"Scanning protein structure databases for serine proteases with novel catalytic triads".
serine proteases; catalytic triad; protein structure; protein structure database scanning
The catalytic mechanism of serine proteases, the most abundant group of proteolytic enzymes, is based on three structural components: the catalytic triad, the oxyanion hole, and the specificity pockets. The catalytic triad residues (serine, histidine, and aspartate) are arranged in four different ways in the polypeptide sequence of classical serine protease clans, indicating at least four different evolutionary origins of this machinery and allowing separation of these enzymes according to their triad layout. However, in theory, the triad residues can be disposed in six different orders. The reason for the other two possible arrangements not being present in known serine proteases has not been discovered. Therefore, to investigate this phenomenon, the purpose of this work is to find and/or to construct serine proteases that possess novel catalytic triads. The structures deposited in the Protein Data Bank and AlphaFold Protein Database were analyzed by an algorithm that measured the distances between specific atoms of all serine, histidine and aspartate residues throughout the protein structure and compared the results with the cut-off distance value calculated from the mean distances found in the active sites of known serine proteases. Triads that met this criterion were separated according to the order in which the three residues appear in the protein sequence. Some of those triads had the two arrangements that are not found in serine proteases, so their accessible surface area was compared to the pattern found in serine protease catalytic residues. The only triad that met this criterion was accessible to solvent but far from the surface and, therefore, inaccessible to peptides. To model a serine protease with an undocumented triad using this protein would involve removing part of this protein so substrate can reach the triad, and then including an oxyanion hole and substrate binding residues, so this protein was discarded as a model.