Engineering Escherichia coli for production of C₁₂-C₁₄ polyhydroxyalkanoate from glucose

Abstract

Demand for sustainable materials motivates the development of microorganisms capable of synthesizing products from renewable substrates. A challenge to commercial production of polyhydroxyalkanoates (PHA), microbially derived polyesters, is engineering metabolic pathways to produce a polymer with the desired monomer composition from an unrelated and renewable source. Here, we demonstrate a metabolic pathway for converting glucose into medium-chain-length (mcl)-PHA composed primarily of 3-hydroxydodecanoate monomers. This pathway combines fatty acid biosynthesis, an acyl-ACP thioesterase to generate desired C₁₂ and C₁₄ fatty acids, β-oxidation for conversion of fatty acids to (R)-3-hydroxyacyl-CoAs, and a PHA polymerase. A key finding is that Escherichia coli expresses multiple copies of enzymes involved in β-oxidation under aerobic conditions. To produce polyhydroxydodecanoate, an acyl-ACP thioesterase (BTE), an enoyl-CoA hydratase (phaJ3), and mcl-PHA polymerase (phaC2) were overexpressed in E. coli ΔfadRABIJ. Yields were improved through expression of an acyl-CoA synthetase resulting in production over 15% CDW--the highest reported production of mcl-PHA of a defined composition from an unrelated carbon source.

Fig. 1

Metabolic pathway for mcl-PHA biosynthesis in Escherichia coli. A carbon source (i.e., glucose) is catabolized to acetyl-CoA which enters fatty acid biosynthesis for production of fatty acyl-ACPs. C₁₂ and C₁₄ acyl-ACPs are substrates for the thioesterase, BTE, which catalyzes FFA formation. An acyl-CoA synthetase (e.g., FadD) activates the FFAs for degradation via a partially intact β-oxidation cycle generating enoyl-CoAs which PhaJ hydrates to produce mcl-PHA monomers for polymerization by PhaC. The resulting monomer composition is therefore identical to that of the FFA pool generated by the thioesterase. FadR represses expression of β-oxidation genes in the absence of acyl-CoAs.

Fig. 2

Metabolism of dodecanoic acid by a library of E. coli β-oxidation knock-out strains harboring the specific fad deletion(s) indicated on the horizontal axis (e.g., K12=E. coli K-12 MG1655; R=E. coli K-12 MG1655 ΔfadR; etc.). (a) Metabolism of exogenously fed dodecanoic acid after 24 and 48 h of shake flask cultivation as a percent of the initial fatty acid concentration. (b) Metabolism of endogenously synthesized fatty acids in strains with plasmid-based expression of BTE after 48 h of cultivation. Data for both saturated (C_12:0) and total C₁₂ (including unsaturated and hydroxy) species are presented.

Fig. 3

Comparison of the effect of a fadR deletion with fadD overexpression via a chromosomal fusion of the trc promoter (Φ(P_trc-fadD)) on exogenous dodecanoic acid metabolism in E. coli over a 24 h period. Data are presented as a percent of the initial fatty acid concentration.

Fig. 4

Production of mcl-PHA in E. coli in the presence of exogenously fed dodecanoic acid or endogenously produced FFA. (a) Titer of PHA as a percentage of cell dry weight determined by quantifying 3-hydroxy fatty acid methyl esters from a PHA extraction. (b) Titer of fatty acids determined by quantifying fatty acid methyl esters (FAME) from a total lipid extraction. Strain ΔfadRABIJ was cultured in the presence of dodecanoic acid while SA01 (expressing BTE) was capable of endogenous FFA production in glucose minimal media. Please see Table S2 for individual CDW and PHA titer values.