43HIGHLIGHTS 2020

was too low for industrial use, was optimised in order to create a more efficient PETase. Structural biology and molecular modelling techniques were used to understand key molecular determinants and to target modifications to improve two essential characteristics of LCC; PET depolymerisation activity and enzyme thermostability. The reaction of PET depolymerisation must be carried out as close as possible to the glass transition temperature of PET (70°C), offering the enzyme a facilitated access to the polymer. Protein crystals were prepared at the IPBS crystallisation facility (PICT Platform, Toulouse), and atomic-resolution crystal structures were obtained at beamline ID30B and at ALBA synchrotron, Spain. These were used for molecular modelling and dynamics studies that led to the design of an enzyme with considerably improved catalytic performance and stability (Figure 28). This optimised PETase (3 mg enzyme/g PET) achieves a rate of PET depolymerisation of 90% in just ten hours, with a

productivity of 16.7 g/L/h terephthalic acid from a suspension of 200 g/kg post-consumer PET waste. With the help of the CRITT-Bioindustries of Toulouse, Leitat and Pivert, the monomers were purified and polymerised into bottle-grade PET that was ultimately blown into bottles. These bottles showed properties that were identical to those produced by polymerisation of petrochemical monomers, thus demonstrating the circularity of the process.

The rational engineering of an optimised PETase, made possible by molecular modelling and structural biology combined with process development, has resulted in a productivity 100-fold superior to the previously published bioprocesses for PET depolymerisation. This constitutes a world first, paving the way for the deployment of a circular economy technology applicable to all PET waste and thus representing a breakthrough solution to the environmental and industrial challenges of plastic recycling.

Fig. 28: a) 2-HE (MHET)3 molecule (coloured rods) docked in the active site of wild-type LCC. The putative active site can be subdivided into three subsites (-2, -1, +1), each in contact with the numbered MHET units relative to the point of ester bond cleavage (red triangle).

Residues from the first shell of interaction with the substrate are shown as grey bars and the catalytic residues as magenta bars. b) Superposition of the structures of wild-type LCC (4EB0) and LCC variant F243I/D238C/S283C/Y127G/S165A (6THT).

The optimised variant (beige) is compared to the wild-type enzyme (green). Close-ups show mutations introduced and the surrounding residues. None of the mutations affect the overall folding of the enzyme. Composite omit electron

density maps are represented by a grey grid and contoured at 1.5σ.

An engineered PET depolymerase to break down and recycle plastic bottles, V. Tournier (a), C.M. Topham (a), A. Gilles (a), B. David (a), C. Folgoas (a), E. Moya-Leclair (a), E. Kamionka (a), M.-L. Desrousseaux (a), H. Texier (a),

S. Gavalda (a), M. Cot (c), E. Guémard (b), M. Dalibey (b), J. Nomme (a), G. Cioci (a), S. Barbe (a), M. Château (b), I. André (a), S. Duquesne (a) and A. Marty (a,b), Nature 580, 216-219 (2020); https://doi. org/10.1038/s41586-020-2149-4.

(a) TBI - Toulouse Biotechnology Institute, Bio & Chemical Engineering, Toulouse (France) (b) Carbios, Saint-Beauzire (France) (c) CRITT Bio-Industries, Toulouse (France)

[1] PlasticsEurope. Plastics - the Facts 2019 (2019). [2] R. Wei & W. Zimmermann, Microb. Biotechnol. 10, 1308-1322 (2017). [3] S. Yoshida et al., Science 351, 1196-1199 (2016). [4] B. Liu et al., ChemBioChem 19, 1471-1475 (2018). [5] H.P. Austin et al., PNAS USA 115, 4350-4357 (2018). [6] S. Sulaiman et al., Appl. Environ. Microbiol. 78, 1556-1562 (2012).

PRINCIPAL PUBLICATION AND AUTHORS

REFERENCES