Understanding activity-stability tradeoffs in biocatalysts by enzyme proximity sequencing

Abstract

Understanding the complex relationships between enzyme sequence, folding stability and catalytic activity is crucial for applications in industry and biomedicine. However, current enzyme assay technologies are limited by an inability to simultaneously resolve both stability and activity phenotypes and to couple these to gene sequences at large scale. Here we present the development of enzyme proximity sequencing, a deep mutational scanning method that leverages peroxidase-mediated radical labeling with single cell fidelity to dissect the effects of thousands of mutations on stability and catalytic activity of oxidoreductase enzymes in a single experiment. We use enzyme proximity sequencing to analyze how 6399 missense mutations influence folding stability and catalytic activity in a D-amino acid oxidase from Rhodotorula gracilis. The resulting datasets demonstrate activity-based constraints that limit folding stability during natural evolution, and identify hotspots distant from the active site as candidates for mutations that improve catalytic activity without sacrificing stability. Enzyme proximity sequencing can be extended to other enzyme classes and provides valuable insights into biophysical principles governing enzyme structure and function.

Fig. 1. Schematic depicting the EP-Seq workflow.

(Top) A pooled library of enzyme variants is displayed on yeast. (Left) The cell population is sorted into bins based on the expression level of the displayed enzyme. (Right) The pooled variant library is assayed for DAOx activity using a cascade peroxidase-mediated proximity labeling reaction with single cell fidelity and sorted into bins. The genetic composition of cells in the sorted bins is quantified via high-throughput sequencing and the distribution of each variant along the expression and activity axes is converted into a fitness score. Joint analysis of the two independent datasets provides insights into the effects of mutations on folding stability and activity of the enzyme.

Fig. 2. Deep mutational scanning of DAOx expression and catalytic activity by EP-Seq.

A Sorting gates for analysis of display levels. B Linear regression (Pearson r, two-tailed) between expression scores calculated from two biological replicates. C Sorting gates for catalytic activity screening. D Linear regression (Pearson r, two-tailed) between activity scores calculated from two biological replicates. E Linear regression (Pearson r, two-tailed) between variant surface display level measured in monogenic yeast culture vs. DMS expression fitness analyzed by EP-Seq for 12 DAOx single mutant variants. F Distribution of expression fitness effects measured by EP-Seq. Dashed lines represent the range of fitness score for synonymous variants. G Linear regression (Pearson r, two-tailed) between variant activity level measured in monogenic yeast culture via peroxidase assay (Amplex Red) vs. DMS activity fitness analyzed by EP-Seq for 12 DAOx single mutant variants. H Distribution of activity fitness effects measured by EP-Seq. Dashed lines represent the range of fitness score for synonymous variants. I Expression fitness scores for each variant represented as a heatmap. J Number of variants analyzed per position in the expression dataset, and secondary structure classification per position (PDB: 1C0P). K Activity fitness scores for each variant obtained by EP-Seq represented as a heatmap. L Number of variants analyzed per position in the activity dataset and secondary structure classification per position (PDB: 1C0P). Links to interactive and color blind accessible heatmaps can be found in the data availability statement section of the manuscript.

Fig. 3. EP-Seq reveals biophysical properties that differentially influence expression and activity.

A Relative impact of mutant residue identity on DAOx expression. B Relative impact of mutant residue identity on the DAOx enzymatic activity. C Plot of Pearson correlation coefficients (r) calculated for expression (x-axis) or activity (y-axis) fitness scores with respect to several biophysical properties of mutated sites in DAOx. D DAOx structure colored by expression or activity fitness scores: (i) Residues surrounding FAD cofactor colored by expression score; (ii) Residues contacting the substrate D-alanine colored by expression score; (iii) Expression (left) or activity (right) scores used to color the surface of DAOx monomers. The monomers are shown separated from each other to improve visibility of the interface; (iv) Residues surrounding FAD cofactor colored by activity score; (v) Residues contacting the substrate D-alanine colored by activity score.

Fig. 4. EP-Seq and sequence conservation analysis reveal functional sites of DAOx.

A The distance to the catalytic site for positions that are evolutionarily conserved (conservation score, CONS = 100) were plotted as a function of EP-Seq expression scores (average per position) and analyzed through Pearson linear correlation function with a two-sided statistical test. The nine highly conserved residues associated with the most positive expression scores (light gray right corner) are visualized in (B). B Structural details of DAOx residues with CONS = 100 and expression score > 0. H-bonds are shown in green. The FAD cofactor and D-alanine carbon backbones are colored in yellow and green, respectively. DAOx residue carbon backbones are colored in light blue with colored elements: Hydrogen (white), Nitrogen (dark blue), Oxygen (red), Phosphorus (orange). C Expression and activity scores for each of the conserved positions linked to positive expression scores.

Fig. 5. EP-Seq predicts activity enhancing mutations in DAOx.

A Average expression and activity fitness scores per position in the DAOx enzyme. The coloring indicates relative distance to the active site. B Normalized activity values are depicted on the 3D structure of a DAOx monomer (right). Areas of the enzyme where mutations on average led to increased activity are represented in green, while regions where mutations on average had a negative impact on activity are shown in red. The most frequent 6 positions among the top 1000 mutants with highest normalized activity scores are indicated.