Skip to main content

In Vivo Metabolite Modification in Hybrid Biological and Chemical Synthesis Systems

Final Report Summary - IVMMHBCSS (In Vivo Metabolite Modification in Hybrid Biological and Chemical Synthesis Systems)

Summary overview of main results:
This fellowship set-out to develop a hybridised multi-disciplinary approach towards small molecule synthesis by interfacing transition metal catalysis with engineered microbial metabolism. This was an extension of preliminary results generated by the Balskus Lab in 2012, where they discovered that a biocompatible Pd catalyst could engage microbially-generated H2 in a non-enzymatic alkene hydrogenation reaction.
Our initial aim for the Outgoing Phase of the fellowship was to extend this methodology to the production of new biofuels by interfacing microbial H2-production with engineered terpene biosynthesis in the organism Escherichia coli. This was then to be integrated into a continuous flow set-up in the Ley Lab during the Return Phase.
Preliminary studies in Harvard (ca. 6 months) could not identify a transition metal complex that could reduce farnesene to farnesane in the presence of E. coli BL21(DE3) cells using externally-added H2 (over 150 metal complexes screened). During this initial screen we observed that the majority of the metal complexes were also resulting in the lysis of the bacterial cells. This indicated that many of these complexes were not biocompatible, irrespective of their reductive capabilities. This was further confirmed by performing serial dilutions and plate-count assays from spent reaction mixtures. At this point, we felt it was necessary to step-back and re-evaluate the biocompatibility of various reaction manifolds under microbial growth conditions. In doing so, our aim was to identify relationships between different modes of reactivity, classes of catalysts, reagents, etc., and the biocompatibility of the overall reaction. During this screen, we discovered that the FeTPPCl-catalyzed cyclopropanation of 4-vinylanisole using ethyl diazoacetate (EDA) was remarkably high yielding under biological conditions (75% yield in M9CA media + E. coli BL21(DE3) cells). Moreover, the yield of the reaction was increased relative to non-biological conditions (46% yield in H2O). This idea that non-enzymatic reactivity was enhanced in the presence of a microorganism was fascinating, and entirely unexpected. We therefore decided to explore this biocompatible cyclopropanaiton reaction further.
Whilst screening the reaction under a variety of microbial growth conditions we observed that the by-product of the reaction, diethyl maleate, was reduced to diethyl succinane under anaerobic culturing conditions. However, this only occurred when E. coli had survived the reaction conditions. This was an enabling discovery as it allowed us to rapidly assess the biocompatibility of various catalysts under a variety of reaction conditions by simply analysing the extracted cultures by 1H-NMR. During this screen we discovered that iron(III) phthalocyanine (FePcCl) was a highly-active and biocompatible cyclopropantion catalyst (90% yield, 2.5 mol% loading). We moved on to examine how to interface this reaction with engineered metabolism by focussing on the metabolite styrene. To do this, we established a collaboration with Prof. David Nielsen at Arizona State University. Using his lab’s metabolic engineering strategy, we transformed an evolved L-phenylalanine overproducing host strain of E. coli (NST74) with a plasmid encoding for a phenylalanine ammonia lyase (PAL2) from A. thaliana and a ferulate decarboxylase (FDC1) from S. cerevisiae. Under optimized fermentation conditions we were able to maximise styrene production from D-glucose in MM1 media to 1.65 mM (ca. 190 mg/L). Gratifyingly, when we added FePcCl and EDA to cultures of E. coli NST74_pTrc99a-PAL2/FDC1 at the point of induction of the styrene-producing pathway this culture transformed into a near-quantitative cyclopropane producing system. By feeding various diazo-containing substrates to this system we were able to generate a small panel of substituted cyclopropane products, demonstrating how the application of biocompatible chemistry in microbial fermentations can allow access to multiple products from a single metabolic pathway. We published this study in 2015 as a novel route to these non-natural, industrially important molecules directly from D-glucose in a way that was only possible by merging engineered microbial metabolism with biocompatible transition metal catalysis. This accompanied the publication of two review articles on chemical and microbiological aspects of non-enzymatic chemistry in vivo.
During my final months in the Balskus Lab, I made the intriguing observation that the addition of the amphiphilic compound TPGS-750-M to cultures of E. coli NST74_pTrc99a-PAL2/FDC1 significantly boosted styrene production to 5.5 mM (635 mg/L). This product titre was significantly above the toxicity threshold to E. coli (2.5 mM) and suggested that we were sequestering styrene from the cell. This experiment was inspired by a lecture I had attended at Harvard on the use of TPGS-750-M micelles for synthetic chemistry in water, and also by meetings I had initiated with Prof. Kristala Prather at MIT where we discussed potential future applications of biocompatible chemistry in synthetic biology. In a subsequent screen, the “designer surfactants” TPGS-750-M and TPGS-1000 outperformed over thirty other amphiphiles that had been previously reported in the literature to increase flux through engineered metabolism. During my Return Phase I developed this project further and found that (i) micelles formed by these designer surfactants associated with the outer membrane of E. coli NST74; (ii) this interaction increased the permeability of the membrane bilayer, enabling styrene to be sequestered from cell and preventing it from accumulating in the periplasm; and (iii) that the rate of our Fe-catalyzed cyclopropnation reaction was accelerated within these E. coli-associated micelles. We published this study in 2016 as the first example of biocompatible micellar catalysis in vivo using designer surfactants. This will shortly accompany a book chapter on biocompatible reaction development in Methods in Molecular Biology, which is due to be published in 2017.
Overall, this fellowship has significantly advanced our knowledge of non-enzymatic catalysis in vivo, as well as hinting at the the broader opportunities that exist at the interface of organic chemistry and synthetic biology. In addition, these studies have enabled me to establish strong and sustainable relationships with chemists, synthetic biologists, microbiologists and metabolic engineers in the USA. I believe these relationships and the research they have supported through the IOF program have not only had a positive impact on this project, but will also continue to contribute to future high impact research within the EU. As a result of this research experience I have recently been appointed as a principle investigator and Lecturer in Biotechnology at the Institute for Quantitative Biology, Biochemistry and Biotechnology in the School of Biological Sciences at the University of Edinburgh, where I am currently in the process of establishing my own independent research laboratory.