Evolution and regulation of dihydrodipicolinate synthase

© 2012 Dr. Benjamin Ross BurgessThe lysine biosynthesis pathway is found in plants, bacteria and some fungi but not mammals. For this reason, it has received much attention from the perspective of antibiotic design but without substantial success. Furthermore, lysine is a limiting nutrient in cereal crops and so this pathway has also received attention from an agricultural perspective to improve cereal crop yields and grain quality. The first committed step of the lysine biosynthesis pathway is catalysed by the enzyme dihydrodipicolinate synthase (DHDPS). At the commencement of this dissertation, DHDPS was understood to be an obligate homotetramer as dimeric mutants of Escherichia coli DHDPS retained only 2.5% activity relative to their wild-type counterparts. This was believed to be the result of increased dynamic fluctuations across the tight-dimer interface that are limited with the formation of a ‘dimer of dimers’. Furthermore, lysine-mediated feedback inhibition of DHDPS is known to be weak or non-existent in many Gram-positive bacteria, moderate in Gram-negative bacteria and strong in plant species. Here we demonstrate that DHDPS from the Gram-positive bacterium methicillin-resistant Staphylococcus aureus (MRSA) exists in a monomer-dimer equilibrium in solution. The dissociation constant of the dimer is ~20-fold tighter in the presence of the substrate pyruvate (KD = 1.6 nM). The dimeric form of MRSA DHDPS demonstrates similar kinetic properties to the well characterised E. coli DHDPS, with a kcat of 70 ± 2 s-1 and K0.5 for the substrates pyruvate and (S)-ASA of 0.11 ± 0.01 mM and 0.22 ± 0.02 mM, respectively, although unlike E. coli DHDPS (S)-ASA substrate inhibition was observed with a Ksi of 2.7 ± 0.9 mM. The enzyme was shown to be not feedback inhibited by lysine up to a concentration of 20 mM. The X-ray crystal structure of MRSA DHDPS was determined to a resolution of 1.45 Å and provides an explanation for the identification of an active DHDPS dimer: a significantly greater number of contacts and solvent inaccessible surface area across the tight-dimer interface. These data suggest a previously unidentified evolutionary strategy to limit flexibility across the tight-dimer interface in the DHDPS system to facilitate catalysis. To further understand the strong lysine-mediated feedback inhibition of plant DHDPS, this enzyme was cloned, expressed and characterised from the monocotyledon plant Triticum aestivum (wheat). Contrary to prior studies, wheat DHDPS exhibits similar kinetic properties to other characterised plant species with a kcat of 28 ± 0.5 s-1 and K0.5 for the substrates pyruvate and (S)-ASA of 0.98 ± 0.04 mM and 0.076 ± 0.004 mM, respectively. The enzyme demonstrates strong, co-operative and stereo-selective mixed inhibition by lysine with respect to both pyruvate and (S)-ASA, with KiEapp constants of 27 ± 2 μM and 20 ± 2 μM and KiESapp contants of 22 ± 0.5 μM and 21 ± 0.7 μM for pyruvate and (S)-ASA, respectively. Wheat DHDPS exists in a dimer-tetramer equilibrium in solution that is driven towards the tetramer in the presence of the substrates pyruvate and (S)-ASA. Interestingly, in the presence and absence of substrates, increasing concentrations of (S)-lysine stabilise the dimeric form of the enzyme in correlation with the loss of activity observed during kinetic analysis. These data suggest a previously unidentified mechanism for inhibition of DHDPS and may provide an explanation for the tight lysine-mediated feedback inhibition of plant DHDPS enzymes