Chemical Science

Rhodamine molecules are setting benchmarks in fluorescence microscopy. Herein, we report the deuterium (d12) congeners of tetramethyl(silicon)rhodamine, obtained by isotopic labelling of the four methyl groups, which improves photophysical (i.e. brightness, lifetimes) and chemical (i.e. bleaching) properties. We explore this finding for SNAP- and Halo-tag labelling, and highlight enhanced properties in several applications, such as Forster resonance energy transfer, fluorescence activated cell sorting, fluorescence lifetime microscopy and stimulated emission depletion nanoscopy. We envision deuteration as a generalizable concept to improve existing and develop new Chemical Biology Probes.

The process of -synuclein aggregation results in the formation of amyloid fibrils, which accumulate in the brain of patients affected by Parkinsons disease. Among possible therapeutic strategies to cure this disease, one approach is based on the development of compounds capable of inhibiting -synuclein aggregation. An effective inhibition could be achieved by blocking the nucleation sites on the surface of the amyloid fibrils that are responsible for their autocatalytic proliferation. Here, we report a strategy based on deep learning to achieve this goal, which uses an E(3)-equivariant conditional diffusion model. By using this approach, we designed and tested experimentally candidate small molecules. We found that one of these small molecules acts as a potent inhibitor of secondary nucleation in -synuclein aggregation. These results provide evidence that generative diffusion models offer effective tools for drug design.

Molecular simulations have become indispensable in biological research. Their accuracy continues to improve, but directly modelling biochemical reactions - central to all life processes - remains computationally challenging. Here, we present a biomolecular reaction emulator that models reactions across conformational ensembles using kinetic Monte Carlo. Our method, KIMMDY, is capable of handling dynamic, large-scale systems with successive, competing reactions, even on the second timescale or slower. It leverages graph neural networks for large-scale prediction of reaction rates, while also being capable of using simpler physics-based or heuristic models. We validate our approach against experimental data and showcase its power and versatility through a series of applications, including radical reactions, nucleophilic substitutions, and photodimerization. Example systems span proteins and DNA. KIMMDY aids the understanding of biochemical reaction cascades in complex systems, helps to re-interpret experimental data, and can inspire future wet-lab experiments.

Xenorhabdus strains, Gram-negative bacteria pathogenic to insects and symbionts to nematodes of the genus Steinernema are prolific producers of various natural products. Here we describe the xisABCDE biosynthesis gene cluster from Xenorhabdus hominickii responsible for the production of xildivalines. These non-ribosomal peptide and polyketide hybrids act as peptide deformylase inhibitor (PDI) and occur also in other Gammaproteobacteria, especially Vibrio. Their structure and biosynthesis were fully elucidated despite their instability, highlighting a rare trans-methylation of their N-terminus. Subsequently, the structure of the responsible methyltransferase XisE and the peptide deformylase XisD, serving as resistance mechanism, were elucidated by X-ray crystallography, allowing insights into the function and the mode of action of this novel class of PDIs.

"Mirror life", self-replicating organisms composed of non-natural-chirality biomacromolecules, presents a severe future threat with potentially global consequences. Consequently, there is strong agreement among experts that it should not be created. However, there is some disagreement over how effective existing medical countermeasures might prove against mirror bacteria in the event that they were created. Here, we leverage computational chemistry methods including docking and molecular dynamics to determine the likely binding efficacy of existing antibiotics against natural and mirror bacterial protein targets. We find that existing most antibiotics fail to bind to mirror bacterial protein targets, unlike their natural chirality targets. This suggests that current medical countermeasures would not successfully exert an antimicrobial activity against mirror bacteria if the latter were created. Our results motivate further policy advocacy to curtail research that directly leads to the creation of mirror life.

Electrolyte conditions in vivo and in vitro are known to influence protein structure and function. Intrinsically disordered proteins (IDPs) are particularly sensitive to their solution conditions such as ionic strength and molecular crowding, and their dynamic structural ensembles rapidly respond to the cellular environment. While structural mass spectrometry (MS) techniques are uniquely able to capture aspects of this structural diversity, technical limitations have largely precluded the use of native MS approaches to interrogate the conformational rearrangements of IDPs in response to high concentrations of non-volatile salts. Here, we overcome this challenge by employing sub 100-nm nanopipette electrospray emitters for more gentle and salt-tolerant analysis to study the conformational distribution of -Synuclein (S) using native MS and ion mobility-MS in varied solution conditions, including in phosphate buffered saline. We show using native MS that it is possible to capture salt and buffer induced changes in the S conformational ensemble when using traditional biochemical buffers, which reflect structural changes from in silico predictions and in-solution measurements. This work demonstrates the power of nanopipette emitters for the study of IDPs, and establishes native MS as a method that can be routinely used to determine how solution conditions tune the conformational landscape of IDPs. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=110 SRC="FIGDIR/small/671163v2_ufig1.gif" ALT="Figure 1"> View larger version (28K): org.highwire.dtl.DTLVardef@35ff69org.highwire.dtl.DTLVardef@113821aorg.highwire.dtl.DTLVardef@1c27b90org.highwire.dtl.DTLVardef@13e4db1_HPS_FORMAT_FIGEXP M_FIG C_FIG

Fumarate-adding enzymes (FAE) are a subset of the glycyl radical enzyme superfamily involved in anaerobic hydrocarbon degradation. Benzylsuccinate synthase (BSS) catalyzes the enantiospecific formation of R-benzylsuccinate from toluene and fumarate, initiating anaerobic toluene degradation. In this paper, we present a detailed theoretical study of the reaction mechanism using classical molecular dynamics and multiscale modelling (QM:MM). We describe the potential energy surface of the reaction, confirming the previously postulated mechanism. However, the multiscale character of our model allowed to elucidate the origins of several experimentally observed catalytic phenomena, such as the inversion of the configuration of the benzylic atom upon C-C bond formation, syn addition of the abstracted H atom back to the benzylsuccinyl radical, or kinetic isotope effects in the range of 1.7-2.1. The obtained model is supported by microkinetic analysis and was able to explain and quantitatively predict the strict R-enantioselectivity of BSS, which is not enforced by the binding orientation of the fumarate, but by dynamic kinetic behaviour of toluene in the active site leading to faster production of the R-enantiomer. We were also able to explain the experimentally observed slow H/D exchange in the product during incubation with BSS in D2O, confirming the partial reversibility of the reaction. Our study contributes to the elucidation of the catalytic processes catalyzed by BSS and its role in the bioremediation of hydrocarbon pollutants.

The Gram-negative bacterial outer membrane poses a major obstacle to the development of much-needed antibiotics against drug-resistant infections. Its chemical composition and porin proteins differ from Gram-positive bacteria and mammalian cells, and heuristics developed for mammalian cell uptake apply poorly. Recently, machinelearning methods have predicted small-molecule uptake into Gram-negative bacteria, offering the possibility to rationally optimize this aspect of antibiotic lead development. Here, we report physics-based methods to prospectively predict Gram-negative bacterial uptake, select, and synthesize promising chemical derivatives targeting E. coli DNA gyrase B. Our methods do not require empirical parameterization and are readily adaptable to new chemical scaffolds. These physics-based predictions well capture experimentally measured uptake (r > 0.95) and are indeed predictive of antimicrobial activity (r > 0.92). These methods can be used prospectively in combination with target-binding simulations to optimize both bacterial uptake and target binding, overcoming important barriers to antibiotic lead generation before small-molecule synthesis.

Renilla luciferase (RLuc) remains one of the most popular bioluminescence reporters, but its molecular principle has yet to be fully understood. Here, we reveal a detailed kinetic mechanism of the RLuc catalytic cycle which uncovers multiple limiting factors: (i) an oxygen-induced irreversible inactivation, (ii) a low oxygen saturation, and (iii) rate-limiting induced-fit conformational dynamics coupled with the product release. Furthermore, we could determine the actual enzyme kcat value at all saturating substrates to be 22 s-1. This value is 5-fold higher than the previously reported apparent kcat values determined at physiological, non-saturating oxygen concentration. Our integrative analysis by transient kinetics, X-ray crystallography, and molecular dynamics linked the rate-limiting flexible enzyme opening to the dynamics of the loops surrounding the active site, which prompted targeted engineering of this limiting step by loop grafting. The resulting variant AncFT-L14 (AncFT7) showed a prolonged stable light emission thanks to the selectively improved induced-fit kinetics. Additional characterization of AncFT-L14 identified increased catalytic efficiency kcat/Km, product inhibition factor Kp/Km, and a glow-type signal characteristic. Our results provide mechanistic details of RLuc catalysis and will govern future enzyme engineering to design the next generations of bioluminescence-based tools.

Non-natural amino acids are increasingly used as building blocks in the development of peptide-based drugs, as they expand the available chemical space to tailor function, half-life and other key properties. However, while the chemical space of modified amino acids (mAAs) is potentially vast, experimental methods for measuring the developability properties of mAA-containing peptides are expensive and time consuming. To facilitate developability programs through computational methods, we present CamSol-PTM, a method that enables the fast and reliable sequence-based prediction of the solubility of mAA-containing peptides. From a computational screening of 50,000 mAA-containing variants of three peptides, we selected five different mAAs for a total number of 30 peptide variants for experimental validation. We demonstrate the accuracy of the predictions by comparing the calculated and experimental solubility values. Our results indicate that the computational screening of mAA-containing peptides can extend by over four orders of magnitude the ability to explore the solubility chemical space of peptides. This method is available as a web server at https://www-cohsoftware.ch.cam.ac.uk/index.php/camsolptm.

Forster resonance energy transfer (FRET), an optical distance ruler, provides the unique ability to monitor molecular interactions and conformational changes in single biomacromolecules and multi-subunit complexes. It has proven to be a powerful tool to study enzymatic reactions, membrane transport, protein folding, but also the properties of nucleic acids, proteins, molecular motors and many other biological systems and processes. A prerequisite for FRET is targeted (covalent) labelling of macromolecules with two distinct fluorophores. Here, we present a strategy for stochastic labeling of protein residues with the required donor and acceptor dyes via anion exchange chromatography. While this technique has been used before for this purpose, we provide a conceptual basis to systematically design a purification protocol for an arbitrary choice of fluorophores. By characterizing the interaction fluorophore-maleimides with the column material, we are able to select (and predict) which pairs of fluorophores allow successful purification of donor-acceptor-labelled protein with yields up to 98%. We demonstrate the capabilities of the method for bulk and single-molecule FRET assays of various bacterial substrate-binding proteins.

The {beta}-lactamase OXA-48 efficiently hydrolyses carbapenem antibiotics, especially imipenem. Carbapenem resistance is a rising clinical concern, and is frequently associated with OXA-48 and its variants. OXA-48 variants carrying different mutations in the {beta}5-{beta}6 loop differ in hydrolytic activity towards imipenem. OXA-517 has a higher KM, but similar kcat for imipenem hydrolysis, compared to OXA-48, whereas OXA-163 and -405, which have similar mutations in the {beta}5-{beta}6 loop, are less active. Multiscale simulations (using quantum mechanics/molecular mechanics, QM/MM) of deacylation of the respective imipenem acylenzymes show this to be most efficient when the deacylating water (DW) acts as a hydrogen bond (H-bond) donor to imipenem, and the carboxylated Lys73 base is less hydrated. Calculated barriers for deacylation correlate very well with experimental data, but for OXA-163 and -405 only when DW acts as a H-bond acceptor. Dynamics simulations of imipenem acylenzyme complexes show that mutations in the {beta}5-{beta}6 loop change the active site H-bond network. In OXA-48, the DW H-bonding pattern linked to high activity is more frequently sampled, and in OXA-517 it is stabilised through H-bonding to Thr213; explaining the higher kcat values compared to OXA-163 and -405, where this is not the case. Furthermore, simulations of non-covalent imipenem complexes indicate that increased KM for OXA-517 is linked to lower binding affinity, caused by repositioning of bound imipenem. Our work identifies the molecular basis for differences in imipenem hydrolytic activity between OXA-48 variants, offering detailed insights into how active site interactions alter the dynamics and reaction efficiencies related to antibiotic resistance.

Recombinant proteins and antibodies are routinely used as drugs to treat prevalent diseases such as diabetes or cancer, while enzyme replacement and gene therapies are the main therapeutic intervention lines in rare diseases. In protein-based therapeutics, optimized in vivo stability is key as intrinsic denaturation and intracellular proteostatic degradation will limit potency, particularly in treatments requiring a sustained action, while clearance mechanisms may limit the amount of circulating protein. In vivo stability is ultimately correlated with the intrinsic thermodynamic stability of the biomolecule, but this is difficult to optimize because it often goes at the expense of reducing protein activity. Here, we have used in silico engineering approaches to thermodynamically stabilize human frataxin, a small mitochondrial protein that acts as an allosteric activator for the biosynthesis of Fe-S clusters, whose genetically-driven impairment results in a rare disease known as Friedreich ataxia. Specifically, we developed an efficient thermostability engineering computational approach that combines information on amino acid conservation, the Rosetta energy function, and two recent artificial intelligence tools - AlphaFold and ProteinMPNN - to produce thermodynamically stabilized variants of human frataxin. Such protein variants rescued the large destabilization exerted by well-known pathological mutations, with an increase over 20 {degrees}C in the melting temperature and a thermodynamic stabilization of more than 3 kcal{middle dot}mol-1 at the physiological temperature. This stability surplus is translated into an enhanced resistance to proteolysis, while maintaining the protein fully functional. This case-study highlights the power of our combined computational approach to generate optimized variants, adequate for protein-based therapeutics.

Protein assembly defects due to enrichment of aberrant conformational variants of proteins are emerging as a new frontier in therapeutics design. Understanding, atomistically, structural elements that remodel the energy landscape of proteins, with the consequence of rewiring the conformational dynamics of proteins and pathologically perturbing functionally-oriented ensembles, is key for development of inhibitors. This is particularly relevant for molecular chaperones, hub proteins for the assembly of large multiprotein complexes, where enrichment of aberrant conformers can have a large impact on the cellular proteome, and in turn, on phenotypes. Here, we integrate computational and experimental tools to unveil how N-glycosylation of specific residues in glucose-regulated protein 94 (GRP94) modulates internal dynamics and alters the conformational fitness of regions fundamental for interaction with the nucleotide and synthetic ligands, and impacts substructures dedicated to recognition of interacting proteins. We show how N-glycosylation plays an active role in modulating the energy landscape of the protein, with specific glycosylation patterns determining specific functionally-oriented dynamic signatures. Our results provide support for leveraging the structural-dynamics knowledge on distinct glycosylation variants to design molecules targeting GRP94 disease-associated conformational states and assemblies. Since glycosylation is the most abundant form of post-translational modification, our results and mechanistic models can readily be transferred to other targets and contexts for cancers and other diseases.

Generative drug design facilitates the creation of compounds effective against pathogenic target proteins. This opens up the potential to discover novel compounds within the vast chemical space and fosters the development of innovative therapeutic strategies. However, the practicality of generated molecules is often limited, as many designs focus on a narrow set of drug-related properties, failing to improve the success rate of subsequent drug discovery process. To overcome these challenges, we develop TamGen, a method that employs a GPT-like chemical language model and enables target-aware molecule generation and compound refinement. We demonstrate that the compounds generated by TamGen have improved molecular quality and viability. Additionally, we have integrated TamGen into a drug discovery pipeline and identified 7 compounds showing compelling inhibitory activity against the Tuberculosis ClpP protease, with the most effective compound exhibiting a half maximal inhibitory concentration (IC50) of 1.9 M. Our findings underscore the practical potential and real-world applicability of generative drug design approaches, paving the way for future advancements in the field.

The rapid rise of antibiotic-resistant bacteria necessitates the search for alternative, unconventional solutions, such as targeting bacterial communication. Signal disruption can be achieved by enzymatic degradation of signaling compounds, reducing the expression of genes responsible for virulence, biofilm formation, and drug resistance while evading common resistance mechanisms. Therefore, enzymes with such activity have considerable potential as antimicrobial agents for medicine, industry, and other areas of life. Here, we designed molecular gates that control the binding site of penicillin G acylase to shift its preference from native substrate to signaling molecules. Using an ensemble-based design, three variants carrying triple-point mutations were proposed and experimentally characterized. Integrated inference from biochemical and computational analyses demonstrated that these three variants had markedly reduced activity towards penicillin and each preferred specific signal molecules of different pathogenic bacteria, exhibiting up to three orders of magnitude shifts in substrate specificity. Curiously, while we could consistently expand the pockets in these mutants, the reactive binding of larger substrates was limited, either by overpromoting or overstabilizing the pocket dynamics. Overall, we demonstrated the designability of this acylase for signal disruption and provided insights into the role of appropriately modulated pocket dynamics for such a function. The improved mutants, the knowledge gained, and the computational workflow developed to prioritize large datasets of promising variants may provide a suitable toolbox for future exploration and design of enzymes tailored to disrupt specific signaling pathways as viable antimicrobial agents.

De novo protein design is delivering new peptide and protein structures at a rapid pace. Many of these synthetic polypeptides form well-defined and hyperthermal-stable structures. Generally, however, less is known about the dynamic properties of the de novo designed structures. Here, we explore one aspect of dynamics in a series of de novo coiled-coil peptide assemblies: namely, peptide exchange within and between different oligomers from dimers through to heptamers. First, we develop a fluorescence-based reporter assay for peptide exchange that is straightforward to implement, and, thus, would be useful to others examining similar systems. We apply this assay to explore both homotypic exchange within single species, and heterotypic exchange between coiled coils of different coiled-coil oligomer states. For the former, we provide detailed study for the dimeric coiled coil CC-Di finding a half-life for exchange of 4.2 {+/-} 0.3 minutes when the concentration of CC-Di is 200 {micro}M. Interestingly, more broadly when assessing exchange across all of the oligomeric states, we find that some of the designs are faithful and only undergo homotypic strand exchange, whereas others are promiscuous and exchange to form unexpected hetero-oligomers. Finally, we develop two design strategies to improve the orthogonality of the different oligomers: (i) using alternate positioning of salt bridge interactions; and (ii) incorporating of non-canonical repeats into the designed sequences. In so doing, we reconcile the promiscuity and deliver a set of faithful homo-oligomeric de novo coiled-coil peptides. Our findings have implications for the application of these and other coiled coils as modules in chemical and synthetic biology.

Antibiotic resistance is an alarming public health concern that affects millions of individuals across the globe each year. A major challenge in the development of effective antibiotics lies in their limited ability to permeate into cells, noting that numerous susceptible antibiotic targets reside within the bacterial cytosol. Consequently, improving cellular permeability is often a key consideration during antibiotic development, underscoring the need for reliable methods to assess the permeability of molecules across cellular membranes. Currently, methods used to measure permeability often fail to discriminate between arrival within the cytoplasm and the overall association of molecules with the cell. Additionally, these techniques typically possess throughput limitations. In this work, we describe a luciferase-based assay designed for assessing the permeability of molecules into the cytosolic compartment of Gram-negative bacteria. Our findings demonstrate a robust system that can elucidate the kinetics of intracellular antibiotics accumulation in live bacterial cells in real time.

Protein aggregation correlates with many human diseases. Protein aggregates differ in shape, ranging from amorphous aggregates to amyloid fibrils. Possibly for such heterogeneity, strategies to develop effective aggregation inhibitors that reach the clinic failed so far. Here, we present a new strategy by which we developed a family of peptides targeting early aggregation stages for both amorphous and fibrillar aggregates of proteins unrelated in sequence and structure. Thus, they act on dynamic precursors before a mechanistic differentiation takes place. Using a peptide array approach, we first identified peptides inhibiting the predominantly amorphous aggregation of a molten globular, aggregation-prone protein, a thermolabile mutant of the Axin tumor suppressor. A series of optimization steps revealed that the peptides activity did not depend on their sequences but rather on their molecular determinants. The key properties that made a peptide active were a composition of 20-30% flexible, 30-40% aliphatic and 20-30% aromatic residues, a hydrophobicity/hydrophilicity ratio close to 1 and an even distribution of residues of different nature throughout the sequence. Remarkably, the optimized peptides also suppressed fibrillation of Tau, a disordered protein that forms amyloids in Alzheimers disease, and entirely unrelated to Axin. Our compounds thus target early aggregation stages, independent of the aggregation mechanism, inhibiting both amorphous and amyloid aggregation. Such cross-mechanistic, multi-targeting aggregation inhibitors may be attractive lead compounds against multiple protein aggregation diseases.

The intrinsically disordered protein -Synuclein is identified as a major toxic aggregate in Parkinsons as well as several other neurodegenerative diseases. Recent work on this protein has focused on the effects of posttranslational modifications on aggregation kinetics. Among these, O-GlcNAcylation of -Synuclein has been observed to inhibit the aggregation propensity of the protein. Here we investigate the monomer dynamics of two O-GlcNAcylated -Synucleins, -Syn(gT72) and -Syn(gS87) and correlate them with the aggregation kinetics. We find that, compared to the unmodified protein, glycosylation at T72 makes the protein less compact and more diffusive while glycosylation at S87 makes the protein more compact and less diffusive. Based on a model of the earliest steps in aggregation, we predict that T72 should aggregate slower than unmodified protein, which is confirmed by ThT fluorescence measurements. In contrast, S87 should aggregate faster, which is not mirrored in ThT kinetics of later fibril formation but does not rule out a higher rate of formation of small oligomers. Together, these results show that posttranslational modifications do not uniformly affect aggregation propensity.