ArticlePublisher preview available

Machine learning-aided engineering of hydrolases for PET depolymerization

Authors:
Hongyuan Lu
Hongyuan Lu
Hongyuan Lu
  • This person is not on ResearchGate, or hasn't claimed this research yet.
Daniel J. Diaz
Daniel J. Diaz
Daniel J. Diaz
  • This person is not on ResearchGate, or hasn't claimed this research yet.
Natalie Czarnecki at University of Texas at Austin
Natalie Czarnecki
Congzhi Zhu
Congzhi Zhu
Congzhi Zhu
  • This person is not on ResearchGate, or hasn't claimed this research yet.
Show all 13 authorsHide
Read publisher preview
Download citation
To read the full-text of this research, you can request a copy directly from the authors.

Abstract and Figures

Plastic waste poses an ecological challenge1–3 and enzymatic degradation offers one, potentially green and scalable, route for polyesters waste recycling⁴. Poly(ethylene terephthalate) (PET) accounts for 12% of global solid waste⁵, and a circular carbon economy for PET is theoretically attainable through rapid enzymatic depolymerization followed by repolymerization or conversion/valorization into other products6–10. Application of PET hydrolases, however, has been hampered by their lack of robustness to pH and temperature ranges, slow reaction rates and inability to directly use untreated postconsumer plastics¹¹. Here, we use a structure-based, machine learning algorithm to engineer a robust and active PET hydrolase. Our mutant and scaffold combination (FAST-PETase: functional, active, stable and tolerant PETase) contains five mutations compared to wild-type PETase (N233K/R224Q/S121E from prediction and D186H/R280A from scaffold) and shows superior PET-hydrolytic activity relative to both wild-type and engineered alternatives¹² between 30 and 50 °C and a range of pH levels. We demonstrate that untreated, postconsumer-PET from 51 different thermoformed products can all be almost completely degraded by FAST-PETase in 1 week. FAST-PETase can also depolymerize untreated, amorphous portions of a commercial water bottle and an entire thermally pretreated water bottle at 50 ºC. Finally, we demonstrate a closed-loop PET recycling process by using FAST-PETase and resynthesizing PET from the recovered monomers. Collectively, our results demonstrate a viable route for enzymatic plastic recycling at the industrial scale.
Top 10 MutCompute predictions ranked by fold change in the probabilities between the predicted and the wild-type amino acid The top 10 mutations predicted using the wild-type PETase (a) and ThermoPETase (b) as scaffolds are presented. MutCompute is an ensembled model that consist of three individually trained 3-dimensional convolutional neural network (3DCNN) models. Thus, the avg_log_ratio column is the average of the three log ratio values obtained from the three 3DCNN models, rather than being the log ratio of the average probability assigned to the wild type and predicted amino acid across the three 3DCNN models.
Top 10 MutCompute predictions ranked by fold change in the probabilities between the predicted and the wild-type amino acid The top 10 mutations predicted using the wild-type PETase (a) and ThermoPETase (b) as scaffolds are presented. MutCompute is an ensembled model that consist of three individually trained 3-dimensional convolutional neural network (3DCNN) models. Thus, the avg_log_ratio column is the average of the three log ratio values obtained from the three 3DCNN models, rather than being the log ratio of the average probability assigned to the wild type and predicted amino acid across the three 3DCNN models.
… 
Thermostability and protein yield of the PETase variants incorporating the predicted mutations and their respective scaffolds—wild-type PETase (WT), ThermoPETase (Thermo), DuraPETase (Dura) Tm of each enzyme (left) was determined by DSC. The protein yield of each enzyme (right) from P. putida purification experiments was evaluated using a Bradford protein assay. All measurements were conducted in triplicate (n = 3). The bars shown represent the average numbers.
Thermostability and protein yield of the PETase variants incorporating the predicted mutations and their respective scaffolds—wild-type PETase (WT), ThermoPETase (Thermo), DuraPETase (Dura) Tm of each enzyme (left) was determined by DSC. The protein yield of each enzyme (right) from P. putida purification experiments was evaluated using a Bradford protein assay. All measurements were conducted in triplicate (n = 3). The bars shown represent the average numbers.
… 
X-ray crystal structure of FAST-PETase a, Overall crystal structure of FAST-PETase. Catalytic triads (S160, D206, H237) are shown in blue sticks. Mutations originating from or shared with ThermoPETase (S121E, D186H, R280A) are shown in pink sticks, and completely novel mutations predicted by MutCompute are shown in green-yellow sticks. b, c, 2Fo-Fc map (contoured at 1.5 σ) shown as grey mesh superimposed on the stick models of novel mutation sites (b) R224Q, (c) N233K.
X-ray crystal structure of FAST-PETase a, Overall crystal structure of FAST-PETase. Catalytic triads (S160, D206, H237) are shown in blue sticks. Mutations originating from or shared with ThermoPETase (S121E, D186H, R280A) are shown in pink sticks, and completely novel mutations predicted by MutCompute are shown in green-yellow sticks. b, c, 2Fo-Fc map (contoured at 1.5 σ) shown as grey mesh superimposed on the stick models of novel mutation sites (b) R224Q, (c) N233K.
… 
Location of the LM site in the crystal structures of homologous PHEs On wild-type PETase from I. Sakaiensis (white ribbon), catalytic residues are shown as blue sticks. LM site is shown as gray sticks on top of cartoon representation. LM site is zoomed in to show superimposed structures of four different homologous PHEs (right top panel: WT): I. Sakaiensis wild-type PETase (gray sticks, PDB code 5XJH), LCC (yellow sticks, PDB code 4EB0), LCCF243I/D238C/S283C/N246M (ICCM) (green sticks, PDB code 6THT - LCCF243I/D238C/S283C/Y127G (ICCG) variant structure), and S. viridis Cut190 (pink sticks, PDB code 4WFI - Cut190S226P variant structure). Based on FAST-PETase structure, the structure of homologous PHEs with LM is modelled (right bottom panel: LM) with residues shown as blue-colored sticks.
Location of the LM site in the crystal structures of homologous PHEs On wild-type PETase from I. Sakaiensis (white ribbon), catalytic residues are shown as blue sticks. LM site is shown as gray sticks on top of cartoon representation. LM site is zoomed in to show superimposed structures of four different homologous PHEs (right top panel: WT): I. Sakaiensis wild-type PETase (gray sticks, PDB code 5XJH), LCC (yellow sticks, PDB code 4EB0), LCCF243I/D238C/S283C/N246M (ICCM) (green sticks, PDB code 6THT - LCCF243I/D238C/S283C/Y127G (ICCG) variant structure), and S. viridis Cut190 (pink sticks, PDB code 4WFI - Cut190S226P variant structure). Based on FAST-PETase structure, the structure of homologous PHEs with LM is modelled (right bottom panel: LM) with residues shown as blue-colored sticks.
… 
Time-course of mass loss and PET monomers released from hydrolyzing the hole-punched films of six representative pc-PET products with FAST-PETase The six pc-PET products represent PET #2, 6, 8, 25, 29, 32 that were randomly selected from the 51 pc-PET products (Supplementary Table 3 and Supplementary Fig. 4). The pc-PET films hole-punched from these PET products were hydrolysed by serial treatment with FAST-PETase at 50 °C until the films were completely degraded (film disappeared). Enzyme solution (200 nM of FAST-PETase in 100 mM KH2PO4-NaOH (pH 8.0) buffer) was replenished every 22 h. All measurements were conducted in triplicate (n = 3). The squares (mass loss) and circles (PET monomers released) shown represent the individual numbers. The line connects mean values of the timepoints. Source data
+8
Time-course of mass loss and PET monomers released from hydrolyzing the hole-punched films of six representative pc-PET products with FAST-PETase The six pc-PET products represent PET #2, 6, 8, 25, 29, 32 that were randomly selected from the 51 pc-PET products (Supplementary Table 3 and Supplementary Fig. 4). The pc-PET films hole-punched from these PET products were hydrolysed by serial treatment with FAST-PETase at 50 °C until the films were completely degraded (film disappeared). Enzyme solution (200 nM of FAST-PETase in 100 mM KH2PO4-NaOH (pH 8.0) buffer) was replenished every 22 h. All measurements were conducted in triplicate (n = 3). The squares (mass loss) and circles (PET monomers released) shown represent the individual numbers. The line connects mean values of the timepoints. Source data
… 
Figures - available from: Nature
This content is subject to copyright. Terms and conditions apply.
A preview of this full-text is provided by Springer Nature.
Learn more
Content available from Nature
This content is subject to copyright. Terms and conditions apply.
662 | Nature | Vol 604 | 28 April 2022
Article
Machine learning-aided engineering of
hydrolases for PET depolymerization
Hongyuan Lu1, Daniel J. Diaz2, Natalie J. Czarnecki1, Congzhi Zhu1, Wantae Kim1,
Raghav Shroff3,4, Daniel J. Acosta3, Bradley R. Alexander3, Hannah O. Cole1,3, Yan Zhang3,
Nathaniel A. Lynd1, Andrew D. Ellington3 & Hal S. Alper1 ✉
Plastic waste poses an ecological challenge1–3 and enzymatic degradation oers one,
potentially green and scalable, route for polyesters waste recycling4. Poly(ethylene
terephthalate) (PET) accounts for 12% of global solid waste5, and a circular carbon
economy for PET is theoretically attainable through rapid enzymatic
depolymerization followed by repolymerization or conversion/valorization into
other products6–10. Application of PET hydrolases, however, has been hampered by
their lack of robustness to pH and temperature ranges, slow reaction rates and
inability to directly use untreated postconsumer plastics11. Here, we use a
structure-based, machine learning algorithm to engineer a robust and active PET
hydrolase. Our mutant and scaold combination (FAST-PETase: functional, active,
stable and tolerant PETase) contains ve mutations compared to wild-type PETase
(N233K/R224Q/S121E from prediction and D186H/R280A from scaold) and shows
superior PET-hydrolytic activity relative to both wild-type and engineered
alternatives12 between 30 and 50 °C and a range of pH levels. We demonstrate that
untreated, postconsumer-PET from 51 dierent thermoformed products can all be
almost completely degraded by FAST-PETase in 1 week. FAST-PETase can also
depolymerize untreated, amorphous portions of a commercial water bottle and an
entire thermally pretreated water bottle at 50 ºC. Finally, we demonstrate a
closed-loop PET recycling process by using FAST-PETase and resynthesizing PET from
the recovered monomers. Collectively, our results demonstrate a viable route for
enzymatic plastic recycling at the industrial scale.
Enzymatic depolymerization of PET was first reported in 2005 and
has been nascently demonstrated using 19 distinct PET-hydrolysing
enzymes (PHEs) derived from esterases, lipases and cutinases
4,11,13
. How-
ever, most of these enzymes only show appreciable hydrolytic activity
at high reaction temperatures (that is, at or exceeding the PET glass
transition temperature of roughly 70 ºC) and with highly processed
substrates. For example, an engineered leaf-branch compost cutinase
(LCC) can degrade 90% of pretreated postconsumer-PET (pc-PET) in
10 h at 72 ºC and a pH of 8.0 (ref.
12
). Most other PHEs similarly show poor
activity at moderate temperatures
14
and more neutral pH conditions
15
,
greatly restricting insitu/microbially enabled degradation solutions for
PET waste. This limitation is of critical concern as 40% of plastic waste
bypasses collection systems and resides in natural environments
16
.
In addition, converting untreated postconsumer plastic waste at near
ambient temperatures would lower net operating costs.
Although the PHE from the PET-assimilating bacterium Ideonella
sakaiensis14 (PETase) can operate at ambient conditions, it is highly
labile and loses activity even at 37 ºC after 24 h (ref.
17
). Nonetheless,
this mesophilic enzyme has previously seen attempts to enhance ther-
mostability, robustness and function17–23. The most notable engineered
PETase variants—ThermoPETase17 and DuraPETase22—were created
through rational protein engineering and computational redesign
strategies, respectively. Although the thermostability and catalytic
activity of these two mutants were improved
17,22
under certain con-
ditions, they nonetheless had overall lower PET-hydrolytic activity
at mild temperatures.
We posited that highly focused protein engineering approaches can-
not consider the evolutionary trade-off between overall stability and
activity, and that a neutral, structure-based, deep learning neural net-
work could generally improve enzyme function. To this end, we used a
three-dimensional (3D) self-supervised, convolutional neural network
(CNN), MutCompute
24
(https://mutcompute.com; Supplementary Fig.1
and Supplementary Discussion) to identify stabilizing mutations. This
algorithm learns the local chemical microenvironments of amino acids
on the basis of training over 19,000 sequence-balanced protein struc-
tures from the Protein Data Bank (PDB) and can readily predict positions
within a protein in which wild-type (WT) amino acids are not optimized
for their local environments. We used MutCompute to obtain a discrete
probability distribution for the structural fit of all 20 canonical amino
acids at every position in both WT PETase and ThermoPETase (crystal
https://doi.org/10.1038/s41586-022-04599-z
Received: 10 October 2021
Accepted: 28 February 2022
Published online: 27 April 2022
Check for updates
1McKetta Department of Chemical Engineering, The University of Texas at Austin, Austin, TX, USA. 2Department of Chemistry, The University of Texas at Austin, Austin, TX, USA. 3Department of
Molecular Biosciences, The University of Texas at Austin, Austin, TX, USA. 4DEVCOM ARL-South, Austin, TX, USA. ✉e-mail: halper@che.utexas.edu
Content courtesy of Springer Nature, terms of use apply. Rights reserved
... Models predicting protein fitness evolution can be mainly divided into sequence-based prediction models and structurebased prediction models 15 . Most ML-based protein fitness prediction models are supervised, using numerical representations of protein sequences as features and the corresponding observed fitness as labels, such as eUniRep, ECNet, and Mutcompute [16][17][18][19] . There are also unsupervised inferring probabilistic models using multiple sequence alignments like Potts-Models. ...
... is a monomeric enzyme that specifically converts glutamine residues of proteins or peptides into glutamate and releases ammonia 22 . Initially, the hot fluctuation regions α1 (amino acids [8][9][10][11][12][13][14][15][16][17][18][19], loop2 (amino acids 20-41), α2 (amino acids [42][43][44][45][46][47][48][49][50][51][52][53][54][55], and loop6 (amino acids 102-113) were selected based on the fluctuations in isothermal compressibility (β T ) (Fig. 2a). The results of molecular docking showed that S35, S72, and S108 formed hydrogen bonds with the ligand. ...
Tailoring industrial enzymes for thermostability and activity evolution by the machine learning-based iCASE strategy
Article
Full-text available
  • Jan 2025
The pursuit of obtaining enzymes with high activity and stability remains a grail in enzyme evolution due to the stability-activity trade-off. Here, we develop an isothermal compressibility-assisted dynamic squeezing index perturbation engineering (iCASE) strategy to construct hierarchical modular networks for enzymes of varying complexity. Molecular mechanism analysis elucidates that the peak of adaptive evolution is reached through a structural response mechanism among variants. Furthermore, this dynamic response predictive model using structure-based supervised machine learning is established to predict enzyme function and fitness, demonstrating robust performance across different datasets and reliable prediction for epistasis. The universality of the iCASE strategy is validated by four sorts of enzymes with different structures and catalytic types. This machine learning-based iCASE strategy provides guidance for future research on the fitness evolution of enzymes.
View
... However, rational mutagenesis has thus far not yielded a superior RuBisCO variant capable of simultaneously enhancing carboxylation while minimizing oxygenation. While recent progress in machine learning, incorporating both sequence and structural data, has enhanced the design of libraries for more thorough exploration of the sequence-function landscape (Bryant et al., 2021;Lu et al., 2022), these applications of such approaches to RuBisCO are in their infancy (Iqbal et al., , 2023. ...
... AI can also help in rational design by identifying key residues and predicting the impact of combinations of mutations. By incorporating structural and biochemical data, AI models can suggest targeted modifications that might enhance RuBisCO's carboxylation rate, CO 2 affinity, or thermal stability without the detrimental trade-offs seen in purely experimental approaches (Bryant et al., 2021;Lu et al., 2022). ...
Engineering carbon assimilation in plants
Article
Full-text available
Carbon assimilation is a crucial part of the photosynthetic process, wherein inorganic carbon, typically in the form of CO2, is converted into organic compounds by living organisms, including plants, algae, and a subset of bacteria. Although several carbon fixation pathways have been elucidated, the Calvin–Benson–Bassham (CBB) cycle remains fundamental to carbon metabolism, playing a pivotal role in the biosynthesis of starch and sucrose in plants, algae, and cyanobacteria. However, Ribulose‐1,5‐bisphosphate carboxylase/oxygenase (RuBisCO), the key carboxylase enzyme of the CBB cycle, exhibits low kinetic efficiency, low substrate specificity, and high temperature sensitivity, all of which have the potential to limit flux through this pathway. Consequently, RuBisCO needs to be present at very high concentrations, which is one of the factors contributing to its status as the most prevalent protein on Earth. Numerous attempts have been made to optimize the catalytic efficiency of RuBisCO and thereby promote plant growth. Furthermore, the limitations of this process highlight the potential benefits of engineering or discovering more efficient carbon fixation mechanisms, either by improving RuBisCO itself or by introducing alternative pathways. Here, we review advances in artificial carbon assimilation engineering, including the integration of synthetic biology, genetic engineering, metabolic pathway optimization, and artificial intelligence in order to create plants capable of performing more efficient photosynthesis. We additionally provide a perspective of current challenges and potential solutions alongside a personal opinion of the most promising future directions of this emerging field.
View
... In recent years, key advances such as directed evolution 2 , enzyme immobilization 3 , multi-enzymatic cascades 4 , and the integration of biocatalysis with chemical catalysis 5 have significantly broadened the scope and efficacy of enzymatic processes, enabling them to meet the growing demands of green chemistry and industrial biotechnology. In particular, enzyme catalysis has demonstrated remarkable potential in environmental, energy, and pharmaceutical applications, such as the degradation of polyethylene terephthalate (PET) plastic [6][7][8] , artificial CO2 fixation 9 , and the synthesis of bioactive compounds like the anti-cancer drug vinblastine 10 and the diabetes medication sitagliptin 11 . Enzyme catalysis stands as a genuinely green and sustainable technology, offering not only significant environmental benefits but also cost-effectiveness, driven by ongoing advancements in biotechnology. ...
A De Novo Design Strategy to Convert FAcD from Dimer to Active Monomer
Preprint
Full-text available
  • Jan 2025
Enzymes largely exist in various oligomeric states, but monomeric enzymes are more conducive to industrial applications. Converting an oligomeric enzyme into an active monomer is a significant challenge. In this study, we present a de novo design strategy to convert fluoroacetate dehalogenase (FAcD) from its native dimeric form to an active monomer. Using the AI-based method ProteinMPNN, we identified critical protein-protein interaction (PPI) sites at the dimer interface. ArDCA, another AI tool, was employed to pinpoint catalytic hotspots. Six mutants, Mu1-Mu6, were designed. Molecular dynamics (MD) simulations, coupled with mass spectrometry, confirmed that these mutants form stable monomers. The pre-reaction state (PRS) model predicted that three of these mutants exhibited catalytic activity. In particular, Mu5 with 11 mutations from the wild-type, was predicted to have high catalytic activity, and was subsequently confirmed by kinetics experiment, with a kcat of 672.2 min-1 and a T5030 > 100 ℃ , comparable to the wild-type enzyme (kcat = 676.3 min-1, T5030 = 84 ℃). Notably, the Y149M mutation increased catalytic activity nearly forty-fold, demonstrating the effectiveness of our design strategy.
View
... Polyethylene terephthalate (PET) plastics chemically comprise two main monomer units of ethylene glycol (EG) and terephthalate, which hold good stability and light weight for their wide utilization in human life [1][2][3][4][5][6]. However, waste PET plastics are generally transported in landfills, which inevitably causes severe waste pollution and dangers to human health [7][8][9][10][11]. ...
Trimetallic mesoporous PdCuRu for efficient formate electrosynthesis from polyethylene terephthalate recycling
Article
Full-text available
  • Jan 2025
Electrochemical recycling of waste polyethylene terephthalate (PET) plastics represents a green and sustainable route to produce valuable fuels and address serious environmental issues. Despite some encouraged achievements, their progress is still slower because of the lack of highly active and selective electrocatalysts. Here, we reported an efficient PET recycling route with trimetallic PdCuRu mesoporous nanospheres (MSs) as a high-performance electrocatalyst that achieved selective electro-synthesis of valuable formate from PET hydrolysate (PETH) electrooxidation. Penetrated metal mesopores and trimetallic compositions of PdCuRu MSs synergistically optimized electronic structures of Pd sites and changed chemisorption properties of intermediates, which disclosed a high formate Faradic efficiency of 97% and impressive yield rate of 0.49 mmol h−1 cm−2 in an alkaline solution. Meanwhile, this electrocatalyst held good electrocatalytic stability, retaining high formate Faradic efficiency and yield rate even being performed for 10 cycles, which highlighted its potential applications for practical PET recycling. Our study provides a powerful strategy for designing high performance electrocatalysts toward selective electro synthesis of valuable fuels from various waste plastics.
View
View
View
View
Machine Learning Approaches in Polymer Science: Progress and Fundamental for a New Paradigm
Article
Full-text available
  • Jan 2025
Machine learning (ML), material genome, and big data approaches are highly overlapped in their strategies, algorithms, and models. They can target various definitions, distributions, and correlations of concerned physical parameters in given polymer systems, and have expanding applications as a new paradigm indispensable to conventional ones. Their inherent advantages in building quantitative multivariate correlations have largely enhanced the capability of scientific understanding and discoveries, thus facilitating mechanism exploration, target prediction, high‐throughput screening, optimization, and rational and inverse designs. This article summarizes representative progress in the recent two decades focusing on the design, preparation, application, and sustainable development of polymer materials based on the exploration of key physical parameters in the composition–process–structure–property–performance relationship. The integration of both data‐driven and physical insights through ML approaches to deepen fundamental understanding and discover novel polymer materials is categorically presented. Despite the construction and application of robust ML models, strategies and algorithms to deal with variant tasks in polymer science are still in rapid growth. The challenges and prospects are then presented. We believe that the innovation in polymer materials will thrive along the development of ML approaches, from efficient design to sustainable applications.
View
Upcycling of polyethylene terephthalate to 1,4-cyclohexanedimethanol in water
Article
  • Jan 2025
The recycling of waste polyethylene terephthalate (PET), an extensively used polymer as plastic and fiber materials, has become an urgent topic due to the rapid generation and accumulation of plastic waste into the environment. Current chemical recycling approaches to depolymerize and convert PET into 1,4-cyclohexanedimethanol (CHDM), an important chemical in the production of advanced polyesters, predominantly use organic solvents, such as alcohols or dioxane. Herein, we developed an environmentally friendly and economically viable pathway for converting PET to CHDM using water as the solvent. In this process, PET is sequentially hydrolyzed in water, hydrogenated to 1,4-cyclohexane dicarboxylic acid (CHDA), and then reduced to CHDM. The PET conversion rate is self-enhanced in aqueous medium via acid-catalyzed hydrolysis by the generated CHDA intermediate. A Pd-based catalyst (e.g., Pd/C) selectively hydrogenates the arene ring of the PET monomer, terephthalic acid intermediate, and a Ru-Sn/C bimetallic catalyst reduces its COOH to CH2OH group. At optimized reaction conditions, a complete conversion of PET and 80.1% yield of CHDM are obtained. Alternatively, this approach can also efficiently produce CHDA from PET, with a yield up to 86.4%.
View
View
The global threat from plastic pollution
Article
Full-text available
  • Jul 2021
Plastic pollution accumulating in an area of the environment is considered “poorly reversible” if natural mineralization processes occurring there are slow and engineered remediation solutions are improbable. Should negative outcomes in these areas arise as a consequence of plastic pollution, they will be practically irreversible. Potential impacts from poorly reversible plastic pollution include changes to carbon and nutrient cycles; habitat changes within soils, sediments, and aquatic ecosystems; co-occurring biological impacts on endangered or keystone species; ecotoxicity; and related societal impacts. The rational response to the global threat posed by accumulating and poorly reversible plastic pollution is to rapidly reduce plastic emissions through reductions in consumption of virgin plastic materials, along with internationally coordinated strategies for waste management.
View
General features to enhance enzymatic activity of poly(ethylene terephthalate) hydrolysis
Article
Full-text available
  • May 2021
Poly(ethylene terephthalate) (PET) is the most abundant polyester plastic and a major contributor to plastic pollution. IsPETase, from the PET-assimilating bacterium Ideonella sakaiensis, is a unique PET-hydrolytic enzyme that shares high sequence identity to canonical cutinases, but shows substrate preference towards PET and exhibits higher PET-hydrolytic activity at ambient temperature. Structural analyses suggest that IsPETase harbours a substrate-binding residue, W185, with a wobbling conformation and a highly flexible W185-locating β6-β7 loop. Here, we show that these features result from the presence of S214 and I218 in IsPETase, whose equivalents are strictly His and Phe, respectively, in all other homologous enzymes. We found that mutating His/Phe residues to Ser/Ile could enhance the PET-hydrolytic activity of several IsPETase-like enzymes. In conclusion, the Ser/Ile mutations should provide an important strategy to improve the activity of potential PET-hydrolytic enzymes with properties that may be useful for various applications.
View
Chemical and biological catalysis for plastics recycling and upcycling
Article
  • Jul 2021
Plastics pollution is causing an environmental crisis, prompting the development of new approaches for recycling, and upcycling. Here, we review challenges and opportunities in chemical and biological catalysis for plastics deconstruction, recycling, and upcycling. We stress the need for rigorous characterization and use of widely available substrates, such that catalyst performance can be compared across studies. Where appropriate, we draw parallels between catalysis on biomass and plastics, as both substrates are low-value, solid, recalcitrant polymers. Innovations in catalyst design and reaction engineering are needed to overcome kinetic and thermodynamic limitations of plastics deconstruction. Either chemical and biological catalysts will need to act interfacially, where catalysts function at a solid surface, or polymers will need to be solubilized or processed to smaller intermediates to facilitate improved catalyst–substrate interaction. Overall, developing catalyst-driven technologies for plastics deconstruction and upcycling is critical to incentivize improved plastics reclamation and reduce the severe global burden of plastic waste. Plastics are invaluable materials for modern society, although they result in the generation of large amounts of litter at the end of their life cycle. This Review explores the challenges and opportunities associated with the catalytic transformation of waste plastics, looking at both chemical and biological approaches to transforming such spent materials into a resource.
View
View
Achieving a circular bioeconomy for plastics
Article
  • Jul 2021
Designing plastics for assembly and disassembly is essential to closing the resource loop
View
Plastic ingestion as an evolutionary trap: Toward a holistic understanding
Article
  • Jul 2021
  • SCIENCE
Human activities are changing our environment. Along with climate change and a widespread loss of biodiversity, plastic pollution now plays a predominant role in altering ecosystems globally. Here, we review the occurrence of plastic ingestion by wildlife through evolutionary and ecological lenses and address the fundamental question of why living organisms ingest plastic. We unify evolutionary, ecological, and cognitive approaches under the evolutionary trap theory and identify three main factors that may drive plastic ingestion: (i) the availability of plastics in the environment, (ii) an individual's acceptance threshold, and (iii) the overlap of cues given by natural foods and plastics.
View
View
Computational Redesign of a PETase for Plastic Biodegradation under Ambient Condition by the GRAPE Strategy
Article
  • Jan 2021
Nature has provided a fantastic array of enzymes that are responsible for essential biochemical functions but not usually suitable for technological applications. Not content with the natural repertoire, protein engineering holds promise to extend the applications of improved enzymes with tailored properties. However, engineering of robust proteins remains a difficult task since the positive mutation library may not cooperate to reach the target function in most cases owing to the ubiquity of epistatic effects. The main demand lies in identifying an efficient path of accumulated mutations. Herein, we devised a computational strategy (greedy accumulated strategy for protein engineering, GRAPE) to improve the robustness of a PETase from Ideonella sakaiensis. A systematic clustering analysis combined with greedy accumulation of beneficial mutations in a computationally derived library enabled the redesign of a variant, DuraPETase, which exhibits an apparent melting temperature that is drastically elevated by 31 °C and a strikingly enhanced degradation toward semicrystalline poly(ethylene terephthalate) (PET) films (30%) at mild temperatures (over 300-fold). Complete biodegradation of 2 g/L microplastics to water-soluble products under mild conditions is also achieved, opening up opportunities to steer the biological degradation of uncollectable PET waste and further conversion of the resulting monomers to high-value molecules. The crystal structure revealed the individual mutation match with the design model. Concurrently, synergistic effects are captured, while epistatic interactions are alleviated during the accumulation process. We anticipate that our design strategy will provide a broadly applicable strategy for global optimization of enzyme performance.
View
Discovery of Novel Gain-of-Function Mutations Guided by Structure-Based Deep Learning
Article
  • Nov 2020
Despite the promise of deep learning accelerated protein engineering, examples of such improved proteins are scarce. Here we report that a 3D convolutional neural network trained to associate amino acids with neighboring chemical microenvironments can guide identification of novel gain-of-function mutations that are not predicted by energetics-based approaches. Amalgamation of these mutations improved protein function in vivo across three diverse proteins by at least 5-fold. Furthermore, this model provides a means to interrogate the chemical space within protein microenvironments and identify specific chemical interactions that contribute to the gain-of-function phenotypes resulting from individual mutations.
View
The highly crystalline PET found in plastic water bottles does not support the growth of the PETase‐producing bacterium Ideonella sakaiensis
Article
  • Aug 2020
Ideonella sakaiensis produces an enzyme, PETase, that is capable of hydrolyzing polyethylene terephthalate (PET) plastic. We demonstrate that although I. sakaiensis can grow on amorphous plastic, it does not grow on highly crystalline plastic under otherwise identical conditions. Both amorphous film and amorphous plastic obtained from commercial food containers support growth of the bacteria, whereas highly crystalline film and the highly crystalline body of a plastic water bottle do not support growth. Highly crystalline PET can be melted and rapidly cooled to make amorphous plastic which then supports bacterial growth, whereas the same plastic can be melted and slowly cooled to make crystalline plastic which does not support growth. We further subject a plastic water bottle to a top‐to‐bottom analysis, finding that only amorphous sections are degraded, namely the finish (threading), the topmost portion of the shoulder which connects to the finish, and the area immediately surrounding the center of the base. Finally, we use these results to estimate that the percentage of non‐degradable plastic in plastic water bottles ranges from 52 to 82% (depending on size), demonstrating that most of the plastic found in PET water bottles will not be degraded by I. sakaiensis. This article is protected by copyright. All rights reserved.
View