Molecular Dynamics Simulations of Nucleic Acids with a Generalized Born Solvation Model

Accurate structure prediction of peptide-MHC complexes for identifying highly immunogenic antigens

Designing an optimal HIV-1 vaccine faces the challenge of identifying antigens that induce a broad immune capacity. One factor to control the breadth of T cell responses is the surface morphology of a peptide-MHC complex. Here, we present an in silico protocol for predicting peptide-MHC structure. A robust signature of a conformational transition was identified during all-atom molecular dynamics, which results in a model with high accuracy. A large test set was used in constructing our protocol and we went another step further using a blind test with a wild-type peptide and two highly immunogenic mutants, which predicted substantial conformational changes in both mutants. The center residues at position five of the analogs were configured to be accessible to solvent, forming a prominent surface, while the residue of the wild-type peptide was to point laterally toward the side of the binding cleft. We then experimentally determined the structures of the blind test set, using high resolution of X-ray crystallography, which verified predicted conformational changes. Our observation strongly supports a positive association of the surface morphology of a peptide-MHC complex to its immunogenicity. Our study offers the prospect of enhancing immunogenicity of vaccines by identifying MHC binding immunogens.

Assessing the accuracy and performance of implicit solvent models for drug molecules: conformational ensemble approaches

The accuracy and performance of implicit solvent methods for solvation free energy calculations were assessed on a set of 20 neutral drug molecules. Molecular dynamics (MD) provided ensembles of conformations in water and water-saturated octanol. The solvation free energies were calculated by popular implicit solvent models based on quantum mechanical (QM) electronic densities (COSMO-RS, MST, SMD) as well as on molecular mechanical (MM) point-charge models (GB, PB). The performance of the implicit models was tested by a comparison with experimental water-octanol transfer free energies (ΔG(ow)) by using single- and multiconformation approaches. MD simulations revealed difficulties in a priori estimation of the flexibility features of the solutes from simple structural descriptors, such as the number of rotatable bonds. An increasing accuracy of the calculated ΔG(ow) was observed in the following order: GB1 ~ PB < GB7 ≪ MST < SMD ~ COSMO-RS with a clear distinction identified between MM- and QM-based models, although for the set excluding three largest molecules, the differences among COSMO-RS, MST, and SMD were negligible. It was shown that the single-conformation approach applied to crystal geometries provides a rather accurate estimate of ΔG(ow) for rigid molecules yet fails completely for the flexible ones. The multiconformation approaches improved the performance, but only when the deformation contribution was ignored. It was revealed that for large-scale calculations on small molecules a recent GB model, GB7, provided a reasonable accuracy/speed ratio. In conclusion, the study contributes to the understanding of solvation free energy calculations for physical and medicinal chemistry applications.

Improved Generalized Born Solvent Model Parameters for Protein Simulations

The generalized Born (GB) model is one of the fastest implicit solvent models and it has become widely adopted for Molecular Dynamics (MD) simulations. This speed comes with tradeoffs, and many reports in the literature have pointed out weaknesses with GB models. Because the quality of a GB model is heavily affected by empirical parameters used in calculating solvation energy, in this work we have refit these parameters for GB-Neck, a recently developed GB model, in order to improve the accuracy of both the solvation energy and effective radii calculations. The data sets used for fitting are significantly larger than those used in the past. Comparing to other pairwise GB models like GB-OBC and the original GB-Neck, the new GB model (GB-Neck2) has better agreement to Poisson-Boltzmann (PB) in terms of reproducing solvation energies for a variety of systems ranging from peptides to proteins. Secondary structure preferences are also in much better agreement with those obtained from explicit solvent MD simulations. We also obtain near-quantitative reproduction of experimental structure and thermal stability profiles for several model peptides with varying secondary structure motifs. Extension to non-protein systems will be explored in the future.

PowerBorn: A Barnes-Hut Tree Implementation for Accurate and Efficient Born Radii Computation

Implicit solvent models are one of the standard tools in computational biophysics. While Poisson-Boltzmann methods offer highly accurate results within this framework, generalized Born models have been used due to their higher computational efficiency in many (bio)molecular simulations, where computational power is a limiting factor. In recent years, there have been remarkable advances to reduce some deficiencies in the generalized Born models. On the other hand, these advances come at an increased computational cost that contrasts the reasons for choosing generalized Born models over Poisson-Boltzmann methods. To address this performance issue, we present a new algorithm for Born radii computation, one performance critical part in the evaluation of generalized Born models, which is based on a Barnes-Hut tree code scheme. We show that an implementation of this algorithm provides accurate Born radii and polar solvation free energies in comparison to Poisson-Boltzmann computations, while delivering up to an order of magnitude better performance over existing, similarly accurate methods. The C++ implementation of this algorithm will be available at http://www.int.kit.edu/nanosim/ .

The C-terminal region Mesd peptide mimics full-length Mesd and acts as an inhibitor of Wnt/β-catenin signaling in cancer cells

While Mesd was discovered as a specialized molecular endoplasmic reticulum chaperone for the Wnt co-receptors LRP5 and LRP6, recombinant Mesd protein is able to bind to mature LRP5 and LRP6 on the cell surface and acts as a universal antagonist of LRP5/6 modulators. In our previous study, we found that the C-terminal region of Mesd, which is absent in sequences from invertebrates, is necessary and sufficient for binding to mature LRP6 on the cell surface. In the present studies, we further characterized the interaction between the C-terminal region Mesd peptide and LRP5/6. We found that Mesd C-terminal region-derived peptides block Mesd binding to LRP5 at the cell surface too. We also showed that there are two LRP5/6 binding sites within Mesd C-terminal region which contain several positively charged residues. Moreover, we demonstrated that the Mesd C-terminal region peptide, like the full-length Mesd protein, blocked Wnt 3A- and Rspodin1-induced Wnt/β-catenin signaling in LRP5- and LRP6- expressing cells, suppressed Wnt/β-catenin signaling in human breast HS578T cells and prostate cancer PC-3 cells, and inhibited cancer cell proliferation, although the full-length Mesd protein is more potent than its peptide. Finally, we found that treatment of the full-length Mesd protein and its C-terminal region peptide significantly increased chemotherapy agent Adriamycin-induced cytotoxicity in HS578T and PC-3 cells. Together, our results suggest that Mesd C-terminal region constitutes the major LRP5/6-binding domain, and that Mesd protein and its C-terminal region peptide have a potential therapeutic value in cancer.

Structural integrity of the B24 site in human insulin is important for hormone functionality

Despite the recent first structural insight into the insulin-insulin receptor complex, the role of the C terminus of the B-chain of insulin in this assembly remains unresolved. Previous studies have suggested that this part of insulin must rearrange to reveal amino acids crucial for interaction with the receptor. The role of the invariant Phe(B24), one of the key residues of the hormone, in this process remains unclear. For example, the B24 site functionally tolerates substitutions to D-amino acids but not to L-amino acids. Here, we prepared and characterized a series of B24-modified insulin analogues, also determining the structures of [D-HisB24]-insulin and [HisB24]-insulin. The inactive [HisB24]-insulin molecule is remarkably rigid due to a tight accommodation of the L-His side chain in the B24 binding pocket that results in the stronger tethering of B25-B28 residues to the protein core. In contrast, the highly active [D-HisB24]-insulin is more flexible, and the reverse chirality of the B24C(α) atom swayed the D-His(B24) side chain into the solvent. Furthermore, the pocket vacated by Phe(B24) is filled by Phe(B25), which mimics the Phe(B24) side and main chains. The B25→B24 downshift results in a subsequent downshift of Tyr(B26) into the B25 site and the departure of B26-B30 residues away from the insulin core. Our data indicate the importance of the aromatic L-amino acid at the B24 site and the structural invariance/integrity of this position for an effective binding of insulin to its receptor. Moreover, they also suggest limited, B25-B30 only, unfolding of the C terminus of the B-chain upon insulin activation.

Two-phase stretching of molecular chains

Although stretching of most polymer chains leads to rather featureless force-extension diagrams, some, notably DNA, exhibit nontrivial behavior with a distinct plateau region. Here, we propose a unified theory that connects force-extension characteristics of the polymer chain with the convexity properties of the extension energy profile of its individual monomer subunits. Namely, if the effective monomer deformation energy as a function of its extension has a nonconvex (concave up) region, the stretched polymer chain separates into two phases: the weakly and strongly stretched monomers. Simplified planar and 3D polymer models are used to illustrate the basic principles of the proposed model. Specifically, we show rigorously that, when the secondary structure of a polymer is mostly caused by weak noncovalent interactions, the stretching is two phase, and the force-stretching diagram has the characteristic plateau. We then use realistic coarse-grained models to confirm the main findings and make direct connection to the microscopic structure of the monomers. We show in detail how the two-phase scenario is realized in the α-helix and DNA double helix. The predicted plateau parameters are consistent with single-molecules experiments. Detailed analysis of DNA stretching shows that breaking of Watson-Crick bonds is not necessary for the existence of the plateau, although some of the bonds do break as the double helix extends at room temperature. The main strengths of the proposed theory are its generality and direct microscopic connection.

Structure and interactions of the human programmed cell death 1 receptor

PD-1, a receptor expressed by T cells, B cells, and monocytes, is a potent regulator of immune responses and a promising therapeutic target. The structure and interactions of human PD-1 are, however, incompletely characterized. We present the solution nuclear magnetic resonance (NMR)-based structure of the human PD-1 extracellular region and detailed analyses of its interactions with its ligands, PD-L1 and PD-L2. PD-1 has typical immunoglobulin superfamily topology but differs at the edge of the GFCC′ sheet, which is flexible and completely lacks a C″ strand. Changes in PD-1 backbone NMR signals induced by ligand binding suggest that, whereas binding is centered on the GFCC′ sheet, PD-1 is engaged by its two ligands differently and in ways incompletely explained by crystal structures of mouse PD-1·ligand complexes. The affinities of these interactions and that of PD-L1 with the costimulatory protein B7-1, measured using surface plasmon resonance, are significantly weaker than expected. The 3–4-fold greater affinity of PD-L2 versus PD-L1 for human PD-1 is principally due to the 3-fold smaller dissociation rate for PD-L2 binding. Isothermal titration calorimetry revealed that the PD-1/PD-L1 interaction is entropically driven, whereas PD-1/PD-L2 binding has a large enthalpic component. Mathematical simulations based on the biophysical data and quantitative expression data suggest an unexpectedly limited contribution of PD-L2 to PD-1 ligation during interactions of activated T cells with antigen-presenting cells. These findings provide a rigorous structural and biophysical framework for interpreting the important functions of PD-1 and reveal that potent inhibitory signaling can be initiated by weakly interacting receptors.

Protein l-isoaspartyl-O-methyltransferase of Vibrio cholerae: interaction with cofactors and effect of osmolytes on unfolding

Protein l-isoaspartyl-O-methyltransferase (PIMT) is an ubiquitous enzyme widely distributed in cells and plays a role in the repair of deamidated and isomerized proteins. In this study, we show that this enzyme is present in cytosolic extract of Vibrio cholerae, an enteric pathogenic Gram-negative bacterium and is enzymatically active. Additionally, we focus on the detailed biophysical characterization of the recombinant PIMT from V. cholerae to gain insight into its structure, stability and the cofactor binding. The equilibrium denaturation of PIMT has been studied using tryptophan fluorescence and CD spectroscopy. The far- and near-UV CD, as well as fluorescence experiments reveal the presence of a non-native intermediate in the folding pathway. Binding of the hydrophobic fluorescent probe, bis-ANS, to the intermediate occurs with high affinity because of the exposure of the hydrophobic clusters during the unfolding process. The existence of the probable intermediate has also been confirmed from limited tryptic digestion and DLS experiments. The protein shows higher binding affinity for AdoHcy, in comparison to AdoMet, and the binding increases the midpoint of thermal unfolding by 6 and 5 °C, respectively. Modeling and molecular dynamics simulations also support the higher stability of the protein in presence of AdoHcy.

Computational study of the effects of protein tyrosine nitrations on the catalytic activity of human thymidylate synthase

Tyrosine nitration is a widespread post-translational modification capable of affecting both the function and structure of the host protein molecule. Enzyme thymidylate synthase (TS), a homodimer, is a molecular target for anticancer therapy. Recently purified TS preparations, isolated from mammalian tissues, were found to be nitrated, suggesting this modification to appear endogenously in normal and tumor tissues. Moreover, human TS (hTS) nitration in vitro led to a by twofold lowered catalytic activity following nitration in average of 1 tyrosine residue per monomer (Dąbrowska-Maś et al. in Org Biomol Chem 10:323-331, 2012), with the modification identified by mass spectrometry at seven different sites (Y33, Y65, Y135, Y213, Y230, Y258 and Y301). In the present paper, combined computational approach, including molecular and essential dynamics and free energy computations, was used to predict the influence on the activity of hTS of nitration of each of the seven tyrosine residues. The simulations were based on the crystal structure of hTS ternary complex with dUMP and Tomudex (PDB code: 1I00), with the Tomudex molecule replaced by the molecule of TS cofactor analogue, tetrahydrofolate. The present results indicate that while with nitration of five out of seven residues (Y33, Y135, Y230, Y258 and Y301), single residue modification appears to have a strong reducing effect on the activity, with the remaining two, Y65 and Y213, no or a weaker influence is apparent. Taken together, these results demonstrate that tyrosine nitrations in the hTS enzyme show clear tendency to influence the structure and dynamics and, in turn, catalytic properties of the host enzyme. These effects are overall distance-dependent.

Studies of base pair sequence effects on DNA solvation based on all-atom molecular dynamics simulations

Detailed analyses of the sequence-dependent solvation and ion atmosphere of DNA are presented based on molecular dynamics (MD) simulations on all the 136 unique tetranucleotide steps obtained by the ABC consortium using the AMBER suite of programs. Significant sequence effects on solvation and ion localization were observed in these simulations. The results were compared to essentially all known experimental data on the subject. Proximity analysis was employed to highlight the sequence dependent differences in solvation and ion localization properties in the grooves of DNA. Comparison of the MD-calculated DNA structure with canonical A- and B-forms supports the idea that the G/C-rich sequences are closer to canonical A- than B-form structures, while the reverse is true for the poly A sequences, with the exception of the alternating ATAT sequence. Analysis of hydration density maps reveals that the flexibility of solute molecule has a significant effect on the nature of observed hydration. Energetic analysis of solute-solvent interactions based on proximity analysis of solvent reveals that the GC or CG base pairs interact more strongly with water molecules in the minor groove of DNA that the AT or TA base pairs, while the interactions of the AT or TA pairs in the major groove are stronger than those of the GC or CG pairs. Computation of solvent-accessible surface area of the nucleotide units in the simulated trajectories reveals that the similarity with results derived from analysis of a database of crystallographic structures is excellent. The MD trajectories tend to follow Manning's counterion condensation theory, presenting a region of condensed counterions within a radius of about 17 A from the DNA surface independent of sequence. The GC and CG pairs tend to associate with cations in the major groove of the DNA structure to a greater extent than the AT and TA pairs. Cation association is more frequent in the minor groove of AT than the GC pairs. In general, the observed water and ion atmosphere around the DNA sequences is the MD simulation is in good agreement with experimental observations.

In silico docking reveals possible Riluzole binding sites on Nav1.6 sodium channel: implications for amyotrophic lateral sclerosis therapy

Amyotrophic lateral sclerosis (ALS) is a common neurodegenerative disorder characterized mainly by a progressive loss of motor neurons. Glutamate excitotoxicity is likely the main cause of neuronal death, and Riluzole interferes with glutamate-mediated transmission. Thus, in such independent pathway, these effects may be partly due to inactivation of voltage-dependent sodium channels. Here we predict the structural model of the interaction and report the possible binding sites of Riluzole on Nav1.6 channel. The docked complexes were subjected to minimization and we further investigated the key interacting residues, binding free energies, pairing bridge determination, folding pattern, hydrogen bounding formation, hydrophobic contacts and flexibilities. Our results demonstrate that Riluzole interacts with the Nav1.6 channel, more specifically in the key residues TYR 1787, LEU 1843 and GLN 1799, suggesting possible cellular implications driven by these amino acids on Riluzole-Nav1.6 interaction, which may serve as an important output for a more specific and experimental drug design therapy against ALS.

Discrimination of ligands with different flexibilities resulting from the plasticity of the binding site in tubulin

Tubulin, an α,β heterodimer, has four distinct ligand binding sites (for paclitaxel, peloruside/laulimalide, vinca, and colchicine). The site where colchicine binds is a promising drug target for arresting cell division and has been observed to accommodate compounds that are structurally diverse but possess comparable affinity. This investigation, using two such structurally different ligands as probes (one being colchicine itself and another, TN16), aims to provide insight into the origin of this diverse acceptability to provide a better perspective for the design of novel therapeutic molecules. Thermodynamic measurements reveal interesting interplay between entropy and enthalpy. Although both these parameters are favourable for TN16 binding (ΔH < 0, ΔS > 0), but the magnitude of entropy has the determining role for colchicine binding as its enthalpic component is destabilizing (ΔH > 0, ΔS > 0). Molecular dynamics simulation provides atomistic insight into the mechanism, pointing to the inherent flexibility of the binding pocket that can drastically change its shape depending on the ligand that it accepts. Simulation shows that in the complexed states both the ligands have freedom to move within the binding pocket; colchicine can switch its interactions like a "flying trapeze", whereas TN16 rocks like a "swing cradle", both benefiting entropically, although in two different ways. Additionally, the experimental results with respect to the role of solvation entropy correlate well with the computed difference in the hydration: water molecules associated with the ligands are released upon complexation. The complementary role of van der Waals packing versus flexibility controls the entropy-enthalpy modulations. This analysis provides lessons for the design of new ligands that should balance between the "better fit" and "flexibility"', instead of focusing only on the receptor-ligand interactions.

Combining H/D exchange mass spectroscopy and computational docking reveals extended DNA-binding surface on uracil-DNA glycosylase

X-ray crystallography provides excellent structural data on protein-DNA interfaces, but crystallographic complexes typically contain only small fragments of large DNA molecules. We present a new approach that can use longer DNA substrates and reveal new protein-DNA interactions even in extensively studied systems. Our approach combines rigid-body computational docking with hydrogen/deuterium exchange mass spectrometry (DXMS). DXMS identifies solvent-exposed protein surfaces; docking is used to create a 3-dimensional model of the protein-DNA interaction. We investigated the enzyme uracil-DNA glycosylase (UNG), which detects and cleaves uracil from DNA. UNG was incubated with a 30 bp DNA fragment containing a single uracil, giving the complex with the abasic DNA product. Compared with free UNG, the UNG-DNA complex showed increased solvent protection at the UNG active site and at two regions outside the active site: residues 210-220 and 251-264. Computational docking also identified these two DNA-binding surfaces, but neither shows DNA contact in UNG-DNA crystallographic structures. Our results can be explained by separation of the two DNA strands on one side of the active site. These non-sequence-specific DNA-binding surfaces may aid local uracil search, contribute to binding the abasic DNA product and help present the DNA product to APE-1, the next enzyme on the DNA-repair pathway.

Binding mechanism between Hsp90 and Sgt1 explored by homology modeling and molecular dynamics simulations in rice

The Hsp90 (for heat shock protein90) and the Sgt1 (for suppressor of the G2 allele of skp1) are widely distributed in animals, yeast, and plants. The former functions as molecular chaperon activating a series of client proteins, the latter functions as an adaptor protein participating in multiple biological processes such as immunity response through interactions with different protein complexes. In the present study, we have constructed a homology model of Hsp90-Sgt1 complex in rice based on a recently resolved structure from barley and Arabidopsis to explore its binding mechanisms and to understand the detailed interaction profile. A total of 20 ns explicit solvent molecular dynamics simulations combined with MM-GBSA computations and virtual alanine scanning were performed for the modeled complex. In the final structure, three strong salt bridges were found between OsHsp90 and OsSgt1, D217(OsHsp90)-K186(OsSgt1), D218(OsHsp90)-K237(OsSgt1) and K161(OsHsp90)-E239(OsSgt1). Besides, residue Y173 of OsSgt1 played a vital role in the interactions with OsHsp90, the detailed interactions were discussed. These results would help us understand the critical features determining the Hsp90-Sgt1 binding process.

Differential geometry based solvation model. III. Quantum formulation

Solvation is of fundamental importance to biomolecular systems. Implicit solvent models, particularly those based on the Poisson-Boltzmann equation for electrostatic analysis, are established approaches for solvation analysis. However, ad hoc solvent-solute interfaces are commonly used in the implicit solvent theory. Recently, we have introduced differential geometry based solvation models which allow the solvent-solute interface to be determined by the variation of a total free energy functional. Atomic fixed partial charges (point charges) are used in our earlier models, which depends on existing molecular mechanical force field software packages for partial charge assignments. As most force field models are parameterized for a certain class of molecules or materials, the use of partial charges limits the accuracy and applicability of our earlier models. Moreover, fixed partial charges do not account for the charge rearrangement during the solvation process. The present work proposes a differential geometry based multiscale solvation model which makes use of the electron density computed directly from the quantum mechanical principle. To this end, we construct a new multiscale total energy functional which consists of not only polar and nonpolar solvation contributions, but also the electronic kinetic and potential energies. By using the Euler-Lagrange variation, we derive a system of three coupled governing equations, i.e., the generalized Poisson-Boltzmann equation for the electrostatic potential, the generalized Laplace-Beltrami equation for the solvent-solute boundary, and the Kohn-Sham equations for the electronic structure. We develop an iterative procedure to solve three coupled equations and to minimize the solvation free energy. The present multiscale model is numerically validated for its stability, consistency and accuracy, and is applied to a few sets of molecules, including a case which is difficult for existing solvation models. Comparison is made to many other classic and quantum models. By using experimental data, we show that the present quantum formulation of our differential geometry based multiscale solvation model improves the prediction of our earlier models, and outperforms some explicit solvation model.

The binding affinities of proteins interacting with the PDZ domain of PICK1

Protein interacting with C kinase (PICK1) is well conserved throughout evolution and plays a critical role in synaptic plasticity by regulating the trafficking and posttranslational modification of its interacting proteins. PICK1 contains a single PSD95/DlgA/Zo-1 (PDZ) protein-protein interaction domain, which is promiscuous and shown to interact with over 60 proteins, most of which play roles in neuronal function. Several reports have suggested the role of PICK1 in disorders such as epilepsy, pain, brain trauma and stroke, drug abuse and dependence, schizophrenia and psychosis. Importantly, lead compounds that block PICK1 interactions are also now becoming available. Here, a new modeling approach was developed to investigate binding affinities of PDZ interactions. Using these methods, the binding affinities of all major PICK1 interacting proteins are reported and the effects of PICK1 mutations on these interactions are described. These modeling methods have important implications in defining the binding properties of proteins interacting with PICK1 as well as the general structural requirements of PDZ interactions. The study also provides modeling methods to support in the drug design of ligands for PDZ domains, which may further aid in development of the family of PDZ domains as a drug target.

Differential geometry based solvation model II: Lagrangian formulation

Solvation is an elementary process in nature and is of paramount importance to more sophisticated chemical, biological and biomolecular processes. The understanding of solvation is an essential prerequisite for the quantitative description and analysis of biomolecular systems. This work presents a Lagrangian formulation of our differential geometry based solvation models. The Lagrangian representation of biomolecular surfaces has a few utilities/advantages. First, it provides an essential basis for biomolecular visualization, surface electrostatic potential map and visual perception of biomolecules. Additionally, it is consistent with the conventional setting of implicit solvent theories and thus, many existing theoretical algorithms and computational software packages can be directly employed. Finally, the Lagrangian representation does not need to resort to artificially enlarged van der Waals radii as often required by the Eulerian representation in solvation analysis. The main goal of the present work is to analyze the connection, similarity and difference between the Eulerian and Lagrangian formalisms of the solvation model. Such analysis is important to the understanding of the differential geometry based solvation model. The present model extends the scaled particle theory of nonpolar solvation model with a solvent-solute interaction potential. The nonpolar solvation model is completed with a Poisson-Boltzmann (PB) theory based polar solvation model. The differential geometry theory of surfaces is employed to provide a natural description of solvent-solute interfaces. The optimization of the total free energy functional, which encompasses the polar and nonpolar contributions, leads to coupled potential driven geometric flow and PB equations. Due to the development of singularities and nonsmooth manifolds in the Lagrangian representation, the resulting potential-driven geometric flow equation is embedded into the Eulerian representation for the purpose of computation, thanks to the equivalence of the Laplace-Beltrami operator in the two representations. The coupled partial differential equations (PDEs) are solved with an iterative procedure to reach a steady state, which delivers desired solvent-solute interface and electrostatic potential for problems of interest. These quantities are utilized to evaluate the solvation free energies and protein-protein binding affinities. A number of computational methods and algorithms are described for the interconversion of Lagrangian and Eulerian representations, and for the solution of the coupled PDE system. The proposed approaches have been extensively validated. We also verify that the mean curvature flow indeed gives rise to the minimal molecular surface and the proposed variational procedure indeed offers minimal total free energy. Solvation analysis and applications are considered for a set of 17 small compounds and a set of 23 proteins. The salt effect on protein-protein binding affinity is investigated with two protein complexes by using the present model. Numerical results are compared to the experimental measurements and to those obtained by using other theoretical methods in the literature.

Fast Analytical Methods for Macroscopic Electrostatic Models in Biomolecular Simulations

We review recent developments of fast analytical methods for macroscopic electrostatic calculations in biological applications, including the Poisson-Boltzmann (PB) and the generalized Born models for electrostatic solvation energy. The focus is on analytical approaches for hybrid solvation models, especially the image charge method for a spherical cavity, and also the generalized Born theory as an approximation to the PB model. This review places much emphasis on the mathematical details behind these methods.

Proteome-wide detection of Abl1 SH3-binding peptides by integrating computational prediction and peptide microarray

Protein-protein interactions are essential for regulating almost all aspects of cellular functions. Many of these interactions are mediated by weak and transient protein domain-peptide binding, but they are often under-represented in high throughput screening of protein-protein interactions using techniques such as yeast two-hybrid and mass spectrometry. On the other hand, computational predictions and in vitro binding assays are valuable in providing clues of in vivo interactions. We present here a systematic approach that integrates computer modeling and a peptide microarray technology to identify binding peptides of the SH3 domain of the tyrosine kinase Abl1 in the human proteome. Our study provides a comprehensive list of candidate interacting partners for the Abl1 protein, among which the presence of numerous methyltransferases and RNA splicing proteins may suggest a novel function of Abl1 in chromatin remodeling and RNA processing. This study illustrates a powerful approach for integrating computational and experimental methods to detect protein interactions mediated by domain-peptide recognition.

Evaluation of DNA Force Fields in Implicit Solvation

DNA structural deformations and dynamics are crucial to its interactions in the cell. Theoretical simulations are essential tools to explore the structure, dynamics, and thermodynamics of biomolecules in a systematic way. Molecular mechanics force fields for DNA have benefited from constant improvements during the last decades. Several studies have evaluated and compared available force fields when the solvent is modeled by explicit molecules. On the other hand, few systematic studies have assessed the quality of duplex DNA models when implicit solvation is employed. The interest of an implicit modeling of the solvent consists in the important gain in the simulation performance and conformational sampling speed. In this study, respective influences of the force field and the implicit solvation model choice on DNA simulation quality are evaluated. To this end, extensive implicit solvent duplex DNA simulations are performed, attempting to reach both conformational and sequence diversity convergence. Structural parameters are extracted from simulations and statistically compared to available experimental and explicit solvation simulation data. Our results quantitatively expose the respective strengths and weaknesses of the different DNA force fields and implicit solvation models studied. This work can lead to the suggestion of improvements to current DNA theoretical models.

Surveying implicit solvent models for estimating small molecule absolute hydration free energies

Implicit solvent models are powerful tools in accounting for the aqueous environment at a fraction of the computational expense of explicit solvent representations. Here, we compare the ability of common implicit solvent models (TC, OBC, OBC2, GBMV, GBMV2, GBSW, GBSW/MS, GBSW/MS2 and FACTS) to reproduce experimental absolute hydration free energies for a series of 499 small neutral molecules that are modeled using AMBER/GAFF parameters and AM1-BCC charges. Given optimized surface tension coefficients for scaling the surface area term in the nonpolar contribution, most implicit solvent models demonstrate reasonable agreement with extensive explicit solvent simulations (average difference 1.0-1.7 kcal/mol and R(2)=0.81-0.91) and with experimental hydration free energies (average unsigned errors=1.1-1.4 kcal/mol and R(2)=0.66-0.81). Chemical classes of compounds are identified that need further optimization of their ligand force field parameters and others that require improvement in the physical parameters of the implicit solvent models themselves. More sophisticated nonpolar models are also likely necessary to more effectively represent the underlying physics of solvation and take the quality of hydration free energies estimated from implicit solvent models to the next level.

Α-mangostin, a polyphenolic xanthone derivative from mangosteen, attenuates β-amyloid oligomers-induced neurotoxicity by inhibiting amyloid aggregation

Alzheimer's disease (AD) is a progressive neurodegenerative disease characterized by the accumulation of β-sheet-rich amyloid oligomers or fibrils which are associated with cellular toxicity in the brain. Inhibition of Aβ aggregation could be a viable therapeutic strategy for slowing and/or preventing the progress of AD. Here we reported that α-mangostin (α-M), a polyphenolic xanthone derivative from mangosteen, concentration-dependently attenuated the neurotoxicity induced by Aβ-(1-40) or Aβ-(1-42) oligomers (EC(50) = 3.89 nM, 4.14 nM respectively) as observed by decreased cell viability and impaired neurite outgrowth in primary rat cerebral cortical neurons. Molecular docking and dynamics simulations demonstrated that α-M could potentially bind to Aβ and stabilize α-helical conformation. α-M was found to directly dissociate Aβ-(1-40) and Aβ-(1-42) oligomers by blotting with oligomer-specific antibodies. ThioflavinT fluorescence assay and electron microscopy imaging further demonstrated that α-M blocked the fibril formation as well as disturbed the pre-formed fibrils. Taken together, our results indicate that α-M is capable to inhibit and dissociate the Aβ aggregation, which could contribute to its effect of attenuating Aβ oligomers-induced neurotoxicity. Thus, α-M could be a great potential candidate for AD treatment. This article is part of a Special Issue entitled 'Post-Traumatic Stress Disorder'.

A strategy for reducing gross errors in the generalized Born models of implicit solvation

The "canonical" generalized Born (GB) formula [C. Still, A. Tempczyk, R. C. Hawley, and T. Hendrickson, J. Am. Chem. Soc. 112, 6127 (1990)] is known to provide accurate estimates for total electrostatic solvation energies ΔG(el) of biomolecules if the corresponding effective Born radii are accurate. Here we show that even if the effective Born radii are perfectly accurate, the canonical formula still exhibits significant number of gross errors (errors larger than 2k(B)T relative to numerical Poisson equation reference) in pairwise interactions between individual atomic charges. Analysis of exact analytical solutions of the Poisson equation (PE) for several idealized nonspherical geometries reveals two distinct spatial modes of the PE solution; these modes are also found in realistic biomolecular shapes. The canonical GB Green function misses one of two modes seen in the exact PE solution, which explains the observed gross errors. To address the problem and reduce gross errors of the GB formalism, we have used exact PE solutions for idealized nonspherical geometries to suggest an alternative analytical Green function to replace the canonical GB formula. The proposed functional form is mathematically nearly as simple as the original, but depends not only on the effective Born radii but also on their gradients, which allows for better representation of details of nonspherical molecular shapes. In particular, the proposed functional form captures both modes of the PE solution seen in nonspherical geometries. Tests on realistic biomolecular structures ranging from small peptides to medium size proteins show that the proposed functional form reduces gross pairwise errors in all cases, with the amount of reduction varying from more than an order of magnitude for small structures to a factor of 2 for the largest ones.

Molecular mechanism of δ-dendrotoxin-potassium channel recognition explored by docking and molecular dynamic simulations

δ-Dendrotoxin, isolated from mamba snake venom, has 57 residues cross-linked by three disulfide bridges. The protein shares a pharmacological activity with other animal toxins, the potent blockade of potassium channels, but is structurally unrelated to toxins of different species. We employed alanine-scanning mutagenesis to explore the molecular mechanism of δ-dendrotoxin binding to potassium channels, using protein-protein docking and molecular dynamic simulations. In our reasonable model of the δ-dendrotoxin-ShaKv1.1 complex, δ-dendrotoxin interacted mainly with the N-terminal region and the turn of two antiparallel β-sheets of the channel. This binding mode could well explain the functional roles of critical residues in δ-dendrotoxin and the ShaKv1.1 channel. Structural analysis indicated that the critical Lys6 residue of δ-dendrotoxin plugged its side chain into a channel selectivity filter. Another two critical δ-dendrotoxin residues, Lys3 and Arg10, were found to contact channel residues through strong polar and nonpolar interactions, especially salt-bridge interactions. As for the ShaKv1.1 channel, the channel turrets were found in the "half-open state," and two of four Glu423 in the turrets of the channel B and D chains could interact, respectively, with Lys3 and Lys26 of δ-dendrotoxin through electrostatic interactions. The essential Asp431 channel residue was found to associate electrostatically with Arg10 of δ-dendrotoxin, and a critical Tyr449 channel residue was just under the channel-interacting surface of δ-dendrotoxin. Together, these novel data may accelerate the structure-function research of toxins in the dendrotoxin family and be of significant value in revealing the diverse interactions between animal toxins and potassium channels.

Heat conductivity of DNA double helix

Thermal conductivity of isolated single molecule DNA fragments is of importance for nanotechnology, but has not yet been measured experimentally. Theoretical estimates based on simplified (1D) models predict anomalously high thermal conductivity. To investigate thermal properties of single molecule DNA we have developed a 3D coarse-grained (CG) model that retains the realism of the full all-atom description, but is significantly more efficient. Within the proposed model each nucleotide is represented by 6 particles or grains; the grains interact via effective potentials inferred from classical molecular dynamics (MD) trajectories based on a well-established all-atom potential function. Comparisons of 10 ns long MD trajectories between the CG and the corresponding all-atom model show similar root-mean-square deviations from the canonical B-form DNA, and similar structural fluctuations. At the same time, the CG model is 10 to 100 times faster depending on the length of the DNA fragment in the simulation. Analysis of dispersion curves derived from the CG model yields longitudinal sound velocity and torsional stiffness in close agreement with existing experiments. The computational efficiency of the CG model makes it possible to calculate thermal conductivity of a single DNA molecule not yet available experimentally. For a uniform (polyG-polyC) DNA, the estimated conductivity coefficient is 0.3 W/mK which is half the value of thermal conductivity for water. This result is in stark contrast with estimates of thermal conductivity for simplified, effectively 1D chains ("beads on a spring") that predict anomalous (infinite) thermal conductivity. Thus, full 3D character of DNA double-helix retained in the proposed model appears to be essential for describing its thermal properties at a single molecule level.

Comparison of virtual high-throughput screening methods for the identification of phosphodiesterase-5 inhibitors

Reliable and effective virtual high-throughput screening (vHTS) methods are desperately needed to minimize the expenses involved in drug discovery projects. Here, we present an improvement to the negative image-based (NIB) screening: the shape, the electrostatics, and the solvation state of the target protein's ligand-binding site are included into the vHTS. Additionally, the initial vHTS results are postprocessed with molecular mechanics/generalized Born surface area (MMGBSA) calculations to estimate the favorability of ligand-protein interactions. The results show that docking produces very good early enrichment for phosphodiesterase-5 (PDE-5); however, in general, the NIB and the ligand-based screening performed better with or without the added electrostatics. Furthermore, the postprocessing of the NIB screening results using MMGBSA calculations improved the early enrichment for the PDE-5 considerably, thus, making hit discovery affordable.

Conformational flexibility and binding interactions of the G protein βγ heterodimer

Previous NMR experiments on unbound G protein βγ heterodimer suggested that particular residues in the binding interface are mobile on the nanosecond timescale. In this work we performed nanosecond-timescale molecular dynamics simulations to investigate conformational changes and dynamics of Gβγ in the presence of several binding partners: a high-affinity peptide (SIGK), phosducin, and the GDP-bound α subunit. In these simulations, the high mobility of GβW99 was reduced by SIGK, and it appeared that a tyrosine might stabilize GβW99 by hydrophobic or aromatic stacking interactions in addition to hydrogen bonds. Simulations of the phosducin-Gβγ complex showed that the mobility of GβW99 was restricted, consistent with inferences from NMR. However, large-scale conformational changes of Gβγ due to binding, which were hypothesized in the NMR study, were not observed in the simulations, most likely due to their short (nanosecond) duration. A pocket consisting of hydrophobic amino acids on Gα appears to restrict GβW99 mobility in the crystal structure of the Gαβγ? heterotrimer. The simulation trajectories are consistent with this idea. However, local conformational changes of residues GβW63, GβW211, GβW297, GβW332, and GβW339 were detected during the MD simulations. As expected, the magnitude of atomic fluctuations observed in simulations was greater for α than for the βγ subunits, suggesting that α has greater flexibility. These observations support the notion that to maintain the high mobility of GβW99 observed by solution NMR requires that the Gβ-α interface must open up on time scale longer than can be observed in nanosecond scale simulations.

Treecode-based generalized Born method

We have developed a treecode-based O(N log N) algorithm for the generalized Born (GB) implicit solvation model. Our treecode-based GB (tGB) is based on the GBr6 [J. Phys. Chem. B 111, 3055 (2007)], an analytical GB method with a pairwise descreening approximation for the R6 volume integral expression. The algorithm is composed of a cutoff scheme for the effective Born radii calculation, and a treecode implementation of the GB charge-charge pair interactions. Test results demonstrate that the tGB algorithm can reproduce the vdW surface based Poisson solvation energy with an average relative error less than 0.6% while providing an almost linear-scaling calculation for a representative set of 25 proteins with different sizes (from 2815 atoms to 65456 atoms). For a typical system of 10k atoms, the tGB calculation is three times faster than the direct summation as implemented in the original GBr6 model. Thus, our tGB method provides an efficient way for performing implicit solvent GB simulations of larger biomolecular systems at longer time scales.

Isolation of thylakoid membrane complexes from rice by a new double-strips BN/SDS-PAGE and bioinformatics prediction of stromal ridge subunits interaction

Thylakoid membrane complexes of rice (Oryza sativa L.) play crucial roles in growth and crop production. Understanding of protein interactions within the complex would provide new insights into photosynthesis. Here, a new "Double-Strips BN/SDS-PAGE" method was employed to separate thylakoid membrane complexes in order to increase the protein abundance on 2D-gels and to facilitate the identification of hydrophobic transmembrane proteins. A total of 58 protein spots could be observed and subunit constitution of these complexes exhibited on 2D-gels. The generality of this new approach was confirmed using thylakoid membrane from spinach (Spinacia oleracea) and pumpkin (Cucurita spp). Furthermore, the proteins separated from rice thylakoid membrane were identified by the mass spectrometry (MS). The stromal ridge proteins PsaD and PsaE were identified both in the holo- and core- PSI complexes of rice. Using molecular dynamics simulation to explore the recognition mechanism of these subunits, we showed that salt bridge interactions between residues R19 of PsaC and E168 of PasD as well as R75 of PsaC and E91 of PsaD played important roles in the stability of the complex. This stromal ridge subunits interaction was also supported by the subsequent analysis of the binding free energy, the intramolecular distances and the intramolecular energy.

Molecular-level examination of Cu2+ binding structure for amyloid fibrils of 40-residue Alzheimer's β by solid-state NMR spectroscopy

Cu(2+) binding to Alzheimer's β (Aβ) peptides in amyloid fibrils has attracted broad attention, as it was shown that Cu ion concentration elevates in Alzheimer's senile plaque and such association of Aβ with Cu(2+) triggers the production of neurotoxic reactive oxygen species (ROS) such as H(2)O(2). However, detailed binding sites and binding structures of Cu(2+) to Aβ are still largely unknown for Aβ fibrils or other aggregates of Aβ. In this work, we examined molecular details of Cu(2+) binding to amyloid fibrils by detecting paramagnetic signal quenching in 1D and 2D high-resolution (13)C solid-state NMR (SSNMR) for full-length 40-residue Aβ(1-40). Selective quenching observed in (13)C SSNMR of Cu(2+)-bound Aβ(1-40) suggested that primary Cu(2+) binding sites in Aβ(1-40) fibrils include N(ε) in His-13 and His-14 and carboxyl groups in Val-40 as well as in Glu sidechains (Glu-3, Glu-11, and/or Glu-22). (13)C chemical shift analysis demonstrated no major structural changes upon Cu(2+) binding in the hydrophobic core regions (residues 18-25 and 30-36). Although the ROS production via oxidization of Met-35 in the presence of Cu(2+) has been long suspected, our SSNMR analysis of (13)C(ε)H(3)-S- in M35 showed little changes after Cu(2+) binding, excluding the possibility of Met-35 oxidization by Cu(2+) alone. Preliminary molecular dynamics (MD) simulations on Cu(2+)-Aβ complex in amyloid fibrils confirmed binding sites suggested by the SSNMR results and the stabilities of such bindings. The MD simulations also indicate the coexistence of a variety of Cu(2+)-binding modes unique in Aβ fibril, which are realized by both intra- and intermolecular contacts and highly concentrated coordination sites due to the in-register parallel β-sheet arrangements.

Molecular dynamics of ribosomal elongation factors G and Tu

Translation on the ribosome is controlled by external factors. During polypeptide lengthening, elongation factors EF-Tu and EF-G consecutively interact with the bacterial ribosome. EF-Tu binds and delivers an aminoacyl-tRNA to the ribosomal A site and EF-G helps translocate the tRNAs between their binding sites after the peptide bond is formed. These processes occur at the expense of GTP. EF-Tu:tRNA and EF-G are of similar shape, share a common binding site, and undergo large conformational changes on interaction with the ribosome. To characterize the internal motion of these two elongation factors, we used 25 ns long all-atom molecular dynamics simulations. We observed enhanced mobility of EF-G domains III, IV, and V and of tRNA in the EF-Tu:tRNA complex. EF-Tu:GDP complex acquired a configuration different from that found in the crystal structure of EF-Tu with a GTP analogue, showing conformational changes in the switch I and II regions. The calculated electrostatic properties of elongation factors showed no global similarity even though matching electrostatic surface patches were found around the domain I that contacts the ribosome, and in the GDP/GTP binding region.

RNA stability under different combinations of amber force fields and solvation models

The proper matching of force field and solvent is critical to obtain correct result in molecular dynamics simulation of bio-molecules. This problem has been intensively investigated for protein but not for RNA yet. In this paper, we use standard molecular dynamics and replica exchange molecular dynamics to take a series of tests on the RNA stability under different combinations of Amber force field parameters (ff98, ff99 and ff99bsc0) and the general Born implicit solvent models (igb1, igb2 and igb5). It is found that only ff98 and ff99bsc0 with igb1 can keep the native conformations of RNA hairpin and duplex. Our results suggest that ff98 plus igb1 may be reasonable choice for molecular dynamics simulation of RNA dynamics.

An n log n Generalized Born Approximation

Molecular dynamics (MD) simulations based on the generalized Born (GB) model of implicit solvation offer a number of important advantages over the traditional explicit solvent based simulations. Yet, in MD simulations, the GB model has not been able to reach its full potential partly due to its computational cost, which scales as ∼n(2), where n is the number of solute atoms. We present here an ∼n log n approximation for the generalized Born (GB) implicit solvent model. The approximation is based on the hierarchical charge partitioning (HCP) method (Anandakrishnan and Onufriev J. Comput. Chem. 2010 , 31 , 691 - 706 ) previously developed and tested for electrostatic computations in gas-phase and distant dependent dielectric models. The HCP uses the natural organization of biomolecular structures to partition the structures into multiple hierarchical levels of components. The charge distribution for each of these components is approximated by a much smaller number of charges. The approximate charges are then used for computing electrostatic interactions with distant components, while the full set of atomic charges are used for nearby components. To apply the HCP concept to the GB model, we define the equivalent of the effective Born radius for components. The component effective Born radius is then used in GB computations for points that are distant from the component. This HCP approximation for GB (HCP-GB) is implemented in the open source MD software, NAB in AmberTools, and tested on a set of representative biomolecular structures ranging in size from 632 atoms to ∼3 million atoms. For this set of test structures, the HCP-GB method is 1.1-390 times faster than the GB computation without additional approximations (the reference GB computation), depending on the size of the structure. Similar to the spherical cutoff method with GB (cutoff-GB), which also scales as ∼n log n, the HCP-GB is relatively simple. However, for the structures considered here, we show that the HCP-GB method is more accurate than the cutoff-GB method as measured by relative RMS error in electrostatic force compared to the reference (no cutoff) GB computation. MD simulations of four biomolecular structures on 50 ns time scales show that the backbone RMS deviation for the HCP-GB method is in reasonable agreement with the reference GB simulation. A critical difference between the cutoff-GB and HCP-GB methods is that the cutoff-GB method completely ignores interactions due to atoms beyond the cutoff distance, whereas the HCP-GB method uses an approximation for interactions due to distant atoms. Our testing suggests that completely ignoring distant interactions, as the cutoff-GB does, can lead to qualitatively incorrect results. In general, we found that the HCP-GB method reproduces key characteristics of dynamics, such as residue fluctuation, χ1/χ2 flips, and DNA flexibility, more accurately than the cutoff-GB method. As a practical demonstration, the HCP-GB simulation of a 348 000 atom chromatin fiber was used to refine the starting structure. Our findings suggest that the HCP-GB method is preferable to the cutoff-GB method for molecular dynamics based on pairwise implicit solvent GB models.

Differential molecular information of maurotoxin peptide recognizing IK(Ca) and Kv1.2 channels explored by computational simulation

BACKGROUND

Scorpion toxins are invaluable tools for ion channel research and are potential drugs for human channelopathies. However, it is still an open task to determine the molecular basis underlying the diverse interactions between toxin peptides and ion channels. The inhibitory peptide Maurotoxin (MTX) recognized the distantly related IK(Ca) and Kv1.2 channel with approximately the same potency and using the same functional residues, their differential binding mechanism remain elusive. In this study, we applied computational methods to explore the differential binding modes of MTX to Kv1.2 and IK(Ca) channels, which would help to understand the diversity of channel-toxin interactions and accelerate the toxin-based drug design.

RESULTS

A reasonably stable MTX-IK(Ca) complex was obtained by combining various computational methods and by in-depth comparison with the previous model of the MTX-Kv1.2 complex. Similarly, MTX adopted the β-sheet structure as the interacting surface for binding both channels, with Lys23 occluding the pore. In contrast, the other critical residues Lys27, Lys30, and Tyr32 of MTX adopted distinct interactions when associating with the IK(Ca) channel. In addition, the residues Gln229, Ala230, Ala233, and Thr234 on the IK(Ca) channel turret formed polar and non-polar interactions with MTX, whereas the turret of Kv1.2 was almost not involved in recognizing MTX. In all, the pairs of interacting residues on MTX and the IK(Ca) channel of the bound complex indicated that electrostatic and Van der Waal interactions contributed equally to the formation of a stable MTX-IK(Ca) complex, in contrast to the MTX-Kv1.2 binding that is dominantly mediated by electrostatic forces.

CONCLUSIONS

Despite sharing similar pharmacological profiles toward both IK(Ca) and Kv1.2 channels, MTX adopted totally diverging modes in the two association processes. All the molecular information unveiled here could not only offer a better understanding about the structural differences between the IK(Ca) and Kv1.2 channels, but also provide novel structural clues that will help in the designing of more selective molecular probes to discriminate between these two channels.

Mechanistic insight into human ether-à-go-go-related gene (hERG) K+ channel deactivation gating from the solution structure of the EAG domain

Human ether-à-go-go-related gene (hERG) K⁺ channels have a critical role in cardiac repolarization. hERG channels close (deactivate) very slowly, and this is vital for regulating the time course and amplitude of repolarizing current during the cardiac action potential. Accelerated deactivation is one mechanism by which inherited mutations cause long QT syndrome and potentially lethal arrhythmias. hERG deactivation is highly dependent upon an intact EAG domain (the first 135 amino acids of the N terminus). Importantly, deletion of residues 2–26 accelerates deactivation to a similar extent as removing the entire EAG domain. These and other experiments suggest the first 26 residues (NT1–26) contain structural elements required to slow deactivation by stabilizing the open conformation of the pore. Residues 26–135 form a Per-Arnt-Sim domain, but a structure for NT1–26 has not been forthcoming, and little is known about its site of interaction on the channel. In this study, we present an NMR structure for the entire EAG domain, which reveals that NT1–26 is structurally independent from the Per-Arnt-Sim domain and contains a stable amphipathic helix with one face being positively charged. Mutagenesis and electrophysiological studies indicate that neutralizing basic residues and breaking the amphipathic helix dramatically accelerate deactivation. Furthermore, scanning mutagenesis and molecular modeling studies of the cyclic nucleotide binding domain suggest that negatively charged patches on its cytoplasmic surface form an interface with the NT1–26 domain. We propose a model in which NT1–26 obstructs gating motions of the cyclic nucleotide binding domain to allosterically stabilize the open conformation of the pore.

Improving the description of salt bridge strength and geometry in a Generalized Born model

The Generalized Born (GB) solvent model is widely used in molecular dynamics simulations because it can be less computationally expensive and it samples conformational changes more efficiently than explicit solvent simulations. Meanwhile, great efforts have been made in the past to improve its precision and accuracy. Previous studies have shown that reducing intrinsic GB radii of some hydrogen atoms would improve AMBER GB-HCT solvent model's accuracy on salt bridges. Here we present our finding that similar correction also shows dramatic improvement for the AMBER GB-OBC solvent model. Potential of mean force and cluster analysis for small peptide replica exchange molecular dynamics simulations suggested that new radii GB simulation with ff99SB/GB-OBC corrected salt bridge strength and achieved significantly higher geometry similarity with TIP3P simulation. Improved performance in 60 ns HIV-1 protease GB simulation further validated this approach for large systems.