Human Quinone Reductase 2 (NQO2) is a pharmacological target and has appeared in numerous screening efforts as an off-target interactor with kinase-targeted drugs. However the cellular functions of NQO2 are not known. To gain insight into the potential cellular functions of NQO2, we have carried out a detailed evolutionary analysis. One of the most striking characteristics of NQO2 is that it uses conventional dihydronicotinamide cosubstrates, NADH and NADPH, extremely inefficiently, raising questions about an enzymatic function in cells. To characterize the ability of NQO2 to serve as an enzyme, the NQO2 gene was disrupted in HCT116 cells. These NQO2 knockouts along with the parental cells were used to demonstrate that cellular NQO2 is unable to catalyze the activation of the DNA cross-linking reagent, CB1954, without the addition of exogenous dihydronicotinamide riboside (NRH). To find whether the unusual cosubstrate specificity of NQO2 has been conserved in the amniotes, recombinant NQO2 from a reptile, Alligator mississippiensis, and a bird, Anas platyrhynchos, were cloned, purified, and their catalytic activity characterized. Like the mammalian enzymes, the reptile and bird NQO2 were efficient catalysts with the small and synthetic cosubstrate N-benzyl-1,4-dihydronicotinamide but were inefficient in their use of NADH and NADPH. Therefore, the unusual cosubstrate preference of NQO2 appears to be conserved throughout the amniotes; however, we found that NQO2 is not well-conserved in the amphibians. A phylogenetic analysis indicates that NQO1 and NQO2 diverged at the time, approximately 450 MYA, when tetrapods were beginning to evolve.

Quinone reductases 1 and 2 (NQO1 and NQO2) are paralogous FAD-linked enzymes found in all amniotes. NQO1 and NQO2 have similar structures and can both catalyze reduction of quinones and other electrophiles. The two enzymes differ in their cosubstrate specificity, with NQO1 using cellular redox couples NAD(H) and NADP(H), while NQO2 is almost completely inactive with these cosubstrates, and instead uses dihydronicotinamide riboside (NRH) and small synthetic cofactors such as N-benzyl-dihydronicotinamide (BNAH). Ancient sequence reconstruction was used to investigate the catalytic properties of a predicted common ancestor and 2 additional ancestors from each of the evolutionary pathways to extant NQO1 and NQO2. We found that in all cases, the small nicotinamide cofactors NRH and BNAH were good cosubstrates for the common ancestor and the enzymes along the NQO1 and NQO2 lineages. In the case of NADH, however, extant NQO1 evolved a catalytic efficiency 100x higher than the common ancestor, while NQO2 evolved a catalytic efficiency 1000x lower than the common ancestor. These results suggest a selective pressure for evolution of NQO1 towards greater efficiency with NADH, and for NQO2 towards extremely low efficiency with NADH. These divergent trajectories have implications for the cellular functions of both enzymes, but particularly for NQO2 whose cellular functions are only beginning to be uncovered.