Enhanced biosynthesis of poly(3-hydroxybutyrate) in engineered strains of Pseudomonas putida via increased malonyl-CoA availability

Abstract

Malonyl-coenzyme A (CoA) is a key precursor for the biosynthesis of multiple value-added compounds by microbial cell factories, including polyketides, carboxylic acids, biofuels, and polyhydroxyalkanoates. Owing to its role as a metabolic hub, malonyl-CoA availability is limited by competition in several essential metabolic pathways. To address this limitation, we modified a genome-reduced Pseudomonas putida strain to increase acetyl-CoA carboxylation while limiting malonyl-CoA utilization. Genes involved in sugar catabolism and its regulation, the tricarboxylic acid (TCA) cycle, and fatty acid biosynthesis were knocked-out in specific combinations towards increasing the malonyl-CoA pool. An enzyme-coupled biosensor, based on the rppA gene, was employed to monitor malonyl-CoA levels in vivo. RppA is a type III polyketide synthase that converts malonyl-CoA into flaviolin, a red-colored polyketide. We isolated strains displaying enhanced malonyl-CoA availability via a colorimetric screening method based on the RppA-dependent red pigmentation; direct flaviolin quantification identified four engineered strains had a significant increase in malonyl-CoA levels. We further modified these strains by adding a non-canonical pathway that uses malonyl-CoA as precursor for poly(3-hydroxybutyrate) biosynthesis. These manipulations led to increased polymer accumulation in the fully engineered strains, validating our general strategy to boost the output of malonyl-CoA-dependent pathways in P. putida.

FIGURE 1

Genetic modifications to boost malonyl‐CoA availability in Pseudomonas putida. (A) P. putida SEM11 was modified by knocking‐out various combinations of the four genes shown in red and reactivating a dormant FabF‐2 via promoter insertion in order to increase the malonyl‐CoA pool. In addition, the native gene cluster encoding the enzymes for PHA production and degradation was deleted to eliminate endogenous biopolymer production. Abbreviations: G6P, glucose‐6‐phosphate; 6PG, 6‐phosphogluconate; KDPG, 2‐keto‐3‐deoxy‐6‐phosphogluconate; F6P, fructose‐6‐phosphate; FBP, fructose‐1,6‐bisphosphate; G3P, glyceraldehyde‐3‐phosphate; 3PG, 3‐phosphoglycerate; DHAP, dihydroxyacetone phosphate; PEP, phosphoenolpyruvate; Cit, citrate; Acon, aconitate; Icit, isocitrate; 2‐KG, 2‐ketoglutarate; Suc‐CoA, succinyl‐CoA; Succ, succinate; Fum, fumarate; Mal, malate; and Oaa, oxaloacetate. (B) Growth curves of the resulting engineered strains. Cell density was estimated as the optical density measured at 600 nm (OD₆₀₀). QurvE software was used to analyze growth curves (Wirth, Funk, et al., ; Wirth, Rohr, et al., 2023), and the maximum specific growth rate (μ_max) was derived from the OD₆₀₀ measurements over time. GraphPad Prism 9 (GraphPad Software, Inc.) was used to perform all statistical analyses; the levels of significance are indicated as *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. The error bars represent standard deviations; n = 3.

FIGURE 2

Analysis of malonyl‐CoA levels with a biosensor. (A) A semi‐quantitative analysis of malonyl‐CoA levels in the engineered strains was carried out with an enzyme‐coupled biosensor based on the rppA gene. This gene encodes RppA, a type III polyketide synthase that converts five molecules of malonyl‐CoA into one molecule of flaviolin, which displays a red color. Malonyl‐CoA is first converted to THN by RppA (THNS), and the metabolite undergoes non‐enzymatic oxidation to flaviolin. In our design, the rppA gene was placed under the control of the constitutive P_14f promoter and the module was integrated in the genome at the attTn7 site. A workflow for assessing malonyl‐CoA levels in the engineered strain is shown; this illustration was created with BioRender.com. (B) Engineered Pseudomonas putida strains endowed with enhanced malonyl‐CoA turnover were identified through a colorimetric screening method, isolating clones that displayed increased red pigmentation in the supernatant. (C) HPLC analysis was used to quantify the flaviolin produced by the different engineered strains. Arbitrary units (A.U.) indicate the intensity of the peak corresponding to flaviolin, measured at 310 nm. GraphPad Prism 9 (GraphPad Software, Inc.) was used to perform all statistical analyses; the levels of significance are indicated as *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. The error bars represent standard deviations; n = 3.

FIGURE 3

Adopting engineered Pseudomonas putida strains for malonyl‐CoA–dependent PHA production. (A) A non‐canonical PHA biosynthesis pathway, based on NphT7 (an acetoacetyl‐CoA synthase from Streptomyces sp.), was used in this study. PhaB catalyzes the NADPH‐dependent reduction of acetoacetyl‐CoA, followed by polymerization by PhaC, a PHA synthase; both PhaB and PhaC are enzymes from C. necator. The genes encoding these three enzymes are encoded in plasmid pSEVA2311·PHAS under control of the cyclohexanone‐inducible ChnR/P _chnB expression system. (B) Maximum cell density (estimated as the optical density at 600 nm, OD₆₀₀) reached by the strains transformed with plasmid pSEVA2311·PHAS, in the presence or absence of the inducer (cyclohexanone). (C) Nile Red staining for visualization of PHB granules. Nile Red is a lipophilic fluorescent dye that binds to PHB granules and can be readily detected through fluorescence microscopy, offering qualitative evidence of biopolymer accumulation. (D) Nile red staining for semi‐quantitative assessment of PHB levels in engineered P. putida. Nile red was added to the cultures after 24 h of growth, when the cultures had reached stationary phase, and the fluorescence was read after 30 min of incubation. Nile red fluorescence values for each strain were normalized to the OD₆₀₀ of the corresponding culture; normalized fluorescence values were compared with those of the parental strain, SEM11 Δpha. The relative fluorescence of SEM11 Δpha containing an empty pSEVA2311 vector is indicated by the dotted gray line. GraphPad Prism 9 (GraphPad Software, Inc.) was used to perform all statistical analyses; the levels of significance are indicated as *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. In all cases, the error bars represent standard deviations; n = 3.