Structural characterization of the FcDFR1-Dihydroflavonols interactions using Molecular dynamic symulation

Abstract

Dihydroflavonol 4-reductase (DFR) is a pivotal enzyme in the flavonoid biosynthesis pathway catalyzing the last common step that leads to anthocyanins and proanthocyanidins. DFR promotes the reduction of three dihydroflavonols: dihydrokaempferol (DHK), dihydroquercetin (DHQ) and dihydromyricetin (DHM) to leucoanthocyanidins. These substrates differ only in the number of hydroxyl groups on the B phenyl ring: only one, two or three to DHK, DHQ and DHM respectively. Recently, a new variant of DFR (DFR1), which showed an unusual preference for only DHK was identified in strawberry, meanwhile DFR2 can convert any of the three dihydroflavonols. A region of 26 amino acid residues could be relevant to identify the substrates, proposed as the binding pocket of B phenyl ring of dihydroflavonols, where an asparagine residue could be critical. To determine the importance of these differences in both proteins, a characterization at structural level by homology model methodology was carried out to FcDFR1 and FcDFR2 from the Chilean strawberry (Fragaria chiloensis). Additionally, by molecular dynamics simulation we identify differences in substrate binding mode of the proteins with DHK and DHQ. Phylogenetic analyses grouped FcDFR1 and FcDFR2 into separate clades. FcDFR1 and FcDFR2 sequences consist of 341 and 350 amino acid residues respectively, and share 78.6% sequence identity. The most important differences were found in the region that is important for substrate identification. FcDFR1 and FcDFR2 structures were obtained through comparative modeling, showing a RMSD of 2.39 A. Regarding protein-ligand interactions, in FcDFR2 a strong and stable interaction between Asn133 and the 3'-OH group on ring B of DHQ was determined by molecular dynamics simulations, but not in FcDFR1, where the equivalent residue is Ala135. In contrast, DHK without 3'-OH group could be transformed by both enzymes as stable interactions were determined. The data provides an explanation of why DFR1 could interact with DHK and not with DHQ.