l-Serine overproduction with minimization of by-product synthesis by engineered Corynebacterium glutamicum

Abstract

The direct fermentative production of L-serine by Corynebacterium glutamicum from sugars is attractive. However, superfluous by-product accumulation and low L-serine productivity limit its industrial production on large scale. This study aimed to investigate metabolic and bioprocess engineering strategies towards eliminating by-products as well as increasing L-serine productivity. Deletion of alaT and avtA encoding the transaminases and introduction of an attenuated mutant of acetohydroxyacid synthase (AHAS) increased both L-serine production level (26.23 g/L) and its productivity (0.27 g/L/h). Compared to the parent strain, the by-products L-alanine and L-valine accumulation in the resulting strain were reduced by 87 % (from 9.80 to 1.23 g/L) and 60 % (from 6.54 to 2.63 g/L), respectively. The modification decreased the metabolic flow towards the branched-chain amino acids (BCAAs) and induced to shift it towards L-serine production. Meanwhile, it was found that corn steep liquor (CSL) could stimulate cell growth and increase sucrose consumption rate as well as L-serine productivity. With addition of 2 g/L CSL, the resulting strain showed a significant improvement in the sucrose consumption rate (72 %) and the L-serine productivity (67 %). In fed-batch fermentation, 42.62 g/L of L-serine accumulation was achieved with a productivity of 0.44 g/L/h and yield of 0.21 g/g sucrose, which was the highest production of L-serine from sugars to date. The results demonstrated that combined metabolic and bioprocess engineering strategies could minimize by-product accumulation and improve L-serine productivity.

Increased production of L-serine in Escherichia coli through Adaptive Laboratory Evolution

“…The most promising strain generated using this approach achieved a threefold higher titer compared to what has previously been achieved for E. coli (Mundhada et al, 2016). The value reported for the strain developed in the present study is close to highest titer and yield reported for any organism (Zhu et al, 2015).…”

A novel aceE mutation leading to a better growth profile and a higher l-serine production in a high-yield l-serine-producing Corynebacterium glutamicum strain

“…It minimized the carbon flow through PDHC and increased the accumulation of l-alanine and l-valine. According to a previous study, stepwise metabolic engineering of the l-alanine and l-valine synthesis pathways resulted in a gradual increase in l-serine production [31]. This result indicates that the carbon flux in the l-alanine and l-valine pathways is closely related to the carbon flux in the l-serine pathway.…”

“…Regular sampling at different intervals was performed for different strains, and the fermentation broth was stored to measure biomass, sugar, and amino acid concentrations. The analytical methods were the same as that in the previous report [31].…”

“…Moreover, alaT and avtA were deleted and ilvBN was attenuated in the genome of 33a∆SS to minimize the l-alanine and l-valine synthesis, resulting in strain 33a∆SSAAI, which is, to date, the best strain regarding l-serine production [31]. It is worth to note that 33a∆SSAAI still produced 1.23 g/L of l-alanine and 2.63 g/L of l-valine, and showed a similar growth rate compared with the parent strain 33a∆SS [31]. These results are not consistent with the report about ATCC13032, where they found that with alaT and avtA deletions, the resulting strain showed l-alanine auxotrophy [14].…”

“…Besides l-serine, this mutant also produces more other amino acids (such as l-alanine and l-valine) compared with the parent strain. Subsequently, metabolic engineering was performed on SYPS-062 and SYPS-062-33a [27,28,31], and the l-serine titer was further improved to 42.62 g/L in a 5 L bioreactor, which was the highest production to date [31]. Although metabolic engineering of the C. glutamicum strains resulted in improved l-serine production, the genetic characteristics of the l-serine-producing strains, such as SYPS-062 and SYPS-062-33a, have not been actively exploited.…”