Further improvement of erythromycin production in E3 by engineering the molecular targets
Specific genes functioning to supply precursors for erythromycin were chosen as targets in the present study for metabolic engineering to further enhance the production of erythromycin by the overproducer E3. 2-oxoglutarate was considered one of the reporter metabolites identified by transcriptional analysis between E3 and NRRL23338 (Fig. 4). Additionally, nonsynonymous or intergenic variations were also found in genes (icd , sucA and sucB ) related to 2-oxoglutarate metabolism (Table S3). Several genes surrounding 2-oxoglutarate, i.e., icd , sucA and sucB , were stimulated in E3 with respect to NRRL23338 (Fig. 7a). As such, the metabolic flux from isocitrate to succinyl-CoA was supposed to be increased in E3. Although this still requires metabolomic data to support, metabolic engineering regarding 2-oxoglutarate was worthwhile.
Metabolic manipulations upon pathways around 2-oxoglutarate were performed by straightforwardly overexpressing icd , sucAand/or sucB in E3 (Fig. 7b). The overexpression of icd in E3 decreased the production of erythromycin by 45% and the seperate overexpression of sucA or sucB exhibited no significant effects on the production. Inspiringly,the simultaneous overexpression of sucB and sucA boosted the erythromycin titer by 40% in shake flask culture. Based on the result, bacth experiments were porformed in 5-L bioreactors with minimum essential medium (Fig. 8). Growth curves revealed that overexpression of sucBA led to a significant delaying on growth. However, the growth of E3::sucBA expedited with a faster rate after 36 h, and reached into the stationary phase at a similar time to E3. Similar trends were observed on the carbon dioxide evolution rate (CER) too. Fig.8c reveals that a rapid increase in CER occurred with both strains, with a lag of approximately 12h. After 60 h when both strains exhibited CER peaks, the CER of E3::sucBA droped rapidly, and maintained at lower levels than E3 during the stational phase. The change of CER indicated that the combinational expression of sucB and sucA accelerate the metabolic transition between exponential phase and stational phase of S. erythraea . The erythromycin titer of E3::sucBA was increased by 71% compated to E3, reaching to 967 mg/L. In consideration of glucose consumption rates, the overexpression of sucB and sucA led to a higher yield of erythromycin of E3::sucBA compared to E3.
The overexpression of icd is expected to enhance the intracellular 2-oxoglutarate pool generated from isocitrate . 2-oxoglutarate plays a role of indicating the cellular nitrogen status, and the anticipated larger 2-oxoglutarate pool in E3::icd relative to E3 signaled nitrogen deficiency that would lead to a stimulated nitrogen metabolism (Radchenko, Thornton, & Merrick, 2013), which exerts negative effects on the biosynthesis of erythromycin (Liao et al., 2015; Zhang et al., 2014). However, the overexpression of sucBA could avoid the accumulation of 2-oxoglutarate. 2-oxoglutarate dehydrogenase contributes to the degeneration of 2-oxoglutarate to succinyl-CoA. Via the catalysis by the methylmalonyl-CoA mutase, succinyl-CoA is then converted to methylmalonyl-CoA and subsequently to the biosynthesis of erythromycin. The anticipated increase of CoA-related precursors for erythromycin and decrease of 2-oxoglutarate for nitrogen metabolism might account for the titer enhancement and growth retardation by sucBA overexpression, respectively. It suggests that the overexpression of sucBAeffectively reproduced the evolution between the wild type NRRL23338 and industrial overproducer E3.