Different protein sequences can give rise to highly similar folds through different stabilizing interactions

Abstract

We report an interesting case of structural similarity between 2 small, nonhomologous proteins, the third domain of ovomucoid (ovomucoid) and the C-terminal fragment of ribosomal L7/L12 protein (CTF). The region of similarity consists of a 3-stranded 0-sheet and an a-helix. This region is highly similar; the corresponding elements of secondary structure share a common topology, and the RMS difference for "equivalent" C a atoms is 1.6 A.Surprisingly, this common structure arises from completely different sequences. For the common core, the sequence identity is less than 3%, and there is neither significant sequence similarity nor similarity in the position or orientation of conserved hydrophobic residues. This superposition raises the question of how 2 entirely different sequences can produce an identical structure. Analyzing this common region in ovomucoid revealed that it is stabilized by disulfide bonds. In contrast, the corresponding structure in CTF is stabilized in the a-helix by a composition of residues with high helix-forming propensities. This result suggests that different sequences and different stabilizing interactions can produce an identical structure.Keywords: CTF; evolution; ovomucoid; protein design; protein folding; sequence; structure; structure superpositionThe tremendous growth of the number of solved protein structures has created a need for new automatic methods for structural comparison (Bowie et al., 1991). Such methods are useful because they allow the detection of homology in the many cases where distantly related proteins have lost sequence similarity but still retain a common structure (Rossmann et al., 1974;Matthews et al., 1981). Protein structure comparison can also offer insights into the interrelationships between sequence, structure, and function. To fill this need, we have developed a rapid, quantitative method for comparing pairs of protein structures. In the course of examining the results of all possible pairwise comparisons of protein structures in the Protein Data Bank (PDB), we were first startled to find structural similarity between proteins of the repressor and globin families (Subbiah et al., 1993). Here, we report a second case of notable structural similarity, between the third domain of ovomucoid (ovomucoid) and the C-terminal fragment of ribosomal L7/L12 protein (CTF).Ovomucoid and CTF are small proteins belonging to the a+P structure class. The ovomucoid domain consists of about 56 residues (Laskowski et al., 1987). The fold, as revealed by a 1.9-A crystal structure (PDB entry 20VO) begins with a 20-residue

Are knowledge-based potentials derived from protein structure sets discriminative with respect to amino acid types?

“…It is widely accepted that the only problem accounting for the lack of prediction success consists in the insufficient conservation of the environment variable value at equivalent spatial position of protein family members. [47][48][49][50][51][52] It was never explored whether the potential is able to place correctly, for example, a histidine if accessibility, secondary structure, or a set of residue-residue contacts is conserved. We have seen that neither potential is sensitive for this amino acid type as well as for a few others.…”

Recognition of analogous and homologous protein folds: analysis of sequence and structure conservation 1 1Edited by F. E. Cohen

“…see Lesk & Chothia, 1980Martin, 1995;Ollis et al, 1992) or analogous folds (e.g. see Murzin et al, 1992;Hazes & Hol, 1992;Pickett et al, 1992;Laurents et al, 1994). More general analysis identi®ed principles of structural variation in relation to sequence similarity (Chothia & Lesk, 1986;Hubbard & Blundell, 1987).…”

Complementarity of network and sequence information in homologous proteins

“…Moreover, while the vast majority of protein pairs with sequence identity higher than 30%-35% are found to be structurally similar, the most similar protein structure pairs appear to have less than 12% pairwise sequence identity; the average sequence identity between all pairs of similar structures is between 8% and 10% [26]. Thus, entirely different protein sequences can produce very similar structures [32,30]. When such proteins are expected to share a common function, a sequence-based function prediction would fail, where a network topology-based one would not.…”