Periodic Reporting for period 1 - ROAD (Harnessing Rubisco oxygenation reaction for advancing sustainable biotechnology)
Reporting period: 2023-05-01 to 2025-04-30
The project ROAD aimed at exploring a novel biomanufacturing concept for the conversion of CO2 into the platform chemical glycolate (precursor of different polymers) by exploiting the promiscuous oxygenase reaction of the enzyme ribulose-1,5-bisphosphate carboxylate/oxygenase (Rubisco). This reaction, generally considered as a wasteful error by the enzyme, could in principle be used to catalyse the synthesis of a key monomer for the chemical industry. To achieve this overarching goal, ROAD included different complementary approaches, encompassing creation of auxotrophic metabolic sensors for determining the enzyme’s oxygenation reaction in vivo through growth-coupled selection strategies, the development of genome editing tools to domesticate a promising bacterium (the CO2-fixing bacterium Cupriavidus necator) for performing such bioconversions, and its engineering towards the conversion of CO2 into glycolate.
1.2.1 Work Package 1 - [Screening Rubisco variants with high O2 affinity]
• Progress: satisfactory progress.
• Brief summary. We managed to create six AMS strains to sense glycolate, a downstream product of Rubisco oxygenation. Creation of these strains was a joint effort involving the group of Prof. Tobias Erb at the Max Planck Institute for Terrestrial Microbiology (Germany) and the group of Prof. Ron Milo at the Weizmann Institute of Science (Israel). We decided not to perform a single AMS as described originally in the proposal, but rather a suite of several ones, with incremental glycolate demand. Then, we adapted the most sensitive strain (the one requiring the least amount of glycolate for growth) to become a Rubisco oxygenation sensor, by deleting the glycine cleavage system (gcvTHP). Then, we created a dedicated expression vector (named pCBB2.1) where we tested 14 different phosphoribulokinases (prk) and 10 Rubisco (rbc) variants, which were chosen among fastest Rubisco isoforms present in nature (PMID: 32500941). We did not observe significant changes in the oxygenation rates determined between the different isoforms when performing our screenings at ambient CO2 conditions (0.04% CO2 v/v). Nevertheless, we observed differences in their growth rates when exposed to increasing CO2 concentrations, suggesting a different dose-response behaviour to CO2.
• Outcome and results. The WP delivered six AMS strains which can be used to sense Rubisco oxygenation and-more generally-the molecules glyoxylate and glycolate. Also, we cloned several variants of pCBB2.1 with 14 prk combinations (cloned with the archetypal Rubisco from Synechococcus elongatus), and 10 rbc variants together with the prk from C. necator. We observed that they differ in their dose-response dynamics to Rubisco.
• Achievement of scientific deliverables and milestones. The first milestone of creating the AMS strains was fully achieved and resulted in one high-impact publication in the journal Nature Communications (PMID: 40038270). The second milestone on the Rubisco screening was also fully achieved, although at the current stage we did not identify a superior isoform to use for bioproduction. We expect to deliver a second high-impact scientific publication on the development of the oxygenation screening platform and its use to monitor Rubisco oxygenation in vivo.
1.2.2 Work Package 2 - [Directed evolution of Rubisco to adjust its carboxylation-to-oxygenation ratio]
• Progress: satisfactory progress.
• Brief summary. Using the AMS sensor developed in WP1, we explored its potential as an evolutionary platform to enhance the oxygenation rate of Rubisco. We therefore conducted a traditional adaptive laboratory evolution (ALE) experiment using the archetypal Rubisco isoform from S. elongatus by performing a continuous cultivation (turbidostat) of the AMS equipped with the corresponding pCBB2.1 to improve the growth rate. We identified two mutations, one on prk and one on rbc gene, respectively. We are now retroengineering these mutations to see if they result in enhanced oxygenation. As alternative approach, we recently started experimenting directed evolution approaches-like error-prone PCR (epPCR) and site-specific mutagenesis-to scramble the rbc sequence and see if there are mutations which favour the rate of oxygenation. This second research line was created in collaboration with the group of Dr. Carlos Acevedo-Rocha, which is expert in protein engineering at our institute. This approach began later in the project and has not yet yielded favorable mutations.
• Outcome and results. Five evolved populations of AMS strains which have been genome sequenced and deep-sequenced in the pCBB2.1 plasmid. Some coherent mutations have been found on the pCBB2.1 from independent lineages that seem to converge. We are currently retroengineering them to see if there is an effect of these mutations on the oxygenation rate. Also, we performed a first round of epPCR and identified some clones with an enhanced growth rate, which likely will mean an improved oxygenation rate of Rubisco.
• Achievement of scientific deliverables and milestones. The milestone of performing directed evolution to enhance Rubisco oxygenation was partially achieved, as it resulted in strains with an improved behaviour, but we are still characterizing them. The ALE setup was also partially completed for the same reason. Moreover, in terms of deliverable we wrote a perspective article in the journal Nature Communications to discuss the benefits of using AMS strains as evolutionary platforms for enzymes, also in the context of automation and machine learning (PMID: 38658554).
1.2.3 Work Package 3 - [Bacterial strain engineering for glycolate production from CO2]
• Progress: satisfactory progress.
• Brief summary. To upgrade C. necator as an effective microbial cell factory, we first improved the state-of-the-art in its genome editing toolkit. We created the Self-splicing Intron-Based Riboswitch system for the organism and validated it for CRISPR/Cas9 and Cas12a usage. With this tool, we could generate targeted gene deletions with high efficiencies (above 70%). We therefore used the tool to delete the glycolate dehydrogenase (GlcDEF) complex in the microorganism, so that the glycolate generated from Rubisco oxygenation is not redirected into the central carbon metabolism through glyoxylate. Then, this mutant strain was used to test the ability to produce glycolate from the one-carbon feedstock formate as only carbon and energy source. We could confirm that glycolate is produced in the mg/L range when the strain is exposed at ambient CO2 concentration (0.04%), whereas not when exposed to high CO2 conditions (10%). We also transformed the pCBB2.1 plasmids measured in the WP1 and WP2 to assess the ability to produce more glycolate. We are planning to perform the same experiments also on CO2 and H2, to then inform a techno-economic assessment.
• Outcome and results. Improvement of the genome editing toolkit of C. necator to enable CRISPR/Cas9 and Cas12a counter-selection mediated by the SIBR technology. Moreover, we demonstrated the ability to convert the one-carbon feedstock formate into glycolate as effect of the native Rubisco oxygenation. When expressing heterologous Rubiscos with enhanced oxygenation, we observed an increase in glycolate titers and yields.
• Achievement of scientific deliverables and milestones. Expansion of the genome editing toolkit was fully achieved, resulting in one high-impact factor publication in the journal Trends in Biotechnology (PMID: 40087133). Demonstration of glycolate production from formate was fully achieved, but its scaling up is not completely achieved yet. This milestone would result in another scientific publication.
• Progress: satisfactory progress.
• Brief summary. We managed to create six AMS strains to sense glycolate, a downstream product of Rubisco oxygenation. Creation of these strains was a joint effort involving the group of Prof. Tobias Erb at the Max Planck Institute for Terrestrial Microbiology (Germany) and the group of Prof. Ron Milo at the Weizmann Institute of Science (Israel). We decided not to perform a single AMS as described originally in the proposal, but rather a suite of several ones, with incremental glycolate demand. Then, we adapted the most sensitive strain (the one requiring the least amount of glycolate for growth) to become a Rubisco oxygenation sensor, by deleting the glycine cleavage system (gcvTHP). Then, we created a dedicated expression vector (named pCBB2.1) where we tested 14 different phosphoribulokinases (prk) and 10 Rubisco (rbc) variants, which were chosen among fastest Rubisco isoforms present in nature (PMID: 32500941). We did not observe significant changes in the oxygenation rates determined between the different isoforms when performing our screenings at ambient CO2 conditions (0.04% CO2 v/v). Nevertheless, we observed differences in their growth rates when exposed to increasing CO2 concentrations, suggesting a different dose-response behaviour to CO2.
• Outcome and results. The WP delivered six AMS strains which can be used to sense Rubisco oxygenation and-more generally-the molecules glyoxylate and glycolate. Also, we cloned several variants of pCBB2.1 with 14 prk combinations (cloned with the archetypal Rubisco from Synechococcus elongatus), and 10 rbc variants together with the prk from C. necator. We observed that they differ in their dose-response dynamics to Rubisco.
• Achievement of scientific deliverables and milestones. The first milestone of creating the AMS strains was fully achieved and resulted in one high-impact publication in the journal Nature Communications (PMID: 40038270). The second milestone on the Rubisco screening was also fully achieved, although at the current stage we did not identify a superior isoform to use for bioproduction. We expect to deliver a second high-impact scientific publication on the development of the oxygenation screening platform and its use to monitor Rubisco oxygenation in vivo.
1.2.2 Work Package 2 - [Directed evolution of Rubisco to adjust its carboxylation-to-oxygenation ratio]
• Progress: satisfactory progress.
• Brief summary. Using the AMS sensor developed in WP1, we explored its potential as an evolutionary platform to enhance the oxygenation rate of Rubisco. We therefore conducted a traditional adaptive laboratory evolution (ALE) experiment using the archetypal Rubisco isoform from S. elongatus by performing a continuous cultivation (turbidostat) of the AMS equipped with the corresponding pCBB2.1 to improve the growth rate. We identified two mutations, one on prk and one on rbc gene, respectively. We are now retroengineering these mutations to see if they result in enhanced oxygenation. As alternative approach, we recently started experimenting directed evolution approaches-like error-prone PCR (epPCR) and site-specific mutagenesis-to scramble the rbc sequence and see if there are mutations which favour the rate of oxygenation. This second research line was created in collaboration with the group of Dr. Carlos Acevedo-Rocha, which is expert in protein engineering at our institute. This approach began later in the project and has not yet yielded favorable mutations.
• Outcome and results. Five evolved populations of AMS strains which have been genome sequenced and deep-sequenced in the pCBB2.1 plasmid. Some coherent mutations have been found on the pCBB2.1 from independent lineages that seem to converge. We are currently retroengineering them to see if there is an effect of these mutations on the oxygenation rate. Also, we performed a first round of epPCR and identified some clones with an enhanced growth rate, which likely will mean an improved oxygenation rate of Rubisco.
• Achievement of scientific deliverables and milestones. The milestone of performing directed evolution to enhance Rubisco oxygenation was partially achieved, as it resulted in strains with an improved behaviour, but we are still characterizing them. The ALE setup was also partially completed for the same reason. Moreover, in terms of deliverable we wrote a perspective article in the journal Nature Communications to discuss the benefits of using AMS strains as evolutionary platforms for enzymes, also in the context of automation and machine learning (PMID: 38658554).
1.2.3 Work Package 3 - [Bacterial strain engineering for glycolate production from CO2]
• Progress: satisfactory progress.
• Brief summary. To upgrade C. necator as an effective microbial cell factory, we first improved the state-of-the-art in its genome editing toolkit. We created the Self-splicing Intron-Based Riboswitch system for the organism and validated it for CRISPR/Cas9 and Cas12a usage. With this tool, we could generate targeted gene deletions with high efficiencies (above 70%). We therefore used the tool to delete the glycolate dehydrogenase (GlcDEF) complex in the microorganism, so that the glycolate generated from Rubisco oxygenation is not redirected into the central carbon metabolism through glyoxylate. Then, this mutant strain was used to test the ability to produce glycolate from the one-carbon feedstock formate as only carbon and energy source. We could confirm that glycolate is produced in the mg/L range when the strain is exposed at ambient CO2 concentration (0.04%), whereas not when exposed to high CO2 conditions (10%). We also transformed the pCBB2.1 plasmids measured in the WP1 and WP2 to assess the ability to produce more glycolate. We are planning to perform the same experiments also on CO2 and H2, to then inform a techno-economic assessment.
• Outcome and results. Improvement of the genome editing toolkit of C. necator to enable CRISPR/Cas9 and Cas12a counter-selection mediated by the SIBR technology. Moreover, we demonstrated the ability to convert the one-carbon feedstock formate into glycolate as effect of the native Rubisco oxygenation. When expressing heterologous Rubiscos with enhanced oxygenation, we observed an increase in glycolate titers and yields.
• Achievement of scientific deliverables and milestones. Expansion of the genome editing toolkit was fully achieved, resulting in one high-impact factor publication in the journal Trends in Biotechnology (PMID: 40087133). Demonstration of glycolate production from formate was fully achieved, but its scaling up is not completely achieved yet. This milestone would result in another scientific publication.
Overall, during the execution of ROAD, we achieved the following results: i) creation and validation of a suite of auxotrophic metabolic sensors (AMS) capable of sensing glycolate as target molecule; ii) use of such AMS to sense the oxygenation reaction of different Rubisco isoforms in vivo; iii) development of a new genome editing toolkit based on CRISPR/Cas9 for engineering C. necator; iv) using such CRISPR/Cas9 tool for deleting a key enzymatic activity to prevent glycolate consumption through wasteful pathways; v) demonstration of glycolate production in the mg/L range under CO2-fixing conditions in the engineered C. necator strain as effect of Rubisco oxygenation.