Using High-Throughput Experiments To Screen N-Glycosyltransferases with Altered Specificities.

The important roles that protein glycosylation plays in modulating the activities and efficacies of protein therapeutics have motivated the development of synthetic glycosylation systems in living bacteria and in vitro. A key challenge is the lack of glycosyltransferases that can efficiently and site-specifically glycosylate desired target proteins without the need to alter primary amino acid sequences at the acceptor site. Here, we report an efficient and systematic method to screen a library of glycosyltransferases capable of modifying comprehensive sets of acceptor peptide sequences in parallel. This approach is enabled by cell-free protein synthesis and mass spectrometry of self-assembled monolayers and is used to engineer a recently discovered prokaryotic N-glycosyltransferase (NGT). We screened 26 pools of site-saturated NGT libraries to identify relevant residues that determine polypeptide specificity and then characterized 122 NGT mutants, using 1052 unique peptides and 52,894 unique reaction conditions. We define a panel of 14 NGTs that can modify 93% of all sequences within the canonical X-1-N-X+1-S/T eukaryotic glycosylation sequences as well as another panel for many noncanonical sequences (with 10 of 17 non-S/T amino acids at the X+2 position). We then successfully applied our panel of NGTs to increase the efficiency of glycosylation for three protein therapeutics. Our work promises to significantly expand the substrates amenable to in vitro and bacterial glycoengineering.

Surface potential modulation as a tool for mitigating challenges in SERS-based microneedle sensors.

Raman spectroscopic-based biosensing strategies are often complicated by low signal and the presence of multiple chemical species. While surface-enhanced Raman spectroscopy (SERS) nanostructured platforms are able to deliver high quality signals by focusing the electromagnetic field into a tight plasmonic hot-spot, it is not a generally applicable strategy as it often depends on the specific adsorption of the analyte of interest onto the SERS platform. This paper describes a strategy to address this challenge by using surface potential as a physical binding agent in the context of microneedle sensors. We show that the potential-dependent adsorption of different chemical species allows scrutinization of the contributions of different chemical species to the final spectrum, and that the ability to cyclically adsorb and desorb molecules from the surface enables efficient application of multivariate analysis methods. We demonstrate how the strategy can be used to mitigate potentially confounding phenomena, such as surface reactions, competitive adsorption and the presence of molecules with similar structures. In addition, this decomposition helps evaluate criteria to maximize the signal of one molecule with respect to others, offering new opportunities to enhance the measurement of analytes in the presence of interferants.

Tyrosine phosphatase activity is restricted by basic charge substituting mutation of substrates.

Phosphorylation controls important cellular signals and its dysregulation leads to disease. While most phospho-regulation studies are focused on kinases, phosphatases are comparatively overlooked. Combining peptide arrays with SAMDI mass spectrometry, we show that tyrosine phosphatase activity is restricted by basic amino acids adjacent to phosphotyrosines. We validate this model using two ß-catenin mutants associated with cancer (T653R/K) and a mouse model for intellectual disability (T653K). These mutants introduce a basic residue next to Y654, an established phosphorylation site where modification shifts ß-catenin from cell-cell adhesions and towards its essential nuclear role as Wnt-signaling effector. We show that T653-basic mutant ß-catenins are less efficiently dephosphorylated by phosphatases, leading to sustained Y654 phosphorylation and elevated Wnt signals, similar to those observed for Y654E phospho-mimic mutant mice. This model rationalizes how basic mutations proximal to phosphotyrosines can restrict counter-regulation by phosphatases, providing new mechanismistic and treatment insights for 6000+ potentially relevant cancer mutations.

Scanning Transmission Electron Microscopy in a Scanning Electron Microscope for the High-Throughput Imaging of Biological Assemblies.

Electron microscopy of soft and biological materials, or "soft electron microscopy", is essential to the characterization of macromolecules. Soft microscopy is governed by enhancing contrast while maintaining low electron doses, and sample preparation and imaging methodologies are driven by the length scale of features of interest. While cryo-electron microscopy offers the highest resolution, larger structures can be characterized efficiently and with high contrast using low-voltage electron microscopy by performing scanning transmission electron microscopy in a scanning electron microscope (STEM-in-SEM). Here, STEM-in-SEM is demonstrated for a four-lobed protein assembly where the arrangement of the proteins in the construct must be examined. STEM image simulations show the theoretical contrast enhancement at SEM-level voltages for unstained structures, and experimental images with multiple STEM modes exhibit the resolution possible for negative-stained proteins. This technique can be extended to complex protein assemblies, larger structures such as cell sections, and hybrid materials, making STEM-in-SEM a valuable high-throughput imaging method.

Cell-free prototyping enables implementation of optimized reverse ß-oxidation pathways in heterotrophic and autotrophic bacteria.

Carbon-negative synthesis of biochemical products has the potential to mitigate global CO2 emissions. An attractive route to do this is the reverse ß-oxidation (r-BOX) pathway coupled to the Wood-Ljungdahl pathway. Here, we optimize and implement r-BOX for the synthesis of C4-C6 acids and alcohols. With a high-throughput in vitro prototyping workflow, we screen 762 unique pathway combinations using cell-free extracts tailored for r-BOX to identify enzyme sets for enhanced product selectivity. Implementation of these pathways into Escherichia coli generates designer strains for the selective production of butanoic acid (4.9 ± 0.1 gL-1), as well as hexanoic acid (3.06 ± 0.03 gL-1) and 1-hexanol (1.0 ± 0.1 gL-1) at the best performance reported to date in this bacterium. We also generate Clostridium autoethanogenum strains able to produce 1-hexanol from syngas, achieving a titer of 0.26 gL-1 in a 1.5 L continuous fermentation. Our strategy enables optimization of r-BOX derived products for biomanufacturing and industrial biotechnology.

High-Throughput Microfluidics Platform for Intracellular Delivery and Sampling of Biomolecules from Live Cells.

Nondestructive cell membrane permeabilization systems enable the intracellular delivery of exogenous biomolecules for cell engineering tasks as well as the temporal sampling of cytosolic contents from live cells for the analysis of dynamic processes. Here, we report a microwell array format live-cell analysis device (LCAD) that can perform localized-electroporation induced membrane permeabilization, for cellular delivery or sampling, and directly interfaces with surface-based biosensors for analyzing the extracted contents. We demonstrate the capabilities of the LCAD via an automated high-throughput workflow for multimodal analysis of live-cell dynamics, consisting of quantitative measurements of enzyme activity using self-assembled monolayers for MALDI mass spectrometry (SAMDI) and deep-learning enhanced imaging and analysis. By combining a fabrication protocol that enables robust assembly and operation of multilayer devices with embedded gold electrodes and an automated imaging workflow, we successfully deliver functional molecules (plasmid and siRNA) into live cells at multiple time-points and track their effect on gene expression and cell morphology temporally. Furthermore, we report sampling performance enhancements, achieving saturation levels of protein tyrosine phosphatase activity measured from as few as 60 cells, and demonstrate control over the amount of sampled contents by optimization of electroporation parameters using a lumped model. Lastly, we investigate the implications of cell morphology on electroporation-induced sampling of fluorescent molecules using a deep-learning enhanced image analysis workflow.

Efficient Enzymatic Incorporation of Dehydroalanine Based on SAMDI-Assisted Identification of Optimized Tags for OspF/SpvC.

Site-specific modification of proteins has important applications in biological research and drug development. Reactive tags such as azide, alkyne, and tetrazine have been used extensively to achieve the abovementioned goal. However, bulky side-chain "ligation scars" are often left after the labeling and may hinder the biological application of such engineered protein products. Conjugation chemistry via dehydroalanine (Dha) may provide an opportunity for "traceless" ligation because the activated alkene moiety on Dha can then serve as an electrophile to react with radicalophile, thiol/amine nucleophile, and reactive phosphine probe to introduce a minimal linker in the protein post-translational modifications. In this report, we present a mild and highly efficient enzymatic approach to incorporate Dha with phosphothreonine/serine lyases, OspF and SpvC. These lyases originally catalyze an irreversible elimination reaction that converts a doubly phosphorylated substrate with phosphothreonine (pT) or phosphoserine (pS) to dehydrobutyrine (Dhb) or Dha. To generate a simple monophosphorylated tag for these lyases, we conducted a systematic approach to profile the substrate specificity of OspF and SpvC using peptide arrays and self-assembled monolayers for matrix-assisted laser desorption/ionization mass spectrometry. The optimized tag, [F/Y/W]-pT/pS-[F/Y/W] (where [F/Y/W] indicates an aromatic residue), results in a ∼10-fold enhancement of the overall peptide labeling efficiency via Dha chemistry and enables the first demonstration of protein labeling as well as live cell labeling with a minimal ligation linker via enzyme-mediated incorporation of Dha.

Characterizing Enzyme Cooperativity with Imaging SAMDI-MS.

This paper describes a method that combines a microfluidic device and self-assembled monolayers for matrix-assisted laser desorption/ionization mass spectrometry (SAMDI) mass spectrometry to calculate the cooperativity in binding of calcium ions to peptidylarginine deiminase type 2 (PAD2). This example uses only 120 µL of enzyme solution and three fluidic inputs. This microfluidic device incorporates a self-assembled monolayer that is functionalized with a peptide substrate for PAD2. The enzyme and different concentrations of calcium ions are flowed through each of eight channels, where the position along the channel corresponds to reaction time and position across the channel corresponds to the concentration of Ca2+ . Imaging SAMDI (iSAMDI) is then used to determine the yield for the enzyme reaction at each 200 µm pixel on the monolayer, providing a time course for the reactions. Analysis of the peptide conversion as a function of position and time gives the degree of cooperativity (n) and the concentration of ligand required for half maximal activity (K0.5 ) for the Ca2+ - dependent activation of PAD2. This work establishes a high-throughput and label-free method for studying enzyme-ligand binding interactions and widens the applicability of microfluidics and matrix-assisted laser desorption/ionization mass spectrometry (MALDI) imaging mass spectrometry.

NADH inhibition of SIRT1 links energy state to transcription during time-restricted feeding.

In mammals, circadian rhythms are entrained to the light cycle and drive daily oscillations in levels of NAD+, a cosubstrate of the class III histone deacetylase sirtuin 1 (SIRT1) that associates with clock transcription factors. Although NAD+ also participates in redox reactions, the extent to which NAD(H) couples nutrient state with circadian transcriptional cycles remains unknown. Here we show that nocturnal animals subjected to time-restricted feeding of a calorie-restricted diet (TRF-CR) only during night-time display reduced body temperature and elevated hepatic NADH during daytime. Genetic uncoupling of nutrient state from NADH redox state through transduction of the water-forming NADH oxidase from Lactobacillus brevis (LbNOX) increases daytime body temperature and blood and liver acyl-carnitines. LbNOX expression in TRF-CR mice induces oxidative gene networks controlled by brain and muscle Arnt-like protein 1 (BMAL1) and peroxisome proliferator-activated receptor alpha (PPARα) and suppresses amino acid catabolic pathways. Enzymatic analyses reveal that NADH inhibits SIRT1 in vitro, corresponding with reduced deacetylation of SIRT1 substrates during TRF-CR in vivo. Remarkably, Sirt1 liver nullizygous animals subjected to TRF-CR display persistent hypothermia even when NADH is oxidized by LbNOX. Our findings reveal that the hepatic NADH cycle links nutrient state to whole-body energetics through the rhythmic regulation of SIRT1.

Morphological features of single cells enable accurate automated classification of cancer from non-cancer cell lines.

Accurate cancer detection and diagnosis is of utmost importance for reliable drug-response prediction. Successful cancer characterization relies on both genetic analysis and histological scans from tumor biopsies. It is known that the cytoskeleton is significantly altered in cancer, as cellular structure dynamically remodels to promote proliferation, migration, and metastasis. We exploited these structural differences with supervised feature extraction methods to introduce an algorithm that could distinguish cancer from non-cancer cells presented in high-resolution, single cell images. In this paper, we successfully identified the features with the most discriminatory power to successfully predict cell type with as few as 100 cells per cell line. This trait overcomes a key barrier of machine learning methodologies: insufficient data. Furthermore, normalizing cell shape via microcontact printing on self-assembled monolayers enabled better discrimination of cell lines with difficult-to-distinguish phenotypes. Classification accuracy remained robust as we tested dissimilar cell lines across various tissue origins, which supports the generalizability of our algorithm.

Storing and Reading Information in Mixtures of Fluorescent Molecules.

The rapidly increasing use of digital technologies requires the rethinking of methods to store data. This work shows that digital data can be stored in mixtures of fluorescent dye molecules, which are deposited on a surface by inkjet printing, where an amide bond tethers the dye molecules to the surface. A microscope equipped with a multichannel fluorescence detector distinguishes individual dyes in the mixture. The presence or absence of these molecules in the mixture encodes binary information (i.e., "0" or "1"). The use of mixtures of molecules, instead of sequence-defined macromolecules, minimizes the time and difficulty of synthesis and eliminates the requirement of sequencing. We have written, stored, and read a total of approximately 400 kilobits (both text and images) with greater than 99% recovery of information, written at an average rate of 128 bits/s (16 bytes/s) and read at a rate of 469 bits/s (58.6 bytes/s).

Development of an Enzyme-Inhibitor Reaction Using Cellular Retinoic Acid Binding Protein II for One-Pot Megamolecule Assembly.

This paper presents an enzyme building block for the assembly of megamolecules. The system is based on the inhibition of the human-derived cellular retinoic acid binding protein II (CRABP2) domain. We synthesized a synthetic retinoid bearing an arylfluorosulfate group, which uses sulfur fluoride exchange click chemistry to covalently inhibit CRABP2. We conjugated both the inhibitor and a fluorescein tag to an oligo(ethylene glycol) backbone and measured a second-order rate constant for the protein inhibition reaction of approximately 3,600 M-1 s-1 . We used this new enzyme-inhibitor pair to assemble multi-protein structures in one-pot reactions using three orthogonal assembly chemistries to demonstrate exact control over the placement of protein domains within a single, homogeneous molecule. This work enables a new dimension of control over specificity, orientation, and stoichiometry of protein domains within atomically precise nanostructures.

Synthesis, Characterization, and Simulation of Four-Armed Megamolecules.

This paper describes the synthesis, characterization, and modeling of a series of molecules having four protein domains attached to a central core. The molecules were assembled with the "megamolecule" strategy, wherein enzymes react with their covalent inhibitors that are substituted on a linker. Three linkers were synthesized, where each had four oligo(ethylene glycol)-based arms terminated in a para-nitrophenyl phosphonate group that is a covalent inhibitor for cutinase. This enzyme is a serine hydrolase and reacts efficiently with the phosphonate to give a new ester linkage at the Ser-120 residue in the active site of the enzyme. Negative-stain transmission electron microscopy (TEM) images confirmed the architecture of the four-armed megamolecules. These cutinase tetramers were also characterized by X-ray crystallography, which confirmed the active-site serine-phosphonate linkage by electron-density maps. Molecular dynamics simulations of the tetracutinase megamolecules using three different force field setups were performed and compared with the TEM observations. Using the Amberff99SB-disp + pH7 force field, the two-dimensional projection distances of the megamolecules were found to agree with the measured dimensions from TEM. The study described here, which combines high-resolution characterization with molecular dynamics simulations, will lead to a comprehensive understanding of the molecular structures and dynamics for this new class of molecules.

Exploring the Ligand Preferences of the PHD1 Domain of Histone Demethylase KDM5A Reveals Tolerance for Modifications of the Q5 Residue of Histone 3.

Understanding the ligand preferences of epigenetic reader domains enables identification of modification states of chromatin with which these domains associate and can yield insight into recruitment and catalysis of chromatin-acting complexes. However, thorough exploration of the ligand preferences of reader domains is hindered by the limitations of traditional protein-ligand binding assays. Here, we evaluate the binding preferences of the PHD1 domain of histone demethylase KDM5A using the protein interaction by SAMDI (PI-SAMDI) assay, which measures protein-ligand binding in a high-throughput and sensitive manner via binding-induced enhancement in the activity of a reporter enzyme, in combination with fluorescence polarization. The PI-SAMDI assay was validated by confirming its ability to accurately profile the relative binding affinity of a set of well-characterized histone 3 (H3) ligands of PHD1. The assay was then used to assess the affinity of PHD1 for 361 H3 mutant ligands, a select number of which were further characterized by fluorescence polarization. Together, these experiments revealed PHD1's tolerance for H3Q5 mutations, including an unexpected tolerance for aromatic residues in this position. Motivated by this finding, we further demonstrate a high-affinity interaction between PHD1 and recently identified Q5-serotonylated H3. This work yields interesting insights into permissible PHD1-H3 interactions and demonstrates the value of interfacing PI-SAMDI and fluorescence polarization in investigations of protein-ligand binding.

Synthetic Tuning of Domain Stoichiometry in Nanobody-Enzyme Megamolecules.

This paper presents a method to synthetically tune atomically precise megamolecule nanobody-enzyme conjugates for prodrug cancer therapy. Previous efforts to create heterobifunctional protein conjugates suffered from heterogeneity in domain stoichiometry, which in part led to the failure of antibody-enzyme conjugates in clinical trials. We used the megamolecule approach to synthesize anti-HER2 nanobody-cytosine deaminase conjugates with tunable numbers of nanobody and enzyme domains in a single, covalent molecule. Linking two nanobody domains to one enzyme domain improved avidity to a human cancer cell line by 4-fold but did not increase cytotoxicity significantly due to lowered enzyme activity. In contrast, a megamolecule composed of one nanobody and two enzyme domains resulted in an 8-fold improvement in the catalytic efficiency and increased the cytotoxic effect by over 5-fold in spheroid culture, indicating that the multimeric structure allowed for an increase in local drug activation. Our work demonstrates that the megamolecule strategy can be used to study structure-function relationships of protein conjugate therapeutics with synthetic control of protein domain stoichiometry.

High Throughput Screening with SAMDI Mass Spectrometry for Directed Evolution.

Advances in directed evolution have led to an exploration of new and important chemical transformations; however, many of these efforts still rely on the use of low-throughput chromatography-based screening methods. We present a high-throughput strategy for screening libraries of enzyme variants for improved activity. Unpurified reaction products are immobilized to a self-assembled monolayer and analyzed by mass spectrometry, allowing for direct evaluation of thousands of variants in under an hour. The method was demonstrated with libraries of randomly mutated cytochrome P411 variants to identify improved catalysts for C-H alkylation. The technique may be tailored to evolve enzymatic activity for a variety of transformations where higher throughput is needed.

Computational planning of the synthesis of complex natural products.

Training algorithms to computationally plan multistep organic syntheses has been a challenge for more than 50 years1-7. However, the field has progressed greatly since the development of early programs such as LHASA1,7, for which reaction choices at each step were made by human operators. Multiple software platforms6,8-14 are now capable of completely autonomous planning. But these programs 'think' only one step at a time and have so far been limited to relatively simple targets, the syntheses of which could arguably be designed by human chemists within minutes, without the help of a computer. Furthermore, no algorithm has yet been able to design plausible routes to complex natural products, for which much more far-sighted, multistep planning is necessary15,16 and closely related literature precedents cannot be relied on. Here we demonstrate that such computational synthesis planning is possible, provided that the program's knowledge of organic chemistry and data-based artificial intelligence routines are augmented with causal relationships17,18, allowing it to 'strategize' over multiple synthetic steps. Using a Turing-like test administered to synthesis experts, we show that the routes designed by such a program are largely indistinguishable from those designed by humans. We also successfully validated three computer-designed syntheses of natural products in the laboratory. Taken together, these results indicate that expert-level automated synthetic planning is feasible, pending continued improvements to the reaction knowledge base and further code optimization.

Design and Synthesis of Megamolecule Mimics of a Therapeutic Antibody.

This communication describes the design, synthesis, and biological activity of a megamolecule mimic of an anti-HER2 antibody. The antibody mimic was prepared by linking two Fabs from the therapeutic antibody trastuzumab, which are fused through the heavy chain variable domain to either cutinase or SnapTag, with a linker terminated in an irreversible inhibitor for each enzyme. This mimic binds HER2 with comparable avidity to trastuzumab, has similar activity in a cell-based assay, and can arrest tumor growth in a mouse xenograft BT474 tumor model. A panel of 16 bivalent anti-HER2 antibodies were prepared wherein each varied in the orientation of the fusion domain on the Fabs. The analogs displayed a range of cytotoxic activity, and surprisingly, the most active mimic binds to cells with a 10-fold lower avidity than the least active variant suggesting that structure plays a large role in their efficacy. This work suggests that the megamolecule approach can be used to prepare antibody mimics having a broad structural diversity.

Outer-Membrane Protease (OmpT) Based E. coli Sensing with Anionic Polythiophene and Unlabeled Peptide Substrate.

E. coli and Salmonella are two of the most common bacterial pathogens involved in foodborne and waterborne related deaths. Hence, it is critical to develop rapid and sensitive detection strategies for near-outbreak applications. Reported is a simple and specific assay to detect as low as 1 CFU mL-1 of E. coli in water within 6 hours by targeting the bacteria's surface protease activity. The assay relies on polythiophene acetic acid (PTAA) as an optical reporter and a short unlabeled peptide (LL37FRRV ) previously optimized as a substrate for OmpT, an outer-membrane protease on E. coli. LL37FRRV interacts with PTAA to enhance its fluorescence while also inducing the formation of a helical PTAA-LL37FRRV construct, as confirmed by circular dichroism. However, in the presence of E. coli LL37FRRV is cleaved and can no longer affect the conformations and optical properties of PTAA. This ability to distinguish between an intact and cleaved peptide was investigated in detail using LL37FRRV sequence variants.

High-throughput photocapture approach for reaction discovery.

Modern organic reaction discovery and development relies on the rapid assessment of large arrays of hypothesis-driven experiments. The time-intensive nature of reaction analysis presents the greatest practical barrier for the execution of this iterative process that underpins the development of new bioactive agents. Toward addressing this critical bottleneck, we report herein a high-throughput analysis (HTA) method of reaction mixtures by photocapture on a 384-spot diazirine-terminated self-assembled monolayer, and self-assembled monolayers for matrix-assisted laser desorption/ionization mass spectrometry (SAMDI-MS) analysis. This analytical platform has been applied to the identification of a single-electron-promoted reductive coupling of acyl azolium species.