Combinatorial synthetic pathway fine-tuning and comparative transcriptomics for metabolic engineering of Raoultella ornithinolytica BF60 to efficiently synthesize 2,5-furandicarboxylic acid

摘要

The compound 5-hydroxymethylfurfural (HMF) has attracted much attention due to its versatility as an important bio-based platform chemical. Here, we engineered Raoultella ornithinolytica BF60 as a whole-cell biocatalyst for a highly efficient synthesis of 2,5-furandicarboxylic acid (FDCA) from HMF. Specifically, various expression cassettes of key genes, such as hmfH (gene encoding HMF/furfural oxidoreductase [HmfH]) and hmfo (gene encoding HMF oxidase), were designed and constructed for fine-tuning FDCA synthesis from HMF. The FDCA titer reached 108.9 mM with a yield of 73% when 150 mM HMF was used as the substrate. This yield was 16% higher than that without balancing key gene expression in FDCA synthetic pathways. Additionally, to strengthen HmfH expression at the translational level, ribosomal binding site (RBS) sequences, which were computationally designed using the RBS calculator, were assembled into HmfH expression cassettes. The HmfH expression in the presence of these sequences enhanced FDCA titer to 139.6 mM with a yield of 93%. Next, previously unknown candidate genes, such as aldR, dkgA, akR, AdhP1, and AdhP2, which encode enzymes that catalyze the reactions leading to the formation of the undesired product 2,5-bis(hydroxymethyl)furan (HMF alcohol) from HMF, were identified by RNA-sequencing-based transcriptomics. Combinatorial deletion of these five candidate genes led to an 88% reduction in HMF alcohol formation and 12% enhancement in FDCA production (175.6 mM). Finally, FDCA synthesis was further improved by the substrate pulse-feeding strategy, and 221.5 mM FDCA with an 88.6% yield was obtained. The combinatorial synthetic pathway fine-tuning and comparative transcriptomics approach may be useful for improving the biocatalysis efficiency of other industrially useful compounds.