Angewandte Chemie

Investigating the conformational dynamics of intrinsically disordered proteins (IDPs) is essential to understanding how their structural heterogeneity underlies function and how their dysregulation contributes to diseases. Here, we utilized an MspA nanopore-based approach for studying the conformational dynamics and interactions of IDPs at the single-molecule level. The platform was demonstrated using the intrinsically disordered transactivation domain of tumor suppressor p53 (p53-TAD), one of the important proteins in cancer biology. We showed that MspA can stably capture p53-TAD and resolve up to six distinct current states with frequent interconversions, revealing a rich conformational landscape. The nanopore also detected the effect of a cancer-associated double mutational variant, N29K/N30D. Combining experiments with steered molecular dynamics simulations, we showed that the mutant sampled compact conformational states more frequently than wild type, consistent with previous NMR studies. Importantly, the MspA platform enabled direct monitoring of E3 ligase MDM2 binding to p53-TAD and resolved how this interaction is inhibited by anti-cancer compound epigallocatechin gallate (EGCG). Notably, EGCG stabilizes one of the six states sampled by p53-TAD, providing a mechanistic explanation for its inhibitory effect. Together, these findings demonstrate the promise of the nanopore platform for label-free monitoring of IDP conformational dynamics, modulation, binding and inhibition at single-molecule resolution.

Methods for installing synthetic functions on living cell surfaces provide powerful approaches for imaging, sensing, and manipulating cell behavior, but many require genetic modification of the target cell or chemical modification of the plasma membrane. Here, we repurpose the glycosylphosphatidylinositol-anchored protein (GPI-AP)-binding toxin aerolysin as a modular chassis for non-genetic cell-surface functionalization. We show that a non-cytotoxic, monomeric aerolysin mutant retains high-affinity and GPI-AP-dependent cell binding when genetically fused to diverse protein cargos. Fluorescent protein-aerolysin fusions robustly label multiple cell types and remain predominantly associated with the cell surface for at least 24 h, in contrast to wheat germ agglutinin, which is extensively internalized. Aerolysin can also be equipped with SpyTag/SpyCatcher to enable modular assembly with independently expressed protein cargos. Importantly, aerolysin supports functional rather than solely optical modification of the cell surface: fusion to the proximity-labeling enzyme APEX2 enables extracellular protein biotinylation, while fusion to HUH endonuclease tags enables covalent attachment of synthetic DNA to living cells. Using this latter architecture, we developed a DNA hairpin sensor that converts cell-surface nuclease activity into a fluorescent signal and distinguishes cells with different levels of extracellular nuclease activity. Together, these results establish non-cytotoxic aerolysin as a genetically encoded, soluble adapter for installing proteins, enzymes, and programmable nucleic acids onto living cells without modification of the target-cell genome.

Obtaining sufficient amounts of pure protein for downstream applications such as native mass spectrometry (nMS) is often challenging, especially when expression yields are low or proteins are unstable. In these cases, the commonly required buffer-exchange step is a major bottleneck, as it often leads to substantial protein loss and compromises biophysical characterization. These challenges are exacerbated in insect or eukaryotic expression systems, where protein yields are typically lower than in bacteria, making protein loss during purification particularly detrimental. Standard lysis and purification buffers contain non-volatile components such as Tris, phosphate, HEPES and sodium chloride, which form adducts during electrospray ionization (ESI) interfering with the signal and therefore must be re-moved prior to nMS. To address protein loss associated with this mandatory buffer-exchange, we evaluated an affinity-purification workflow, in which non-volatile salts are excluded throughout purification and proteins are directly eluted into nMS-compatible ammonium acetate-based buffers. This approach eliminates the need for a separate buffer exchange step and enables rapid nMS analysis immediately after affinity purification. We show that common eluents used in His- and Strep- based affinity purification, such as imidazole, biotin, and desthiobiotin, are well tolerated at relevant concentrations, allowing acquisition of high-quality spectra suitable for determining protein stoichiometry and for monitoring enzymatic or assembly processes. Together, this fast-track affinity workflow increases protein recovery, shortens sample preparation and complements online exchange protocols, which are less suited for monitoring processes. It hence expands the applicability of nMS to proteins and protein complexes that are difficult to obtain in sufficient quantity using conventional purification and buffer exchange strategies.

Nanodiscs and bicelles are widely used as membrane mimetics for structural studies of membrane-associated systems. Studies have reported that their magnetic alignment behavior and phase stability are highly sensitive to composition and temperature. In this study, we systematically investigate the effects of cholesterol on bicelle formation and magnetic alignment in DMPC + 0.2 glycyrrhizic acid (GA) systems using a combined 31P and 14N solid-state NMR experimental and simulation-based approach. Temperature dependent 31P NMR spectra reveal a clear transition from vesicle dominant to aligned bicelles/nanodsics phase, while 1N quadrupolar splitting and lineshape analysis provides quantitative insights into heterogeneous lipid bilayer populations, distinguishing large aligned nanodiscs (B(L)), small nanodiscs (B(S)), and isotropic/random components (B(R)). A strong correlation is observed between the 31P derived bicelle fraction and the 14N B(L) population, confirming that macroscopic alignment in the presence of an external magnetic field directly reflects the growth of large, well-ordered nanodiscs. Cholesterol is found to play a critical dual role by modulating membrane order and curvature. At low cholesterol concentration (0 to 5 mole percent), nanodiscs alignment occurs gradually with increasing temperature, while at higher cholesterol concentration (15 to 25 mole percent), the alignment is delayed and accompanied by broader spectral features, indicating structural heterogeneity. Notably, 10 mole percent cholesterol consistently provides the optimal balance, enabling efficient temperature dependent conversion to aligned bicelles while maintaining high B(L) populations (about 70-80 percentage) and minimal isotropic fractions. In contrast, higher cholesterol maintains significant B(S) and B(R) populations, even at elevated temperature. The 14N quadrupolar coupling (Cq is approximately 8.5 to 9.2 kHz for aligned nanodiscs) remains nearly invariant across compositions, showing that cholesterol does not change local headgroup dynamics but instead redistributes lipid populations. These findings establish a combined 31P and 14N solid -state NMR approach provides a valuable platform for quantitatively correlating membrane structure, dynamics, and alignment, offering practical guidelines for optimizing bicelle systems for high resolution solid-state NMR studies of membrane associated biomolecules.

Lysine acylations, including lactylation (Klac), are pivotal regulators of cellular physiology. However, their analysis is currently bottlenecked by antibody enrichment strategies that suffer from sequence bias and require milligram-scale protein inputs, severely precluding the profiling of scarce clinical biopsies and rare cell populations. Here we present ChemIntelligence, an acyl-NHS chemistry-empowered derivatization strategy that rapidly generates unprecedented acylation-specific spectral libraries, exemplified by over 2.5x10^9 human Klac peptides, enabling cross-species reference atlases. Integrated with Prosit-based rescoring, these libraries substantially increase Klac identifications across diverse proteomic datasets. Leveraging this spectral resource, we devised ChemIntelligence Scope, a reproducible, multiplexed parallel reaction monitoring (PRM) platform that quantifies hundreds of Klac peptides per injection from as little as ~200 ng of cell lysates, clinical biopsies, and even true single cells - revealing functional Klac signatures inaccessible to conventional methods. The ChemIntelligence pipeline also extends seamlessly to lysine nicotinylation, underscoring its broad adaptability for discovering and profiling new acylations. Together, these chemical and computational advances establish a scalable, antibody-free framework for acyl-proteome mapping that overcomes input constraints and enables deep functional insights from otherwise intractable biological samples.

Cereblon (CRBN) is widely used in targeted protein degradation, but its ligand space has remained dominated by a narrow set of cyclic imide chemotypes. Here, we show that the accessible CRBN ligand space extends substantially beyond this canonical space. A high-throughput screen of > 40,000 compounds, followed by orthogonal biophysical validation, X-ray crystallography and SAR analyses, identified several chemically distinct ligand classes. These include linear acetyl-based motifs, a phthalide-derived scaffold, steroidal compounds, and a range of bicyclic ligands. They engage CRBN through distinct recognition modes, several of which deviate from the canonical hydrogen-bonding pattern. Steroidal scaffolds were particularly notable: cortisone binds the human CRBN thalidomide-binding domain with an affinity comparable to thalidomide, with its A-ring occupying the tri-tryptophan pocket in a glutarimide-like orientation despite lacking the canonical imide NH donor. SAR within this series showed substantial tolerance for chemical modification and scaffold simplification, raising the possibility that endogenous steroidal metabolites may contribute to the physiological ligand landscape of CRBN. Bicyclic lactams additionally provided synthetically accessible scaffolds with tunable affinity and promising sites for linker attachment. Across the identified ligand classes, none of the tested representatives induced detectable degradation of canonical CRBN neosubstrates, and several showed largely clean proteomic profiles. Together, these findings broaden the chemical, mechanistic and potential physiological landscape of CRBN recognition and provide diverse starting points for alternative, potentially neosubstrate-sparing CRBN recruiters.

Amyloid fibrils are highly ordered protein aggregates characterized by a conserved cross-{beta}-sheet architecture despite originating from structurally diverse precursor proteins. Growing evidence suggests that interactions between different amyloidogenic proteins can modulate aggregation pathways through heterologous cross-seeding; however, the influence of seed polymorphism on the structure and biological properties of cross-seeded fibrils remains poorly understood. Here, we investigated the cross-seeding of native human insulin by two structurally distinct polymorphs of hen egg-white lysozyme (HEWL): flexible fibrils (FFs) and rigid fibrils (RFs). Native insulin remained stable under physiological conditions and underwent spontaneous fibrillation only under acidic conditions. In contrast, both HEWL polymorphs efficiently induced insulin aggregation at physiological pH, bypassing the nucleation barrier. Thioflavin T fluorescence, circular dichroism spectroscopy, and transmission electron microscopy revealed that lysozyme FFs templated the formation of insulin flexible fibrils (IFFs), whereas lysozyme RFs produced insulin rigid fibrils (IRFs), demonstrating that the structural characteristics of the parental HEWL polymorphs were propagated during heterologous cross-seeding. The toxicity of the resulting insulin fibrils was evaluated in SH-SY5Y neuronal cells and CCF-STTG1 astrocytes. IFFs exhibited minimal cytotoxicity and only subtle morphological alterations, whereas IRFs caused modest reductions in cell viability accompanied by more pronounced cellular damage. These findings demonstrate that the structural polymorphism of HEWL fibrils governs both the architecture and biological activity of cross-seeded insulin fibrils, highlighting amyloid polymorphism as an important determinant of heterologous amyloid propagation and a potential design principle for engineering functional amyloid-based biomaterials and protein delivery platforms.

The immune checkpoint protein PD-L1 plays a pivotal role in tumor immune evasion by binding to PD-1 on immune cells, including T lymphocytes. While the expression and function of PD-L1 have been well studied, the importance of its spatial organization on the cell surface of tumor cells remains poorly understood. In this study, we used super-resolution microscopy combined with biochemical perturbations to investigate the factors regulating PD-L1 clustering and its effects on PD-1 binding and T cell inhibition. We found that PD-L1 is organized into nanoscale clusters at the plasma membrane, with distinct regulatory roles for the actin cytoskeleton, galectin-3, and cholesterol. Disruption of cortical actin increased PD-L1 cluster size, while galectin-3 promoted smaller, denser clusters and increased PD-L1 lateral mobility. Cholesterol depletion reduced PD-L1 cluster size and number and impaired PD-1 binding. These findings indicate that PD-L1 surface organization is collectively regulated by the actin cytoskeleton, galectin-3, and membrane cholesterol within the plasma membrane of tumour cells. Our results provide new insights into the dynamic regulation of PD-L1 and its potential as a therapeutic target in cancer immunotherapy.

Accurate determination of kinase inhibitor binding modes could provide essential information for understanding resistance mechanisms and accelerating drug discovery. While conventional structural methods such as X-ray crystallography, cryo-EM and NMR provide high-resolution information but are low-throughput and capture largely static snapshots of dynamic protein-ligand interactions Here, we introduce a single-molecule nanopore tweezer platform that functionally subtypes ATP-competitive Abl kinase inhibitors by resolving distinct ionic current signatures of Abl-inhibitor complexes. This approach distinguishes Type I, Type IIA, and Type IIB inhibitors without structural determination. We further show how clinically relevant Abl variants (T315I and E255V) reshape inhibitor engagement and binding modes. By combining baseline probability features with wavelet-based time-frequency descriptors, ensemble machine-learning models achieved 97.5% classification accuracy across seven kinase inhibitor binding modes at sub-angstrom resolution and enabled deconvolution of mixed-inhibitor samples at nanomolar concentrations. These results establish nanopore tweezers as a label-free, super-resolution platform for profiling kinase conformational states and inhibitor binding modes, complementing structural approaches and supporting precision oncology.

The binding affinities of a chimeric analog of a riboflavin derivative linked to biotin, (6- (7,8-dimethyl-2,4-dioxo-3,4-dihydrobenzo[g]pteridin-10(2H)-yl)hexyl 5-((3aS,4S,6aR)-2- oxohexahydro-1H-thieno[3,4-d]imidazol-4-yl)pentanoate), referred to as C6-Rf-biotin-tag, to the riboflavin binding retain or streptavidin are in the M range, 1.29 {+/-} 0.277 and 3.00 {+/-} 0.459, respectively. These values suggest that C6-Rf-biotin-tag has potential applications in diagnostic assay and labelling target flavin binding proteins. The C6-Rf-biotin-tag which was characterized with respect to physical and biochemical properties retains UV/Vis spectroscopic and fluorescence behavior similar to riboflavin.

To combat refractory diseases, such as cancer, multitarget-directed ligands (MTDLs) have become an emerging area of research to exploit synthetic lethality (SL) relationships associated with drug resistance. Herein, we present the in silico design of MTDLs for the polypharmaceutical treatment of endometrial adenocarcinoma (EAC) and our discovery of a novel SL in EAC; PTEN loss of function (LOF) and the inhibition of CDK9. We used high-resolution x-ray crystallographic data to chemically engineer, LCI133, to inhibit CDK9, CDK4/6-and AURKA/B kinases. PTEN LOF in EAC results in augmented deregulated transcription and a massive increase in nascent RNA, a phenotype which encodes a high level of apoptotic sensitivity to LCI133 and CDK9 inhibitors. Treatment with LCI133 results in a rapid decline nose-dive in global nRNA, MYC nRNA levels and TS elongation (TE) in PTEN LOF EAC. PTEN LOF is necessary and sufficient to confer sensitivity of EAC cells to LCI133 and other CDK9 inhibitors.

Enterococci are opportunistic pathogens classified by the World Health Organization as high-priority microorganisms. They cause a broad spectrum of infections, and their intrinsic and acquired resistance to antimicrobials makes these infections particularly difficult to treat and eradicate. In Enterococcus faecalis, the most frequently isolated enterococcal pathogen in humans, antimicrobial resistance and innate immune evasion are largely driven by the Enterococcal Polysaccharide Antigen (EPA). This surface polymer underpins key virulence traits, including resistance to host defence mechanisms, reduced susceptibility to multiple classes of antimicrobials, and susceptibility to bacteriophage infection. EPA consists of a rhamnan backbone decorated with strain-specific substituents that are essential for its biological activity. Here, we show that epaB encodes the enzyme responsible for the first committed step in assembling the EPA rhamnan chain. Using NMR spectroscopy, we demonstrate that E. faecalis lacking epaB produces an EPA polymer composed solely of decorations directly anchored to the peptidoglycan, with no detectable rhamnan backbone. The absence of this rhamnan moiety profoundly alters cell wall architecture, as revealed by atomic force microscopy of the mutant cell walls. The epaB mutation also abolishes innate immune evasion and virulence in the zebrafish infection model, while conferring resistance to bacteriophages. Collectively, these findings demonstrate that both the rhamnan backbone and its decorations are required for EPAs full biological activity, establishing the structural and functional interdependence of these two components.

SynNotch-CAR circuits improve T cell specificity by coupling antigen recognition to inducible CAR expression. However, basal CAR expression without receptor activation, termed here as leakiness, can reduce the separation between killing of intended target cells and sparing of antigen-positive off-target cells, limiting target-cell discrimination. Here, we systematically quantified basal CAR expression for several synNotch-CAR designs and developed a coupled ordinary differential equation model to show that discrimination depends on basal output, CAR potency, and effector-to-target ratio. We introduced C-terminal tags such as fluorescent proteins, degron domains, endocytosis signals, and endoplasmic reticulum retention motifs as a strategy to reduce CAR leakiness. We found that fluorescent proteins and degron-containing tags reduced basal CAR surface expression while preserving antigen-induced CAR expression, improving discrimination of antigen-density sensing and combinatorial circuits in vitro. In xenograft models, fluorescent protein-tagged CARs improved discrimination by reducing activity against off-target cells while retaining activity against high-antigen tumors. Degron-containing constructs reduced basal CAR expression in vitro but showed suboptimal performance in vivo, revealing a trade-off between basal CAR suppression and induced CAR persistence. Together, these findings demonstrate that basal output expression is a key parameter for inducible genetic circuit designs and establish layered transcriptional and post-translational regulation as a strategy to improve the fidelity of inducible T cell circuits.

More than one thousand bifunctional terpene synthases combining prenyltransferase and terpene cyclase activities have been identified in bacteria and fungi, but only a handful of enzymes have been identified that combine terpene cyclase activity with a downstream processing activity. Drimenol synthase from the marine bacterium Aquimarina spongiae (AsDMS) consists of a class II terpene cyclase that converts farnesyl diphosphate into drimenyl diphosphate, and a haloacid dehalogenase-like phosphatase that hydrolyzes drimenyl diphosphate to generate the sesquiterpene alcohol drimenol. The first crystal structure of AsDMS to be reported revealed the architecture of domain assembly as well as dimeric quaternary structure, establishing a structural chemical foundation for cyclization and hydrolysis mechanisms [K. R. Osika, M. N. Gaynes, D. W. Christianson (2025) Proc. Natl. Acad. Sci. U.S.A. 122, e2506584122]. Here, we report crystal structures of the catalytically-inactive double mutant, D33A-D323A AsDMS, complexed with farnesyl diphosphate, geranyl diphosphate, and dimethylallyl diphosphate, which bind in the active sites of both the cyclase and phosphatase domains. Molecular recognition of the diphosphate group dominates binding interactions in both active sites. In the cyclase active site, only farnesyl diphosphate is sufficiently long for its terminal isoprenoid C=C bond to bind adjacent to the catalytic general acid that would initiate the cyclization cascade in the wild-type enzyme. In the phosphatase active site, all isoprenoid diphosphate groups bind similarly, but isoprenoid chain conformations vary. These structures provide a foundation for understanding substrate recognition and catalysis in both active sites. Finally, we present kinetic evidence suggesting that substrate channeling is operative in wild-type AsDMS.

Reflectin proteins drive dynamic structural coloration in cephalopods by organizing into dense intracellular lamellar structures that dictate local refractive index. While reconstituted reflectins readily undergo liquid-liquid phase separation in vitro, these assemblies frequently undergo dynamic arrest, vitrifying into non-dynamic condensates. Here, we investigate the primary sequence features, post-translational modifications, and heterotypic interactions that regulate the material properties of reflectin condensates within the crowded cellular environment of mammalian HeLa cells. Using confocal microscopy and fluorescence recovery after photobleaching (FRAP), we demonstrate that canonical block copolymeric A-type reflectins readily form dynamically arrested condensates, with the linker blocks primarily responsible for the observed arrest. In contrast, non-canonical B/C reflectin variants exhibit significantly greater fluidity and rapid recovery kinetics. We show that phosphomimetic substitutions progressively fluidize some reflectin condensates. Lastly, we find that heterotypic condensates composed of canonical and non-canonical reflectins in combinations associated with reversible iridescence in squid substantially enhance canonical mobility. Our findings establish a biophysical framework in which phosphorylation and heterotypic mixing cooperatively suppress dynamic arrest, enabling the reversible material transitions required for active cephalopod camouflage and communication.

Pseudomonas aeruginosa is an opportunistic human pathogen capable of infecting a wide range of tissues and organs. Its persistence during chronic infection is strongly associated with biofilm formation, which depends on extracellular polysaccharides such as alginate. The HtrA-like periplasmic serine protease MucD is a key regulator of bacterial virulence, stress response, and alginate production, yet its molecular mechanism has remained largely unclear. Here, we discovered the alginate acetylation and export proteins AlgX and AlgK as MucD substrates, and characterized their degradation by mass spectrometry and bioinformatic analysis. We further determined the cryo-EM structure of MucD bound to an AlgK-derived substrate peptide, offering atomic insights into MucD oligomerization assembly, substrate recognition, and specificity. Together with structure-guided mutagenesis and biochemical assays, our results revealed that MucD proteolytic activity is governed by an equilibrium between a resting 12-mer and an active trimer. Crucially, we demonstrate that MucD represses alginate biosynthesis post-translationally, in addition to its previously implicated role in transcriptional regulation. These findings define a distinct activation mechanism and regulatory function for MucD and provide new insight into bacterial HtrA-like serine proteases.

The Gram-negative bacterial cell envelope features an asymmetric outer membrane, that confers intrinsic resistance to toxins. Maintenance of this barrier relies on the Mla system, which mediates retrograde transport of mislocalized phospholipids. In Escherichia coli, this system comprises the lipoprotein MlaA, the periplasmic shuttle protein MlaC, and the ABC transporter complex MlaFEDB. Intriguingly, in Campylobacter jejuni, mlaA and mlaC share an operon with an encoded Resistance-Nodulation-cell Division antiporter potentially involved in anterograde phospholipid transport. Here, we describe the functional and mechanistical characterization of Cj MlaC. Complementation experiments in E. coli show that Cj MlaC functions independently of the native Mla system. Native mass spectrometry revealed that Cj MlaC uniquely exists as both monomer and dimer. Lipid binding stabilized the dimer and ion mobility mass spectrometry showed that conformational transitions precede phospholipid release, suggesting a cycle between a low-affinity monomer and a higher-lipid-affinity dimer. Cj MlaC binds phospholipid species distinct from Ec MlaC, showing an increased propensity for lysophospholipids, consistent with the unusually lysophospholipid-rich lipidome of C. jejuni, indicative of evolutionary adaptation to this unique lipid environment. Collectively, these findings uncover structural and mechanistic features of Cj MlaC and support divergent physiological roles for Cj and Ec MlaC in phospholipid trafficking.

O-linked {beta}-N-acetylglucosamine (O-GlcNAc) transferase (OGT) is the sole writer for intracellular O-GlcNAcylation. It catalyzes O-GlcNAcylation of thousands of protein substrates, but relatively less is known about the post-translational modifications that occur on OGT itself. Herein, we demonstrate that OGT is S-palmitoylated at Cys-472 and Cys-477, which is mediated by the S-acyltransferase Zinc Finger DHHC-Type Palmitoyl transferase 14 (zDHHC14) and removed by acyl protein thioesterase 2 (APT2). S-Palmitoylation stabilizes OGT by shunting it away from the lysosomal chaperone-mediated autophagy (CMA) pathway, as S-palmitoylation decreases the interaction between OGT and heat shock cognate 70 kDa protein (HSC70), the CMA chaperone. Via label-free quantitative mass spectrometry, we find that S- palmitoylation elevates the affinity between OGT and protein phosphatase 1 catalytic subunit gamma (PPP1CC), but not PPP1CB. We further demonstrate that S-palmitoylation of OGT augments binding with Yes-associated protein-1 (YAP), a protein that associates with PPP1CC, and subsequently enhances YAP O-GlcNAcylation. Our work unearths S-palmitoylation of OGT and CMA-mediated degradation of lysosomal OGT, the orchestration of which finetunes the activity of key OGT complexes, such as OGT-PPP1CC, and contributes to OGT substrate selectivity.

Background: Polymorphisms in the N-acetyltransferase 2 (NAT2) gene explain much of the interindividual variation in isoniazid (INH) metabolism and determine risk of toxicities. However, there is limited evidence to guide INH dose adjustment according to the NAT2 acetylator profile in weekly rifapentine-INH tuberculosis preventive therapy (TPT). Methods: In a prospective, multicenter, within-subject PK trial (NCT05413551), adults initiating 3HP in Brazil were assigned genotype-guided INH doses (slow: 5 mg/kg <=300 mg; intermediate: 15 mg/kg <=900 mg; rapid: 25 mg/kg <=1,500 mg) alongside a standard 900 mg flat dose on an alternate occasion. AUC0-24 and C24 were estimated from serial blood samples; a two-compartment Michaelis-Menten population PK model characterized NAT2 effects on clearance. Results: Among 228 participants, 47.4% (108/228) were intermediate, 43.4% (99/228) slow, and 9.2% (21/228) rapid acetylators. Genotype-guided dosing reduced AUC0-24 variability approximately two-fold versus standard dosing (CV 58.8% vs 76.8%) and increased exposure uniformity (median AUC0-24 27.2 [IQR 18.8-41.3] vs 43.2 [27.3-71.0] mg h/L). Among slow acetylators, C24 >0.15 ug/mL decreased from 27/42 (64%) with standard dosing to 1/42 (2%) with genotype-guided dosing (P<0.0001). In 104 participants with intensive PK sampling, rapid acetylators receiving guided doses had AUC0-24 similar to standard-dose intermediate acetylators (42.8 vs 39.5 mg h/L; P=.63). Monte Carlo simulations supported doses of 600, 900, and 1,200 mg for slow, intermediate, and rapid acetylators, respectively. Conclusions: NAT2-guided isoniazid dosing reduced variation in drug levels, averting very low and high AUC and C24. These findings inform genotype-stratified dosing of INH for TPT, which might reduce toxicities and improve outcomes.

Predicting disease risk from prior diagnoses is fundamental to clinical decision-making, particularly during health emergencies such as the COVID-19 pandemic, when individuals with long-term conditions may be disproportionately vulnerable to adverse outcomes. Despite intense interest in developing models to predict disease risk from prior diagnoses (1-3), most prediction models do not estimate effects of each prior diagnosis on disease risk conditional on other diagnoses, limiting interpretability and clinical utility. We developed the Comorbidity Risk Score (CRS), trained on 13 million individuals (age 40-69) from linked electronic health record (EHR) datasets of the entire population of England, to predict COVID-19 hospitalisation and 87 other disease outcomes. CRS was trained at close to saturated sample size and precisely estimated the effects of 212 prior diagnoses on the 88 disease outcomes, conditional on all other prior diagnoses. Correlations of CRS effect sizes across outcomes (e.g. 0.76 for myocardial infarction vs. hyperlipidaemia) matched the corresponding genetic correlations (e.g. 0.79 for myocardial infarction vs. hyperlipidaemia), confirming that comorbidity architectures capture disease aetiology. On average, CRS identified 5% of the population with 3.4-fold higher disease risk, including myocardial infarction (4.4-fold), lung cancer (6.5-fold), and COVID-19 hospitalisation (6.3-fold). Using prior diagnoses alone, CRS outperformed state-of-the-art clinical COVID-19 models (4). Furthermore, CRS (N=13 million) substantially outperformed state-of-the-art AI (1) (N=0.5 million) and linear (3) (N=0.5 million) models in predicting disease risk, suggesting that training sample size outweighs model complexity. CRS attained near-perfect transferability across self-reported ethnicities (e.g., Black vs. White: AUROC ratio = 97.3%). Finally, CRS distinguished independently predictive comorbidities from indirect associations, e.g., lipid metabolism disorder was a strong predictor of myocardial infarction risk but not ischaemic stroke, after conditioning on other prior diagnoses. In conclusion, CRS provides a comprehensive resource for understanding the impact of comorbidities on COVID-19 and other future diseases, revealing disease aetiology while enabling powerful prediction of disease risk.