state and water solution was made. Three amide I components observed in the polypeptide spectra were attributed to three different peptide CO groups in each triplet. On the basis of this assignment, the interpretation of the amide I multicomponent structure in collagen and isomorphous oligo- and polypeptides was attempted. The ordering of intra- and intermolecular hydrogen bonds involving peptide CO groups in collagen and related compounds was discussed.     

Unusual conformational transitions of leucine-containing basic polytripeptides

The conformational transitions of synthetic basic polytripeptides (Lys-Leu-Gly)_n, (A₂bu-Leu-Gly)_n, (Lys-Leu-Ala)_n, and (A₂bu-Leu-Ala)_n induced by high salt concentrations and elevated pH were investigated by CD, ir, and ¹H-nmr spectroscopy, sedimentation analysis, viscometry, and light scattering. Sheet aggregates of chains in a conformation similar to the polyglycine II (polyproline II) helix, bound together by hydrogen bonds, are the most probable form of (Lys-Leu-Gly)_n and also, partly, of (A₂bu-Leu-Gly)_n in a high-pH or high-salt solutions. The conformation (Lys-Leu-Ala)_n, in a low-salt concentration, is an α-helix. Since (A₂bu-Leu-Ala)_n is disordered under similar conditions, it appears that this α-helix is stabilized by hydrophobic interactions between Lys and Leu side chains. In a high concentration of water structure-making ions, CD data for (Lys-Leu-Ala)_n indicate distortion of the α-helix, with a parallel increase in the average molecular weight corresponding to trimer formation. Hydrodynamic data are consistent with a model of bundles of three closely touching spherocylinders. (A₂bu-Leu-Ala)_n shows a limited tendency to α-helix formation.     

An energetic evaluation of a "Smith" collagen microfibril model.

An energy minimized three-dimensional structure of a collagen microfibril template was constructed based on the five-stranded model of Smith (1968), using molecular modeling methods and Kollman force fields (Weiner and Kollman, 1981). For this model, individual molecules were constructed with three identical polypeptide chains ((Gly-Pro-Pro) _n , (Gly-Prop-Hyp) _n , or (Gly-Ala-Ala) _n , wheren=4, 12, and 16) coiled into a right-handed triple-helical structure. The axial distance between adjacent amino acid residues is about 0.29 nm per polypeptide chain, and the pitch of each chain is approximately 3.3 residues. The microfibril model consists of five parallel triple helices packed so that a left-handed superhelical twist exists. The structural characteristics of the computed microfibril are consistent with those obtained for collagen by X-ray diffraction and electron microscopy. The energy minimized Smith microfibril model for (Gly-Pro-Pro)₁₂ has an axial length of about 10.2 nm (for a 36 amino acid residue chain), which gives an estimated D-spacing (234 amino acids per chain) of approximately 66.2 nm. Studies of the microfibril models (Gly-Pro-Pro)₁₂, (Gly-Pro-Hyp)₁₂, and (Gly-Ala-Ala)₁₂ show that nonbonded van der Waals interactions are important for microfibril formation, while electrostatic interactions contribute to the stability of the microfibril structure and determine the specificity by which collagen molecules pack within the microfibril.     

Helix-coil transition of Poly(L-glutamic acid) in N-methylacetamide

The folding of randomly coiled poly(L -glutamic acid) to the helical state has been studied in N-methylacetamide by titration methods. Since this solvent would be expected to form amide-peptide group hydrogen bonds with the unfolded form of the polymer, to a first approximation no helix stabilization could come from intrapolymer hydrogen bonds. The titration data, collected from 30 to 70°C yield the following values per residue for the thermodynamic parameters governing the coil-helix reaction for the uncharged polymer: ΔG_30°C°, ?1. 9 ± 0.1 kcal; Δ H°, 0 ± 0.1 kcal; ΔS_30°C°, 6.3 ± 0.6 eu. In N-methyl acetamide, the helix is an order of magnitude more stable than in water, and this stabilization appears to be entirely the result of the entropy gained by solvent molecules which are released from the polymer upon folding.     

Electronic structures of collagen model polymers: (Gly-Pro)n, (Gly-Hyp)n, (Ala-Pro)n, (Ala-Hyp)n, (Gly-Pro-Gly)n, (Gly-Hyp-Gly)n, (Gly-Pro-Pro)n, and (Gly-Pro-Hyp)n

In order to investigate the relationship existing between the electronic structures of collagen and its biochemical functions in vivo, the semiempirical CNDO/2 SCF MO calculations were carried out on several model polymers of collagen, (Gly-Pro)_n, (Gly-Hyp)_n, (Ala-Pro)_n, (Ala-Hyp)_n, (Gly-Pro-Gly)_n, (Gly-Hyp-Gly)_n, (Gly-Pro-Pro)_n and (Gly-Pro-Hyp)_n. Geometries of the skeleton of these polymers were assumed to be the same as those of poly(l-proline) I (cis) and II (trans) and the calculations were performed only on infinite polymers in a single chain. The results show that the cis form is always more stable than the trans form for all the polymers treated. This energy difference between the cis and trans forms depends, for example, on the kind of amino acid residue, Gly or Ala, but this could not be seen in the Pro or Hyp residue. The flexibility or mobility of the collagen structure was explained using the energy difference between the cis and trans forms of the polymers, i.e. the cis-trans conversion of the collagen was discussed in connection with the energy difference. The reason why the collagen has the constitution of (Gly-Pro-Hyp)_n is briefly discussed.     

Two types of tripeptide conformation in collagen. Calculation of the structure of (Gly-Pro-Ser)n and (Gly-Val-Hyp)n polytripeptides

Conformational analysis of polypeptides (Gly-Pro-Ser)n and (Gly-Val-Hyp)n was carried out for collagen-like triple helical complexes (coiled coils with screw symmetry). The lowest energy structure of the first polymer (helical parameters t 52,8, h 0,282 nm) is very close to that of (Gly-Pro-Hyp)n. The hydroxyl group of a serine residue does not form any intramolecular hydrogen bonds in this structure. (Gly-Val-Hyp)n triple complex is shown to unwind to t 7,7, h 0,297 nm as a result of optimization procedure. These findings confirm the assumption, made earlier on the basis of conformational analysis of (Gly-Pro-Hyp)n, (Gly-Pro-Ala)n, (Gly-Ala-Hyp)n, (Gly-Ala-Ala)n, that the collagen triple helix contains stable wound triplets with proline in the second position, while the absence of imino acid in the 2nd position facilitates the unwinding of the triple helix. Thus, a collagen helix appears to have different parameters for the sites differing in the amino acid sequence. The values measured in the X-ray experiments (h 0,29 nm, t' 36) should be considered as a result of averaging. The model allows to reconcile the X-ray data for collagen and crystalline (Gly-Pro-Pro)10 oligomer.     

Conformational change of the triple-helical structure. III. Stabilizing forces in the triple helix

The analysis of thermal melting curves of (PPG)_n (n = 10, 12, 14, and 15) and (PPG)_n(APG)_m (PPG)_n (2n + m = 15; m = 1, 3, and 5) revealed that the enthalpy and entropy changes accompanying the transition from the random coil to the triple helix are ?2500 cal and ?6.3 e.u. per one mole of the tripeptide of the form of Pro-Pro-Gly, and ?3100 cal and ?11.2 e.u. per one mole of the tripeptide of the form of Ala-Pro-Gly. The thermal instability of the triple helix composed of Ala-Pro-Gly sequences, compared to the helix of Pro-Pro-Gly sequences, is due to the larger entropy change of Ala-Pro-Gly (?11.2 e.u.) compared to that of Pro-Pro-Gly (?6.3 e.u.), not from the difference in the enthalpy change. The difference in the enthalpy change between Pro-Pro-Gly and Ala-Pro-Gly arises from the hydrophobic bond between two pyrrolidine rings of proline residues formed in the triple helix. Since the enthalpy change for the formation of hydrophobic bonds is positive, it is also concluded that only one hydrogen bond is formed in a tripeptide unit, regardless of the amino acid sequence. The enthalpy change for the formation of this hydrogen bond is ?3100 cal/mol, and that of the hydrophobic bond between two pyrrolidine rings is +600 cal/mol.     

Omega amino acids in peptide design: incorporation into helices

Incorporation of easily available achiral ω-amino acid residues into an oligopeptide results in substitution of amide bonds by polymethylene units of an aliphatic chain, thereby providing a convenient strategy for constructing a peptidomimetic. The central Gly-Gly segment of the helical octapeptide Boc-Leu-Aib-Val-Gly-Gly-Leu-Aib-Val-Ome(1) has been replaced by δ-amino-valeric acid (δ-Ava) residue in the newly designed peptide Boc-Leu-Aib-Val-δ-Ava-Leu-Aib-Val-OMe(2). ¹H-nmr results clearly suggest that in the apolar solvent CDCl₃, the δ-Ava residue is accommodated into a folded helical conformation, stabilized by successive hydrogen bonds involving the NH groups of Val(3), δ-Ava(4), and Leu(5). The δ-Ava residue must adopt a gauche-gauche-trans-gauche-gauche conformation along the central polymethylene unit of the aliphatic segment, a feature seen in an energy-minimized model conformation based on nmr parameters. The absence of hydrogen bonding functionalities, however, limits the elongation of the helix. In fact, in CDCl₃, the folded conformation consists of an N-terminal helix spanning residues 1–4, followed by a Type II β-turn at residues 5 and 6, whereas in strongly solvating media like (CD₃)₂SO, the unfolding of the N-terminal helix results in β-turn conformations at Leu(1)-Aib(2). The Type II β-turn at the Leu(5)-Aib(6) segment remains intact even in (CD₃)₂SO. CD comparisons of peptides 1 and 2 reveal a “nonhelical” spectrum for 2 in 2,2,2-trifluoroethanol. © 1996 John Wiley & Sons, Inc.     

Studying the Structural and Folding Features of Long‐Sequence Trichobrachin Peptides

In this theoretical study, the folding processes of long‐sequence trichobrachin peptides (i.e., TB IIb peptides) were investigated by molecular dynamics methods. The formation of various helical structures (i.e., 3₁₀‐, α‐, and left‐handed α‐helices) was studied with regard to the entire sequence of peptides, as well as to each amino acid. The results pointed out that TB IIb molecules showed a propensity to form helical conformations, and they could be characterized by 3₁₀‐helical structure rather than by α‐helical structure. The formation of local (i.e., i←i+3 and i←i+4) as well as of non‐local (i.e., i←i+n, where n>4; and all i→i+n) H‐bonds was also examined. The results revealed that the occurrence of local, helix‐stabilizing H‐bonds was in agreement with the appearance of helical conformations, and the non‐local H‐bonds did not produce relevant effects on the evolution of helical structures. Based on the data obtained by our structural investigation, differences were observed between the TB IIb peptides, according to the type of amino acid located in the 17th position of their sequences. In summary, the folding processes were explored for TB IIb molecules, and our theoretical study led to the conclusion that these long‐sequence peptaibols showed characteristic structural and folding features.     

Common interruptions in the repeating tripeptide sequence of non-fibrillar collagens: sequence analysis and structural studies on triple-helix peptide models

Interruptions in the repeating (Gly-X1-X2)_n amino acid sequence pattern are found in the triple-helix domains of all non-fibrillar collagens, and perturbations to the triple-helix at such sites are likely to play a role in collagen higher-order structure and function. This study defines the sequence features and structural consequences of the most common interruption, where one residue is missing from the tripeptide pattern, Gly-X1-X2-Gly-AA₁-Gly-X1-X2, designated G1G interruptions. Residues found within G1G interruptions are predominantly hydrophobic (70%), followed by a significant amount of charged residues (16%), and the Gly-X1-X2 triplets flanking the interruption are atypical. Studies on peptide models indicate the degree of destabilization is much greater when Pro is in the interruption, GP, than when hydrophobic residues (GF, GY) are present, and a rigid Gly-Pro-Hyp tripeptide adjacent to the interruption leads to greater destabilization than a flexible Gly-Ala-Ala sequence. Modeling based on NMR data indicates the Phe residue within a GF interruption is located on the outside of the triple helix. The G1G interruptions resemble a previously studied collagen interruption GPOGAAVMGPO, designated G4G-type, in that both are destabilizing, but allow continuation of rod-like triple helices and maintenance of the single residue stagger throughout the imperfection, with a loss of axial register of the superhelix on both sides. Both kinds of interruptions result in a highly localized perturbation in hydrogen bonding and dihedral angles, but the hydrophobic residue of a G4G interruption packs near the central axis of the superhelix, while the hydrophobic residue of a G1G interruption is located on the triple-helix surface. The different structural consequences of G1G and G4G interruptions in the repeating tripeptide sequence pattern suggest a physical basis for their differential susceptibility to matrix metalloproteinases in type X collagen.     

Proton magnetic resonance of the triple helix of poly(prolylprolylglycyl)n with defined degree of polymerization

Proton magnetic resonance spectra at 220 MHz were obtained for deuterium oxide and aqueous solutions of the polytripeptides (Pro-Pro-Gly)_n, which were synthesized as collagen models by the modified solid phase method. At higher temperatures, the signals of the proline C_a-protons for the peptides with n ≦ 5 and for those with n = 10 and 15 demonstrate the presence of cis and trans isomers with respect to the Gly-Pro or Pro-Pro C-N bonds. Glycine C_a-protons give typical AB type patterns. At lower temperatures, as the peptides with n = 5, 10 and 15 form triple helices, all of the resonance peaks become broad, but the whole form of the spectrum is quite similar to that of poly(l-proline) form II. The glycine C_a-proton resonances become barely detectable and the upfield peak of the two proline C_a-proton resonances fade away. At the same time, a new glycine NH resonance appears at a field slightly higher than that of a random coil. It seems to suggest that the formation of triple helices accompanies the conversion of cis proline peptide bonds into all trans bonds, and that the glycine residue environment completely changes in the helix.     

Synthesis and characterization of four sequential glycyl/glutamate polypeptides

Poly(Glu(OBzl)-Gly)_n, poly(Glu-Gly)_n, poly(Gly)-(Glu(OBzl)-Gly), and poly(Gly-Glu-Gly) were synthesized from the pentachlorophenyl esters of the sequential monomer. Both of the polymers containing free glumatic-acid residues are soluble in water, as is the lower molecular weight fraction of the polytripeptides with the benzyl ester in place. Circular dichroism studies and infrared dichroism studies suggest that the 2₁ helix is favored for the polydipeptide with removal of the benzyl ester reducing the conformational integrity. The polytripeptide showed evidence of 3₁ helix in addition to the 2₁ form, depending on solvent. A rationale for the conformations observed is developed based on the bulkiness of the side-chain residues and conformational stabilization, in certain cases, by hydrophobic interactions between the benzyl ester groups.     

Séquences tripeptidiques typiques du collagène: étude conformationnelle quantique

Theoretical and experimental investigations on the conformations of synthetic polytripeptides, with Gly-R₂-R₃ general repetitive sequence, are of great interest for the comprehension of the factors wich contribute to the triple helix stabilization.Using appropriate molecules as models, we have investigated by the PCILO method the intrinsic conformational possibilities of each « in situresidue for the following cases: Gly-imino-imino, Gly-amino-imino and Gly-imino-amino which represent the most important sequences occuring in the primary structure of collagen. The computations were carried out taking into account the two pyrrolidine ring puckerings and the preferential side-chains orientations.Extending our previous work concerning the isomeric polytripeptides Gly-X-Pro and Gly-Pro-X with X {Pro, Phe} we have studied the cases of X {Ala, Ser}. In this brief communication, we present a preliminary synthesis of these recent results.     

An energetic evaluation of a “Smith” collagen microfibril model

An energy minimized three-dimensional structure of a collagen microfibril template was constructed based on the five-stranded model of Smith (1968), using molecular modeling methods and Kollman force fields (Weiner and Kollman, 1981). For this model, individual molecules were constructed with three identical polypeptide chains ((Gly-Pro-Pro) _n , (Gly-Prop-Hyp) _n , or (Gly-Ala-Ala) _n , wheren=4, 12, and 16) coiled into a right-handed triple-helical structure. The axial distance between adjacent amino acid residues is about 0.29 nm per polypeptide chain, and the pitch of each chain is approximately 3.3 residues. The microfibril model consists of five parallel triple helices packed so that a left-handed superhelical twist exists. The structural characteristics of the computed microfibril are consistent with those obtained for collagen by X-ray diffraction and electron microscopy. The energy minimized Smith microfibril model for (Gly-Pro-Pro)₁₂ has an axial length of about 10.2 nm (for a 36 amino acid residue chain), which gives an estimated D-spacing (234 amino acids per chain) of approximately 66.2 nm. Studies of the microfibril models (Gly-Pro-Pro)₁₂, (Gly-Pro-Hyp)₁₂, and (Gly-Ala-Ala)₁₂ show that nonbonded van der Waals interactions are important for microfibril formation, while electrostatic interactions contribute to the stability of the microfibril structure and determine the specificity by which collagen molecules pack within the microfibril.     

Structurally symmetrical,easily polarizable hydrogen bonds between side chains in proteins and proton-conducting mechanisms. III

(L -Cys)_n, (L -Lys)_n, and (L -Glu)_n were studied by ir spectroscopy in terms of their degree of deprotonation or protonation. It is shown that structurally symmetrical, easily polarizable SH ?S^? ? ^?S ?HS, N⁺H ?N ? N ?H⁺N, and OH ?O^? ? ^?O ?HO hydrogen bonds are formed between the side chains. The different wave number distributions of the ir continua caused by these hydrogen bonds show that the barrier in the double-minimum proton potential decreases in the series of these hydrogen bonds. The stability of these hydrogen bonds against hydration increases in this series. The OH ?O^? ? ^?O ?HO bonds are not broken by small amounts of water. With (L -Cys)_n the formation of easily polarizable hydrogen bonds and a β-structure–coil transition are strongly interdependent. As a result of this coupling effect, the β-structure–coil transition becomes cooperative. With (L -Glu)_n, the formation of the polarizable hydrogen bonds and the observed conformational change are independent processes. The (L -Glu)_n conformation changes from α-helix to coil only if more than 80% of the residues are deprotonated. Finally, on the basis of the various types of easily polarizable hydrogen bonds, charge shifts in active centers of enzymes and the proton-conducting mechanism through hydrophobic regions of biological membranes are discussed.     

Peptide design: Crystal structure of a helical peptide module attached to a potentially nonhelical amino terminal segment

The peptide Boc-Gly-Dpg-Gly-Val-Ala-Leu-Aib-Val-Ala-Leu-OMe has been designed to examine the structural consequences of placing a short segment with a low helix propensity at the amino terminus of a helical heptapeptide module. The Gly-Dpg-Gly segment is a potential connecting element in the synthetic construction of a helix-linker-helix motif. Crystal parameters for the peptide are P2₁, a = 8.651(3) Å, b = 46.826(13) Å, c = 16.245 Å, β = 90.13(3)*, Z = 4; 2 independent molecules/asymmetric unit. The structure reveals almost identical conformations for the two independent molecules. The backbone is completely helical for residues 2–9, with one 4 → 1 hydrogen bond and six 5 → 1 hydrogen bonds. The α,α-di-n-propylglycine residue adopts a helical conformation. Gly(1) adopts an extended conformation resulting in a nonhelical N-terminus, with the Boc group swinging away from the helix. The lateral association of helices in the b axis direction is unusual in that the helix axes are directed up or down (parallel or antiparallel) by pairs: ↓↓↑↑↓↓, etc. © 1996 John Wiley & Sons, Inc.     

Crystal structure of human thioredoxin revealing an unraveled helix and exposed S-nitrosation site

Thioredoxins reduce disulfide bonds and other thiol modifications in all cells using a CXXC motif. Human thioredoxin 1 is unusual in that it codes for an additional three cysteines in its 105 amino acid sequence, each of which have been implicated in other reductive activities. Cys 62 and Cys 69 are buried in the protein interior and lie at either end of a short helix (helix 3), and yet can disulfide link under oxidizing conditions. Cys 62 is readily S‐nitrosated, giving rise to a SNO modification, which is also buried. Here, we present two crystal structures of the C69S/C73S mutant protein under oxidizing (1.5 Å) and reducing (1.1 Å) conditions. In the oxidized structure, helix 3 is unraveled and displays a new conformation that is stabilized by a series of new hydrogen bonds and a disulfide link with Cys 62 in a neighboring molecule. The new conformation provides an explanation for how a completely buried residue can participate in SNO exchange reactions.