Computational tools play an important role in the description of biological systems. Scientists describe and study structure, conformational changes and interactions between molecules in silico, often as a cheaper and faster alternative for biosynthesis. The simulated dynamic behavior in time of a molecular system is a straight forward source of information about substrate-enzyme interactions at the atomic level, and a powerful tool for the identification of molecular properties important in enzymatic reactions. Our study is focused on the computational investigation of structure and substrate specificity of hydrolases important in biotransformation. The computational work was performed in close collaboration with biochemists-experimentalists from Charles University and the Microbiological Institute of the Academy of Sciences of the Czech Republic. Hydrolases have great a potential in the chemoenzymatic synthesis of modified carbohydrates with regulated properties. Carbohydrates, as substrates of hydrolases, are important in normal functionality of many organisms. They have a dual role in immune response regulation: some carbohydrates (like GlcNAc and ManNAc) participate in activation and some (like GalNAc) in suppressing immunity; glycosidase deficiency is associated with a number of lysosomal disorders. We used homology modeling, computational docking and molecular dynamics simulation (MD) methods for the complex study of fungal hydrolases: alpha-galactosidase/alpha-N-acetylgalactosaminidase from Aspergillus niger; beta-N-acetylhexosaminidases (HEX) (from Aspergillus oryzae and Penicillium oxalicum); nitrilase from Aspergillus niger. Our structural study unambigously demonstrates that the enzyme encoded by genes variant A (aglA) from A. niger is able to accept alpha-N-acetylgalactosamine as its substrate and explains structural features responsible for its specificity. Homology models of HEXs from P. oxalicum and A. oryzae were built and compared. Homology models were used to study the role of protein glycosylation, disulfide bonds, dimer formation and interaction with natural and modified substrates. Model of nitrilase from Aspergillus niger helped to analyze multimer formation.

Nitric oxide (NO) is an important signaling molecule in plants. In the pollen of Arabidopsis thaliana, NO causes re-orientation of the growing tube and this response is mediated by 3′,5′-cyclic guanosine monophosphate (cGMP). However, in plants, NO-sensors have remained somewhat elusive. Here, the findings of an NO-binding candidate, Arabidopsis thaliana DIACYLGLYCEROL KINASE 4 (ATDGK4; AT5G57690) is presented. In addition to the annotated diacylglycerol kinase domain, this molecule also harbors a predicted heme-NO/oxygen (H-NOX) binding site and a guanylyl cyclase (GC) catalytic domain which have been identified based on the alignment of functionally conserved amino acid residues across species. A 3D model of the molecule was constructed, and from which the locations of the kinase catalytic center, the ATP-binding site, the GC and H-NOX domains were estimated. Docking of ATP to the kinase catalytic center was also modeled. The recombinant ATDGK4 demonstrated kinase activity in vitro, catalyzing the ATP-dependent conversion of sn-1,2-diacylglycerol (DAG) to phosphatidic acid (PA). This activity was inhibited by the mammalian DAG kinase inhibitor R59949 and importantly also by the NO donors diethylamine NONOate (DEA NONOate) and sodium nitroprusside (SNP). Recombinant ATDGK4 also has GC activity in vitro, catalyzing the conversion of guanosine-5'-triphosphate (GTP) to cGMP. The catalytic domains of ATDGK4 kinase and GC may be independently regulated since the kinase but not the GC, was inhibited by NO while Ca2+ only stimulates the GC. It is likely that the DAG kinase product, PA, causes the release of Ca2+ from the intracellular stores and Ca2+ in turn activates the GC domain of ATDGK4 through a feedback mechanism. Analysis of publicly available microarray data has revealed that ATDGK4 is highly expressed in the pollen. Here, the pollen tubes of mis-expressing atdgk4 recorded slower growth rates than the wild-type (Col-0) and importantly, they showed altered NO responses. Specifically, the mis-expressing atdgk4 pollen tubes have growth rates that were less affected by NO and showed reduced bending angles when challenged by an NO source. Further works on atdgk4 knockout/knockdown mutants will reveal the biological functions of ATDGK4 in NO and/or cGMP signaling in the pollen, and in the broader fertilization process.