ИСТИНА

Neuraminidases play a key role in the host-pathogen interactions during influenza virus life cycle and are used as targets to design antiviral drugs. Their bacterial homologs within the corresponding sialidase superfamily of proteins remain poorly studied; however they attract increasing attention in the context of the emerging bacterial infections of the lower respiratory tract. In order to understand the molecular mechanisms of protein function and regulation it is important to study organization of the catalytic sites of enzymes as well as the poorly studied or so far non-characterized pockets and cavities within functional and regulatory binding sites in protein structures [1]. In this work the systematic bioinformatic analysis of the spatial organization of the active sites in the sialidase superfamily of proteins has been carried out. Multiple structure-guided sequence alignment has been constructed [2] based on sequence and structural information of 774 sialidases from viruses, bacteria, animals and human obtained from public databases. Two strategies were implemented to analyze the collected data. First, the Fpocket program was used to analyze topology of active sites and their location on the surface of homologous proteins from different organisms in order to establish the common features as well as differences of their structural organization. It was shown that the cavity which binds structural fragment of the sialic acid in the catalytic center is preserved throughout the superfamily and the key catalytic Arginine residue is highly conserved. At the same time the organization of the loop between residues 423-441 in viral neuraminidases was found to be different when compared to the corresponding region of other homologs. Based on the results obtained, enzymes within the sialidase superfamily were classified into three classes. The first class is represented by N1-N9 neuraminidases of influenza virus A which possess the largest active site where the 423-441 loop was found to form the so-called 430-cavity [3] characterized by high sequence and structural similarity. The second class includes influenza virus B neuraminidase and human sialidase which share some similarity in the active site region with influenza virus A neuraminidases, however several crucial differences in amino acid sequence and spatial organization of the 423-441 loop increase the exposure of the corresponding cavity to solvent. The third group of enzymes according to our classification includes trans-sialidase from Trypanosoma cruzi as well as bacterial sialidases from Salmonella typhimurium and Streptococcus pneumoniae, which all are known as causative agents of human infectious diseases. The active sites of these proteins have low sequence similarity to enzymes of the first and second classes and do not contain typical subsites. Remarkable differences observed in the structural organization of the catalytic sites of viral, bacterial and human enzymes can serve as a basis for construction of selective inhibitors. [1] Suplatov, D., & Švedas, V. (2015). Acta Naturae, 7(4), 34. [2] Suplatov, D., Kirilin, E., Arbatsky, M., Takhaveev, V., & Švedas, V. (2014). Nucleic acids research, 42(W1), W344-W349. [3] Amaro, R.E., et al. (2007). JACS, 129(25), 7764-7765. This work was supported by the Russian Science Foundation (grant #15-14-00069)