Integrated platform for the accelerated engineering of microorganisms: application to industrial bioprocessing

Abstract

Due to climate change and uncertainties in global fuel prices, there is a need to adopt biomass derived feed stocks for sustainable manufacturing of fuels, chemicals and pharmaceuticals. As a result, many major industrial manufacturers are now seeking routes to their products that are sustainable, more efficient, and less waste or energy intensive. While bioprocesses to produce compounds ranging from therapeutic drugs to fuels have already been widely implemented, the current microbes being employed are often relatively inefficient and limited in the feedstocks they can utilise. Inherent to successful bioprocess development is the ability to rapidly and predictably engineer microbes for the efficient flux of simple biomass towards compounds of industrial significance. Current iterative and empirical processes for microbial strain improvement are limited and therefore improved enabling technologies to accelerate these processes are required. To address these issues, this thesis describes the development of a platform for the rapid and predictable engineering of microbes for industrial bioprocesses. This has been achieved through complementing an accelerated DNA assembly technique for biosynthetic pathway construction with quantitative proteomics to identify pathway bottlenecks and guide subsequent rounds of pathway optimisation. Only through the ability to rapidly construct biosynthetic pathways and then assess the failure or success of the introduced pathways can microbes be engineered in an intuitive and predictable manner. A central theme of this thesis is the optimisation and implementation of a DNA assembly technique for the construction of multicomponent pathways. Despite being a fundamental aspect of strain engineering, DNA assembly is often unreliable and time consuming. One limitation of this technique is the reduced efficiency observed in the assembly of multiple DNA fragments (as is often the case when constructing a heterologous pathway). To overcome this issue a ‘nested’ DNA assembly methodology has been developed for the predictable construction of combinatorial vector libraries and complex vectors resulting in the successful assembly of up to 10 fragments. Appropriate shake flask and microtiter plate assays were additionally developed to characterise these constructs. A parallel strand of this work has been the optimisation of the methodology to maximise throughput and efficiency whilst also ensuring the method is amenable to process automation. To exemplify the power of proteomics in guiding strain engineering the reverse glyoxylate shunt was selected as a simple benchmark heterologous pathway in the commonly used host, Escherichia coli. This pathway allows for the conversion of tricarboxylic acid cycle intermediates malate and succinate to oxaloacetate and two molecules of acetyl-CoA. Strains have been engineered to overexpress the pathway genes and tryptic digestions of cell lysates carried out. Liquid chromatography, mass spectrometry and data analysis methods have been developed for the identification of over 100 proteins from these lysates. Work was then focused on developing quantitative acquisitions which will allow for the identification of pathway bottlenecks. The coupling of techniques for pathway engineering and pathway analysis will create a step change in the speed and predictability with which microbes can be engineered for industrial application.