Detection of common three-dimensional substructures in proteins.

We present a fully automatic algorithm for three-dimensional alignment of protein structures and for the detection of common substructures and structural repeats. Given two proteins, the algorithm first identifies all pairs of structurally similar fragments and subsequently clusters into larger units pairs of fragments that are compatible in three dimensions. The detection of similar substructures is independent of insertion/deletion penalties and can be chosen to be independent of the topology of loop connections and to allow for reversal of chain direction. Using distance geometry filters and other approximations, the algorithm, implemented in the WHAT IF program, is so fast that structural comparison of a single protein with the entire database of known protein structures can be performed routinely on a workstation. The method reproduces known non-trivial superpositions such as plastocyanin on azurin. In addition, we report surprising structural similarity between ubiquitin and a (2Fe-2S) ferredoxin.     

What is the minimum number of letters required to fold a protein?

Experimental studies have shown that the full sequence complexity of naturally occurring proteins is not required to generate rapidly folding and functional proteins, i.e. proteins can be designed with fewer than 20 letters. This raises the question of what is the minimum number of amino acid types required to encode complex protein folds? Here, we investigate this issue from three aspects. First, we study the minimum sequence complexity that can reserve the necessary structural information for detection of distantly related homologues. Second, we compare the ability of designing foldable model sequences over a wide range of reduced amino acid alphabets, which find the minimum number of letters that have the similar design ability as 20. Finally, we survey the lower bound of alphabet size of globular proteins in a non-redundant protein database. These different approaches give a remarkably consistent view, that the minimum number of letters required to fold a protein is around ten.     

COMPASS: a tool for comparison of multiple protein alignments with assessment of statistical significance

We present a novel method for the comparison of multiple protein alignments with assessment of statistical significance (COMPASS). The method derives numerical profiles from alignments, constructs optimal local profile-profile alignments and analytically estimates E-values for the detected similarities. The scoring system and E-value calculation are based on a generalization of the PSI-BLAST approach to profile-sequence comparison, which is adapted for the profile-profile case. Tested along with existing methods for profile-sequence (PSI-BLAST) and profile-profile (prof_sim) comparison, COMPASS shows increased abilities for sensitive and selective detection of remote sequence similarities, as well as improved quality of local alignments. The method allows prediction of relationships between protein families in the PFAM database beyond the range of conventional methods. Two predicted relations with high significance are similarities between various Rossmann-type folds and between various helix-turn-helix-containing families. The potential value of COMPASS for structure/function predictions is illustrated by the detection of an intricate homology between the DNA-binding domain of the CTF/NFI family and the MH1 domain of the Smad family.     

Symmetric primary and tertiary structure mutations within a symmetric superfold: a solution, not a constraint, to achieve a foldable polypeptide

In previous studies designed to increase the primary structure symmetry within the hydrophobic core of human acidic fibroblast growth factor (FGF-1) a combination of five mutations were accommodated, resulting in structure, stability and folding kinetic properties similar to wild-type (despite the symmetric constraint upon the set of core residues). A sixth mutation in the core, involving a highly conserved Met residue at position 67, appeared intolerant to substitution. Structural analysis suggested that the local packing environment of position 67 involved two regions of apparent insertions that distorted the tertiary structure symmetry inherent in the beta-trefoil architecture. It was postulated that a symmetric constraint upon the primary structure within the core could only be achieved after these insertions had been deleted (concomitantly increasing the tertiary structure symmetry). The deletion of these insertions is now shown to permit mutation of position 67, thereby increasing the primary structure symmetry relationship within the core. Furthermore, despite the imposed symmetric constraint upon both the primary and tertiary structure, the resulting mutant form of FGF-1 is substantially more stable. The apparent inserted regions are shown to be associated with heparin-binding functionality; however, despite a marked reduction in heparin-binding affinity the mutant form of FGF-1 is surprisingly approximately 70 times more potent in 3T3 fibroblast mitogenic assays. The results support the hypothesis that primary structure symmetry within a symmetric protein superfold represents a possible solution, rather than a constraint, to achieving a foldable polypeptide.     

Sequence variations within protein families are linearly related to structural variations

It is commonly believed that similarities between the sequences of two proteins infer similarities between their structures. Sequence alignments reliably recognize pairs of protein of similar structures provided that the percentage sequence identity between their two sequences is sufficiently high. This distinction, however, is statistically less reliable when the percentage sequence identity is lower than 30% and little is known then about the detailed relationship between the two measures of similarity. Here, we investigate the inverse correlation between structural similarity and sequence similarity on 12 protein structure families. We define the structure similarity between two proteins as the cRMS distance between their structures. The sequence similarity for a pair of proteins is measured as the mean distance between the sequences in the subsets of sequence space compatible with their structures. We obtain an approximation of the sequence space compatible with a protein by designing a collection of protein sequences both stable and specific to the structure of that protein. Using these measures of sequence and structure similarities, we find that structural changes within a protein family are linearly related to changes in sequence similarity.     

基于知识的蛋白质结构预测

介绍了近几年基于知识的蛋白质三维结构预测方法及其进展．目前,基于知识的结构预测方法主要有两类,一类是同源蛋白模建,这种技术比较成熟,模建的结果可靠性比较高,但只适用于同源性比较高的目标序列的模建;另一类方法即蛋白质逆折叠技术,主要包括3D profile方法和基于势函数的方法,给出的是目标蛋白质的空间走向,它主要可用于序列同源性比较低的蛋白质的结构预测．     

Evolutionary Aspects of Protein Structure and Folding

Four basic stages of evolution of protein structure are described, basing on recent work of the authors aimed specifically to reconstruct the earliest events in the protein evolution. According to this reconstruction, the initial stage of short peptides comprising, probably, only a few amino acid residues had been followed by formation of closed loops of 25–30 residues, which corresponds to the polymer-statistically optimal ring closure size for mixed polypeptide chains. The next stage involved fusion of relatively small linear genes and formation of protein structures consisting of several closed loops of a nearly standard size, with 4–6 loops (100–200 amino acid residues) in a typical protein fold. The last, modern stage began with combinatorial fusion of the presumably circular 300–600 bp DNA units and, accordingly, formation of multidomain proteins.     

Transformation of an alpha-helix peptide into a beta-hairpin induced by addition of a fragment results in creation of a coexisting state

Intrinsic rules of determining the tertiary structure of a protein have been unknown partly because physicochemical factors that contribute to stabilization of a protein structure cannot be represented as a linear combination of local interactions. To clarify the rules on the nonlinear term caused by nonlocal interaction in a protein, we tried to transform a peptide that has a fully helical structure (Target Peptide or TP) into a peptide that has a beta-hairpin structure (Designed Peptide or DP) by adding seven residues to the C terminus of TP. According to analyses of nuclear magnetic resonance measurements, while the beta-hairpin structure is stabilized in some DPs, it is evident that the helical structure observed in TP is also persistent and even extended throughout the length of the molecule. As a result, we have produced a peptide molecule that contains both the alpha-helix and beta-hairpin conformation at an almost equally populated level. The helical structures contained in these DPs were more stable than the helix in TP, suggesting that stabilizing one conformation does not result in destabilizing the other conformation. These DPs can thus be regarded as an isolated peptide version of the chameleon sequence, which has the capability of changing the secondary structure depending on the context of the surrounding environment in a protein structure. The fact that the transformation of one secondary structure caused stabilization of both the original and the induced structure would shed light on the mechanism of protein folding.     

Recapitulation of protein family divergence using flexible backbone protein design

We use flexible backbone protein design to explore the sequence and structure neighborhoods of naturally occurring proteins. The method samples sequence and structure space in the vicinity of a known sequence and structure by alternately optimizing the sequence for a fixed protein backbone using rotamer based sequence search, and optimizing the backbone for a fixed amino acid sequence using atomic-resolution structure prediction. We find that such a flexible backbone design method better recapitulates protein family sequence variation than sequence optimization on fixed backbones or randomly perturbed backbone ensembles for ten diverse protein structures. For the SH3 domain, the backbone structure variation in the family is also better recapitulated than in randomly perturbed backbones. The potential application of this method as a model of protein family evolution is highlighted by a concerted transition to the amino acid sequence in the structural core of one SH3 domain starting from the backbone coordinates of an homologous structure.