Analysis of electrostatic coupling throughout the laboratory evolution of a designed retroaldolase

Coulther, Timothy A.; Zeymer, Cathleen; Hilvert, Donald; Ondrechen, Mary Jo

doi:10.1002/pro.4099

Cited by 4 publications

(3 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Distal residues that interact with catalytic residues or as part of networks connecting to the catalytic site are essential for fine tuning and optimization of enzyme activity (Dudev et al, 2003; Somarowthu, Brodkin, et al 2011; Parasuram et al, 2018; Brodkin et al, 2015; Tiwari et al, 2014; Coulther, Ko, & Ondrechen, 2021; Coulther et al 2021; Ngu et al, 2020). The MAHOMES II burial of ionizable residues feature subgroup differentiates enzyme sites based on buried second shell polar and charged residues, which would be a direct result of crucial coupled interactions that enhance enzyme activity.…”

Section: Discussionmentioning

confidence: 99%

MAHOMES II: A webserver for predicting if a metal binding site is enzymatic

et al. 2023

View full text Add to dashboard Cite

Recent advances have enabled high‐quality computationally generated structures for proteins with no solved crystal structures. However, protein function data remains largely limited to experimental methods and homology mapping. Since structure determines function, it is natural that methods capable of using computationally generated structures for functional annotations need to be advanced. Our laboratory recently developed a method to distinguish between metalloenzyme and nonenzyme sites. Here we report improvements to this method by upgrading our physicochemical features to alleviate the need for structures with sub‐angstrom precision and using machine learning to reduce training data labeling error. Our improved classifier identifies protein bound metal sites as enzymatic or nonenzymatic with 94% precision and 92% recall. We demonstrate that both adjustments increased predictive performance and reliability on sites with sub‐angstrom variations. We constructed a set of predicted metalloprotein structures with no solved crystal structures and no detectable homology to our training data. Our model had an accuracy of 90%–97.5% depending on the quality of the predicted structures included in our test. Finally, we found the physicochemical trends that drove this model's successful performance were local protein density, second shell ionizable residue burial, and the pocket's accessibility to the site. We anticipate that our model's ability to correctly identify catalytic metal sites could enable identification of new enzymatic mechanisms and improve de novo metalloenzyme design success rates.

show abstract

Section: Discussionmentioning

confidence: 99%

MAHOMES II: A webserver for predicting if a metal binding site is enzymatic

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Distal residues that interact with catalytic residues or as part of networks connecting to the catalytic site are essential for fine tuning and optimization of enzyme activity(Dudev et al 2003; Somarowthu, Brodkin, et al 2011; Parasuram et al 2018; Brodkin et al 2015; Tiwari et al 2014; Coulther, Ko, and Ondrechen 2021; Coulther et al 2021; Ngu et al 2020). The MAHOMES II burial of ionizable residues feature subgroup differentiates enzyme sites based on buried second shell polar and charged residues, which would be a direct result of crucial coupled interactions that enhance enzyme activity.…”

Section: Discussionmentioning

confidence: 99%

MAHOMES II: A webserver for predicting if a metal binding site is enzymatic

Feehan

Copeland

Franklin

et al. 2023

Preprint

View full text Add to dashboard Cite

Recent advances have enabled high-quality computationally generated structures for proteins with no solved crystal structures. However, protein function data remains largely limited to experimental methods and homology mapping. Since structure determines function, it is natural that methods capable of using computationally generated structures for functional annotations need to be advanced. Our laboratory recently developed a method to distinguish between metalloenzyme and non-enzyme sites. Here we report improvements to this method by upgrading our physicochemical features to alleviate the need for structures with sub-angstrom precision and using machine learning to reduce training data labeling error. Our improved classifier identifies protein bound metal sites as enzymatic or non-enzymatic with 94% precision and 92% recall. We demonstrate that both adjustments increased predictive performance and reliability on sites with sub-angstrom variations. We constructed a set of predicted metalloprotein structures with no solved crystal structures and no detectable homology to our training data. Our model had an accuracy of 90 - 97.5% depending on the quality of the predicted structures included in our test. Finally, we found the physicochemical trends that drove this model's successful performance were local protein density, second shell ionizable residue burial, and the pocket's accessibility to the site. We anticipate that our model's ability to correctly identify catalytic metal sites could enable identification of new enzymatic mechanisms and improve de novo metalloenzyme design success rates.

show abstract

“…Then again, MD simulations were performed using the YASARA software package. [56][57][58][59][60][61] The AMBER14N force field was employed with periodic boundary conditions. [62] Nonbonded interactions were cut off at 1.05 nm.…”

Section: Methodsmentioning

confidence: 99%

Understanding Self‐Assembly of Silica‐Precipitating Peptides to Control Silica Particle Morphology

et al. 2023

View full text Add to dashboard Cite

The most advanced materials are those found in nature. These evolutionary optimized substances provide highest efficiencies, e.g., in harvesting solar energy or providing extreme stability, and are intrinsically biocompatible. However, the mimicry of biological materials is limited to a few successful applications since there is still a lack of the tools to recreate natural materials. Herein, such means are provided based on a peptide library derived from the silaffin protein R5 that enables rational biomimetic materials design. It is now evident that biomaterials do not form via mechanisms observed in vitro. Instead, the material's function and morphology are predetermined by precursors that self‐assemble in solution, often from a combination of protein and salts. These assemblies act as templates for biomaterials. The RRIL peptides used here are a small part of the silica‐precipitation machinery in diatoms. By connecting RRIL motifs via varying central bi‐ or trifunctional residues, a library of stereoisomers is generated, which allows characterization of different template structures in the presence of phosphate ions by combining residue‐resolved real‐time NMR spectroscopy and molecular dynamics (MD) simulations. Understanding these templates in atomistic detail, the morphology of silica particles is controlled via manipulation of the template precursors.

show abstract

Analysis of electrostatic coupling throughout the laboratory evolution of a designed retroaldolase

Cited by 4 publications

References 42 publications

MAHOMES II: A webserver for predicting if a metal binding site is enzymatic

MAHOMES II: A webserver for predicting if a metal binding site is enzymatic

MAHOMES II: A webserver for predicting if a metal binding site is enzymatic

Understanding Self‐Assembly of Silica‐Precipitating Peptides to Control Silica Particle Morphology

Contact Info

Product

Resources

About