Metabolic engineering of Escherichia coli for itaconate production

Interest in sustainable development together with limited amounts of fossil resources have increased the demand for production of chemicals and fuels from renewable resources. The market potential for bio-based products is growing and a transition from petrochemicals to biomass-based chemicals is ongoing. Itaconic acid is a C5-dicarboxylic acid which can be produced by microbial conversion processes. It can be easily polymerized and is an appealing building block for the chemical industry with many potential applications. However, biobased chemicals have to compete with their petrochemical counterparts, and yield and productivity of the microbial processes are therefore of the utmost importance. Traditionally itaconic acid is produced using the ascomycete Aspergillus terreus. This process is not competitive with petrochemical processes due to high production costs caused by low yields, and difficult and expensive product recovery. Maximizing product yield is important to lower production costs. This thesis looked at ways to reach theoretical maximum yield in a recombinant production host, Escherichia coli.

Chapter 2 describes the construction of an itaconate biosynthesis pathway in E. coli. The key enzyme of microbial itaconate production is cis-Aconitate decarboxylase (CadA) that converts the citric acid cycle intermediate cis-aconitate into itaconate. We focused on optimizing heterologous expression of cadA from Aspergillus terreus in E. coli. Initially this resulted in low CadA activities and production of trace amounts of itaconate. CadA was primarily present as inclusion bodies, explaining the low activity. The activity was significantly improved by using lower cultivation temperatures and mineral medium and this resulted in enhanced itaconate titres. The itaconate titre was further increased in aerobic bioreactor cultures by introducing citrate synthase and aconitase from Corynebacterium glutamicum and by deleting genes encoding phosphate acetyltransferase and lactate dehydrogenase. The maximum itaconate yield from glucose obtained in this study was only 0.09 mol/mol, due to high flux of carbon to by-products such as acetate and pyruvate. Pyruvate is a precursor molecule for itaconate biosynthesis and its accumulation suggested that the activity of CadA might be one of the rate limiting steps. It was concluded that further optimization of cadA expression, and reduction of acetate formation should be achieved to obtain higher itaconate yield.

As sufficient cis-aconitate decarboxylase activity is crucial for itaconate production, in chapter 3 ways to increase the activity of CadA were investigated. A recently characterized cis-aconitate decarboxylase of mammalian origin was therefore expressed in E.coli. The novel cis-aconitate decarboxylase from Mus musculus encoded by immunoresponsive gene 1 (irg1) produced comparable amounts of itaconate as CadA from A. terreus. In addition, the effects of codon optimization and harmonization on enzymatic activities of heterologously expressed cadA and irg1 were studied. Codon harmonization increased the activity of CadA in cell free extracts, but this did not result in higher itaconate production in bioreactor cultures. This suggests that other factors such as itaconate transport may limit the production.

In chapter 4, proof of principle for an anaerobic fermentation process for the production of itaconic acid was obtained by using the mixed acid fermentation pathway of E. coli. Itaconic acid production was redox balanced by co-producing succinate or ethanol with H2 and CO2. Expression of cadA together with citrate synthase (gltA) and aconitase (acnA) from Corynebacterium glutamicum resulted in 0.66 mM (1.2 % Cmol) itaconate under anaerobic conditions. Unexpectedly, strains started to produce significant amounts of glutamate when the itaconate pathway was introduced. As glutamate production depends on the availability of nitrogen in the medium, a nitrogen-limited medium was tested to diminish glutamate production. This enhanced the production of itaconate to up to 2.9 mM (5.4 % C mol %). Here, anaerobic production of itaconate from glucose was reported for the first time. The observed itaconate yields and productivities were still modest. Eliminating the pathways to major by-products like glutamate, succinate, and acetate, and enhancing the pathway between pyruvate and itaconate is crucial to obtain a cost-competitive anaerobic itaconic acid process production.

To investigate how itaconate production can be improved, the insights from the previous chapters together with existing scientific literature were combined with our pathway design proposals in chapter 5. The tricarboxylic acid (TCA) cycle is an important source of precursors for biobased chemicals. The opinion article takes a closer look at the metabolic engineering of TCA cycle for the production of chemicals high yield. For most TCA cycle products the maximum pathway yield is much lower than the theoretical maximum yield. For succinate, this was solved by creating two pathways to the product, using both branches of the TCA cycle, connected by the glyoxylate shunt. A similar solution cannot be applied directly for production of compounds from the oxidative branch of the TCA cycle because irreversible reactions are involved: the conversion of acetyl-CoA and glyoxylate to malate in the glyoxylate shunt and the conversion of 2-oxoglutarate into succinyl-CoA in the TCA cycle. This way, the pathway yield for products originating from the oxidative TCA cycle branch such as citrate, itaconate and L-glutamate becomes identical to the theoretical maximum. Future research should focus on implementing these solutions in suitable production hosts, and increasing the ATP yield of the production pathways. This will minimize the oxygen requirement of the process, or even allow for anaerobic operation, and should lead to reduced operational costs and maximal product yields.

In chapter 6 the implications of the overall results of this thesis for the current research status of itaconate production are presented. Solutions to optimize itaconate production strains and production process were proposed.