Engineering improved plant Rubisco

Abstract

Rubisco is the central CO2 fixing enzyme of photosynthesis and the primary gateway through which inorganic carbon enters the biosphere. Despite its essential role in sustaining life on Earth, the catalytic inefficiencies of Rubisco severely limit plant productivity. This has made improving plant Rubisco's catalytic efficiency one of the most important, yet technically challenging goals in crop engineering. Until recently, the inaccessibility of plant Rubisco to high-throughput engineering pipelines has limited experimental progress. However, solving its complex biogenesis requirements has allowed the development of SynBio tools to produce plant Rubisco in E. coli - offering transformative opportunities for rational and directed evolution approaches for catalytic optimisation.

This thesis aimed to improve plant Rubisco by directed evolution. This was achieved by addressing four key aims: (1) developing a novel site-saturation variant library (SSVL) and nanopore sequencing based directed evolution pipeline to comprehensively map catalytically accessible sequence space for 15% of the tobacco Rubisco large subunit (RbcL), identifying new catalytic enhancing mutations, (2) developing an E. coli expression system for producing canola Rubisco, (3) testing the transferability of catalytic enhancing mutations identified in aim 1 from tobacco to canola Rubisco and (4) identifying new catalytic enhancing mutations in canola Rubisco via directed evolution. Chapter 3 introduced a new directed evolution framework that combined site-saturation variant libraries (SSVLs) targeted to key catalytic regions of tobacco RbcL with deep nanopore sequencing, establishing a high-resolution screening platform. Unlike conventional random mutagenesis, this approach enabled systematic testing of all non-wild type amino acid substitutions at each RbcL locus of interest. Functional screening in Rubisco-dependent E. coli (RDE) identified a range of mutations that enhanced catalytic activity, including up to 65% improvement in the rate of carboxylation (kcatc) and 20% improvement in carboxylation efficiency. Comprehensive biochemical characterisation of these in chapter 4 enabled predictive photosynthetic modelling that showed some mutants could support up to 21% improvements in Rubisco activity-limited photosynthesis under sub-saturating CO2 conditions. Together, chapters 3 and 4 demonstrated the utility of carefully designed directed evolution screens for optimising Rubisco beyond natural evolutionary limits. Given the inability of the tobacco Rubisco E. coli expression system to support production of canola Rubisco, chapter 5 describes the development of a canola-compatible expression system. This two-plasmid system comprised a plasmid encoding canola Rubisco (with Rubisco activase) and another encoding its cognate chloroplast folding and assembly proteins. Its capacity to produce canola Rubisco was validated along with confirmation that several catalytically beneficial mutations identified in tobacco RbcL (chapters 3 and 4) conveyed comparable improvements in canola. This motivated efforts to leverage the E. coli expression system for the directed evolution of canola Rubisco using a conventional error-prone PCR approach coupled with deep nanopore sequencing in chapter 6. An initial round of evolution primarily selected for Rubisco biogenesis (solubility) mutants, including an M309I catalytic switch, which improved kcatc but increased Kc. A second round of evolution on top of M309I identified mutations that restored Kc to improve carboxylation efficiency up to 31%.

This work provides the most comprehensive directed evolution of crop Rubiscos to date and delivers a scalable, analytically rigorous framework for future Rubisco bioengineering, laying the foundation for realising the long-standing goal of engineering more efficient Rubiscos to enhance photosynthetic performance and crop yields.