l-Lysine production independent of the oxidative pentose phosphate pathway by Corynebacterium glutamicum with the Streptococcus mutans gapN gene.

We have recently developed a Corynebacterium glutamicum strain that generates NADPH via the glycolytic pathway by replacing endogenous NAD-dependent glyceraldehyde 3-phosphate dehydrogenase (GapA) with a nonphosphorylating NADP-dependent glyceraldehyde 3-phosphate dehydrogenase (GapN) from Streptococcus mutans. Strain RE2, a suppressor mutant spontaneously isolated for its improved growth on glucose from the engineered strain, was proven to be a high-potential host for l-lysine production (Takeno et al., 2010). In this study, the suppressor mutation was identified to be a point mutation in rho encoding the transcription termination factor Rho. Strain RE2 still showed retarded growth despite the mutation rho696. Our strategy for reconciling improved growth with a high level of l-lysine production was to use GapA together with GapN only in the early growth phase, and subsequently shift this combination-type glycolysis to one that depends only on GapN in the rest of the growth phase. To achieve this, we expressed gapA under the myo-inositol-inducible promoter of iolT1 encoding a myo-inositol transporter in strain RE2. The resulting strain RE2A(iol) was engineered into an l-lysine producer by introduction of a plasmid carrying the desensitized lysC, followed by examination for culture conditions with myo-inositol supplementation. We found that as a higher concentration of myo-inositol was added to the seed culture, the following fermentation period became shorter while maintaining a high level of l-lysine production. This finally reached a fermentation period comparable to that of the control GapA strain, and yielded a 1.5-fold higher production rate compared with strain RE2. The transcript level of gapA, as well as the GapA activity, in the early growth phase increased in proportion to the myo-inositol concentration and then fell to low levels in the subsequent growth phase, indicating that improved growth was a result of increased GapA activity, especially in the early growth phase. Moreover, blockade of the pentose phosphate pathway through a defect in glucose 6-phosphate dehydrogenase did not significantly affect l-lysine production in the engineered GapN strains, while a drastic decrease in l-lysine production was observed for the control GapA strain. Determination of the intracellular NADPH/NADP(+) ratios revealed that the ratios in the engineered strains were significantly higher than the ratio of the control GapA strain irrespective of the pentose phosphate pathway. These results demonstrate that our strain engineering strategy allows efficient l-lysine production independent of the oxidative pentose phosphate pathway.