Coenzyme Q10 production in recombinant Escherichia coli strains engineered with a heterologous decaprenyl diphosphate synthase gene and foreign mevalonate pathway

In the present work, Escherichia coli DH5α was metabolically engineered for CoQ10 production by the introduction of decaprenyl diphosphate synthase gene (ddsA) from Agrobacterium tumefaciens. Grown in 2YTG medium (1.6% tryptone, 1% yeast extract, 0.5% NaCl, and 0.5% glycerol) with an initial pH of 7, the recombinant E. coli was capable of CoQ10 production up to 470 μg/gDCW (dry cell weight). This value could be further elevated to 900 μg/gDCW simply by increasing the initial culture pH from 7 to 9. Supplementation of 4-hydroxy benzoate did not improve the productivity any further. However, engineering of a lower mevalonate semi-pathway so as to increase the isopentenyl diphosphate (IPP) supply of the recombinant strain using exogenous mevalonate efficiently increased the CoQ10 production. Lower mevalonate semi-pathways of Staphylococcus aureus, Streptococcus pyogenes, Streptococcus pneumoniae, Enterococcus faecalis, and Saccharomyces cerevisiae were tested. Among these, the pathway of Streptococcus pneumoniae proved to be superior, yielding CoQ10 production of 2700±115 μg/gDCW when supplemented with exogenous mevalonate of 3 mM. In order to construct a complete mevalonate pathway, the upper semi-pathway of the same bacterium, Streptococcus pneumoniae, was recruited. In a recombinant E. coli DH5α harboring three plasmids encoding for upper and lower mevalonate semi-pathways as well as DdsA enzyme, the heterologous mevalonate pathway could convert endogenous acetyl-CoA to IPP, resulting in CoQ10 production of up to 2428±75 μg/gDCW, without mevalonate supplementation. In contrast, a whole mevalonate pathway constructed in a single operon was found to be less efficient. However, it provided CoQ10 production of up to 1706±86 μg/gDCW, which was roughly 1.9 times higher than that obtained by ddsA alone.