Nature Chemical Biology

Biomedical research overlooks most genes in favor of a well-studied minority, yet whether analogous blind spots exist in metabolomics remains unknown. We show that reductive amination, forming secondary amines from aldehydes or ketones and amines, generates a previously hidden class of metabolites we term alkamines. Multiplexed synthesis of 8,475 alkamines combined with MS/MS searches across 1.7 billion spectra identified 1,626 candidates across multiple species and organs. Of these, 56 were confirmed in biological samples, including 27 steroid- and 12 drug-derived alkamines matching prescription patterns. Notably, 77% of synthesized alkamines are absent from PubChem. This combinatorial logic likely explains why alkamines have evaded detection and suggests drug metabolism frameworks substantially underestimate drug-derived metabolite diversity. Reductive amination is an overlooked route modifying steroids, bile acids, and xenobiotics.

Drug-metabolizing enzymes determine therapeutic exposure, efficacy and toxicity, but defining their isoform-specific functions in vivo remains challenging. Cytochrome P450 enzymes (P450s) are central to drug metabolism and pharmacokinetics (DMPK) and mediate the phase I metabolism of [~]75% of all marketed drugs. However, conventional knockout models can induce develop-mental and compensatory adaptations, and selective inhibitors are unavailable for many P450 isoforms. Here, we report the use of an inducible chemical-genetic platform for acute and specific degradation of the endogenous P450 enzyme Cyp1a2 in mice. Using CRISPR-Cas9-mediated knock-in editing, we introduced an FKBP12F36V degron into the endogenous Cyp1a2 locus to generate Cyp1a2dTAG mice. Treatment with the dTAG degrader dTAG-13 recruited an E3 ubiquitin ligase to CYP1A2dTAG, resulting in rapid and reversible proteasomal depletion of CYP1A2dTAG in vivo. Temporally controlled CYP1A2dTAG loss altered caffeine pharmacokinetics as expected, validating this model as a functional tool for DMPK studies. By enabling reversible suppression of drug-metabolizing enzymes without permanent deletion or chronic inhibitor exposure, this work establishes targeted protein degradation as a broadly adaptable strategy for studying drug metabolism in vivo and provides a foundation for extending inducible DMPK control to other P450s, conjugating enzymes and transporters.

Proximity labeling (PL) technologies like APEX2 have transformed spatial multi-omics in live cells, but their long-standing dependence on hydrogen peroxide (H2O2) disrupts redox signaling and prevents use in live animals. Here we introduce H2O2-independent APEX2 (Hi-APEX), which uses a clickable tetrazine-phenol probe, requiring no enzyme engineering. We show that APEX2 directly catalyzes TP radical formation without H2O2 via a mechanism requiring the probes tetrazine group and a key histidine residue. We benchmarked Hi-APEX-based spatial multi-omics by mapping the mitochondrial matrix and dynamic secretomes. Hi-APEX significantly outperforms traditional APEX in capturing redox-sensitive processes such as stress response and ferroptosis, enabling discovering authentic stress granule components and protein interaction networks for mitochondria-localized GPx4. One mGPx4 interactor TRMT61B--known to regulate mitochondrial m{superscript 1}A modifications--promotes ferroptosis. Crucially, Hi-APEX achieves full in vivo compatibility, enabling direct PL in tumor xenografts and hippocampal neurons, thereby expanding PL-based spatial multi-omics from cellular systems to living organisms.

Molecular glue degraders (MGDs) offer a sophisticated, proximity-based approach to protein modulation. In this study, we introduce LJY-3-60, a novel proximity-inducing agent that unexpectedly triggers the potent and selective autodegradation of CRBN. Evidence from CRISPR-Cas9 screening and IP-MS reveals that this degradation process is strictly governed by the intrinsic CRL4CRBN machinery, independent of any extrinsic E3 recruitment. Through a combination of cellular and biophysical characterizations, we demonstrate that LJY-3-60 acts as a molecular bridge to template CRBN homodimerization. This mechanism is unequivocally elucidated by the atomic-resolution co-crystal structure of the CRBNMidi-LJY-3-60 complex. The structure explicitly delineates the homodimerization interface, revealing how the ligand reorganizes the protein surface to stabilize a non-canonical architecture that drives trans-autoubiquitination and subsequent proteasomal degradation. Furthermore, LJY-3-60 serves as a highly effective, controllable off-switch to mitigate PROTAC-induced toxicity. Ultimately, this work delivers a robust chemical tool for modulating CRBN stability. By demonstrating how a small molecule can functionally mimic an endogenous E3 substrates degron to catalyse targeted autodegradation, this study establishes a rational structural framework for designing the next generation of self-destructive modulators in targeted protein degradation (TPD) therapeutics.

Life is predicated on chirality, a molecular asymmetry akin to the left and right versions of human hands. Here we show that privileged protein residues are predisposed for chiral regulation. We developed enantiomeric oxaziridine reagents that systematically identify pro-(S) and pro-(R) methionine oxidation sites across proteomes that can be erased by stereospecific methionine sulfoxide reductase enzymes A and B, respectively. These probes reveal that chiral regulation of methionine oxidation-reduction processes can allosterically regulate protein function, as shown in cell and murine models of oxidative stress where selective (R)-methionine sulfoxide formation on M69 of biphenyl hydrolase-like protein leads to hydrolase inhibition and amplification of proteome N-homocysteinylation modifications. This work introduces a platform for characterizing sites of asymmetric methionine oxidation and the functional consequences concomitant with an individual chiral single-atom modification.

The regulation of post-translational modifications (PTMs) is central to cellular biology and disease. Induced-proximity strategies enable manipulation of PTMs by recruiting modifying enzymes to proteins of interest, but identifying effective effector enzymes typically requires extensive heterobifunctional molecule synthesis before biological validation. Here we report a modular platform that enables rapid evaluation of PTM editing enzymes against defined protein substrates in living cells using compound-dependent or nanobody-mediated induced proximity. Using lysine acetylation as a model system, we demonstrate programmable acetylation of GFP, histone H3, and p53 through recruitment of diverse acetyltransferases. Effector identity dictates site-specific acetylation patterns, enabling selective PTM deposition across substrates and cellular compartments. This platform enables rapid identification of productive effector-substrate relationships prior to heterobifunctional molecule development, accelerating the design of induced-proximity chemical probes for targeted PTM editing.

Different small-molecule drugs targeting the same protein can produce divergent clinical outcomes through poorly characterized interactome changes. Existing proximity labeling approaches for target identification suffer from background biotinylation independent of small-molecule recruitment, obscuring true drug targets and their binding partners. Here, we incorporate a destabilizing domain (DD) into the biotin targeting chimera (BioTAC) framework to create ddBioTAC, wherein the proximity labeling enzyme TurboID is selectively stabilized only upon binding of a bifunctional targeting molecule. Using the bromodomain-targeting molecule NICE-01 in HeLa cells, we demonstrate that, in the absence of the bifunctional targeting molecule the destabilized TurboID enzyme (TurboID-DD) exhibits reduced protein levels and biotinylation activity compared to the control TurboID-FKBP (FK506-binding protein), while recovering comparable activity upon NICE-01 treatment. This results in an eightfold improvement in specific enrichment of the known target bromodomain containing protein 4 (BRD4) and its interactors, including MED1 and EF1D. Proteome-wide mass spectrometry confirms that ddBioTAC more accurately discriminates drug targets and proximal interactors from non-specific background, advancing unbiased drug-induced interactome profiling.

Ligand dimerization represents a powerful strategy to enhance avidity, potency, and selectivity. Leveraging the natural-product molecular glue Rocaglamide (RocA), we identified BisRoc, a dimeric rocaglate ligand that potently and durably suppresses translation and exhibits greater specificity across a cancer cell line panel than the monomeric RocA. CRISPRi screening revealed that BisRoc activity is influenced by cellular context, including IFITM-mediated uptake, ABC-type efflux transporters, and the translation initiation factor eIF4A2. Mechanistic studies showed that the paralogs eIF4A1 and eIF4A2 are differentially sensitive to BisRoc-induced dimerization. Owing to the presence of multiple binding sites on RNAs, BisRoc-bridged eIF4A-RNA motifs assemble into higher-order complexes that promote stress-granule formation more efficiently than monomeric RocA. Given the widespread multivalency of RNA-RBP interactions, this ligand dimerization strategy may be extended to modulate the higher-order assembly of other RNA-binding proteins. GRAPHICAL ABSTRACT O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=95 SRC="FIGDIR/small/710667v1_ufig1.gif" ALT="Figure 1"> View larger version (26K): org.highwire.dtl.DTLVardef@12b4dbborg.highwire.dtl.DTLVardef@1fc6e23org.highwire.dtl.DTLVardef@1a2c0f4org.highwire.dtl.DTLVardef@29e158_HPS_FORMAT_FIGEXP M_FIG C_FIG

Neurons and glial cells are biochemically coupled through the exchange of nutrients, but our knowledge of which metabolites are transferred between them remains limited due to technical challenges. Here, we introduce a strategy to label specific cell types with isotopic tracers so that metabolite transfer can be measured directly in the intact brain. By engineering neurons in mice to metabolize 13C-labeled cellobiose, a glucose dimer that wild-type cells cannot catabolize, we selectively track neuron-derived metabolites by using mass spectrometry-based metabolomics. Applying this approach enabled us to identify myo-inositol as a critical metabolite synthesized by neurons and transferred to oligodendrocyte progenitor cells (OPCs) via the SLC5A3 transporter. The transfer of myo-inositol from neurons to OPCs promotes OPC proliferation and differentiation by enhancing phosphatidylinositol synthesis and upregulating expression of myelin-associated genes. During demyelination, deficient nutrient transfer can be rescued by dietary supplementation of myo-inositol, which accelerates myelin repair. These findings establish a generalizable technology for tracing intercellular metabolite transfer in vivo and identify a previously unrecognized mechanism of myo-inositol transfer from neurons to glial cells in support of CNS regeneration, revealing a potential metabolic target for therapeutic intervention in neurodegenerative disease.

Rhodopsin misfolding underlies rhodopsin-linked retinitis pigmentosa, and small-molecule pharmacochaperones represent a promising therapeutic strategy. However, the mechanisms by which these compounds interact with and stabilize rhodopsin remain poorly understood. Here, we combine backbone amide hydrogen-deuterium exchange mass spectrometry (amide HDX-MS), histidine-specific HDX (His-HDX), protein structure network (PSN) analysis, molecular docking, and functional spectroscopy to define ligand-induced conformational signatures in this receptor elicited by three non-retinoid small molecules, quercetin, myricetin, and the chromenone CR5, and to compare them with those of the native chromophore 11-cis-retinal. Binding of 11-cis-retinal to ligand-free opsin establishes a benchmark orthosteric conformational signature, characterized by strong backbone HDX protection across TM4-TM7 and adjacent loops, suppression of EX1-like hydrogen-deuterium exchange kinetics at the N-terminal ends of TM1 and TM4, and reorganization of PSN hubs that stabilizes an inactive-state residue interaction network. All three non-retinoid ligands generate HDX footprints that closely track this retinal-induced pattern within the chromophore pocket, consistent with direct orthosteric engagement, but they confer weaker and ligand-specific stabilization. Among them, quercetin most closely reproduces the retinal-like backbone protection and His-HDX microenvironment changes, whereas myricetin and CR5 only partially recapitulate retinal-induced stabilization and redistribute conformational flexibility toward TM1 and intradiscal regions, without fully suppressing EX1-like gating. In addition, all three compounds induce weak cytoplasmic allosteric effects in retinal-bound rhodopsin, indicating secondary interactions in addition to a primary orthosteric mechanism. Together, these results provide the first residue-level experimental framework for understanding the differential pharmacochaperoning capacity of non-retinoid ligands and highlight key conformational principles for future optimization of opsin stabilizers.

Polyketides are prized for their structural complexity and therapeutic potential, yet the incorporation of amide bonds into their frameworks typically relies on linear, nonribosomal peptide synthetase (NRPS)-dependent assembly. The direct, convergent coupling of distinct polyketide chains via amide bond formation--an ideal strategy for combinatorial biosynthesis--has remained largely elusive. Here, we report the discovery of a novel family of arylamine N-acetyltransferases (NATs) from manumycin-type biosynthetic pathways that catalyze an unprecedented intermolecular amidative chain transfer/condensation between an acyl carrier protein (ACP)-tethered polyketide donor and a free polyketide acceptor. Biochemical, structural, and molecular dynamics studies reveal that the representative enzyme, ColC2, possesses a distinctive substrate-binding pocket that diverges from canonical arylamine NATs, conferring exceptional promiscuity toward diverse acyl donors and acceptors. We demonstrate the utility of this biocatalyst by coupling arylamines with either synthetic acyl-thioesters or polyketide synthase (PKS) machinery to generate a library of non-natural polyketide amides and manumycin derivatives. These findings establish a new paradigm for amide bond formation in polyketide biosynthesis and position arylamine NATs as powerful tools for the development of novel therapeutics through combinatorial synthesis.

Covalent chemistry has transformed small-molecule drug discovery, yet analogous strategies for proteins remain largely inaccessible because covalent warheads cannot be readily integrated into biologics. Conventional genetic code expansion requires engineering a dedicated aminoacyl-tRNA synthetase for each new amino acid, rendering broad warhead screening impractical. Here we introduce AminoX, a platform that bypasses this limitation through direct tRNA acylation, enabling site-specific incorporation of chemically diverse non-standard amino acids (nsAAs), including covalent warhead nsAAs compatible with scalable biologic manufacturing and multifunctional nsAAs. Using a pooled mRNA display workflow, we screened more than 2,000 warhead-position combinations in machine learning-designed de novo miniproteins targeting CTLA-4, enabling parallel interrogation of covalent chemistry, linker geometry, and incorporation site. We confirmed covalent engagement on cells together with enhanced functional blockade. Finally, we demonstrate multifunctional nsAAs that combine covalent warheads with fluorogenic reporters for real-time detection of target engagement, as well as dual nsAA incorporation for macrocyclization and fluorescent imaging of covalent binding on cell surfaces. By uniting synthetic biology, chemical biology, generative protein design, and high-throughput functional selection, AminoX compresses covalent protein engineering timelines by orders of magnitude, accelerating the development of next-generation therapeutics, biosensors, and chemical probes.

Paclitaxel biosynthesis is limited by the instability of taxadiene-4(5)-epoxide, which readily diverts to the non-productive byproduct 5(12)-oxa-3(11)-cyclotaxane (OCT) instead of rearranging to taxadiene-5-ol. Although FoTO1 suppresses OCT accumulation, its molecular function has been unclear. Here we identify FoTO1 as a dedicated epoxide isomerase that directs productive rearrangement. Biochemical characterization, site-directed mutagenesis, and QM/MM calculations reveal a pre-organized D68-D149 dyad that electrostatically activates epoxide ring opening and stereospecific rearrangement. Modular dissection of the C-terminal extension further reveals a functional partition between catalytic integrity and productive coupling with T5OH, mediated by specific hydrophobic contacts that enforce precise geometric complementarity at the binary complex interface. These results demonstrate how electrostatic activation and enzyme association cooperate to control the fate of a highly reactive intermediate in paclitaxel biosynthesis.

Adenine is a ubiquitous nucleobase found in nucleic acids, cofactors, and signaling molecules and mediates diverse molecular interactions. Here, we identify TvAPT, an adenine prenyltransferase from the cyanobacterium Trichormus variabilis NIES-23. Unlike canonical enzymes limited to C5 dimethylallylation, TvAPT efficiently catalyzes the unprecedented N6-prenylation of adenine-containing substrates using extended prenyl donors (C10 and C15), markedly increasing the hydrophobicity of the adenine moiety. X-ray structural analysis and protein engineering revealed that an enlarged prenyl-binding pocket enables this donor promiscuity, allowing rational tuning of prenyl-donor preference. These findings establish TvAPT as a versatile biocatalytic platform that expands the chemical space of adenine-containing molecules for biomolecular engineering, as demonstrated by the synthesis of membrane-permeable nucleotides and analogues of plant signaling molecules.

Identifying the direct molecular targets of bioactive natural products remains a central challenge in chemical biology. Here we present an integrated experimental-computational framework, that combines matrix-augmented thermal proteomics with HoloGNN, a holistic graph neural network, to systematically prioritize and validate protein-ligand interactions. Benchmarking with PDBbind datasets HoloGNN achieves state-of-the-art performance. Applying this framework to 50 structurally diverse natural products identified Demethylzeylasteral as a direct interactor of ACLY. Orthogonal biochemical assays confirmed micromolar binding and enzymatic inhibition. In an imiquimod-induced psoriasis-like mouse model, Demethylzeylasteral reduced disease severity and inflammatory cytokine expression. Single-cell transcriptomics revealed that Demethylzeylasteral reverses keratinocyte hyperproliferation and suppresses ACLY-dependent lipid metabolic reprogramming. Together, this scalable, closed-loop strategy integrates thermal proteomics and machine learning to uncover direct targets of natural products and provides mechanistic evidence linking ACLY inhibition to therapeutic modulation of inflammatory pathology.

Multicomponent Rieske oxygenases catalyze diverse oxidative transformations but require precisely matched redox partners to sustain efficient electron transfer, severely limiting their modularity and biocatalytic application. Yet, the molecular logic underlying this specificity remains poorly defined. Here we decode the molecular principles governing redox partner specificity in representative three-component Rieske oxygenase systems. Through systematic mutagenesis analysis and cross-component reconstitution assays, we identify a single ferredoxin residue that acts as a class-defining determinant of oxygenase recognition. Guided by this insight, we reprogram electron transfer between non-cognate components by complementary engineering of the oxygenase interface, creating an unnatural redox chain with substantially enhanced catalytic turnover compared to the native system. Spectroscopic, binding and computational analyses reveal that productive electron transfer arises from optimized electrostatic complementarity and redox potential alignment rather than maximal binding affinity. Extending this strategy to another oxygenase system demonstrates its generality. Together, these results establish transferable design rules for rationally engineering electron transfer pathways in multicomponent oxygenases, enabling their predictable adaptation as customizable biocatalysts.

Cereblon (CRBN)-based molecular glue degraders (MGDs) induce the degradation of diverse disease-relevant proteins, underscoring their broad therapeutic potential. Here we systematically expand the CRBN neosubstrate landscape using a target-agnostic discovery approach. By integrating deep proteomic and ubiquitinomic profiling of a 960-compound library, we identify compound-induced ubiquitination and depletion of over 230 endogenous proteins. Among these, 124 represent previously unreported CRBN neosubstrates, with over half lacking a predicted G-loop degron. We provide this dataset via an interactive resource, NeosubstratesDB. Complementary cellular and biochemical assays mechanistically define the interaction domain of IRAK1 and establish G-loop-dependent degradation for BCL6. Interpretable machine learning (iML) integrating proteomic profiles with chemical structures highlights key molecular fingerprints driving neosubstrate selectivity for targets such as CSNK1A1, ZFP91 and WEE1. Together, these findings significantly expand the repertoire of CRBN neosubstrates and provide a framework for rational design of next-generation MGDs.

Cell-surface degrader platforms typically require target-specific engineering and have therefore been applied to a relatively small set of protein targets. Here we report Z-TAC, a strategy that enables plug-and-play conversion of existing IgG antibodies into cell-surface protein degraders. Across multiple targets from distinct protein families, Z-TAC induced efficient and sustained degradation of both individual receptors and receptor combinations. For a multi-pass membrane receptor lacking selective antagonists, Z-TAC mediated complete receptor degradation and functional inhibition, demonstrating the ability of this platform to overcome the limitations of conventional pharmacological approaches. This study delineates a generalizable and scalable strategy for functional perturbation of the cell-surface proteome.

Activity-based probes (ABPs) are widely used to profile serine protease activity - enzymes central to diverse physiological and pathological processes - but most rely on covalent modification of the conserved catalytic serine residue, often resulting in poor selectivity across related proteases. Here, we introduce covalent macrocyclic activity-based probes (cmABPs) that selectively target non-catalytic residues within serine protease active sites. By combining phage display with systematic electrophile scanning, we identify macrocyclic scaffolds that position sulfur(VI) fluoride (SuFEx) electrophiles to covalently engage alternative nucleophiles such as lysine and tyrosine. Applied to plasma kallikrein, this approach yielded a macrocyclic scaffold that was converted into covalent probes via fluorosulfate scanning. Remarkably, small changes in electrophile structure produced large, tuneable differences in covalent kinetics, with benzenesulfonyl fluoride derivative 23 achieving rapid and complete protein modification. Biochemical and mass spectrometry analyses confirmed selective modification of an active-site lysine by 23, along with robust performance in complex biological samples. Extension to urokinase plasminogen activator further demonstrates the generality of this strategy. More broadly, this work establishes electrophile scanning within macrocyclic scaffolds as a general approach for tuning covalent reactivity and provides a blueprint for designing selective probes that move beyond catalytic-residue targeting.

Targeted protein degradation (TPD) is a therapeutic strategy to remove disease-causing proteins by routing them to the ubiquitin-proteasome, autophagy, or lysosme machineries. For instance, proteolysis-targeting chimeras (PROTACs) are synthetic hetero-bifunctional small molecules that simultaneously bind the target and an E3 ubiquitin ligase to drive ubiquitination and degradation by the proteasome. Despite considerable success, designing such molecules is challenging and the number of currently addressable ubiquitin E3 ligases is limited. Here we demonstrate hetero-bifunctional de novo designed proteins as alternatives for TPD to access more targets and ligases. First, we develop a stable and highly adaptable helix-turn-helix scaffold for presenting different binding sites. Next, we use computational protein design to incorporate and embellish hot-spot-binding sites to target BCL-xL, plus short linear motifs (SLiMs) for KLHL20 ligase recruitment. The resulting mono- and bi-functionalised proteins bind the targets in vitro, and the latter degrade BCL-xL in cells leading to apoptosis.