How a bacterial enzyme discovered in a Japanese recycling facility is reshaping our understanding of plastic degradation — and opening unexpected doors in human health.

Polyethylene terephthalate, known universally as PET, is one of the most widely produced synthetic polymers on Earth — present in everything from single-use water bottles to polyester textiles — and it accumulates in landfills and ecosystems at a rate that outpaces every conventional recycling strategy currently deployed. With over 70 million tons manufactured annually and the predominant thermomechanical recycling route degrading material quality with each cycle, the case for a fundamentally different approach has grown increasingly urgent. Into this gap stepped an unlikely protagonist: Ideonella sakaiensis 201-F6, isolated from a PET film sample at a Japanese recycling facility. This organism had evolved a two-enzyme system capable of using PET as its primary carbon and energy source, and at its center was a structurally unique serine hydrolase, later known as PETase.

The catalytic logic of PETase is elegant in its precision. The enzyme operates through a canonical serine-histidine-aspartate catalytic triad, the same arrangement seen in serine proteases, and executes a two-step acylation-deacylation mechanism that cleaves the ester bonds that hold PET polymer chains together. What makes the enzyme structurally distinctive is the architecture of its substrate-binding cleft — wider and more open than that of related cutinases — allowing the bulky aromatic PET chain to access the active site. PETase attacks the solid polymer surface in an interfacial biocatalytic process, releasing soluble oligomers including mono(2-hydroxyethyl) terephthalate (MHET) as the primary product; a partner enzyme, MHETase, then completes the depolymerization cascade, cleaving MHET into terephthalic acid and ethylene glycol.

The wild-type enzyme was constrained by poor thermostability and modest efficiency against the densely packed crystalline regions of post-consumer waste. Two engineering milestones changed the trajectory. In 2020, Tournier et al. reported an engineered thermophilic cutinase variant achieving at least 90% PET depolymerization within 10 hours at industrially relevant loadings. Then, in 2022, Lu et al. used a structure-based machine learning algorithm to engineer FAST-PETase — five mutations that collectively enabled near-complete degradation of untreated post-consumer PET from 51 different thermoformed products within one week, and recovery of monomers suitable for repolymerisation into virgin-quality PET. A comprehensive structural and mechanistic review published the following year detailed how each of these rational and computational strategies exploited the enzyme’s binding-groove geometry to improve catalytic turnover.

Perhaps the most striking recent development lies not in industrial recycling but in biomedicine. Microplastics have now been detected in human blood, placental tissue, and gastrointestinal samples, with PET among the most frequently identified polymer types. In a 2024 study, Lopez-Lorenzo et al. at KTH Royal Institute of Technology demonstrated that an engineered PETase variant can depolymerize microplastic PET substrates directly in human serum at 37°C, producing terephthalic acid and MHET without measurable cytotoxicity in HEK293 cell viability assays. What began as an observation about bacterial metabolism in a recycling facility has matured, in under a decade, into a scientific program spanning polymer chemistry, protein engineering, environmental remediation, and biomedicine.