The folding pathway of T4 lysozyme: the high-resolution structure and folding of a hidden intermediate

Folding intermediates have been detected and characterized for many proteins. However, their structures at atomic resolution have only been determined for two small single domain proteins: Rd-apocytochrome b(562) and engrailed homeo domain. T4 lysozyme has two easily distinguishable but energetically coupled domains: the N and C-terminal domains. An early native-state hydrogen exchange experiment identified an intermediate with the C-terminal domain folded and the N-terminal domain unfolded. We have used a native-state hydrogen exchange-directed protein engineering approach to populate this intermediate and demonstrated that it is on the folding pathway and exists after the rate-limiting step. Here, we determined its high-resolution structure and the backbone dynamics by multi-dimensional NMR methods. We also characterized the folding behavior of the intermediate using stopped-flow fluorescence, protein engineering, and native-state hydrogen exchange. Unlike the folding intermediates of the two single-domain proteins, which have many non-native side-chain interactions, the structure of the hidden folding intermediate of T4 lysozyme is largely native-like. It folds like many small single domain proteins. These results have implications for understanding the folding mechanism and evolution of multi-domain proteins.     

Structural examination of phi-value analysis in protein folding

Protein folding intermediates and transition states are commonly characterized using a protein engineering procedure (Phi-value analysis) based on several assumptions, including (1) intermediates and transition states have native-like conformations and (2) single mutations from larger hydrophobic residues to smaller ones do not perturb their structures. Although Phi-value analysis has been widely used, these assumptions have not been tested to date because of the lack of high-resolution structures of intermediates and transition states. We recently have determined the structure of a folding intermediate for a four-helix bundle protein (Rd-apocytochrome b(562)) using NMR. The intermediate has the N-terminal helix unfolded. The other three helices fold in a native-like topology with extensive non-native hydrophobic interactions. Here, we have determined the Phi values for 14 hydrophobic core residues, including those with significant non-native interactions. All of the Phi values are in the normal range from 0 to 1, indicating that these non-native interactions cannot be identified by the common Phi-value analysis, and therefore, the first assumption is not valid for this intermediate. We also determined the structure of a mutant (F65A) of the intermediate and found that the structure of the intermediate is not perturbed by the mutation, supporting the second assumption. Together, these results suggest that Phi-value analysis may be valid for characterizing the energetics of the interactions between the mutated residue and others, but not for determining the detailed structures of intermediates and transition states because non-native interactions may exist and may not be identifiable by the common Phi-value analysis.     

Probing the folding intermediate of Rd-apocyt b562 by protein engineering and infrared T-jump

Small proteins often fold in an apparent two-state manner with the absence of detectable early-folding intermediates. Recently, using native-state hydrogen exchange, intermediates that exist after the rate-limiting transition state have been identified for several proteins. However, little is known about the folding kinetics from these post-transition intermediates to their corresponding native states. Herein, we have used protein engineering and a laser-induced temperature-jump (T-jump) technique to investigate this issue and have applied it to Rd-apocyt b(562) , a four-helix bundle protein. Previously, it has been shown that Rd-apocyt b(562) folds via an on-pathway hidden intermediate, which has only the N-terminal helix unfolded. In the present study, a double mutation (V16G/I17A) in the N-terminal helix of Rd-apocyt b(562) was made to further increase the relative population of this intermediate state at high temperature by selectively destabilizing the native state. In the circular dichroism thermal melting experiment, this mutant showed apparent two-state folding behavior. However, in the T-jump experiment, two kinetic phases were observed. Therefore, these results are in agreement with the idea that a folding intermediate is populated on the folding pathway of Rd-apocyt b(562) . Moreover, it was found that the exponential growth rate of the native state from this intermediate state is roughly (25 microsec)(-1) at 65 degrees C.     

The folding pathway of barnase: the rate-limiting transition state and a hidden intermediate under native conditions

The nature of the rate-limiting transition state at zero denaturant (TS(1)) and whether there are hidden intermediates are the two major unsolved problems in defining the folding pathway of barnase. In earlier studies, it was shown that TS(1) has small phi values throughout the structure of the protein, suggesting that the transition state has either a defined partially folded secondary structure with all side chains significantly exposed or numerous different partially unfolded structures with similar stability. To distinguish the two possibilities, we studied the effect of Gly mutations on the folding rate of barnase to investigate the secondary structure formation in the transition state. Two mutations in the same region of a beta-strand decreased the folding rate by 20- and 50-fold, respectively, suggesting that the secondary structures in this region are dominantly formed in the rate-limiting transition state. We also performed native-state hydrogen exchange experiments on barnase at pD 5.0 and 25 degrees C and identified a partially unfolded state. The structure of the intermediate was investigated using protein engineering and NMR. The results suggest that the intermediate has an omega loop unfolded. This intermediate is more folded than the rate-limiting transition state previously characterized at high denaturant concentrations (TS(2)). Therefore, it exists after TS(2) in folding. Consistent with this conclusion, the intermediate folds with the same rate and denaturant dependence as the wild-type protein, but unfolds faster with less dependence on the denaturant concentration. These and other results in the literature suggest that barnase folds through partially unfolded intermediates that exist after the rate-limiting step. Such folding behavior is similar to those of cytochrome c and Rd-apocyt b(562). Together, we suggest that other small apparently two-state proteins may also fold through hidden intermediates.     

The chicken-egg scenario of protein folding revisited

What is the first step in protein folding - hydrophobic collapse (compaction) or secondary structure formation? It is still not clear if the major driving force in protein folding is hydrogen bonding or hydrophobic interactions or both. We analyzed data on the conformational characteristics of 41 globular proteins in native and partially folded conformational states. Our analysis shows that a good correlation exists between relative decrease in hydrodynamic volume and increase in secondary structure content. No compact equilibrium intermediates lacking secondary structure, or highly ordered non-compact species, were found. This correlation provides experimental support for the hypothesis that hydrophobic collapse occurs simultaneously with formation of secondary structure in the early stages of the protein folding.     

Identification of native and non-native structure in kinetic folding intermediates of apomyoglobin

Site-directed mutagenesis has been used to probe the interactions that stabilize the equilibrium and burst phase kinetic intermediates formed by apomyoglobin. Nine bulky hydrophobic residues in the A, E, G and H helices were replaced by alanine, and the effects on protein stability and kinetic folding pathways were determined. Hydrogen exchange pulse-labeling experiments, with NMR detection, were performed for all mutants. All of the alanine substitutions resulted in changes in proton occupancy or an increased rate of hydrogen-deuterium exchange for amides in the immediate vicinity of the mutation. In addition, most mutations affected residues in distant parts of the amino acid sequence, providing insights into the topology of the burst phase intermediate and the interactions that stabilize its structure. Differences between the pH 4 equilibrium molten globule and the kinetic intermediate are evident: the E helix region plays no discernible role in the equilibrium intermediate, but contributes significantly to stabilization of the ensemble of compact intermediates formed during kinetic refolding. Mutations that interfere with docking of the E helix onto the preformed A/B/G/H helix core substantially decrease the folding rate, indicating that docking and folding of the E helix region occurs prior to formation of the apomyoglobin folding transition state. The results of the mutagenesis experiments are consistent with rapid formation of an ensemble of compact burst phase intermediates with an overall native-like topological arrangement of the A, B, E, G, and H helices. However, the experiments also point to disorder in docking of the E helix and to non-native contacts in the kinetic intermediate. In particular, there is evidence for translocation of the H helix by approximately one helical turn towards its N terminus to maximize hydrophobic interactions with helix G. Thus, the burst phase intermediate observed during kinetic refolding of apomyoglobin consists of an ensemble of compact, kinetically trapped states in which the helix docking appears to be topologically correct, but in which there are local non-native interactions that must be resolved before the protein can fold to the native structure.     

Aromatic residues engineered into the beta-turn nucleation site of ubiquitin lead to a complex folding landscape, non-native side-chain interactions, and kinetic traps

The fast folding of small proteins is likely to be the product of evolutionary pressures that balance the search for native-like contacts in the transition state with the minimum number of stable non-native interactions that could lead to partially folded states prone to aggregation and amyloid formation. We have investigated the effects of non-native interactions on the folding landscape of yeast ubiquitin by introducing aromatic substitutions into the beta-turn region of the N-terminal beta-hairpin, using both the native G-bulged type I turn sequence (TXTGK) as well as an engineered 2:2 XNGK type I' turn sequence. The N-terminal beta-hairpin is a recognized folding nucleation site in ubiquitin. The folding kinetics for wt-Ub (TLTGK) and the type I' turn mutant (TNGK) reveal only a weakly populated intermediate, however, substitution with X = Phe or Trp in either context results in a high propensity to form a stable compact intermediate where the initial U-->I collapse is visible as a distinct kinetic phase. The introduction of Trp into either of the two host turn sequences results in either complex multiphase kinetics with the possibility of parallel folding pathways, or formation of a highly compact I-state stabilized by non-native interactions that must unfold before refolding. Sequence substitutions with aromatic residues within a localized beta-turn capable of forming non-native hydrophobic contacts in both the native state and partially folded states has the undesirable consequence that folding is frustrated by the formation of stable compact intermediates that evolutionary pressures at the sequence level may have largely eliminated.     

Design and folding of a multidomain protein

To test whether the folding process of a large protein can be understood on the basis of the folding behavior of the domains that constitute it, we coupled two well-studied small -helical proteins, the B-domain of protein A (60 amino acids) and Rd-apocytochrome b562 (Rd-apocyt b562, 106 amino acids), by fusing the C-terminal helix of the B-domain of protein A with the N-terminal helix of Rd-apocyt b562 without changing their hydrophobic core residues. The success of the design was confirmed by determining the structure of the engineered protein with multidimensional NMR methods. Kinetic studies showed that the logarithms of the folding/unfolding rate constants of the engineered protein are linearly dependent on concentrations of guanidinium chloride in the measurable range from 1.7 to 4 M. Their slopes (m-values) are close to those of Rd-apocyt b562. In addition, the 1H-15N HSQC spectrum taken at 1.5 M guanidinium chloride reveals that only the Rd-apocyt b562 domain in the designed protein remained folded. These results suggest that the two domains have weak energetic coupling. Interestingly, the redesigned protein folds faster than Rd-apocyt b562, suggesting that the fused helix stabilizes the rate-limiting transition state.     

Fluorescence energy transfer indicates similar transient and equilibrium intermediates in staphylococcal nuclease folding

Fluorescence resonance energy transfer (FRET) is one of the few methods available to measure the rate at which a folding protein collapses. Using staphylococcal nuclease in which a cysteine residue was engineered in place of Lys64, permitted FRET measurements of the distance between the donor tryptophan 140 and 5-[[2-[(iodoacetyl)-amino]ethyl]amino]naphthalene-1-sulfonic acid-labeled Cys64. These measurements were undertaken on both equilibrium partially folded intermediates at low pH (A states), as well as transient intermediates during stopped-flow refolding. The results indicate that there is an initial collapse of the protein in the deadtime of the stopped-flow instrument, corresponding to a regain of approximately 60% of the native signal, followed by three slower transients. This is in contrast to circular dichroism measurements which show only 20-25% regain of the native secondary structure in the burst phase. Thus hydrophobic collapse precedes the formation of substantial secondary structure. The first two detected transient intermediate species have FRET properties essentially identical with those of the previously characterized equilibrium A state intermediates, suggesting similar structures between the equilibrium and transient intermediates.The effects of anions on the folding of acid-unfolded staphylococcal nuclease, and urea on the unfolding of the resulting A states, indicates that in folding the protein becomes compact prior to formation of major secondary structure, whereas in unfolding the protein expands prior to major loss of secondary structure. Comparison of the kinetics of refolding of staphylococcal nuclease, monitored by FRET, and for a proline-free variant, indicate that folding occurs via two partially folded intermediates leading to a native-like species with one (or more) proline residues in a non-native conformation. For the A states an excellent correlation between compactness measured by FRET, and compactness determined from small-angle X-ray scattering, was observed. Further, a linear relationship between compactness and free energy of unfolding was noted. Formation of soluble aggregates of the A states led to dramatic enhancement of the FRET, consistent with intermolecular fluorescence energy transfer.     

Conformational heterogeneity of an equilibrium folding intermediate quantified and mapped by scanning mutagenesis

It is challenging to experimentally define an energy landscape for protein folding that comprises multiple partially unfolded states. Experimental results are often ambiguous as to whether a non-native state is conformationally homogeneous. Here, we tested an approach combining systematic mutagenesis and a Br?nsted-like analysis to reveal and quantify conformational heterogeneity of folding intermediate states. Using this method, we resolved an otherwise apparently homogeneous equilibrium folding intermediate of Borrelia burgdorferi OspA into two conformationally distinct species and determined their relative populations. Furthermore, we mapped the structural differences between these intermediate species, which are consistent with the non-native species that we previously proposed based on native-state hydrogen exchange studies. When treated as a single state, the intermediate ensemble exhibited fractional Phi-values for mutations and Hammond-type behaviors that are often observed for folding transition states. We found that a change in relative population of the two species within the intermediate ensemble explains these properties well, suggesting that fractional Phi-values and Hammond-type behaviors exhibited by folding intermediates and transition states may arise more often from conformational heterogeneity than from a single partial structure. Our results are consistent with the presence of multiple minima in a rugged energy landscape predicted from theoretical studies. The method described here provides a promising means to probe a complex folding energy landscape.     

Characterization of a quaternary-structured folding intermediate of an antibody Fab-fragment.

Antibody folding is a complex process comprising folding and association reactions. Although it is usually difficult to characterize kinetic folding intermediates, in the case of the antibody Fab fragment, domain-domain interactions lead to a rate-limiting step of folding, thus accumulating folding intermediates at a late step of folding. Here, we analyzed a late folding intermediate of the Fab fragment of the monoclonal antibody MAK 33 from mouse (kappa/IgG1). As a strategy for accumulation of this intermediate we used partial denaturation of the native Fab by guanidinium chloride. This denaturation intermediate, which can be populated to about 90%, is indistinguishable from a late-folding intermediate with respect to denaturation and renaturation kinetics. The spectroscopic analysis reveals a native-like secondary structure of this intermediate with aromatic side chains only slightly more solvent exposed than in the native state. The respective partner domains are weekly associated. From these data we conclude that the intramolecular association of the two chains during folding, with all domains in a native-like structure, follows a two-step mechanism. In this mechanism, presumably hydrophobic interactions are followed by rearrangements leading to the exact complementarity of the contact sites of the respective domains.     

Structural and kinetic characterization of early folding events in beta-lactoglobulin

We have defined the structural and dynamic properties of an early folding intermediate of beta-lactoglobulin known to contain non-native alpha-helical structure. The folding of beta-lactoglobulin was monitored over the 100 micros--10 s time range using ultrarapid mixing techniques in conjunction with fluorescence detection and hydrogen exchange labeling probed by heteronuclear NMR. An initial increase in Trp fluorescence with a time constant of 140 micros is attributed to formation of a partially helical compact state. Within 2 ms of refolding, well protected amide protons indicative of stable hydrogen bonded structure were found only in a domain comprising beta-strands F, G and H, and the main alpha-helix, which was thus identified as the folding core of beta-lactoglobulin. At the same time, weak protection (up to approximately 10-fold) of amide protons in a segment spanning residues 12--21 is consistent with formation of marginally stable non-native alpha-helices near the N-terminus. Our results indicate that efficient folding, despite some local non-native structural preferences, is insured by the rapid formation of a native-like alpha/beta core domain.     

Role of the B helix in early folding events in apomyoglobin: evidence from site-directed mutagenesis for native-like long range interactions

The folding pathways of four mutants in which bulky hydrophobic residues in the B helix of apomyoglobin (ApoMb) are replaced by alanine (I28A, L29A, I30A, and L32A) have been analyzed using equilibrium and kinetic methods employing NMR, CD, fluorescence and mass spectrometry. Hydrogen exchange pulse-labeling followed by mass spectrometry reveals detectable intermediates in the kinetic folding pathways of each of these mutants. Comparison of the quench-flow data analyzed by NMR for the wild-type protein and the mutants showed that the substitutions I28A, L29A and L32A lead to destabilization of the B helix in the burst phase kinetic intermediate, relative to wild-type apomyoglobin. In contrast, the I30A mutation apparently has a slight stabilizing effect on the B helix in the burst phase intermediate; under weak labeling conditions, residues in the C helix region were also relatively stabilized in the mutant compared to the wild-type protein. This suggests that native-like helix B/helix C packing interactions occur in the folding intermediate. The L32A mutant showed significantly lower proton occupancies in the burst phase for several residues in the G helix, specifically F106, I107, E109 and A110, which are in close proximity to L32 in the X-ray structure of myoglobin, providing direct evidence that native-like helix B/helix G contacts are formed in the apomyoglobin burst phase intermediate. The L29A mutation resulted in an increase in burst phase proton occupancies for several residues in the E helix. Since these regions of the B and E helices are not in contact in the native myoglobin structure, these effects suggest the possibility of non-native B/E packing interactions in the kinetic intermediate. The differing effects of these B helix mutations on the apomyoglobin folding process suggests that each side-chain plays a different and important role in forming stable structure in the burst phase intermediate, and points to a role for both native-like and non-native contacts in stabilization of the folding intermediate.     

Substrate-induced activation of a trapped IMC-mediated protein folding intermediate

While several unfolded proteins acquire native structures through distinct folding intermediates, the physiological relevance and importance of such states in the folding kinetics remain controversial. The intramolecular chaperone (IMC) of subtilisin was used to trap a partially folded, stable crosslinked intermediate conformer (CLIC) through a disulfide bond between mutated IMC and subtilisin. The trapped CLIC contains non-native interactions. Here we show that CLIC can be induced into a catalytically active form by incubating it with small peptide substrates. The structure and catalytic properties of the activated crosslinked intermediate conformer (A-CLIC) differ from those of the fully folded enzyme in that A-CLIC lacks any endopeptidase activity toward a large protein substrate. Our results show that a disulfide-linked partially folded protein can be induced to acquire catalytic activity with a substrate specificity that is different from completely folded subtilisin. These results also suggest that protein folding intermediates may also participate in catalytic reactions.     

A thiol labelling competition experiment as a probe for sidechain packing in the kinetic folding intermediate of N-PGK

Protein folding is directed by the sequence of sidechains along the polypeptide backbone, but despite this the developement of sidechain interactions during folding is not well understood. Here, the thiol-active reagent, dithio-nitrobenzoic acid (DTNB), is used to probe the exposure of the cysteine sidechain thiols in the kinetic folding intermediates of the N-terminal domain of phosphoglycerate kinase (N-PGK) and a number of conservative (I-, L-, or V-to-C) single cysteine variants. Rapid dilution of chemically denatured protein into folding conditions in the presence of DTNB allowed the degree of sidechain protection in any rapidly formed intermediate to be determined through the analysis of the kinetics of labelling. The protection factors derived for the intermediate(s) were generally small (<25), indicating only partial burial of the sidechains. The distribution of protection parallels the previously reported backbone amide protection for the folding intermediate of N-PGK. These observations are consistent with the hypothesis that such intermediates resemble molten globule states; i.e. with native-like backbone hydrogen bonding and overall tertiary structure, but with the sidechains that make up the hydrophobic protein core dynamic and intermittently solvent exposed. The success of the competition technique in characterizing this kinetic intermediate invites application to other model systems.     

Im7 folding mechanism: misfolding on a path to the native state

Many proteins populate collapsed intermediate states during folding. In order to elucidate the nature and importance of these species, we have mapped the structure of the on-pathway intermediate of the four-helix protein, Im7, together with the conformational changes it undergoes as it folds to the native state. Kinetic data for 29 Im7 point mutants show that the intermediate contains three of the four helices found in the native structure, packed around a specific hydrophobic core. However, the intermediate contains many non-native interactions; as a result, hydrophobic interactions become disrupted in the rate-limiting transition state before the final helix docks onto the developing structure. The results of this study support a hierarchical mechanism of protein folding and explain why the misfolding of Im7 occurs. The data also demonstrate that non-native interactions can play a significant role in folding, even for small proteins with simple topologies.     

Protein Vivisection Reveals Elusive Intermediates in Folding

Although most folding intermediates escape detection, their characterization is crucial to the elucidation of folding mechanisms. Here, we outline a powerful strategy to populate partially unfolded intermediates: A buried aliphatic residue is substituted with a charged residue (e.g., Leu → Glu⁻) to destabilize and unfold a specific region of the protein. We applied this strategy to ubiquitin, reversibly trapping a folding intermediate in which the β5-strand is unfolded. The intermediate refolds to a native-like structure upon charge neutralization under mildly acidic conditions. Characterization of the trapped intermediate using NMR and hydrogen exchange methods identifies a second folding intermediate and reveals the order and free energies of the two major folding events on the native side of the rate-limiting step. This general strategy may be combined with other methods and have broad applications in the study of protein folding and other reactions that require trapping of high-energy states.     

Independent nucleation and heterogeneous assembly of structure during folding of equine lysozyme.

The refolding of equine lysozyme from guanidinium chloride has been studied using hydrogen exchange pulse labelling in conjunction with NMR spectroscopy and stopped flow optical methods. The stopped flow optical experiments indicate that extensive hydrophobic collapse occurs rapidly after the initiation of refolding. Pulse labelling experiments monitoring nearly 50 sites within the protein have enabled the subsequent formation of native-like structure to be followed in considerable detail. They reveal that an intermediate having persistent structure within three of the four helices of the alpha-domain of the protein is formed for the whole population of molecules within 4 ms. Subsequent to this event, however, the hydrogen exchange protection kinetics are complex and highly heterogeneous. Analysis of the results by fitting to stretched exponential functions shows that a series of other intermediates is formed as a consequence of the stepwise assembly of independently nucleated local regions of structure. In some molecules the next step in folding involves the stabilisation of the remaining helix in the alpha-domain, whilst in others persistent structure begins to form in the beta-domain. The formation of native-like structure throughout the beta-domain is itself heterogeneous, involving at least three kinetically distinguishable steps. Residues in loop regions throughout the protein attain persistent structure more slowly than regions of secondary structure. There is in addition evidence for locally misfolded regions of structure that reorganise on much longer timescales. The results reveal that the native state of the protein is generated by the heterogeneous assembly of a series of locally cooperative regions of structure. This observation has many features in common with the findings of recent theoretical simulations of protein folding.     

Intimate view of a kinetic protein folding intermediate: residue-resolved structure, interactions, stability, folding and unfolding rates, homogeneity

A cytochrome c kinetic folding intermediate was studied by hydrogen exchange (HX) pulse labeling. Advances in the technique and analysis made it possible to define the structured and unstructured regions, equilibrium stability, and kinetic opening and closing rates, all at an amino acid-resolved level. The entire N-terminal and C-terminal helices are formed and docked together at their normal native positions. They fray in both directions from the interaction region, due to a progression in both unfolding and refolding rates, leading to the surprising suggestion that helix propagation may proceed very slowly in the condensed milieu. Several native-like beta turns are formed. Some residues in the segment that will form the native 60s helix are protected but others are not, suggesting energy minimization to some locally non-native conformation in the transient intermediate. All other regions are unprotected, presumably dynamically disordered. The intermediate resembles a partially constructed native state. It is early, on-pathway, and all of the refolding molecules pass through it. These and related results consistently point to distinct, homogeneous, native-like intermediates in a stepwise sequential pathway, guided by the same factors that determine the native structure. Previous pulse labeling efforts have always assumed EX2 exchange during the labeling pulse, often leading to the suggestion of heterogeneous intermediates in alternative parallel pathways. The present work reveals a dominant role for EX1 exchange in the high pH labeling pulse, which will mimic heterogeneous behavior when EX2 exchange is assumed. The general problem of homogeneous versus heterogeneous intermediates and pathways is discussed.     

Energy barriers, cooperativity, and hidden intermediates in the folding of small proteins

Current theoretical views of the folding process of small proteins (< approximately 100 amino acids) postulate that the landscape of potential mean force (PMF) for the formation of the native state has a funnel shape and that the free energy barrier to folding arises from the chain configurational entropy only. However, recent theoretical studies on the formation of hydrophobic clusters with explicit water suggest that a barrier should exist on the PMF of folding, consistent with the fact that protein folding generally involves a large positive activation enthalpy at room temperature. In addition, high-resolution structural studies of the hidden partially unfolded intermediates have revealed the existence of non-native interactions, suggesting that the correction of the non-native interactions during folding should also lead to barriers on PMF. To explore the effect of a PMF barrier on the folding behavior of proteins, we modified Zwanzig's model for protein folding with an uphill landscape of PMF for the formation of transition states. We found that the modified model for short peptide segments can satisfy the thermodynamic and kinetic criteria for an apparently two-state folding. Since the Levinthal paradox can be solved by a stepwise folding of short peptide segments, a landscape of PMF with a locally uphill search for the transition state and cooperative stabilization of folding intermediates/native state is able to explain the available experimental results for small proteins. We speculate that the existence of cooperative hidden folding intermediates in small proteins could be the consequence of the highly specific structures of the native state, which are selected by evolution to perform specific functions and fold in a biologically meaningful time scale.