Bioinformatics-Guided Experimental Characterization of Mismatch-Repair Enzymes and Their Relatives

Abstract

The increasing information in databases of protein sequences and structures together with the development of bioinformatic tools has helped the biochemists to identify and validate the function of many different proteins. In this chapter, we will show the successful application of two methods, proteinfold recognition (FR) and evolutionary-trace (ET) analysis to learn about the function of a group of proteins which belong to the class restriction endonucleases, namely the type II restriction endonuclease (REase) Sau3AI and the mismatch repair (MMR) protein MutH. The first method, fold recognition, makes use of sequence information to predict the secondary structure, the topology, and finally the tertiary structure of a protein. These methods are of great value for many purposes in modern biology since the available sequence information has by far exceeded the available structural information and the knowledge about the fold of a given protein is an important step towards the understanding of its function. Here, we applied several fold-recognition programs and the consensus server Pcons available via “metaservers” (Bujnicki et al. 2001; Kurowski and Bujnicki 2003) to predict the structure of the C-terminal domain (CTD) of Sau3AI. The second method, the evolutionary-trace analysis, makes use of phylogenetic and structural information of a protein family in order to identify functional sites in proteins. This method has been successfully used to predict the functional sites in a variety of different proteins (review: Lichtarge et al. 2002). Here, we use the evolutionary-trace method to identify amino acid residues in MutH, which are involved in sensing the methylation status of its recognition sequence.