A goal of several projects was to determine sequence-structure-function relationship of proteins with internal tandem repeats. In particular, sequence profiles were made to search sequence databases for arrays of tandem repeats in bacterial secretory proteins. Based on this analysis beta-helix models were suggested for large domains of related bacterial secretory proteins. The detailed three-dimensional structure of filamentous hemagglutinin of Bordatella pertussis, the etiological agent of whooping cough, has been predicted and modeled. The model was confirmed by subsequent electron microscopy. This result provides an insight on molecular mechanism of the pertussis infection of the respiratory tract and has implications for vaccine development (1). In the other project, an inspection of the known structures of leucine-rich repeat proteins has been performed. These proteins form a large superfamily of more than 300 different proteins with versatile functions such as hormone receptors, enzymes, virulence factors, cell cycle regulators etc. They were classified and the relationships between their sequences, structure and function have been reviewed (2). One of the projects deals with structural arrangement of human cell envelopes, cornified epithelial cells of skin, Majority of the cell envelope proteins have sequence repeats. Possible three-dimensional structures of known proteins of human cell envelope and their interactions with each other are analyzed. A mode of interaction between involucrin and small proline-rich proteins of the skin is suggested. Latest advances and trends in prediction and modeling of proteins with repetitive sequences were summarized in recent reviews (3, 4). In particular, correlations between amino acid composition and the ability to fold into a unique protein structure were described. A classification of the protein structures based on the repeat length was proposed. The analysis suggests that deciphering of the sequence-structure-function relationship of proteins with repeats will be a fertile research subject of structural bioinformatics. Several projects on molecular modeling of globular proteins and docking of their substrates were performed. For example, a red fluorescence protein was modeled in order to understand molecular processes occurring in ?fluorescent timers?, proteins that change color with time (5). One of the ongoing projects concerns molecular docking of peptide substrates in the active site of transglutaminase enzyme. Based on this docking, a specificity of the substrate binding is suggested. This will allow understanding the regulation of cross-linking of cell envelope proteins by this enzyme. In the other project, the three-dimensional structure of C-terminal part of atypical transcription factor Prospero was predicted and modeled. The relation between the model and known experimental data on its nuclear transport was analyzed.
Showing the most recent 10 out of 17 publications
