Multi-condensate state as a functional strategy to optimize the cell signaling output.

The existence of multiple biomolecular condensates inside living cells is a peculiar phenomenon not compatible with the predictions of equilibrium statistical mechanics. In this work, we address the problem of multiple condensates state (MCS) from a functional perspective. We combine Langevin dynamics, reaction-diffusion simulation, and dynamical systems theory to demonstrate that MCS can indeed be a function optimization strategy. Using Arp2/3 mediated actin nucleation pathway as an example, we show that actin polymerization is maximum at an optimal number of condensates. For a fixed amount of Arp2/3, MCS produces a greater response compared to its single condensate counterpart. Our analysis reveals the functional significance of the condensate size distribution which can be mapped to the recent experimental findings. Given the spatial heterogeneity within condensates and non-linear nature of intracellular networks, we envision MCS to be a generic functional solution, so that structures of network motifs may have evolved to accommodate such configurations.

Entropy-driven Denaturation for Sustainable Regeneration of Protein Waste.

Electrolytes are essential parts of the environment for all life forms, where proteins, water, and solutes interplay to support vital activities. However, a fundamental understanding of the effect of ionic solutes on proteins remains elusive for more than a century. Here we show how some ionic solutes can serve as potent denaturants despite the absence of direct protein-ion interactions. We demonstrate dramatic differences between denaturation potency of different ionic solutes with lithium bromide (LiBr) being the strongest denaturant and sodium bromide (NaBr) being the least potent. Experiments and simulations indicate the presence of certain ions disrupts the structure of water network, thereby induce protein denaturation indirectly via an entropy-driven mechanism. We further introduce a scalable strategy for protein waste revalorization, distinguished by the closed-loop recycling of denaturants, straightforward protein separation, and facile manufacturing, all enabled by the entropy-driven denaturation by LiBr. Through successful isolation and systematic study of indirect solute effects, our findings suggest a unified and generally applicable framework for decoding of the protein-water-solute nexus, where all current studies can be easily incorporated. Besides, our regeneration approach underscores the feasibility of repurposing protein waste into valuable biomaterials in a sustainable way with wide-reaching application potential.

Biophysical principles predict fitness of SARS-CoV-2 variants.

SARS-CoV-2 employs its spike protein's receptor binding domain (RBD) to enter host cells. The RBD is constantly subjected to immune responses, while requiring efficient binding to host cell receptors for successful infection. However, our understanding of how RBD's biophysical properties contribute to SARS-CoV-2's epidemiological fitness remains largely incomplete. Through a comprehensive approach, comprising large-scale sequence analysis of SARS-CoV-2 variants and the identification of a fitness function based on binding thermodynamics, we unravel the relationship between the biophysical properties of RBD variants and their contribution to viral fitness. We developed a biophysical model that uses statistical mechanics to map the molecular phenotype space, characterized by dissociation constants of RBD to ACE2, LY-CoV016, LY-CoV555, REGN10987, and S309, onto an epistatic fitness landscape. We validate our findings through experimentally measured and machine learning (ML) estimated binding affinities, coupled with infectivity data derived from population-level sequencing. Our analysis reveals that this model effectively predicts the fitness of novel RBD variants and can account for the epistatic interactions among mutations, including explaining the later reversal of Q493R. Our study sheds light on the impact of specific mutations on viral fitness and delivers a tool for predicting the future epidemiological trajectory of previously unseen or emerging low-frequency variants. These insights offer not only greater understanding of viral evolution but also potentially aid in guiding public health decisions in the battle against COVID-19 and future pandemics.

Multi-condensate state as a functional strategy to optimize the cell signaling output.

The existence of multiple biomolecular condensates inside living cells is a peculiar phenomenon not compatible with the predictions of equilibrium statistical mechanics. In this work, we address the problem of multiple condensates state (MCS) from a functional perspective. We combined Langevin dynamics, reaction-diffusion simulation, and dynamical systems theory to demonstrate that MCS can indeed be a function optimization strategy. Using Arp2/3 mediated actin nucleation pathway as an example, we show that actin polymerization is maximum at an optimal number of condensates. For a fixed amount of Arp2/3, MCS produces a greater response compared to its single condensate counterpart. Our analysis reveals the functional significance of the condensate size distribution which can be mapped to the recent experimental findings. Given the spatial heterogeneity within condensates and non-linear nature of intracellular networks, we envision MCS to be a generic functional solution, so that structures of network motifs may have evolved to accommodate such configurations.

Biophysical principles predict fitness of SARS-CoV-2 variants.

SARS-CoV-2 employs its spike protein's receptor binding domain (RBD) to enter host cells. The RBD is constantly subjected to immune responses, while requiring efficient binding to host cell receptors for successful infection. However, our understanding of how RBD's biophysical properties contribute to SARS-CoV-2's epidemiological fitness remains largely incomplete. Through a comprehensive approach, comprising large-scale sequence analysis of SARS-CoV-2 variants and the discovery of a fitness function based on binding thermodynamics, we unravel the relationship between the biophysical properties of RBD variants and their contribution to viral fitness. We developed a biophysical model that uses statistical mechanics to map the molecular phenotype space, characterized by binding constants of RBD to ACE2, LY-CoV016, LY-CoV555, REGN10987, and S309, onto a epistatic fitness landscape. We validate our findings through experimentally measured and machine learning (ML) estimated binding affinities, coupled with infectivity data derived from population-level sequencing. Our analysis reveals that this model effectively predicts the fitness of novel RBD variants and can account for the epistatic interactions among mutations, including explaining the later reversal of Q493R. Our study sheds light on the impact of specific mutations on viral fitness and delivers a tool for predicting the future epidemiological trajectory of previously unseen or emerging low frequency variants. These insights offer not only greater understanding of viral evolution but also potentially aid in guiding public health decisions in the battle against COVID-19 and future pandemics.

Epistasis and pleiotropy shape biophysical protein subspaces associated with drug resistance.

Protein space is a rich analogy for genotype-phenotype maps, where amino acid sequence is organized into a high-dimensional space that highlights the connectivity between protein variants. It is a useful abstraction for understanding the process of evolution, and for efforts to engineer proteins towards desirable phenotypes. Few mentions of protein space consider how protein phenotypes can be described in terms of their biophysical components, nor do they rigorously interrogate how forces like epistasis-describing the nonlinear interaction between mutations and their phenotypic consequences-manifest across these components. In this study, we deconstruct a low-dimensional protein space of a bacterial enzyme (dihydrofolate reductase; DHFR) into "subspaces" corresponding to a set of kinetic and thermodynamic traits [k_{cat}, K_{M}, K_{i}, and T_{m} (melting temperature)]. We then examine how combinations of three mutations (eight alleles in total) display pleiotropy, or unique effects on individual subspace traits. We examine protein spaces across three orthologous DHFR enzymes (Escherichia coli, Listeria grayi, and Chlamydia muridarum), adding a genotypic context dimension through which epistasis occurs across subspaces. In doing so, we reveal that protein space is a deceptively complex notion, and that future applications to bioengineering should consider how interactions between amino acid substitutions manifest across different phenotypic subspaces.

Enzymatic metabolons dramatically enhance metabolic fluxes of low-efficiency biochemical reactions.

In this work, we investigate how spatial proximity of enzymes belonging to the same pathway (metabolon) affects metabolic flux. Using off-lattice Langevin dynamics simulations in tandem with a stochastic reaction-diffusion protocol and a semi-analytical reaction-diffusion model, we systematically explored how strength of protein-protein interactions, catalytic efficiency, and protein-ligand interactions affect metabolic flux through the metabolon. Formation of a metabolon leads to a greater speedup for longer pathways and especially for reaction-limited enzymes, whereas, for fully optimized diffusion-limited enzymes, the effect is negligible. Notably, specific cluster architectures are not a prerequisite for enhancing reaction flux. Simulations uncover the crucial role of optimal nonspecific protein-ligand interactions in enhancing catalytic efficiency of a metabolon. Our theory implies, and bioinformatics analysis confirms, that longer catalytic pathways are enriched in less optimal enzymes, whereas most diffusion-limited enzymes populate shorter pathways. Our findings point toward a plausible evolutionary strategy where enzymes compensate for less-than-optimal efficiency by increasing their local concentration in the clustered state.

Separation of sticker-spacer energetics governs the coalescence of metastable biomolecular condensates.

Biopolymer condensates often emerge as a multi-droplet state and never coalesce into one large droplet within the experimental timespan. This contradicts the prediction of classical polymer physics which suggests the existence of one large droplet beyond the phase transition. Previous work revealed that the sticker-spacer architecture of biopolymers may dynamically stabilize the multi-droplet state. Here, we simulate the condensate coalescence using metadynamics approach and reveal two distinct physical mechanisms underlying the fusion of droplets. Condensates made of sticker-spacer polymers readily undergo a kinetic arrest when stickers exhibit slow exchange while fast exchanging stickers at similar levels of saturation allow merger to equilibrium states. On the other hand, condensates composed of homopolymers fuse readily until they reach a threshold density. We also show that the inter-condensate exchange of chains offers a general mechanism that drives the fusion. We map the range of mechanisms of kinetic arrest from slow sticker exchange dynamics to density mediated in terms of energetic separation of stickers and spacers. Our predictions appear to be in excellent agreement with recent experiments probing dynamic nature of protein-RNA condensates.

Mapping the Space of Photoswitchable Ligands and Photodruggable Proteins with Computational Modeling.

Light-activated drugs are a promising way to localize biological activity and minimize side effects. However, their development is complicated by the numerous photophysical and biological properties that must be simultaneously optimized. To accelerate the design of photoactive drugs, we describe a procedure that combines ligand-protein docking with chemical property prediction based on machine learning (ML). We apply this procedure to 58 proteins and 9000 photo-drug candidates based on azobenzene cis-trans isomerism. We find that most proteins display a preference for trans isomers over cis and that the binding affinities of nominally active/inactive pairs are in fact highly correlated. These findings have significant value for photopharmacology research, and reinforce the need for virtual screening to identify compounds with rare desirable properties. Further, we combine our procedure with quantum chemical validation to identify promising candidates for the photoactive inhibition of PARP1, an enzyme that is over-expressed in cancer cells. The top compounds are predicted to have long-lived active forms, differential bioactivity, and absorption in the near-infrared therapeutic window.

Co-translational formation of disulfides guides folding of the SARS-CoV-2 receptor binding domain.

Many secreted proteins, including viral proteins, contain multiple disulfide bonds. How disulfide formation is coupled to protein folding in the cell remains poorly understood at the molecular level. Here, we combine experiment and simulation to address this question as it pertains to the SARS-CoV-2 receptor binding domain (RBD). We show that the RBD can only refold reversibly if its native disulfides are present before folding. But in their absence, the RBD spontaneously misfolds into a nonnative, molten-globule-like state that is structurally incompatible with complete disulfide formation and that is highly prone to aggregation. Thus, the RBD native structure represents a metastable state on the protein's energy landscape with reduced disulfides, indicating that nonequilibrium mechanisms are needed to ensure native disulfides form before folding. Our atomistic simulations suggest that this may be achieved via co-translational folding during RBD secretion into the endoplasmic reticulum. Namely, at intermediate translation lengths, native disulfide pairs are predicted to come together with high probability, and thus, under suitable kinetic conditions, this process may lock the protein into its native state and circumvent highly aggregation-prone nonnative intermediates. This detailed molecular picture of the RBD folding landscape may shed light on SARS-CoV-2 pathology and molecular constraints governing SARS-CoV-2 evolution.

Systematic conformation-to-phenotype mapping via limited deep sequencing of proteins.

Non-native conformations drive protein-misfolding diseases, complicate bioengineering efforts, and fuel molecular evolution. No current experimental technique is well suited for elucidating them and their phenotypic effects. Especially intractable are the transient conformations populated by intrinsically disordered proteins. We describe an approach to systematically discover, stabilize, and purify native and non-native conformations, generated in vitro or in vivo, and directly link conformations to molecular, organismal, or evolutionary phenotypes. This approach involves high-throughput disulfide scanning (HTDS) of the entire protein. To reveal which disulfides trap which chromatographically resolvable conformers, we devised a deep-sequencing method for double-Cys variant libraries of proteins that precisely and simultaneously locates both Cys residues within each polypeptide. HTDS of the abundant E. coli periplasmic chaperone HdeA revealed distinct classes of disordered hydrophobic conformers with variable cytotoxicity depending on where the backbone was cross-linked. HTDS can bridge conformational and phenotypic landscapes for many proteins that function in disulfide-permissive environments.

Empowering drug off-target discovery with metabolic and structural analysis.

Elucidating intracellular drug targets is a difficult problem. While machine learning analysis of omics data has been a promising approach, going from large-scale trends to specific targets remains a challenge. Here, we develop a hierarchic workflow to focus on specific targets based on analysis of metabolomics data and growth rescue experiments. We deploy this framework to understand the intracellular molecular interactions of the multi-valent dihydrofolate reductase-targeting antibiotic compound CD15-3. We analyse global metabolomics data utilizing machine learning, metabolic modelling, and protein structural similarity to prioritize candidate drug targets. Overexpression and in vitro activity assays confirm one of the predicted candidates, HPPK (folK), as a CD15-3 off-target. This study demonstrates how established machine learning methods can be combined with mechanistic analyses to improve the resolution of drug target finding workflows for discovering off-targets of a metabolic inhibitor.

Phosphorylation sites are evolutionary checkpoints against liquid-solid transition in protein condensates.

Assemblies of multivalent RNA-binding protein fused in sarcoma (FUS) can exist in the functional liquid-like state as well as less dynamic and potentially toxic amyloid- and hydrogel-like states. How could then cells form liquid-like condensates while avoiding their transformation to amyloids? Here, we show how posttranslational phosphorylation can provide a "handle" that prevents liquid-solid transition of intracellular condensates containing FUS. Using residue-specific coarse-grained simulations, for 85 different mammalian FUS sequences, we show how the number of phosphorylation sites and their spatial arrangement affect intracluster dynamics preventing conversion to amyloids. All atom simulations further confirm that phosphorylation can effectively reduce the ß-sheet propensity in amyloid-prone fragments of FUS. A detailed evolutionary analysis shows that mammalian FUS PLDs are enriched in amyloid-prone stretches compared to control neutrally evolved sequences, suggesting that mammalian FUS proteins evolved to self-assemble. However, in stark contrast to proteins that do not phase-separate for their function, mammalian sequences have phosphosites in close proximity to these amyloid-prone regions. These results suggest that evolution uses amyloid-prone sequences in prion-like domains to enhance phase separation of condensate proteins while enriching phosphorylation sites in close proximity to safeguard against liquid-solid transitions.

Epistasis meets pleiotropy in shaping biophysical protein subspaces associated with antimicrobial resistance.

Protein space is a rich analogy for genotype-phenotype maps, where amino acid sequence is organized into a high-dimensional space that highlights the connectivity between protein variants. It is a useful abstraction for understanding the process of evolution, and for efforts to engineer proteins towards desirable phenotypes. Few framings of protein space consider how higher-level protein phenotypes can be described in terms of their biophysical dimensions, nor do they rigorously interrogate how forces like epistasis-describing the nonlinear interaction between mutations and their phenotypic consequences-manifest across these dimensions. In this study, we deconstruct a low-dimensional protein space of a bacterial enzyme (dihydrofolate reductase; DHFR) into "subspaces" corresponding to a set of kinetic and thermodynamic traits [(kcat, KM, Ki, and Tm (melting temperature)]. We then examine how three mutations (eight alleles in total) display pleiotropy in their interactions across these subspaces. We extend this approach to examine protein spaces across three orthologous DHFR enzymes (Escherichia coli, Listeria grayi, and Chlamydia muridarum), adding a genotypic context dimension through which epistasis occurs across subspaces. In doing so, we reveal that protein space is a deceptively complex notion, and that the process of protein evolution and engineering should consider how interactions between amino acid substitutions manifest across different phenotypic subspaces.

Allosteric communication between ligand binding domains modulates substrate inhibition in adenylate kinase.

Enzymes play a vital role in life processes; they control chemical reactions and allow functional cycles to be synchronized. Many enzymes harness large-scale motions of their domains to achieve tremendous catalytic prowess and high selectivity for specific substrates. One outstanding example is provided by the three-domain enzyme adenylate kinase (AK), which catalyzes phosphotransfer between ATP to AMP. Here we study the phenomenon of substrate inhibition by AMP and its correlation with domain motions. Using single-molecule FRET spectroscopy, we show that AMP does not block access to the ATP binding site, neither by competitive binding to the ATP cognate site nor by directly closing the LID domain. Instead, inhibitory concentrations of AMP lead to a faster and more cooperative domain closure by ATP, leading in turn to an increased population of the closed state. The effect of AMP binding can be modulated through mutations throughout the structure of the enzyme, as shown by the screening of an extensive AK mutant library. The mutation of multiple conserved residues reduces substrate inhibition, suggesting that substrate inhibition is an evolutionary well conserved feature in AK. Combining these insights, we developed a model that explains the complex activity of AK, particularly substrate inhibition, based on the experimentally observed opening and closing rates. Notably, the model indicates that the catalytic power is affected by the microsecond balance between the open and closed states of the enzyme. Our findings highlight the crucial role of protein motions in enzymatic activity.

Systematic conformation-to-phenotype mapping via limited deep-sequencing of proteins.

Non-native conformations drive protein misfolding diseases, complicate bioengineering efforts, and fuel molecular evolution. No current experimental technique is well-suited for elucidating them and their phenotypic effects. Especially intractable are the transient conformations populated by intrinsically disordered proteins. We describe an approach to systematically discover, stabilize, and purify native and non-native conformations, generated in vitro or in vivo, and directly link conformations to molecular, organismal, or evolutionary phenotypes. This approach involves high-throughput disulfide scanning (HTDS) of the entire protein. To reveal which disulfides trap which chromatographically resolvable conformers, we devised a deep-sequencing method for double-Cys variant libraries of proteins that precisely and simultaneously locates both Cys residues within each polypeptide. HTDS of the abundant E. coli periplasmic chaperone HdeA revealed distinct classes of disordered hydrophobic conformers with variable cytotoxicity depending on where the backbone was cross-linked. HTDS can bridge conformational and phenotypic landscapes for many proteins that function in disulfide-permissive environments.

Thermal Half-Lives of Azobenzene Derivatives: Virtual Screening Based on Intersystem Crossing Using a Machine Learning Potential.

Molecular photoswitches are the foundation of light-activated drugs. A key photoswitch is azobenzene, which exhibits trans-cis isomerism in response to light. The thermal half-life of the cis isomer is of crucial importance, since it controls the duration of the light-induced biological effect. Here we introduce a computational tool for predicting the thermal half-lives of azobenzene derivatives. Our automated approach uses a fast and accurate machine learning potential trained on quantum chemistry data. Building on well-established earlier evidence, we argue that thermal isomerization proceeds through rotation mediated by intersystem crossing, and incorporate this mechanism into our automated workflow. We use our approach to predict the thermal half-lives of 19,000 azobenzene derivatives. We explore trends and trade-offs between barriers and absorption wavelengths, and open-source our data and software to accelerate research in photopharmacology.

Different states and the associated fates of biomolecular condensates.

Biomolecular condensates are functional assemblies, which can enrich intrinsically disordered proteins (IDPs) and/or RNAs at concentrations that are orders of magnitude higher than the bulk. In their native functional state, these structures can exist in multiple physical states including liquid-droplet phase, hydrogels, and solid assemblies. On the other hand, an aberrant transition between these physical states can result in loss-of-function or a gain-of-toxic-function. A prime example of such an aberrant transition is droplet aging-a phenomenon where some condensates may progressively transition into less dynamic material states at biologically relevant timescales. In this essay, we review structural and viscoelastic roots of aberrant liquid-solid transitions. Also, we highlight the different checkpoints and experimentally tunable handles, both active (ATP-dependent enzymes, post-translational modifications) and passive (colocalization of RNA molecules), that could alter the material state of assemblies.

Cotranslational formation of disulfides guides folding of the SARS COV-2 receptor binding domain.

Many secreted proteins contain multiple disulfide bonds. How disulfide formation is coupled to protein folding in the cell remains poorly understood at the molecular level. Here, we combine experiment and simulation to address this question as it pertains to the SARS-CoV-2 receptor binding domain (RBD). We show that, whereas RBD can refold reversibly when its disulfides are intact, their disruption causes misfolding into a nonnative molten-globule state that is highly prone to aggregation and disulfide scrambling. Thus, non-equilibrium mechanisms are needed to ensure disulfides form prior to folding in vivo. Our simulations suggest that co-translational folding may accomplish this, as native disulfide pairs are predicted to form with high probability at intermediate lengths, ultimately committing the RBD to its metastable native state and circumventing nonnative intermediates. This detailed molecular picture of the RBD folding landscape may shed light on SARS-CoV-2 pathology and molecular constraints governing SARS-CoV-2 evolution.

Excited state non-adiabatic dynamics of large photoswitchable molecules using a chemically transferable machine learning potential.

Light-induced chemical processes are ubiquitous in nature and have widespread technological applications. For example, photoisomerization can allow a drug with a photo-switchable scaffold such as azobenzene to be activated with light. In principle, photoswitches with desired photophysical properties like high isomerization quantum yields can be identified through virtual screening with reactive simulations. In practice, these simulations are rarely used for screening, since they require hundreds of trajectories and expensive quantum chemical methods to account for non-adiabatic excited state effects. Here we introduce a diabatic artificial neural network (DANN), based on diabatic states, to accelerate such simulations for azobenzene derivatives. The network is six orders of magnitude faster than the quantum chemistry method used for training. DANN is transferable to azobenzene molecules outside the training set, predicting quantum yields for unseen species that are correlated with experiment. We use the model to virtually screen 3100 hypothetical molecules, and identify novel species with high predicted quantum yields. The model predictions are confirmed using high-accuracy non-adiabatic dynamics. Our results pave the way for fast and accurate virtual screening of photoactive compounds.