Predicting protein folding pathways

A structured folding pathway, which is a time ordered sequence of folding events, plays an important role in the protein folding process and hence, in the conformational search. Pathway prediction, thus gives more insight into the folding process and is a valuable guiding tool to search the conformation space. In this paper, we propose a novel 'unfolding' approach to predict the folding pathway. We apply graph-based methods on a weighted secondary structure graph of a protein to predict the sequence of unfolding events. When viewed in reverse this yields the folding pathway. We demonstrate the success of our approach on several proteins whose pathway is partially known.     

The ion-induced folding of the hammerhead ribozyme: core sequence changes that perturb folding into the active conformation.

The hammerhead ribozyme undergoes an ion-dependent folding process into the active conformation. We find that the folding can be blocked at specific stages by changes of sequence or functionality within the core. In the the absence of added metal ions, the global structure of the hammerhead is extended, with a large angle subtended between stems I and II. No core sequence changes appear to alter this geometry, consistent with an unstructured core under these conditions. Upon addition of low concentrations of magnesium ions, the hammerhead folds by an association of stems II and III, to include a large angle between them. This stage is inhibited or altered by mutations within the oligopurine sequence lying between stems II and III, and folding is completely prevented by an A14G mutation. Further increase in magnesium ion concentration brings about a second stage of folding in the natural sequence hammerhead, involving a reorientation of stem I, which rotates around into the same direction of stem II. Because this transition occurs over the same range of magnesium ion concentration over which the hammerhead ribozyme becomes active, it is likely that the final conformation is most closely related to the active form of the structure. Magnesium ion-dependent folding into this conformation is prevented by changes at G5, notably removal of the 2'-hydroxyl group and replacement of the base by cytidine. The ability to dissect the folding process by means of sequence changes suggests that two separate ion-dependent stages are involved in the folding of the hammerhead ribozyme into the active conformation.     

The Cancer Mutation D83V Induces an α-Helix to β-Strand Conformation Switch in MEF2B

MEF2B is a major target of somatic mutations in non-Hodgkin lymphoma. Most of these mutations are non-synonymous substitutions of surface residues in the MADS-box/MEF2 domain. Among them, D83V is the most frequent mutation found in tumor cells. The link between this hotspot mutation and cancer is not well understood. Here we show that the D83V mutation induces a dramatic α-helix to β-strand switch in the MEF2 domain. Located in an α-helix region rich in β-branched residues, the D83V mutation not only removes the extensive helix stabilization interactions but also introduces an additional β-branched residue that further shifts the conformation equilibrium from α-helix to β-strand. Cross-database analyses of cancer mutations and chameleon sequences revealed a number of well-known cancer targets harboring β-strand favoring mutations in chameleon α-helices, suggesting a commonality of such conformational switch in certain cancers and a new factor to consider when stratifying the rapidly expanding cancer mutation data.     

Loop propensity of the sequence YKGQP from staphylococcal nuclease: implications for the folding of nuclease.

Recently we performed molecular dynamics (MD) simulations on the folding of the hairpin peptide DTVKLMYKGQPMTFR from staphylococcal nuclease in explicit water. We found that the peptide folds into a hairpin conformation with native and nonnative hydrogen-bonding patterns. In all the folding events observed in the folding of the hairpin peptide, loop formation involving the region YKGQP was an important event. In order to trace the origins of the loop propensity of the sequence YKGQP, we performed MD simulations on the sequence starting from extended, polyproline II and native type I' turn conformations for a total simulation length of 300 ns, using the GROMOS96 force field under constant volume and temperature (NVT) conditions. The free-energy landscape of the peptide YKGQP shows minima corresponding to loop conformation with Tyr and Pro side-chain association, turn and extended conformational forms, with modest free-energy barriers separating the minima. To elucidate the role of Gly in facilitating loop formation, we also performed MD simulations of the mutated peptide YKAQP (Gly --> Ala mutation) under similar conditions starting from polyproline II conformation for 100 ns. Two minima corresponding to bend/turn and extended conformations were observed in the free-energy landscape for the peptide YKAQP. The free-energy barrier between the minima in the free-energy landscape of the peptide YKAQP was also modest. Loop conformation is largely sampled by the YKGQP peptide, while extended conformation is largely sampled by the YKAQP peptide. We also explain why the YKGQP sequence samples type II turn conformation in these simulations, whereas the sequence as part of the hairpin peptide DTVKLMYKGQPMTFR samples type I' turn conformation both in the X-ray crystal structure and in our earlier simulations on the folding of the hairpin peptide. We discuss the implications of our results to the folding of the staphylococcal nuclease.     

Structure and folding of four putative kink turns identified in structured RNA species in a test of structural prediction rules

k-Turns are widespread key architectural elements that occur in many classes of RNA molecules. We have shown previously that their folding properties (whether or not they fold into their tightly kinked structure on addition of metal ions) and conformation depend on their local sequence, and we have elucidated a series of rules for prediction of these properties from sequence. In this work, we have expanded the rules for prediction of folding properties, and then applied the full set to predict the folding and conformation of four probable k-turns we have identified amongst 224 structured RNA species found in bacterial intergenenic regions by the Breaker lab (1). We have analyzed the ion-dependence of folding of the four k-turns using fluorescence resonance energy transfer, and determined the conformation of two of them using X-ray crystallography. We find that the experimental data fully conform to both the predicted folding and conformational properties. We conclude that our folding rules are robust, and can be applied to new k-turns of unknown characteristics with confidence.     

Structure of GroEL in Complex with an Early Folding Intermediate of Alanine Glyoxylate Aminotransferase

Primary hyperoxaluria type 1 is a rare autosomal recessive disease caused by mutations in the alanine glyoxylate aminotransferase gene (AGXT). We have previously shown that P11L and I340M polymorphisms together with I244T mutation (AGXT-LTM) represent a conformational disease that could be amenable to pharmacological intervention. Thus, the study of the folding mechanism of AGXT is crucial to understand the molecular basis of the disease. Here, we provide biochemical and structural data showing that AGXT-LTM is able to form non-native folding intermediates. The three-dimensional structure of a complex between the bacterial chaperonin GroEL and a folding intermediate of AGXT-LTM mutant has been solved by cryoelectron microscopy. The electron density map shows the protein substrate in a non-native extended conformation that crosses the GroEL central cavity. Addition of ATP to the complex induces conformational changes on the chaperonin and the internalization of the protein substrate into the folding cavity. The structure provides a three-dimensional picture of an in vivo early ATP-dependent step of the folding reaction cycle of the chaperonin and supports a GroEL functional model in which the chaperonin promotes folding of the AGXT-LTM mutant protein through forced unfolding mechanism.     

Neighborhood properties are important determinants of temperature sensitive mutations

Temperature-sensitive (TS) mutants are powerful tools to study gene function in vivo. These mutants exhibit wild-type activity at permissive temperatures and reduced activity at restrictive temperatures. Although random mutagenesis can be used to generate TS mutants, the procedure is laborious and unfeasible in multicellular organisms. Further, the underlying molecular mechanisms of the TS phenotype are poorly understood. To elucidate TS mechanisms, we used a machine learning method-logistic regression-to investigate a large number of sequence and structure features. We developed and tested 133 features, describing properties of either the mutation site or the mutation site neighborhood. We defined three types of neighborhood using sequence distance, Euclidean distance, and topological distance. We discovered that neighborhood features outperformed mutation site features in predicting TS mutations. The most predictive features suggest that TS mutations tend to occur at buried and rigid residues, and are located at conserved protein domains. The environment of a buried residue often determines the overall structural stability of a protein, thus may lead to reversible activity change upon temperature switch. We developed TS prediction models based on logistic regression and the Lasso regularized procedure. Through a ten-fold cross-validation, we obtained the area under the curve of 0.91 for the model using both sequence and structure features. Testing on independent datasets suggested that the model predicted TS mutations with a 50% precision. In summary, our study elucidated the molecular basis of TS mutants and suggested the importance of neighborhood properties in determining TS mutations. We further developed models to predict TS mutations derived from single amino acid substitutions. In this way, TS mutants can be efficiently obtained through experimentally introducing the predicted mutations.     

Cis and trans requirements for rolling circle replication of a satellite RNA

Satellite RNAs usurp the replication machinery of their helper viruses, even though they bear little or no sequence similarity to the helper virus RNA. In Cereal yellow dwarf polerovirus serotype RPV (CYDV-RPV), the 322-nucleotide satellite RNA (satRPV RNA) accumulates to high levels in the presence of the CYDV-RPV helper virus. Rolling circle replication generates multimeric satRPV RNAs that self-cleave via a double-hammerhead ribozyme structure. Alternative folding inhibits formation of a hammerhead in monomeric satRPV RNA. Here we determine helper virus requirements and the effects of mutations and deletions in satRPV RNA on its replication in oat cells. Using in vivo selection of a satRPV RNA pool randomized at specific bases, we found that disruption of the base pairing necessary to form the non-self-cleaving conformation reduced satRPV RNA accumulation. Unlike other satellite RNAs, both the plus and minus strands proved to be equally infectious. Accordingly, very similar essential replication structures were identified in each strand. A different region is required only for encapsidation. The CYDV-RPV RNA-dependent RNA polymerase (open reading frames 1 and 2), when expressed from the nonhelper Barley yellow dwarf luteovirus, was capable of replicating satRPV RNA. Thus, the helper virus's polymerase is the sole determinant of the ability of a virus to replicate a rolling circle satellite RNA. We present a framework for functional domains in satRPV RNA with three types of function: (i) conformational control elements comprising an RNA switch, (ii) self-functional elements (hammerhead ribozymes), and (iii) cis-acting elements that interact with viral proteins.     

Smooth Functional Transition along a Mutational Pathway with an Abrupt Protein Fold Switch

Recent protein design experiments have demonstrated that proteins can migrate between folds through the accumulation of substitution mutations without visiting disordered or nonfunctional points in sequence space. To explore the biophysical mechanism underlying such transitions we use a three-letter continuous protein model with seven atoms per amino acid to provide realistic sequence-structure and sequence-function mappings through explicit simulation of the folding and interaction of model sequences. We start from two 16-amino-acid sequences folding into an α-helix and a β-hairpin, respectively, each of which has a preferred binding partner with 35 amino acids. We identify a mutational pathway between the two folds, which features a sharp fold switch. By contrast, we find that the transition in function is smooth. Moreover, the switch in preferred binding partner does not coincide with the fold switch. Discovery of new folds in evolution might therefore be facilitated by following fitness slopes in sequence space underpinned by binding-induced conformational switching.     

Molecular mechanisms of "off-on switch" of activities of human IDH1 by tumor-associated mutation R132H

Human cytosolic NADP-IDH (IDH1) has recently been found to be involved in tumorigenesis. Notably, the tumor-derived IDH1 mutations identified so far mainly occur at Arg132, and mutation R132H is the most prevalent one. This mutation impairs the oxidative IDH activity of the enzyme, but renders a new reduction function of converting α-ketoglutarate (αKG) to 2-hydroxyglutarate. Here, we report the structures of the R132H mutant IDH1 with and without isocitrate (ICT) bound. The structural data together with mutagenesis and biochemical data reveal a previously undefined initial ICT-binding state and demonstrate that IDH activity requires a conformational change to a closed pre-transition state. Arg132 plays multiple functional roles in the catalytic reaction; in particular, the R132H mutation hinders the conformational changes from the initial ICT-binding state to the pre-transition state, leading to the impairment of the IDH activity. Our results describe for the first time that there is an intermediate conformation that corresponds to an initial ICT-binding state and that the R132H mutation can trap the enzyme in this conformation, therefore shedding light on the molecular mechanism of the "off switch" of the potentially tumor-suppressive IDH activity. Furthermore, we proved the necessity of Tyr139 for the gained αKG reduction activity and propose that Tyr139 may play a vital role by compensating the increased negative charge on the C2 atom of αKG during the transfer of a hydride anion from NADPH to αKG, which provides new insights into the mechanism of the "on switch" of the hypothetically oncogenic reduction activity of IDH1 by this mutation.     

Kinetic partitioning of protein folding and aggregation.

We have systematically studied the effects of 40 single point mutations on the conversion of the denatured form of the alpha/beta protein acylphosphatase (AcP) into insoluble aggregates. All the mutations that significantly perturb the rate of aggregation are located in two regions of the protein sequence, residues 16-31 and 87-98, each of which has a relatively high hydrophobicity and propensity to form beta-sheet structure. The measured changes in aggregation rate upon mutation correlate with changes in the hydrophobicity and beta-sheet propensity of the regions of the protein in which the mutations are located. The two regions of the protein sequence that determine the aggregation rate are distinct from those parts of the sequence that determine the rate of protein folding. Dissection of the protein into six peptides corresponding to different regions of the sequence indicates that the kinetic partitioning between aggregation and folding can be attributed to the intrinsic conformational preferences of the denatured polypeptide chain.     

Smooth Functional Transition along a Mutational Pathway with an Abrupt Protein Fold Switch

Recent protein design experiments have demonstrated that proteins can migrate between folds through the accumulation of substitution mutations without visiting disordered or nonfunctional points in sequence space. To explore the biophysical mechanism underlying such transitions we use a three-letter continuous protein model with seven atoms per amino acid to provide realistic sequence-structure and sequence-function mappings through explicit simulation of the folding and interaction of model sequences. We start from two 16-amino-acid sequences folding into an α-helix and a β-hairpin, respectively, each of which has a preferred binding partner with 35 amino acids. We identify a mutational pathway between the two folds, which features a sharp fold switch. By contrast, we find that the transition in function is smooth. Moreover, the switch in preferred binding partner does not coincide with the fold switch. Discovery of new folds in evolution might therefore be facilitated by following fitness slopes in sequence space underpinned by binding-induced conformational switching.     

Rational design of a conformation-switchable Ca2+- and Tb3+-binding protein without the use of multiple coupled metal-binding sites

Ca2+, as a messenger of signal transduction, regulates numerous target molecules via Ca2+-induced conformational changes. Investigation into the determinants for Ca2+-induced conformational change is often impeded by cooperativity between multiple metal-binding sites or protein oligomerization in naturally occurring proteins. To dissect the relative contributions of key determinants for Ca2+-dependent conformational changes, we report the design of a single-site Ca2+-binding protein (CD2.trigger) created by altering charged residues at an electrostatically sensitive location on the surface of the host protein rat Cluster of Differentiation 2 (CD2).CD2.trigger binds to Tb3+ and Ca2+ with dissociation constants of 0.3 +/- 0.1 and 90 +/- 25 microM, respectively. This protein is largely unfolded in the absence of metal ions at physiological pH, but Tb3+ or Ca2+ binding results in folding of the native-like conformation. Neutralization of the charged coordination residues, either by mutation or protonation, similarly induces folding of the protein. The control of a major conformational change by a single Ca2+ ion, achieved on a protein designed without reliance on sequence similarity to known Ca2+-dependent proteins and coupled metal-binding sites, represents an important step in the design of trigger proteins.     

Striking HIV-1 Entry by Targeting HIV-1 gp41. But,Where Should We Target?

HIV-1 gp41 facilitates the viral fusion through a conformational switch involving the association of three C-terminal helices along the conserved hydrophobic grooves of three N-terminal helices coiled-coil. The control of these structural rearrangements is thought to be central to HIV-1 entry and, therefore, different strategies of intervention are being developed. Herewith, we describe a procedure to simulate the folding of an HIV-1 gp41 simplified model. This procedure is based on the construction of plausible conformational pathways, which describe protein transition between non-fusogenic and fusogenic conformations. The calculation of the paths started with 100 molecular dynamics simulations of the non-fusogenic conformation, which were found to converge to different intermediate states. Those presenting defined criteria were selected for separate targeted molecular dynamics simulations, subjected to a force constant imposing a movement towards the gp41 fusogenic conformation. Despite significant diversity, a preferred sequence of events emerged when the simulations were analyzed in terms of the formation, breakage and evolution of the contacts. We pointed out 29 residues as the most relevant for the movement of gp41; also, 2696 possible interactions were reduced to only 48 major interactions, which reveals the efficiency of the method. The analysis of the evolution of the main interactions lead to the detection of four main behaviors for those contacts: stable, increasing, decreasing and repulsive interactions. Altogether, these results suggest a specific small cavity of the HIV-1 gp41 hydrophobic groove as the preferred target to small molecules.     

"DFG-flip" in the insulin receptor kinase is facilitated by a helical intermediate state of the activation loop

We have characterized a large-scale inactive-to-active conformational change in the activation-loop of the insulin receptor kinase domain at the atomistic level via untargeted temperature-accelerated molecular dynamics (TAMD) and free-energy calculations using the string method. TAMD simulations consistently show folding of the A-loop into a helical conformation followed by unfolding to an active conformation, causing the highly conserved DFG-motif (Asp(1150), Phe(1151), and Gly(1152)) to switch from the inactive "D-out/F-in" to the nucleotide-binding-competent "D-in/F-out" conformation. The minimum free-energy path computed from the string method preserves these helical intermediates along the inactive-to-active path, and the thermodynamic free-energy differences are consistent with previous work on various other kinases. The mechanisms revealed by TAMD also suggest that the regulatory spine can be dynamically assembled/disassembled either by DFG-flip or by movement of the αC-helix. Together, these findings both broaden our understanding of kinase activation and point to intermediates as specific therapeutic targets.     

Evolutionary Dynamics on Protein Bi-stability Landscapes can Potentially Resolve Adaptive Conflicts

Experimental studies have shown that some proteins exist in two alternative native-state conformations. It has been proposed that such bi-stable proteins can potentially function as evolutionary bridges at the interface between two neutral networks of protein sequences that fold uniquely into the two different native conformations. Under adaptive conflict scenarios, bi-stable proteins may be of particular advantage if they simultaneously provide two beneficial biological functions. However, computational models that simulate protein structure evolution do not yet recognize the importance of bi-stability. Here we use a biophysical model to analyze sequence space to identify bi-stable or multi-stable proteins with two or more equally stable native-state structures. The inclusion of such proteins enhances phenotype connectivity between neutral networks in sequence space. Consideration of the sequence space neighborhood of bridge proteins revealed that bi-stability decreases gradually with each mutation that takes the sequence further away from an exactly bi-stable protein. With relaxed selection pressures, we found that bi-stable proteins in our model are highly successful under simulated adaptive conflict. Inspired by these model predictions, we developed a method to identify real proteins in the PDB with bridge-like properties, and have verified a clear bi-stability gradient for a series of mutants studied by Alexander et al. (Proc Nat Acad Sci USA 2009, 106:21149–21154) that connect two sequences that fold uniquely into two different native structures via a bridge-like intermediate mutant sequence. Based on these findings, new testable predictions for future studies on protein bi-stability and evolution are discussed.