Stable Platform for Mevalonate Bioproduction from CO2

Abstract

Stable production of value-added products using a microbial chassis is pivotal for determining the industrial suitability of the engineered biocatalyst. Microbial cells often lose the multicopy expression plasmids during long-term cultivations. Owing to the advantages related to titers, yields, and productivities when using a multicopy expression system compared with genomic integrations, plasmid stability is essential for industrially relevant biobased processes. Cupriavidus necator H16, a facultative chemolithoautotrophic bacterium, has been successfully engineered to convert inorganic carbon obtained from CO₂ fixation into value-added products. The application of this unique capability in the biotech industry has been hindered by C. necator H16 inability to stably maintain multicopy plasmids. In this study, we designed and tested plasmid addiction systems based on the complementation of essential genes. Among these, implementation of a plasmid addiction tool based on the complementation of mutants lacking RubisCO, which is essential for CO₂ fixation, successfully stabilized a multicopy plasmid. Expressing the mevalonate pathway operon (MvaES) using this addiction system resulted in the production of ∼10 g/L mevalonate with carbon yields of ∼25%. The mevalonate titers and yields obtained here using CO₂ are the highest achieved to date for the production of C6 compounds from C1 feedstocks.

Figure 1

Schematic representation of the upper MVA pathway and summary of the C. necator H16 derivative strains used in this study. (A) The upper mevalonate pathway facilitates the conversion of three acetyl-CoA molecules to one molecule of (R)-mevalonate via a three-step pathway. The conversion is catalyzed by an acetyl-CoA acetyltransferase/HMG-CoA reductase (Ef-MvaE) and an HMG-CoA synthase (Ef-MvaS), where HMG stands for 3-hydroxy-3-methylglutarate. Both the mvaE and mvaS genes used in this study were obtained from Enterococcus faecalis (Ef) and their DNA sequences were codon-optimized for Escherichia coli (see the Supporting Information for more details). (B) C. necator H16 derivative names and genotypes, alongside a schematic representation of their corresponding MVA synthetic operon organization and localization (on plasmid/genome integrated) and the description of their addiction systems (where applicable), are reported. RBS1 and RBS2 are synthetic ribosome binding sites generated using the RBS calculator tool (Salis et al., 2009); RBS3 is the native RBS found upstream of the C. necator H16 panC gene; RBS4 is the Cn-phaC gene native RBS; RBS5 is the native RBS found upstream of the Cn-cbbS2 gene in the cbbLS2 operon located on C. necator H16 chromosome 2; RBS6 is the Cn-phaA gene native RBS; Term is the Rho-independent terminator found downstream of the Clostridium pasteurianum fdx gene; T500 is a synthetic terminator from Yarnell and Roberts.

Figure 2

Viable counts and plasmid stability in CTRL and PAN. Number of cfu/mL normalized by the OD₆₀₀ values (cfu/mL/OD₆₀₀) generated by (A) C. necator H16/pMTL71301::araC-PBAD-mvaES (CTRL) and (B) C. necator H16 ΔpanC/pMTL71301::araC-PBAD-mvaES::panC (PAN) on LB; Pan and Tet at the time of inoculation (t = 0 h), at the time of induction with 0.2% l-arabinose (IND) and at the 24, 48, 72, 96, 120, and 168 h postinduction time points and percentage of CTRL (dark blue) and PAN (red) cells retaining the plasmid calculated at the time of inoculation (0 h), at the time of induction with 0.2% l-arabinose (IND) and at the 24, 48, 72, 96, 120, and 168 h postinduction time points, as the ratios of (C) cfu/mL on Tet/cfu/mL on LB and (D) cfu/mL on Tet/cfu/ml on Pan. The mean values were calculated based on three biological replicates and the error bars represent standard deviation. Data were analyzed using the multiple t tests statistical method. *, **, and *** indicate statistically significant differences between the CTRL and PAN strains.

Figure 3

Growth and MVA production in CTRL and PAN. Round dots represent the average of 3 OD₆₀₀ values measured at each time point, while squares correspond to the MVA titers (g/L) produced at each time point by (A) CTRL (dark blue) and (B) PAN (red). MVA production is highlighted by the shaded area. The times of induction with 0.2% l-arabinose are indicated by black arrows.

Figure 4

Viable counts and plasmid stability in CBB strain. (A) Number of cfu/mL normalized by the OD₆₀₀ values (cfu/mL/OD₆₀₀) generated by the CBB strain on LB and Tet at the time of inoculation (t = 0 h), at the time of induction with 0.2% l-arabinose (IND) and at the 24, 48, 72, 96, 120, and 168 h postinduction time points and (B) percentage of CTRL (dark blue) and CBB (dark green) cells retaining the plasmid calculated at the time of inoculation (0 h), the time of induction (IND) and at 24, 48, 72, 96, 120, and 168 h after induction as the ratios of cfu/mL Tet/cfu/mL on LB. The mean values were calculated based on three biological replicates, and the error bars represent standard deviation. The data were analyzed using the multiple t tests statistical method. *** indicates statistically significant differences between the CTRL and CBB strains.

Figure 5

Growth and MVA production by the CBB and KI strains. Round dots represent the average of 3 OD₆₀₀ values measured at each time point, while squares correspond to the MVA titers produced at each time point by (A) CBB (dark green) and (B) KI (orange). MVA production is highlighted by the shaded area. Times of induction with 0.2% l-arabinose are indicated by black arrows.

Figure 6

Viable counts and plasmid stability in the CBB_phaA strain. (A) Number of cfu/mL normalized by the OD₆₀₀ values (cfu/mL/OD₆₀₀) generated by the CBB_phaA strain on LB and Tet at the time of inoculation (t = 0 h), at the time of induction with 0.2% l-arabinose (IND) and at the 24, 48, 72, 96, 120, and 168 h postinduction time points and (B) percentage of CTRL (dark blue) and CBB_phaA (light blue) cells retaining the plasmid calculated at the time of inoculation (0 h), the time of induction (IND) and at 24, 48, 72, 96, 120, and 168 h after induction as the ratios of cfu/mL on Tet/cfu/mL on LB. The mean values were calculated based on three biological replicates and the error bars represent standard deviation The data were analyzed using the multiple t tests statistical method. *** indicates statistically significant differences between the CTRL and CBB_phaA strains.

Figure 7

Growth and MVA production were calculated by CBB_phaA. Light blue dots represent the average of 3 OD₆₀₀ values measured at each time point, while light blue squares correspond to the MVA titers produced at each time point by CBB_phaA. MVA production is highlighted by the shaded area. The time of induction with 0.2% l-arabinose is indicated by the black arrow.