Engineering Enzymes for Environmental Sustainability

Radley, Emily; Davidson, John; Foster, Jake; Obexer, Richard; Bell, Elizabeth L.; Green, Anthony P.

doi:10.1002/anie.202309305

Cited by 5 publications

(2 citation statements)

References 100 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the current time of climate change and increasing resource depletion, enzyme technology has emerged as a more environmentally friendly and potentially low-impact approach to industrial processes traditionally mediated by conventional chemistry (Buller et al, 2023;Hauer, 2020;Radley et al, 2023;Reetz et al, 2024;Sheldon and Woodley, 2018;Wu et al, 2021). Instead of complicated pathways with a plethora of reagents, extreme conditions, and protection groups, enzymes offer a renewable alternative with high selectivity and tunability (Sheldon and Woodley, 2018;Woodley, 2022;Wu et al, 2021).…”

Section: Introductionmentioning

confidence: 99%

Protein Representations: Encoding Biological Information for Machine Learning in Biocatalysis

Harding-Larsen,

Funk,

Madsen

et al. 2024

Preprint

View full text Add to dashboard Cite

Enzymes offer a more environmentally friendly and low-impact solution to conventional chemistry, but they often require additional engineering for industrial settings, an endeavor that is challenging and laborious. To address this issue, the power of machine learning can be harnessed to produce predictive models that facilitate in silico study and engineering of novel enzymatic properties. However, the conversion from the biological domain to the computational realm requires special attention to ensure the training of accurate and precise models. In this review, we examine the critical step of encoding protein information to numeric representations for use in machine learning. We selected the most important approaches for encoding the three distinct biological protein representations — primary sequence, 3D structure, and dynamics — to explore their requirements for employment and inherent biases. Combined representations of proteins and substrates are also introduced as emergent tools in biocatalysis. We propose the division of fixed representations, a collection of rule-based encoding strategies, and learned representations extracted from the latent spaces of large neural networks. To select the most suitable protein representation, we propose two main factors governing this choice. The first one is the model setup, being influenced by the size of the training dataset and the choice of architecture. The second factor is the model objectives, concerning the assayed property, the difference between wild-type models and mutant predictors, and requirements for explainability. This review is aimed at serving as a source of information and guidance for properly representing enzymes in future machine learning models for biocatalysis.

show abstract

Section: Introductionmentioning

confidence: 99%

Protein Representations: Encoding Biological Information for Machine Learning in Biocatalysis

Harding-Larsen,

Funk,

Madsen

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Enzyme engineering campaigns rely on functional screening for the discovery of initial hits and subsequent directed evolution [1] . In light of the increasing demand for biocatalysts in the transition towards a sustainable bioeconomy [2] , sensitive and high-throughput-compatible assays are crucial for overcoming the challenges arising from target functions that are hard to find in screening libraries to sample the proverbial vastness of sequence space [3,4] in a timeefficient manner [5,6] .…”

Section: Introductionmentioning

confidence: 99%

Sub-single-turnover quantification of enzyme catalysis at ultrahigh throughput via a versatile NAD(P)H coupled assay in microdroplets

Penner,

Klein,

Gantz

et al. 2023

Preprint

View full text Add to dashboard Cite

Enzyme engineering and discovery are crucial for a future sustainable bioeconomy, and harvesting new biocatalysts from large libraries through directed evolution or functional metagenomics requires accessible, rapid assays. Ultra-high throughput screening can often require an optical readout, leading to the use of model substrates that may not accurately report on activity for the target reaction and may require bespoke synthesis. In contrast, coupled assays represent a modular ‘plug-and-play’ system, where any pairing of enzyme/substrate may be investigated, if the reaction can produce a common intermediate which links the catalytic reaction to a detection cascade readout. Here we establish a detection cascade, producing a fluorescent readout in response to NAD(P)H via glutathione reductase and a subsequent thiol-mediated uncaging reaction, with a 30 nM detection limit. We demonstrate its utility for the glycosidaseAxyAgu115A (producing monosaccharides from a natural biofuel feedstock) and report a three orders of magnitude improved sensitivity compared to absorbance-based systems, so that less than one catalytic turnover per enzyme molecule expressed from a single cell is detectable. These advantages are brought to bear in plate formats, but also in picoliter emulsion droplets, where enrichments of 950-fold suggest that large libraries can be interrogated against a specific query substrate.

show abstract

Natural diversity screening, assay development, and characterization of nylon-6 enzymatic depolymerization

Bell,

Rosetto,

Ingraham

et al. 2024

Nat Commun

View full text Add to dashboard Cite

Successes in biocatalytic polyester recycling have raised the possibility of deconstructing alternative polymers enzymatically, with polyamide (PA) being a logical target due to the array of amide-cleaving enzymes present in nature. Here, we screen 40 potential natural and engineered nylon-hydrolyzing enzymes (nylonases), using mass spectrometry to quantify eight compounds resulting from enzymatic nylon-6 (PA6) hydrolysis. Comparative time-course reactions incubated at 40-70 °C showcase enzyme-dependent variations in product distributions and extent of PA6 film depolymerization, with significant nylon deconstruction activity appearing rare. The most active nylonase, a NylCK variant we rationally thermostabilized (an N-terminal nucleophile (Ntn) hydrolase, NylCK-TS, Tm = 87.4 °C, 16.4 °C higher than the wild-type), hydrolyzes 0.67 wt% of a PA6 film. Reactions fail to restart after fresh enzyme addition, indicating that substrate-based limitations, such as restricted enzyme access to hydrolysable bonds, prohibit more extensive deconstruction. Overall, this study expands our understanding of nylonase activity distribution, indicates that Ntn hydrolases may have the greatest potential for further development, and identifies key targets for progressing PA6 enzymatic depolymerization, including improving enzyme activity, product selectivity, and enhancing polymer accessibility.

show abstract

Engineering Enzymes for Environmental Sustainability

Cited by 5 publications

References 100 publications

Protein Representations: Encoding Biological Information for Machine Learning in Biocatalysis

Protein Representations: Encoding Biological Information for Machine Learning in Biocatalysis

Sub-single-turnover quantification of enzyme catalysis at ultrahigh throughput via a versatile NAD(P)H coupled assay in microdroplets

Natural diversity screening, assay development, and characterization of nylon-6 enzymatic depolymerization

Contact Info

Product

Resources

About