Multiscale simulations of protein folding: application to formation of secondary structures

A multiscale simulation method of protein folding is proposed, using atomic representation of protein and solvent, combing genetic algorithms to determine the key protein structures from a global view, with molecular dynamic simulations to reveal the local folding pathways, thus providing an integrated landscape of protein folding. The method is found to be superior to previously investigated global search algorithms or dynamic simulations alone. For secondary structure formation of a selected peptide, RN24, the structures and dynamics produced by this method agree well with corresponding experimental results. Three most populated conformations are observed, including hairpin, β-sheet and α-helix. The energetic barriers separating these three structures are comparable to the kinetic energy of the atoms of the peptide, implying that the transition between these states can be easily triggered by kinetic perturbations, mainly through electrostatic interactions between charged atoms. Transitions between α-helix and β-sheet should jump over at least two energy barriers and may stay in the energetic trap of hairpin. It is proposed that the structure of proteins should be jointly governed by thermodynamic and dynamic factors; free energy is not the exclusive dominant for stability of proteins.     

Folding type-specific secondary structure propensities of amino acids,derived from α-helical, β-sheet, α/β, and α+β proteins of known structures

Folding type-specific secondary structure propensities of 20 naturally occurring amino acids have been derived from α-helical, β-sheet, α/β, and α+β proteins of known structures. These data show that each residue type of amino acids has intrinsic propensities in different regions of secondary structures for different folding types of proteins. Each of the folding types shows markedly different rank ordering, indicating folding type-specific effects on the secondary structure propensities of amino acids. Rigorous statistical tests have been made to validate the folding type-specific effects. It should be noted that α and β proteins have relatively small α-helices and β-strands forming propensities respectively compared with those of α+β and α/β proteins. This may suggest that, with more complex architectures than α and β proteins, α+β and α/β proteins require larger propensities to distinguish from interacting α-helices and β-strands. Our finding of folding type-specific secondary structure propensities suggests that sequence space accessible to each folding type may have differing features. Differing sequence space features might be constrained by topological requirement for each of the folding types. Almost all strong β-sheet forming residues are hydrophobic in character regardless of folding types, thus suggesting the hydrophobicities of side chains as a key determinant of β-sheet structures. In contrast, conformational entropy of side chains is a major determinant of the helical propensities of amino acids, although other interactions such as hydrophobicities and charged interactions cannot be neglected. These results will be helpful to protein design, class-based secondary structure prediction, and protein folding. © 1998 John Wiley & Sons, Inc. Biopoly 45: 35–49, 1998     

Folding simulations of three proteins having all α-helix,all β-strand and α/β-structures

We performed folding simulations of three proteins using four force fields, AMBER parm96, AMBER parm99, CHARMM 27 and OPLS-AA/L, in order to examine the features of these force fields. We studied three proteins, protein A (all α-helix), cold-shock protein (all β-strand) and protein G (α/β-structures), for the folding simulations. For the simulation, we used the simulated annealing molecular dynamics method, which was performed 50 times for each protein using the four force fields. The results showed that the secondary-structure-forming tendencies are largely different among the four force fields. AMBER parm96 favours β-bridge structures and extended β-strand structures, and AMBER parm99 favours α-helix structures and 3₁₀-helix structures. CHARMM 27 slightly favours α-helix structures, and there are also π-helix and β-bridge structures. OPLS-AA/L favours α-helix structures and 3₁₀-helix structures.     

Folding Simulations of a De Novo Designed Protein with a βαβ Fold

βαβ structural motifs are commonly used building blocks in protein structures containing parallel β-sheets. However, to our knowledge, no stand-alone βαβ structure has been observed in nature to date. Recently, for the first time that we know of, a small protein with an independent βαβ structure (DS119) was successfully designed in our laboratory. To understand the folding mechanism of DS119, in the study described here, we carried out all-atom molecular dynamics and coarse-grained simulations to investigate its folding pathways and energy landscape. From all-atom simulations, we successfully observed the folding event and got a stable folded structure with a minimal root mean-square deviation of 2.6 Å with respect to the NMR structure. The folding process can be described as a fast collapse phase followed by rapid formation of the central helix, and then slow formation of a parallel β-sheet. By using a native-centric Gō-like model, the cooperativity of the system was characterized in terms of the calorimetric criterion, sigmoidal transitions, conformation distribution shifts, and free-energy profiles. DS119 was found to be an incipient downhill folder that folds more cooperatively than a downhill folder, but less cooperatively than a two-state folder. This may reflect the balance between the two structural elements of DS119: the rapidly formed α-helix and the slowly formed parallel β-sheet. Folding times estimated from both the all-atom simulations and the coarse-grained model were at microsecond level, making DS119 another fast folder. Compared to fast folders reported previously, DS119 is, to the best of our knowledge, the first that exhibits a parallel β-sheet.     

Aggregation of amphipathic peptides at an aqueous–organic interface using coarse-grained simulations

The effect of an aqueous/organic interface on the folding and aggregation of amphipathic peptides is examined by applying discontinuous molecular dynamics (DMD) simulations combined with an intermediate resolution protein model, PRIME20, to a peptide/interface system. The systems contain 48 (KLLK)₄ peptides in random coil or α-helical conformations interacting with both strong and weak interfaces. In the absence of an interface, most of the oligomers form helical bundles, a small fraction of which convert to β-sheets when the temperature is above the folding transition. Adding a weak interface decreases oligomer formation above the folding temperature and increases it below. Little monolayer formation is observed at the weak interface; instead reversible adsorption increases the local peptide concentration near the interface, promoting helical bundle formation in the aqueous phase below the folding temperature and β-sheet formation above the folding temperature. Introducing a strong interface leads to irreversible adsorption, promoting formation of helical monolayers below the folding temperature and mixed β-sheet/amorphous monolayers above the folding temperature. The (KLLK)₄ peptide is more likely to adsorb to the interface when it is in an α-helical conformation, as opposed to a random coil, because of its larger hydrophobic moment.     

Evolution rescues folding of human immunodeficiency virus-1 envelope glycoprotein GP120 lacking a conserved disulfide bond

The majority of eukaryotic secretory and membrane proteins contain disulfide bonds, which are strongly conserved within protein families because of their crucial role in folding or function. The exact role of these disulfide bonds during folding is unclear. Using virus-driven evolution we generated a viral glycoprotein variant, which is functional despite the lack of an absolutely conserved disulfide bond that links two antiparallel β-strands in a six-stranded β-barrel. Molecular dynamics simulations revealed that improved hydrogen bonding and side chain packing led to stabilization of the β-barrel fold, implying that β-sheet preference codirects glycoprotein folding in vivo. Our results show that the interactions between two β-strands that are important for the formation and/or integrity of the β-barrel can be supported by either a disulfide bond or β-sheet favoring residues.     

Fast in silico protein folding by introduction of alternating hydrogen bond potentials

We accelerate protein folding in all-atom molecular dynamics simulations by introducing alternating hydrogen bond potentials as a supplement to the force field. The alternating hydrogen bond potentials result in accelerated hydrogen bond reordering, which leads to rapid formation of secondary structure elements. The method does not require knowledge of the native state but generates the potentials based on the development of the tertiary structure in the simulation. In protein folding, the formation of secondary structure elements, especially α-helix and β-sheet, is very important, and we show that our method can fold both efficiently and with great speed.     

The surface of β-sheet proteins contains amphiphilic regions which may provide clues about protein folding

A major bottleneck in the field of biochemistry is our limited understanding of the processes by which a protein folds into its native conformation. Much of the work on this issue has focused on the conserved core of the folded protein. However, one might imagine that a ubiquitous motif for unaided folding or for the recognition of chaperones may involve regions on the surface of the native structure. We explore this possibility by an analysis of the spatial distribution of regions with amphiphilic α-helical potential on the surface of β-sheet proteins. All proteins, Including β-sheet proteins, contain regions with amphiphilic α-helical potential. That is, any α-helix formed by that region would be amphiphilic, having both hydrophobic and hydrophilic surfaces. In the three-dimensional structure of all β-sheet proteins analyzed, we have found a distinct pattern in the spatial distribution of sequences with amphiphilic α-helical potential. The amphiphilic regions occur in ring shaped clusters approximately 20 to 30 Å in diameter on the surface of the protein. In addition, these regions have a strong preference for positively charged amino acids and a lower preference for residues not favorable to α-helix formation. Although the purpose of these amphiphilic regions which are not associated with naturally occurring α-helix is unknown, they may play a critical role in highly conserved processes such as protein folding. © 1996 Wiley-Liss, Inc.     

Analysis and prediction of the packing of α-helices against a β-sheet in the tertiary structure of globular proteins

The packing of α-helices and β-sheets in six $\text{α}\text{β}$ proteins (e.g. flavodoxin) has been analysed. The results provide the basis for a computer algorithm to predict the tertiary structure of an $\text{α}\text{β}$ protein from its amino acid sequence and actual assignment of secondary structure.The packing of an individual α-helix against a β-sheet generally involves two adjacent ± 4 rows of non-polar residues on the α-helix at the positions i, i + 4, i + 8, i + 1, i + 5, i + 9. The pattern of interacting β-sheet residues results from the twisted nature of the sheet surface and the attendant rotation of the side-chains. At a more detailed level, four of the α-helical residues (i + 1, i + 4, i + 5 and i + 8) form a diamond that surrounds one particular β-sheet residue, generally isoleucine, leucine or valine. In general, the α-helix sits 10 Å above the sheet and lies parallel to the strand direction.The prediction follows a combinational approach. First, a list of possible β-sheet structures (10⁶ to 10¹⁴) is constructed by the generation of all β-sheet topologies and β-strand alignments. This list is reduced by constraints on topology and the location of non-polar residues to mediate the sheet/helix packing, and then rank-ordered on the extent of hydrogen bonding. This algorithm was uniformly applied to 16 $\text{α}\text{β}$ domains in 13 proteins. For every structure, one member of the reduced list was close to the crystal structure; the root-mean-square deviation between equivalenced C^α atoms averaged 5.6 Å for 100 residues. For the $\text{α}\text{β}$ proteins with pure parallel β-sheets, the total number of structures comparable to or better than the native in terms of hydrogen bonds was between 1 and 148. For proteins with mixed β-sheets, the worst case is glyceraldehyde-3-phosphate dehydrogenase, where as many as 3800 structures would have to be sampled. The evolutionary significance of these results as well as the potential use of a combinatorial approach to the protein folding problem are discussed.     

Secondary structure perturbations in salt-induced protein precipitates

The secondary structure implications of precipitation induced by a chaotropic salt, KSCN, and a structure stabilizing salt, Na₂SO₄, were studied for twelve different proteins. α-helix and β-sheet content of precipitate and native structures were estimated from the analysis of amide I band Raman spectra. A statistical analysis of the estimated perturbations in the secondary structure contents indicated that the most significant event is the formation of β-sheet structures with a concomitant loss of α-helix on precipitation with KSCN. The conformational changes for each protein were also analyzed with respect to elements of primary, secondary and tertiary structure existing in the native protein; primary structure was quantified by the fractions of hydrophobic and charged amino acids, secondary structure by x-ray estimates of α-helix and β-sheet contents of native proteins and tertiary structure by the dipole moment and solvent-accessible surface area. For the KSCN precipitates, factors affecting β-sheet content included the fraction of charged amino acids in the primary sequence and the surface area. Changes in α-helix content were influenced by the initial helical content and the dipole moment. The enhanced β-sheet contents of precipitates observed in this work parallel protein structural changes occurring in other aggregative phenomena.     

Determination method of the balance of the secondary-structure-forming tendencies of force fields

We propose a new method of optimisation of backbone torsion-energy parameters in the force field for molecular simulations of protein systems. This method is based on the idea of balancing the secondary-structure-forming tendencies, namely, those of α-helix and β-sheet structures. We perform a minimisation of the backbone dihedral angle-based root-mean-square deviation of the helix and β structure regions in many protein structures. As an example, we optimised the backbone torsion-energy parameters of AMBER parm96 force field using 100 protein molecules from the Protein Data Bank. We then performed folding simulations of α-helical and β-hairpin peptides, using the optimised force field. The results imply that the new force-field parameters give structures more consistent with the experimental implications than the original AMBER parm96 force field.     

1H, 13C and 15N backbone resonance assignments for the BS3 class A β-lactamase from Bacillus licheniformis

Class A β-lactamases (260–280 amino acids; M _r  ~ 29,000) are among the largest proteins studied in term of their folding properties. They are composed of two structural domains: an all-α domain formed by five to eight helices and an α/β domain consisting of a five-stranded antiparallel β-sheet covered by three to four α-helices. The α domain (~150 residues) is made up of the central part of the polypeptide chain whereas the α/β domain (111–135 residues) is constituted by the N- and C-termini of the protein. Our goal is to determine in which order the different secondary structure elements are formed during the folding of BS3. With this aim, we will use pulse-labelling hydrogen/deuterium exchange experiments, in combination with 2D-NMR measurements, to monitor the time-course of formation and stabilization of secondary structure elements. Here we report the backbone resonance assignments as the requirement for further hydrogen/deuterium exchange studies.     

Computational and Experimental Studies on β-Sheet Breakers Targeting Aβ1–40 Fibrils

In this work we present and compare the results of extensive molecular dynamics simulations of model systems comprising an Aβ_1–40 peptide in water in interaction with short peptides (β-sheet breakers) mimicking the 17–21 region of the Aβ_1–40 sequence. Various systems differing in the customized β-sheet breaker structure have been studied. Specifically we have considered three kinds of β-sheet breakers, namely Ac-LPFFD-NH₂ and two variants thereof, one obtained by substituting the acetyl group with the sulfonic amino acid taurine (Tau-LPFFD-NH₂) and a second novel one in which the aspartic acid is substituted by an asparagine (Ac-LPFFN-NH₂). Thioflavin T fluorescence, circular dichroism, and mass spectrometry experiments have been performed indicating that β-sheet breakers are able to inhibit in vitro fibril formation and prevent the β sheet folding of portions of the Aβ_1–40 peptide. We show that molecular dynamics simulations and far UV circular dichroism provide consistent evidence that the new Ac-LPFFN-NH₂ β-sheet breaker is more effective than the other two in stabilizing the native α-helix structure of Aβ_1–40. In agreement with these results thioflavin T fluorescence experiments confirm the higher efficiency in inhibiting Aβ_1–40 aggregation. Furthermore, mass spectrometry data and molecular dynamics simulations consistently identified the 17–21 Aβ_1–40 portion as the location of the interaction region between peptide and the Ac-LPFFN-NH₂ β-sheet breaker.     

Insights into thermal resistance of proteins from the intrinsic stability of their α-helices

To investigate the role of α helices in protein thermostability, we compared energy characteristics of α helices from thermophilic and mesophilic proteins belonging to four protein families of known three-dimensional structure, for at least one member of each family. The changes in intrinsic free energy of α-helix formation were estimated using the statistical mechanical theory for describing helix/coil transitions in peptide helices [Munoz, V., Serrano, L. Nature Struc. Biol. 1:399–409, 1994; Munoz, V., Serrano, L. J. Mol. Biol. 245:275–296, 1995; Munoz, V., Serrano, L. J. Mol. Biol. 245:297–308, 1995]. Based on known sequences of mesophilic and thermophilic RecA proteins we found that (1) a high stability of α helices is necessary but is not a sufficient condition for thermostability of RecA proteins, (2) the total helix stability, rather than that of individual helices, is the factor determining protein thermostability, and (3) two facets of intrahelical interactions, the intrinsic helical propensities of amino acids and the side chain–side chain interactions, are the main contributors to protein thermostability. Similar analysis applied to families of L-lactate dehydrogenases, seryl-tRNA synthetases, and aspartate amino transferases led us to conclude that an enhanced total stability of α helices is a general feature of many thermophilic proteins. The magnitude of the observed decrease in intrinsic free energy on α-helix formation of several thermoresistant proteins was found to be sufficient to explain the experimentally determined increase of their thermostability. Free energies of intrahelical interactions of different RecA proteins calculated at three temperatures that are thought to be close to its normal environmental conditions were found to be approximately equal. This indicates that certain flexibility of RecA protein structure is an essential factor for protein function. All RecA proteins analyzed fell into three temperature-dependent classes of similar α-helix stability (ΔG_int = 45.0 ± 2.0 kcal/mol). These classes were consistent with the natural origin of the proteins. Based on the sequences of protein α helices with optimized arrangement of stabilizing interactions, a natural reserve of RecA protein thermoresistance was estimated to be sufficient for conformational stability of the protein at nearly 200°C. Proteins 29:309–320, 1997. © 1997 Wiley-Liss, Inc.     

Raman study of crystalline peptides containing beta turns

The Raman spectra of crystalline H-ProLeuGlyNH₂ which has a type II β turn, crystalline S-benzylCysProLeuGlyNH₂ which has a type I β-turn, and crystalline gramicidin S which has two β turns and β-sheet structure in its conformation, were investigated. The amide I and amide III bands of the peptides with β turns were generally different from those which are diagnostic for α-helix and β-sheet conformations. The patterns of the amide I and amide III bands, when examined together, indicate that Raman spectra can provide diagnostic evidence for β-turn structure in peptides.     

Reversible thermal denaturation of a 60-kDa genetically engineered beta-sheet polypeptide

A de novo 687-amino-acid residue polypeptide with a regular 32-amino-acid repeat sequence, (GA)₃GY(GA)₃GE(GA)₃GH(GA)₃GK, forms large β-sheet assemblages that exhibit remarkable folding properties and, as well, form fibrillar structures. This construct is an excellent tool to explore the details of β-sheet formation yielding intimate folding information that is otherwise difficult to obtain and may inform folding studies of naturally occurring materials. The polypeptide assumes a fully folded antiparallel β-sheet/turn structure at room temperature, and yet is completely and reversibly denatured at 125°C, adopting a predominant polyproline II conformation. Deep ultraviolet Raman spectroscopy indicated that melting/refolding occurred without any spectroscopically distinct intermediates, yet the relaxation kinetics depend on the initial polypeptide state, as would be indicative of a non-two-state process. Thermal denaturation and refolding on cooling appeared to be monoexponential with characteristic times of ~1 and ~60 min, respectively, indicating no detectable formation of hairpin-type nuclei in the millisecond timescale that could be attributed to nonlocal “nonnative” interactions. The polypeptide folding dynamics agree with a general property of β-sheet proteins, i.e., initial collapse precedes secondary structure formation. The observed folding is much faster than expected for a protein of this size and could be attributed to a less frustrated free-energy landscape funnel for folding. The polypeptide sequence suggests an important balance between the absence of strong nonnative contacts (salt bridges or hydrophobic collapse) and limited repulsion of charged side chains.     

Cotranslational Folding of a Pentarepeat β-Helix Protein

It is becoming increasingly clear that many proteins start to fold cotranslationally before the entire polypeptide chain has been synthesized on the ribosome. One class of proteins that a priori would seem particularly prone to cotranslational folding is repeat proteins, that is, proteins that are built from an array of nearly identical sequence repeats. However, while the folding of repeat proteins has been studied extensively in vitro with purified proteins, only a handful of studies have addressed the issue of cotranslational folding of repeat proteins. Here, we have determined the structure and studied the cotranslational folding of a β-helix pentarepeat protein from the human pathogen Clostridium botulinum—a homolog of the fluoroquinolone resistance protein MfpA—using an assay in which the SecM translational arrest peptide serves as a force sensor to detect folding events. We find that cotranslational folding of a segment corresponding to the first four of the eight β-helix coils in the protein produces enough force to release ribosome stalling and that folding starts when this unit is ~ 35 residues away from the P-site, near the distal end of the ribosome exit tunnel. An additional folding transition is seen when the whole PENT moiety emerges from the exit tunnel. The early cotranslational formation of a folded unit may be important to avoid misfolding events in vivo and may reflect the minimal size of a stable β-helix since it is structurally homologous to the smallest known β-helix protein, a four-coil protein that is stable in solution.     

Ramachandran maps for side chains in globular proteins

The Ramachandran plot for backbone ϕ,ψ-angles in a blocked monopeptide has played a central role in understanding protein structure. Curiously, a similar analysis for side chain χ-angles has been comparatively neglected. Instead, efforts have focused on compiling various types of side chain libraries extracted from proteins of known structure. Departing from this trend, the following analysis presents backbone-based maps of side chains in blocked monopeptides. As in the original ϕ,ψ-plot, these maps are derived solely from hard-sphere steric repulsion. Remarkably, the side chain biases exhibit marked similarities to corresponding biases seen in high-resolution protein structures. Consequently, some of the entropic cost for side chain localization in proteins is prepaid prior to the onset of folding events because conformational bias is built into the chain at the covalent level. Furthermore, side chain conformations are seen to experience fewer steric restrictions for backbone conformations in either the α or β basins, those map regions where repetitive ϕ,ψ-angles result in α-helices or strands of β-sheet, respectively. Here, these α and β basins are entropically favored for steric reasons alone; a blocked monopeptide is too short to accommodate the peptide hydrogen bonds that stabilize repetitive secondary structure. Thus, despite differing energetics, α/β-basins are favored for both monopeptides and repetitive secondary structure, underpinning an energetically unfrustrated compatibility between these two levels of protein structure.     

From Discrete to Continuum Models of Three-Dimensional Deformations in Epithelial Sheets

Metamorphic proteins, including proteins with high levels of sequence identity but different folds, are exceptions to the long-standing rule-of-thumb that proteins with as little as 30% sequence identity adopt the same fold. Which topologies can be bridged by these highly identical sequences remains an open question. Here we bridge two 3-α-helix bundle proteins with two radically different folds. Using a straightforward approach, we engineered the sequences of one subdomain within maltose binding protein (MBP, α/β/α-sandwich) and another within outer surface protein A (OspA, β-sheet) to have high sequence identity (80 and 77%, respectively) with engineered variants of protein G (G_A, 3-α-helix bundle). Circular dichroism and nuclear magnetic resonance spectra of all engineered variants demonstrate that they maintain their native conformations despite substantial sequence modification. Furthermore, the MBP variant (80% identical to G_A) remained active. Thermodynamic analysis of numerous G_A and MBP variants suggests that the key to our approach involved stabilizing the modified MBP and OspA subdomains via external interactions with neighboring substructures, indicating that subdomain interactions can stabilize alternative folds over a broad range of sequence variation. These findings suggest that it is possible to bridge one fold with many other topologies, which has implications for protein folding, evolution, and misfolding diseases.     

Structural trees for protein superfamilies

Structural trees for large protein superfamilies, such as β proteins with the aligned β sheet packing, β proteins with the orthogonal packing of α helices, two-layer and three-layer α/β proteins, have been constructed. The structural motifs having unique overall folds and a unique handedness are taken as root structures of the trees. The larger protein structures of each superfamily are obtained by a stepwise addition of α helices and/or β strands to the corresponding root motif, taking into account a restricted set of rules inferred from known principles of the protein structure. Among these rules, prohibition of crossing connections, attention to handedness and compactness, and a requirement for α helices to be packed in α-helical layers and β strands in β layers are the most important. Proteins and domains whose structures can be obtained by stepwise addition of α helices and/or β strands to the same root motif can be grouped into one structural class or a superfamily. Proteins and domains found within branches of a structural tree can be grouped into subclasses or subfamilies. Levels of structural similarity between different proteins can easily be observed by visual inspection. Within one branch, protein structures having a higher position in the tree include the structures located lower. Proteins and domains of different branches have the structure located in the branching point as the common fold. Proteins 28:241–260, 1997. © 1997 Wiley-Liss Inc.