The ensemble folding kinetics of the FBP28 WW domain revealed by an all-atom Monte Carlo simulation in a knowledge-based potential

In this work, we apply a detailed all‐atom model with a transferable knowledge‐based potential to study the folding kinetics of Formin‐Binding protein, FBP28, which is a canonical three‐stranded β‐sheet WW domain. Replica exchange Monte Carlo simulations starting from random coils find native‐like (Cα RMSD of 2.68 Å) lowest energy structure. We also study the folding kinetics of FBP28 WW domain by performing a large number of ab initio Monte Carlo folding simulations. Using these trajectories, we examine the order of formation of two β‐hairpins, the folding mechanism of each individual β‐hairpin, and transition state ensemble (TSE) of FBP28 WW domain and compare our results with experimental data and previous computational studies. To obtain detailed structural information on the folding dynamics viewed as an ensemble process, we perform a clustering analysis procedure based on graph theory. Further, a rigorous P_fold analysis is used to obtain representative samples of the TSEs showing good quantitative agreement between experimental and simulated Φ values. Our analysis shows that the turn structure between first and second β strands is a partially stable structural motif that gets formed before entering the TSE in FBP28 WW domain and there exist two major pathways for the folding of FBP28 WW domain, which differ in the order and mechanism of hairpin formation. Proteins 2011. © 2011 Wiley‐Liss, Inc.     

Solution structure of the single-domain prolyl cis/trans isomerase PIN1At from Arabidopsis thaliana

The 119-amino acid residue prolyl cis/trans isomerase from Arabidopsis thaliana (PIN1At) is similar to the catalytic domain of the human hPIN1. However, PIN1At lacks the N-terminal WW domain that appears to be essential for the hPIN1 function. Here, the solution structure of PIN1At was determined by three-dimensional nuclear magnetic resonance spectroscopy. The PIN1At fold could be superimposed on that of the catalytic domain of hPIN1 and had a 19 residue flexible loop located between strand beta1 and helix alpha1. The dynamical features of this beta1/alpha1-loop, which are characteristic for a region involved in protein-protein interactions, led to exchange broadening in the NMR spectra. When sodium sulfate salt was added to the protein sample, the beta1/alpha1 loop was stabilized and, hence, a complete backbone resonance assignment was obtained. Previously, with a phospho-Cdc25 peptide as substrate, PIN1At had been shown to catalyze the phosphoserine/phosphothreonine prolyl cis/trans isomerization specifically. To map the catalytic site of PIN1At, the phospho-Cdc25 peptide or sodium sulfate salt was added in excess to the protein and chemical shift changes in the backbone amide protons were monitored in the (1)H(N)-(15)N heteronuclear single quantum coherence spectrum. The peptide caused perturbations in the loops between helix alpha4 and strand beta3, between strands beta3 and beta4, in the alpha3 helix, and in the beta1/alpha1 loop. The amide groups of the residues Arg21 and Arg22 showed large chemical shift perturbations upon phospho-Cdc25 peptide or sulfate addition. We conclude that this basic cluster formed by Arg21 and Arg22, both located in the beta1/alpha1 loop, is homologous to that found in the hPIN1 crystal structure (Arg68 and Arg69), which also is involved in sulfate ion binding. We showed that the sulfate group competed for the interaction between PIN1At and the phospho-Cdc25 peptide. In the absence of the WW domain, three hydrophobic residues (Ile33, Ile34, and Leu35) located in the long flexible loop and specific for the plant PIN-type peptidyl prolyl cis/trans isomerases (PPIases) could be an additional interaction site in PIN1At. However, phospho-peptide addition did not affect the resonances of these residues significantly. Electrostatic potential calculations revealed a negatively charged area not found in hPIN1 on the PIN1At molecular surface, which corresponds to the surface shielded by the WW domain in hPIN1. Based on our experimental results and the molecular specificities of the PIN1At enzyme, functional implications of the lack of WW domains in this plant PIN-type PPIase will be discussed.     

p13(SUC1) and the WW domain of PIN1 bind to the same phosphothreonine-proline epitope

The WW domain of the human PIN1 and p13(SUC1), a subunit of the cyclin-dependent kinase complex, were previously shown to be involved in the regulation of the cyclin-dependent kinase complex activity at the entry into mitosis, by an unresolved molecular mechanism. We report here experimental evidence for the direct interaction of p13(SUC1) with a model CDC25 peptide, dependent on the phosphorylation state of its threonine. Chemical shift perturbation of backbone (1)H(N), (15)N, and (13)Calpha resonances during NMR titration experiments allows accurate identification of the binding site, primarily localized around the anion-binding site, occupied in the crystal structure of the homologous p9(CKSHs2) by a sulfate molecule. The epitope recognized by p13(SUC1) includes the proline at position +1 of the phosphothreonine, as was shown by the decrease in affinity for a mutated CDC25 phosphopeptide, containing an alanine/proline substitution. No direct interaction between the PIN1 WW domain or its catalytic proline cis/trans-isomerase domain and p13(SUC1) was detected, but our study showed that in vitro the WW domain of the human PIN1 antagonizes the binding of the p13(SUC1) to the CDC25 phosphopeptide, by binding to the same phosphoepitope. We thus propose that the full cyclin-dependent kinase complex stimulates the phosphorylation of CDC25 through binding of its p13(SUC1) module to the phosphoepitope of the substrate and that the reported WW antagonism of p13(SUC1)-stimulated CDC25 phosphorylation is caused by competitive binding of both protein modules to the same phosphoepitope.     

The Arabidopsis thaliana PIN1At gene encodes a single-domain phosphorylation-dependent peptidyl prolyl cis/trans isomerase

A homologue of the human site-specific prolyl cis/trans isomerase PIN1 was identified in Arabidopsis thaliana. The PIN1At gene encodes a protein of 119 amino acids that is 53% identical with the catalytic domain of the human PIN1 parvulin. Steady-state PIN1At mRNA is found in all plant tissues tested. We show by two-dimensional NMR spectroscopy that the PIN1At is a prolyl cis/trans isomerase with specificity for phosphoserine-proline bonds. PIN1At is the first example of an eukaryotic parvulin without N- or C-terminal extensions. The N-terminal WW domain of 40 amino acids, typical of all the phosphorylation-dependent eukaryotic parvulins, is absent. However, triple-resonance NMR experiments showed that PIN1At contained a hydrophobic helix similar to the alpha1 helix observed in PIN1 that could mediate the protein-protein interactions.     

Domain formation in a fluid mixed lipid bilayer modulated through binding of the C2 protein motif

The role and mechanism of formation of lipid domains in a functional membrane have generally received limited attention. Our approach, based on the hypothesis that thermodynamic coupling between lipid-lipid and protein-lipid interactions can lead to domain formation, uses a combination of an experimental lipid bilayer model system and Monte Carlo computer simulations of a simple model of that system. The experimental system is a fluid bilayer composed of a binary mixture of phosphatidylcholine (PC) and phosphatidylserine (PS), containing 4% of a pyrene-labeled anionic phospholipid. Addition of the C2 protein motif (a structural domain found in proteins implicated in eukaryotic signal transduction and cellular trafficking processes) to the bilayer first increases and then decreases the excimer/monomer ratio of the pyrene fluorescence. We interpret this to mean that protein binding induces anionic lipid domain formation until the anionic lipid becomes saturated with protein. Monte Carlo simulations were performed on a lattice representing the lipid bilayer to which proteins were added. The important parameters are an unlike lipid-lipid interaction term and an experimentally derived preferential protein-lipid interaction term. The simulations support the experimental conclusion and indicate the existence of a maximum in PS domain size as a function of protein concentration. Thus, lipid-protein coupling is a possible mechanism for both lipid and protein clustering on a fluid bilayer. Such domains could be precursors of larger lipid-protein clusters ('rafts'), which could be important in various biological processes such as signal transduction at the level of the cell membrane.     

Folding, misfolding, and amyloid protofibril formation of WW domain FBP28

We study the folding mechanism of a triple beta-strand WW domain from the Formin binding protein 28 (FBP28) at atomic resolution with explicit water model using replica exchange molecular dynamics computer simulations. Extended sampling over a wide range of temperatures to obtain the free energy, enthalpy, and entropy surfaces as a function of structural coordinates has been performed. Simulations were started from different configurations covering the folded and unfolded states. In the free energy landscape a transition state is identified and its structures and -values are compared with experimental data from a homologous protein, the prolyl-isomerase Pin1 WW domain. A stable intermediate state is found to accumulate during the simulation characterized by the carboxyl-terminal beta-strand 3 having misregistered hydrogen bonds and where the structural heterogeneity is due to nonnative turn II formation. Furthermore, the aggregation behavior of the FBP28 WW domain may be related to one such misfolded structure, which has a much lower free energy of dimer formation than that of the native dimer. Based on the misfolded dimer, aggregation to form protofibril structure is discussed.     

Simulated self-assembly of the HIV-1 capsid: protein shape and native contacts are sufficient for two-dimensional lattice formation

We report Monte Carlo simulations of the initial stages of self-assembly of the HIV-1 capsid protein (CA), using a coarse-grained representation that mimics the CA backbone structure and intermolecular contacts observed experimentally. A simple representation of N-terminal domain/N-terminal domain and N-terminal domain/C-terminal domain interactions, coupled with the correct protein shape, is sufficient to drive formation of an ordered lattice with the correct hexagonal symmetry in two dimensions. We derive an approximate concentration/temperature phase diagram for lattice formation, and we investigate the pathway by which the lattice develops from initially separated CA dimers. Within this model, lattice formation occurs in two stages: 1), condensation of CA dimers into disordered clusters; and 2), nucleation of the lattice by the appearance of one hexamer unit within a cluster. Trimers of CA dimers are important early intermediates, and pentamers are metastable within clusters. Introduction of a preformed hexamer at the beginning of a Monte Carlo run does not directly seed lattice formation, but does facilitate the formation of large clusters. We discuss possible connections between these simulations and experimental observations concerning CA assembly within HIV-1 and in vitro.     

The effect of context on the folding of β-hairpins

Small β-hairpin peptides have been widely used as models for the folding of β-sheets. But how applicable is the folding of such models to β-structure in larger proteins with conventional hydrophobic cores? Here we present multiple unfolding simulations of three such proteins that contain the WW domain double hairpin β-sheet motif: cold shock protein A (CspA), cold shock protein B (CspB) and glucose permease IIA domain. We compare the behavior of the free motif in solution and in the context of proteins of different size and architecture. Both Csp proteins lost contacts between the double-hairpin motif and the protein core as the first step of unfolding and proceeded to unfold with loss of the third β-strand, similar to the isolated WW domain. The glucose permease IIA domain is a larger protein and the contacts between the motif and the core were not lost as quickly. Instead the unfolding pathway of glucose permease IIA followed a different pathway with β1 pulling away from the sheet first. Interestingly, when the double hairpin motif was excised from the glucose permease IIA domain and simulated in isolation in water it unfolded by the same pathway as the WW domain, indicating that it is tertiary interactions with the protein that alter the motif’s unfolding not a sequence dependent effect on its intrinsic unfolding behavior. With respect to the unfolding of the hairpins, there was no consistent order to the loss of hydrogen bonds between the β-strands in the hairpins in any of the systems. Our results show that while the folding behavior of the isolated WW domain is generally consistent with the double hairpin motif’s behavior in the cold shock proteins, it is not the case for the glucose permease IIA domain. So, one must be cautious in extrapolating findings from model systems to larger more complicated proteins where tertiary interactions can overwhelm intrinsic behavior.     

Assembly of protein structure from sparse experimental data: An efficient Monte Carlo model

A new, efficient method for the assembly of protein tertiary structure from known, loosely encoded secondary structure restraints and sparse information about exact side chain contacts is proposed and evaluated. The method is based on a new, very simple method for the reduced modeling of protein structure and dynamics, where the protein is described as a lattice chain connecting side chain centers of mass rather than Cαs. The model has implicit built-in multibody correlations that simulate short- and long-range packing preferences, hydrogen bonding cooperativity and a mean force potential describing hydrophobic interactions. Due to the simplicity of the protein representation and definition of the model force field, the Monte Carlo algorithm is at least an order of magnitude faster than previously published Monte Carlo algorithms for structure assembly. In contrast to existing algorithms, the new method requires a smaller number of tertiary restraints for successful fold assembly; on average, one for every seven residues as compared to one for every four residues. For example, for smaller proteins such as the B domain of protein G, the resulting structures have a coordinate root mean square deviation (cRMSD), which is about 3 Å from the experimental structure; for myoglobin, structures whose backbone cRMSD is 4.3 Å are produced, and for a 247-residue TIM barrel, the cRMSD of the resulting folds is about 6 Å. As would be expected, increasing the number of tertiary restraints improves the accuracy of the assembled structures. The reliability and robustness of the new method should enable its routine application in model building protocols based on various (very sparse) experimentally derived structural restraints. Proteins 32:475–494, 1998. © 1998 Wiley-Liss, Inc.     

Implicit solvent models for flexible protein-protein docking by molecular dynamics simulation

The suitability of three implicit solvent models for flexible protein-protein docking by procedures using molecular dynamics simulation is investigated. The three models are (i) the generalized Born (GB) model implemented in the program AMBER6.0; (ii) a distance-dependent dielectric (DDD) model; and (iii) a surface area-dependent model that we have parameterized and call the NPSA model. This is a distance-dependent dielectric model modified by neutralizing the ionizable side-chains and adding a surface area-dependent solvation term. These solvent models were first tested in molecular dynamics simulations at 300 K of the native structures of barnase, barstar, segment B1 of protein G, and three WW domains. These protein structures display a range of secondary structure contents and stabilities. Then, to investigate the performance of the implicit solvent models in protein docking, molecular dynamics simulations of barnase/barstar complexation, as well as PIN1 WW domain/peptide complexation, were conducted, starting from separated unbound structures. The simulations show that the NPSA model has significant advantages over the DDD and GB models in maintaining the native structures of the proteins and providing more accurate docked complexes.     

Structural basis for controlling the dimerization and stability of the WW domains of an atypical subfamily

The second WW domain in mammalian Salvador protein (SAV1 WW2) is quite atypical, as it forms a beta-clam-like homodimer. The second WW domain in human MAGI1 (membrane associated guanylate kinase, WW and PDZ domain containing 1) (MAGI1 WW2) shares high sequence similarity with SAV1 WW2, suggesting comparable dimerization. However, an analytical ultracentrifugation study revealed that MAGI1 WW2 (Leu355-Pro390) chiefly exists as a monomer at low protein concentrations, with an association constant of 1.3 x 10(2) M(-1). We determined its solution structure, and a structural comparison with the dimeric SAV1 WW2 suggested that an Asp residue is crucial for the inhibition of the dimerization. The substitution of this acidic residue with Ser resulted in the dimerization of MAGI1 WW2. The spin-relaxation data suggested that the MAGI1 WW2 undergoes a dynamic process of transient dimerization that is limited by the charge repulsion. Additionally, we characterized a longer construct of this WW domain with a C-terminal extension (Leu355-Glu401), as the formation of an extra alpha-helix was predicted. An NMR structural determination confirmed the formation of an alpha-helix in the extended C-terminal region, which appears to be independent from the dimerization regulation. A thermal denaturation study revealed that the dimerized MAGI1 WW2 with the Asp-to-Ser mutation gained apparent stability in a protein concentration-dependent manner. A structural comparison between the two constructs with different lengths suggested that the formation of the C-terminal alpha-helix stabilized the global fold by facilitating contacts between the N-terminal linker region and the main body of the WW domain.     

Simulation methods for protein structure fluctuations

Three numerical techniques for generating thermally accessible configurations of globular proteins are considered; these techniques are the molecular dynamics method, the Metropolis Monte Carlo method, and a modified Monte Carlo method which takes account of the forces acting on the protein atoms. The molecular dynamics method is shown to be more efficient than either of the Monte Carlo methods. Because it may be necessary to use Monte Carlo methods in certain important types of sampling problems, the behavior of these methods is examined in some detail. It is found that an acceptance ratio close to 1/6 yields optimum efficiency for the Metropolis method, in contrast to what is often assumed. This result, together with the overall inefficiency of the Monte Carlo methods, appears to arise from the anisotropic forces acting on the protein atoms due to their covalent bonding. Possible ways of improving the Monte Carlo methods are suggested.     

Proteins wriggle

We propose an algorithmic strategy for improving the efficiency of Monte Carlo searches for the low-energy states of proteins. Our strategy is motivated by a model of how proteins alter their shapes. In our model, when proteins fold under physiological conditions, their backbone dihedral angles change synchronously in groups of four or more to avoid steric clashes and respect the kinematic conservation laws. They wriggle; they do not thrash. We describe a simple algorithm that can be used to incorporate wriggling in Monte Carlo simulations of protein folding. We have tested this wriggling algorithm against a code in which the dihedral angles are varied independently (thrashing). Our standard of success is the average root-mean-square distance (rmsd) between the alpha-carbons of the folding protein and those of its native structure. After 100,000 Monte Carlo sweeps, the relative decrease in the mean rmsd, as one switches from thrashing to wriggling, rises from 11% for the protein 3LZM with 164 amino acids (aa) to 40% for the protein 1A1S with 313 aa and 47% for the protein 16PK with 415 aa. These results suggest that wriggling is useful and that its utility increases with the size of the protein. One may implement wriggling on a parallel computer or a computer farm.     

Worm-like Ising model for protein mechanical unfolding under the effect of osmolytes

We show via single-molecule mechanical unfolding experiments that the osmolyte glycerol stabilizes the native state of the human cardiac I27 titin module against unfolding without shifting its unfolding transition state on the mechanical reaction coordinate. Taken together with similar findings on the immunoglobulin-binding domain of streptococcal protein G (GB1), these experimental results suggest that osmolytes act on proteins through a common mechanism that does not entail a shift of their unfolding transition state. We investigate the above common mechanism via an Ising-like model for protein mechanical unfolding that adds worm-like-chain behavior to a recent generalization of the Wako-Saitô-Muñoz-Eaton model with support for group-transfer free energies. The thermodynamics of the model are exactly solvable, while protein kinetics under mechanical tension can be simulated via Monte Carlo algorithms. Notably, our force-clamp and velocity-clamp simulations exhibit no shift in the position of the unfolding transition state of GB1 and I27 under the effect of various osmolytes. The excellent agreement between experiment and simulation strongly suggests that osmolytes do not assume a structural role at the mechanical unfolding transition state of proteins, acting instead by adjusting the solvent quality for the protein chain analyte.     

A single point mutation in a group I WW domain shifts its specificity to that of group II WW domains.

WW domains can be divided into three groups based on their binding specificity. By random mutagenesis, we switched the specificity of the Yes-associated protein (YAP) WW1 domain, a Group I WW domain, to that of the FE65 WW domain, which belongs to Group II. We showed that a single mutation, leucine 190 (betaB5) to tryptophan, is required to switch from Group I to Group II. Although this single substitution in YAP WW1 domain is sufficient to precipitate the two protein isoforms of Mena, an in vivo ligand of FE65, we showed that an additional substitution, histidine 192 (betaB7) to glycine, significantly increased the ability of YAP to mimic FE65. This double mutant (L190W/H192G) precipitates eight of the nine protein bands that FE65 pulls down from rat brain protein lysates. Based on both our data and a sequence comparison between Group I and Group II WW domains, we propose that a block of three consecutive aromatic amino acids within the second beta-sheet of the domain is required, but not always sufficient, for a WW domain to belong to Group II. These data deepen our understanding of WW domain binding specificity and provide a basis for the rational design of modified WW domains with potential therapeutic applications.