NAD-dependent dehydrogenases facilitate a diverse range of hydride transfer reactions in many areas of cellular metabolism. Dehydrogenases catalyze about 12% of metabolic reactions and are classified into 4 different groups: the medium-chain dehydrogenase superfamily, the short-chain dehydrogenase superfamily, the long-chain dehydrogenases, and the family III metal-dependent polyol dehydrogenases. The broad range of reactions catalyzed makes them attractive catalysts for synthetic process, especially since many produce chiral products. However, their use as catalysts has been limited by the requirement that the NAD cofactor be supplied in stoichiometric amounts. Enzymatic cofactor regeneration has been used for synthesis of high value products, but is not applicable to many processes because providing an extra enzyme increases costs. Electrochemical cofactor regeneration could be a lower cost alternative to enzymatic regeneration. Our overall goal is to immobilize Thermotoga maritima glycerol dehydrogenase (TmGlyDH) and its NAD cofactor onto an electrode for catalytic production of dihydroxyacetone (DHA) from glycerol where the electrode regenerates NAD+. Glycerol is a waste product of the biodiesel industry, while DHA is a more valuable synthetic precursor and sunless tanning agent.TmGlyDH’s kinetics, stability and activity were characterized to provide acceptable operating conditions for an electrochemical reactor. TmGlyDH showed cooperative rather than Michaelis-Menten kinetics with glycerol and DHA, so the Hill equation was used to determine limiting rate (Vmax), half saturating substrate concentration (K0.5) and Hill coefficient (n). The optimum pH for glycerol oxidation was 7.9. We tested alternative substrates similar to glycerol and TmGlyDH was able to produce 1,2-propanediol from hydroxyacetone at greater than 99% enantiomeric excess. To test if TmGlyDH can use immobilized NAD, the NAD analogue N6-carboxymethyl-NAD (N6-CM-NAD) was synthesized and immobilized on amino-linker-modified sepharose beads (NAD-sepharose). TmGlyDH had low activity with N6-CM-NAD and mutants were produced to increase activity with the NAD analog 10-fold, but both TmGlyDH mutants and wild-type showed similar activity with NAD-sepharose. The length of the linker between NAD and sepharose had no effect on coenzymic activity. To see if other dehydrogenases can use N6-immobilized NAD, we tested 6 different dehydrogenases and 5 of the 6 used NAD-sepharose as a cofactor; structural analysis of the enzymes binding pockets predicted activity with soluble N6-carboxymethyl-NAD, but not with NAD-sepharose.In a catalytic system, the longevity of the catalyst is important because renewing the catalyst is costly. We observed that TmGlyDH gets inactivated by its product, DHA. TmGlyDH inactivation by DHA is a result of the enzyme getting modified by Maillard reactions between the Lys and Arg residues and DHA. We identified which Lys and Arg residues get modified by DHA and prepared mutants to improve stability in the presence of DHA. The mutant most stable to DHA at 50°C was K361Q, maintaining activity twice as long as the wild-type enzyme in the presence of DHA at 50°C.In summary, TmGlyDH has been characterized and engineered towards use in a bioelectronic catalytic system. Our immobilization method of NAD is suitable for use by a wide variety of dehydrogenases, indicating a working bioelectronic system would be adaptable to many novel applications.
Read
