Engineering proteins for environmental applications

Generative models for protein sequence modeling: recent advances and future directions

The widespread adoption of high-throughput omics technologies has exponentially increased the amount of protein sequence data involved in many salient disease pathways and their respective therapeutics and diagnostics. Despite the availability of large-scale sequence data, the lack of experimental fitness annotations underpins the need for self-supervised and unsupervised machine learning (ML) methods. These techniques leverage the meaningful features encoded in abundant unlabeled sequences to accomplish complex protein engineering tasks. Proficiency in the rapidly evolving fields of protein engineering and generative AI is required to realize the full potential of ML models as a tool for protein fitness landscape navigation. Here, we support this work by (i) providing an overview of the architecture and mathematical details of the most successful ML models applicable to sequence data (e.g. variational autoencoders, autoregressive models, generative adversarial neural networks, and diffusion models), (ii) guiding how to effectively implement these models on protein sequence data to predict fitness or generate high-fitness sequences and (iii) highlighting several successful studies that implement these techniques in protein engineering (from paratope regions and subcellular localization prediction to high-fitness sequences and protein design rules generation). By providing a comprehensive survey of model details, novel architecture developments, comparisons of model applications, and current challenges, this study intends to provide structured guidance and robust framework for delivering a prospective outlook in the ML-driven protein engineering field.

Atomic resolution studies of haloalkane dehalogenases DhaA04, DhaA14 and DhaA15 with engineered access tunnels

The haloalkane dehalogenase DhaA from Rhodococcus rhodochrous NCIMB 13064 is a bacterial enzyme that shows catalytic activity for the hydrolytic degradation of the highly toxic industrial pollutant 1,2,3-trichloropropane (TCP). Mutagenesis focused on the access tunnels of DhaA produced protein variants with significantly improved activity towards TCP. Three mutants of DhaA named DhaA04 (C176Y), DhaA14 (I135F) and DhaA15 (C176Y + I135F) were constructed in order to study the functional relevance of the tunnels connecting the buried active site of the protein with the surrounding solvent. All three protein variants were crystallized using the sitting-drop vapour-diffusion technique. The crystals of DhaA04 belonged to the orthorhombic space group P2(1)2(1)2(1), while the crystals of DhaA14 and DhaA15 had triclinic symmetry in space group P1. The crystal structures of DhaA04, DhaA14 and DhaA15 with ligands present in the active site were solved and refined using diffraction data to 1.23, 0.95 and 1.22 A, resolution, respectively. Structural comparisons of the wild type and the three mutants suggest that the tunnels play a key role in the processes of ligand exchange between the buried active site and the surrounding solvent.

Natural genetic transformation in monoculture Acinetobacter sp. strain BD413 biofilms

Horizontal gene transfer by natural genetic transformation in Acinetobacter sp. strain BD413 was investigated by using gfp carried by the autonomously replicating plasmid pGAR1 in a model monoculture biofilm. Biofilm age, DNA concentration, and biofilm mode of growth were evaluated to determine their effects on natural genetic transformation. The highest transfer frequencies were obtained in young and actively growing biofilms when high DNA concentrations were used and when the biofilm developed during continuous exposure to fresh medium without the presence of a significant amount of cells in the suspended fraction. Biofilms were highly amenable to natural transformation. They did not need to advance to an optimal growth phase which ensured the presence of optimally competent biofilm cells. An exposure time of only 15 min was adequate for transformation, and the addition of minute amounts of DNA (2.4 fg of pGAR1 per h) was enough to obtain detectable transfer frequencies. The transformability of biofilms lacking competent cells due to growth in the presence of cells in the bulk phase could be reestablished by starving the noncompetent biofilm prior to DNA exposure. Overall, the evidence suggests that biofilms offer no barrier against effective natural genetic transformation of Acinetobacter sp. strain BD413.

Combined QM/MM study of the mechanism and kinetic isotope effect of the nucleophilic substitution reaction in haloalkane dehalogenase

Combined QM/MM molecular dynamics simulations have been carried out for the dehalogenation reaction of the nucleophilic displacement of dichloroethane catalyzed by haloalkane dehalogenase. The computed chlorine kinetic isotope effects and free energies of activation in the wild-type and the Phe172Trp mutant enzyme are found to be consistent with experiment. In comparison with the uncatalyzed model reaction in water, the enzyme lowers the activation barrier by about 16 kcal/mol. The enormous enzymatic action was attributed to a combination of contributions from a change in the solvation effect and transition state stabilization. The unique features of tryptophan's ability to interact favorably with hydrophobic substrates and to form hydrogen bonds to the leaving group chloride ion at the transition state enable both factors to make significant contributions to the barrier lowering mechanism in the enzyme. This is in contrast to the reference reaction in water, in which hydrogen bonding interactions are weakened at the transition state because of dispersed charge distribution at the transition state relative to that in the reactant and product states.

Analysis of the reaction mechanism and substrate specificity of haloalkane dehalogenases by sequential and structural comparisons

Haloalkane dehalogenases catalyse environmentally important dehalogenation reactions. These microbial enzymes represent objects of interest for protein engineering studies, attempting to improve their catalytic efficiency or broaden their substrate specificity towards environmental pollutants. This paper presents the results of a comparative study of haloalkane dehalogenases originating from different organisms. Protein sequences and the models of tertiary structures of haloalkane dehalogenases were compared to investigate the protein fold, reaction mechanism and substrate specificity of these enzymes. Haloalkane dehalogenases contain the structural motifs of alpha/beta-hydrolases and epoxidases within their sequences. They contain a catalytic triad with two different topological arrangements. The presence of a structurally conserved oxyanion hole suggests the two-step reaction mechanism previously described for haloalkane dehalogenase from Xanthobacter autotrophicus GJ10. The differences in substrate specificity of haloalkane dehalogenases originating from different species might be related to the size and geometry of an active site and its entrance and the efficiency of the transition state and halide ion stabilization by active site residues. Structurally conserved motifs identified within the sequences can be used for the design of specific primers for the experimental screening of haloalkane dehalogenases. Those amino acids which were predicted to be functionally important represent possible targets for future site-directed mutagenesis experiments.

Protein engineering studies of dichloromethane dehalogenase/glutathione S-transferase from Methylophilus sp. strain DM11. Ser12 but not Tyr6 is required for enzyme activity

The structural gene for dichloromethane dehalogenase/glutathione S-transferase (GST, EC 2.5.1.18) from Methylophilus sp. strain DM11 was subcloned into a multicopy plasmid under the control of the T7 polymerase promoter, allowing expression in Escherichia coli and easy purification of the enzyme in good yield. Several point mutations leading to amino acid changes at residues Tyr6, His8 and Ser12 of the protein were introduced in this gene. Mutations at Tyr6, the N-terminal tyrosine known to be essential for enzymatic activity in glutathione S-transferases of the alpha, mu, and pi classes, had little effect on the activity of dichloromethane dehalogenase. The same applied for mutations at residue His8, which from multiple alignments of GST sequences may also correspond to the conserved N-terminal tyrosine residue of GST enzymes. The higher turnover rate of the wild-type enzyme with dibromomethane compared with dichloromethane was lost in mutants with amino acid replacements at residue His8, but retained in mutant proteins at Tyr6. Mutations at Ser12 led to mutants with drastically reduced enzymatic activity, pinpointing this residue as an essential determinant of catalytic efficiency.