Periodic Reporting for period 3 - RADICALZ (RAPID DISCOVERY AND DEVELOPMENT OF ENZYMES FOR NOVEL AND GREENER CONSUMER PRODUCTS)
Periodo di rendicontazione: 2024-06-01 al 2025-05-31
The application of enzymes in industrial processes is increasingly important to achieve the EU’s sustainability goals and strengthen the circular bioeconomy, because of their capacity to transform (bio)molecules selectively and efficiently. Thus, enzymes can help replace oil-based feedstocks for bio-based ones, preferably from waste streams, transitioning from linear to circular consumption models.
However, enzymes still find hurdles for their industrial application: tedious and expensive methods of enzyme discovery and engineering with low success rates; and limited activity/stability in the final application. The interdisciplinary and intersectoral RADICALZ consortium has delivered faster, more versatile and affordable tools for enzyme discovery and engineering, enabling the development of novel enzymes, new formulations and ingredients for more environment-friendly and healthier consumer products in a timeframe compatible with industrial development. Specifically, RADICALZ has developed: i) new droplet microfluidic tools to find suitable enzymes in large variant libraries; ii) best-in-class software solutions for faster enzyme engineering based on machine learning (ML); iii) novel enzymes and bio-based, bio-catalytically synthesized ingredients for consumer products (glycosides, wash-enhancing enzymes, bio-based thickeners, natural antioxidants and fragrances); iv) condition-responsive capsules for the formulation and triggered release of enzymes and ingredients in consumer products; v) equations for the rapid and early assessment of biocatalytic reactions.
However, enzymes still find hurdles for their industrial application: tedious and expensive methods of enzyme discovery and engineering with low success rates; and limited activity/stability in the final application. The interdisciplinary and intersectoral RADICALZ consortium has delivered faster, more versatile and affordable tools for enzyme discovery and engineering, enabling the development of novel enzymes, new formulations and ingredients for more environment-friendly and healthier consumer products in a timeframe compatible with industrial development. Specifically, RADICALZ has developed: i) new droplet microfluidic tools to find suitable enzymes in large variant libraries; ii) best-in-class software solutions for faster enzyme engineering based on machine learning (ML); iii) novel enzymes and bio-based, bio-catalytically synthesized ingredients for consumer products (glycosides, wash-enhancing enzymes, bio-based thickeners, natural antioxidants and fragrances); iv) condition-responsive capsules for the formulation and triggered release of enzymes and ingredients in consumer products; v) equations for the rapid and early assessment of biocatalytic reactions.
The Consortium’s vast resources, including large collections, metagenomes acquired in past EU-funded actions, private and public databases have been efficiently mined with in silico or in vitro methods, such as the proprietary 3DM™ systems of Bio-Prodict, sequence similarity network-based mining at BRAIN and INSA Toulouse or the microfluidic workflows developed at University of Exeter, Universidad Autónoma de Madrid and INSA Toulouse. Specifically, RADICALZ has established enzyme-catalyzed reactions for >30 bio-based ingredients (all starting from renewable or waste feedstocks) such as aroma compounds, texturizers and antioxidants for personal care and cosmetics, precursors of additives for liquid laundry detergents, taste modifiers and prebiotic compounds for nutraceuticals. Moreover, the techno-economic analysis of the 4 best reactions across different types of molecules, yielded competitive prices already at this preliminary stage, further supporting the advantages of industrial biocatalysis for consumer products.
Enabling technologies for enzyme discovery and engineering were a pillar of RADICALZ, by putting forth machine learning (ML) and microfluidics to speed up finding and improving enzyme candidates. In this respect, RADICALZ has developed 8 new droplet workflows to measure polymer synthesis or degradation, oxidations or hydrolysis at high temperatures or with single-molecule sensitivity, enzyme stability in the presence of additives commonly found in consumer products or food-grade, condition-responsive formulation of enzymes and ingredients. On the other hand, the developed ML tools developed have outperformed the state of the art in prediction competitions since their initial version. Moreover, we interfaced the experimental data obtained in the enzyme engineering campaigns and the ML tools, resulting in improved prediction capacities, as evidenced by more stable lyase variants for the synthesis of texture modifiers.
Another pillar of RADICALZ was creating a “culture of metrics” in biocatalysis. Process metrics and modelling were used project-wide to pinpoint potential pitfalls of the envisioned enzymatic processes. In the former aspect, Life Cycle Analysis (LCA) has outlined the critical relevance of process intensification and the process of raw stuff extraction (upstream) as well as wastewater treatment after the reaction (downstream). Also, Sustainable Momentum created and published a set of simple equations to be used by non-specialists early in process development, when there is not enough data for accurate LCAs.
Finally, targeted and timely innovation and exploitation analyses in tandem with awareness raising via advanced communication tools have maximized the impact of the project, which after 4 years includes 4 patents, 1 technology maturation project, 15 peer-reviewed publications (8 more manuscripts and 1 more patent under review/in preparation), 1 joint newsletter, 2 policy briefs and outstanding engagement metrics in our outreach and dissemination activities. Publications, datasets, diverse communication materials and summaries of deliverables can be found at https://zenodo.org/communities/radicalz(si apre in una nuova finestra).
Enabling technologies for enzyme discovery and engineering were a pillar of RADICALZ, by putting forth machine learning (ML) and microfluidics to speed up finding and improving enzyme candidates. In this respect, RADICALZ has developed 8 new droplet workflows to measure polymer synthesis or degradation, oxidations or hydrolysis at high temperatures or with single-molecule sensitivity, enzyme stability in the presence of additives commonly found in consumer products or food-grade, condition-responsive formulation of enzymes and ingredients. On the other hand, the developed ML tools developed have outperformed the state of the art in prediction competitions since their initial version. Moreover, we interfaced the experimental data obtained in the enzyme engineering campaigns and the ML tools, resulting in improved prediction capacities, as evidenced by more stable lyase variants for the synthesis of texture modifiers.
Another pillar of RADICALZ was creating a “culture of metrics” in biocatalysis. Process metrics and modelling were used project-wide to pinpoint potential pitfalls of the envisioned enzymatic processes. In the former aspect, Life Cycle Analysis (LCA) has outlined the critical relevance of process intensification and the process of raw stuff extraction (upstream) as well as wastewater treatment after the reaction (downstream). Also, Sustainable Momentum created and published a set of simple equations to be used by non-specialists early in process development, when there is not enough data for accurate LCAs.
Finally, targeted and timely innovation and exploitation analyses in tandem with awareness raising via advanced communication tools have maximized the impact of the project, which after 4 years includes 4 patents, 1 technology maturation project, 15 peer-reviewed publications (8 more manuscripts and 1 more patent under review/in preparation), 1 joint newsletter, 2 policy briefs and outstanding engagement metrics in our outreach and dissemination activities. Publications, datasets, diverse communication materials and summaries of deliverables can be found at https://zenodo.org/communities/radicalz(si apre in una nuova finestra).
As a non-comprehensive summary, RADICALZ has gone beyond the state of the art and created impact by:
• reducing the time and cost for enzyme discovery and engineering. We have developed 8 novel enzyme analysis workflows in microdroplet format, with 1000x-fold more throughput, 1 million times cheaper, using 20000 times less plastic that microplate assays. This reduces the cost of enzymes, which are the most expensive component of biocatalytic reactions.
• reducing the time and democratizing enzyme engineering. The innovative machine learning tools of Bio-Prodict developed in RADICALZ are truly best-in-class as publicly demonstrated in competitions. They have been further refined with experimental training data. As an example, with minimal expert intervention and in a few hours, we produced accurate predictions for 2 more stable lyases. The protein language model generated can be re-trained for different problems, turning uncountable months of work into days.
• achieving previously undescribed enzymatic reactions, e.g. several glycosylations and aqueous acylation reactions, involving bulky molecules and polyphenols. This expands the portfolio of enzymes and ingredients for consumer products, for e.g. patented antioxidants and novel dietary fibers from dairy waste streams. By biosourcing feedstocks, these greener, more sustainable consumer products contribute to sustainability and circularity.
• demonstrating economic feasibility and sustainability of reactions. By applying complex or simplified methodologies on 3 case studies, we improved our processes to achieve up to 60% reduction in CO2/kg product compared with oil-based, chemical catalyzed counterparts. Moreover, general lessons emerged, such as the need for medium recycling or process intensification. Our work has established a general roadmap for enzymatic process development.
• expanding the range of stakeholders that the project can reach. Optimizing the communication and dissemination of the project over different channels resulted in sustained interest across diverse stakeholders. Via >50 dissemination actions, >12 training actions, >15 outreach activities, >3500 social media followers, >24 blog entries, 6 educational materials, we have promoted our results, the undeniable possibilities of an enzyme-powered circular bioeconomy and congtributed to the European policy efforts towards competitive and resilient research and innovation.
• reducing the time and cost for enzyme discovery and engineering. We have developed 8 novel enzyme analysis workflows in microdroplet format, with 1000x-fold more throughput, 1 million times cheaper, using 20000 times less plastic that microplate assays. This reduces the cost of enzymes, which are the most expensive component of biocatalytic reactions.
• reducing the time and democratizing enzyme engineering. The innovative machine learning tools of Bio-Prodict developed in RADICALZ are truly best-in-class as publicly demonstrated in competitions. They have been further refined with experimental training data. As an example, with minimal expert intervention and in a few hours, we produced accurate predictions for 2 more stable lyases. The protein language model generated can be re-trained for different problems, turning uncountable months of work into days.
• achieving previously undescribed enzymatic reactions, e.g. several glycosylations and aqueous acylation reactions, involving bulky molecules and polyphenols. This expands the portfolio of enzymes and ingredients for consumer products, for e.g. patented antioxidants and novel dietary fibers from dairy waste streams. By biosourcing feedstocks, these greener, more sustainable consumer products contribute to sustainability and circularity.
• demonstrating economic feasibility and sustainability of reactions. By applying complex or simplified methodologies on 3 case studies, we improved our processes to achieve up to 60% reduction in CO2/kg product compared with oil-based, chemical catalyzed counterparts. Moreover, general lessons emerged, such as the need for medium recycling or process intensification. Our work has established a general roadmap for enzymatic process development.
• expanding the range of stakeholders that the project can reach. Optimizing the communication and dissemination of the project over different channels resulted in sustained interest across diverse stakeholders. Via >50 dissemination actions, >12 training actions, >15 outreach activities, >3500 social media followers, >24 blog entries, 6 educational materials, we have promoted our results, the undeniable possibilities of an enzyme-powered circular bioeconomy and congtributed to the European policy efforts towards competitive and resilient research and innovation.