DEVELOPMENT OF ADVANCED MICROBIAL ENGINEERING PLATFORMS FOR ISOPRENOID PRODUCTION

Abstract

This thesis focused on developing novel microbial engineering platforms to advance isoprenoid production. To date, the production of a number of valuable isoprenoids have been demonstrated in model microorganisms. However, as many isoprenoids are highly hydrophobic, the production of isoprenoid has been limited by their low solubility in aqueous cell culture and inhibitory effects on cell viability. In the first project, a novel cell-free system was developed to produce desired isoprenoids without facing these issues. To avoid utilization of expensive ATP as energy supply, our system used a chemically produced precursor, dimethylallyl diphosphate [DMAPP], obtained from Wu Jie Laboratory (NUS) as the substrate. By using geraniol as a model compound, we managed to establish the workflow, but obtained low production yield. This might be due to several technical issues, including unidentified impurities in the substrate, challenges in producing and isolating desired enzymes and characterizing substrates and intermediates. With the above challenges, we concluded that producing isoprenoids in vivo would still be advantageous at this stage. This project motivated us to focus on developing new in vivo isoprenoid producing methods. In conventional in vivo isoprenoid productions, biphasic systems, consisting of aqueous cell culture media and non-aqueous extraction layers, were commonly X used to harvest hydrophobic isoprenoids that can cross cellular membranes. Such systems enabled the extraction of the isoprenoids from aqueous phase, whereby reducing the toxicity to cells and preventing product loss from air stripping. However, recovering isoprenoids from the organic phase in the biphasic systems could be a complicated process due to their similar physical properties. In the second project, we demonstrated a closed Escherichia coli fermentation system for producing amorphadiene, a precursor of antimalarial artemisinin. The closed system employed an anaerobic fermentation condition in the absence of the organic phase. With optimization, we managed to obtain ~1 g/L amorphadiene through supplementing isopentenol. This titer was comparable to the highest reported titer from a batch fermentation by E. coli. We further scaled up the system and demonstrated a high yield (80%) purification of amorphadiene. Subsequently, we attempted to functionalize the purified amorphadiene by using Cytochrome P450 (CYP450)-expressing S. cerevisiae strains under aerobic conditions. However, the results were unsatisfactory. This inspired us to develop new tools for improving heterologous protein expression in S. cerevisiae. In the third project, we generated a library of intron-aided promoters in the format of [Promoter]-[Start codon]-[Intron]-[Remaining coding sequence]. Screening this library revealed that the introns can improve the dynamic range of expression levels under the control of the same promoter. With the intron- aided promoters, we managed to improve the expression of CYP450 to enhance the conversion of XI

valencene into nootkatol (a precursor of the natural insect repellent nootkatone) by 150%. However, the production yield was still extremely low (<1%). In the future work, we proposed to collaborate with the Wu Laboratory to functionalize biosynthetic amorphadiene and valencene from our anaerobic fermentation by visible light photocatalysis. From preliminary results, we managed to develop a novel isoprenoid production system by combining biocatalysis with photocatalysis (without using CYP450s). The yield of photocatalysis of converting valencene into nootkatone was 31%.