Multiple Folding Research Articles

Hierarchy of folding and unfolding events of protein G, CI2, and ACBP from explicit-solvent simulations

The study of the mechanism which is at the basis of the phenomenon of protein folding requires the knowledge of multiple folding trajectories under biological conditions. Using a biasing molecular-dynamics algorithm based on the physics of the ratchet-and-pawl system, we carry out all-atom, explicit solvent simulations of the sequence of folding events which proteins G, CI2, and ACBP undergo in evolving from the denatured to the folded state. Starting from highly disordered conformations, the algorithm allows the proteins to reach, at the price of a modest computational effort, nativelike conformations, within a root mean square deviation (RMSD) of approximately 1 Å. A scheme is developed to extract, from the myriad of events, information concerning the sequence of native contact formation and of their eventual correlation. Such an analysis indicates that all the studied proteins fold hierarchically, through pathways which, although not deterministic, are well-defined with respect to the order of contact formation. The algorithm also allows one to study unfolding, a process which looks, to a large extent, like the reverse of the major folding pathway. This is also true in situations in which many pathways contribute to the folding process, like in the case of protein G.

Flower micelles of poly(N-isopropylacrylamide) with azobenzene moieties regularly inserted into the main chain

Hydrophobically modified poly(N-isopropylacrylamide) (PNIPAM) with an azobenzene-containing short segment repeatedly inserted into the main chain was synthesized. The reversible trans–cisphotoisomerization of azobenzene was found to exert little effect on the lower critical solution temperature (LCST) or the cloud point of PNIPAM in aqueous solution. The characterization results of variable-temperature 1H NMR, TEM and DLS indicate the formation of flower micelles in cold water with hydrated PNIPAM. The segregation of azobenzene moieties in the micelle core from PNIPAM chains in micelle corona may explain the absence of an effect of photoisomerization of the chromophore on the LCST of PNIPAM. While the micelle core could react during the photoisomerization by swelling (with azobenzene in the cis form) or contracting (in the trans form) due to a polarity change, PNIPAM undergoes the hydration–dehydration transition according to its LCST. Another interesting finding is that the flower micelles resulting from this type of hydrophobically modified PNIPAM exhibit a much reduced propensity for inter-micellar association and could remain well dispersed at T > LCST of PNIAM, in sharp contrast with flower micelles formed by α,ω-telechelic PNIPAM. The multiple intrachain folding arising from this new PNIPAM structure appears to be the origin of the enhanced dispersion stability. With azobenzene groups stacked together forming the micelle core, the multiple loops for individual PNIPAM chains could reduce the number of inter-micellar bridge chains and restrict the chain entanglements upon dehydration of PNIPAM, which, consequently, reduce the propensity for micellar aggregation.

Folding Partitions, Components Therefor and Related Methods

Folding partitions and related methods and components are provided and may include accordion-style folding doors or partitions having multiple folding panels. In one embodiment, a folding partition includes one or more baffling devices. The baffling device may include an elongated enclosure having a first opening proximate a first longitudinal end of the elongated enclosure, a second opening proximate a second longitudinal end of the elongated enclosure, and a flow path extending from the first opening to the second opening. A sound attenuation chamber may be disposed adjacent the at least one insulative channel disposed adjacent at least a portion of the flow path. In one embodiment, the baffling device may be disposed in, or otherwise associated with, a jamb component of the partition. In another embodiment, the baffling device may be disposed in, or otherwise associated with, the hinged connection of panel sections of a partition panel.

PACE Force Field for Protein Simulations. 2. Folding Simulations of Peptides

We present the application of our recently developed PACE force field to the folding of peptides. These peptides include α-helical (AK17 and Fs), β-sheet (GB1m2 and Trpzip2), and mixed helical/coil (Trp-cage) peptides. With replica exchange molecular dynamics (REMD), our force field can fold the five peptides into their native structures while maintaining their stabilities reasonably well. Our force field is also able to capture important thermodynamic features of the five peptides that have been observed in previous experimental and computational studies, such as different preferences for a helix-turn-helix topology for AK17 and Fs, the relative contribution of four hydrophobic side chains of GB1p to the stability of β-hairpin, and the distinct role of a hydrogen bond involving Trp-Hε and a D9/R16 salt bridge in stabilizing the Trp-cage native structure. Furthermore, multiple folding and unfolding events are observed in our microsecond-long normal MD simulations of AK17, Trpzip2, and Trp-cage. These simulations provide mechanistic information such as a "zip-out" pathway of the folding mechanism of Trpzip2 and the folding times of AK17 and Trp-cage, which are estimated to be about 51 ± 43 ns and 270 ± 110 ns, respectively. A 600 ns simulation of the peptides can be completed within one day. These features of our force field are potentially applicable to the study of thermodynamics and kinetics of real protein systems.

Multiple Unfolding Events during Native Folding of the Tetrahymena Group I Ribozyme

Despite the ubiquitous nature of misfolded intermediates in RNA folding, little is known about their physical properties or the folding transitions that allow them to continue folding productively. Folding of the Tetrahymena group I ribozyme includes sequential accumulation of two intermediates, termed I trap and misfolded (M). Here, we probe the structure and folding transition of I trap and compare them to those of M. Hydroxyl radical and dimethyl sulfate footprinting show that both I trap and M are extensively structured and crudely resemble the native RNA. However, regions of the core P3–P8 domain are more exposed to solvent in I trap than in M. I trap rearranges to continue folding nearly 1000-fold faster than M, and urea accelerates folding of I trap much less than M. Thus, the rate-limiting transition from I trap requires a smaller increase in exposed surface. Mutations that disrupt peripheral tertiary contacts give large and nearly uniform increases in re-folding of M, whereas the same mutations give at most modest increases in folding from I trap. Intriguingly, mutations within the peripheral element P5abc give 5- to 10-fold accelerations in escape from I trap, whereas ablation of P13, which lies on the opposite surface in the native structure, near the P3–P8 domain, has no effect. Thus, the unfolding required from I trap appears to be local, whereas the unfolding of M appears to be global. Further, the modest effects from several mutations suggest that there are multiple pathways for escape from I trap and that escape is aided by loosening nearby native structural constraints, presumably to facilitate local movements of nucleotides or segments that have not formed native contacts. Overall, these and prior results suggest a model in which the global architecture and peripheral interactions of the RNA are achieved relatively early in folding. Multiple folding and re-folding events occur on the predominant pathway to the native state, with increasing native core interactions and cooperativity as folding progresses.

Free Energy Landscape of the FBP28 WW Domain by All-Atom Direct Folding Simulation

Using multiplexed Q-replica exchange molecular dynamics simulation (mQ-REMD) with a modified all-atom force field (param99MOD5/GBSA), direct folding simulation at the all-atom level was performed to investigate the folding free energy landscapes of the FBP28 WW domain: a 37-residue three-stranded beta-sheet. Starting from a fully extended conformation of the FBP28 WW domain, a total of 400 ns simulation was run for each replica with this simulation protocol. Free energy analysis showed that the folding of this protein substantially proceeded through the formation of folded turn 1, followed by the partial or misfolded formation of turn 2. More importantly, the folding process of turn 2 exhibited complicated folding behaviors in the presence of several intermediate states along multiple folding routes.

Identification of Multiple Folding Pathways Shared by Three-Helix Bundle Proteins

Protein domains can be clustered on the basis of shared structural similarity into folds, and a limited number of protein folds have been observed. The small number of proteins folds may imply a similarly constrained number of folding pathways. The degree to which dynamics vary within and between folds may provide broader insights to the protein folding problem. As an initial step, we are interested in the breadth of native dynamics and unfolding behavior within a single fold. Three members (EnHD, c-Myb, and hTRF1) of the three-helix bundle engrailed homeodomain family exhibit some outwardly different folding behavior linked by an shared underlying mechanism dependent on their relative helical propensity. However, further sampling is needed to extract consistent residue-level determinants of protein folding. Consequently, ten additional members of this three-helix bundle fold were selected for simulation based on their low sequence similarity. We have identified multiple unique initial unfolding events shared between three-helix bundles and are characterizing the source of the divergent folding events.

The HD-exchange motions of ribosomal protein S6 are insensitive to reversal of the protein-folding pathway

An increasing number of protein structures are found to encompass multiple folding nuclei, allowing their structures to be formed by several competing pathways. A typical example is the ribosomal protein S6, which comprises two folding nuclei (sigma1 and sigma2) defining two competing pathways in the folding energy landscape: sigma1 --> sigma2 and sigma2 --> sigma1. The balance between the two pathways, and thus the order of folding events, is easily controlled by circular permutation. In this study, we make use of this ability to manipulate the folding pathway to demonstrate that the dynamic motions of the S6 structure are independent of how the protein folds. The HD-exchange protection factors remain the same upon complete reversal of the folding order. The phenomenon arises because the HD-exchange motions and the high-energy excitations controlling the folding pathway occur at separated free-energy levels: the Boltzmann distribution of unproductive unfolding attempts samples all unfolding channels in parallel, even those that end up in excessively high barriers. Accordingly, the findings provide a simple rationale for how to interpret native-state dynamics without the need to invoke fluctuations off the normal unfolding reaction coordinate.

Multiple Folding Structures Mediated by Metal Coordination of Acyclic Multidentate Ligand

Four kinds of folded structures are formed upon the metal complexation of a bis(N(2)O(2)) ligand in which two oxime-type N(2)O(2) chelate ligands are connected by a flexible diethyleneoxy linker. The N(2)O(2) coordination sites are intended for d-block transition-metal ions, and the diethyleneoxy linker can interact with hard metal cations. Meso double helical, folded Omega-shaped, S-shaped helical, and single helical structures were formed depending on the metal combination. The difference in the affinity to metal cations resulted in variation of the folding modes and enabled the structural conversion between the folded structures.

Tectonic significance of structural successions preserved within low-strain pods: Implications for thin- to thick-skinned tectonics vs. multiple near-orthogonal folding events in the Palaeo-Mesoproterozoic Mount Isa Inlier (NE Australia)

Event-wise structural successions and their regional correlation across the orogenic belts are always been problematic, resulting in construction of conflicting tectonic models. However, in recent years, it has been demonstrated that complex deformation and metamorphic overprinting relationships and their correlation across large regions can be resolved through careful quantitative analysis of fabrics preserved within low-strain pods at different scales. In this study, integrated macro, meso and microstructural analyses of the southern Selwyn Range of the Palaeo-Mesoproterozoic Mount Isa Inlier, NE Australia, have revealed that the tectonic evolution involved four dominant fabric-generating events. These differ from those reported previously based on interpretation of the Mount Isa deep seismic section. D 1 is preserved in low-strain zones as outcrop-scale recumbent folds (S 1 subparallel S 0) with E–W trending fold axes. D 2 is characterized by upright to inclined N–S trending folds with a pervasive regional foliation (S 2), subparallel to S 0/S 1. D 3 was very heterogeneously developed in the region and appears to be controlled by rheological contrast. Three structural domains are recognized. The Eastern Domain (ED) and Central Domain (CD) are characterized by shallow dipping S 3 foliation. S 3 foliation is rare in the Western Domain (WD). Deformation during D 4 was intense in the WD and produced complex overprinting patterns. All these observations from macro to microscopic scale suggest that the regional S 2 initially formed as a steep foliation and was deflected to shallow to moderate dips due to the effect of subvertical shortening (D 3). These observations are difficult to reconcile with previously suggested thin- to thick-skinned nappe-style tectonics in the region. Metamorphic field gradient is higher (sillimanite zone) in the ED, which represents deeper crustal levels than the rocks exposed in the CD and WD separated by pre-existing faults. Andalusite porphyroblasts in the WD preserve steeply dipping S 2, whereas staurolite porphyroblasts preserve gently dipping S 3, suggesting staurolite grew after andalusite. This interpretation is consistent with the appearance of mineral phases in the KFMASH system along a prograde P– T path (M 2) for D 2 and D 3. An earlier higher pressure P– T path (M 1) is proposed for D 1 based on the metamorphic mineral assemblages in the MnNCKFMASH system.

Probing Early Events in Ferrous Cytochrome c Folding with Time- Resolved Natural and Magnetic Circular Dichroism Spectroscopies.

In a 1998 collaboration with Tony Fink, we coupled nanosecond circular dichroism methods (TRCD) with a CO-photolysis system for quickly triggering folding in cytochrome c (cyt c) in order to make the first time-resolved far-UV CD measurement of early secondary structure formation in a protein. The small signal observed in that initial study, approximately 10% of native helicity, became the seed for increasingly robust results from subsequent studies bringing additional natural and magnetic circular polarization dichroism and optical rotatory dispersion detection methods (e.g., TRORD, TRMCD, and TRMORD), coupled to fast photolysis and photoreduction triggers, to the study of early folding events. Nanosecond polarization methods are reviewed here in the context of the range of initiation methods and structure-sensitive probes currently available for fast folding studies. We also review the impact of experimental results from fast polarization studies on questions in folding dynamics such as the possibility of multiple folding pathways implied by energy landscape models, the sequence dependence of ultrafast helix formation, and the simultaneity of chain collapse and secondary structure formation implicit in molten globule models for kinetic folding intermediates.

How Does a Simplified-Sequence Protein Fold?

To investigate a putatively primordial protein we have simplified the sequence of a 56-residue α/ β fold (the immunoglobulin-binding domain of protein G) by replacing it with polyalanine, polythreonine, and diglycine segments at regions of the sequence that in the folded structure are α-helical, β-strand, and turns, respectively. Remarkably, multiple folding and unfolding events are observed in a 15- μs molecular dynamics simulation at 330 K. The most stable state (populated at ∼20%) of the simplified-sequence variant of protein G has the same α/ β topology as the wild-type but shows the characteristics of a molten globule, i.e., loose contacts among side chains and lack of a specific hydrophobic core. The unfolded state is heterogeneous and includes a variety of α/ β topologies but also fully α-helical and fully β-sheet structures. Transitions within the denatured state are very fast, and the molten-globule state is reached in <1 μs by a framework mechanism of folding with multiple pathways. The native structure of the wild-type is more rigid than the molten-globule conformation of the simplified-sequence variant. The difference in structural stability and the very fast folding of the simplified protein suggest that evolution has enriched the primordial alphabet of amino acids mainly to optimize protein function by stabilization of a unique structure with specific tertiary interactions.

Structural characteristics of the beta-sheet-like human and rat islet amyloid polypeptides as determined by scanning tunneling microscopy

We demonstrate in this work that scanning tunneling microscopy (STM) provides a useful approach to obtaining structural information about human islet amyloid polypeptide (hIAPP) and rat islet amyloid polypeptide (rIAPP) assembly on highly oriented pyrolytic graphite (HOPG) with sub-molecular resolution. The observed hIAPP and rIAPP lamellae consisted of parallel stripes. The STM images of hIAPPs show multiple molecular folding structures, with an average of 11 amino acid residues for the core regions. In addition, the STM images also reveal the assembly characteristics of rIAPP lamellae and may indicate a secondary structural conformation from random coil to beta-sheet-like on the graphite surface.

Dynamic folding pathway models of the villin headpiece subdomain (HP‐36) structure

We have investigated the folding pathway of the 36-residue villin headpiece subdomain (HP-36) by action-derived molecular dynamics simulations. The folding is initiated by hydrophobic collapse, after which the concurrent formation of full tertiary structure and alpha-helical secondary structure is observed. The collapse is observed to be associated with a couple of specific native contacts contrary to the conventional nonspecific hydrophobic collapse model. Stable secondary structure formation after the collapse suggests that the folding of HP-36 follows neither the framework model nor the diffusion-collision model. The C-terminal helix forms first, followed by the N-terminal helix positioned in its native orientation. The short middle helix is shown to form last. Signs for multiple folding pathways are also observed.

Multiple Contact Network Is a Key Determinant to Protein Folding Rates

Understanding the relationship between amino acid sequences and folding rates of proteins is an important task in computational and molecular biology. It has been shown that topological parameters, contact order, long-range order, and total contact distance relate well with protein folding rates. In this work, we have systematically analyzed the influence of amino acid residues that form multiple contacts in protein structures to folding rates of proteins. We observed an inverse relationship between the number of residues with multiple contacts and protein folding rates. Further analysis indicates that multiple contacts are influenced by hydrophobic residues, whereas the role is minimal between the residues that are capable of forming hydrogen bonds. The propensity of multiple contacts forming residues showed that aromatic and hydrophobic residues are dominant in two-state proteins, whereas the polar residues Ser and Thr are also preferred ones in three-state proteins. In addition, multiple contact forming residues are interconnected with each other through contact networks in protein structures. The comparison between slow and fast folding proteins demonstrated the presence of more multiple contact forming residues in slow folding proteins with a limit of 4-6 contacts/residue. These results have been reflected in amino acid sequences in the form of short-, medium-, and long-range contacts, which could discriminate slow and fast folding proteins with an accuracy of 96% using a 5-fold cross-validation method.

BETASCAN: Probable β-amyloids Identified by Pairwise Probabilistic Analysis

Amyloids and prion proteins are clinically and biologically important β-structures, whose supersecondary structures are difficult to determine by standard experimental or computational means. In addition, significant conformational heterogeneity is known or suspected to exist in many amyloid fibrils. Recent work has indicated the utility of pairwise probabilistic statistics in β-structure prediction. We develop here a new strategy for β-structure prediction, emphasizing the determination of β-strands and pairs of β-strands as fundamental units of β-structure. Our program, BETASCAN, calculates likelihood scores for potential β-strands and strand-pairs based on correlations observed in parallel β-sheets. The program then determines the strands and pairs with the greatest local likelihood for all of the sequence's potential β-structures. BETASCAN suggests multiple alternate folding patterns and assigns relative a priori probabilities based solely on amino acid sequence, probability tables, and pre-chosen parameters. The algorithm compares favorably with the results of previous algorithms (BETAPRO, PASTA, SALSA, TANGO, and Zyggregator) in β-structure prediction and amyloid propensity prediction. Accurate prediction is demonstrated for experimentally determined amyloid β-structures, for a set of known β-aggregates, and for the parallel β-strands of β-helices, amyloid-like globular proteins. BETASCAN is able both to detect β-strands with higher sensitivity and to detect the edges of β-strands in a richly β-like sequence. For two proteins (Aβ and Het-s), there exist multiple sets of experimental data implying contradictory structures; BETASCAN is able to detect each competing structure as a potential structure variant. The ability to correlate multiple alternate β-structures to experiment opens the possibility of computational investigation of prion strains and structural heterogeneity of amyloid. BETASCAN is publicly accessible on the Web at http://betascan.csail.mit.edu.

Amyloid β (1–42) Folding Multiplicity and Single-Molecule Binding Behavior Studied with STM

The fine folding and assembling characteristics of amyloid β (Aβ) peptides are important to pharmaceutical studies of drug molecules and to the pathological analysis of neurodegenerative disorders such as Alzheimer's disease at the molecular level. Here we present observations of the multiple folding characteristics of amyloid peptide Aβ 42 lamellae using scanning tunneling microscopy. Molecularly resolved core regions of Aβ 42 hairpins and unfolded peptide assembly structures are identified. The parallel assembling characteristics of Aβ 42 hairpins can be confirmed in the study. In addition, single-molecule binding characteristics of Congo red and thioflavin T have been shown to bind at the groove regions of peptide assemblies. This study demonstrates a complementary venue for studying molecular heterogeneity of peptide assemblies, as well as the binding characteristics of molecular modulators.

Thermodynamics and Folding Pathways of Trpzip2: An Accelerated Molecular Dynamics Simulation Study

In this paper, we apply an enhanced sampling method introduced earlier to study the folding mechanism of a beta-hairpin, trpzip2, using an all-atom potential for the protein and an implicit model for the solvent. The enhanced sampling method allows us to obtain multiple protein folding and unfolding trajectories in relatively short simulations. The sufficient sampling of folding and unfolding events of trpzip2 makes possible a more detailed investigation of its folding landscape and thermodynamics, leading to the identification of folding pathways. The analysis of the thermodynamics involved in the folding of trpzip2 showed that this polypeptide folds by two stages: a downhill hydrophobic collapse followed by formation of native hydrogen bonds. During the hydrogen bond formation, a transition state can be identified with only one native hydrogen bond being formed, which is more consistent with a 'zip-out' mechanism. To address the dynamics in a more reliable way, explicit solvent was used for the estimation of the diffusion constant, which was used in the Kramer's theory, together with the free energy profile calculated using the enhanced sampling and the implicit solvent, to calculate the folding rate. The calculated folding time agrees well with the experimental value of 2.5 micros.

Relative Stability of Helices Determines the Folding Landscape of Adenine Riboswitch Aptamers

Riboswitches, whose folding is controlled by binding of metabolites to the aptamer domain, regulate downstream gene expression. Folding properties of the aptamer strongly influence the conformation of the downstream expression platform, which controls transcription termination or translation initiation. We have characterized the energy landscape of the add riboswitch aptamer quantitatively by unfolding and refolding the molecule with mechanical force using the coarse-grained self-organized polymer model and Brownian dynamics simulation. Multiple folding states have been found during the folding process of the aptamer, both with and without adenine, consistent with single molecule studies of purine riboswitches. Adenine binding stabilizes the folded structure and significantly decreases the unfolding rate of the aptamer, the folding of which is in competition with the formation of the downstream stem-loop structure in the complete riboswitch. These results provide insights into the mechanism of gene regulation by the RNA switches.

Subdomain Competition, Cooperativity, and Topological Frustration in the Folding of CheY

The folding of multidomain proteins often proceeds in a hierarchical fashion with individual domains folding independent of one another. A large single-domain protein, however, can consist of multiple modules whose folding may be autonomous or interdependent in ways that are unclear. We used coarse-grained simulations to explore the folding landscape of the two-subdomain bacterial response regulator CheY. Thermodynamic and kinetic characterization shows the landscape to be highly analogous to the four-state landscape reported for another two-subdomain protein, T4 lysozyme. An on-pathway intermediate structured in the more stable nucleating subdomain was observed, as were transient states frustrated in off-pathway contacts prematurely structured in the weaker subdomain. Local unfolding, or backtracking, was observed in the frustrated state before the native conformation could be reached. Nonproductive frustration was attributable to competition for van der Waals contacts between the two subdomains. In an accompanying article, stopped-flow kinetic measurements support an off-pathway burst-phase intermediate, seemingly consistent with our prediction of early frustration in the folding landscape of CheY. Comparison of the folding mechanisms for CheY, T4 lysozyme, and interleukin-1β leads us to postulate that subdomain competition is a general feature of large single-domain proteins with multiple folding modules.