Preventing Plastic Pollution with Engineering Biology (P3EB)

The Preventing Plastic Pollution with Engineering Biology (P3EB) Hub is one of 6 major UKRI-funded Mission Hubs, designed to tackle major global challenges and transform our health and environment. The Hub will see researchers at the University of Cambridge collaborating with leading scientists across the country to discover and engineer enzymes to more efficiently degrade some of the most widely used and environmentally problematic plastics: PET, polyurethanes, polycarbonates and nylons. The team in Cambridge is led by Prof. Florian Hollfelder and will focus on the directed evolution and high-throughput screening of these novel, plastic degrading enzymes.

Plastics are man-made materials, so finding a natural enzyme able to catalyse their degradation can be challenging. Typically, new enzymes for such processes have been discovered by analysing the DNA from bacteria and other microbes and studying the how the natural enzymes they produce are able to degrade different materials. However, when no natural enzymes are available for degrading a particular material, it may be necessary to use an enzyme which catalyses a similar reaction instead. Based on the plastics targeted in this project, natural small molecule esterases, amidases and urethanases could make appropriate parent enzymes for evolution.

A crucial element of screening for successful mutants is the assay used to characterise them. Typically, researchers use a modified version of the material they want to degrade (the substrate) that produces florescent light when it is degraded. This lets them measure the amount of fluorescence emitted and when more light is emitted this indicates that the enzyme is better at degrading the substrate. However, this can bias the evolution process towards improving activity for the modified version of the substrate rather than the substrate itself. In plastic degradation this is especially important as enzymes interact differently with high molecular weight polymers than with small molecules containing the same functional groups (such as esters or amides). To address this issue, the group has designed a patented ultrahigh-throughput (kHz) nanoparticle-based assay to ensure the evolution directly improves enzyme activity on the polymers. This also allows assessment of the enzymes’ ability to deal with different material properties of the plastics, such as particle sizes and crystallinity, which are otherwise hard to screen for.

A screening method for a successful directed evolution campaign must also be fast, efficient and reliable. The pool of known, naturally occurring enzymes is vast and ever-increasing, and introducing mutations expands this pool exponentially. The Hollfelder group are experts in droplet microfluidics, a technique which can screen the activity of enzymes inside picolitre scale microreactors, minimising the time, cost and material requirements for identifying suitable enzymes. The ‘lab-on-a-chip’ designs allow the screening of up to 108 different variants per day, several orders of magnitude higher than other widely used techniques. Since microfluidics generates large data sets, it is also extremely powerful for training machine learning models. By collaborating with computational scientists through the P3EB Mission Hub, specifically Christine Orengo’s group at UCL, the plastic degrading activity of enzymes can be linked to their structural and sequence data. This allows the development of algorithms which identify distant relationships between proteins and can predict mutants with improved plastic degradation capabilities, creating a design-build-test-learn cycle which is central to the framework of synthetic biology and allows for rapid and efficient design of plastic-degrading enzymes.

2024-12-preventing-plastic-pollution-fig2.jpg

Figure 2: How droplet microfluidics enables a massive scale-down of reaction volumes, from milliliters in test tubes, beyond microliters used in plate formats and robotic liquid handling systems to picoliters in in vitro compartments. The corresponding increase in throughput possible is shown.