For reductive amination by L AmDH the
For reductive amination by L-AmDH, the term related to the enzyme-ammonia complex is excluded rather than the enzyme-pentanone term (Table 1). If this were an ordered mechanism, this would imply ammonia binds second, after NADH and before 2-pentanone. However, a large KM,NH4 value renders ammonia binding to the enzyme itself unlikely. More likely, the mechanism calls for ordered addition of ammonia after 2-pentanone, whereas NADH and 2-pentanone bind randomly. Product inhibition patterns (Table 4) gave further insight into substrate binding order for LeuDH and L-AmDH. For L-AmDH, competitive inhibition observed for NAD+ on NADH is consistent with previous observations for other amino ap4 australia dehydrogenases , , . However, competitive inhibition observed by 2-aminopentane on 2-pentanone at both half and full saturation of NADH has not been reported previously for amino acid dehydrogenases and indicates random binding of NADH and 2-pentanone. Additionally, the presence of noncompetitive inhibition by NAD+ on 2-pentanone and 2-aminopentane on NADH suggests the presence of dead-end complexes of enzyme-NADH-aminopentane and enzyme-NAD+-pentanone. Because an uncompetitive inhibitor cannot bind to the free enzyme, the lack of uncompetitive inhibition by 2-aminopentane on NADH at saturated 2-pentanone in the reductive amination direction rules out a kinetic mechanism where the cofactor must bind first . For LeuDH, an ordered sequential mechanism was expected based on the literature , , , . For a strictly ordered mechanism, uncompetitive inhibition is expected for all pairs listed in Table 4 except for NAD+ versus NADH, as observed for the LeuDH from Bacillus sphaericus . Instead, we see competitive inhibition for leucine versus ketoleucine and noncompetitive inhibition for leucine versus NADH, which indicates that cofactor binding is not necessarily required for leucine binding. L-AmDH and parent LeuDH differed in response to increasing solution viscosity, indicating a change in the rate-limiting steps of catalysis. For LeuDH, both the amination and deamination directions showed an inverse hyperbolic viscosity effect on normalized appkcat/KM value for the amino/amino acid substrate. An inverse hyperbolic KSVE indicates that an isomerization of the enzyme-substrate or enzyme-product complex contributes to catalysis . The kcat value encompasses all first-order rate constants in the reaction while the kcat/KM value reflects all kinetic steps starting from binding of the substrate to the first irreversible step. If the effects on the kcat and kcat/KM values were the same, then the rate-determining step would occur between substrate binding and the first irreversible step . However, because the KSVEs were different, kcat reflects a step that occurs after the first irreversible step (likely the product release step). Because the effect on appkcat/KM was hyperbolic rather than linear, diffusion effects limiting substrate capture are ruled out . A similar hyperbolic KSVE pattern was seen for the deamination of D-histidine by D-arginine dehydrogenase . Recently, a conformational change upon binding of NADPH and α-ketoglutarate by the GluDH from Aspergillus niger (An-GluDH) was demonstrated through x-ray crystallography . LeuDH and L-AmDH have a similar clamshell-like structure to An-GluDH, with two domains connected by a hinge region. The active site lies in the cleft between the two domains. Residues on both sides of the cleft have been identified as being important for either substrate binding or catalysis , . Upon binding of the substrate and cofactor, the two halves of An-GluDH move closer together by as much as 15 Å. Likely, the equilibrium relationship between open and closed conformations of LeuDH is perturbed by an increase in viscosity and favors the closed conformation and thus the formation of products . For LeuDH, a normal viscosity effect on the appkcat value was shown for reductive amination, meaning that increasing solvent viscosity decreases the overall turnover rate at substrate saturation. This effect indicates that overall turnover is partially limited by product release . The diffusion dependence is much stronger for reductive amination than for oxidative deamination. In the amination direction, a strong viscosity limitation on product release was not surprising given the large appkcat value of 330 s−1, as other enzymes with similar kcat values show similar results , , . In the deamination direction at pH 9.5, the appkcat value is 20 s−1 (data not shown); as a result, the viscosity effect in this direction is weaker. Interestingly, an inverse hyperbolic effect on the appkcat/KM,AmPent value was not observed for L-AmDH. Instead, a strong normal viscosity effect with slope of 0.75 was seen, despite the low appkcat/KM,AmPent value of 8.7 M−1 s−1. This effect is the opposite of what was seen for the reductive amination direction, and demonstrates that a viscosity effect drives an increase in concentration of the Michaelis complex in both directions. A decreased appkcat/KM,AmPent in the deamination direction indicates slow substrate release while a decreasing appkcat in the amination direction suggests slow product release; viscosity is likely affecting the same common step in each reaction direction as the substrate for the deamination direction is the product for the amination direction.