Molecular specifications of a mineral modulation sequence derived from the aragonite-promoting protein n16

In the nacre layer of the mollusk, proteins play an important role in regulating the morphology and lattice structure of calcium carbonate minerals. However, this process remains elusive due to the fact that we do not understand how protein sequences control the structure and morphology of biominerals. To take us a step further in this direction, we report the molecular structure of a 30 AA N-terminal mineral interactive sequence (n16N) of the aragonite-promoting protein, n16, and contrast these findings to those previously reported for two "calcite-blocker" nacre-associated sequences, AP7N and AP24N. We find that n16N is conformationally labile and adopts a random-coil conformation that possesses short, dispersed extended beta-strand segments that are located at the A1-Y2, K5-Y9, Y11-I14, and D21-N25 sequence blocks. Like AP7N and AP24N, Ca(II) ion interactions with n16N alter chain dynamics and local structure, and n16N is adsorbed onto calcite crystals and cannot easily be displaced via differential washing techniques. Furthermore, all three sequences have planar surface regions that could serve as putative sites for mineral interactions or ion cluster formation. However, what sets n16N apart from AP7N and AP24N are different folding propensities as well as unique molecular surface features and amino acid composition. n16N has a more condensed structure that, in the presence of TFE, folds into a beta-strand. This contrasts with the more open structures of AP7N and AP24N that are induced by TFE to fold into alpha-helices. Mapping of the n16N molecular surface reveals significant cationic regions and diffuse anionic charge, which contrasts with the small anionic "pocket" regions of AP7N/AP24N. Finally, n16N has 50% fewer sites for mineral surface- or ion cluster-associated water interactions compared to AP7N and AP24N. Overall, the structure of n16N is "tuned" to a different function within the in vitro mineralization scheme. The different features found in AP7N, AP24N, and n16N could be exploited for engineering polypeptides that recognize and bind to different surface features of inorganic crystalline solids.     

Structural characterization of the N-terminal mineral modification domains from the molluscan crystal-modulating biomineralization proteins, AP7 and AP24

The AP7 and AP24 proteins represent a class of mineral-interaction polypeptides that are found in the aragonite-containing nacre layer of mollusk shell (H. rufescens). These proteins have been shown to preferentially interfere with calcium carbonate mineral growth in vitro. It is believed that both proteins play an important role in aragonite polymorph selection in the mollusk shell. Previously, we demonstrated the 1-30 amino acid (AA) N-terminal sequences of AP7 and AP24 represent mineral interaction/modification domains in both proteins, as evidenced by their ability to frustrate calcium carbonate crystal growth at step edge regions. In this present report, using free N-terminal, C(alpha)-amide "capped" synthetic polypeptides representing the 1-30 AA regions of AP7 (AP7-1 polypeptide) and AP24 (AP24-1 polypeptide) and NMR spectroscopy, we confirm that both N-terminal sequences possess putative Ca (II) interaction polyanionic sequence regions (2 x -DD- in AP7-1, -DDDED- in AP24-1) that are random coil-like in structure. However, with regard to the remaining sequences regions, each polypeptide features unique structural differences. AP7-1 possesses an extended beta-strand or polyproline type II-like structure within the A11-M10, S12-V13, and S28-I27 sequence regions, with the remaining sequence regions adopting a random-coil-like structure, a trait common to other polyelectrolyte mineral-associated polypeptide sequences. Conversely, AP24-1 possesses random coil-like structure within A1-S9 and Q14-N16 sequence regions, and evidence for turn-like, bend, or loop conformation within the G10-N13, Q17-N24, and M29-F30 sequence regions, similar to the structures identified within the putative elastomeric proteins Lustrin A and sea urchin spicule matrix proteins. The similarities and differences in AP7 and AP24 N-terminal domain structure are discussed with regard to joint AP7-AP24 protein modification of calcium carbonate growth.     

A C-RING-like domain participates in protein self-assembly and mineral nucleation

AP7 is a nacre-associated protein of the mollusk shell that forms supramolecular assemblies that nucleate single-crystal aragonite in vitro. AP7 possesses two major sequence regions: a random coil 30-amino acid N-terminal domain (AP7N) and a partially disordered 36-amino acid C-terminal domain (AP7C) that exhibits imperfect sequence homology to the C subclass of the intracellular RING domain family. We report here new findings that implicate the C-RING domain in AP7-mediated supramolecular assembly and single-crystal mineral formation. AP7 protein spontaneously self-assembles over a pH range of 4-9 and is monomeric at pH >9.5. AP7N and AP7C both oligomerize over the pH range of 4-9, with the AP7C sequence closely resembling AP7 in terms of particle morphology and size. In vitro mineralization experiments demonstrate that both AP7N and AP7C form supramolecular assemblies that nucleate single-crystal calcium carbonates. Comparison of previously published nuclear magnetic resonance-based structures of AP7C and AP7N reveals the significant presence of complementary anionic-cationic electrostatic molecular surfaces on AP7C that are not found on AP7N, and this may explain the noted discrepancies between the two domains in terms of self-assembly and single-crystal nucleation. We conclude that the C-RING-like sequence is an important site for AP7 self-association and mineral nucleation, and this represents the first known instance of a RING-like sequence performing these functions within an extracellular protein.     

Recognition schemes for protein-nucleic acid interactions

The molecular forces involved in protein-nucleic acid interaction are electrostatic, stacking and hydrogen-bonding. These interactions have a certain amount of specificity due to the directional nature of such interactions and the spatial contributions of the steric effects of different substituent groups. Quantum chemical calculations on these interactions have been reported which clearly bring out such features. While the binding energies for electrostatic interactions are an order of magnitude higher, the differences in interaction energies for structures stabilised by hydrogen-bonding and stacking are relatively small. Thus, the molecular interactions alone cannot explain the highly specific nature of binding observed in certain segments of proteins and nucleic acids. It is therefore logical to assume that the sequence dependent three dimensional structures of these molecules help to place the functional groups in the correct geometry for a favourable interaction between the two molecules. We have carried out 2D-FT nuclear magnetic resonance studies on the oligonucleotide d-GGATCCGGATCC. This oligonucleotide sequence has two binding sites for the restriction enzyme Bam H1. Our studies indicate that the conformation of this DNA fragment is predominantly B-type except near the binding sites where the ribose ring prefers a³E conformation. This interesting finding raises the general question about the presence of specificity in the inherent backbone structures of proteins and nucleic acids as opposed to specific intermolecular interactions which may induce conformational changes to facilitate such binding.     

Characterization of two molluscan crystal-modulating biomineralization proteins and identification of putative mineral binding domains

Ethylenediamine-tetraacetic acid extracted water-soluble matrix proteins in molluscan shells secreted from the mantle epithelia are believed to control crystal nucleation, morphology, orientation, and phase of the deposited mineral. Previously, atomic force microscopy demonstrated that abalone nacre proteins bind to growing step edges and to specific crystallographic faces of calcite, suggesting that inhibition of calcite growth may be one of the molecular processes required for growth of the less thermodynamically stable aragonite phase. Previous experiments were done with protein mixtures. To elucidate the role of single proteins, we have characterized two proteins isolated from the aragonitic component of nacre of the red abalone, Haliotis rufescens. These proteins, purified by hydrophobic interaction chromatography, are designated AP7 and AP24 (aragonitic protein of molecular weight 7 kDa and 24 kDa, respectively). Degenerate oligonucleotide primers corresponding to N-terminal and internal peptide sequences were used to amplify cDNA clones by a polymerase chain reaction from a mantle cDNA library; the deduced primary amino acid sequences are presented. Preliminary crystal growth experiments demonstrate that protein fractions enriched in AP7 and AP24 produced CaCO(3) crystals with morphology distinct from crystals grown in the presence of the total mixture of soluble aragonite-specific proteins. Peptides corresponding to the first 30 residues of the N-terminal sequences of both AP7 and AP24 were generated. The synthetic peptides frustrate the progression of step edges of a growing calcite surface, indicating that sequence features within the N-termini of AP7 and AP24 include domains that interact with CaCO(3). CD analyses demonstrate that the N-terminal peptide sequences do not possess significant percentages of alpha-helix or beta-strand secondary structure in solution. Instead, in both the presence and absence of Ca(II), the peptides retain unfolded conformations that may facilitate protein-mineral interaction.     

Three-dimensional structure of the purple intermediate of porcine kidney D-amino acid oxidase. Optimization of the oxidative half-reaction through alignment of the product with reduced flavin

The three-dimensional structure of the purple intermediate of porcine kidney D-amino acid oxidase (DAO) was solved by cryo-X-ray crystallography; the purple intermediate is known to comprise a complex between the dehydrogenated product, an imino acid, and the reduced form of DAO. The crystalline purple intermediate was obtained by anaerobically soaking crystals of oxidized DAO in a buffer containing excess D-proline as the substrate. The dehydrogenated product, delta(1)-pyrrolidine-2-carboxylate (DPC), is found sandwiched between the phenol ring of Tyr 224 and the planar reduced flavin ring. The cationic protonated imino nitrogen is within hydrogen-bonding distance of the backbone carbonyl oxygen of Gly 313. The carboxyl group of DPC is recognized by the Arg 283 guanidino and Tyr 228 hydroxyl groups through ion-pairing and hydrogen-bonding, respectively. The (+)HN=C double bond of DPC overlaps the N(5)-C(4a) bond of reduced flavin. The electrostatic effect of the cationic nitrogen of DPC is suggested to shift the resonance hybridization of anionic reduced flavin toward a canonical form with a negative charge at C(4a), thereby augmenting the electron density at C(4a), from which electrons are transferred to molecular oxygen during reoxidation of reduced flavin. The reactivity of reduced flavin in the purple intermediate, therefore, is enhanced through the alignment of DPC with respect to reduced flavin.     

Hydrogen bonding of iron-coordinated histidine in heme proteins.

The hydrogen-bonding motifs of the proton on the N delta atom of iron-coordinated histidine residues in heme proteins have been classified into three categories: (1) Those in which the hydrogen-bond acceptor is either an amino acid residue (serine) directly adjacent to the histidine or a carbonyl group of the polypeptide chain less than five residues away from the histidine; (2) those in which the hydrogen-bonding acceptor is a carbonyl group of the polypeptide backbone associated with an amino acid residue 8 to 17 residues away from the histidine; and (3) those in which the hydrogen-bonding acceptor is an exogenous water molecule or an amino acid residue located far from the histidine in the amino acid sequence. Some biological functions are defined by this classification, whereas others span all classes.     

Basis of substrate binding by the chaperonin GroEL.

The molecular chaperonins are essential proteins involved in protein folding, complex assembly, and polypeptide translocation. While there is abundant structural information about the machinery and the mechanistic details of its action are well studied, it is yet unresolved how chaperonins recognize a large number of structurally unrelated polypeptides in their unfolded or partially folded forms. To determine the nature of chaperonin-substrate recognition, we have characterized by NMR methods the interactions of GroEL with synthetic peptides that mimic segments of unfolded proteins. In previous work, we found using transferred nuclear Overhauser effect (trNOE) analysis that two 13 amino acid peptides bound GroEL in an amphipathic alpha-helical conformation. By extending the study to a variety of peptides with differing sequence motifs, we have observed that peptides can adopt conformations other than alpha-helix when bound to GroEL. Furthermore, peptides of the same composition exhibited significantly different affinities for GroEL as manifested by the magnitude of trNOEs. Binding to GroEL correlates well with the ability of the peptide to cluster hydrophobic residues on one face of the peptide, as determined by the retention time on reversed-phase (RP) HPLC. We conclude that the molecular basis of GroEL-substrate recognition is the presentation of a hydrophobic surface by an incompletely folded polypeptide and that many backbone conformations can be accommodated.     

Comparison of energy components of proteins from thermophilic and mesophilic organisms

In order to infer the energetic determinants of thermophilic proteins, molecular mechanics calculations were applied to five proteins from thermophilic eubacteria and their mesophilic homologs. The energy function includes a hydration term as well as the electrostatic contribution from the solvent in addition to the usual conformational energy terms. We calculated energy values for three different states of each protein: the native, near-native, and unfolded structures. The energy difference and its components between pairs of these states were compared. The hypothetical near-native structures have almost the same backbone conformation as the native structure but with largely distorted side-chain packing, thus enabling us to extract the energy components important for stabilizing the native backbone topology itself, irrespective of structural details. It was found that the sum of the electrostatic and hydration energies, although of large positive values, were consistently lower for the thermophilic proteins than for their mesophilic counterparts. This trend was observed in the energy difference not only between the native and unfolded structures, but also between the near-native and unfolded structures. In contrast, the energy components regarding side-chain packing did not show any clear tendency. These results suggest that the thermophilic proteins are stabilized so that the precise packing of the native structure does not significantly affect the stability. Implications of this conclusion are also discussed.     

Roles of key residues and lipid dynamics reveal pHLIP-membrane interactions at intermediate pH

The pH-low insertion peptide (pHLIP) and its analogs sense the microenvironmental pH variations in tumorous cells and serve as useful anticancer drug deliveries. The pHLIP binds peripherally to membranes and adopts random coil conformation at the physiological pH. The peptide switches from random coil to α-helical conformation and inserts unidirectionally into membrane bilayers when pH drops below a critical transition value that has been routinely determined by the Trp fluorescence spectroscopy. Recent high-resolution studies using solid-state NMR spectroscopy revealed the presence of thermodynamically stable intermediate states of membrane-associated pHLIP around the fluorescence-based transition pH-value. However, the molecular structural features and their mechanistic roles of these intermediate states in the pH-driven membrane insertion process of pHLIP remain largely unknown. This work utilizes solid-state NMR spectroscopy to explore 1) the mechanistic roles of key proline and arginine residues within the pHLIP sequence at intermediate pH-values, and 2) the changes in lipid dynamics at intermediate pH-values in multiple types of model bilayers with anionic phospholipid and/or cholesterol. Our results demonstrate several molecular structural and dynamics changes at around the transition pH-values, including the isomerization of proline-threonine backbone configuration, breaking of arginine-aspartic acid salt bridge and the formation of arginine-lipid interactions, and a universal decreasing of dynamics in lipid headgroups and alkyl chains. Overall, the outcomes provide important insights on the molecular interactions between pHLIP and membrane bilayers at intermediate pH-values and, therefore, prompt the understanding of pH-driven membrane insertion process of this anticancer drug-delivering peptide.     

Identification and structural characterization of an unusual RING-like sequence within an extracellular biomineralization protein, AP7

The RING or Really Interesting New Gene represents a family of eukaryotic sequences that bind Zn (II) ions and participate in intracellular processes involving protein-protein interaction. Although found in over 400 different proteins, very little is known regarding the structure-function properties of these domains because of the aggregation problems associated with RING sequences. To augment this data set, we report an unusual 36 AA C-terminal sequence of an extracellular matrix mollusk shell protein, AP7, that exhibits partial homology to the RING family. This Cys, His-containing sequence, termed AP7C, binds Zn (II) and other multivalent ions, but does not utilize a tetracoordinate complexation scheme for binding such as that found in Zn (II) finger polypeptides. Moreover, unlike Zn (II) finger and RING domains, this 36 AA can fold into a relatively stable structure in the absence of Zn (II). This folded structure consists of three short helical segments (A, B, and C), with segments A and B separated by a 4 AA type I beta-turn region and segments B and C separated by a 7 AA loop-like region. Interestingly, the putative RING-like region, -RRPFHECALCYSI-, experiences slow conformational exchange between two structural states in solution, most likely in response to imido ring interconversion at P8 and P21. Poisson-Boltzmann solvation calculations reveal that the AP7C molecular surface possesses a cationic region near its N-terminus, which lies adjacent to the 30 AA mineral modification domain in the AP7 protein. Given that the AP7C sequence does not influence mineralization, it is probable that this cationic pseudo-RING region is utilized by the AP7 protein for other tasks such as protein-protein interaction within the mollusk shell matrix.     

Polyproline II helical structure in protein unfolded states: lysine peptides revisited

The left-handed polyproline II (PPII) helix gives rise to a circular dichroism spectrum that is remarkably similar to that of unfolded proteins. This similarity has been used as the basis for the hypothesis that unfolded proteins possess considerable PPII helical content. It has long been known that homopolymers of lysine adopt the PPII helical conformation at neutral pH, presumably a result of electrostatic repulsion between side chains. It is shown here that a seven-residue lysine peptide also adopts the PPII conformation. In contrast with homopolymers of lysine, this short peptide is shown to retain PPII helical character under conditions in which side-chain charges are heavily screened or even neutralized. The most plausible explanation for these observations is that the peptide backbone favors the PPII conformation to maximize favorable interactions with solvent. These data are evidence that unfolded proteins do indeed possess PPII content, indicating that the ensemble of unfolded states is significantly smaller than is commonly assumed.     

Native and nonnative conformational preferences in the urea-unfolded state of barstar

The refolding of barstar from its urea-unfolded state has been studied extensively using various spectroscopic probes and real-time NMR, which provide global and residue-specific information, respectively, about the folding process. Here, a preliminary structural characterization by NMR of barstar in 8 M urea has been carried out at pH 6.5 and 25 degrees C. Complete backbone resonance assignments of the urea-unfolded protein were obtained using the recently developed three-dimensional NMR techniques of HNN and HN(C)N. The conformational propensities of the polypeptide backbone in the presence of 8 M urea have been estimated by examining deviations of secondary chemical shifts from random coil values. For some residues that belong to helices in native barstar, 13C(alpha) and 13CO secondary shifts show positive deviations in the urea-unfolded state, indicating that these residues have propensities toward helical conformations. These residues are, however, juxtaposed by residues that display negative deviations indicative of propensities toward extended conformations. Thus, segments that are helical in native barstar are unlikely to preferentially populate the helical conformation in the unfolded state. Similarly, residues belonging to beta-strands 1 and 2 of native barstar do not appear to show any conformational preferences in the unfolded state. On the other hand, residues belonging to the beta-strand 3 segment show weak nonnative helical conformational preferences in the unfolded state, indicating that this segment may possess a weak preference for populating a helical conformation in the unfolded state.     

A comprehensive library of blocked dipeptides reveals intrinsic backbone conformational propensities of unfolded proteins

Despite prolonged scientific efforts to elucidate the intrinsic peptide backbone preferences of amino-acids based on understanding of intermolecular forces, many open questions remain, particularly concerning neighboring peptide interaction effects on the backbone conformational distribution of short peptides and unfolded proteins. Here, we show that spectroscopic studies of a complete library of 400 dipeptides reveal that, irrespective of side-chain properties, the backbone conformation distribution is narrow and they adopt polyproline II and β-strand, indicating the importance of backbone peptide solvation and electronic effects. By directly comparing the dipeptide circular dichroism and NMR results with those of unfolded proteins, the comprehensive dipeptides form a complete set of structural motifs of unfolded proteins. We thus anticipate that the present dipeptide library with spectroscopic data can serve as a useful database for understanding the nature of unfolded protein structures and for further refinements of molecular mechanical parameters.     

Structural effects of nucleobase variations at key active site residue Ade38 in the hairpin ribozyme

The hairpin ribozyme requires functional groups from Ade38 to achieve efficient bond cleavage or ligation. To identify molecular features that contribute to catalysis, structures of position 38 base variants 2,6-diaminopurine (DAP), 2-aminopurine (AP), cytosine (Cyt), and guanine (Gua) were determined between 2.2 and 2.8 A resolution. For each variant, two substrate modifications were compared: (1) a 2'-O-methyl-substituent at Ade-1 was used in lieu of the nucleophile to mimic the precatalytic state, and (2) a 3'-deoxy-2',5'-phosphodiester linkage between Ade-1 and Gua+1 was used to mimic a reaction-intermediate conformation. While the global fold of each variant remained intact, the results revealed the importance of Ade38 N1 and N6 groups. Absence of N6 resulting from AP38 coincided with failure to localize the precatalytic scissile phosphate. Cyt38 severely impaired catalysis in a prior study, and its structures here indicated an anti base conformation that sequesters the imino moiety from the scissile bond. Gua38 was shown to be even more deleterious to activity. Although the precatalytic structure was nominally affected, the reaction-intermediate conformation indicated a severe electrostatic clash between the Gua38 keto oxygen and the pro-Rp oxygen of the scissile bond. Overall, position 38 modifications solved in the presence of 2'-OMe Ade-1 deviated from in-line geometry, whereas variants with a 2',5' linkage exhibited S-turn destabilization, as well as base conformational changes from syn to anti. These findings demonstrate the importance of the Ade38 Watson-Crick face in attaining a reaction-intermediate state and the sensitivity of the RNA fold to restructuring when electrostatic and shape features fail to complement.     

The dominant interaction between peptide and urea is electrostatic in nature: a molecular dynamics simulation study

The conformational equilibrium of a blocked valine peptide in water and aqueous urea solution is studied using molecular dynamics simulations. Pair correlation functions indicate enhanced concentration of urea near the peptide. Stronger hydrogen bonding of urea-peptide compared to water-peptide is observed with preference for helical conformation. The potential of mean force, computed using umbrella sampling, shows only small differences between urea and water solvation that are difficult to quantify. The changes in solvent structure around the peptide are explained by favorable electrostatic interactions (hydrogen bonds) of urea with the peptide backbone. There is no evidence for significant changes in hydrophobic interactions in the two conformations of the peptide in urea solution. Our simulations suggest that urea denatures proteins by preferentially forming hydrogen bonds to the peptide backbone, reducing the barrier for exposing protein residues to the solvent, and reaching the unfolded state.     

Solvent accessibilities in glycyl, alanyl and seryl dipeptides.

Theoretical studies on glycyl-alanyl and seryl dipeptides were performed to determine the probable backbone and side-group conformations that are preferred for solvent interaction. By following the method of Lee & Richards [(1971) J. Mol. Biol. 55, 379-400], a solute molecule is represented by a set of interlocking spheres of appropriate van der Waals radii assigned to each atom, and a solvent (water) molecule is rolled along the envelope of the van der Waals surface, and the surface accessible to the solvent molecule, and hence the solvent accessibility for a particular conformation of the solute molecule, is computed. From the calculated solvent accessibilities for various conformations, solvation maps for dipeptides were constructed. These solvation maps suggest that the backbone polar atoms could interact with solvent molecules selectively, depending on the backbone conformation. A conformation in the right-handed bridge (zetaR) region is favoured for both solvent interaction and intrachain hydrogen-bonding. Also the backbone side-chain hydrogen-bonding within the same dipeptide fragment in proteins is less favoured than hydrogen-bonding between side chain and water and between side chain and atoms of other residues. Solvent accessibilities suggest that very short distorted alphaR-helical and extended-structural parts may be stabilized via solvent interaction, and this could easily be possible at the surface of the protein molecules, in agreement with protein-crystal data.     

UV resonance Raman investigation of a 3(10)-helical peptide reveals a rough energy landscape

We used UVRRS at 194 and 204 nm excitation to examine the backbone conformation of a 13-residue polypeptide (gp41(659-671)) that has been shown by NMR to predominantly fold into a 3(10)-helix. Examination of the conformation sensitive AmIII(3) region indicates the peptide has significant populations of beta-turn, PPII, 3(10)-helix, and pi-helix-like conformations but little alpha-helix. We estimate that at 1 degree C on average six of the 12 peptide bonds are in folded conformations (predominantly 3(10)- and pi-helix), while the other six are in unfolded (beta-turn/PPII) conformations. The folded and unfolded populations do not change significantly as the temperature is increased from 1 to 60 degrees C, suggesting a unique energy landscape where the folded and unfolded conformations are essentially degenerate in energy and exhibit identical temperature dependences.     

AbDesign: An algorithm for combinatorial backbone design guided by natural conformations and sequences

Computational design of protein function has made substantial progress, generating new enzymes, binders, inhibitors, and nanomaterials not previously seen in nature. However, the ability to design new protein backbones for function—essential to exert control over all polypeptide degrees of freedom—remains a critical challenge. Most previous attempts to design new backbones computed the mainchain from scratch. Here, instead, we describe a combinatorial backbone and sequence optimization algorithm called AbDesign, which leverages the large number of sequences and experimentally determined molecular structures of antibodies to construct new antibody models, dock them against target surfaces and optimize their sequence and backbone conformation for high stability and binding affinity. We used the algorithm to produce antibody designs that target the same molecular surfaces as nine natural, high‐affinity antibodies; in five cases interface sequence identity is above 30%, and in four of those the backbone conformation at the core of the antibody binding surface is within 1 Å root‐mean square deviation from the natural antibodies. Designs recapitulate polar interaction networks observed in natural complexes, and amino acid sidechain rigidity at the designed binding surface, which is likely important for affinity and specificity, is high compared to previous design studies. In designed anti‐lysozyme antibodies, complementarity‐determining regions (CDRs) at the periphery of the interface, such as L1 and H2, show greater backbone conformation diversity than the CDRs at the core of the interface, and increase the binding surface area compared to the natural antibody, potentially enhancing affinity and specificity. Proteins 2015; 83:1385–1406. © 2015 Wiley Periodicals, Inc.     

Characterisation of the conformational properties of urea-unfolded Im7: implications for the early stages of protein folding

The colicin immunity protein Im7 folds from its unfolded state in 6 M urea to its native four-helix structure through an on-pathway intermediate that lacks one of the helices of the native structure (helix III). In order to further characterize the folding mechanism of Im7, we have studied the conformational properties of the protein unfolded in 6 M urea in detail using heteronuclear NMR. Triple-resonance experiments with 13C/15N-labelled Im7 in 6 M urea provided almost complete resonance assignments for the backbone nuclei, and measurement of backbone 15N relaxation parameters allowed dynamic ordering of the unfolded polypeptide chain to be investigated. Reduced spectral density mapping and fitting backbone R2 relaxation rates to a polymer dynamics model identified four clusters of interacting residues, each predicted by the average area buried upon folding for each residue. Chemical shift analyses and measurement of NOEs detected with a long mixing-time 1H-1H-15N NOESY-HSQC spectrum confirmed the formation of four clusters. Each cluster of interacting side-chains in urea-unfolded Im7 occurs in a region of the protein that forms a helix in the protein, with the largest clusters being associated with the three long helices that are formed in the on-pathway folding intermediate, whilst the smallest cluster forms a helix only in the native state. NMR studies of a Phe15Ala Im7 variant and a protein in which residues 51-56 are replaced by three glycine residues (H3G3 Im7*), indicated that the clusters do not interact with each other, possibly because they are solvated by urea, as indicated by analysis of NOEs between the protein and the solvent. Based on these data, we suggest that dilution of the chaotrope to initiate refolding will result in collapse of the clusters, leading to the formation of persistent helical structure and the generation of the three-helix folding intermediate.