A citric acid cycle-deficient Escherichia coli as an efficient chassis for aerobic fermentations

Abstract

Tricarboxylic acid cycle (TCA cycle) plays an important role for aerobic growth of heterotrophic bacteria. Theoretically, eliminating TCA cycle would decrease carbon dissipation and facilitate chemicals biosynthesis. Here, we construct an E. coli strain without a functional TCA cycle that can serve as a versatile chassis for chemicals biosynthesis. We first use adaptive laboratory evolution to recover aerobic growth in minimal medium of TCA cycle-deficient E. coli. Inactivation of succinate dehydrogenase is a key event in the evolutionary trajectory. Supply of succinyl-CoA is identified as the growth limiting factor. By replacing endogenous succinyl-CoA dependent enzymes, we obtain an optimized TCA cycle-deficient E. coli strain. As a proof of concept, the strain is engineered for high-yield production of four separate products. This work enhances our understanding of the role of the TCA cycle in E. coli metabolism and demonstrates the advantages of using TCA cycle-deficient E. coli strain for biotechnological applications.

Fig. 1. The TCA cycle-defective Escherichia coli strain dTCA showed aerobic growth recovery in glucose minimal medium when a succinate dehydrogenase mutation was introduced.

a Growth rates of evolved TCA cycle-deficient strains comparing to iTCA strain. Values are means ± SD (n  =  3 biological replicates). b Acetate yields of evolved TCA cycle-deficient strains comparing to iTCA strain. Values are means ± SD (n  =  3 biological replicates). c Biomass yields of evolved TCA cycle-deficient strains comparing to iTCA strain. Values are means ± SD (n  =  3 biological replicates). d Glucose uptake rates of evolved TCA cycle-deficient strains comparing to iTCA strain. Values are means ± SD (n  =  3 biological replicates). e Mutated genes encoding TCA cycle enzymes in two groups of independent ALE experiments. GltA: citrate synthase; AcnB: aconitate hydratase B; SdhA: succinate dehydrogenase subunit A; FumA: fumarase A; Mqo: malate: quinone oxidoreductase. f Enzyme activities of mutated succinate dehydrogenases in evolved strains. Values are means ± SD (n  =  4 biological replicates). g Enzyme activities of mutated citrate synthases in evolved strains. Values are means ± SD (n  =  4 biological replicates). h Aerobic growth of unevolved dTCA strain after introducing succinate dehydrogenase or citrate synthase mutations. Values are means ± SD (n  =  3 biological replicates). i Aerobic growth of unevolved dTCA ΔsdhA strain after introducing citrate synthase mutations. Values are means ± SD (n  =  3 biological replicates). All above P values were calculated by unpaired two-tailed t test. dTCA: BW25113 ΔaceAΔsucAΔgadAΔgadBΔpoxB::acs; iTCA: BW25113 ΔpoxB::acs; dTCA-E1, dTCA-E2, dTCA-E3, dTCA-E4: evolved dTCA strains. Source data are provided as a Source Data file.

Fig. 2. ¹³C-metabolic flux analysis reveals rewired central carbon metabolism to counter a TCA cycle defect.

a Central carbon metabolism flux distributions (normalized to 100 mol glucose uptake) of the evolved strain dTCA-E1 and the unevolved strain iTCA. Values are means ± SD (n  =  3 biological replicates). b Carbon balance analysis of dTCA-E1 and iTCA. Values are means ± SD (n  =  3 biological replicates). c NADPH metabolism balance analysis of dTCA-E1 and iTCA. Values are means ± SD (n  =  3 biological replicates). d ATP metabolism balance analysis of dTCA-E1 and iTCA. Values are means ± SD (n  =  3 biological replicates). e NADH/FADH₂ metabolism balance analysis of dTCA-E1 and iTCA. Values are means ± SD (n  =  3 biological replicates). The co-factor metabolism balance analysis was shown as the normalized production (positive value) and consumption (negative value) per 100 mol of glucose consumed. dTCA-E1: evolved TCA cycle deficient dTCA strain (Fig. 1, Supplementary Table 1); iTCA: strain BW25113 ΔpoxB::acs with an intact TCA cycle (Fig. 1, Supplementary Table 1). Source data are provided as a Source Data file.

Fig. 3. Strategies to recover aerobic growth of TCA cycle-deficient E. coli in glucose minimal medium.

a Aerobic growth of TCA cycle-deficient E. coli BW25113 ΔaceAΔsucA with supplements. Values are means ± SD (n  =  6 biological replicates). Aerobic growth was conducted in 96-well plates shaking under 800 rpm and 37 °C. Suc: 2 mmol/L succinate; 3AAs: 1 mmol/L for each of meso-2,6-diaminoheptanedioate (meso-DAP), lysine and methionine. b Schematic diagram of reprogramming E. coli to bypass succinyl-CoA requirement. c Aerobic growth of TCA cycle-deficient E. coli DDPYM strains without supplements. Values are means ± SD (n  =  6 biological replicates). Aerobic growth was conducted in 96-well plates shaking under 800 rpm and 37 °C. d Intracellular succinyl-CoA level in TCA cycle-deficient E. coli strains comparing to strain BW25113. Values are means ± SD (n  =  3 biological replicates). Aerobic growth was conducted with 50 mL culture in 250 mL flasks shaking under 200 rpm and 37 °C. 3AAs: 1 mmol/L for each of meso-2,6-diaminoheptanedioate (meso-DAP), lysine and methionine. All above P values were calculated by unpaired two-tailed t test. Source data are provided as a Source Data file.

Fig. 4. High yield biosynthesis of chemicals using TCA cycle-deficient E. coli strains.

a Overview of chemicals biosynthesis using engineered DDPYM strains. Ppc: phosphoenolpyruvate carboxylase; GdhA: glutamate dehydrogenase; AlsS: alpha-acetolactate synthase; AlsD: alpha-acetolactate decarboxylase; AckA: acetate kinase A; Pta: phosphate acetyltransferase; DAOCS: deacetoxycephalosporin C synthase. b Overexpression of deacetoxycephalosporin C synthase (DAOCS) using TCA cycle-deficient ZH40 strain as chasiss. Inducible expression of DAOCS was conducted via 6 h cultivation after adding 0.5 mmol/L IPTG at 0.5 OD₆₀₀. c Production of G-7-ADCA using TCA cycle-deficient ZH40 strain as chasiss. Values are means ± SD (n  =  3 biological replicates). d Production of glutamate in TCA cycle-deficient chassis ZH40 with plasmid pSC2s-gdhA-ppc. Values are means ± SD (n  =  3 biological replicates). e Production of acetate in TCA cycle-deficient chassis strains comparing to BW25113 and DDPYM. Values are means ± SD (n  =  3 biological replicates). f Fraction of carbon flux from glucose to metabolites during whole-cell catalysis. Values are means ± SD (n  =  3 biological replicates). DDPYM: BW25113 ΔdapD::dapH(Bs)-dapL(Bs)-patA(Bs) ΔmetA::yjcI(Bs)-metA(Bs). Detailed information on all strains is listed in Supplementary Table 1. All above P values were calculated by unpaired two-tailed t test. Source data are provided as a Source Data file.