Enzymatic carbon-fluorine bond cleavage by human gut microbes

Abstract

Fluorinated compounds are used for agrochemical, pharmaceutical, and numerous industrial applications, resulting in global contamination. In many molecules, fluorine is incorporated to enhance the half-life and improve bioavailability. Fluorinated compounds enter the human body through food, water, and xenobiotics including pharmaceuticals, exposing gut microbes to these substances. The human gut microbiota is known for its xenobiotic biotransformation capabilities, but it was not previously known whether gut microbial enzymes could break carbon-fluorine bonds, potentially altering the toxicity of these compounds. Here, through the development of a rapid, miniaturized fluoride detection assay for whole-cell screening, we identified active gut microbial defluorinases. We biochemically characterized enzymes from diverse human gut microbial classes including Clostridia, Bacilli, and Coriobacteriia, with the capacity to hydrolyze (di)fluorinated organic acids and a fluorinated amino acid. Whole-protein alanine scanning, molecular dynamics simulations, and chimeric protein design enabled the identification of a disordered C-terminal protein segment involved in defluorination activity. Domain swapping exclusively of the C-terminus conferred defluorination activity to a nondefluorinating dehalogenase. To advance our understanding of the structural and sequence differences between defluorinating and nondefluorinating dehalogenases, we trained machine learning models which identified protein termini as important features. Models trained on 41-amino acid segments from protein C termini alone predicted defluorination activity with 83% accuracy (compared to 95% accuracy based on full-length protein features). This work is relevant for therapeutic interventions and environmental and human health by uncovering specificity-determining signatures of fluorine biochemistry from the gut microbiome.

Keywords: defluorination; haloacid dehalogenases; human gut microbiome; molecular dynamics; protein engineering.

Fig. 1.

Haloacid dehalogenases are widespread across human gut bacterial phyla and exhibit broad substrate specificity. (A) Overview of the identification pipeline for the selected haloacid dehalogenases. We reduced initial homology search hits using sequence similarity networking and multiple sequence alignment, resulting in eight candidates. (B) IC traces of standards colored for distinction. Blue shade: Chloride; Green shade: Glycolate; Pink shade: Chloroacetate; Red shade: Fluoride; Yellow shade: Fluoroacetate. The last four traces show protein 3 and protein 6 tested with fluoroacetate and chloroacetate. Green line: boiled protein control; Blue line: sample at t0; Black: sample after 24 h incubation. The different standard peaks are color coded and dotted lines are extending the colored shading toward the sample peaks. (C) Substrates specificity of ten different substrates with the eight selected proteins. The substrate conversion rate ranges from dark blue (0% conversion) to yellow (100% conversion) (SI Appendix, Figs. S5–S12) gray: not tested. ClAc: Chloroacetate; FAc: Fluoroacetate, FP: Fluoropropionate; FBAL: α-fluoro-β-alanine; DFA: Difluoroacetate; TFA: Trifluoroacetate; DFP: Difluoropropionate; TFP: Trifluoropropionate; TFB: Trifluorbutanoate; HFB: Heptafluorobutanoate. Predicted enzymatic degradability of the substrates based on literature is indicated by the broadness of the bar.

Fig. 2.

Defluorination capability does not correlate with the active site compactness or amino acid sequence identity. (A) HBS size in nm during molecular simulations. Distribution of the Euclidean distances of the 3 C_α atoms of the HBS during simulation from 500 to 1,000 ns. *Asterisks indicate the two enzymes used for protein engineering in this study. (B) Visualization of the active site of the inactive protein P1 (WP_118709078, E. aldenensis) black and the defluorinating protein P6 (WP_178618037, Guopingia tenuis) teal. The variants introduced in P1 (E. aldenensis) are F54I (blue), T57M (orange), H74F (green), and F187S (purple). Substrate specificity of the different protein variants with the two substrates chloroacetate (ClAc) and fluoroacetate (FAc).

Fig. 3.

The intrinsically disordered C-terminal region is important for defluorination activity. (A) Results of Experiment 1 colorimetric assays for defluorination and dechlorination activity, respectively, mapped onto the P6 protein model. The color gradient from white (low activity) to yellow (medium activity) to red (high activity) and line thickness is used to visualize activity differences based on location in the wild-type protein structure. (B) Mean relative feature importance plot extracted from random forest regression model using alanine scanning data to predict defluorination activity. (C) Protein “hotspots” important for defluorination and dechlorination based on the alanine scanning data, where line thickness and coloring correspond to a higher importance for defluorination alone. (D) Top five residue positions with the highest mean Gini importance in random forest classification for defluorinating vs. nondefluorinating proteins. (E) RMS fluctuations (RMSF, reported in nm) of the C_α atoms indicates high movement in the two regions of the protein. The two boxes highlight the more flexible regions in the proteins and lines connect the corresponding amino acids in the protein visualization. Maroon = amino acids 36–69, Red = amino acids 197–235. The surface of the binding pocket location is shaded by proximity to the two regions.

Fig. 4.

Swapping the C-terminus of a nondefluorinating HAD with the C-terminus of a defluorinating homologous enzyme enables defluorination activity. (A) Representation of the defluorinating protein P6. The colored parts represent the exchanged segments used for generating the chimeric protein variants with P1. (B) The protein variants with the colored segment from P6 showed no defluorination activity. (C) The engineered chimeric protein with a swapped C-terminal segment enables weak conversion of fluoroacetate to glycolate and fluoride ion. (D) Protein defluorination activity of the alanine scanning library with all data points shown for two independent biological replicates (Experiment 2 results are visualized, both Experiment 1 and Experiment 2 results are provided in SI Appendix, Tables S13, S14, and S18). The x-axis indicates the position in P6 and the y-axis indicates the change in min-max normalized activity for each alanine scanning variant relative to wild-type activity (dotted gray line). Points are colored by chimeric segments as depicted in other figure panels. The C-terminal chimeric segment exhibits a significantly different distribution of active to inactive alanine variants from the other chimeric segments of the protein (P < 0.05, Kruskal–Wallis test). (E) Boxplot of RMS deviation (RMSD) in nm for different protein segments derived from MD simulations with the C-terminal chimeric segment exhibiting the highest overall movement.