Tilley, Clemency
(2022)
Understanding carbon flux of a heterologous metabolic pathway in Escherichia coli for sustainable methyl methacrylate biosynthesis.
PhD thesis, University of Nottingham.
Abstract
Three thousand kilotons of methyl methacrylate (MMA) are produced each year. The polymer of MMA, also known as Perspex®, has become ubiquitous in applications from construction to medical technologies since it was first synthesised in 1936. Presently, MMA is exclusively manufactured via chemical synthesis from petrochemical feedstocks, 96% of which is acetone. However, in recent years biosynthesis of an MMA precursor, butyl methacrylate (BMA), has been demonstrated in Escherichia coli. This is characterised by the up-regulation of 2-ketoisovalerate production, followed by expression of heterologous branched-chain β-ketoacid dehydrogenase, acyl-CoA oxidase (ACX), and alcohol acyltransferase (AAT). Decoupling industrial MMA production from the petrochemical industry by achieving industrial fermentation of BMA would provide a green alternative to Perspex® at a time where a shift towards low carbon technologies is increasingly supported politically, economically, and societally.
Microbial fermentation is a powerful platform for sustainable resource production to displace current fossil fuel-based manufacturing practices and has gradually gained traction in recent years. In this project, I worked to understand and improve the industrial viability of BMA biosynthesis in E. coli strain BW25113. Despite the assumption that product toxicity would be the primary hindrance to high BMA productivity, the maximum titres that were synthesised from a series of BMA-resistant E. coli mutants were 1.4 mM from biotransformation and 0.13 mM in logarithmic growth, well below the IC50 for BMA in E. coli. In examining BMA producer strains, I determined that extraneous (off target) butyl esters are formed during BMA biosynthesis: butyl acetate, butyl isobutyrate (BIB), butyl propionate, and butyl isovalerate being the most prevalent. This led me to develop a series of experiments to identify and circumvent bottlenecks, by assessing carbon flux through individual stages of BMA biosynthesis. I identified a production bottleneck at the final two steps in synthesis. These concern the oxidation of isobutyryl-CoA to methacrylyl-CoA by acyl-CoA oxidase 4 from Arabidopsis thaliana (AtACX4) and methacrylyl-CoA conversion to BMA as catalysed by a mutant alcohol acyltransferase (AATm4). I determined that AtACX4 has a low Ki of 32.8 µM for its product methacrylyl-CoA, which prevents adequate intracellular methacrylyl-CoA accumulation to ensure sufficient specificity and activity from AATm4. I used bioinformatics to identify phylogenetically related but diverse ACX and AAT enzymes to replace AtACX4 and AATm4. I followed this with Golden Gate assembly to subsequently generate an ACX-AATm4 variant library, and 8 combinatorial ACX-AAT libraries. To facilitate screening of such genomic diversity I took advantage of a novel BMA-reactive fluorescent probe to develop a semi-quantitative screening approach. Using this plate-based screen, I isolated two new ACX4s with comparable activity to ACX4 from A. thaliana, from Zea mays and Vigna radiata, as well as identifying an ACX3 from Spinacea oleracea with improved ester selectivity.
During this project I determined that the highest product titre achievable using the existing BMA production pathway was 0.175 mM. Furthermore, de-bottlenecking experiments revealed a significant carbon flux hold up at methacrylyl-CoA: AATm4 preferentially produces off target esters with a ratio of 1:74 BMA:BIB, whilst failing to compete with endogenous esterase isobutyryl-CoA consumption, at a scale of 63 mM isobutyric acid as compared to 0.026 mM BMA. My in vitro assay work on AtACX4 revealed it is substantially inhibited by low methacrylyl-CoA concentrations. My subsequent bioinformatics and screening approaches resulted in the identification of 6 novel ACX4 enzymes active on isobutyryl-CoA, as well as an ACX3 enzyme. I also developed and implemented a Golden Gate assembly system capable of more efficiently swapping in alternative oxidases and transferases into the BMA pathway. To date, the industrial target of 2 g L-1 h-1 BMA has not yet been achieved, but understanding of the limitations on carbon flux through the BMA pathway in E. coli has been expanded. This provides guidance for future engineering towards industrial bioproduction of BMA.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
![[img]](/style/images/fileicons/application_pdf.png)