A computational spatial whole-Cell model for hepatitis B viral infection and drug interactions.

Despite a vaccine, hepatitis B virus (HBV) remains a world-wide source of infections and deaths. We develop a whole-cell computational platform combining spatial and kinetic models describing the infection cycle of HBV in a hepatocyte host. We simulate key parts of the infection cycle with this whole-cell platform for 10 min of biological time, to predict infection progression, map out virus-host and virus-drug interactions. We find that starting from an established infection, decreasing the copy number of the viral envelope proteins shifts the dominant infection pathway from capsid secretion to re-importing the capsids into the nucleus, resulting in more nuclear-localized viral covalently closed circular DNA (cccDNA) and boosting transcription. This scenario can mimic the consequence of drugs designed to manipulate viral gene expression. Mutating capsid proteins facilitates capsid destabilization and disassembly at nuclear pore complexes, resulting in an increase in cccDNA copy number. However, excessive destabilization leads to premature cytoplasmic disassembly and does not increase the cccDNA counts. Finally, our simulations can predict the best drug dosage and its administration timing to reduce the cccDNA counts. Our adaptable computational platform can be parameterized to study other viruses and identify the most central viral pathways that can be targeted by drugs.

Spliceosomal SL1 RNA binding to U1-70K: the role of the extended RRM.

The RNA recognition motif (RRM) occurs widely in RNA-binding proteins, but does not always by itself support full binding. For example, it is known that binding of SL1 RNA to the protein U1-70K in the U1 spliceosomal particle is reduced when a region flanking the RRM is truncated. How the RRM flanking regions that together with the RRM make up an 'extended RRM' (eRRM) contribute to complex stability and structural organization is unknown. We study the U1-70K eRRM bound to SL1 RNA by thermal dissociation and laser temperature jump kinetics; long-time molecular dynamics simulations interpret the experiments with atomistic resolution. Truncation of the helix flanking the RRM on its N-terminal side, 'N-helix,' strongly reduces overall binding, which is further weakened under higher salt and temperature conditions. Truncating the disordered region flanking the RRM on the C-terminal side, 'C-IDR', affects the local binding site. Surprisingly, all-atom simulations show that protein truncation enhances base stacking interactions in the binding site and leaves the overall number of hydrogen bonds intact. Instead, the flanking regions of the eRRM act in a distributed fashion via collective interactions with the RNA when external stresses such as temperature or high salt mimicking osmotic imbalance are applied.

Molecular mechanism of capsid disassembly in hepatitis B virus.

The disassembly of a viral capsid leading to the release of its genetic material into the host cell is a fundamental step in viral infection. In hepatitis B virus (HBV), the capsid consists of identical protein monomers that dimerize and then arrange themselves into pentamers or hexamers on the capsid surface. By applying atomistic molecular dynamics simulation to an entire solvated HBV capsid subjected to a uniform mechanical stress protocol, we monitor the capsid-disassembly process and analyze the process down to the level of individual amino acids in 20 independent simulation replicas. The strain of an isotropic external force, combined with structural fluctuations, causes structurally heterogeneous cracks to appear in the HBV capsid. Analysis of the monomer-monomer interfaces reveals that, in contrast to the expectation from purely mechanical considerations, the cracks mainly occur within hexameric sites, whereas pentameric sites remain largely intact. Only a small subset of the capsid protein monomers, different in each simulation, are engaged in each instance of disassembly. We identify specific residues whose interactions are most readily lost during disassembly; R127, I139, Y132, N136, A137, and V149 are among the hot spots at the interfaces between dimers that lie within hexamers, leading to disassembly. The majority of these hot-spot residues are conserved by evolution, hinting to their importance for disassembly by avoiding overstabilization of capsids.

An in-silico human cell model reveals the influence of spatial organization on RNA splicing.

Spatial organization is a characteristic of all cells, achieved in eukaryotic cells by utilizing both membrane-bound and membrane-less organelles. One of the key processes in eukaryotes is RNA splicing, which readies mRNA for translation. This complex and highly dynamical chemical process involves assembly and disassembly of many molecules in multiple cellular compartments and their transport among compartments. Our goal is to model the effect of spatial organization of membrane-less organelles (specifically nuclear speckles) and of organelle heterogeneity on splicing particle biogenesis in mammalian cells. Based on multiple sources of complementary experimental data, we constructed a spatial model of a HeLa cell to capture intracellular crowding effects. We then developed chemical reaction networks to describe the formation of RNA splicing machinery complexes and splicing processes within nuclear speckles (specific type of non-membrane-bound organelles). We incorporated these networks into our spatially-resolved human cell model and performed stochastic simulations for up to 15 minutes of biological time, the longest thus far for a eukaryotic cell. We find that an increase (decrease) in the number of nuclear pore complexes increases (decreases) the number of assembled splicing particles; and that compartmentalization is critical for the yield of correctly-assembled particles. We also show that a slight increase of splicing particle localization into nuclear speckles leads to a disproportionate enhancement of mRNA splicing and a reduction in the noise of generated mRNA. Our model also predicts that the distance between genes and speckles has a considerable effect on the mRNA production rate, with genes located closer to speckles producing mRNA at higher levels, emphasizing the importance of genome organization around speckles. The HeLa cell model, including organelles and sub-compartments, provides a flexible foundation to study other cellular processes that are strongly modulated by spatiotemporal heterogeneity.

The Conformational Change in Elongation Factor Tu Involves Separation of Its Domains.

Elongation factor Tu (EF-Tu) is a highly conserved GTPase that is responsible for supplying the aminoacylated tRNA to the ribosome. Upon binding to the ribosome, EF-Tu undergoes GTP hydrolysis, which drives a major conformational change, triggering the release of aminoacylated tRNA to the ribosome. Using a combination of molecular simulation techniques, we studied the transition between the pre- and post-hydrolysis structures through two distinct pathways. We show that the transition free energy is minimal along a non-intuitive pathway that involves "separation" of the GTP binding domain (domain 1) from the OB folds (domains 2 and 3), followed by domain 1 rotation, and, eventually, locking the EF-Tu conformation in the post-hydrolysis state. The domain separation also leads to a slight extension of the linker connecting domain 1 to domain 2. Using docking tools and correlation-based analysis, we identified and characterized the EF-Tu conformations that release the tRNA. These calculations suggest that EF-Tu can release the tRNA before the domains separate and after domain 1 rotates by 25°. We also examined the EF-Tu conformations in the context of the ribosome. Given the high degrees of sequence similarity with other translational GTPases, we predict a similar separation mechanism is followed.

Role of Electrostatics in Protein-RNA Binding: The Global vs the Local Energy Landscape.

U1A protein-stem loop 2 RNA association is a basic step in the assembly of the spliceosomal U1 small nuclear ribonucleoprotein. Long-range electrostatic interactions due to the positive charge of U1A are thought to provide high binding affinity for the negatively charged RNA. Short range interactions, such as hydrogen bonds and contacts between RNA bases and protein side chains, favor a specific binding site. Here, we propose that electrostatic interactions are as important as local contacts in biasing the protein-RNA energy landscape toward a specific binding site. We show by using molecular dynamics simulations that deletion of two long-range electrostatic interactions (K22Q and K50Q) leads to mutant-specific alternative RNA bound states. One of these states preserves short-range interactions with aromatic residues in the original binding site, while the other one does not. We test the computational prediction with experimental temperature-jump kinetics using a tryptophan probe in the U1A-RNA binding site. The two mutants show the distinct predicted kinetic behaviors. Thus, the stem loop 2 RNA has multiple binding sites on a rough RNA-protein binding landscape. We speculate that the rough protein-RNA binding landscape, when biased to different local minima by electrostatics, could be one way that protein-RNA interactions evolve toward new binding sites and novel function.

Estimation of Relative Protein-RNA Binding Strengths from Fluctuations in the Bound State.

Protein-RNA complexes are increasingly important in our understanding of cell signaling, metabolism, and transcription. Electrostatic interactions play dominant role in stabilizing such complexes. Using conventional computational approaches, very long simulations of both bound and unbound states are required to obtain accurate estimates of complex dissociation constants (Kd). Here, we derive a simple formula that offers an alternative approach based on the theory of fluctuations. Our method extracts a strong correlate to experimental Kd values using short molecular dynamics simulations of the bound complex only. To test our method, we compared the computed relative Kd values to our experimentally measured values for the U1A-Stem Loop 2 (SL2) RNA complex, which is one of the most-studied protein-RNA complexes. Additionally we also included several experimental values from the literature, to enlarge the data set. We obtain a correlation of r = 0.93 between theoretical and measured estimates of Kd values of the mutated U1A protein-RNA complexes relative to the wild type dissociation constant. The proposed method increases the efficiency of relative Kd values estimation for multiple protein mutants, allowing its applicability to protein engineering projects.

Permeability Coefficients of Lipophilic Compounds Estimated by Computer Simulations.

The ability of a drug to cross the intestine-blood barrier is a key quantity for drug design and employment and is normally quantified by the permeability coefficient P, often evaluated in the so-called Caco-2 assay. This assay is based on measuring the initial growth rate of the concentration of the drug beyond the cellular barrier but not its steady-state flux through the membrane. This might lead to confusion since, in the case of lipophilic drugs, the initial slope is strongly affected by the retention of the drug in the membrane. This effect is well known but seldom considered in the assay. Here, we exploit all-atoms molecular dynamics and bias exchange metadynamics to calculate the concentration of two lipophilic drugs across a model membrane as a function of time. This allows estimating both the steady-state flux and the initial slope of the concentration growth and comparing Caco-2 and steady-state estimates of P. We show that our computational procedure is able to reproduce the experimental values, although these may differ from the permeability coefficients by orders of magnitude. Our findings are generalized by a simplified one-dimensional model of the permeation process that may act as a roadmap to assess which measure of membrane permeability would be more appropriate and, consequently, whether retention corrections should be included in estimates based on Caco-2 assays.

Analysis of the Resistance Mechanism of a Benzoxaborole Inhibitor Reveals Insight into the Leucyl-tRNA Synthetase Editing Mechanism.

A new class of antimicrobial benzoxaborole compounds was identified as a potent inhibitor of leucyl-tRNA synthetase (LeuRS) and therefore of protein synthesis. In a novel mechanism, AN2690 (5-fluoro-1,3-dihydro-1-hydroxy-2,1-benzoxaborole) blocks fungal cytoplasmic LeuRS by covalently trapping tRNA(Leu) in the editing site of the enzyme's CP1 domain. However, some resistant mutation sites are located outside of the CP1 hydrolytic editing active site. Thus, their mode of action that undermines drug inhibition was not understood. A combination of X-ray crystallography, molecular dynamics, metadynamics, biochemical experiments, and mutational analysis of a distal benzoxaborole-resistant mutant uncovered a eukaryote-specific tyrosine "switch" that is critical to tRNA-dependent post-transfer editing. The tyrosine "switch" has three states that shift between interactions with a lysine and the 3'-hydroxyl of the tRNA terminus, to inhibit or promote post-transfer editing. The oxaborole's mechanism of action capitalizes upon one of these editing active site states. This tunable editing mechanism in eukaryotic and archaeal LeuRSs is proposed to facilitate precise quality control of aminoacylation fidelity. These mechanistic distinctions could also be capitalized upon for development of the benzoxaboroles as a broad spectrum antibacterial.

Native conformational dynamics of the spliceosomal U1A protein.

The complex of spliceosomal U1A protein and its cognate SL2 RNA is a prototype system for protein-RNA binding studies. A major question is whether U1A protein alone is capable of undergoing conformational dynamics similar to structural rearrangements upon RNA binding. Using a fast temperature jump and tryptophan fluorescence detection, we uncover a ∼20 µs conformational transition for the Lys22Gln/Phe56Trp-only mutant of U1A, yet a Phe56Trp-only control mutant does not show the transition. To explain this observation, we performed extensive molecular dynamics (MD) simulations. The simulations explain why only the Lys22Gln/Phe56Trp-only mutant shows a fluorescence signal: in the other mutant, the tryptophan probe is not quenched upon structural rearrangement. The simulations support helix C movement as the underlying structural rearrangement, although the simulated time scale is faster than experimentally detected. On the basis of our MD results, we propose a reversible two-pathway three-state transition for the helix C movement and assign T-jump kinetics to a closed to semi-closed transition of the helix. Our result provides a specific example of how alternative protein conformations on the native side of the folding barrier can be functionally important, for example in conformational selection by a binding partner.

A novel approach to the investigation of passive molecular permeation through lipid bilayers from atomistic simulations.

Predicting the permeability coefficient (P) of drugs permeating through the cell membrane is of paramount importance in drug discovery. We here propose an approach for calculating P based on bias-exchange metadynamics. The approach allows constructing from atomistic simulations a model of permeation taking explicitly into account not only the "trivial" reaction coordinate, the position of the drug along the direction normal to the lipid membrane plane, but also other degrees of freedom, for example, the torsional angles of the permeating molecule, or variables describing its solvation/desolvation. This allows deriving an accurate picture of the permeation process, and constructing a detailed molecular model of the transition state, making a rational control of permeation properties possible. We benchmarked this approach on the permeation of ethanol molecules through a POPC membrane, showing that the value of P calculated with our model agrees with the one calculated by a long unbiased molecular dynamics of the same system.