In silico characterization of the phosphoenolpyruvate carboxylase from Methylorubrum sp. through molecular modeling, phylogenetic analysis and sequence similarity networks

OTERINO, MARÍA BELÉN; GONZALEZ, JAVIER M.

Congreso; Biofísica en tiempos de COVID-19 : Primeras Jornadas Virtuales SAB 2020; 2020

Sociedad Argentina de Biofísica

Some extremophilic bacterial strains of the genus Methylorubrum, native of arid areas of northern Argentina, use methanol as the only source of carbon and energy, and are ideal for biotechnological production of biomass from C1 compounds. In particular, they express a carbon-fixing phosphoenolpyruvate (PEP) carboxylase (PEPC), which catalyzes the irreversible carboxylation of PEP with HCO3?, to form oxaloacetate and phosphate. Unlike RuBisCO, the most widespread carbon-fixing carboxylase on Earth, PEPC is present in most life forms as an anaplerotic enzyme, it is not inhibited by O2, it does not require ATP or cofactors other than Mg2+, and is the most efficient carboxylase known to work in aerobic conditions. Methylorubrum extorquens (Mex) is a prototypical methylotroph; the closest MexPEPC homolog with known crystallographic structure is the enzyme from E. coli, which is herein used to model the three-dimensional structure of MexPEPC. For molecular modeling, the following steps were followed: search for the best template (PDB code 1fiy), alignment with the target protein with Clustal, construction of models using Modeller, and model refinement to find the best model. The quality of the model was analyzed using the Ramachandran plot, indicating that 92 % of amino acid residues are in favorable regions, therefore it was considered acceptable. For sequence similarity network (SSN) calculation, the EFI-EST online tool was used with Pfam family PF00311. The obtained networks were displayed and analyzed with Cytoscape. A sequence alignment was calculated with Mafft using sequences from selected clusters in the SSN, which were then analyzed with Jalview, and a maximum-likelihood phylogenetic tree was calculated with RAxML.From the analysis of the complete SSN network, composed of 1,494 nodes, it was found that only 54 possess a verified SwissProt description, and 5 present structures deposited in the PDB. Such clustering was combined with gene neighborhood analysis with the EFI-GNT online tool in order to identify PEPC enzymes from (C3, C4, CAM) plants, cyanobacteria, bacteria, and those specific from Methylorubrum strains. In this way, conserved residues were identified and mapped onto the MexPEPC structural model. We found a high degree of conservation in the amino acid residues that participate in the binding of the substrate in the active site, along with important differences in binding sites of regulatory ligands. This work was partly funded by grants from ANPCyT PICT 2017-4590 to J.M.G. and CONICET institutional grant PUE 2018-0035. Geordano Ropón-Palacios and his entire team at Umbrella Bioinformatics are acknowledged for their advice on molecular modeling.