Reports of plastic-degrading enzymes can sound deceptively similar. One enzyme may release identifiable products from a well-characterised polymer, while another has only been predicted to work from its sequence. PAZy, the Plastics-Active Enzymes Database, helps researchers separate those claims and find the evidence behind them.
Introduced by Patrick C. F. Buchholz and colleagues in a 2022 paper in Proteins, the resource brings together information about proteins associated with plastic breakdown. Its value lies in connecting enzyme sequences, structures and publications so that researchers can investigate what was actually tested.
Experimental evidence comes first
The founding study assembled 44 protein sequences with biochemical or structural information from an earlier resource: 34 associated with polyethylene terephthalate, or PET, and ten with polyurethane, or PUR. These were the starting data for that study, not a permanent count of everything the database would contain.
Sequence searches then identified thousands of related proteins. Those matches broadened the search for promising enzymes, but similarity alone did not establish that each protein could degrade plastic. The distinction between a characterised enzyme and a computational candidate is central to interpreting the work.
PAZy’s current documentation describes a focus on biochemically characterised plastics-active enzymes and related structural evidence. It excludes uncharacterised predictions from that collection. The database has developed since the founding paper, so anyone reporting its size should identify the date and the particular collection being counted.
The plastic matters as much as the enzyme
PET is a polyester used in products including bottles and textiles. Its susceptibility to enzymatic attack depends on properties such as crystallinity, surface area and processing history. A result obtained with a finely divided, low-crystallinity sample cannot automatically be transferred to an intact bottle or a mixed waste stream.
Polyurethane presents a different challenge. It is a family of materials with varied building blocks and formulations. Activity against one polyester-based polyurethane does not demonstrate that an enzyme will break down every polyurethane product.
The experimental readout also matters. A change in a material’s mass can be informative, but the founding PAZy paper cautions that loss of additives can complicate its interpretation. Identifying the products released from the polymer provides a different kind of evidence from observing surface changes alone.
These differences help explain why comparisons of plastic-degrading enzymes need careful attention to materials and test conditions.
A guide to the underlying experiments
For a researcher choosing an enzyme, the useful next step is to open its record and follow the cited publication. Which polymer was used? At what temperature and pH? How much enzyme was added, and how was breakdown measured? Those details determine whether a reported result is relevant to another experiment.
PAZy’s activity-data documentation now describes records organised around specific combinations of protein, publication, material and assay conditions. It also explains its use of language-model-assisted extraction, validation and sampled manual checks. The original publication remains the authority for interpreting an experiment; a database entry makes that evidence easier to find.
A research resource, not a recycling process
PAZy does not by itself demonstrate an industrial solution to plastic pollution. Finding an active enzyme is one part of a larger problem that includes waste preparation, enzyme production, reaction conditions and recovery of useful products.
Its contribution is more practical: it helps researchers compare evidence, identify gaps and choose better starting points for experiments. That is valuable precisely because “plastic-degrading” can otherwise hide very different levels of proof.
Sources
- Buchholz P. C. F. et al. (2022). “Plastics degradation by hydrolytic enzymes: The plastics-active enzymes database—PAZy.” Proteins, 90, 1443–1456. Original study.
- PAZy: About the resource and activity-data documentation. Accessed 27 September 2026.
- Sulaiman S. et al. (2014). Structural study associated with the illustrated LCC enzyme. DOI: 10.1021/bi401561p; PDB 4EB0.
Update note: This archived article has been revised to distinguish experimental evidence from sequence-based predictions and to explain the limits of comparing plastic-degradation results. The original publication date and URL have been retained.
