Alternative routes for production of the drug candidate d-chiro-inositol with Corynebacterium glutamicum using endogenous or promiscuous plant enzymes

d-chiro-Inositol (DCI) is a promising drug candidate for treating insulin resistance and associated diseases such as type 2 diabetes or polycystic ovary syndrome. In this study, we developed two production processes for DCI using Corynebacterium glutamicum as host. In the first process, myo-inositol (MI) is oxidized to 2-keto-myo-inositol (2KMI) by the inositol dehydrogenase (IDH) IolG and then isomerized to 1-keto-d-chiro-inositol (1KDCI) by the isomerases Cg0212 or Cg2312, both of which were identified in this work. 1KDCI is then reduced to DCI by IolG. Overproduction of IolG and Cg0212 in a chassis strain unable to degrade inositols allowed the production of 1.1 g/L DCI from 10 g/L MI. As both reactions involved are reversible, only a partial conversion of MI to DCI can be achieved. To enable higher conversion ratios, a novel route towards DCI was established by utilizing the promiscuous activity of two plant-derived enzymes, the NAD+-dependent d-ononitol dehydrogenase MtOEPa and the NADPH-dependent d-pinitol dehydrogenase MtOEPb from Medicago truncatula (barrelclover). Heterologous production of these enzymes in the chassis strain led to the production of 1.6 g/L DCI from 10 g/L MI. For replacing the substrate MI by glucose, the two plant genes were co-expressed with the endogenous myo-inositol-1-phosphate synthase gene ino1 either as a synthetic operon or using a novel, bicistronic T7-based expression vector. With the single operon construct, 0.75 g/L DCI was formed from 20 g/L glucose, whereas with the bicistronic construct 1.2 g/L DCI was obtained, disclosing C. glutamicum as an attractive host for of d-chiro-inositol production.