COPE

Winner of the 2024 call for projects.

The COPE (Chassis Optimization by Proteome-allocation Engineering) project addresses the critical need to optimize proteome allocation to enhance bioproduction. The proteome allocation regulation controls two processes important for biosynthesis :

• Expression of heterologous genes.
• Metabolic flux to maintain precursor pools.

The constraints and requirements for heterologous expression and precursors can vary with the substrates and products. Therefore, the central hypothesis of COPE is that each bioproduction application or substrate and product combination has a specific proteome allocation optimum.

However, there are no standard methods to optimize proteome allocation. Intrinsic regulation has evolved to allocate proteome between ribosome synthesis and anabolic and catabolic enzymes to determine growth. The alarmone ppGpp interacts with RNAP alongside the protein DksA to regulate allocation between the growth and the other sectors. The messenger cAMP interacts with the transcriptional regulator CRP to regulate allocations between anabolism and catabolism.

Previous approaches to target these regulators to improve allocation faced limitations:

• On the one hand complex genetic circuits and costly inducers are used to control the levels of key genes or ligands, which are not suitable for an industrial scale.
• On the other hand, discovery of stable and industrially exploitable genomic mutations necessitates laboratory evolution, a slow and biased approach that cannot be applied to non-selectable traits such as increasing product titers.

Further, the directed evolution of the key targets has been previously limited by our inability to introduce genomic mutations at a library scale. In COPE, we will combine deep-mutational scanning (DMS), based on a high throughput protocol of targeted library construction (CREPE), and artificial intelligence (AI) to overcome these limitations and develop a standard method to identify proteome allocation-optimizing mutations for any bioproduction application.

The project is organized in 4 work areas: 

• Implementation of genetic circuits to alter the intracellular ppGpp and cAMP concentrations to determine optimum levels for maximum growth rate, biomass yields or product titers on a panel of carbohydrate substrates. We will thus establish proteome allocation-phenotype relationships. We will then perform adaptive evolution of E. coli on different substrates with the GM3 platform for continuous culture and evaluate the impact of adaptation on proteome allocation.
• Employment of high-throughput Cas9-mediated genome editing method CREPE to construct mutation libraries targeting key regulatory proteins. We will use the libraries to measure the fitness of thousands of mutations in these proteins for improved growth, biomass yields, and product titers on multiple carbohydrate growth substrates. From these DMS data, we will establish genotype-phenotype relationships.
• Use of artificial intelligence (AI) to integrate data from DMS and the regulator optima identified using genetic circuits, to develop predictive models linking mutations to resource regulation. These models will allow us to identify mutations for any bioproduction application, even non-selectable traits like yields of non-growth-associated products, thus providing a powerful tool for strain engineering.
• Apply this strategy to engineer E. coli strains for industrial applications such as production of the platform chemical 2, 4-dihydroxybutyric acid (DHB) from carbohydrate feedstock.

In summary, we will develop methods to determine optimal proteome allocation for different bioproduction applications, characterize thousands of mutations in key regulatory proteins through DMS and correlate proteome allocation with DMS data using AI to develop prediction algorithms. The final objective is the development of a standardized chassis optimization platform that is strain and application agnostic, and therefore industry-ready.
 

Project lifetime:
 

2024 - 2028

Scientific manager:
 

Alaksh Choudhury (CEA)