An engineered PET depolymerase to break down and recycle plastic bottles


Present estimates suggest that of the 359 million tons of plastics produced annually worldwide1, 150–200 million tons accumulate in landfill or in the natural environment2. Poly(ethylene terephthalate) (PET) is the most abundant polyester plastic, with almost 70 million tons manufactured annually worldwide for use in textiles and packaging3. The main recycling process for PET, via thermomechanical means, results in a loss of mechanical properties4. Consequently, de novo synthesis is preferred and PET waste continues to accumulate. With a high ratio of aromatic terephthalate units—which reduce chain mobility—PET is a polyester that is extremely difficult to hydrolyse5. Several PET hydrolase enzymes have been reported, but show limited productivity6,7. Here we describe an improved PET hydrolase that ultimately achieves, over 10 hours, a minimum of 90 per cent PET depolymerization into monomers, with a productivity of 16.7 grams of terephthalate per litre per hour (200 grams per kilogram of PET suspension, with an enzyme concentration of 3 milligrams per gram of PET). This highly efficient, optimized enzyme outperforms all PET hydrolases reported so far, including an enzyme8,9 from the bacterium Ideonella sakaiensis strain 201-F6 (even assisted by a secondary enzyme10) and related improved variants11,12,13,14 that have attracted recent interest. We also show that biologically recycled PET exhibiting the same properties as petrochemical PET can be produced from enzymatically depolymerized PET waste, before being processed into bottles, thereby contributing towards the concept of a circular PET economy.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Fig. 1: LCC outperformed all other evaluated PET hydrolases during PET-depolymerization assays.
Fig. 2: Improvement of the PET-depolymerization specific activity of LCC after mutagenesis by saturation of the residues in contact with a 2-HE(MHET)3 substrate.
Fig. 3: Improvement of LCC thermostability by addition of a disulfide bridge.
Fig. 4: Improved performance of LCC variants in enzymatic depolymerization of post-consumer PET waste.

Data availability

The authors declare that all data supporting the findings of this study are available within the article, its Extended Data, its Source Data or from the corresponding authors upon reasonable request. The atomic coordinates and structure factors of the reported structures have been deposited in the Protein Data Bank under accession codes 6THS for LCC-S165A and 6THT for ICCG-S165A.


  1. 1.

    PlasticsEurope. Plastics—the facts 2019. An analysis of European plastics production, demand and waste data. PlasticsEurope (2019).

  2. 2.

    Geyer, R., Jambeck, J. R. & Law, K. L. Production, use, and fate of all plastics ever made. Sci. Adv. 3, e1700782 (2017).

    ADS  Article  Google Scholar 

  3. 3.

    PET polymer: chemical economics handbook. IHS Markit (2018).

  4. 4.

    Ragaert, K., Delva, L. & Van Geem, K. Mechanical and chemical recycling of solid plastic waste. Waste Manag. 69, 24–58 (2017).

    CAS  Article  Google Scholar 

  5. 5.

    Marten, E., Müller, R.-J. & Deckwer, W.-D. Studies on the enzymatic hydrolysis of polyesters. II. Aliphatic-aromatic copolyesters. Polym. Degrad. Stabil. 88, 371–381 (2005).

    CAS  Article  Google Scholar 

  6. 6.

    Wei, R. & Zimmermann, W. Microbial enzymes for the recycling of recalcitrant petroleum-based plastics: how far are we? Microb. Biotechnol. 10, 1308–1322 (2017).

    CAS  Article  Google Scholar 

  7. 7.

    Kawai, F., Kawabata, T. & Oda, M. Current knowledge on enzymatic PET degradation and its possible application to waste stream management and other fields. Appl. Microbiol. Biotechnol. 103, 4253–4268 (2019).

    CAS  Article  Google Scholar 

  8. 8.

    Yoshida, S. et al. A bacterium that degrades and assimilates poly(ethylene terephthalate). Science 351, 1196–1199 (2016).

    ADS  CAS  Article  Google Scholar 

  9. 9.

    Bornscheuer, U. T. Feeding on plastic. Science 351, 1154–1155 (2016).

    ADS  CAS  Article  Google Scholar 

  10. 10.

    Palm, G. J. et al. Structure of the plastic-degrading Ideonella sakaiensis MHETase bound to a substrate. Nat. Commun. 10, 1717 (2019).

    ADS  Article  Google Scholar 

  11. 11.

    Han, X. et al. Structural insight into catalytic mechanism of PET hydrolase. Nat. Commun. 8, 2106 (2017).

    ADS  Article  Google Scholar 

  12. 12.

    Joo, S. et al. Structural insight into molecular mechanism of poly(ethylene terephthalate) degradation. Nat. Commun. 9, 382 (2018).

    ADS  Article  Google Scholar 

  13. 13.

    Austin, H. P. et al. Characterization and engineering of a plastic-degrading aromatic polyesterase. Proc. Natl Acad. Sci. USA 115, E4350–E4357 (2018).

    CAS  Article  Google Scholar 

  14. 14.

    Taniguchi, I. et al. Biodegradation of PET: current status and application aspects. ACS Catal. 9, 4089–4105 (2019).

    CAS  Article  Google Scholar 

  15. 15.

    Brueckner, T., Eberl, A., Heumann, S., Rabe, M. & Guebitz, G. M. Enzymatic and chemical hydrolysis of poly (ethylene terephthalate) fabrics. J. Polym. Sci. A 46, 6435–6443 (2008).

    CAS  Article  Google Scholar 

  16. 16.

    Vertommen, M. A., Nierstrasz, V. A., van der Veer, M. & Warmoeskerken, M. M. Enzymatic surface modification of poly(ethylene terephthalate). J. Biotechnol. 120, 376–386 (2005).

    CAS  Article  Google Scholar 

  17. 17.

    Wei, R. et al. Biocatalytic degradation efficiency of postconsumer polyethylene terephthalate packaging determined by their polymer microstructures. Adv. Sci. 6, 1900491 (2019).

    Article  Google Scholar 

  18. 18.

    Ronkvist, A. S. M., Xie, W., Lu, W. & Gross, R. A. Cutinase-catalyzed hydrolysis of poly(ethylene terephthalate). Macromolecules 42, 5128–5138 (2009).

    ADS  CAS  Article  Google Scholar 

  19. 19.

    Zimmermann, W. & Billig, S. Enzymes for the biofunctionalization of poly(ethylene terephthalate). Adv. Biochem. Eng. Biotechnol. 125, 97–120 (2010).

    Google Scholar 

  20. 20.

    Kitadokoro, K. et al. Crystal structure of cutinase Est119 from Thermobifida alba AHK119 that can degrade modified polyethylene terephthalate at 1.76 Å resolution. Polym. Degrad. Stabil. 97, 771–775 (2012).

    CAS  Article  Google Scholar 

  21. 21.

    Chen, S., Su, L., Chen, J. & Wu, J. Cutinase: characteristics, preparation, and application. Biotechnol. Adv. 31, 1754–1767 (2013).

    CAS  Article  Google Scholar 

  22. 22.

    Wei, R., Oeser, T. & Zimmermann, W. Synthetic polyester-hydrolyzing enzymes from thermophilic actinomycetes. Adv. Appl. Microbiol. 89, 267–305 (2014).

    Article  Google Scholar 

  23. 23.

    Then, J. et al. Ca2+ and Mg2+ binding site engineering increases the degradation of polyethylene terephthalate films by polyester hydrolases from Thermobifida fusca. Biotechnol. J. 10, 592–598 (2015).

    CAS  Article  Google Scholar 

  24. 24.

    Kawabata, T., Oda, M. & Kawai, F. Mutational analysis of cutinase-like enzyme, Cut190, based on the 3D docking structure with model compounds of polyethylene terephthalate. J. Biosci. Bioeng. 124, 28–35 (2017).

    CAS  Article  Google Scholar 

  25. 25.

    Sulaiman, S., You, D. J., Kanaya, E., Koga, Y. & Kanaya, S. Crystal structure and thermodynamic and kinetic stability of metagenome-derived LC-cutinase. Biochemistry 53, 1858–1869 (2014).

    CAS  Article  Google Scholar 

  26. 26.

    Then, J. et al. A disulfide bridge in the calcium binding site of a polyester hydrolase increases its thermal stability and activity against polyethylene terephthalate. FEBS Open Bio 6, 425–432 (2016).

    CAS  Article  Google Scholar 

  27. 27.

    Sowdhamini, R. et al. Stereochemical modeling of disulfide bridges. Criteria for introduction into proteins by site-directed mutagenesis. Protein Eng. 3, 95–103 (1989).

    CAS  Article  Google Scholar 

  28. 28.

    Awaja, F. & Pavel, D. Recycling of PET. Eur. Polym. J. 41, 1453–1477 (2005).

    CAS  Article  Google Scholar 

  29. 29.

    Barboza Neto, E. S., Coelho, L. A. F., Forte, M. M. C., Amico, S. C. & Ferreira, C. A. Processing of a LLDPE/HDPE pressure vessel liner by rotomolding. Mater. Res. 17, 236–241 (2014).

    CAS  Article  Google Scholar 

  30. 30.

    Fullbrook, P. D. in Glucose Syrups, Science and Technology (eds Dziedzic, S. Z. & Kearsley, M. W.) 65–115 (Elsevier, 1984).

  31. 31.

    Gusakov, A. V. et al. Design of highly efficient cellulase mixtures for enzymatic hydrolysis of cellulose. Biotechnol. Bioeng. 97, 1028–1038 (2007).

    CAS  Article  Google Scholar 

  32. 32.

    Liu, G., Zhang, J. & Bao, J. Cost evaluation of cellulose enzyme for industrial-scale cellulosic ethanol production based on rigorous Aspen Plus modeling. Bioprocess Biosyst. Eng. 39, 133–140 (2016).

    Article  Google Scholar 

  33. 33.

    Mohammad-Khah, A. & Ansari, R. Activated charcoal: preparation, characterization and applications: a review article. Int. J. Chemtech Res. 1, 859–864 (2014).

    Google Scholar 

  34. 34.

    Meyer, D. H. Process for purifying terephthalic acid. US patent 3,288,849 (1966).

  35. 35.

    Merchant Research and Consulting. Sodium sulfate: 2020 world market outlook and forecast up to 2029. (2019).

  36. 36.

    Müller, R. J., Schrader, H., Profe, J., Dresler, K. & Deckwer, W.-D. Enzymatic degradation of poly(ethylene terephthalate): rapid hydrolyse using a hydrolase from T. fusca. Macromol. Rapid Commun. 26, 1400–1405 (2005).

    Article  Google Scholar 

  37. 37.

    Sulaiman, S. et al. Isolation of a novel cutinase homolog with polyethylene terephthalate-degrading activity from leaf-branch compost by using a metagenomics approach. Appl. Environ. Microbiol. 78, 1556–1562 (2012).

    CAS  Article  Google Scholar 

Download references


We thank the ICEO facility of the Toulouse Biotechnology Institute (TBI), which is part of the Integrated Screening Platform of Toulouse (PICT, IBiSA), for providing access to ultrahigh-performance liquid chromatography (UHPLC) and protein-purification equipment; and Toulouse White Biotechnology (TWB, UMS INRAE 1337/UMS CNRS 3582) for providing access to Minibio bioreactors. We acknowledge Carbios (Saint-Beauzire, France), CRITT Bio-Industries (Toulouse, France), Pivert (Venette, France) and Leitat Technological Center (Barcelona, Spain) for their contribution to the purification of the terephthalic acid and for PET and bottle production. We also thank the Structural Biophysics Team of the Institute of Pharmacology and Structural Biology (IPBS, Toulouse, France) for access to the crystallization facility, as well as the ALBA (Barcelona, Spain) and ESRF (Grenoble, France) synchrotrons for data collection. We also acknowledge the use of High-Performance Computing resources on the Occigen (CINES, Montpellier, France) and Curie (TGCC, Paris-Saclay, France) supercomputers as well as on the Computing Mesocenter of Région Midi-Pyrénées (CALMIP, Toulouse, France). This study was supported by Truffle Capital (P. Pouletty) and a grant-in-aid for scientific research (THANAPLAST project, OSEO ISI contract number I 1206040W).

Author information




I.A., S.D., M. Chateau, V.T. and A.M. designed and directed the research. This work was further conceptualized by G.C., C.M.T., B.D. and J.N. In investigation and validation, E.M.-L., A.G., V.T. and H.T. performed enzyme engineering, purification and variant kinetic analysis. I.A., C.M.T., B.D., S.B. and C.F. performed molecular modelling. S.G. and J.N. carried out structural and physical characterization of variants. M.-L.D., M. Cot, E.G. and E.K. carried out reactions in Minibio reactors. M.D. characterized PET powders. E.G., M.D., M. Chateau and M. Cot developed the scheme for purifying terephthalic acid and supervised the production of PET and bottles. M. Chateau, A.M., I.A., V.T. and S.D. wrote the original draft. All authors reviewed and accepted the manuscript.

Corresponding authors

Correspondence to I. André or S. Duquesne or A. Marty.

Ethics declarations

Competing interests

E.G., M.D., M. Chateau and A.M. are employees of Carbios. V.T. has been an employee of Carbios since January 2019. C.M.T., H.T., V.T., M.-L.D., S.D., I.A., S.B. and A.M. have filed patents WO 2018/011284 and WO 2018/011281 for ‘Novel esterases and uses thereof’. H.T., M.-L.D., S.D., A.M., M.D. and M. Chateau have filed patent WO 2017/198786, ‘A process for degrading plastic products’, for protection of part of the work described herein. Confidentiality agreements prevent them from disclosing any newly submitted declaration of invention. All other authors declare no competing interests.

Additional information

Peer review information Nature thanks Peter Rem and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data figures and tables

Extended Data Fig. 1 LCC outperformed all other evaluated PET hydrolases during PET-depolymerization assays.

a, Enzymes used here that are reported to hydrolyse PET8,16,23,36,37. Melting temperatures (Tm) were assessed by DSF; values correspond to means ± s.d. (n = 3). b, Hydrolysis of amorphous Gf-PET using equimolar amounts of purified Is-PETase, FsC, BTA-hydrolase 1, BTA-hydrolase 2 or LCC (6.9 nmolprotein gPET−1 and 2 gPET lbuffer−1) in their respective buffers at various temperatures. The highest specific activities obtained are in bold. The specific activity of LCC towards Pf-PET is also shown. Means ± s.d. (n = 3) are indicated; n.d., not determined.

Extended Data Fig. 2 The thermostability of LCC at 65 °C is the limiting factor in PET depolymerization.

a, Depolymerization of PET by wild-type LCC is not affected by the products of hydrolysis. The graphs compare the kinetics of Pf-PET depolymerization at 65 °C with or without further addition of ethylene glycol (EG; calculated yields (percentages) are obtained from the quantity of terephthalic acid equivalents (TAeq.) released during the reaction), and with or without further addition of TA (calculated yields obtained from the quantity of EG released during the reaction). The amount of EG or TA added at reaction initiation corresponds to the quantity of products released with 100% PET depolymerization. Each symbol shows a mean ± s.d. (n = 3). b, The thermostability of LCC is a limiting factor for the PET-depolymerization yield. No change is observed upon adding 100 mg amorphized Pf-PET after 3 days of reaction at 65 °C (empty diamond) compared with the kinetics of Pf-PET depolymerization by wild-type LCC (filled diamond). However, adding 0.69 nmol of wild-type LCC after 3 days of reaction at 65 °C restarts the previously stopped assay (filled circle) with an identical specific activity to that determined originally (shown in the table). Each symbol represents a mean ± s.d. (n = 3); n.a., not applicable; n.d., not determined.

Extended Data Fig. 3 Comparative analysis of the LCC variants generated here.

a, Boxplot distribution of the saturation results. We produced 209 variants by semi-purification, and determined their respective percentage of specific activity with respect to the wild-type LCC, measured in the same Pf-PET-depolymerization conditions. Q1 and Q3 correspond to the first and third quartiles of the distribution, respectively. The median value is shown as a red line. b, Comparison of specific activities and melting temperatures of wild-type LCC and the variants used here. Experiments were performed through preparative production of enzymes followed by a Pf-PET-depolymerization assay (see Supplementary Method 1). Means ± s.d. (n = 3) are shown.

Extended Data Fig. 4 Formation of an additional disulfide bridge within the D238C/S283C variant of LCC.

First derivatives were calculated from DSF thermal denaturation curves for wild-type LCC (solid line) and the D238C/S283C variant (dashed line) in the presence of 0 mM, 1 mM or 100 mM dithiothreitol (DTT). First-derivative peaks correspond to protein melting temperatures. Each curve is representative of a triplicate test. The increasing concentrations of DTT reduced disulfide bridges, resulting in proteins with reduced thermostability. Grey highlighting represents populations with 0, 1 or 2 formed disulfide bridges. Low DTT concentrations (blue curves) were not sufficient to completely reduce all disulfide bridges, resulting in a mixed population of proteins with intermediate melting temperatures. RFU, relative fluorescence units.

Extended Data Fig. 5 Comparative analysis of the kinetics of PcW-PET depolymerization in Minibio bioreactors by the LCC variants generated here.

a, Effect of temperature on PcW-PET depolymerization by the LCC variant F243I/D238C/S283C/Y127G at 72 °C or 75 °C, with 1 mgenzyme gPET−1. b, Effect of enzyme concentration on PcW-PET depolymerization by the LCC variant F243I/D238C/S283C/Y127G at 72 °C with 1 mgenzyme gPET−1, 2 mgenzyme gPET−1 or 3 mgenzyme gPET−1. Percentages of PET depolymerization were calculated on the basis of NaOH consumption. c, Comparison of wild-type LCC and variants in assays of PcW-PET depolymerization in Minibio bioreactors. The first two columns show the parameters used during PcW-PET depolymerization (temperature and enzyme concentration). The next four columns show the calculated depolymerization yields (after 24 h), based on either NaOH consumption, EG produced, TAeq. produced, or the weight of residual PET. The last column represents the calculated initial rate of the reaction, based on NaOH consumption.

Extended Data Fig. 6 Kinetics of PET crystallization.

Evolution of the crystallinity level of PET (from coloured PcW-PET) at 65 °C, 70 °C, 72 °C and 75 °C.

Extended Data Fig. 7 Overall structures of the catalytic inactive LCC variant S165A and the catalytic inactive F243I/D238C/S283C/Y127G variant S165A.

a, Left, wild-type LCC (PDB ID 4EB0; green) and the catalytic inactive variant S165A (cyan) are superimposed (with a root mean square deviation (RMSD) of 0.25 Å over 214 Cα atoms). Catalytic residues are represented as magenta rods. The close-up on the right focuses on the catalytic serine (S165) and neighbouring residues. Also shown is an unbiased composite omit map (grey mesh, 2Fo − Fc) contoured at 2.0σ around residues 164–166. The S165A mutation does not structurally affect the folding of the protein around this position. Moreover, the inactivated enzyme is more liable to crystallize and generate better-quality crystals, so we introduced the S165A mutation to our most efficient LCC variant, namely F243I/D238C/S283C/Y127G (ICCG). b, Wild-type LCC (PDB ID 4EB0; green) and the catalytic inactive S165A mutant of the ICCG variant (tan) are superimposed (RMSD = 0.27 Å over 220 Cα atoms). Catalytic (magenta) and mutated (tan) residues are represented as rods. Close-ups show the different mutations and their surrounding residues. Residues from wild-type LCC are represented as thinner rods by comparison with the ICCG variant. Unbiased composite omit maps (grey mesh, 2Fo − Fc) contoured at 1.5σ are shown. None of the introduced Y127G, S165A, F243I or D238C–S283C (engineered disulfide bridge) mutations affected the overall structure of LCC. The asterisked close-up shows that alternative conformations were observed for the cysteine residues, labelled as conformers a and b.

Extended Data Fig. 8 Analysis of 30-ns molecular-dynamics simulations carried out with wild-type LCC (blue) and the ICCG variant (red).

a, Comparison of protein backbone flexibility using average root mean square fluctuations (RMSF) of Cα atoms calculated per residue along molecular-dynamics simulations of enzymes in the apo conformation. The RMSF is linked to the crystallographic B-factor (B) as follows: \(\text{RMSF}=\surd \left(\frac{3B}{8{\pi }^{2}}\right)\). Red arrows, β-strands; black rectangles, α-helices; yellow rectangles, loops in the X-ray crystal structure of the wild-type LCC (PDB ID 4EB0); dashed lines, positions of catalytic residues. b, Monitoring of key catalytic interatomic distances that characterize the catalytic events occurring during molecular-dynamics simulations of enzymes in complex with the model substrate 2-HE(MHET)3. At the right is a representation of the catalytic triad (residues S165 (Ser 165), H242 (His 242) and D210 (Asp 210)) and 2-HE(MHET)3, highlighting three relevant interatomic distances (d1, d2, d3). The three graphs show the distributions of these three distances over the first 30 ns of molecular-dynamics simulations of wild-type and ICCG LCC in complex with 2-HE(MHET)3 (represented as histograms and Gaussian kernel densities), starting from the same initial conformation. Red arrows show changes occurring during the nucleophilic attack of the catalytic serine on the substrate reactive centre; dashed blue lines show hydrogen bonds that assist the catalytic mechanism. The graphs highlight the favoured catalytically productive state adopted by 2-HE(MHET)3 in variant ICCG. Substantial changes are observed for d1 and d2. Whereas ICCG mainly sampled conformations near the catalytically productive state (average d1 is approximately 3.2 Å; average d2 is approximately 2.8 Å), the wild-type LCC showed a pronounced bimodal distribution with the major conformational population centred on higher distance values, indicating less efficient catalysis. Overall, along the first 30 ns of simulations of these enzymes in complex with 2-HE(MHET)3, the average distance separating the substrate cleavage site from the catalytic serine (S165) hydroxyl oxygen was substantially shorter in ICCG than in parental LCC, suggesting that formation of the covalent intermediate during catalysis would be facilitated. c, Occurrence of key hydrogen bonds (HBs) between pairs of catalytic residues. The third and fourth columns show the proportion of snapshots in which an HB interaction is observed between the pairs of catalytic residues S165/H242 and H242/D210 during the first 30 ns of simulations. The higher occurrence of HBs in the ICCG simulation between the S165 hydroxyl oxygen and the catalytic H242 ε nitrogen could assist in the abstraction of the S165 hydroxyl hydrogen by H242, and thus enhance the catalytic performance of this variant.

Extended Data Table 1 Sequence analysis of eight prokaryotic cutinase homologues of known crystal structure
Extended Data Table 2 Data collection and refinement statistics

Supplementary information

Supplementary Information

Supplementary Method 1 | Materials, methods and associated references from the study.

Reporting Summary

Supplementary Information

Supplementary Method 2 | List of nucleotide sequences and expressed amino acid sequences of the genes used in this study. Genetic code specifically used to generate protein variants of LCC is provided.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Tournier, V., Topham, C.M., Gilles, A. et al. An engineered PET depolymerase to break down and recycle plastic bottles. Nature 580, 216–219 (2020).

Download citation


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing