The barriers in protein folding

Ultrafast signals in protein folding and the polypeptide contracted state

To test the significance of ultrafast protein folding signals (<<1 msec), we studied cytochrome c (Cyt c) and two Cyt c fragments with major C-terminal segments deleted. The fragments remain unfolded under all conditions and so could be used to define the unfolded baselines for protein fluorescence and circular dichroism (CD) as a function of denaturant concentration. When diluted from high to low denaturant in kinetic folding experiments, the fragments readjust to their new baseline values in a "burst phase" within the mixing dead time. The fragment burst phase reflects a contraction of the polypeptide from a more extended unfolded condition at high denaturant to a more contracted unfolded condition in the poorer, low denaturant solvent. Holo Cyt c exhibits fluorescence and CD burst phase signals that are essentially identical to the fragment signals over the whole range of final denaturant concentrations, evidently reflecting the same solvent-dependent, relatively nonspecific contraction and not the formation of a specific folding intermediate. The significance of fast folding signals in Cyt c and other proteins is discussed in relation to the hypothesis of an initial rate-limiting search-nucleation-collapse step in protein folding [Sosnick, T. R., Mayne, L. & Englander, S. W. (1996) Proteins Struct. Funct. Genet. 24, 413-426].

Protein folding and intermediates

With the exception of the discovery of the rate of formation of the earliest intermediates, there have been no major conceptual leaps in our understanding of protein folding reactions over the past two years. Rather, this period has seen an extension of two established techniques: first, mutational analysis combined with a kinetic definition of the energy landscape of the reaction; and second, the use of hydrogen/deuterium exchange of backbone amide groups combined with NMR. Owing to the application of these methods to a wider range of proteins, it is now possible to draw some general conclusions about the physical processes that direct a protein to its native fold.

A comparison of the folding kinetics and thermodynamics of two homologous fibronectin type III modules

The homologous ninth and tenth type III modules of human fibronectin share identical topologies and nearly identical core structures. Despite these structural similarities, the refolding characteristics of the two modules, which have a sequence identity of less than 30 %, are very different; in the absence of denaturant the ninth module folds several hundred times more slowly than the tenth and, although both modules contain numerous proline residues, only the ninth exhibits a slow, proline isomerization-limited folding phase. The different folding kinetics of the two modules coincide with a large difference in their thermodynamic stability, with the folding free energy of the tenth being approximately five fold greater than that of the ninth. This may be the reason why the ninth module, unlike the rapidly folding tenth module, is apparently unable to overcome characteristics of the fibronectin type III modules that can slow the folding process. The non-proline-limited folding kinetics are, however, very similar for the two modules when compared under conditions where their overall stabilities are similar. The significance of this finding is discussed in terms of possible determinants of the kinetics of protein folding.

Touring the landscapes: partially folded proteins examined by hydrogen exchange

Recent studies on Escherichia coli ribonuclease H and several other proteins reveal a specific region in each protein that remains structured in partially folded conformations. These regions play a dominant role in determining the fold and stability of the protein.

Three-state model for lysozyme folding: triangular folding mechanism with an energetically trapped intermediate

We investigated the role of a partially folded intermediate that transiently accumulates during lysozyme folding. Previous studies had shown that the partially folded intermediate is located on a slow-folding pathway and that an additional fast direct pathway from the unfolded state to the native state exists. Kinetic double-jump experiments showed that the two folding pathways are not caused by slow equilibration reactions in the unfolded state. Rather, kinetic partitioning occurs very early in lysozyme refolding, giving the molecules the chance to enter the direct pathway or a slow-folding channel. Fitting the guanidinium chloride dependencies of the refolding and unfolding reactions to analytical solutions for different folding scenarios enables us to propose a triangular mechanism as the minimal model for lysozyme folding explaining all observed kinetic reactions: [diagram in text]. All microscopic rate constants and their guanidinium chloride dependencies could be obtained from the experimental data. The results suggest that population of the intermediate during refolding increases the free energy of activation of the folding process. This effect is due to the increased stability of the intermediate state compared to the unfolded state leading to an increase in the free energy of activation (deltaG0) compared to folding in the absence of populated intermediate states. The absolute energy of the transition state is identical on both pathways. The results imply that pre-formed secondary structure in the folding intermediate obstructs formation of the transition state of folding but does not change the nature of the rate-limiting step in the folding process.

Detection of residue contacts in a protein folding intermediate

Protein folding can be described in terms of the development of specific contacts between residues as a highly disordered polypeptide chain converts into the native state. Here we describe an NMR based strategy designed to detect such contacts by observation of nuclear Overhauser effects (NOEs). Experiments with alpha-lactalbumin reveal the existence of extensive NOEs between aromatic and aliphatic protons in the archetypal molten globule formed by this protein at low pH. Analysis of their time development provides direct evidence for near-native compactness of this state. Through a rapid refolding procedure the NOE intensity can be transferred efficiently into the resolved and assigned spectrum of the native state. This demonstrates the viability of using this approach to map out time-averaged interactions between residues in a partially folded protein.

Absence of a stable intermediate on the folding pathway of protein A

The B-domain of protein A has one of the simplest protein topologies, a three-helix bundle. Its folding has been studied as a model for elementary steps in the folding of larger proteins. Earlier studies suggested that folding might occur by way of a helical hairpin intermediate. Equilibrium hydrogen exchange measurements indicate that the C-terminal helical hairpin could be a potential folding intermediate. Kinetic refolding experiments were performed using stopped-flow circular dichroism and NMR hydrogen-deuterium exchange pulse labeling. Folding of the entire molecule is essentially complete within the 6 ms dead time of the quench-flow apparatus, indicating that the intermediate, if formed, progresses rapidly to the final folded state. Site-directed mutagenesis of the isoleucine residue at position 16 was used to generate a variant protein containing tryptophan (the 116 W mutant). The formation of the putative folding intermediate was expected to be favored in this mutant at the expense of the native folded form, due to predicted unfavorable steric interactions of the bulky tryptophan side chain in the folded state. The 116 W mutant refolds completely within the dead time of a stopped-flow fluorescence experiment. No partly folded intermediate could be detected by either kinetic or equilibrium measurements. Studies of peptide fragments suggest that the protein A sequence has an intrinsic propensity to form a helix II/helix III hairpin. However, its stability appears to be marginal (of the order of 1/2 kT) and it could not be an obligatory intermediate on a defined folding pathway. These results explicitly demonstrate that the protein A B domain folds extremely rapidly by an apparent two-state mechanism without formation of stable partly folded intermediates. Similar mechanisms may also be involved in the rapid folding of subdomains of larger proteins to form the compact molten globule intermediates that often accumulate during the folding process.

Temperature control for kinetic refolding of heat-denatured ovalbumin

The folding of heat-denatured ovalbumin, a non-inhibitory serpin with a molecular size of 45 kDa, was examined. Ovalbumin was heat-denatured at 80 degrees C under nonreducing conditions at pH 7.5 and then cooled either slowly or rapidly. Slow cooling allowed the heat-denatured ovalbumin to refold to its native structure with subsequent resistance to digestion by trypsin. Upon rapid cooling, by contrast, the heat-denatured molecules assumed the metastable non-native conformations that were susceptible to trypsin. The non-native species were marginally stable for several days at a low temperature, but the molecules were transformed slowly into the native conformation. Considering data from size-exclusion chromatography and from analyses of CD, intrinsic tryptophan fluorescence, and adsorption of the dye 1-anilinonaphthalene-8-sulfonate, we postulated that the non-native species that accumulated upon rapid cooling were compact but structureless globules with disordered side chains collectively as a folding intermediate. Temperature-jumped CD experiments revealed biphasic kinetics for the refolding process of heat-denatured ovalbumin, with the features of increasing and subsequently decreasing amplitude of the rapid and the slow phases, respectively, with the decrease in folding temperature. The temperature dependence of the refolding kinetics indicated that the yield of renaturation was maximal at about 55 degrees C. These findings suggested the kinetic partitioning of heat-denatured ovalbumin between alternative fates, slow renaturation to the native state and rapid collapse to the metastable intermediate state. Analysis of disulfide pairing revealed the formation of a scrambled form with non-native disulfide interactions in both the heat-denatured state and the intermediate state that accumulated upon rapid cooling, suggesting that non-native disulfide pairing is responsible for the kinetic barriers that retard the correct folding of ovalbumin.

Mutational analysis of the BPTI folding pathway: I. Effects of aromatic-->leucine substitutions on the distribution of folding intermediates

The roles of aromatic residues in determining the folding pathway of bovine pancreatic trypsin inhibitor (BPTI) were analyzed mutationally by examining the distribution of disulfide-bonded intermediates that accumulated during the refolding of protein variants in which tyrosine or phenylalanine residues were individually replaced with leucine. The eight substitutions examined all caused significant changes in the intermediate distribution. In some cases, the major effect was to decrease the accumulation of intermediates containing two of the three disulfides found in the native protein, without affecting the distribution of earlier intermediates. Other substitutions, however, led to much more random distributions of the intermediates containing only one disulfide. These results indicate that the individual residues making up the hydrophobic core of the native protein make clearly distinguishable contributions to conformation and stability early in folding: The early distribution of intermediates does not appear to be determined by a general hydrophobic collapse. The effects of the substitutions were generally consistent with the structures of the major intermediates determined by NMR studies of analogs, confirming that the distribution of disulfide-bonded species is determined by stabilizing interactions within the ordered regions of the intermediates. The plasticity of the BPTI folding pathway implied by these results can be described using conformational funnels to illustrate the degree to which conformational entropy is lost at different stages in the folding of the wild-type and mutant proteins.

Protein folding: does diffusion determine the folding rate?

A major question about protein folding is whether the coming together by diffusion of different segments of the polypeptide chain is rate-determining. This seemingly simple question has been very difficult to answer experimentally, but a positive result has now been obtained with one small model protein.

Diffusion control in an elementary protein folding reaction

The cold-shock protein CspB (from Bacillus subtilis), a very small protein of 67 residues, folds extremely fast in a reversible N &lrharr; U two-state reaction. Both unfolding and refolding are strongly decelerated when the viscosity of the solvent is increased by adding ethylene glycol or sucrose. The folding of CspB thus seems to follow Kramers' model for reactions in which the reactants must diffuse together. It indicates that the compaction of the protein chain occurs in the rate-limiting step of folding. Chain diffusion to a productively collapsed form and the crossing of a high energy barrier are thus tightly coupled in this folding reaction, and the measured reaction rate depends on both the diffusion of the protein chain in the solvent and the magnitude of the activation energy. We suggest that in protein folding an energetic barrier is essential to separate the native from the unfolded conformations of a protein. This barrier protects the ordered structure of a native protein against continuous unfolding by diffusive chain motions and leads to apparent two-state behavior.

Folding of the disulfide-bonded beta-sheet protein tendamistat: rapid two-state folding without hydrophobic collapse

We investigated the reversible folding and unfolding reactions of the small 74 amino acid residue protein tendamistat. The secondary structure of tendamistat contains only beta-sheets and loop regions and the protein contains two disulfide bonds. Fluorescence-detected refolding kinetics of tendamistat (disulfide bonds intact) comprise of a major rapid fast reaction (tau = 10 ms in water) and two minor slow reactions. In the fast reaction 80% of the unfolded molecules are converted to native protein. The two slow reactions are part of a parallel slow folding pathway. On this pathway the rate-limiting step in the formation of native molecules is cis to trans isomerization of at least one of the three trans Xaa-Pro peptide bonds. This reaction is catalyzed efficiently by the enzyme peptidyl-prolyl cis-trans isomerase. Comparison of kinetic data with equilibrium unfolding transitions shows that the fast folding pathway follows a two-state process without populated intermediate states. Additionally, various sensitive tests did not detect any rapid chain collapse during tendamistat folding prior to the acquisition of the native three-dimensional structure. These results show that pre-formed disulfide bonds do not prevent efficient and rapid protein folding.

Fast and slow tracks in lysozyme folding: insight into the role of domains in the folding process

The folding of lysozyme involves parallel events in which hydrogen exchange kinetics indicate the development of persistent structure at very different rates. We have monitored directly the kinetics of formation of the native molecule by the binding of a fluorescently labelled inhibitor, MeU-diNAG (4-methylumbelliferyl-N,N'-diacetyl-beta-D-chitobioside). The data show that native character monitored in this way also develops with different timescales. Although the rate determining step on the slow pathway (approximately 75% of molecules at pH 5.5, 20 degrees C) can be attributed to the need to reorganise structure formed early in the folding process, the data indicate that the rate determining step on the fast track (involving approximately 25% of molecules) involves the docking of the two constituent domains of the protein. In the fast folding track the data are consistent with a model in which each domain forms persistent structure prior to their docking in a locally cooperative manner on a timescale comparable to the folding of small single domain proteins.

The kinetic folding intermediate of ribonuclease H resembles the acid molten globule and partially unfolded molecules detected under native conditions

Folding of ribonuclease HI from Escherichia coli populates a kinetic intermediate detectable by stopped-flow circular dichroism. Pulse labelling hydrogen exchange reveals that this intermediate consists of a structured core region of the protein, namely helices A and D and beta-strand 4. This kinetic intermediate resembles both the acid molten globule of ribonuclease HI and rarely populated, partially unfolded forms detected under native conditions. These results indicate that the first portion of ribonuclease HI to fold is the most thermodynamically stable region of the native state, and that folding of this protein follows a hierarchical process.

Submillisecond protein folding kinetics studied by ultrarapid mixing

An ultrarapid-mixing continuous-flow method has been developed to study submillisecond folding of chemically denatured proteins. Turbulent flow created by pumping solutions through a small gap dilutes the denaturant in tens of microseconds. We have used this method to study cytochrome c folding kinetics in the previously inaccessible time range 80 micros to 3 ms. To eliminate the heme-ligand exchange chemistry that complicates and slows the folding kinetics by trapping misfolded structures, measurements were made with the imidazole complex. Fluorescence quenching due to excitation energy transfer from the tryptophan to the heme was used to monitor the distance between these groups. The fluorescence decrease is biphasic. There is an unresolved process with tau < 50 micros, followed by a slower, exponential process with tau = 600 micros at the lowest denaturant concentration (0.2 M guanidine hydrochloride). These kinetics are interpreted as a barrier-free, partial collapse to the new equilibrium unfolded state at the lower denaturant concentration, followed by slower crossing of a free energy barrier separating the unfolded and folded states. The results raise several fundamental issues concerning the dynamics of collapse and barrier crossings in protein folding.

Fast folding of cytochrome c

Native iso-2 cytochrome c contains two residues (His 18, Met 80) coordinated to the covalently attached heme. On unfolding of iso-2, the His 18 ligand remains coordinated to the heme iron, whereas Met 80 is displaced by a non-native heme ligand, His 33 or His 39. To test whether non-native His-heme ligation slows folding, we have constructed a double mutant protein in which the non-native ligands are replaced by asparagine and lysine, respectively (H33N,H39K iso-2). The double mutant protein, which cannot form non-native histidine-heme coordinate bonds, folds significantly faster than normal iso-2 cytochrome c: gamma = 14-26 ms for H33N,H39K iso-2 versus gamma = 200-1,100 ms for iso-2. These results with iso-2 cytochrome c strongly support the hypothesis that non-native His-heme ligation results in a kinetic barrier to fast folding of cytochrome c. Assuming that the maximum rate of a conformational search is about 10(11) s-1, the results imply that the direct folding pathway of iso-2 involves passage through on the order of 10(9) or fewer partially folded conformers.

Folding kinetics of Che Y mutants with enhanced native alpha-helix propensities

In this work we study the folding kinetics of Che Y mutants in which the helical propensity of each of its five alpha-helices has been greatly enhanced by local interactions (between residues close in sequence). This constitutes an experimental test on the role of local interactions in protein folding, as well as providing new information on the details of the folding pathway of the protein Che Y. With respect to the first issue, our results show that the enhancement of helical propensities by native-like local interactions in Che Y has the following general effects: (1) the energetics of the whole Che Y folding energy landscape (folded state, intermediate, denatured state and main transition state) are affected by the enhancement of helical propensities, thus, native-like local interactions appear to have a low specificity for the native conformation; (2) our results support the idea, proposed from thermodynamic analysis of the mutants, that the denatured state under native conditions becomes more compact upon enhancement of helical propensities; (3) the rate of folding in aqueous solution decreases in all the mutants, suggesting that the optimization of the folding rate in this protein requires low secondary structure propensities. Regarding the description of the folding pathway of Che Y, we find evidence that the folding transition state of Che Y is constituted by two sub-domains with different degree of helical structure. The first includes helices 1 and 2 which are rather structured, while the second encompasses the last three helices, which are very unstructured. On the other hand, the same analysis for the folding intermediate indicates that all the five alpha-helices are, on average, rather structured. Thus, suggesting that a large structural reorganization of the last three alpha-helices must take place before folding can be completed. This conclusion indicates that the folding intermediate of Che Y is a misfolded species.

Submillisecond kinetics of protein folding

New experimental methods permit observation of protein folding and unfolding on the previously inaccessible nanosecond-microsecond timescale. These studies are beginning to establish times for the elementary motions in protein folding - secondary structure and loop formation, local hydrophobic collapse, and global collapse to the compact denatured state. They permit an estimate of about one microsecond for the shortest time in which a protein can possibly fold.

Kinetic role of early intermediates in protein folding

The traditional view that partly folded intermediates are important for directing a protein toward the native state has been challenged by the notion that proteins can intrinsically fold rapidly in a single step if kinetic complications due to slow conformational events are avoided. Intermediates that accumulate within the first few milliseconds of folding are, however, a common observation even for small single-domain proteins. Recent spectroscopic studies, coupled with quantitative kinetic analysis, suggest that folding is facilitated by the rapid formation of compact intermediates with some native-like structural features.

Nucleation mechanisms in protein folding

Experiment and theory are converging on the importance of nucleation mechanisms in protein folding. These mechanisms do not use classic nuclei, which are well formed elements of structure present in ground states, but they use diffuse, extended regions, which are observed in transition states.

Theoretical studies of protein-folding thermodynamics and kinetics

Recently, protein-folding models have advanced to the point where folding simulations of protein-like chains of reasonable length (up to 125 amino acids) are feasible, and the major physical features of folding proteins, such as cooperativity in thermodynamics and nucleation mechanisms in kinetics, can be reproduced. This has allowed deep insight into the physical mechanism of folding, including the solution of the so-called 'Levinthal paradox'.

Structural characterization of the molten globule of alpha-lactalbumin by solution X-ray scattering

A compact denatured state is often observed under a mild denaturation condition for various proteins. A typical example is the alpha-lactalbumin molten globule. Although the molecular compactness and shape are the essential properties for defining the molten globule, there have been ambiguities of these properties for the molten globule of alpha-lactalbumin. Using solution X-ray scattering, we have examined the structural properties of two types of molten globule of alpha-lactalbumin, the apo-protein at neutral pH and the acid molten globule. The radius of gyration for the native holo-protein was 15.7 A, but the two different molten globules both had a radius of gyration of 17.2 A. The maximum dimension of the molecule was also increased from 50 A for the native state to 60 A for the molten globule. These values clearly indicate that the molten globule is not as compact as the native state. The increment in the radius of gyration was less than 10% for the alpha-lactalbumin molten globule, compared with up to 30% for the molten globules of other globular proteins. Intramolecular disulfide bonds restrict the molecular expansion of the molten globule. The distance distribution function of the alpha-lactalbumin molten globule is composed of a single peak suggesting a globular shape, which is simply swollen from the native state. The scattering profile in the high Q region of the molten globule indicates the presence of a significant amount of tertiary fold. Based on the structural properties obtained by solution X-ray scattering, general and conceptual structural images for the molten globules of various proteins are described and compared with the individual, detailed structural model obtained by nuclear magnetic resonance.

Characterization of a partially structured state in an all-beta-sheet protein

Cardiotoxin analogue III (CTX III) is a low-molecular-mass all-beta-sheet protein isolated from the Taiwan cobra (Naja naja atra) venom. A stable partially structured state similar to the "molten globule' state has been identified for CTX III in a 3% (w/v) solution of 2,2,2-trichloroacetic acid at 298 K. This stable state has been structurally characterized using a variety of techniques such as CD, 1-anilinonaphthalene-8-sulphonate fluorescence binding, Fourier transform IR and two-dimensional NMR spectroscopy techniques. Direct assignment of the homonuclear two-dimensional NMR spectra of the protein in 3% trichloroacetic acid showed that drastic structural perturbation had not taken place in the protein and that the 'intermediate' state retained a significant portion of the native secondary-structural interactions. It is found that about 65% of the native beta-sheet structural contacts are maintained in the partially structured state of CTX III in 3% trichloroacetic acid.

Favourable native-like helical local interactions can accelerate protein folding

BACKGROUND

Extensive studies of peptide conformation have provided reasonable knowledge of the rules determining helix stability. This knowledge can be used to stabilize proteins against chemical and thermal denaturation. This has been done in two proteins: the chemotactic protein from Escherichia coli, Che Y (a 129 aa alpha/beta parallel protein with five alpha-helices, which shows an accumulating intermediate during refolding) and the activation domain of human procarboxypeptidase A2, ADA2h (a 81 aa alpha + beta protein domain, with two alpha-helices, which follows a two-state mechanism). As the introduced stabilizing interactions are local in nature, the energy balance between the contribution of local and nonlocal interactions changes considerably. Recent theoretical analyses of protein folding using simplified models have indicated that optimization of folding speed requires this balance to be biased towards nonlocal interactions. To determine whether this is the case, we study here the folding kinetics of two ADA2h mutants in which alpha-helix 1 (mutant M1) or 2 (mutant M2) has been stabilized through local interactions, as well as the equilibrium and kinetic behaviour of a double mutant (DM) in which both helices have been stabilized.

RESULTS

The stability of DM is considerably enhanced with respect to wild type (WI) and this mutant can be considered as a thermoresistant protein (Tm > 363 K). The thermodynamic parameters obtained by chemical denaturation (urea and GdnHCl) show that DM is approximately 2.6 kcal mol-1 more stable than WT. The effects on folding kinetics are different in each of the single mutants. M1 shows very little effect in refolding, while its unfolding is greatly decelerated with respect to WT. M2 shows, together with a deceleration in unfolding, a significant acceleration in refolding. As with equilibrium parameters, the kinetics of the double mutant can be explained by the simple addition of the effects found in each single mutant. Interestingly enough, the refolding slope mkf in mutants M2 and DM is smaller than in the wild-type and M1 mutant.

CONCLUSIONS

Thermoresistance can be achieved, in some cases, by increasing favourable native local interactions. The balance between local and nonlocal interactions can be significantly changed in some proteins and still keep a cooperative unfolding transition similar to that of the wild type. The introduction of favourable local interactions by mutational redesign can also be used to increase the folding speed of certain proteins, showing that not all proteins in nature have been optimized for rapid folding, contrary to what has been theoretically indicated. This behaviour is probably also shared by other polypeptides with highly unstructured denatured states. All these phenomena have been shown experimentally in ADA2h by mutations that increase helix stability. However, the effects promoted for such an approach in proteins with residual structure and/or intermediates in the denatured ensemble could be different. This has been shown by experiments performed on CheY in which the cooperativity of the folding process was greatly affected.

Protein folding: the endgame

The last stage of protein folding, the "endgame," involves the ordering of amino acid side-chains into a well defined and closely packed configuration. We review a number of topics related to this process. We first describe how the observed packing in protein crystal structures is measured. Such measurements show that the protein interior is packed exceptionally tightly, more so than the protein surface or surrounding solvent and even more efficiently than crystals of simple organic molecules. In vitro protein folding experiments also show that the protein is close-packed in solution and that the tight packing and intercalation of side-chains is a final and essential step in the folding pathway. These experimental observations, in turn, suggest that a folded protein structure can be described as a kind of three-dimensional jigsaw puzzle and that predicting side-chain packing is possible in the sense of solving this puzzle. The major difficulty that must be overcome in predicting side-chain packing is a combinatorial "explosion" in the number of possible configurations. There has been much recent progress towards overcoming this problem, and we survey a variety of the approaches. These approaches differ principally in whether they use ab initio (physical) or more knowledge-based methods, how they divide up and search conformational space, and how they evaluate candidate configurations (using scoring functions). The accuracy of side-chain prediction depends crucially on the (assumed) positioning of the main-chain. Methods for predicting main-chain conformation are, in a sense, not as developed as that for side-chains. We conclude by surveying these methods. As with side-chain prediction, there are a great variety of approaches, which differ in how they divide up and search space and in how they score candidate conformations.

From Levinthal to pathways to funnels

While the classical view of protein folding kinetics relies on phenomenological models, and regards folding intermediates in a structural way, the new view emphasizes the ensemble nature of protein conformations. Although folding has sometimes been regarded as a linear sequence of events, the new view sees folding as parallel microscopic multi-pathway diffusion-like processes. While the classical view invoked pathways to solve the problem of searching for the needle in the haystack, the pathway idea was then seen as conflicting with Anfinsen's experiments showing that folding is pathway-independent (Levinthal's paradox). In contrast, the new view sees no inherent paradox because it eliminates the pathway idea: folding can funnel to a single stable state by multiple routes in conformational space. The general energy landscape picture provides a conceptual framework for understanding both two-state and multi-state folding kinetics. Better tests of these ideas will come when new experiments become available for measuring not just averages of structural observables, but also correlations among their fluctuations. At that point we hope to learn much more about the real shapes of protein folding landscapes.

Folding of cytochrome c initiated by submillisecond mixing

Cytochrome c folding was initiated using a new solution mixer that provides a time window which covers over 90% of the burst phase unresolved by conventional stop-flow measurements. Folding was followed by resonance Raman scattering. Kinetic analysis of the high frequency Raman data indicates that a nascent phase occurs within the mixing dead time of 100 microseconds. A significant fraction of the protein was found to be trapped in a misfolded bis-histidine form during the nascent phase at pH 4.5, thereby preventing the protein from folding rapidly and homogeneously. The nascent phase was followed by a haem-ligand exchange phase that populates the native histidine-methionine coordinated form through a thermodynamically controlled equilibrium.

Protein folding kinetics: timescales, pathways and energy landscapes in terms of sequence-dependent properties

BACKGROUND

Recent experimental and theoretical studies have revealed that protein folding kinetics can be quite complex and diverse depending on various factors such as size of the protein sequence and external conditions. For example, some proteins fold apparently in a kinetically two-state manner, whereas others follow complex routes to the native state. We have set out to provide a theoretical basis for understanding the diverse behavior seen in the refolding kinetics of proteins in terms of properties that are intrinsic to the sequence.

RESULTS

The folding kinetics of a number of sequences for off-lattice continuum models of proteins is studied using Langevin simulations at two different values of the friction coefficient. We show for these models that there is a remarkable correlation between folding time, tau F, and sigma = (T theta - TF)/T theta, where T theta and TF are the equilibrium collapse and folding transition temperatures, respectively. The microscopic dynamics reveals that several scenarios for the kinetics of refolding arise depending on the range of values of sigma. For relatively small sigma, the chain reaches the native conformation by a direct native conformation nucleation collapse (NCNC) mechanism without being trapped in any detectable intermediates. For moderate and large values of sigma, the kinetics is described by the kinetic partitioning mechanism, according to which a fraction of molecules phi (kinetic partition factor) reach the native conformation via the NCNC mechanism. The remaining fraction attains the native state by off-pathway processes that involve trapping in several misfolded structures. The rate-determining step in the off-pathway processes is the transition from the misfolded structures to the native state. The partition factor phi is also determined by sigma: the smaller the value of sigma, the larger is phi. The qualitative aspects of our results are found to be independent of the friction coefficient. The simulation results and theoretical arguments are used to obtain estimates for timescales for folding via the NCNC mechanism in small proteins, those with less than about 70 amino acid residues.

CONCLUSIONS

We have shown that the various scenarios for folding of proteins, and possibly other biomolecules, can be classified solely in terms of sigma. Proteins with small values of sigma reach the native conformation via a nucleation collapse mechanism and their energy landscape is characterized by having one dominant native basin of attraction (NBA). On the other hand, proteins with large sigma get trapped in competing basins of attraction (CBAs) in which they adopt misfolded structures. Only a small fraction of molecules access the native state rapidly when sigma is large. For these sequences, the majority of the molecules approach the native state by a three-stage multipathway mechanism in which the rate-determining step involves a transition from one of the CBAs to the NBA.

Ligand exchange during cytochrome c folding

Submillisecond folding of cytochrome c reveals that a nascent phase appears within the mixing dead time of 100 microseconds, followed by a ligand exchange reaction during which His 26/33, water and Met 80 are inter-exchanged as haem ligands through a thermodynamically controlled equilibrium. In the ligand exchange phase, the rate of formation of a misfolded histidine-histidine coordinated state (HH) decreases by two orders of magnitude as the pH is reduced from 5.9 to 4.5 due to the protonation of the misligated His 26/33. The activation energy barriers for the transitions from the histidine-water coordinated form (HW) to the histidine-methionine coordinated form and the HH form are 18 and 4 kcal mol-1 respectively, at pH 4.8. The activation energy barrier for protein to escape from the misligated HH to the HW form was measured to be 12 kcal mol-1, demonstrating the kinetic trapping effect of the misligated bis-histidine form. The development of the polypeptide tertiary structure near the haem is concomitant with the coordination of the native haem axial ligand.

Kinetic intermediates in the formation of the cytochrome c molten globule

The relationship between molten globules and transient intermediates in protein folding has been explored by equilibrium and kinetic analysis of the compact acid-denatured A-state of cytochrome c. The chloride-induced formation of the A-state is a complex reaction with structural intermediates resembling those found under native refolding conditions, including a rapidly formed compact state and a subsequent intermediate with interacting N- and C-terminal helices. Together with mutational evidence for specific helix-helix packing interactions, this shows that the A-state is a stable analogue of a late folding intermediate. The L94A mutation blocks all folding steps after the initial collapse and its equilibrium state resembles early kinetic intermediates.

Direct electrochemical evidence for an equilibrium intermediate in the guanidine-induced unfolding of cytochrome c

This paper reports a voltammetric and spectroscopic investigation of the guanidine-induced unfolding of cytochrome c at neutral pH and 25 degrees C. Electrochemical data provide direct evidence for the presence of an equilibrium intermediate (form I) strictly dependent on the denaturant concentration. The midpoint potential of form I has been determined (E1/2 = +0.010 V vs. NHE) and its structural features defined from analysis of the circular dichroism and absorbance spectroscopy data obtained under the same experimental conditions. From the correlation of electrochemical and spectroscopic data, we propose that the features detected by the intermediate conform to the molten globule state.

Fast events in protein folding

Understanding how proteins fold is one of the central problems in biochemistry. A new generation of kinetic experiments has emerged to investigate the mechanisms of protein folding on the previously inaccessible submillisecond time scale. These experiments provide the first glimpse of processes such as secondary structure formation, local hydrophobic collapse, global collapse to compact denatured states, and fast barrier crossings to the native state. Key results are summarized and discussed in terms of the statistical energy landscape theory of protein folding.

Diffusion-limited contact formation in unfolded cytochrome c: estimating the maximum rate of protein folding

How fast can a protein fold? The rate of polypeptide collapse to a compact state sets an upper limit to the rate of folding. Collapse may in turn be limited by the rate of intrachain diffusion. To address this question, we have determined the rate at which two regions of an unfolded protein are brought into contact by diffusion. Our nanosecond-resolved spectroscopy shows that under strongly denaturing conditions, regions of unfolded cytochrome separated by approximately 50 residues diffuse together in 35-40 microseconds. This result leads to an estimate of approximately (1 microsecond)-1 as the upper limit for the rate of protein folding.

A fast conformational search strategy for finding low energy structures of model proteins

We describe a new computer algorithm for finding low-energy conformations of proteins. It is a chain-growth method that uses a heuristic bias function to help assemble a hydrophobic core. We call it the Core-directed chain Growth method (CG). We test the CG method on several well-known literature examples of HP lattice model proteins [in which proteins are modeled as sequences of hydrophobic (H) and polar (P) monomers], ranging from 20-64 monomers in two dimensions, and up to 88-mers in three dimensions. Previous nonexhaustive methods--Monte Carlo, a Genetic Algorithm, Hydrophobic Zippers, and Contact Interactions--have been tried on these same model sequences. CG is substantially better at finding the global optima, and avoiding local optima, and it does so in comparable or shorter times. CG finds the global minimum energy of the longest HP lattice model chain for which the global optimum is known, a 3D 88-mer that has only been reachable before by the CHCC complete search method. CG has the potential advantage that it should have nonexponential scaling with chain length. We believe this is a promising method for conformational searching in protein folding algorithms.

Different folding transition states may result in the same native structure

The crystal structures of two circular permutants of the alpha-spectrin SH3 domain with new termini within the RT loop (S19-P20s) and the distal loop (N47-D48s) have been determined at 2.02 and 1.77 A resolution respectively. Both fold into the same three-dimensional structure as the wild-type SH3 domain except for the engineered loop that fuses the wild-type termini. The cleaved RT loop in S19-P20s loses nine conserved hydrogen bonds through local hydrogen bond unzipping; no hydrogen bond unzipping occurs in N47-D48s. The structures of the transition states for folding of wild-type alpha-spectrin SH3 domain and the two circular permutants have been examined by analysis of the folding kinetics of eight strategically distributed point mutants. Unlike the native structures, the transition states of the three proteins are considerably different, suggesting that there is no direct relationship between these two states in a protein.

Protein secondary structural types are differentially coded on messenger RNA

Tricodon regions on messenger RNAs corresponding to a set of proteins from Escherichia coli were scrutinized for their translation speed. The fractional frequency values of the individual codons as they occur in mRNAs of highly expressed genes from Escherichia coli were taken as an indicative measure of the translation speed. The tricodons were classified by the sum of the frequency values of the constituent codons. Examination of the conformation of the encoded amino acid residues in the corresponding protein tertiary structures revealed a correlation between codon usage in mRNA and topological features of the encoded proteins. Alpha helices on proteins tend to be preferentially coded by translationally fast mRNA regions while the slow segments often code for beta strands and coil regions. Fast regions correspondingly avoid coding for beta strands and coil regions while the slow regions similarly move away from encoding alpha helices. Structural and mechanistic aspects of the ribosome peptide channel support the relevance of sequence fragment translation and subsequent conformation. A discussion is presented relating the observation to the reported kinetic data on the formation and stabilization of protein secondary structural types during protein folding. The observed absence of such strong positive selection for codons in non-highly expressed genes is compatible with existing theories that mutation pressure may well dominate codon selection in non-highly expressed genes.

Insights into protein folding from NMR

NMR has emerged as an important tool for studies of protein folding because of the unique structural insights it can provide into many aspects of the folding process. Applications include measurements of kinetic folding events and structural characterization of folding intermediates, partly folded states, and unfolded states. Kinetic information on a time scale of milliseconds or longer can be obtained by real-time NMR experiments and by quench-flow hydrogen-exchange pulse labeling. Although NMR cannot provide direct information on the very rapid processes occurring during the earliest stages of protein folding, studies of isolated peptide fragments provide insights into likely protein folding initiation events. Multidimensional NMR techniques are providing new information on the structure and dynamics of protein folding intermediates and both partly folded and unfolded states.

Requirements for perpendicular helix pairing

Cassette mutagenesis was used to produce a library of mutations at the interface of the N- and C-terminal helices of Saccharomyces cerevisiae iso-1-cytochrome c. The library is random and comprises > 98% mutations. Over 11,000 candidates were assayed for function by selecting for the ability of yeast, with the mutated gene as their sole cytochrome c source, to grow on nonfermentable carbon sources. We estimate that approximately 0.5% of the 160,000 total amino acid combinations at these four residues result in a functional cytochrome c. Significant correlations are found between the phenotype of yeast harboring the alleles and both the Dayhoff mutation matrix and transfer free energies (cyclohexane-to-water and n-octanol-to-water). Similar correlations are observed with respect to growth rate. Finally, sequences that are consistent with function follow a binary amino acid pattern.

Detection of rare partially folded molecules in equilibrium with the native conformation of RNaseH

Despite the general observation that single domain proteins denature in a completely cooperative manner, amide hydrogen exchange of ribonuclease H in low levels of denaturant demonstrates the existence of two partially folded species. The structures of these marginally stable species resemble kinetic folding intermediates and the molten globule state of the protein. These data suggest that the first region to fold is the thermodynamically most stable portion of the protein and that the molten globule is a high free energy conformation present at equilibrium in the native state.

Fast-folding experiments and the topography of protein folding energy landscapes

The rapid folding of certain proteins can be described theoretically using an energy landscape in the shape of a folding funnel. New techniques have allowed the examination of fast-folding events that occur in microseconds or less, and have tested the predictions of the theoretical models.

Cytochrome c folding triggered by electron transfer

BACKGROUND

Experimental and theoretical studies of protein folding suggest that the free-energy change associated with the folding process is a primary factor in determining folding rates. We have recently developed a photochemical electron-transfer-triggering method to study protein-folding kinetics over a wide range of folding free energies. Here, we have used this technique to investigate the relationship between folding rate and free-energy change using cytochromes c from horse (h-cyt c) and yeast (y-cyt c), which have similar backbone folds but different amino-acid sequences and, consequently, distinct folding energies.

RESULTS

The folding free energies for oxidized and reduced h-cyt c and y-cyt c are linear functions of the denaturant (guanidine hydrochloride) concentration, but the concentration required to unfold half of the protein is 1.5 M lower for y-cyt c. We measured the folding rates of reduced h-cyt c and y-cyt c over a range of guanidine hydrochloride concentrations at two temperatures. When driving forces are matched at the appropriate denaturant concentrations, the two homologs have comparable folding rates. The activation free energies for folding h-cyt c and y-cyt c are linearly dependent on the folding free energies. The slopes of these lines are similar (approximately 0.4) for the two proteins, suggesting an early transition state along the folding reaction coordinate.

CONCLUSIONS

The free-energy relationships found for h-cyt c and y-cyt c folding kinetics imply that the height of the barrier to folding depends upon the relative stabilities of the unfolded and folded states. The striking correspondence in rate/free-energy profiles for h-cyt c and y-cyt c suggests that, despite low sequence homology, they follow similar folding pathways.

Chaperonin-facilitated protein folding: optimization of rate and yield by an iterative annealing mechanism

We develop a heuristic model for chaperonin-facilitated protein folding, the iterative annealing mechanism, based on theoretical descriptions of "rugged" conformational free energy landscapes for protein folding, and on experimental evidence that (i) folding proceeds by a nucleation mechanism whereby correct and incorrect nucleation lead to fast and slow folding kinetics, respectively, and (ii) chaperonins optimize the rate and yield of protein folding by an active ATP-dependent process. The chaperonins GroEL and GroES catalyze the folding of ribulose bisphosphate carboxylase at a rate proportional to the GroEL concentration. Kinetically trapped folding-incompetent conformers of ribulose bisphosphate carboxylase are converted to the native state in a reaction involving multiple rounds of quantized ATP hydrolysis by GroEL. We propose that chaperonins optimize protein folding by an iterative annealing mechanism; they repeatedly bind kinetically trapped conformers, randomly disrupt their structure, and release them in less folded states, allowing substrate proteins multiple opportunities to find pathways leading to the most thermodynamically stable state. By this mechanism, chaperonins greatly expand the range of environmental conditions in which folding to the native state is possible. We suggest that the development of this device for optimizing protein folding was an early and significant evolutionary event.

Barriers to protein folding: formation of buried polar interactions is a slow step in acquisition of structure

In the MYL mutant of the Arc repressor dimer, sets of partially buried salt-bridge and hydrogen-bond interactions mediated by Arg-31, Glu-36, and Arg-40 in each subunit are replaced by hydrophobic interactions between Met-31, Tyr-36, and Leu-40. The MYL refolding/dimerization reaction differs from that of wild type in being 10- to 1250-fold faster, having an earlier transition state, and depending upon viscosity but not ionic strength. Formation of the wild-type salt bridges in a hydrophobic environment clearly imposes a kinetic barrier to folding, which can be lowered by high salt concentrations. The changes in the position of the transition state and viscosity dependence can be explained if denatured monomers interact to form a partially folded dimeric intermediate, which then continues folding to form the native dimer. The second step is postulated to be rate limiting for wild type. Replacing the salt bridge with hydrophobic interactions lowers this barrier for MYL. This makes the first kinetic barrier rate limiting for MYL refolding and creates a downhill free-energy landscape in which most molecules which reach the intermediate state continue to form native dimers.

Molecular collapse: the rate-limiting step in two-state cytochrome c folding

Experiments with cytochrome c (cyt c) show that an initial folding event, molecular collapse, is not an energetically downhill continuum as commonly presumed but represents a large-scale, time-consuming, cooperative barrier-crossing process. In the absence of later misfold-reorganization barriers, the early collapse barrier limits cyt c folding to a time scale of milliseconds. The collapse process itself appears to be limited by an uphill search for some coarsely determined transition state structure that can nucleate subsequent energetically downhill folding events. An earlier "burst phase" event at strongly native conditions appears to be a non-specific response of the unfolded chain to reduced denaturant concentration. The molecular collapse process may or may not require the co-formation of the amino- and carboxyl-terminal helices, which are present in an initial metastable intermediate directly following the rate-limiting collapse. After the collapse-nucleation event, folding can proceed rapidly in an apparent two-state manner, probably by way of a predetermined sequence of metastable intermediates that leads to the native protein structure (Bai et al., Science 269:192-197, 1995).

The role of helix formation in the folding of a fully alpha-helical coiled coil

To determine when secondary structure forms as two chains coalesce to form an alpha-helical dimer, the folding rates of variants of the coiled coil region of GCN4 were compared. Residues at non-perturbing positions along the exterior length of the helices were substituted one at a time with alanine and glycine to vary helix propensity and therefore dimer stability. For all variants, the bimolecular folding rate remains largely unchanged; the unfolding rate changes to largely account for the change in stability. Thus, contrary to most folding models, widespread helix is not yet formed at the rate-limiting step in the folding pathway. The high-energy transition state is a collapsed form that contains little if any secondary structure, as suggested for the globular protein cytochrome c (Sosnick et al., Proteins 24: 413-426, 1996).

Molecular dynamics simulations of N-terminal peptides from a nucleotide binding protein

Molecular dynamics (MD) simulations of N-terminal peptides from lactate dehydrogenase (LDH) with increasing length and individual secondary structure elements were used to study their stability in relation to folding. Ten simulations of 1-2 ns of different peptides in water starting from the coordinates of the crystal structure were performed. The stability of the peptides was compared qualitatively by analyzing the root mean square deviation (RMSD) from the crystal structure, radius of gyration, secondary and tertiary structure, and solvent accessible surface area. In agreement with earlier MD studies, relatively short (< 15 amino acids) peptides containing individual secondary structure elements were generally found to be unstable; the hydrophobic alpha 1-helix of the nucleotide binding fold displayed a significantly higher stability, however. Our simulations further showed that the first beta alpha beta supersecondary unit of the characteristic dinucleotide binding fold (Rossmann fold) of LDH is somewhat more stable than other units of similar length and that the alpha 2-helix, which unfolds by itself, is stabilized by binding to this unit. This finding suggests that the first beta alpha beta unit could function as an N-terminal folding nucleus, upon which the remainder of the polypeptide chain can be assembled. Indeed, simulations with longer units (beta-alpha-beta-alpha and beta-alpha-beta-alpha beta-beta) showed that all structural elements of these units are rather stable. The outcome of our studies is in line with suggestions that folding of the N-terminal portion of LDH in vivo can be a cotranslational process that takes place during the ribosomal peptide synthesis.