New enzymes found in landfill and compost bacteria break down plastic foam and nylon

Researchers looking for new plastic-degrading enzymes collected bacteria from a Kenyan landfill, a tropical zoo, a compost bin enriched with polyurethane, and the guts of wax worms that had been fed polyurethane foam. They were hunting for microbes that could break down two plastics with few practical recycling options: polyurethane (PUR), used in foams, coatings and elastane fibers, and nylon, a polyamide plastic used in textiles and nets. Both plastics contain chemical bonds — carbamate bonds in polyurethane, amide bonds in nylon — that are harder for enzymes to attack than the ester bonds in more commonly recycled plastics such as PET.

The effort identified 21 bacterial species positive for polyurethane-degrading activity and 8 positive for nylon-degrading activity, drawn from landfill polyurethane foam and wire, zoo polyurethane, compost polyurethane foam and elastane, nylon nets and fibers, and the wax-worm guts. No bacterial species from polyethylene or polypropylene samples showed consistent activity against the probes used. The isolates spanned several bacterial groups, including Actinomycetia, alpha-proteobacteria, gamma-proteobacteria and Bacilli, with Actinobacteria showing up disproportionately often on both plastic types.

To find which bacteria were actively breaking down these plastics, the researchers built fluorescent probes: small molecules built to mimic the chemical structure of polyurethane or nylon. One probe, called PURp1, contains a carbamate bond similar to the one found in polyurethane; cleaving it shifts the color of light it emits. The other, NYLp1, contains an amide bond like the one in nylon, and cleaving it releases a quencher that makes the molecule glow brighter. Bacteria that had taken up and cleaved these probes lit up, and were then sorted out from the rest using a technique called fluorescence-assisted cell sorting (FACS), which can process up to 10,000 cells per second.

The two probes behave differently once cleaved. The fragment released from cleaved PURp1 can diffuse freely into bacterial cells, while the fragment released from cleaved NYLp1 carries a positive charge that traps it either inside the cell that cleaved it or stuck to the outside of a nearby cell's membrane. This difference let the two probes reveal different patterns of enzyme activity. In a control test, the researchers mixed bacteria engineered to produce a known polyurethane-degrading enzyme into a much larger population of bacteria without it, at a 100-fold dilution, then sorted with the probe; every colony recovered by the sort carried the enzyme, confirming that the sorting step reliably picks out active plastic-degrading bacteria from a mixed population.

From the sorted bacteria, the enzymes responsible for the activity were purified through repeated rounds of chromatography and probe testing, then identified using mass spectrometry, a technique for identifying molecules by their mass. Ten polyurethane-degrading enzymes and two nylon-degrading enzymes were identified this way; additional related enzymes were found by searching the isolates' genomes and public databases for similar sequences, bringing the total to 12 newly reported enzymes.

Compared with previously known polyurethane- and nylon-degrading enzymes, the newly identified enzymes showed higher heat tolerance and stronger activity against a range of high-molecular-weight polyurethane polymers and nylon textiles. Their melting temperature reached up to 68.3 degrees Celsius, plus or minus 0.1 degrees, putting their heat tolerance above that of most previously described polyurethane-degrading enzymes and roughly on par with existing nylon-degrading enzymes. According to the researchers, this is the first reported case of enzymes able to break down long chains of untreated polyurethane and nylon, as well as crosslinked polyurethane, a chemically bonded network form of the plastic used in foams that is generally harder to break apart than simpler, linear versions.

This matters because most previously identified polyurethane-degrading enzymes have only been tested against the ester-based form of the plastic, or have shown activity only against a simpler related compound called ethyl carbamate, leaving very few enzymes with proven ability to break down the carbamate bond in actual polyurethane material. Among those few, only one, called UMG-SP1, had been shown to degrade the ether-based, thermoplastic form of polyurethane. Similarly, many enzymes labeled as nylon-degrading enzymes have only been tested on soluble, simpler amide compounds or on biodegradable polyamides, rather than the tougher nylon-6 or nylon-6,6 used in textiles and nets; enzymes from a family called Ntn hydrolases, the most active nylon-degraders reported to date, have shown less than 1% activity on actual nylon polymer, an outcome researchers attribute to the polymer's structure limiting the enzyme's access to bonds it could otherwise cleave.

Separate review work on polyurethane biodegradation has cataloged bacteria and fungi that break the plastic down through biofilm formation and hydrolysis of its ester and urethane bonds, using enzymes including cutinases, esterases and urethanases; that review also notes that whether a polyurethane sample degrades easily depends heavily on its chemistry, since the ester bonds in polyester-based polyurethane are far more vulnerable to microbial attack than the ether bonds in polyether-based polyurethane, which is more chemically stable. That same review describes genetic engineering approaches producing a 2.3-fold improvement in degradation efficiency in some studies, and nanoparticle-based catalysts retaining 82% of their activity across repeated use cycles, while cautioning that industrial-scale deployment still faces slow degradation speeds, sensitivity to environmental conditions, and regulatory uncertainty.

Enzymatic recycling of a different, more chemically simple plastic, PET, offers a sense of what mature versions of this approach can look like: an engineered enzyme called FAST-PETase, developed using a machine-learning tool called MutCompute to identify helpful mutations, was reported able to break down PET plastic in a matter of hours to days rather than the centuries the material can otherwise persist, according to researchers at the University of Texas at Austin. Polyurethane and nylon remain considerably further behind on that path. The new enzymes described in this work provide, in the researchers' words, starting points for further optimization through protein engineering and computational discovery, rather than an enzyme ready for industrial deployment. No comparable industrial-scale enzymatic process for polyurethane or nylon currently exists.