Discovery of Novel Gain-of-Function Mutations Guided by Structure-Based Deep Learning.

Despite the promise of deep learning accelerated protein engineering, examples of such improved proteins are scarce. Here we report that a 3D convolutional neural network trained to associate amino acids with neighboring chemical microenvironments can guide identification of novel gain-of-function mutations that are not predicted by energetics-based approaches. Amalgamation of these mutations improved protein function in vivo across three diverse proteins by at least 5-fold. Furthermore, this model provides a means to interrogate the chemical space within protein microenvironments and identify specific chemical interactions that contribute to the gain-of-function phenotypes resulting from individual mutations.

Single-Molecule Mechanistic Study of Enzyme Hysteresis.

Hysteresis is an important feature of enzyme-catalyzed reactions, as it reflects the influence of enzyme regulation in the presence of ligands such as substrates or allosteric molecules. In typical kinetic studies of enzyme activity, hysteretic behavior is observed as a "lag" or "burst" in the time course of the catalyzed reaction. These lags and bursts are due to the relatively slow transition from one state to another state of the enzyme molecule, with different states having different kinetic properties. However, it is difficult to understand the underlying mechanism of hysteresis by observing bulk reactions because the different enzyme molecules in the population behave stochastically. In this work, we studied the hysteretic behavior of mutant ß-glucuronidase (GUS) using a high-throughput single-molecule array platform and investigated the effect of thermal treatment on the hysteresis.

Employing 25-Residue Docking Motifs from Modular Polyketide Synthases as Orthogonal Protein Connectors.

The proteins of trans-acyltransferase modular polyketide synthases (PKSs) self-organize into assembly lines, enabling the multienzyme biosynthesis of complex organic molecules. Docking domains comprised of ∼25 residues at the C- and N-termini of these polypeptides (CDDs and NDDs) help drive this association through the formation of four-helix bundles. Molecular connectors like these are desired in synthetic contexts, such as artificial biocatalytic systems and biomaterials, to orthogonally join proteins. Here, the ability of six CDD/NDD pairs to link non-PKS proteins is examined using green fluorescent protein (GFP) variants. As observed through size-exclusion chromatography and Förster resonance energy transfer (FRET), matched but not mismatched pairs of Venus+CDD and NDD+mTurquoise2 fusion proteins associate with low micromolar affinities.

Supercharging enables organized assembly of synthetic biomolecules.

Symmetrical protein oligomers are ubiquitous in biological systems and perform key structural and regulatory functions. However, there are few methods for constructing such oligomers. Here we have engineered completely synthetic, symmetrical oligomers by combining pairs of oppositely supercharged variants of a normally monomeric model protein through a strategy we term 'supercharged protein assembly' (SuPrA). We show that supercharged variants of green fluorescent protein can assemble into a variety of architectures including a well-defined symmetrical 16-mer structure that we solved using cryo-electron microscopy at 3.47 Å resolution. The 16-mer is composed of two stacked rings of octamers, in which the octamers contain supercharged proteins of alternating charges, and interactions within and between the rings are mediated by a variety of specific electrostatic contacts. The ready assembly of this structure suggests that combining oppositely supercharged pairs of protein variants may provide broad opportunities for generating novel architectures via otherwise unprogrammed interactions.

Retroelement-Based Genome Editing and Evolution.

While several genome editing methods exist, few are suitable for the continuous evolution of targeted sequences.  Here we develop bacterial retroelements known as "retrons" for the dynamic,  in vivo editing and mutagenesis of targeted genes. We first optimized retrons' ability to introduce preprogrammed mutations, optimizing both their expression and the host machinery that interacts with them to increase the incorporation frequency of mutations 78-fold over rates previously reported in synthetic systems. The optimized system is capable of simultaneously overwriting 13 separate positions spanning  a 31-base length, and is for the first time shown to yield targeted deletions and insertions. To engineer retrons as a tool to introduce novel, unprogrammed mutations in specific targeted regions, we expressed them under a mutagenic T7 RNA polymerase. This coupled mutagenic T7 RNA polymerase-retron system enabled the evolution of diverse variants of environmentally selected antibiotic resistance genes, producing mutation rates in the targeted region 190-fold higher than  background cellular mutation rates, potentially enabling the dynamic, continuous self-evolution of selected phenotypes.

Evolving Bacterial Fitness with an Expanded Genetic Code.

Since the fixation of the genetic code, evolution has largely been confined to 20 proteinogenic amino acids. The development of orthogonal translation systems that allow for the codon-specific incorporation of noncanonical amino acids may provide a means to expand the code, but these translation systems cannot be simply superimposed on cells that have spent billions of years optimizing their genomes with the canonical code. We have therefore carried out directed evolution experiments with an orthogonal translation system that inserts 3-nitro-L-tyrosine across from amber codons, creating a 21 amino acid genetic code in which the amber stop codon ambiguously encodes either 3-nitro-L-tyrosine or stop. The 21 amino acid code is enforced through the inclusion of an addicted, essential gene, a beta-lactamase dependent upon 3-nitro-L-tyrosine incorporation. After 2000 generations of directed evolution, the fitness deficit of the original strain was largely repaired through mutations that limited the toxicity of the noncanonical. While the evolved lineages had not resolved the ambiguous coding of the amber codon, the improvements in fitness allowed new amber codons to populate protein coding sequences.

Portable platform for rapid in-field identification of human fecal pollution in water.

Human fecal contamination of water is a public health risk. However, inadequate testing solutions frustrate timely, actionable monitoring. Bacterial culture-based methods are simple but typically cannot distinguish fecal host source. PCR assays can identify host sources but require expertise and infrastructure. To bridge this gap we have developed a field-ready nucleic acid diagnostic platform and rapid sample preparation methods that enable on-site identification of human fecal contamination within 80 min of sampling. Our platform relies on loop-mediated isothermal amplification (LAMP) of human-associated Bacteroides HF183 genetic markers from crude samples. Oligonucleotide strand exchange (OSD) probes reduce false positives by sequence specifically transducing LAMP amplicons into visible fluorescence that can be photographed by unmodified smartphones. Our assay can detect as few as 17 copies/ml of human-associated HF183 targets in sewage-contaminated water without cross-reaction with canine or feline feces. It performs robustly with a variety of environmental water sources and with raw sewage. We have also developed lyophilized assays and inexpensive 3D-printed devices to minimize cost and facilitate field application.