Progesterone 5β-reductases/iridoid synthases (PRISE): gatekeeper role of highly conserved phenylalanines in substrate preference and trapping is supported by molecular dynamics simulations
Vein Patterning 1 (VEP1)-encoded progesterone 5β-reductases/iridoid synthases (PRISE) belong to the short-chain dehydrogenase/reductase superfamily of proteins. They are characterized by a set of highly conserved amino acids in the substrate-binding pocket. All PRISEs are capable of reducing the activated C=C double bond of various enones enantioselectively and therefore have a potential as biocatalysts in bioorganic synthesis. Here, recombinant forms of PRISEs of Arabidopsis thaliana and Digitalis lanata were modified using site-directed mutagenesis (SDM). In rDlP5βR, a set of highly conserved amino acids in the vicinity of the catalytic center was individually substituted for alanine resulting in considerable to complete loss of enone reductase activity. F153 and F343, which can be found in most PRISEs known, are located at the outer rim of the catalytic cavity and seem to be involved in substrate binding and their role was addressed in a series of SDM experiments. The wild-type PRISE accepted progesterone (large hydrophobic 1,4-enone) as well as 2-cyclohexen-1-one (small hydrophilic 1,4-enone), whereas the double mutant rAtP5βR_F153A_F343A converted progesterone much better than the wild-type enzyme but almost lost its capability of reducing 2-cyclohexen-1-one. Recombinant Draba aizoides P5βR (rDaP5βR) has a second pair of phenylalanines at position 156 and 345 at the rim of the binding site. These two phenylalanines were introduced into rAtP5βR_F153A_F343A and the resulting quadruple mutant rAtP5βR_F153A_F343A_V156F_V345F partly recovered the ability to reduce 2-cyclohexen-1-one. These results can best be explained by assuming a trapping mechanism in which phenylalanines at the rim of the substrate-binding pocket are involved. The dynamic behavior of individual P5βRs and mutants thereof was investigated by molecular dynamics simulations and all calculations supported the ‘gatekeeper’ role of phenylalanines at the periphery of the substrate-binding pocket. Our findings provide structural and mechanistic explanations for the different substrate preferences seen among the natural PRISEs and help to explain the large differences in catalytic efficiency found for different types of 1,4-enones.