Folding dynamics of Trp-cage in the presence of chemical interference and macromolecular crowding. I

Proteins fold and function in the crowded environment of the cell's interior. In the recent years it has been well established that the so-called "macromolecular crowding" effect enhances the folding stability of proteins by destabilizing their unfolded states for selected proteins. On the other hand, chemical and thermal denaturation is often used in experiments as a tool to destabilize a protein by populating the unfolded states when probing its folding landscape and thermodynamic properties. However, little is known about the complicated effects of these synergistic perturbations acting on the kinetic properties of proteins, particularly when large structural fluctuations, such as protein folding, have been involved. In this study, we have first investigated the folding mechanism of Trp-cage dependent on urea concentration by coarse-grained molecular simulations where the impact of urea is implemented into an energy function of the side chain and/or backbone interactions derived from the all-atomistic molecular dynamics simulations with urea through a Boltzmann inversion method. In urea solution, the folding rates of a model miniprotein Trp-cage decrease and the folded state slightly swells due to a lack of contact formation between side chains at the terminal regions. In addition, the equilibrium m-values of Trp-cage from the computer simulations are in agreement with experimental measurements. We have further investigated the combined effects of urea denaturation and macromolecular crowding on Trp-cage's folding mechanism where crowding agents are modeled as hard-spheres. The enhancement of folding rates of Trp-cage is most pronounced by macromolecular crowding effect when the extended conformations of Trp-cast dominate at high urea concentration. Our study makes quantitatively testable predictions on protein folding dynamics in a complex environment involving both chemical denaturation and macromolecular crowding effects.     

Measuring pK(a) values in protein folding transition state ensembles by NMR spectroscopy

Protein folding kinetic data have been obtained for the marginally stable N-terminal SH3 domain of the Drosophila protein drk as a function of pH in order to investigate the electrostatic properties of Asp8 in the folding transition state ensemble. The slow exchange between folded and unfolded forms of the protein gives rise to separate NMR resonances for both folded and unfolded states at equilibrium. As a result, kinetic data can be derived from magnetization transfer between these two states without the need for denaturants. Using the fact that ionization of Asp8 dominates the electrostatic behavior of the protein between pH 2 and 3, along with pKa values for titrating groups in both folded and unfolded states that have been determined in a previous study, values of 2.9 +/- 0.1 and 3.3 +/- 0.2 are obtained for the pKa of Asp8 in the transition state for the wild-type protein and for a His7Ala mutant, respectively. The data are consistent with the partial formation in the transition state ensemble of an Asp8 side chain carboxylate-a Lys21 backbone amide interaction that represents a highly conserved contact in folded SH3 domains.     

Influence of water-protein hydrogen bonding on the stability of Trp-cage miniprotein. A comparison between the TIP3P and TIP4P-Ew water models

We report extensive replica exchange molecular dynamics (REMD) simulations on the folding/unfolding equilibrium of Trp-cage miniprotein using the Amber ff99SB all atom forcefield and TIP3P and TIP4P-Ew explicit water solvent models. REMD simulation-lengths in the 500 ns to the microsecond regime per replica are required to adequately sample the folding/unfolding equilibrium. We observe that this equilibrium is significantly affected by the choice of the water model. Compared with experimental data, simulations using the TIP3P solvent describe the stability of the Trp-cage quite realistically, providing a melting point which is just a few Kelvins above the experimental transition temperature of 317 K. The TIP4P-Ew model shifts the equilibrium towards the unfolded state and lowers the free energy of unfolding by about 3 kJ mol(-1) at 280 K, demonstrating the need to fine-tune the protein-forcefield depending on the chosen water model. We report evidence that the main difference between the two water models is mostly due to the different solvation of polar groups of the peptide. The unfolded state of the Trp-cage is stabilized by an increasing number of hydrogen bonds, destabilizing the α-helical part of the molecule and opening the R-D salt bridge. By reweighting the strength of solvent-peptide hydrogen bonds by adding a hydrogen bond square well potential, we can fully recover the effect of the different water models and estimate the shift in population as due to a difference in hydrogen bond-strength of about 0.4 kJ mol(-1) per hydrogen bond.     

Computational design of thermostabilizing D-amino acid substitutions

Judicious incorporation of D-amino acids in engineered proteins confers many advantages such as preventing degradation by endogenous proteases and promoting novel structures and functions not accessible to homochiral polypeptides. Glycine to D-alanine substitutions at the carboxy termini can stabilize α-helices by reducing conformational entropy. Beyond alanine, we propose additional side chain effects on the degree of stabilization conferred by D-amino acid substitutions. A detailed, molecular understanding of backbone and side chain interactions is important for developing rational, broadly applicable strategies in using D-amino acids to increase protein thermostability. Insight from structural bioinformatics combined with computational protein design can successfully guide the selection of stabilizing D-amino acid mutations. Substituting a key glycine in the Trp-cage miniprotein with D-Gln dramatically stabilizes the fold without altering the protein backbone. Stabilities of individual substitutions can be understood in terms of the balance of intramolecular forces both at the α-helix C-terminus and throughout the protein.     

Contribution of charged groups to the enthalpic stabilization of the folded states of globular proteins

In this paper we use the results from all-atom molecular dynamics (MD) simulations of proteins and peptides to assess the individual contribution of charged atomic groups to the enthalpic stability of the native state of globular proteins and investigate how the distribution of charged atomic groups in terms of solvent accessibility relates to protein enthalpic stability. The contributions of charged groups is calculated using a comparison of nonbonded interaction energy terms from equilibrium simulations of charged amino acid dipeptides in water (the "unfolded state") and charged amino acids in globular proteins (the "folded state"). Contrary to expectation, the analysis shows that many buried, charged atomic groups contribute favorably to protein enthalpic stability. The strongest enthalpic contributions favoring the folded state come from the carboxylate (COO(-)) groups of either Glu or Asp. The contributions from Arg guanidinium groups are generally somewhat stabilizing, while N(+)(3) groups from Lys contribute little toward stabilizing the folded state. The average enthalpic gain due to the transfer of a methyl group in an apolar amino acid from solution to the protein interior is described for comparison. Notably, charged groups that are less exposed to solvent contribute more favorably to protein native-state enthalpic stability than charged groups that are solvent exposed. While solvent reorganization/release has favorable contributions to folding for all charged atomic groups, the variation in folded state stability among proteins comes mainly from the change in the nonbonded interaction energy of charged groups between the unfolded and folded states. A key outcome is that the calculated enthalpic stabilization is found to be inversely proportional to the excess charge density on the surface, in support of an hypothesis proposed previously.     

Measurement of side-chain carboxyl pK(a) values of glutamate and aspartate residues in an unfolded protein by multinuclear NMR spectroscopy

A triple-resonance NMR pulse scheme is presented for measuring aspartic and glutamic acid side-chain pK(a) values in unfolded protein states where chemical shift overlap is limiting. The experiment correlates side-chain carboxyl carbon chemical shifts of these residues with the backbone amide proton chemical shift of the following residue. The methodology is applied to an (15)N, (13)C labeled sample of the N-terminal SH3 domain of the Drosophila protein drk, which exists in equilibrium between folded (F(exch)) and unfolded (U(exch)) states under nondenaturing conditions. Residue-specific pK(a) values of side-chain carboxyl groups are presented for the first time for an unfolded protein (drk U(exch) state), determined from a pH titration. Results indicate that deviations from pK(a) values measured for model compounds are likely due to local effects, while long-range electrostatic interactions appear to be of minor importance for this protein.     

UV-resonance raman thermal unfolding study of Trp-cage shows that it is not a simple two-state miniprotein

Trp-cage, a synthetic 20 residue polypeptide, is proposed to be an ultrafast folding synthetic miniprotein which utilizes tertiary contacts to define its native conformation. We utilized UV resonance Raman spectroscopy (UVRS) with 204 and 229 nm excitation to follow its thermal melting. Our results indicate that Trp-cage melting is complex, and it is not a simple two-state process. Using 204 nm excitation we probe the peptide secondary structure and find the Trp-cage's alpha-helix shows a broad melting curve where on average four alpha-helical amide bonds melt upon a temperature increase from 4 to 70 degrees C. Using 229 nm excitation we probe the environment of the Trp side chain and find that its immediate environment becomes more compact as the temperature is increased from 4 to 20 degrees C; however, further temperature increases lead to exposure of the Trp to water. The chi(2) angle of the Trp side chain remains invariant throughout the entire temperature range. Previous kinetic results indicated a single-exponential decay in the 4-70 degrees C temperature range, suggesting that Trp-cage behaves as a two-state folder. However, this miniprotein does not show clear two-state behavior in our steady-state studies. Rather it shows a continuous distribution of steady-state spectral parameters. Only the alpha-helix melting curve even hints of a cooperative transition. Possibly, the previous kinetic results monitor only a small region of the Trp-cage which locally appears two-state. This would then argue for spatially decoupled folding even for this small peptide.     

Simulation studies of protein folding/unfolding equilibrium under polar and nonpolar confinement

We study the equilibrium folding/unfolding thermodynamics of a small globular miniprotein, the Trp cage, that is confined to the interior of a 2 nm radius fullerene ball. The interactions of the fullerene surface are changed from nonpolar to polar to mimic the interior of the GroEL/ES chaperonin that assists proteins to fold in vivo. We find that nonpolar confinement stabilizes the folded state of the protein due to the effects of volume reduction that destabilize the unfolded state and also due to interactions with the fullerene surface. For the Trp cage, polar confinement has a net destabilizing effect that results from the stabilizing confinement and the competitive exclusion effect that keeps the protein away from the surface hydration shell and stronger interactions between charged side chains in the protein and the polar surface that compete against the formation of an ion pair that stabilizes the protein folded state. We show that confinement effects due to volume reduction can be overcome by sequence-specific interactions of the protein side chains with the encapsulating surface. This study shows that there is a complex balance among many competing effects that determine the mechanism of GroEL chaperonin in enhancing the folding rate of polypeptide inside its cavity.     

Insights into the phosphoryl-transfer mechanism of cAMP-dependent protein kinase from quantum chemical calculations and molecular dynamics simulations

To investigate the molecular details of the phosphoryl-transfer mechanism catalyzed by cAMP-dependent protein kinase, we performed quantum mechanical (QM) calculations on a cluster model of the active site and molecular dynamics (MD) simulations of a ternary complex of the protein with Mg(2)ATP and a 20-residue peptide substrate. Overall, our theoretical results confirm the participation of the conserved aspartic acid, Asp(166), as an acid/base catalyst in the reaction mechanism catalyzed by protein kinases. The MD simulation shows that the contact between the nucleophilic serine side chain and the carboxylate group of Asp(166) is short and dynamically stable, whereas the QM study indicates that an Asp(166)-assisted pathway is structurally and energetically feasible and is in agreement with previous experimental results.     

Primary and secondary locations of charge sites in angiotensin II (M + 2H)2+ ions formed by electrospray ionization

High-energy tandem mass spectrometry and molecular dynamics calculations are used to determine the locations of charge in metastably decomposing (M + 2H)2+ ions of human angiotensin II. Charge-separation reactions provide critical information regarding charge sites in multiple charged ions. The most probable kinetic energy released (Tm.p.) from these decompositions are obtained using kinetic energy release distributions (KERDs) in conjunction with MS/MS (MS2), MS/MS/MS (MS3), and MS/MS/MS/MS (MS4) experiments. The most abundant singly and doubly charged product ions arise from precursor ion structures in which one proton is located on the arginine (Arg) side chain and the other proton is located on a distal peptide backbone carbonyl oxygen. The MS3 KERD experiments show unequivocally that neither the N-terminal amine nor the aspartic acid (Asp) side chain are sites of protonation. In the gas phase, protonation of the less basic peptide backbone instead of the more proximal and basic histidine (His) side chain is favored as a result of reduced coulomb repulsion between the two charge sites. The singly and doubly charged product ions of lesser abundance arise from precursor ion structures in which one proton is located on the Arg side chain and the other on the His side chain. This is demonstrated in the MS3 and MS4 mass-analyzed ion kinetic energy spectrometry experiments. Interestingly, (b7" + OH)2+ product ions, like the (M + 2H)2+ ions of angiotensin II, are observed to have at least two different decomposing structures in which charge sites have a primary and secondary location.     

Side chain dynamics in unfolded protein states: an NMR based 2H spin relaxation study of delta131delta

NMR relaxation data on disordered proteins can provide insight into both structural and dynamic properties of these molecules. Because of chemical shift degeneracy in correlation spectra, detailed site-specific analyses of side chain dynamics have not been possible. Here, we present new experiments for the measurement of side chain dynamics in methyl-containing residues in unfolded protein states. The pulse schemes are similar to recently proposed methods for measuring deuterium spin relaxation rates in (13)CH(2)D methyl groups in folded proteins.(1) However, because resolution in (1)H-(13)C correlation maps of unfolded proteins is limiting, relaxation data are recorded as a series of (1)H-(15)N spectra. The methodology is illustrated with an application to the study of side chain dynamics in delta131delta, a large disordered fragment of staphylococcal nuclease containing residues 1-3 and 13-140 of the wide-type protein. A good correlation between the order parameters of the symmetry axes of the methyl groups and the backbone (1)H-(15)N bond vectors of the same residue is observed. Simulations establish that such a correlation is only possible if the unfolded state is comprised of an ensemble of structures which are not equiprobable. A motional model, which combines wobbling-in-a-cone and Gaussian axial fluctuations, is proposed to estimate chi(1) torsion angle fluctuations, sigma(chi)()1, of Val and Thr residues on the basis of the backbone and side chain order parameters. Values of sigma(chi)()1 are approximately 10 degrees larger than what has previously been observed in folded proteins. Of interest, the value of sigma(chi)()1 for Val 104 is considerably smaller than for other Val or Thr residues, suggesting that it may be part of a hydrophobic cluster. Notably large (15)N transverse relaxation rates are observed in this region. To our knowledge, this is the first time that side chain dynamics in an unfolded state have been studied in detail by NMR.     

The dehydroalanine effect in the fragmentation of ions derived from polypeptides

The fragmentation of peptides and proteins upon collision‐induced dissociation (CID) is highly dependent on sequence and ion type (e.g. protonated, deprotonated, sodiated, odd electron, etc.). Some amino acids, for example aspartic acid and proline, have been found to enhance certain cleavages along the backbone. Here, we show that peptides and proteins containing dehydroalanine, a non‐proteinogenic amino acid with an unsaturated side‐chain, undergo enhanced cleavage of the N—Cα bond of the dehydroalanine residue to generate c‐ and z‐ions. Because these fragment ion types are not commonly observed upon activation of positively charged even‐electron species, they can be used to identify dehydroalanine residues and localize them within the peptide or protein chain. While dehydroalanine can be generated in solution, it can also be generated in the gas phase upon CID of various species. Oxidized S‐alkyl cysteine residues generate dehydroalanine upon activation via highly efficient loss of the alkyl sulfenic acid. Asymmetric cleavage of disulfide bonds upon collisional activation of systems with limited proton mobility also generates dehydroalanine. Furthermore, we show that gas‐phase ion/ion reactions can be used to facilitate the generation of dehydroalanine residues via, for example, oxidation of S‐alkyl cysteine residues and conversion of multiply‐protonated peptides to radical cations. In the latter case, loss of radical side‐chains to generate dehydroalanine from some amino acids gives rise to the possibility for residue‐specific backbone cleavage of polypeptide ions. Copyright © 2016 John Wiley & Sons, Ltd.     

Exploration of the secondary structure specific differential solvation dynamics between the native and molten globule states of the protein HP-36

Recent experiments have shown that the time dependence of fluorescence Stokes shift of a chromophore is substantially different when the chromophore is located in a molten globule (MG) state and in the native state of the same protein. To understand the origin of this difference, particularly the role of water in the differential solvation of the protein in the native and the MG states, we have carried out fully atomistic molecular dynamics simulations with explicit water of a partially unfolded MG state of the protein HP-36 and compared the results with the solvation dynamics of the protein in the folded native state. It is observed that the polar solvation dynamics of the three helical segments of the protein is influenced in a nonuniform heterogeneous manner in the MG state. While the equilibrium solvation time correlation function for helix-3 has been found to relax faster in the MG state as compared to that in the native state, the decay of the corresponding function for the other two helices slows down in the MG state. A careful analysis shows that the origin of such heterogeneous relative solvation behavior lies in the differential location of the polar probe residues and their exposure to bulk solvent. We find a significant negative cross-correlation between the contribution (to the solvation energy of a tagged amino acid residue) of water and the other groups of the protein, indicating a competing role in solvation. The sensitivity of solvation dynamics to the secondary structure and the immediate environment can be used to discriminate the partially unfolded and folded states. These results therefore should be useful in explaining recent solvation dynamics experiments on native and MG states of proteins.     

A quadrupole/time-of-flight mass spectrometry study of trp-cage’s conformation

Trp-cage is a synthetic 20-residue miniprotein that uses tertiary contacts to stabilize its native conformation. NMR, circular dichroism (CD), and UV-resonance Raman spectroscopy were used to probe its energy landscape. In this quadrupole/time-of-flight study, electrospray ionization charge state distribution (CSD) and solution-phase H/D exchange are used to probe Trp-cage's tertiary structure. The CSDs of Trp-cage and its mutant provide spectra showing a pH-dependent conformation change. Solution-phase H/D exchange in 30% deuterated trifluoroethanol solution of the wild type shows increased protection of one labile hydrogen in the native state. Together, CSDs and solution-phase H/D exchange are demonstrated to constitute a simple but effective means to follow conformation changes in a small tertiary protein.     

The effect of neighboring charges on the helix forming ability of charged amino acids in proteins.

It has been found that the fraction of glutamic acid residues which are helical in proteins is larger than might be expected from the experimentally determined value of the helical stability constants of glutamic acid. In order to understand this difference, the effect of neighboring charged side chains on the glutamic acid residues in proteins of known structure is examined. It is found that a positively charged side chain four residues away from a glutamic acid greatly enhances its probability to be helical. Similar results are obtained for aspartic acid, lysine, arginine, and histidine.     

Fluorescence probe of Trp-cage protein conformation in solution and in gas phase

Measurements of protein unfolding in the absence of solvent, when combined with unfolding studies in solution, offer a unique opportunity to measure the effects of solvent on protein structure and dynamics. The experiments presented here rely on the fluorescence of an attached dye to probe the local conformational dynamics through interactions with a Trp residue and fields originating on charge sites. We present fluorescence measurements of thermal fluctuations accompanying conformational change of a miniprotein, Trp-cage, in solution and in gas phase. Molecular dynamics (MD) simulations are performed as a function of temperature, charge state, and charge location to elucidate the dye-protein conformational dynamics leading to the changes in measured fluorescence. The results indicate that the stability of the unsolvated protein is dominated by hydrogen bonds. Substituting asparagine for aspartic acid at position 9 results in a dramatic alteration of the solution unfolding curve, indicating that the salt bridge involving Lys8, Asp9, and Arg16 (+ - +) is essential for Trp-cage stability in solution. In contrast, this substitution results in minor changes in the unfolding curve of the unsolvated protein, showing that hydrogen bonds are the major contributor to the stability of Trp-cage in gas phase. Consistent with this hypothesis, the decrease in the number of hydrogen bonds with increasing temperature indicated by MD simulations agrees reasonably well with the experimentally derived enthalpies of conformational change. The simulation results display relatively compact conformations compared with NMR structures that are generally consistent with experimental results. The measured unfolding curves of unsolvated Trp-cage ions are invariant with the acetonitrile content of the solution from which they are formed, possibly as a result of conformational relaxation during or after desolvation. This work demonstrates the power of combined solution and gas-phase studies and of single-point mutations to identify specific noncovalent interactions which contribute to protein-fold stability. The combination of experiment and simulation is particularly useful because these approaches yield complementary information which can be used to deduce the details of structural changes of proteins in the gas phase.     

Potential of mean force and pKa profile calculation for a lipid membrane-exposed arginine side chain

The issue of ionizable protein side chains interacting with lipid membranes has been the focus of much attention since the proposal of the paddle model of voltage-gated ion channels, which suggested multiple arginine (Arg) side chains may move through the hydrocarbon core of a lipid membrane. Recent cell biology experiments have also been interpreted to suggest that these side chains would face only small free energy penalties to cross membranes, challenging a long-standing view in membrane biophysics. Here, we employ side chain analog and transmembrane helix models to determine the free energy of an Arg side chain, as a function of protonation state, across a membrane. We observe high free energy barriers for both the charged and neutral states that would prohibit lipid-exposed movement. The mechanisms for charged and neutral Arg transport are, however, very different, with the neutral state experiencing simple dehydration, whereas the charged state experiences a complex mechanism involving connections to the bilayer interfaces that deform the local membrane structure. We employ special methods to ensure sampling of these interfacial connections and decompose the free energy to shed light on the mechanisms. These deformations are found to preferentially stabilize the protonated form, such that the Arg side chain remains almost exclusively charged inside the membrane, with a pKa shift of 相似文献   

Characterizing the protonation states of the catalytic residues in apo and substrate-bound human T-cell leukemia virus type 1 protease

Human T-cell leukemia virus type 1 (HTLV-1) protease is an attractive target when developing inhibitors to treat HTLV-1 associated diseases. To study the catalytic mechanism and design novel HTLV-1 protease inhibitors, the protonation states of the two catalytic aspartic acid residues must be determined. Free energy simulations have been conducted to study the proton transfer reaction between the catalytic residues of HTLV-1 protease using a combined quantum mechanical and molecular mechanical (QM/MM) molecular dynamics simulation. The free energy profiles for the reaction in the apo-enzyme and in an enzyme – substrate complex have been obtained. In the apo-enzyme, the two catalytic residues are chemically equivalent and are expected to be both unprotonated. Upon substrate binding, the catalytic residues of HTLV-1 protease evolve to a singly protonated state, in which the OD1 of Asp32 is protonated and forms a hydrogen bond with the OD1 of Asp32′, which is unprotonated. The HTLV-1 protease–substrate complex structure obtained from this simulation can serve as the Michaelis complex structure for further mechanistic studies of HTLV-1 protease while providing a receptor structure with the correct protonation states for the active site residues toward the design of novel HTLV-1 protease inhibitors through virtual screening.     

Origin and Prediction of Highly Specific Bond Cleavage Sites in the Thermal Activation of Intact Protein Ions

Predicting the fragmentation patterns of proteins would be beneficial for the reliable identification of intact proteins by mass spectrometry. However, the ability to accurately make such predictions remains elusive. An approach to predict the specific cleavage sites in whole proteins resulting from collision-induced dissociation by use of an improved electrostatic model for calculating the proton configurations of highly-charged protein ions is reported. Using ubiquitin, cytochrome c, lysozyme and β-lactoglobulin as prototypical proteins, this approach can be used to predict the fragmentation patterns of intact proteins. For sufficiently highly charged proteins, specific cleavages occur near the first low-basicity amino acid residues that are protonated with increasing charge state. Hybrid QM/QM′ (QM=quantum mechanics) and molecular dynamics (MD) simulations and energy-resolved collision-induced dissociation measurements indicated that the barrier to the specific dissociation of the protonated amide backbone bond is significantly lower than competitive charge remote fragmentation. Unlike highly charged peptides, the protons at low-basicity sites in highly charged protein ions can be confined to a limited sequence of low-basicity amino acid residues by electrostatic repulsion, which results in highly specific fragmentation near the site of protonation. This research suggests that the optimal charge states to form specific sequence ions of intact proteins in higher abundances than the use of less specific ion dissociation methods can be predicted a priori.     

Probing electrostatic interactions and structural changes in highly charged protein polyanions by conformer-selective photoelectron spectroscopy

We have recorded the first conformer-selective photoelectron spectra of a protein polyanion in the gas-phase. Bovine cytochrome c protein was studied in 8 different negative charge states ranging from 5- to 12-. Electron binding energies were extracted for all charge states and used as a direct probe of intramolecular Coulomb repulsion. Comparison of experimental results with simulations shows that the experimental outcome can be reproduced with a simple electrostatic model. Energetics are consistent with a structural transition from a folded to an unfolded conformational state of the protein as the number of charges increases. Furthermore, the additional ion-mobility data show that the onset of unfolding can be assigned to charge state 6- where three conformers can be distinguished.