Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Some insights into the transition state ensemble of the folding of globular proteins

Some insights into the transition state ensemble of the folding of globular proteins Bio-Algorithms and Med-Systems 2014; 10(4): 167­168 Editorial Jacques Chomilier DOI 10.1515/bams-2014-0020 This issue of Bio-Algorithms and Med-Systems (BAMS) contains several review papers about some aspects of the globular protein folding process. Several models have been proposed over the years for the folding, and we do not intend to give a final answer on this pending issue. Nevertheless, predictions of some positions along the sequence, which are important for the formation of the native structure, have reached the point of maturity. These key residues are involved in intra-chain contacts, and even if they are separated in sequence by a so-called long range, they constitute the core of the globule around which the local regular secondary structures take place. The chronology of the formation of these contacts relative to the secondary structures varies according to the available models. To link these structural key residues, whatever they are called, some fragments of the protein are present, which have been initially proposed by Ittah and Haas [1] in a seminal paper in 1995. The "loop hypothesis", reviewed by Orevi et al. [2], states that, during the early steps of the folding process, random dynamics of the residues may lead to close contacts at the extremities of long fragments, producing in some manner a loop structure. In the present issue, experimental evidence is reported by labeling the ends of such closed loops of adenylate kinase that actually form in the early steps of the folding, in the so-called transition state ensemble. In a similar vein, Bonet et al. [3] propose a classification of fragments composed of two adjacent regular secondary structures and the loop connecting these stems. This analysis at the level of subdomain is somehow related to the notion of super secondary structures, and they call them Smotifs. Smotifs can be used as building blocks of globular (or even trans membrane) domains and therefore constitute a path to the prediction of their structure by molecular modeling, provided the number of constraints (in other words of intra-chain contacts) is sufficient. This method avoids the necessity of a template based on sequence similarity with the query, and the quality of the prediction is about the same order as that of a fragmentbased approach such as Rosetta [4]. In the field of folding simulation, great advances have been made by going from a continuous space to a discrete space. Lattice simulations have been developed over the last decades in various aspects, among which the introduction of a degenerate alphabet, with only two types of amino acids, hydrophobic ones (H) and the rest (P) [5]. These "toy models" allow to reproduce the biphasic aspect of globular proteins: a core (mainly H) and a surrounding shell (mainly P). This approach is the basis of the constrained-based protein structure prediction (CPSP) reviewed by Mann and Backofen [6] in this issue. They thoroughly discuss the virtues and defects of contactbased potentials, in particular in their ability to predict regular secondary structures. They also present the limitations of their method for obtaining an optimal structure without a complete enumeration of the conformation space. Folding simulation on an enhanced (2,1,0) lattice, limited to the early steps of the process, i.e., before the formation of any regular secondary structure, is presented in this issue by Acuna et al. [7]. Displacements of the various amino acids under the constraint of a statistical potential in empty nodes of the lattice produce local aggregation around hydrophobic residues involved in numerous contacts, called MIR (most interacting residues). They constitute, in certain aspects, a signature of the fold of globular proteins and can be predicted from the sequence in a reasonable computing time. Moreover, when compared to the evaluation of the free energy modification of these positions relative to a point mutation, they appear to correspond to the maximum of the stability. Key residues for the structure can also be deciphered by means of analysis of multiple alignments of related proteins. Techner et al. [8] review the algorithms developed to identify intra-chain contacts under the aspect of coevolution. Systematic biases are very difficult to avoid, due mainly to the transitive nature of coevolution; besides, present algorithms need numerous 168Chomilier: Transition state ensemble of folding of globular proteins sequences that are not necessarily available for all types of proteins. This paper covers the last advances in the field of inter-residue contact prediction, and it compares various versions of direct coupling analysis (DCA) [9] with precise structural contact prediction using sparse inverse covariance estimation (PSICOV), based on the inverse of the covariance matrix. It is encouraging to observe that a consensus among the methods can emerge for about half the predicted contacts. This paves the way for an improvement of the molecular modeling by implementing topological constraints in the building of the structures, for instance, by means of fragmentbased approaches. Nevertheless, an incomplete set of these contacts can be sufficient in order to predict the fold of a set of proteins. This editorial would not be complete without mentioning a previous paper, recently published in BAMS by Cruzeiro [10]. It shows molecular dynamics, starting either from a fully extended conformation or from a single helix, directed under the constraint of the experimental structure of two proteins, one from the class and the second one from the mixed class. Although the simulations are performed on a small time scale, 0.5 ns, the results show that the first steps the proteins go through depend on their native fold. These results, in partial contradiction with other simulations, plead for more systematic studies in order to better characterize the transition state ensemble, by combined methods, both on an experimental and on a predictive point of view. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Bio-Algorithms and Med-Systems de Gruyter

Some insights into the transition state ensemble of the folding of globular proteins

Bio-Algorithms and Med-Systems , Volume 10 (4) – Dec 19, 2014

Loading next page...
 
/lp/de-gruyter/some-insights-into-the-transition-state-ensemble-of-the-folding-of-VBhFUtGvw0
Publisher
de Gruyter
Copyright
Copyright © 2014 by the
ISSN
1895-9091
eISSN
1896-530X
DOI
10.1515/bams-2014-0020
Publisher site
See Article on Publisher Site

Abstract

Bio-Algorithms and Med-Systems 2014; 10(4): 167­168 Editorial Jacques Chomilier DOI 10.1515/bams-2014-0020 This issue of Bio-Algorithms and Med-Systems (BAMS) contains several review papers about some aspects of the globular protein folding process. Several models have been proposed over the years for the folding, and we do not intend to give a final answer on this pending issue. Nevertheless, predictions of some positions along the sequence, which are important for the formation of the native structure, have reached the point of maturity. These key residues are involved in intra-chain contacts, and even if they are separated in sequence by a so-called long range, they constitute the core of the globule around which the local regular secondary structures take place. The chronology of the formation of these contacts relative to the secondary structures varies according to the available models. To link these structural key residues, whatever they are called, some fragments of the protein are present, which have been initially proposed by Ittah and Haas [1] in a seminal paper in 1995. The "loop hypothesis", reviewed by Orevi et al. [2], states that, during the early steps of the folding process, random dynamics of the residues may lead to close contacts at the extremities of long fragments, producing in some manner a loop structure. In the present issue, experimental evidence is reported by labeling the ends of such closed loops of adenylate kinase that actually form in the early steps of the folding, in the so-called transition state ensemble. In a similar vein, Bonet et al. [3] propose a classification of fragments composed of two adjacent regular secondary structures and the loop connecting these stems. This analysis at the level of subdomain is somehow related to the notion of super secondary structures, and they call them Smotifs. Smotifs can be used as building blocks of globular (or even trans membrane) domains and therefore constitute a path to the prediction of their structure by molecular modeling, provided the number of constraints (in other words of intra-chain contacts) is sufficient. This method avoids the necessity of a template based on sequence similarity with the query, and the quality of the prediction is about the same order as that of a fragmentbased approach such as Rosetta [4]. In the field of folding simulation, great advances have been made by going from a continuous space to a discrete space. Lattice simulations have been developed over the last decades in various aspects, among which the introduction of a degenerate alphabet, with only two types of amino acids, hydrophobic ones (H) and the rest (P) [5]. These "toy models" allow to reproduce the biphasic aspect of globular proteins: a core (mainly H) and a surrounding shell (mainly P). This approach is the basis of the constrained-based protein structure prediction (CPSP) reviewed by Mann and Backofen [6] in this issue. They thoroughly discuss the virtues and defects of contactbased potentials, in particular in their ability to predict regular secondary structures. They also present the limitations of their method for obtaining an optimal structure without a complete enumeration of the conformation space. Folding simulation on an enhanced (2,1,0) lattice, limited to the early steps of the process, i.e., before the formation of any regular secondary structure, is presented in this issue by Acuna et al. [7]. Displacements of the various amino acids under the constraint of a statistical potential in empty nodes of the lattice produce local aggregation around hydrophobic residues involved in numerous contacts, called MIR (most interacting residues). They constitute, in certain aspects, a signature of the fold of globular proteins and can be predicted from the sequence in a reasonable computing time. Moreover, when compared to the evaluation of the free energy modification of these positions relative to a point mutation, they appear to correspond to the maximum of the stability. Key residues for the structure can also be deciphered by means of analysis of multiple alignments of related proteins. Techner et al. [8] review the algorithms developed to identify intra-chain contacts under the aspect of coevolution. Systematic biases are very difficult to avoid, due mainly to the transitive nature of coevolution; besides, present algorithms need numerous 168Chomilier: Transition state ensemble of folding of globular proteins sequences that are not necessarily available for all types of proteins. This paper covers the last advances in the field of inter-residue contact prediction, and it compares various versions of direct coupling analysis (DCA) [9] with precise structural contact prediction using sparse inverse covariance estimation (PSICOV), based on the inverse of the covariance matrix. It is encouraging to observe that a consensus among the methods can emerge for about half the predicted contacts. This paves the way for an improvement of the molecular modeling by implementing topological constraints in the building of the structures, for instance, by means of fragmentbased approaches. Nevertheless, an incomplete set of these contacts can be sufficient in order to predict the fold of a set of proteins. This editorial would not be complete without mentioning a previous paper, recently published in BAMS by Cruzeiro [10]. It shows molecular dynamics, starting either from a fully extended conformation or from a single helix, directed under the constraint of the experimental structure of two proteins, one from the class and the second one from the mixed class. Although the simulations are performed on a small time scale, 0.5 ns, the results show that the first steps the proteins go through depend on their native fold. These results, in partial contradiction with other simulations, plead for more systematic studies in order to better characterize the transition state ensemble, by combined methods, both on an experimental and on a predictive point of view.

Journal

Bio-Algorithms and Med-Systemsde Gruyter

Published: Dec 19, 2014

References