N Nature Chemical Biology · Dec 05, 2025 Accurate single-domain scaffolding of three nonoverlapping protein epitopes using deep learning De novo protein design has seen major success in scaffolding single functional motifs; however, in nature, most proteins present multiple functional sites. Here, we describe an approach to simultaneously scaffold multiple functional sites in a single-domain protein using deep learning. We designed small single-domain immunogens, under 130 residues, that present three distinct and irregular motifs from respiratory syncytial virus. These motifs together comprise nearly half of the designed proteins; hence, the overall folds are quite unusual with little global similarity to proteins in the Protein Data Bank. Despite this, X-ray crystal structures confirmed the accuracy of presentation of each of the motifs and the multiepitope design yields improved cross-reactive titers and neutralizing response compared to a single-epitope immunogen. The successful presentation of three distinct binding surfaces in a small single-domain protein highlights the power of generative deep learning methods to solve complex protein design problems. Machine learning Protein design Proteins Vaccines biology
N Nature Chemical Biology · Nov 21, 2025 Molecular basis of SAM-AMP synthesis and degradation in the type III-B CRISPR–Cas system Upon sensing nonself target RNA, the CorA-associated type III-B CRISPR–Cas system catalyzesS-adenosyl methionine (SAM) and ATP to synthesize SAM-AMP, which activates the effector CorA and triggers immune responses. SAM-AMP can be degraded by NrN and SAM lyase, potentially deactivating the system. Here we find that the type III-B effector complex fromBacteroidesfragilisuses a specific mechanism to recognize nonself target RNA and synthesize SAM-AMP. The 3′ anti-tag of nonself target RNA induces conformational changes in the Cmr2 subunit, triggering SAM-AMP synthesis independently of the stalk loop of Cmr3 subunit. SAM-AMP binding induces NrN to transit from an open to a closed conformation, enabling hydrolysis of the 3′–5′ phosphodiester bond. SAM lyase forms a triangular trimer that specifically degrades SAM-AMP into 5′-methylthioadenosine-AMP and homoserine lactone. These findings unveil unique mechanisms for SAM-AMP synthesis and degradation and provide deeper insights into the molecular basis of type III CRISPR–Cas signaling. Bacteria Biosynthesis Enzyme mechanisms Small molecules Structural biology biology
N Nature Chemical Biology · Nov 21, 2025 SINE compounds activate exportin 1 degradation through an allosteric mechanism Overexpression of exportin 1 (XPO1/CRM1) in cancer cells mislocalizes numerous cancer-related nuclear export cargoes. Covalent selective inhibitors of nuclear export (SINEs), including the cancer drug selinexor, restore proper nuclear localization by blocking XPO1–cargo interaction. These inhibitors also induce XPO1 degradation through the Cullin–RING E3 ligase (CRL) substrate receptor ASB8. Here we present cryo-electron microscopy structures revealing ASB8 binding to a cryptic XPO1 site that is exposed upon SINE conjugation. Unlike typical molecular glue degraders that directly bridge CRLs and substrates, SINEs bind XPO1 independently of ASB8, triggering an allosteric mechanism that enables high-affinity ASB8 recruitment, leading to XPO1 ubiquitination and degradation. ASB8-mediated degradation is also triggered by the endogenous itaconate derivative 4-octyl itaconate, suggesting that synthetic XPO1 inhibitors exploit a native cellular mechanism. This allosteric XPO1 degradation mechanism expands known modes of targeted protein degradation beyond molecular glue degraders and proteolysis-targeting chimeras of CRL4. Cancer therapy Small molecules Structural biology biology
N Nature Chemical Biology · Nov 21, 2025 Dynamic metal coordination controls chemoselectivity in a radical halogenase The activation of inert C(sp3)–H bonds by nonheme Fe enzymes provides a powerful biocatalytic platform for the chemical synthesis of molecules with increased sp3 complexity. In this context, FeII/α-ketoglutarate-dependent radical halogenases are uniquely capable of carrying out transfer of a diverse array of bound anions following C–H activation. Here, we provide experimental evidence that bifurcation of radical rebound after H-atom abstraction can be driven both by the ability of a dynamic metal coordination sphere to reorganize and by a second-sphere hydrogen-bonding network where only two residues are sufficient. In addition, we present crystallographic data supporting the existence of an early peroxyhemiketal intermediate in the O2 activation pathway of FeII/α-ketoglutarate-dependent enzymes. These data provide a paradigm for understanding the evolution of catalytic plasticity in these enzymes and yields insight into the design principles by which to expand their reaction scope. Radical FeII/α-ketoglutarate-dependent halogenases are powerful biocatalysts for C–H functionalization. Here, the authors reveal the mechanistic basis for chemoselectivity in a lysine halogenase. Biocatalysis Enzyme mechanisms Enzymes biology
N Nature Chemical Biology · Nov 18, 2025 Temporal photoproximity labeling of ligand-activated EGFR neighborhoods using MultiMap Photoproximity labeling proteomics (PLP) methods have recently shown that cell surface receptors can form lateral interactome networks. Here, we present a paired set of PLP workflows that dynamically track neighborhood changes for oncogenic epidermal growth factor receptor (EGFR) over time, both outside and inside of cells. We achieved this by augmenting the multiscale PLP workflow we call MultiMap, where three photoprobes with different labeling ranges were photoactivated by one photocatalyst, eosin Y, anchored extracellularly and intracellularly on EGFR. We identified hundreds of neighboring proteins that changed within minutes to over 1 h after the addition of EGF. These neighborhoods reveal dynamic interactomes during early, middle and late signaling that drive phosphorylation, internalization, degradation and transcriptional regulation. This rapid ‘molecular photographic’ labeling approach provides snapshots of signaling neighborhoods, revealing their dynamic nature and potential for drug targeting. Chemical tools Membrane proteins Proteomics Target identification
N Nature Chemical Biology · Nov 14, 2025 Dynamics driving the precursor in NifEN scaffold during nitrogenase FeMo-cofactor assembly Nitrogenase catalyzes atmospheric nitrogen fixation, a critical biological process that depends on an intricate organometallic cofactor assembled by a dedicated multiprotein system. Here we uncover the structural basis for the function of NifEN, the scaffold protein that mediates the final stages of cofactor biosynthesis before its incorporation into nitrogenase. High-resolution structural analyses reveal that the cofactor precursor initially binds at a surface docking site before being transferred into a specialized cavity for further maturation. This process involves dynamic structural rearrangements, including coordinated domain motions and partial unfolding, enabling the scaffold to alternate between open and closed states. Additionally, a rear channel extends to the precursor-binding cavity, likely facilitating the entry of the modifying components molybdenum and homocitrate. These findings illuminate the dynamic mechanisms underlying FeMo-cofactor assembly and underscore the functional divergence between NifEN, the biosynthetic scaffold, and NifDK, the catalytic component of nitrogenase. Enzyme mechanisms Metals Proteins X-ray crystallography
N Nature Chemical Biology · Nov 13, 2025 Chemical richness and diversity of uncultivated ‘Entotheonella’ symbionts in marine sponges Marine sponges are the source of numerous bioactive natural products that serve as chemical defenses and provide pharmaceutical leads for drug development. For some of the compounds, symbiotic bacteria have been established as the actual producers. Among the known sponge symbionts, ‘CandidatusEntotheonella’ members stand out because of their abundant and variable biosynthetic gene clusters (BGCs). Here, to obtain broader insights into this producer taxon, we conduct a comparative analysis on eight sponges through metagenomic and single-bacterial sequencing and biochemical studies. The data suggest sets of biosynthetic genes that are largely unique in 14 ‘Entotheonella’ candidate species and a member of a sister lineage named ‘CandidatusProxinella’. Four biosynthetic loci were linked in silico or experimentally to cytotoxins, antibiotics and the terpene cembrene A from corals. The results support widespread and diverse bacterial roles in the chemistry of sponges and aid the development of sustainable production methods for sponge-derived therapeutics. Bacteria Biosynthesis Natural products
N Nature Chemical Biology · Nov 03, 2025 Mutant p53 protein accumulation is selectively targetable by proximity-inducing drugs TP53mutant cancers are associated with approximately half of cancer deaths. The most common mechanism of p53 inactivation involves missense mutations. Such mutations inTP53result in a robust upregulation of the p53 protein. Here, we demonstrate an induced proximity approach to selectively killTP53mutant cells. This approach uses the increased abundance of p53 protein inTP53mutant cancer cells to concentrate toxic molecules in these cells. We demonstrate this approach with a molecule that binds the Y220C mutant of p53 and concentrates a PLK1 inhibitor in cells harboringTP53Y220Cmutations. The resulting bifunctional molecule promotes formation of a p53Y220C–PLK1 ternary complex, mislocalizes PLK1, inhibits PLK1 activity, elicits selective G2/M arrest and induces apoptosis inTP53Y220Ccells while sparing wild-typeTP53cells. These data exemplify a potentially generalizable framework for targetingTP53missense mutations by leveraging mutant p53 protein abundance to induce cell death, independent of p53’s transcriptional activity. Cancer therapy Cell death Cell signalling Small molecules
N Nature Chemical Biology · Oct 31, 2025 Click-code-seq reveals strand biases of DNA oxidation and depurination in human genome DNA modifications drive aging, neurodegeneration, carcinogenesis and chemotherapy drug action. Accurate mapping of diverse DNA modifications with single-nucleotide precision in complex genomes remains challenging. We upgraded click-code-seq, a click-chemistry-aided DNA-modification mapping strategy, to enable its first application for sequencing oxidation and depurination in the human genome. We developed a companion fluorescence assay, click-fluoro-quant, to rapidly quantify common DNA modifications and novel adaptors to minimize false positives and assess modification frequency. We uncovered that endogenous DNA oxidation in a human cell line mirrors cancer mutational signatures linked to oxidative stress. The chemotherapy drug irofulven preferentially induces depurination in ApA dimers and promoters. Notably, oxidized guanines and apurinic sites, both irofulven induced and endogenous, are depleted in gene transcribed strands, with the strand bias increasing with gene expression. This work substantially advances click-code-seq for deciphering the impacts of key modifications in human DNA on cellular physiology and toxicological responses. Bioinformatics Cancer therapy Chemical modification DNA DNA damage and repair
N Nature Chemical Biology · Oct 31, 2025 A gasdermin-based life–death evolution system for reprogramming protease specificity Reprogramming the specificity of proteases toward alternative target sequences could enable an array of exciting applications, ranging from proteome editing to therapeutic interventions. Here we report an in vivo life–death selection system for protease reprogramming using the toxic N-terminal domain of gasdermin D (GD-N) protein as a selection marker. The approach is a modular system that can be used to cover the protease mutational diversity in the billions through only a few cycles of directed evolution. By inserting the desired cleavage sequence into the loop region of the GD-N protein, which is toxic to host bacteria cells, the system selects for efficient substrate cleavage—rendering GD-N nontoxic—by enrichment of bacteria in liquid culture. Using the tobacco etch virus protease (TEVp) and corresponding substrate sequence as a model, we demonstrated that our platform could select and enrich an efficient protease variant millionfold after a single round of selection. We also evolve TEVp to cut sequences on target proteins with known pathological roles. Enzymes Proteins
N Nature Chemical Biology · Oct 29, 2025 Carbon-monoxide-driven bioethanol production operates through a tungsten-dependent catalyst Microbial alcohol production from waste gases is a game changer for sustainable carbon cycling and remediation. While the biotechnological process usingClostridiumautoethanogenumto transform syngas (H2, CO2and CO) is blooming, scientific debates remain on the ethanol biosynthesis pathway. Here, we experimentally validated that ethanol production is initiated through a tungsten-dependent aldehyde:ferredoxin oxidoreductase (AFOR), which reduces acetate to acetaldehyde. The reaction, thermodynamically unfavorable under standard conditions, has been considered by many as unsuitable in vivo but is rather approved by metabolic modeling. To answer this riddle, we demonstrated that the thermodynamic coupling of CO oxidation and ethanol synthesis allows acetate reduction. The experiments, performed with native CO dehydrogenase and AFOR, highlighted the key role of ferredoxin in stimulating the activity of both metalloenzymes and electron shuttling. The crystal structure of holo AFOR, refined to 1.59-Å resolution, and its biochemical characterization provide new insights into the cofactor chemistry and the specificities of this enzyme, fundamental to sustainable biofuel production. Biocatalysis Enzymes Industrial microbiology Metabolic pathways X-ray crystallography
N Nature Chemical Biology · Oct 23, 2025 MED1 IDR deacetylation controls stress responsive genes through RNA Pol II recruitment Cells fine-tune gene expression in response to cellular stress, a process critical for tumorigenesis. However, mechanisms governing stress-responsive transcription remain incompletely understood. This study shows that the MED1 subunit of the Mediator coactivator complex is acetylated in its intrinsically disordered region (IDR). Under stress, SIRT1 associates with the super elongation complex to deacetylate MED1 in promoter-proximal regions. The deacetylated (or acetylation-defective mutant) MED1 amplified stress-activated cytoprotective genes and rescued stress-suppressed growth-supportive genes in estrogen-receptor-positive breast cancer (ER+BC) cells. Mechanistically, deacetylated MED1 promotes chromatin incorporation of RNA polymerase II (Pol II) through IDR-mediated interactions. Functionally, ER+BC cells with deacetylated MED1 exhibit faster growth and enhanced stress resistance in culture and in an orthotopic mouse model. These findings advance our understanding of Pol II regulation under cellular stress and potentially suggest therapeutic strategies targeting oncogenic transcription driven by MED1 and Mediator. Cancer Chromatin Post-translational modifications Transcription
N Nature Chemical Biology · Oct 17, 2025 Site-resolved energetic information from HX–MS experiments High-resolution energetic information about protein conformational ensembles is essential for understanding protein function, yet remains challenging to obtain. Here we present PIGEON-FEATHER, a method for calculating ensemble free energies of opening (∆Gop) at single-amino-acid or near-single-amino-acid resolution for proteins of all sizes from hydrogen exchange–mass spectrometry (HX–MS) data. PIGEON-FEATHER disambiguates and reconstructs all experimentally measured HX–MS isotopic mass envelopes using a Bayesian Monte Carlo sampling approach. We applied PIGEON-FEATHER to reveal howEscherichiacoliand human dihydrofolate reductases (ecDHFR and hDHFR) have evolved distinct ensembles. We show how two competitive inhibitors bind these orthologs differently, solving the longstanding mystery of why both therapeutic molecules inhibit ecDHFR but only one inhibits hDHFR. Extending PIGEON-FEATHER to a large protein–DNA complex, we mapped ligand-induced ensemble reweighting in theE.colilacrepressor to describe the functional switching mechanism crucial for transcriptional regulation. Biophysics Proteins
N Nature Chemical Biology · Oct 16, 2025 A small molecule targets LIC1 to suppress lung tumor growth by inducing autophagy Small molecules that induce autophagy in specific biological contexts can provide invaluable chemical probes and potential anticancer therapeutics. Here we identified a potent autophagy inducer 3,4-diisobutyryl derivative of auxarthrol A (DAA) from an endophyte-derived small-molecule library. DAA demonstrates notable antitumor efficacy in non-small cell lung cancer (NSCLC) tumor and sensitizes tumors to anti-PD1 immunotherapy. Using a photoaffinity labeling approach, we identified light intermediate chain 1 (LIC1), a subunit of dynein, as the direct target of DAA. We found that LIC1 is overexpressed in NSCLC tumors and correlates with poor survival. Mechanistically, the targeting of LIC1 by DAA markedly disrupts the interactions between LIC1 and stress-sensing effector RuvB-like AAA ATPase 1, which in turn elevates downstream GCN2–eIF2α–ATF4 axis-mediated integrated stress response, ultimately promoting autophagic cell death. Our findings define LIC1 as a novel therapeutic target for NSCLC and highlight the potential of DAA as a promising autophagy inducer for treatment of this disease. Cancer therapy Cell death Cell signalling Natural products Target identification
N Nature Chemical Biology · Oct 16, 2025 A color-tailored fluorogenic sensor for hydrogen peroxide H2O2is an important signaling molecule and redox regulator of normal cellular metabolism and a major element of oxidative stress. Here we report HyPerFLEX (HyPer with flexible fluorogen excitation), a sensor from the HyPer family designed for high-precision H2O2monitoring in living cells. HyPerFLEX combines the redox-sensitive OxyR domain fromNeisseriameningitidisand circularly permuted fluorogenic protein Y-FAST, yielding oxygen-independent fluorescence upon oxidation of OxyR by H2O2. HyPerFLEX enables imaging H2O2dynamics in living cells, with tunable spectra from green to far red for multicompartment imaging, even under prolonged hypoxia. It surpasses HyPer7 in detecting ultralow H2O2concentrations, such as during early glucose-stimulated insulin production, and can measure H2O2levels in the highly oxidizing endoplasmic reticulum lumen. These advanced features and broad compatibility make HyPerFLEX a powerful tool for studying oxidative stress and cellular signaling. Chemical tools Microscopy Single-molecule biophysics Synthetic biology
N Nature Chemical Biology · Oct 15, 2025 Molecular basis for noncanonical transcription initiation from Np4A alarmones Stress-induced dinucleoside tetraphosphates (Np4Ns, where N is adenosine, guanosine, cytosine or uridine) are ubiquitous in living organisms, yet their function has been largely elusive for over 50 years. Recent studies have revealed that RNA polymerase can influence the cellular lifetime of transcripts by incorporating these alarmones into RNA as 5′-terminal caps. Here we present structural and biochemical data that reveal the molecular basis of noncanonical transcription initiation from Np4As byEscherichia coliandThermus thermophilusRNA polymerases. Our results show the influence of the first two nucleotide incorporation steps on capping efficiency and the different interactions of Np4As with transcription initiation complexes. These data provide critical insights into the substrate selectivity that dictates levels of Np4capping in bacterial cells. RNA metabolism X-ray crystallography
N Nature Chemical Biology · Oct 09, 2025 Boolean logic-gated protein presentation through autonomously compiled molecular topology Stimulus-responsive materials have enabled advanced applications in biosensing, tissue engineering and therapeutic delivery. Although controlled molecular topology has been demonstrated as an effective route toward creating materials that respond to prespecified input combinations, prior efforts suffer from a reliance on complicated and low-yielding multistep organic syntheses that dramatically limit their utility. Harnessing the power of recombinant expression, we integrate emerging chemical biology tools to create topologically specified protein cargos that can be site-specifically tethered to and conditionally released from biomaterials following user-programmable Boolean logic. Critically, construct topology is autonomously compiled during expression through spontaneous intramolecular ligations, enabling direct and scalable synthesis of advanced operators. Using this framework, we specify protein release from biomaterials following all 17 possible YES/OR/AND logic outputs from input combinations of three orthogonal protease actuators, multiplexed delivery of three distinct biomacromolecules from hydrogels, five-input-based conditional cargo liberation and logically defined protein localization on or within living mammalian cells. Biomaterials Proteins
N Nature Chemical Biology · Oct 08, 2025 Efficient and precise inversion of genomic DNA from large to chromosomal scale Chromosomal inversion is a key structural variation impacting cellular fitness and genomic integrity. Here we developed prime-editing-based inversion with enhanced performance (PIE) to efficiently induce large-scale inversions in mammalian cells. PIEv1 uses a prime-editing guide RNA (pegRNA) pair but yields one imprecise junction. PIEv2 and PIEv3 add a second pegRNA pair for precise inversion, with PIEv3b further enhancing coupling precise inversion through improved plasmid design. PIEv3b achieves inversion efficiencies up to 61.7% for 1 Mb and 14.2% for 50 Mb segments and shows 4–20-fold higher efficiency compared to twin prime editing with integrase, across ranges of 100 kb to 30 Mb. Additionally, PIEv3b outperforms nuclease-based approaches in both inversion efficiency and precision. Using PIE, we convert human chromosomes from metacentric to telocentric configurations by inverting 30-Mb and 100-Mb chromosomal segments. Our work represents a powerful tool for engineering chromosomal structural variations, with broad implications for medicine and biotechnology. Biological techniques Genetics
N Nature Chemical Biology · Oct 08, 2025 Stabilization of AFF1 by PARylation ensures transcriptional restart after DNA damage Precise control of transcription is essential for cell survival under stress conditions, including DNA damage. While mechanisms of DNA damage-induced transcriptional silencing are well characterized, how transcription resumes remains less understood. Here we identify a new role for poly(ADP-ribose) polymerase 1 (PARP1) in transcriptional restart during the DNA damage response (DDR) through a mechanism termed poly(ADP-ribose)-mediated stabilization (PARSTA) of AFF1. Upon DNA damage, PARP1 binds to and PARylates AFF1 in a region targeted by the E3 ligase Siah1, preventing AFF1 ubiquitination and promoting its stability. This stabilization supports efficient transcriptional recovery after DNA damage. Notably, cells resistant to genotoxic stress exhibit elevated PARP1 activity and AFF1 levels, while AFF1 depletion impairs DNA repair and survival. Together, these findings expand PARP1’s role to the transcriptional recovery phase in DDR and suggest that targeting the PARSTA pathway may offer therapeutic potential in diseases characterized by hyperactive PARP1 and elevated levels of AFF1. Post-translational modifications Transcription
N Nature Chemical Biology · Oct 06, 2025 Large-scale mining of plant genomes unlocks the diversity of oxidosqualene cyclases The differential cyclization and rearrangement of 2,3-oxidosqualene controlled by oxidosqualene cyclases (OSCs) represents one of the most complex single enzyme transformations in nature and gives rise to a vast array of triterpenoid diversity in the plant kingdom. Here we systematically mine 599 plant genomes representing 387 species and investigate OSC diversity across different plant lineages. From the OSC sequences identified, 20 were selected for functional evaluation. Through analysis of these enzymes, we discover product profiles within clades previously believed to be functionally conserved and OSCs producing triterpenes for which no enzymatic source was known. We also discover OSCs with product profiles that yield mechanistic insights into the control of specific reaction pathways. Our study reveals lineage-specific blooms of OSC subgroups suggestive of adaptation to different environmental niches, opens up previously inaccessible chemistry and provides a framework for systematic investigations of metabolic diversification and underlying enzymatic mechanisms in the plant kingdom. Biochemistry Biosynthesis Enzymes Plant sciences
N Nature Chemical Biology · Oct 02, 2025 Cell-selective multiplexed bioorthogonal noncanonical amino acid tagging for nascent proteomics Cell-selective bioorthogonal noncanonical amino acid tagging (BONCAT) enables time-resolved characterization of newly synthesized proteins using engineered aminoacyl-tRNA synthetases. In bacteria, this is typically accomplished using an engineered methionyl-tRNA synthetase (MetRS-NLL). Here we substantially expand the scope of this technology by introducing an engineered tyrosyl-tRNA synthetase (EcTyrRS) and a tryptophanyl-tRNA synthetase (EcTrpRS). These enzymes enable the ultrafast proteome tagging at much lower expression levels than MetRS-NLL, thereby improving both time resolution and robustness in nonmodel bacteria. Additionally, both enzymes can incorporate multiple different noncanonical amino acids with distinct click handles. This enabled new multiplexing capabilities such as distinct tagging of the nascent proteome generated in a cell in response to different cues, and tighter temporal control through pulse-chase BONCAT. EcTyrRS and EcTrpRS are also mutually orthogonal, enabling distinct proteome tagging of different cell types in mixed populations. We therefore demonstrate the utility of this technology inEscherichiacoliand nonmodel ESKAPE pathogens. Chemical modification Proteomics
N Nature Chemical Biology · Sep 29, 2025 IDR-induced CAR condensation improves the cytotoxicity of CAR-Ts against low-antigen cancers Chimeric antigen receptor (CAR)-T cell therapies have shown remarkable efficacies for treating otherwise intractable cancers. However, current clinically approved CAR-T therapies are limited by low antigen sensitivity, impeding their efficacy against cancers with low antigen expression. Here, to address this issue, we engineered CARs targeting CD19, CD22 and HER2 by including intrinsically disordered regions (IDRs) that promote signaling condensation. We discovered that the CAR fused with an IDR from FUS, EWS or TAF15 promoted the formation of CAR-T conjugation with cancer targets, the mechanical strength of CAR-T synapses and membrane-proximal signaling, which led to an increased release of cytotoxic factors and a higher killing activity toward low-antigen-expressing cancer cells in vitro. Moreover, the FUS IDR CAR-T induced improved antitumor effects in both blood cancer and solid tumor models. No spontaneous activation in the absence of antigen was observed. Together, our work demonstrates IDRs as a new toolset for improving CAR-T function through inducing biomolecular condensation. Cell signalling Immunology
N Nature Chemical Biology · Sep 26, 2025 Molecular basis for the regulation of membrane proteins through preferential lipid solvation The mechanism by which lipids regulate membrane proteins remains an open question. While many protein structures reveal associated lipids, neither binding nor regulatory mechanisms can be gleaned from frozen static snapshots, as these processes occur in the context of a dynamic membrane at equilibrium. In this study, we combine single-molecule experiments with computational analyses of lipid dynamics and lipid-solvation energetics to understand how changes in the lipid composition of the membrane influence the dimerization of the CLC-ec1 chloride/proton antiporter. We find this influence does not result from long-lived lipid binding at specific sites, but instead from an inherently dynamic effect known as preferential lipid solvation, which ultimately determines the relative thermodynamic stability of associated and dissociated dimers. This study provides a foundation for linking lipid composition to the modulation of membrane protein conformational equilibria and a framework for discriminating among different lipid regulation mechanisms in membranes. Biophysics Computational biology and bioinformatics Single-molecule biophysics
N Nature Chemical Biology · Sep 24, 2025 Engineering synthetic agonists for targeted activation of Notch signaling Notch signaling regulates cell fate decisions and has context-dependent tumorigenic or tumor suppressor functions. Although there are several classes of Notch inhibitors, the mechanical force requirement for Notch activation has hindered attempts to generate soluble agonists. To address this problem, we engineered synthetic Notch agonists (SNAGs) by tethering affinity-matured Notch ligands to proteins that internalize their targets. This bispecific format enables SNAGs to ‘pull’ on mechanosensitive Notch receptors, triggering their activation in the presence of desired biomarkers. We successfully developed SNAGs targeting six independent surface markers, including the tumor antigens PDL1, CD19 and HER2 and the immunostimulatory receptor CD40. HER2-SNAGs and CD19-SNAGs increased the expression of T cell activation markers and Notch target genes in cocultures with tumor cells, highlighting their potential for immunotherapeutic applications. These insights have broad implications for the pharmacological activation of mechanoreceptors and will expand our ability to modulate Notch signaling in biotechnology. Biologics Immunology Molecular biology
N Nature Chemical Biology · Sep 18, 2025 Enhancing RNA base editing on mammalian transcripts with small nuclear RNAs Endogenous uridine-rich small nuclear RNAs (U snRNAs) form RNA–protein complexes to process eukaryotic pre-mRNA into mRNA. Previous studies have demonstrated programmable U snRNA guide-targeted exon inclusion and exclusion. Here we investigated whether snRNAs can also enhance RNA base editing over state-of-the-art RNA-targeting technologies in human cells. Compared with adenosine deaminase acting on RNA (ADAR)-recruiting circular RNAs, we find that guided A>I snRNAs consistently increase adenosine-to-inosine editing for higher exon count genes, perturb substantially fewer off-target genes and localize more persistently to the nucleus where ADAR is expressed. A>I snRNAs also more efficiently edit long noncoding RNAs and pre-mRNA 3′ splice sites to promote splicing changes. Lastly, snRNA–H/ACA box snoRNA fusions (U>Ψ snRNAs) increase targeted RNA pseudouridylation without DKC1 overexpression, facilitating improvedCFTRrescue from nonsense-mediated mRNA decay in a cystic fibrosis human bronchial epithelial cell model. Our results advance the endogenous protein-mediated RNA base editing toolbox and RNA-targeting technologies to treat genetic diseases. Chemical modification Non-coding RNAs RNA
N Nature Chemical Biology · Sep 15, 2025 Leaflet-specific phospholipid imaging using genetically encoded proximity sensors The lipid composition of cells varies widely across organelles and between individual membrane leaflets. Transport proteins are thought to generate this heterogeneity, but measuring their functions in vivo has been hampered by limited tools for imaging lipids at relevant spatial resolutions. Here we present fluorogen-activating coincidence encounter sensing (FACES), a chemogenetic tool capable of quantitatively imaging subcellular lipid pools and reporting their transbilayer orientation in living cells. FACES combines bioorthogonal chemistry with genetically encoded fluorogen-activating proteins (FAPs) for reversible proximity sensing of conjugated molecules. We first apply this approach to identify roles for lipid transfer proteins that traffic phosphatidylcholine pools between the ER and mitochondria. We then show that transmembrane domain-containing FAPs can reveal the membrane asymmetry of multiple lipid classes in thetrans-Golgi network and be used to investigate the mechanisms that generate it. Finally, we present that FACES can be applied to measure glycans and other molecule classes. Chemical tools Glycobiology Membrane trafficking Membranes Phospholipids
N Nature Chemical Biology · Sep 15, 2025 ELOVL6 activity attenuation induces mutant KRAS degradation KRASis one of the most frequently mutated oncogenes in cancer. Targeting mutant KRAS directly has been challenging because of minor structural changes caused by mutations. Despite recent success in targeting KRAS-G12C, targeted therapy for another hotspot mutant, KRAS-G12V, has not been described. We used CRISPR–Cas9 genome-wide knockout screens to identify genes that specifically modulate mutant KRAS harboring the G12V substitution. Our top hit, a fatty acid elongase (ELOVL6), showed remarkable selectivity in diminishing KRAS-G12V protein expression and aberrant oncogenic signaling associated with mutant KRAS. Our studies reveal that ELOVL6 can be targeted to control the production of phospholipids exploited by KRAS mutants for function-targeted and trigger-targeted degradation of the protein. Our results demonstrate the basis for a first-in-class small-molecule inhibitor to selectively clear KRAS-G12V from cancer cells. Cancer therapy Cell signalling Membrane lipids Screening
N Nature Chemical Biology · Sep 12, 2025 A Ψ–Ψ codon–anticodon pairing in nonsense suppression and translational recoding Pseudouridine (Ψ) is known for decades but its flexibility in base pairing remains unclear. This study engineers artificial box H/ACA guide RNAs to direct pseudouridylation at the uridine of a premature termination codon (PTC; UAA, UAG or UGA) within an intronless mRNA and U36 of the anticodon of a matching tRNA in yeast and human cells. Targeted pseudouridylation leads to the formation of a Ψ–Ψ codon–anticodon pair, which, together with the other two Watson–Crick base pairs in the codon–anticodon duplex, greatly improves codon–anticodon recognition, robustly promoting PTC readthrough. The intronless mRNA level remains unchanged with or without guide RNAs. Additionally, pseudouridylation does not impact tRNA stability or charging. Our results show that nonsense suppression is promoted by the high affinity of the Ψ–Ψ pair, which is verified by melting curve analysis. This work identifies an unusual Ψ–Ψ base pair, which contributes greatly to codon–anticodon recognition and translational recoding. Biological techniques Genetics Molecular biology Translation
N Nature Chemical Biology · Sep 11, 2025 Real-time imaging of protein microenvironment changes in cells with rotor-based fluorescent amino acids Fluorescent protein fusions with environmentally sensitive fluorophores have been widely used to investigate changes in the protein microenvironment. Unfortunately, these techniques often rely on bulky fluorescent proteins or tags to the N terminus or C terminus of the target protein, which can disrupt the behavior of the target protein and may limit their ability to investigate microenvironment changes with high spatial resolution. Here we develop a strategy to visualize microenvironment changes of protein substructures in real time by genetically incorporating environment-sensitive noncanonical amino acids (ncAAs) containing rotor-based fluorophores at specific positions of the target protein. Through computational redesign of aminoacyl-tRNA synthetase, we successfully incorporated these rotor-based ncAAs into several proteins in mammalian cells. Precise placement of these ncAAs at specific sites of proteins enables the detection of microenvironmental changes around individual residues during events such as aggregation, clustering, cluster dissociation and others. Chemical libraries Chemical tools Imaging Peptides
N Nature Chemical Biology · Sep 09, 2025 Navigating condensate micropolarity to enhance small-molecule drug targeting Many pharmaceutical targets partition into biomolecular condensates, whose microenvironments can significantly influence drug distribution. Nevertheless, it is unclear how drug design principles should adjust for these targets to optimize target engagement. To address this question, we systematically investigated how condensate microenvironments influence drug-targeting efficiency. We found that condensates highlight a notable heterogeneity, with nonpolar-residue-enriched condensates being more hydrophobic and housing more hydrophobic drugs. Furthermore, L1000 dataset analysis revealed a strong positive correlation between inhibitor hydrophobicity and targeting efficiency for phase-separated proteins, represented by estrogen receptor 1 (ESR1) enriched with nonpolar residues. We developed random forest models to predict inhibitor targeting efficiency from molecular properties, with hydrophobicity identified as a key determinant. In cellulo experiments with ESR1 condensates confirmed that both binding affinity and hydrophobicity of inhibitors contribute significantly to potency. These results suggest a new drug design principle for phase-separated proteins by considering condensate micropolarity, potentially leading to drugs with optimal target engagement. Chemical biology Drug discovery Proteins
N Nature Chemical Biology · Aug 29, 2025 ALFA nanobody-guided endogenous labeling Small peptide tags offer advantages as their compact size reduces target protein interference, making them valuable for labeling endogenous proteins. However, the lack of inherent fluorescence poses challenges for post-genome knockin monoclonal clone screening. Here we report an adaptable approach leveraging antigen-stabilizing fluorescent protein-fused nanobodies (Nbs) to selectively illuminate cells with successful ALFA tag knockins, streamlining high-throughput cell screening using fluorescence-activated cell sorting. Through targeted mutations and screening of ALFA Nbs (NbALFA), the fluorescently labeled Nb can be selectively degraded in the absence of the ALFA peptide. Conversely, successful insertion of the ALFA peptide into the genome results in a substantial increase in the fluorescence intensity of the Nb. This technique, termed ALFA Nb-guided endogenous labeling (ANGEL), enables a wide array of versatile applications within the native cellular environment. These applications include precise protein labeling and signal amplification through the tandem arrangement of ALFA tags, dynamic monitoring of protein behavior, initiation of protein degradation processes and analysis of protein interactome. Biological techniques Molecular biology
N Nature Chemical Biology · Aug 29, 2025 Designing small molecules targeting a cryptic RNA binding site through base displacement Most RNA-binding small molecules have limited solubility, weak affinity and/or lack of specificity, restricting the medicinal chemistry often required for lead compound discovery. We reasoned that conjugation of these unfavorable ligands to a suitable ‘host’ molecule can solubilize the ‘guest’ and deliver it site-specifically to an RNA of interest to resolve these issues. Using this framework, we designed a small-molecule library that was hosted by cobalamin (Cbl) to interact with the Cbl riboswitch through a common base displacement mechanism. Combining in vitro binding, cell-based assays, chemoinformatic modeling and structure-based design, we unmasked a cryptic binding site within the riboswitch that was exploited to discover compounds that have affinity exceeding the native ligand, antagonize riboswitch function or bear no resemblance to Cbl. These data demonstrate how a privileged biphenyl-like scaffold effectively targets RNA by optimizingπ-stacking interactions within the binding pocket. RNA Structure-based drug design X-ray crystallography
N Nature Chemical Biology · Aug 27, 2025 Mass spectrometry and enzyme assays refute histone tyrosine sulfation arising fromW. Yu et al.Nature Chemical Biologyhttps://doi.org/10.1038/s41589-023-01267-9(2023) Mass spectrometry Post-translational modifications
N Nature Chemical Biology · Aug 27, 2025 Real-time visualization of STAT activation in live cells using genetically encoded biosensors The signal transducer and activator of transcription (STAT) family proteins are attractive drug targets but tools to monitor their activation are lacking. Now, STAT biosensors have been developed for real-time tracking in live cells and are applied to screen inhibitors and investigate the effects of cancer-associated mutations.
N Nature Chemical Biology · Aug 27, 2025 Reply to: Mass spectrometry and enzyme assays refute histone tyrosine sulfation replying toM. M. Youssef et al.Nature Chemical Biologyhttps://doi.org/10.1038/s41589-025-01994-1(2025) Chemical biology Chemical modification
N Nature Chemical Biology · Aug 27, 2025 pH variations enable guanine crystal formation within iridosomes Many animals display brilliant colors thanks to the precise formation of guanine crystals within specialized organelles. Here, the authors demonstrate that dynamic pH shifts orchestrate this process: an initially acidic lumen stabilizes amorphous, protonated guanine and subsequent alkalinization triggers its crystallization.
N Nature Chemical Biology · Aug 26, 2025 Oncometabolite 5-IP7inhibits inositol 5-phosphatase to license E-cadherin endocytosis E-cadherin downregulation is an epithelial–mesenchymal transition hallmark canonically attributed to transcriptional repression. Here we delineate a metabolite-driven endocytic route of E-cadherin downregulation in inflammation-associated colorectal cancer (CRC). Specifically, IP6kinase-2 (IP6K2), a 5-diphosphoinositol pentakisphosphate (5-IP7) synthase upregulated in patients with CRC, is activated via a ROS–Src phosphorylation axis elicited by dextran sulfate sodium (DSS), generating 5-IP7around adherens junction (AJ) to promote E-cadherin endocytosis and the transcriptional activities of β-catenin. Mechanistically, 5-IP7inhibits inositol 5-phosphatases such as OCRL to promote PI(4,5)P2-mediated endocytic adaptor recruitment. Depleting 5-IP7or overexpressing a 5-IP7binding-deficient OCRL mutant confers resistance to DSS-elicited AJ disruption. Intestinal epithelium-specific IP6K2 deletion attenuates DSS-induced colitis/CRC, whereas an IP6K2 isoform-selective inhibitor protects wild-type but not IP6K2−/−mice against DSS insult. Thus, 5-IP7is an oncometabolite whose stimulus-dependent synthesis relieves a PI(4,5)P2dephosphorylation-based endocytic checkpoint, leading to AJ disassembly and protumorigenic β-catenin activation. Targeting IP6K2 could strengthen intestinal epithelial barrier against inflammation and cancer. Cancer Cell signalling Kinases Membrane trafficking Post-translational modifications
N Nature Chemical Biology · Aug 22, 2025 SARM1 activation promotes axonal degeneration via a two-step phase transition SARM1 is a key executioner of axonal degeneration, acting through NAD⁺ depletion by NADase activity of its TIR domain. Although normally autoinhibited, SARM1 becomes activated in response to axonal damage; however, the underlying mechanism remains unclear. Here, using a class of pyridine-containing compounds that trigger SARM1-dependent axon degeneration, we uncover a two-step activation process. First, NMN primes the base exchange activity of SARM1, generating covalent adducts between ADP-ribose (an NAD⁺ hydrolysis product) and the compounds. These ADP-ribose conjugates then serve as molecular glues to promote the assembly of superhelical SARM1 filaments, in which TIR domains adopt an active configuration. After reaching solubility limits, these filaments condense into stable, phase-separated assemblies with full enzymatic activity. Unexpectedly, several clinical-stage SARM1 inhibitors targeting its TIR domain also form such adducts, paradoxically promoting its activation. These findings reveal a molecular mechanism that spatially restricts SARM1 activation to damaged axons and offer new guidance for therapeutic strategies targeting SARM1. Enzyme mechanisms Molecular neuroscience
N Nature Chemical Biology · Aug 22, 2025 α-Ketoglutarate dictates AMPK protein synthesis for energy sensing in human cancers The energy sensor AMP-activated protein kinase (AMPK) promotes tumor cell survival under stress but how to prevent AMPK activation to blunt tumor progression remains unclear. Here we show that the metabolite α-ketoglutarate (α-KG) dictates AMPK translation through a TET–YBX1 axis, which can be exploited to sensitize human cancer cells to energy stress. α-KG-deficient cells fail to activate AMPK under glucose starvation, which elicits cytosolic NADPH depletion and disulfidptosis. Mechanistically, α-KG insufficiency inhibits TET-dependent transcription of YBX1, an RNA-binding protein required for human-specific AMPK protein synthesis. Similarly, α-KG competitors including succinate and itaconate inhibit the YBX1–AMPK axis and sensitize cancer cells to glucose deprivation. Lastly, cotargeting oncogenic YBX1 and GLUT1 creates synthetic lethality and blunts tumor growth in vivo. Together, our findings link α-KG to energy sensing through AMPK translation and propose that targeting α-KG–YBX1-dependent AMPK translation can sensitize human cancer cells to energy stress for treatment. Cancer therapy Kinases Metabolic pathways Transcription Translation
N Nature Chemical Biology · Aug 22, 2025 KLHL23 and RhoGDI coordinate CDC42 inactivation ensuring membrane homeostasis F-Actin cytoskeleton remodeling is vital for cell migration, organ development and immune responses. The small GTPase CDC42, a key regulator of F-actin dynamics, cycles between inactive GDP- and active GTP-bound states. However, mechanisms governing CDC42 turnover and their biological significance remain unclear. Here we show that KLHL23-mediated polyubiquitylation of CDC42•GTP and RhoGDI-mediated sequestration of CDC42•GDP spatiotemporally co-inactivate CDC42, preserving membrane dynamics and homeostasis during migration. KLHL23–Cul3 acts as the E3 ligase for CDC42 degradation, with KLHL23 and RhoGDI competing for CDC42’s switch II region, enhancing selectivity toward CDC42•GTP and CDC42•GDP, respectively. KLHL23 depletion disrupts membrane homeostasis, inducing excessive protrusions and promoting metastasis. Notably, the CDC42-Y64C germline variant in Takenouchi–Kosaki Syndrome escapes KLHL23-mediated degradation. Fluorescence resonance energy transfer assays reveal that KLHL23 and RhoGDI coordinately inactivate CDC42 in a spatiotemporal manner. These findings highlight the biological and clinical relevance of the KLHL23/RhoGDI–CDC42 axis, presenting new avenues for therapeutic exploration. Membrane trafficking Neurological disorders Post-translational modifications
N Nature Chemical Biology · Aug 22, 2025 Interactions with tau’s microtubule-binding repeats modulate amyloid-β aggregation and toxicity The complicated pathogenesis of Alzheimer’s disease (AD) is characterized by the accumulation of neurofibrillary tangles and senile plaques, primarily composed of tau and amyloid-β (Aβ) aggregates, respectively. While substantial efforts have focused on unraveling the individual roles of tau and Aβ in AD development, the intricate interplay between these components remains elusive. Here we report how the microtubule-binding repeats of tau engage with Aβ in a distinct manner. Crucially, this interaction notably modifies Aβ aggregation behavior, thereby altering Aβ-associated toxicity within both extracellular and intracellular milieus. Our mechanistic investigations at the molecular level manifest specific fragments within tau’s microtubule-binding domain that possess a balance of hydrophobic and hydrophilic properties, promoting the formation of hetero-adducts with Aβ peptides. These findings offer avenues for understanding and treating AD by emphasizing the tau–Aβ interplay in the pathogenesis. Biophysical chemistry Protein aggregation
N Nature Chemical Biology · Aug 22, 2025 Conversion of natural cytokine receptors into orthogonal synthetic biosensors Synthetic receptors enable bioengineers to build cell-based therapies that perform therapeutic functions in a targeted or conditional fashion to enhance specificity and efficacy. Although many synthetic receptors exist, it remains challenging to generate new receptors that sense soluble cues and relay that detection through orthogonal mechanisms independent of native pathways. Here we co-opt natural cytokine receptor ectodomains into modular extracellular sensor architecture (MESA) receptors to form natural ectodomain (NatE) MESA receptors. We generated multiple functional, orthogonal synthetic cytokine receptors, identified design principles and constraints and propose guidance for extending this approach to other natural receptors. We demonstrate the utility of NatE MESA by engineering T cells to sense an immunosuppressive cue and respond with customized transcriptional output to support chimeric antigen receptor T cell activity. Lastly, we multiplex NatE MESA to logically evaluate multiple cues associated with the tumor microenvironment. These technologies and learnings will enable engineering cellular functions for new applications. Protein design Synthetic biology
N Nature Chemical Biology · Aug 21, 2025 Highly specific intracellular ubiquitination of a small molecule Ubiquitin is a small, highly conserved protein that acts as a posttranslational modification in eukaryotes. Ubiquitination of proteins frequently serves as a degradation signal, marking them for disposal by the proteasome. Here we report a novel small molecule from a diversity-oriented synthesis library, BRD1732, that is directly ubiquitinated in cells, resulting in dramatic accumulation of inactive ubiquitin monomers and polyubiquitin chains, which causes broad inhibition of the ubiquitin–proteasome system. Ubiquitination of BRD1732 and its associated cytotoxicity are stereospecific and dependent on two homologous E3 ubiquitin ligases, RNF19A and RNF19B, and their shared E2 conjugating enzyme, UBE2L3. Our finding opens the possibility for indirect ubiquitination of a target through a ubiquitinated bifunctional small molecule and more broadly raises the potential for posttranslational modification intrans. Chemical modification Chemical tools Mechanism of action
N Nature Chemical Biology · Aug 21, 2025 Site-selective protein editing by backbone extension acyl rearrangements Protein and polypeptide heteropolymers containing non-α-backbone monomers are highly desirable as potential materials and therapeutics but many remain difficult or impossible to biosynthesize in cells using traditional genetic code expansion. Here we describe a next-generation approach to such materials that relies instead on proximity-guided intramolecular rearrangements that edit the protein backbone post-translationally. This approach relies on orthogonal aminoacyl-tRNA synthetase enzymes that accept α-hydroxy acid monomers whose side chains contain masked nucleophiles. Introduction of such an α-hydroxy acid into a protein translated in vivo, followed by nucleophile unmasking, sets up a thermodynamically favored intramolecular backbone extension acyl rearrangement (BEAR) reaction that edits the protein to install an extended-backbone monomer. In the examples described here, BEAR reactions are used to generate protein heteropolymers containing a β-backbone, γ-backbone or δ-backbone. This report represents a general strategy to install extended backbones into genetically encoded proteins and peptides expressed in cells. Proteins Synthetic biology
N Nature Chemical Biology · Aug 15, 2025 Structural basis for E4 enzyme Ufd2-catalyzed K48/K29 branched ubiquitin chains E4 enzymes amplify and remodel ubiquitin chain signals beyond the conventional E1–E2–E3 cascade. The first identified E4 enzyme Ufd2 preferentially catalyzes K48/K29 branched ubiquitin chains, yet the structural mechanism remains unknown. Here, we combined chemical biology and cryo-electron microscopy to visualize stable intermediates in Ufd2 loading ubiquitin at K48 of proximal ubiquitin on K29-linked di- and triubiquitin. Our data reveal that the core region of Ufd2 functions as an unprecedented K29 diubiquitin binding domain, interacting extensively with proximal and distal ubiquitin, which orients the K48 site of proximal ubiquitin toward the active site of Ubc4, facilitating K48/K29 branched ubiquitin chain formation. We also identified a unique dimeric conformation where dimerized Ufd2 and Ubc4 stabilize each other’s distal ubiquitin during branching on K29 triubiquitin. Our findings provide mechanistic insights into the assembly of K48/K29 branched ubiquitin chains by the E4 enzyme Ufd2 and highlight the spatial cooperation among multiple pairs of ubiquitin-related enzymes on longer ubiquitin chains. Chemical modification Enzyme mechanisms Post-translational modifications
N Nature Chemical Biology · Aug 15, 2025 Covalently constrained ‘Di-Gembodies’ enable parallel structure solutions by cryo-EM Whilst cryo-electron microscopy(cryo-EM) has become a routine methodology in structural biology, obtaining high-resolution cryo-EM structures of small proteins (<100 kDa) and increasing overall throughput remain challenging. One approach to augment protein size and improve particle alignment involves the use of binding proteins or protein-based scaffolds. However, a given imaging scaffold or linking module may prove inadequate for structure solution and availability of such scaffolds remains limited. Here, we describe a strategy that exploits covalent dimerization of nanobodies to trap an engineered, predisposed nanobody-to-nanobody interface, giving Di-Gembodies as modular constructs created in homomeric and heteromeric forms. By exploiting side-chain-to-side-chain assembly, they can simultaneously display two copies of the same or two distinct proteins through a subunit interface that provides sufficient constraint required for cryo-EM structure determination. We validate this method with multiple soluble and membrane structural targets, down to 14 kDa, demonstrating a flexible and scalable platform for expanded protein structure determination. Chemical modification Chemical tools Membrane proteins Protein design Structural biology
