Chemical protein synthesis enables the construction of specific protein architectures but is limited to millimolar reaction concentrations, restricting access to poorly soluble proteins. Potassium acyltrifluoroboronates (KATs) offer a promising alternative through fast and chemoselective amide bond formation, but their application to protein synthesis has been precluded by the lack of a masking strategy. We report chiral, zwitterionic organoboron complexes that mask amino acid-derived KATs. These molecules exhibit unexpected nitrogen-carbon-boron connectivity and are fully compatible with solid-phase peptide synthesis and stereoretentive deprotection. We synthesized C-terminal KAT peptides and demonstrated KAT ligation at micromolar concentrations for the convergent synthesis of the aggregation-prone programmed death ligand 2 (PD-L2) immunoglobulin V domain. This work establishes organoboron chemistry as an enabling strategy for chemical protein synthesis at low concentrations far more suitable for handling large, aggregation-prone biomolecules.

The development of rapid, chemoselective covalent bond-forming reactions enables the assembly of hydrogel scaffolds suitable for applications in cellular encapsulation. We previously reported that amide-forming ligations between potassium acyltrifluoroborates (KATs) and hydroxylamines produce robust hydrogels that have excellent cytocompatibility, but the requirement for somewhat acidic conditions for efficient hydrogel assemblies limited their application to robust cell types. To overcome this constraint, we have recently found that quinolinium acyltrifluoroborates (QATs) serve as highly efficient reaction partners for amide-forming reactions at neutral pH. In this article, we document the construction of poly(ethylene glycol) (PEG)-derived hydrogels by efficient cross-linking of QAT-functionalized macromers with a partner hydroxylamine-functionalized macromer. Gelation occurs at physiological pH in under 2 min, offering a rapid and facile approach to the immobilization of delicate stromal cells. The cytocompatibility of the cross-linking was demonstrated by in situ gelation in the presence of human mesenchymal stem cells and sustained cell viability for 7 days. Facile incorporation of a cyclic cell adhesion peptide, simply by including the reaction partner in the gelation reactions, illustrated that the desired components can be introduced into the gels without further elaboration.

During normal cellular homeostasis, unfolded and mislocalized proteins are recognized and removed, preventing the build-up of toxic byproducts1. When protein homeostasis is perturbed during ageing, neurodegeneration or cellular stress, proteins can accumulate several forms of chemical damage through reactive metabolites2,3. Such modifications have been proposed to trigger the selective removal of chemically marked proteins3-6; however, identifying modifications that are sufficient to induce protein degradation has remained challenging. Here, using a semi-synthetic chemical biology approach coupled to cellular assays, we found that C-terminal amide-bearing proteins (CTAPs) are rapidly cleared from human cells. A CRISPR screen identified FBXO31 as a reader of C-terminal amides. FBXO31 is a substrate receptor for the SKP1-CUL1-F-box protein (SCF) ubiquitin ligase SCF-FBXO31, which ubiquitylates CTAPs for subsequent proteasomal degradation. A conserved binding pocket enables FBXO31 to bind to almost any C-terminal peptide bearing an amide while retaining exquisite selectivity over non-modified clients. This mechanism facilitates binding and turnover of endogenous CTAPs that are formed after oxidative stress. A dominant human mutation found in neurodevelopmental disorders reverses CTAP recognition, such that non-amidated neosubstrates are now degraded and FBXO31 becomes markedly toxic. We propose that CTAPs may represent the vanguard of a largely unexplored class of modified amino acid degrons that could provide a general strategy for selective yet broad surveillance of chemically damaged proteins.

Low-density lipoprotein (LDL) is the primary natural carrier of lipids in the bloodstream and plays a central role in the development of atherosclerosis. By leveraging LDL's natural tendency to accumulate at sites of plaque formation, LDL can be employed as a carrier to selectively deliver the imaging probes to efficiently detect atherosclerotic plaques. In our previous studies, we reported several LDL-based magnetic resonance imaging contrast agents (MRI-CAs) formed by modifying natural LDL (nLDL) or developing LDL-mimetic (synthetic LDL, sLDL) from lipid nanoparticles (LNPs) utilizing chemical reactions on the nanoparticle surface, including preliminary MRI tests. In this study, we report the in vivo biological functionality of these LDLs (both nLDL and sLDL)-based Gd(III)-based contrast agents (GBCAs) by conducting detailed in vivo studies on two types of atherosclerosis murine models, namely, apoE -/- and LDLr -/- . We provide more comprehensive MRI data accompanied by ex vivo results, including microscopic analysis of aorta segments for LDL accumulation and whole-body cryoVIZ analysis for biodistribution of the probe. We also tested in vitro cellular internalization of sLDL on two cell lines (RAW 264.7 and THP-1), which are derived from macrophages and monocytes, respectively, in order to observe sLDL uptake by macrophages, which are often present at the vulnerable types of atherosclerotic plaques. In conclusion, our current study demonstrates that modified LDLs-both nLDL and sLDL-facilitate MRI detection of atheroplaques by efficient uptake by macrophages. Taken together with the high loading capacity of Gd(III)-chelate molecules on LDL, especially sLDL, the LDL-based MRI contrast agents reported here hold significant potential for the early detection of atherosclerosis, including vulnerable ones, and should be useful for preventive diagnosis strategies.

Interleukin-4 (IL-4) plays a central role in type 2 immune responses. Despite its potential use for allergic and autoimmune diseases, its pleiotropic receptor binding complicates selective targeting of IL-4 signaling pathways. We developed a chemical synthesis of (i) IL-4 variants with atomically tailored side-chain modifications that deter specific receptor interactions and (ii) conditionally activatable IL-4 variants uncaged with 365-nanometer light. In primary cell studies, different variants elicited selective STAT5 or STAT6 phosphorylation in lymphocytes or neutrophils. In murine studies, photocaged IL-4 suppressed inflammation only upon UV irradiation, demonstrating precise on demand control. We accomplished the synthesis and folding of IL-4, a hydrophobic cytokine with three disulfide bonds, using the alpha-ketoacid-hydroxylamine (KAHA) ligation to assemble three segments. We introduced further conjugations, including PEGylation for half-life extension, through orthogonal ligations enabled by functionalized amino acid building blocks. This work highlights the flexibility of chemical protein synthesis to produce therapeutically valuable cytokines, including receptor-biased and spatiotemporally activatable IL-4 variants.

Amide bonds are ubiquitous in molecules of interest such as pharmaceuticals, natural products, and agrochemicals. Over the last decades, significant efforts have been directed at developing efficient methods for the formation of this class of chemical bonds. Herein, we disclose a robust, user-friendly approach for the efficient and rapid parallel synthesis of amide-containing compounds, in good yield and purity. Our approach utilizes automated synthesis technologies and pre-packed capsules containing all the necessary materials for the reaction and work-up. Following manual addition of the amine and carboxylic acid, the reaction and product isolation is achieved automatically in a standardized, integrated manner. This methodology tolerates a wide diversity of carboxylic acids and amines (or amine salts). To accelerate discovery and rapidly create amide-containing compound libraries, the method was miniaturized into an automated parallel synthesis format, using 96-well plate kits, which could be beneficial to medicinal chemists.

Covalent binders to protein targets offer a powerful approach to the generation of tool compounds and an increasingly common strategy for therapeutic development. The installation of electrophiles onto peptide binders, however, is often precluded by standard conditions for peptide synthesis, which involve strong nucleophiles, bases, and acids. The introduction of C-terminal electrophiles is further complicated by the C → N directionality of standard solid-phase peptide synthesis. Here, we employ chemoselective, site-specific functionalization of C-terminal peptide acyl hydrazides to install strong electrophiles on unprotected peptides. Using automated, high-throughput liquid handling and solid-phase extraction techniques, we have established a combinatorial workflow for the synthesis of peptide-derived covalent protease inhibitors. This methodology enables the synthesis and initial screening of inhibitor libraries in a 96-well plate format without the need for chromatographic purification prior to enzyme inhibition studies, leading to the identification of covalent Cathepsin S inhibitors active in the nanomolar range. When tested in cells, the covalent probes revealed strong off-target interactions with the protein disulfide isomerase PDIA1. These findings both underscore the role of chemoselective chemistries for covalent probe synthesis and highlight the utility of the platform for both the rapid identification of potent inhibitors and the detection of potential off-target interactions.

Single chain camelid antigen binding domains, often called nanobodies, have emerged as powerful tools for diagnostics and therapy. Methods for their site-specific modification offer immense potential for enhancing their therapeutic applications, but established approaches, such as fusion proteins, have well-known limitations in the nanobody format. Here, we report a convenient and broadly applicable method for site-specifically functionalizing a single residue near the C-terminus of VHH nanobodies by employing lysine acylation using conjugating enzymes (LACE) to transfer short peptides bearing functional group handles for potassium acyltrifluoroborate (KAT) ligations onto a single lysine residue of the expressed nanobodies. This approach requires a LACE tag (4 residues or 11 residues) in the recombinant nanobodies and enables direct elaboration of the products via a rapid amide forming reaction. In this study, VHH nanobodies expressed in Escherichia coli could be efficiently modified through the transfer of specific chemical handles, enabling their conjugation to small molecules, nanobodies, and antibodies by chemoselective, amide-forming KAT ligations that operate at micromolar concentrations.

Efficient drug discovery depends on reliable synthetic access to candidate molecules, but emerging machine learning approaches to predicting reaction outcomes are hampered by poor availability of high-quality data. Here, we demonstrate an on-demand synthesis platform based on a three-component reaction that delivers drug-like molecules. Miniaturization and automation enable the execution and analysis of 50,000 distinct reactions on a 3-microliter scale from 193 different substrates, producing the largest public reaction outcome dataset. With machine learning, we accurately predict the result of unknown reactions and analyze the impact of dataset size on model training, both enabling accurate outcome predictions even for unseen reactants and providing a sufficiently large dataset to critically evaluate emerging machine learning approaches to chemical reactivity.

The α-ketoacid-hydroxylamine (KAHA) ligation enables the chemoselective coupling of unprotected peptide segments. The most commonly used hydroxylamine building block, (S)-5-oxaproline, yields homoserine residues at ligation sites, limiting applications where the native sequence is essential. To overcome this limitation, we developed cyclic dipeptide-derived hydroxylamine building blocks that enable the formation of canonical amino acids directly under modified KAHA ligation conditions. These building blocks are prepared from dipeptides and are applicable at nonobvious peptide ligation junctions, including Leu-Ile and Lys-Ile. We applied this approach to the synthesis of K48/K63 selectively protected ubiquitin monomers for chemoenzymatic ubiquitin chain formation and the total synthesis of tirzepatide, a GLP-1 receptor agonist peptide therapeutic containing amino-isobutyric acid (Aib) residues and a fatty acid side chain modification. This work establishes a practical approach for KAHA ligation at fully native sites and expands its applicability to the practical synthesis of challenging peptide targets.

Nerve growth factor (NGF) is a powerful neurotrophic protein for treating central nervous system diseases, but its therapeutic utility is limited by severe side effects, including hyperalgesia. These adverse effects arise from pleitropic receptor binding that can, in principle, be modulated by side chain mutations or modificationa task suited for chemical protein synthesis. Despite its small size (13 kDa), the chemical synthesis of NGF has been stymied by exceptional hydrophobicity and the requirement for a 104-residue N-terminal "chaperone peptide" for folding. This study presents a chemical synthesis of NGF using α-ketoacid-hydroxylamine (KAHA) ligations, featuring recombinant production of the chaperone peptide and its chemoselective conversion to a C-terminal α-ketoacid. A novel solubility tag, SOLACE, and ester-forming KAHA ligations enabled assembly of linear proNGF from three synthetic and one recombinant segment. Controlled folding and disulfide-bond formation mediated by the chaperone peptide followed by proteolytic cleavage yielded biologically active synthetic NGF as its noncovalent dimer. The synthetic NGF exhibited comparable activity to recombinant NGF in axon growth assays, establishing a platform for engineering NGF variants with tailored therapeutic properties. This approach provides a versatile framework for the semisynthesis of neurotrophins and related proteins that also require long chaperone peptides for proper folding.

The role of monoclonal antibodies as vehicles to deliver payloads has evolved as a powerful tool in cancer therapy in recent years. The clinical development of therapeutic antibody-conjugates with precise payloads holds great promise for targeted therapeutic interventions. The use of affinity-peptide mediated functionalization of native off-the-shelf antibodies offers an effective approach to selectively modify IgG antibodies with a drug antibody ratio (DAR) of 2. Here, we report the traceless, peptide-directed attachment of two hydroxylamines to native IgGs followed by chemoselective KAT ligation with quinolinium acyltrifluoroborates (QATs), which provide enhanced ligation rates with hydroxylamines under physiological conditions. By applying KAT ligation to the modified antibodies, conjugation of small molecules, proteins, and oligonucleotides to off-the-shelf IgGs proceeds efficiently, in good yields, and with simultaneous cleavage of the affinity peptide-directing moiety.

Interleukin-4 (IL-4) is a cytokine that plays a central role in type 2 immune responses and is involved in regulating pleiotropic actions in our body by engaging multiple different IL-4 receptor (IL-4R) complexes. Targeting the IL-4R system has a high potential for therapeutic intervention for allergic and autoimmune diseases. A challenge in developing this pleiotropic cytokine for clinical application is the construction of variants tailored for engagement with specific receptor IL-4R subunits, which are necessary for selective activation of specific signaling pathways to treat disease with minimum side effects. To establish a platform for preparation of tailored IL-4 variants, we developed a modular and flexible chemical synthesis of IL-4 and applied this approach to the preparation of (i) IL-4 variants that act as receptor antagonist due to presence of unnatural residues that block specific interactions, and (ii) photocaged and in vivo half-life extended IL-4 variants that can be conditionally activated using UV light, achieved by the incorporation of a photocaged Gln116 residue. We were able to show that these different cytokine variants elicit differential STAT5 or STAT6 phosphorylation in lymphocytes or neutrophils in vitro with just one amino acid substitution. Furthermore, we demonstrated that the photocaged IL-4 can be activated by UV light and effectively suppresses neutrophils in an inflammation model in vivo. Collectively, this work demonstrated the flexibility and applicability of chemical protein synthesis by allowing us to broaden the scope of protein variants that can be accessed for the preparation and evaluation of therapeutically valuable proteins.

In contrast to the large body of work on bioactive natural products from individually cultivated bacteria, the chemistry of environmental microbial communities remains largely elusive. Here, we present a comprehensive bioinformatic and functional study on a complex and interaction-rich ecosystem, algal-bacterial (microbial) mats of Lake Chilika in India, Asia’s largest brackish water body. We report the bacterial compositional dynamics over the mat life cycle, >1,300 reconstructed environmental genomes harboring >2,200 biosynthetic gene clusters (BGCs), the successful cultivation of a widespread core microbiome member belonging to the genusRheinheimera, heterologous reconstitution of two silentRheinheimerabiosynthetic pathways, and new compounds with potent protease inhibitory and antiviral activities. The identified substances, posttranslationally modified peptides from the graspetide and spliceotide families, were targeted among the large BGC diversity by applying a strategy focusing on recurring multi-BGC loci identified in diverse samples, suggesting their presence in successful colonizers. In addition to providing broad insights into the biosynthetic potential of a poorly studied community from sampling to bioactive substances, the study highlights the potential of ribosomally synthesized and posttranslationally modified peptides as a large, underexplored resource for antiviral drug discovery.

Inhibition of K -RAS effectors like B-RAF or MEK1/2 is accompanied by treatment resistance in cancer patients via re-activation of PI3K and Wnt signaling. We hypothesized that myotubularin-related-protein-7 (MTMR7), which inhibits PI3K and ERK1/2 signaling downstream of RAS, directly targets RAS and thereby prevents resistance. Using cell and structural biology combined with animal studies, we show that MTMR7 binds and inhibits RAS at cellular membranes. Overexpression of MTMR7 reduced RAS GTPase activities and protein levels, ERK1/2 phosphorylation, c - FOS transcription and cancer cell proliferation in vitro . We located the RAS-inhibitory activity of MTMR7 to its charged coiled coil (CC) region and demonstrate direct interaction with the gastrointestinal cancer-relevant K-RAS G12V mutant, favouring its GDP-bound state. In mouse models of gastric and intestinal cancer, a cell-permeable MTMR7-CC mimicry peptide decreased tumour growth, Ki67 proliferation index and ERK1/2 nuclear positivity. Thus, MTMR7 mimicry peptide(s) could provide a novel strategy for targeting mutant K -RAS in cancers.

Ubiquitin (Ub) is a small, highly conserved protein essential for eukaryotic biology, and is unique in its formation of polyubiquitin chains by conjugation to one of its seven lysine side chains. Here we report that atomic tailoring of Ub side chains – i.e. the insertion, deletion, or replacement of specific atoms – has significant and unexpected consequences on the enzymatic conjugation of Ub oligomers by isopeptide bond formation mediated by E2 conjugating enzymes. These studies employed chemical synthesis and ligation methods to prepare numerous specifically tailored Ub monomers on multi-milligram scales. While some modifications including N-terminal acylation and methionine replacement did not affect protein folding or Ub chain formation with Ube2K, other modifications had a pronounced effect of oligomerization with Ubc13/Mms2. We observed that Ala46Hse mutation obliterates the ability of this Ub monomer to accept another Ub at Lys63 in Ubc13-mediated conjugations. Exhaustive replacement of all seven lysines with shorter surrogates Orn, Dab, or Dap essentially blocks Ub chain formation, and in the case of Dap, precludes proper folding of the Ub protein.

Efficient drug discovery relies on accessing diverse small molecules expediently and reliably. Improvements to reliability through machine learning predictions are hampered by poor availability of high-quality reaction data. Here, we introduce an on-demand synthesis platform based on a three-component reaction that delivers drug-like molecules overnight. Miniaturization and automation enable the execution and analysis of 50,000 reactions on a 3 microliter scale with distinct substrates, producing the largest public reaction outcome dataset. With machine learning, we accurately predict the result of unknown reactions and analyze the impact of data set size on model training. This study advances the on-demand synthesis of drug-like molecules through concatenating chemoselective reactions and provides a sufficiently large data set to critically evaluate emerging machine learning approaches to predicting chemical reactivity.