Engineering bioplastic-degrading enzymes

Enabling self-degrading plastics and recycling by engineering bio-plastic degrading enzymes.

Project duration: November 2022 – October 2025

A pile of plastic bottles with an illustration of an enzymatic process underway.

We are identifying new bioplastic degrading enzymes using advanced engineering biology techniques. Credit: Shutterstock


Project lead

A female scientist with long brown hair smiling at the camera

Dr Hafna Ahmed

Research Scientist


Colin Scott, Andrew Warden, Lygie Esquirol, Alex Caputo

The challenge

Bioplastics are an attractive alternative to petroleum-based plastics. They have been identified as part of the solution to reducing environmental plastic waste accumulation, including by CSIRO’s Ending Plastic Waste Mission.

Made from renewable plant-based materials like corn starch, vegetable fats or cellulose, bioplastics can disintegrate and biodegrade into only carbon dioxide and water. However, this process is dependent on the right conditions (e.g. temperature, or types of microbes present) being met at the waste disposal and accumulation sites where bioplastic material ultimately ends up.

While many microbes are known to produce enzymes that degrade bioplastics, these microbes and natural enzymes are most often not at high concentrations at waste accumulation sites and are too slow in their activity for efficient biodegradation. Mismanaged bioplastics leave microplastic residue, or otherwise become landfill.

The research community is currently investigating the use of enzymes in breaking down bioplastics. For example, it could be possible to ensure bioplastic material is biodegradable by embedding tailor-made enzymes in the bioplastic material as it is being made. This enzyme could then, theoretically, activate under more easily attainable conditions (such as lower temperature) once disposed. Another approach includes using enhanced enzymes to recycle collected bioplastics back into their raw materials (called ‘monomers’). 

Both of these approaches depend on the ready availability of fit-for-purpose enzymes to degrade bioplastics. However, these enzymes need to be specifically engineered using advanced engineering biology techniques.

Our response

To generate fit-for-purpose bioplastic degrading enzymes, we are identifying new bioplastic degrading enzymes by mining data that are already publicly available (e.g. whole-genome sequences) aided by computational enzyme function prediction methods.

Candidate enzymes are then being biochemically characterised for their bioplastic degradation activity using state-of-the art techniques and equipment. This includes determination of the molecular structures of the enzymes in collaboration with the CSIRO Biomolecular Crystallisation and Characterisation (BCC) facility.

We will then engineer the best candidates using high-throughput experimental techniques in collaboration with the CSIRO BioFoundry, in combination with computational methods such as machine learning and artificial intelligence (ML/AI).


We are hoping to use this project as a platform to develop experimental-computational hybrid workflows at CSIRO that can be used to engineer enzymes for other applications as well. This will enable faster and better outcomes in enzyme and protein engineering activities at CSIRO.

Enzyme design-test-build cycles that incorporate computation methods such as ML/AI and protein property prediction is a rapidly evolving field in protein chemistry. Although some successful methodologies have been reported, most are either proprietary and not publicly available, or the methods are not interchangeable between different enzymes and proteins. This lack of generic application is partly due to the difficulty in generating high-quality high-throughput data training and testing computational models, particularly for complex enzyme reactions that involve solid substrates such as plastics.

Our work in the implementation of improved computational-experimental hybrid workflows in CSIRO for protein function identification and engineering brings us a step closer to faster and fully computational protocols, which are still several years away.

The mechanism of action of plastic degrading enzymes is poorly understood, and the biochemical data generated in this project will feed into this knowledge pool, making it more feasible to design targeted enzymes for different plastics to help tackle plastic waste accumulation. The novel fit-for-purpose plastic degrading enzymes developed through this project can enable direct end-of-life outcomes for plastics for zero harm industries, and enable the repurposing of plastic waste via novel resource recovery/modification technologies through the circular economy.


More information

Data-driven molecular design