A kinetic method to evaluate the two-state character of solvent-induced protein denaturation

Combining Ancestral Reconstruction with Folding-Landscape Simulations to Engineer Heterologous Protein Expression

Obligate symbionts typically exhibit high evolutionary rates. Consequently, their proteins may differ considerably from their modern and ancestral homologs in terms of both sequence and properties, thus providing excellent models to study protein evolution. Also, obligate symbionts are challenging to culture in the lab and proteins from uncultured organisms must be produced in heterologous hosts using recombinant DNA technology. Obligate symbionts thus replicate a fundamental scenario of metagenomics studies aimed at the functional characterization and biotechnological exploitation of proteins from the bacteria in soil. Here, we use the thioredoxin from Candidatus Photodesmus katoptron, an uncultured symbiont of flashlight fish, to explore evolutionary and engineering aspects of protein folding in heterologous hosts. The symbiont protein is a standard thioredoxin in terms of 3D-structure, stability and redox activity. However, its folding outside the original host is severely impaired, as shown by a very slow refolding in vitro and an inefficient expression in E. coli that leads mostly to insoluble protein. By contrast, resurrected Precambrian thioredoxins express efficiently in E. coli, plausibly reflecting an ancient adaptation to unassisted folding. We have used a statistical-mechanical model of the folding landscape to guide back-to-ancestor engineering of the symbiont protein. Remarkably, we find that the efficiency of heterologous expression correlates with the in vitro (i.e., unassisted) folding rate and that the ancestral expression efficiency can be achieved with only 1-2 back-to-ancestor replacements. These results demonstrate a minimal-perturbation, sequence-engineering approach to rescue inefficient heterologous expression which may potentially be useful in metagenomics efforts targeting recent adaptations.

The Folding Pathway of 6aJL2

One-third of the reported cases of light chain amyloidosis are related to the germ line λ6 family; remarkably, healthy individuals express this type of protein in just 2% of the peripheral blood and bone marrow B-cells. The appearance of the disease has been related to the inherent properties of this protein family. A recombinant representative model for λ6 proteins called 6aJL2 containing the amino acid sequence encoded by the 6a and JL2 germ line genes was previously designed and synthesized to study the properties of this family. Previous work on 6aJL2 suggested a simple two-state folding model at 25 °C; no intermediate could be identified either by kinetics or by fluorescence and circular dichroism equilibrium studies, although the presence of an intermediate that is populated at ∼2.4 M urea was suggested by size exclusion chromatography. In this study we employed classic equilibrium and kinetic experiments and analysis to elucidate the detailed folding mechanism of this protein. We identify species that are kinetically accessible and/or are populated at equilibrium. We describe the presence of intermediate and native-like species and propose a five-species folding mechanism at 25 °C at short incubation times, similar to and consistent with those observed in other proteins of this fold. The formation of intermediates in the mechanism of 6aJL2 is faster than that proposed for a V_κ light chain, which could be an important distinction in the amyloidogenic potential of both germ lines.

Non-conservation of folding rates in the thioredoxin family reveals degradation of ancestral unassisted-folding

Evolution involves not only adaptation, but also the degradation of superfluous features. Many examples of degradation at the morphological level are known (vestigial organs, for instance). However, the impact of degradation on molecular evolution has been rarely addressed. Thioredoxins serve as general oxidoreductases in all cells. Here, we report extensive mutational analyses on the folding of modern and resurrected ancestral bacterial thioredoxins. Contrary to claims from recent literature, in vitro folding rates in the thioredoxin family are not evolutionarily conserved, but span at least a ∼100-fold range. Furthermore, modern thioredoxin folding is often substantially slower than ancestral thioredoxin folding. Unassisted folding, as probed in vitro, thus emerges as an ancestral vestigial feature that underwent degradation, plausibly upon the evolutionary emergence of efficient cellular folding assistance. More generally, our results provide evidence that degradation of ancestral features shapes, not only morphological evolution, but also the evolution of individual proteins.

pH-dependent relationship between thermodynamic and kinetic stability in the denaturation of human phosphoglycerate kinase 1

Human phosphoglycerate kinase 1 (hPGK1) is a glycolytic enzyme essential for ATP synthesis, and it is implicated in different pathological conditions such as inherited diseases, oncogenesis and activation of drugs for cancer and viral treatments. Particularly, mutations in hPGK1 cause human PGK1 deficiency, a rate metabolic conformational disease. We have recently found that most of these mutations cause protein kinetic destabilization by significant changes in the structure/energetics of the transition state for irreversible denaturation. In this work, we explore the relationships between protein conformation, thermodynamic and kinetic stability in hPGK1 by performing comprehensive analyses in a wide pH range (2.5-8). hPGK1 remains in a native conformation at pH 5-8, but undergoes a conformational transition to a molten globule-like state at acidic pH. Interestingly, hPGK1 kinetic stability remains essentially constant at pH 6-8, but is significantly reduced when pH is decreased from 6 to 5. We found that this decrease in kinetic stability is caused by significant changes in the energetic/structural balance of the denaturation transition state, which diverge from those found for disease-causing mutations. We also show that protein kinetic destabilization by acidic pH is strongly linked to lower thermodynamic stability, while in disease-causing mutations seems to be linked to lower unfolding cooperativity. These results highlight the plasticity of the hPGK1 denaturation mechanism that responds differently to changes in pH and in disease-causing mutations. New insight is presented into the different factors contributing to hPGK1 thermodynamic and kinetic stability and the role of denaturation mechanisms in hPGK1 deficiency.

pH-Dependent urea-induced unfolding of stem bromelain: unusual stability against urea at neutral pH

Equilibrium unfolding of stem bromelain (SB) with urea as a denaturant has been monitored as a function of pH using circular dichroism and fluorescence emission spectroscopy. Urea-induced denaturation studies at pH 4.5 showed that SB unfolds through a two-state mechanism and yields DeltaG (free energy difference between the fully folded and unfolded forms) of approximately 5.0 kcal/mol and C(m) (midpoint of the unfolding transition) of approximately 6.5 M at 25 degrees C. Very high concentration of urea (9.5 M) provides unusual stability to the protein with no more structural loss and transition to a completely unfolded state.

Natural selection for kinetic stability is a likely origin of correlations between mutational effects on protein energetics and frequencies of amino acid occurrences in sequence alignments

It appears plausible that natural selection constrains, to some extent at least, the stability in many natural proteins. If, during protein evolution, stability fluctuates within a comparatively narrow range, then mutations are expected to be fixed with frequencies that reflect mutational effects on stability. Indeed, we recently reported a robust correlation between the effect of 27 conservative mutations on the thermodynamic stability (unfolding free energy) of Escherichia coli thioredoxin and the frequencies of residues occurrences in sequence alignments. We show here that this correlation likely implies a lower limit to thermodynamic stability of only a few kJ/mol below the unfolding free energy of the wild-type (WT) protein. We suggest, therefore, that the correlation does not reflect natural selection of thermodynamic stability by itself, but of some other factor which is linked to thermodynamic stability for the mutations under study. We propose that this other factor is the kinetic stability of thioredoxin in vivo, since( i) kinetic stability relates to irreversible denaturation, (ii) the rate of irreversible denaturation in a crowded cellular environment (or in a harsh extracellular environment) is probably determined by the rate of unfolding, and (iii) the half-life for unfolding changes in an exponential manner with activation free energy and, consequently, comparatively small free energy effects can have deleterious consequences for kinetic stability. This proposal is supported by the results of a kinetic study of the WT form and the 27 single-mutant variants of E. coli thioredoxin based on the global analyses of chevron plots and equilibrium unfolding profiles determined from double-jump unfolding assays. This kinetic study suggests, furthermore, one of the factors that may contribute to the high activation free energy for unfolding in thioredoxin (required for kinetic stability), namely the energetic optimization of native-state residue environments in regions, which become disrupted in the transition state for unfolding.

Guanidine hydrochloride denaturation of human serum albumin originates by local unfolding of some stable loops in domain III

The effect of guanidine hydrochloride (GnHCl) on the global stability of human serum albumin (HSA) has been studied by fluorescence and circular dichroism spectroscopic measurements. The differential stability of native conformation of three HSA domains were explored by using domain-specific ligands, hemin (domain I), chloroform (domain II), bilirubin (at domain I/domain II interface) and diazepam (domain III). GnHCl induced unfolding transition curves as monitored by probes for secondary and tertiary structures were cooperative but noncoincidental. A strong ANS binding to the protein was observed around 1.8 M GnHCl, suggesting existence of intermediate states in the unfolding pathway of HSA. A gradual decrease (in the GnHCl concentration range 0.0-1.8 M) in the binding of diazepam indicates that domain III is the most labile to GnHCl denaturation. A significant increase in the binding of bilirubin up to 1.4 M GnHCl and decrease thereafter leading to complete abolishment of bilirubin binding at around 2.0 M GnHCl suggest favorable rearrangement and separation of domains I and II at 1.4 and 2.0 M GnHCl concentration, respectively. Above 1.6 M GnHCl, decrease of the binding of hemin, a ligand for domain I, chloroform, which binds in domain II and lone tryptophanyl fluorescence (Trp-214 located in domain II) indicate that at higher concentration of GnHCl domains I and II start unfolding simultaneously but the stability of domain I (7.4 Kcal/mol) is much more than domain II (4.3 Kcal/mol). A pictorial model for the unfolding of HSA domains, consistent with all these results, has been formulated, suggesting that domain III is the most labile followed by domain II while domain I is the most stable. A molten globule like state of domain III around 1.8 M GnHCl has also been identified and characterized.

Lower kinetic limit to protein thermal stability: a proposal regarding protein stability in vivo and its relation with misfolding diseases

In vitro thermal denaturation experiments suggest that, because of the possibility of irreversible alterations, thermodynamic stability (i.e., a positive value for the unfolding Gibbs energy) does not guarantee that a protein will remain in the native state during a given timescale. Furthermore, irreversible alterations are more likely to occur in vivo than in vitro because (a) some irreversible processes (e.g., aggregation, "undesirable" interactions with other macromolecular components, and proteolysis) are expected to be fast in the "crowded" cellular environment and (b) in many cases, the relevant timescale in vivo (probably related to the half-life for protein degradation) is expected to be longer than the timescale of the usual in vitro experiments (of the order of minutes). We propose, therefore, that many proteins (in particular, thermophilic proteins and "complex" proteins systems) are designed (by evolution) to have significant kinetic stability when confronted with the destabilizing effect of irreversible alterations. We show that, as long as these alterations occur mainly from non-native states (a Lumry-Eyring scenario), the required kinetic stability may be achieved through the design of a sufficiently high activation barrier for unfolding, which we define as the Gibbs energy barrier that separates the native state from the non-native ensemble (unfolded, partially folded, and misfolded states) in the following generalized Lumry-Eyring model: Native State <--> Non-Native Ensemble --> Irreversibly Denatured Protein. Finally, using familial amyloid polyneuropathy (FAP) as an illustrative example, we discuss the relation between stability and amyloid fibril formation in terms of the above viewpoint, which leads us to the two following tentative suggestions: (a) the hot spot defined by the FAP-associated amyloidogenic mutations of transthyretin reflects the structure of the transition state for unfolding and (b) substances that decrease the in vitro rate of transthyretin unfolding could also be inhibitors of amyloid fibril formation.

Separation and detection techniques for peptides and proteins in stability research and bioanalysis

In this paper, a brief overview of the most commonly used methods for the separation and analysis of peptides and proteins in stability and bioanalysis studies is presented. To investigate the physical stability of peptides and proteins, size-exclusion chromatography and electrophoretic separation techniques are being used, apart from several other methods. To determine the chemical stability of these compounds, separation systems are also important, with informative detection modes, such as various spectroscopic detections, electrochemical detection and mass spectrometric detection. For the bioanalysis of peptides, separation is the most important factor, while the detection must be done at the highest possible level of sensitivity.

Submillisecond unfolding kinetics of apomyoglobin and its pH 4 intermediate

Submillisecond mixing experiments and tryptophan fluorescence spectroscopy are used to address two questions raised in earlier stopped-flow studies of the folding and unfolding kinetics of sperm whale apomyoglobin. A study of the pH 4 folding intermediate (I) revealed, surprisingly, that its folding and unfolding kinetics are measurable and fit the two-state model except for a possible burst phase in unfolding. Submillisecond mixing experiments confirm the unfolding burst phase and show that its properties are consistent with the recently discovered interconversion between two forms of I, Ia equilibrium Ib. In urea-induced unfolding, Ib is converted to Ia before Ia unfolds, and the unfolding kinetics of Ia fit the two-state model when the burst phase is assigned to Ib-->Ia. The second question is whether the Ia, Ib intermediates accumulate transiently when the native protein (N) unfolds to the acid unfolded form (U). Earlier work showed that Ia and Ib accumulate when U refolds to N at pH 6.0 and the results fit the linear folding pathway U equilibrium Ia equilibrium Ib equilibrium N. We report here that either or both Ia and Ib accumulate transiently when N unfolds to U at pH 2.7 and that the position of the rate-limiting step in the pathway changes between unfolding at pH 2. 7 and refolding at pH 6.0. In unfolding as in refolding, we do not detect a fast track that bypasses the Ia, Ib intermediates.

The paradox between m values and deltaCp's for denaturation of ribonuclease T1 with disulfide bonds intact and broken

Urea-induced denaturations of RNase T1 and reduced and carboxyamidated RNase T1 (RTCAM) as a function of temperature were analyzed using the linear extrapolation method, and denaturation m values, deltaCp, deltaH, deltaS, and deltaG quantities were determined. Because both deltaCp and m values are believed to reflect the protein surface area newly exposed on denaturation, the prediction is that the ratio of m values for RNase T1 and RTCAM should equal the deltaCp ratio for the two proteins. This is not the case, for it is found that the m value of RTCAM is 1.5 times that of RNase T1, while the denaturation deltaCp's for the two proteins are identical. The paradox of why the two parameters, m and deltaCp, are not equivalent in their behavior is of importance in the interpretations of their respective molecular-level meanings. It is found that the measured denaturation deltaCp's are consistent with deltaCp's calculated on the basis of empirical relationships between the change in surface area on denaturation (deltaASA), and that the measured m value of RNase T1 agrees with m calculated from empirical data relating m to deltaASA. However, the measured m of RTCAM is so much out of line with its calculated m as to call into question the validity of always equating m with surface area newly exposed on denaturation.

Analytical techniques used to study the degradation of proteins and peptides: physical instability

The physical instability of proteins and peptides as well as the various analytical techniques used to study the various aspects of physical instability have been reviewed. Physical instability of proteins and peptides involve changes in secondary, tertiary and quaternary structures of these compounds. After a general introduction of the subject the literature data of these changes and their analytical aspects have been summarized in a Table.

Guanidine hydrochloride unfolding of a transmembrane beta-strand in FepA using site-directed spin labeling

We have used the electron spin resonance (ESR) site-directed spin-labeling (SDSL) technique to examine the guanidine hydrochloride (Gdn-HCl) induced denaturation of several sites along a transmembrane beta-strand located in the ferric enterobactin receptor, FepA. In addition, we have continued the characterization of the beta-strand previously identified by our group (Klug CS et al., 1997, Biochemistry 36:13027-13033) to extend from the periplasm to the extracellular surface loop in FepA, an integral membrane protein containing a beta-barrel motif comprised of a series of antiparallel beta-strands that is responsible for transport of the iron chelate, ferric enterobactin (FeEnt), across the outer membrane of Escherichia coli and many related enteric bacteria. We have previously shown that a large surface loop in FepA containing the FeEnt binding site denatures independently of the beta-barrel domain (Klug CS et al., 1995, Biochemistry 34:14230-14236). The SDSL approach allows examination of the unfolding at individual residues independent of the global unfolding of the protein. This work shows that sites along the beta-strand that are exposed to the aqueous lumen of the channel denature more rapidly and with higher cooperativity than the surface loop, while sites on the hydrophobic side of the beta-strand undergo a limited degree of noncooperative unfolding and do not fully denature even at high (e.g., 4 M) Gdn-HCl concentrations. We conclude that, in a transmembrane beta-strand, the local environment of a given residue plays a significant role in the loss of structure at each site.

Enhancement of transcriptional activity of T7 RNA polymerase by guanidine hydrochloride

T7 RNA polymerase shows an increase in processive transcription in the presence of low concentrations of guanidine hydrochloride (GdnCl) upto 60 mM, which is not observed when the enzyme is treated with urea. Higher concentrations of the denaturant lead to a progressive loss in the processive transcriptional activity of the enzyme. We have attempted to explain the above phenomenon in terms of the structural change in the enzyme. Fluorescence and CD studies suggest that the tertiary structure of the native enzyme undergoes an alteration upon addition of low concentration of guanidine hydrochloride. This is also indicated from the decreased susceptibility of the enzyme to limited proteolysis by trypsin.

Proteins unfold in steps

I present results from an experiment on the dynamics of folding of a globular protein (bovine serum albumin). Employing a micro-mechanical technique, I perform the measurements on very few molecules (1-100). I observed a sequence of steps in time for both unfolding and refolding. The overall characteristic time of the process is thus built up of waiting times between successive steps. The pattern of steps is reproducible, demonstrating the existence of deterministic pathways for folding and unfolding. Certain symmetries in the patterns of steps may reflect the architecture of the protein's structure.

Conformation of thermally denatured RNase T1 with intact disulfide bonds: a study by small-angle X-ray scattering

Small-angle X-ray scattering of RNase T1 with intact disulfide bonds was measured at 20 degrees and 60 degrees C in order to get insight into the structural changes of the protein caused by thermal denaturation. The radius of gyration increases from R(G)= 1.43 nm to R(G) = 2.21 nm. The conformations of the molecules at 60 degrees C are similar to those of ring-shaped random walk chains. However, the molecules are more compact than one would expect under theta conditions due to attractive interactions between the chain segments. The volume needed for free rotation of the thermally unfolded protein molecules about any axis in solution is five times greater than in the native state whereas the hydrodynamic effective volume is increasing only two times.

Remarkably slow folding of a small protein

Equilibrium denaturation of the 72 amino acid alpha/beta-protein MerP, by acid, guanidine hydrochloride, or temperature, is fully reversible and follows a two-state model in which only the native and unfolded states are populated. A cis-trans equilibrium around a proline peptide bond causes a heterogeneity of the unfolded state and gives rise to a slow- and a fast folding population. With a rate constant of 1.2 s(-1) for the major fast folding population, which has none of the common intrinsically slow steps, MerP is the slowest folding protein of this small size yet reported.

Effects of Mg2+ and denaturants on the unfolding pattern of DNA-T--a replication protein of E. coli

Escherichia coli-DNA-T protein is a key component of a multiprotein complex called the primosome which is involved in the initiation of DNA replication. The thermal and urea induced unfolding transition of this protein in the presence and absence of Mg2+ was studied using circular dichroism (CD) and fluorescence spectroscopy as probes. Quenching of the intrinsic fluorescence of DNA-T was observed in the thermal unfolding while formation of a hyperfluorescent form of the protein was found in the urea induced unfolding process. The CD studies showed a monophasic transition curve for thermal unfolding in the presence and absence of Mg2+. Biphasic curves indicative of the formation of intermediates was observed in the urea induced unfolding. The results suggest that the pathways of unfolding of thermal- and urea-induced transitions are different. MgCl2, which affects the conformation of the protein and stabilises the secondary structure, also affects the unfolding pattern.

On-pathway versus off-pathway folding intermediates

Rapidly formed molten globule intermediates accumulate at the start of the folding reactions of several small proteins. Opinion is sharply divided as to whether they are on-pathway or off-pathway intermediates. I discuss recent experiments aimed at resolving this issue. Specific points include whether a 'rollover' in the plot of folding rate versus denaturant concentration implies that a folding intermediate is or is not on-pathway; whether the failure to observe folding intermediates for some small proteins implies a different folding mechanism or only that the intermediates are less stable; possible interpretation of 'fast-track' folding of hen lysozyme; and the significance of recent results in the search for unfolding intermediates.

Stopped-flow NMR spectroscopy: real-time unfolding studies of 6-19F-tryptophan-labeled Escherichia coli dihydrofolate reductase

Escherichia coli dihydrofolate reductase (DHFR; EC 1.5.1.3) contains five tryptophan residues that have been replaced with 6-19F-tryptophan. The 19F NMR assignments are known in the native, unliganded form and the unfolded form. We have used these assignments with stopped-flow 19F NMR spectroscopy to investigate the behavior of specific regions of the protein in real time during urea-induced unfolding. The NMR data show that within 1.5 sec most of the intensities of the native 19F resonances of the protein are lost but only a fraction (approximately 20%) of the intensities of the unfolded resonances appears. We postulate that the early disappearance of the native resonances indicates that most of the protein rapidly forms an intermediate in which the side chains have considerable mobility. Stopped-flow far-UV circular dichroism measurements indicate that this intermediate retains native-like secondary structure. Eighty percent of the intensities of the NMR resonances assigned to the individual tryptophans in the unfolded state appear with similar rate constants (k approximately 0.14 sec-1), consistent with the major phase of unfolding observed by stopped-flow circular dichroism (representing 80% of total amplitude). These data imply that after formation of the intermediate, which appears to represent an expanded structural form, all regions of the protein unfold at the same rate. Stopped-flow measurements of the fluorescence and circular dichroism changes associated with the urea-induced unfolding show a fast phase (half-time of about 1 sec) representing 20% of the total amplitude in addition to the slow phase mentioned above. The NMR data show that approximately 20% of the total intensity for each of the unfolded tryptophan resonances is present at 1.5 sec, indicating that these two phases may represent the complete unfolding of the two different populations of the native protein.

Extremely rapid protein folding in the absence of intermediates

Here we used the cold-shock protein CspB from Bacillus subtilis to study protein folding at an elementary level. The thermodynamic stability of this small five-stranded beta-barrel protein is low, but unfolding and refolding are extremely rapid reactions. In 0.6 M urea the time constant of refolding is about 1.5 ms, and at the transition midpoint (4 M urea) the folded and unfolded forms equilibrate in less than 100 ms. Both the equilibrium unfolding transition and the folding kinetics are perfectly described by a N<-->U two-state model. The validity of this model was confirmed by several kinetic tests. Folding intermediates could neither be detected at equilibrium nor in the folding kinetics. We suggest that the extremely rapid folding of CspB and the absence of folding intermediates are related phenomena.

Direct NMR evidence for an intermediate preceding the rate-limiting step in the unfolding of ribonuclease A

It is commonly believed that there are no detectable intermediates in the kinetic unfolding reactions of small proteins. If such intermediates could be found, they would give important information about the nature of the transition state for unfolding, which is thought to occur close to the native state. We report here that one-dimensional proton magnetic resonance spectra recorded during the unfolding of ribonuclease A provide direct evidence for at least one unfolding intermediate in which side chains are free to rotate. This intermediate appears to be a 'dry molten globule' of the kind hypothesized by Shakhnovich and Finkelstein.

Folding mechanism of ribonuclease T1 in the absence of the disulfide bonds

In the absence of its two disulfide bonds, ribonuclease T1 can exist in a native-like folded conformation when > or = 2 M NaCl is present. We measured the kinetics of unfolding and refolding of two reduced and carboxymethylated variants of ribonuclease T1 with one cis proline (the Ser54Gly/Pro55Asn variant) and with two cis prolines (the wild-type protein) as a function of the NaCl concentration. Single and double mixing techniques were used. Analysis of the kinetic results demonstrates that the two cis prolyl bonds at Pro39 and Pro55 remain cis in the folded state after the reduction and carboxymethylation of the disulfide bonds. Folded molecules with trans isomers could not be found. The substitution of cis-Pro55 influences the proline-limited folding reaction, and the analysis of the changes in the folding kinetics shows that the trans-->cis isomerizations of both prolines are slow and are rate-determining steps for the refolding of ribonuclease T1 in the presence as well as in the absence of the disulfide bonds. The direct folding reaction of protein chains with correct prolyl isomers is also affected by the Ser54Gly/Pro55Asn mutation. The rate of refolding is decreased, whereas the rate of unfolding is almost unaffected. The kinetic analysis points to two main consequences of the Ser54Gly/Pro55Asn mutation for the stability and the folding mechanism of RNase T1. It is moderately destabilizing, because the deletion of a conformationally restricted residue (Pro55-->Asn) and the insertion of a flexible residue (Ser54-->Gly) both tend to increase the entropy of the unfolded state. The cis<-->trans isomerization of Pro55 is abolished, however, leading to a decrease in the entropy of the unfolded protein. These two entropic contributions seem to partially compensate each other, and the net change in free energy as a consequence of the Ser54Gly/Pro55Asn double mutation is very small.