Importance of the peptide backbone description in modeling the selectivity filter in potassium channels

A Molecular Lid Mechanism of K+ Channel Blocker Action Revealed by a Cone Peptide

Many venomous organisms carry in their arsenal short polypeptides that block K⁺ channels in a highly selective manner. These toxins may compete with the permeating ions directly via a "plug" mechanism or indirectly via a "pore-collapse" mechanism. An alternative "lid" mechanism was proposed but remained poorly defined. Here we study the Drosophila Shaker channel block by Conkunitzin-S1 and Conkunitzin-C3, two highly similar toxins derived from cone venom. Despite their similarity, the two peptides exhibited differences in their binding poses and biophysical assays, implying discrete action modes. We show that while Conkunitzin-S1 binds tightly to the channel turret and acts via a "pore-collapse" mechanism, Conkunitzin-C3 does not contact this region. Instead, Conk-C3 uses a non-conserved Arg to divert the permeant ions and trap them in off-axis cryptic sites above the SF, a mechanism we term a "molecular-lid". Our study provides an atomic description of the "lid" K⁺ blocking mode and offers valuable insights for the design of therapeutics based on venom peptides.

Biobetters From an Integrated Computational/Experimental Approach

Biobetters are new drugs designed from existing peptide or protein-based therapeutics by improving their properties such as affinity and selectivity for the target epitope, and stability against degradation. Computational methods can play a key role in such design problems—by predicting the changes that are most likely to succeed, they can drastically reduce the number of experiments to be performed. Here we discuss the computational and experimental methods commonly used in drug design problems, focusing on the inverse relationship between the two, namely, the more accurate the computational predictions means the less experimental effort is needed for testing. Examples discussed include efforts to design selective analogs from toxin peptides targeting ion channels for treatment of autoimmune diseases and monoclonal antibodies which are the fastest growing class of therapeutic agents particularly for cancers and autoimmune diseases.

Computational Study of Binding of μ-Conotoxin GIIIA to Bacterial Sodium Channels NaVAb and NaVRh

Structures of several voltage-gated sodium (NaV) channels from bacteria have been determined recently, but the same feat might not be achieved for the mammalian counterparts in the near future. Thus, at present, computational studies of the mammalian NaV channels have to be performed using homology models based on the bacterial crystal structures. A successful homology model for the mammalian NaV1.4 channel was recently constructed using the extensive mutation data for binding of μ-conotoxin GIIIA to NaV1.4, which was further validated through studies of binding of other μ-conotoxins and ion permeation. Understanding the similarities and differences between the bacterial and mammalian NaV channels is an important issue, and the NaV1.4-GIIIA system provides a good opportunity for such a comparison. To this end, we study the binding of GIIIA to the bacterial channels NaVAb and NaVRh. The complex structures are obtained using docking and molecular dynamics simulations, and the dissociation of GIIIA is studied through umbrella sampling simulations. The results are compared to those obtained from the NaV1.4-GIIIA system, and the differences in the binding modes arising from the changes in the selectivity filters are highlighted.

Explicit Hydrogen-Bond Potentials and Their Application to NMR Scalar Couplings in Proteins

Hydrogen bonds (H bonds) are fundamental for the stability, structure, and dynamics of chemically and biologically relevant systems. One of the direct means to detect H bonds in proteins is NMR spectroscopy. As H bonds are dynamic in nature, atomistic simulations offer a meaningful way to characterize and analyze properties of hydrogen bonds, provided a sufficiently accurate interaction potential is available. Here, we use explicit H-bond potentials to investigate scalar coupling constants (h3)JNC' and characterize the conformational ensemble for increasingly accurate intermolecular potentials. By considering a range of proteins with different overall topology a general procedure to improve the hydrogen-bonding potential ("morphing potentials") based on experimental information is derived. The robustness of this approach is established through explicit simulations in full solvation and comparison with experimental results. The H-bond potentials used here lead to more directional H bonds than conventional electrostatic representations employed in molecular mechanics potentials. It is found that the optimized potentials lead to H-bond geometries in remarkable agreement with previous ab initio and knowledge-based approaches to H bonds in model systems and in proteins. This suggests that, by combining theory, computation, and experimental data, H-bonding potentials can be improved and are potentially useful to better study coupling, energy transfer, and allosteric communication in proteins.

Mechanism of Ion Permeation in Mammalian Voltage-Gated Sodium Channels

Recent determination of the crystal structures of bacterial voltage-gated sodium (NaV) channels have raised hopes that modeling of the mammalian counterparts could soon be achieved. However, there are substantial differences between the pore domains of the bacterial and mammalian NaV channels, which necessitates careful validation of mammalian homology models constructed from the bacterial NaV structures. Such a validated homology model for the NaV1.4 channel was constructed recently using the extensive mutagenesis data available for binding of μ-conotoxins. Here we use this NaV1.4 model to study the ion permeation mechanism in mammalian NaV channels. Linking of the DEKA residues in the selectivity filter with residues in the neighboring domains is found to be important for keeping the permeation pathway open. Molecular dynamics simulations and potential of mean force calculations reveal that there is a binding site for a Na+ ion just inside the DEKA locus, and 1-2 Na+ ions can occupy the vestibule near the EEDD ring. These sites are separated by a low free energy barrier, suggesting that inward conduction occurs when a Na+ ion in the vestibule goes over the free energy barrier and pushes the Na+ ion in the filter to the intracellular cavity, consistent with the classical knock-on mechanism. The NaV1.4 model also provides a good description of the observed Na+/K+ selectivity.

Computational approaches for designing potent and selective analogs of peptide toxins as novel therapeutics

Peptide toxins provide valuable therapeutic leads for many diseases. As they bind to their targets with high affinity, potency is usually ensured. However, toxins also bind to off-target receptors, causing potential side effects. Thus, a major challenge in generating drugs from peptide toxins is ensuring their specificity for their intended targets. Computational methods can play an important role in solving such design problems through construction of accurate models of receptor-toxin complexes and calculation of binding free energies. Here we review the computational methods used for this purpose and their application to toxins targeting ion channels. We describe ShK and HsTX1 toxins, high-affinity blockers of the voltage-gated potassium channel Kv1.3, which could be developed as therapeutic agents for autoimmune diseases.

Calculation of free energy changes due to mutations from alchemical free energy simulations

How a mutation affects the binding free energy of a ligand is a fundamental problem in molecular biology/biochemistry with many applications in pharmacology and biotechnology, e.g. design of drugs and enzymes. Free energy change due to a mutation can be determined most accurately by performing alchemical free energy calculations in molecular dynamics (MD) simulations. Here we discuss the necessary conditions for success of free energy calculations using toxin peptides that bind to ion channels as examples. We show that preservation of the binding mode is an essential requirement but this condition is not always satisfied, especially when the mutation involves a charged residue. Otherwise problems with accuracy of results encountered in mutation of charged residues can be overcome by performing the mutation on the ligand in the binding site and bulk simultaneously and in the same system. The proposed method will be useful in improving the affinity and selectivity profiles of drug leads and enzymes via computational design and protein engineering.

Systematic study of binding of μ-conotoxins to the sodium channel NaV1.4

Voltage-gated sodium channels (NaV) are fundamental components of the nervous system. Their dysfunction is implicated in a number of neurological disorders, such as chronic pain, making them potential targets for the treatment of such disorders. The prominence of the NaV channels in the nervous system has been exploited by venomous animals for preying purposes, which have developed toxins that can block the NaV channels, thereby disabling their function. Because of their potency, such toxins could provide drug leads for the treatment of neurological disorders associated with NaV channels. However, most toxins lack selectivity for a given target NaV channel, and improving their selectivity profile among the NaV1 isoforms is essential for their development as drug leads. Computational methods will be very useful in the solution of such design problems, provided accurate models of the protein-ligand complex can be constructed. Using docking and molecular dynamics simulations, we have recently constructed a model for the NaV1.4-μ-conotoxin-GIIIA complex and validated it with the ample mutational data available for this complex. Here, we use the validated NaV1.4 model in a systematic study of binding other μ-conotoxins (PIIIA, KIIIA and BuIIIB) to NaV1.4. The binding mode obtained for each complex is shown to be consistent with the available mutation data and binding constants. We compare the binding modes of PIIIA, KIIIA and BuIIIB to that of GIIIA and point out the similarities and differences among them. The detailed information about NaV1.4-μ-conotoxin interactions provided here will be useful in the design of new NaV channel blocking peptides.

Molecular dynamics study of binding of µ-conotoxin GIIIA to the voltage-gated sodium channel Na(v)1.4

Homology models of mammalian voltage-gated sodium (NaV) channels based on the crystal structures of the bacterial counterparts are needed to interpret the functional data on sodium channels and understand how they operate. Such models would also be invaluable in structure-based design of therapeutics for diseases involving sodium channels such as chronic pain and heart diseases. Here we construct a homology model for the pore domain of the NaV1.4 channel and use the functional data for the binding of µ-conotoxin GIIIA to NaV1.4 to validate the model. The initial poses for the NaV1.4-GIIIA complex are obtained using the HADDOCK protein docking program, which are then refined in molecular dynamics simulations. The binding mode for the final complex is shown to be in broad agreement with the available mutagenesis data. The standard binding free energy, determined from the potential of mean force calculations, is also in good agreement with the experimental value. Because the pore domains of NaV1 channels are highly homologous, the model constructed for NaV1.4 will provide an excellent template for other NaV1 channels.

Energetics of Multi-Ion Conduction Pathways in Potassium Ion Channels

Potassium ion channels form pores in cell membranes, allowing potassium ions through while preventing the passage of sodium ions. Despite numerous high-resolution structures, it is not yet possible to relate their structure to their single molecule function other than at a qualitative level. Over the past decade, there has been a concerted effort using molecular dynamics to capture the thermodynamics and kinetics of conduction by calculating potentials of mean force (PMF). These can be used, in conjunction with the electro-diffusion theory, to predict the conductance of a specific ion channel. Here, we calculate seven independent PMFs, thereby studying the differences between two potassium ion channels, the effect of the CHARMM CMAP forcefield correction, and the sensitivity and reproducibility of the method. Thermodynamically stable ion-water configurations of the selectivity filter can be identified from all the free energy landscapes, but the heights of the kinetic barriers for potassium ions to move through the selectivity filter are, in nearly all cases, too high to predict conductances in line with experiment. This implies it is not currently feasible to predict the conductance of potassium ion channels, but other simpler channels may be more tractable.

The mechanism of Na⁺/K⁺ selectivity in mammalian voltage-gated sodium channels based on molecular dynamics simulation

Voltage-gated sodium (Nav) channels and their Na⁺/K⁺ selectivity are of great importance in the mammalian neuronal signaling. According to mutational analysis, the Na⁺/K⁺ selectivity in mammalian Nav channels is mainly determined by the Lys and Asp/Glu residues located at the constriction site within the selectivity filter. Despite successful molecular dynamics simulations conducted on the prokaryotic Nav channels, the lack of Lys at the constriction site of prokaryotic Nav channels limits how much can be learned about the Na⁺/K⁺ selectivity in mammalian Nav channels. In this work, we modeled the mammalian Nav channel by mutating the key residues at the constriction site in a prokaryotic Nav channel (NavRh) to its mammalian counterpart. By simulating the mutant structure, we found that the Na⁺ preference in mammalian Nav channels is collaboratively achieved by the deselection from Lys and the selection from Asp/Glu within the constriction site.

Computational studies of marine toxins targeting ion channels

Toxins from marine animals offer novel drug leads for treatment of diseases involving ion channels. Computational methods could be very helpful in this endeavour in several ways, e.g., (i) constructing accurate models of the channel-toxin complexes using docking and molecular dynamics (MD) simulations; (ii) determining the binding free energies of toxins from umbrella sampling MD simulations; (iii) predicting the effect of mutations from free energy MD simulations. Using these methods, one can design new analogs of toxins with improved affinity and selectivity properties. Here we present a review of the computational methods and discuss their applications to marine toxins targeting potassium and sodium channels. Detailed examples from the potassium channel toxins—ShK from sea anemone and κ-conotoxin PVIIA—are provided to demonstrate capabilities of the computational methods to give accurate descriptions of the channel-toxin complexes and the energetics of their binding. An example is also given from sodium channel toxins (µ-conotoxin GIIIA) to illustrate the differences between the toxin binding modes in potassium and sodium channels.

Why the Drosophila Shaker K+ channel is not a good model for ligand binding to voltage-gated Kv1 channels

The Drosophila Shaker K(+) channel is the first cloned voltage-gated potassium channel and has, therefore, played an important role in structural and functional studies of those channels. While such a role is well justified for ion permeation, it is not clear whether this also extends to ligand binding. Despite the high degree of homology among Shaker and Kv1 channels, κ-conotoxin PVIIA (κ-PVIIA) binds to Shaker with high affinity but not to Kv1 channels. Here we address this issue by studying binding of κ-PVIIA to Shaker and Kv1 channels using molecular dynamics (MD) simulations. The structures of the channel-toxin complexes are constructed via docking and refinement with MD. The binding mode of each complex is characterized and compared to available mutagenesis data to validate the complex models. The potential of mean force for dissociation of the Shaker-κ-PVIIA complex is calculated from umbrella sampling MD simulations, and the corresponding binding free energy is determined, which provides further validation of the complex structure. Comparison of the Shaker and Kv1 complex models shows that a few mutations in the turret and extended regions are sufficient to abolish the observed sensitivity of Shaker to κ-PVIIA. This study demonstrates that Shaker is not always a good model for Kv1 channels for ligand binding. It also provides insights into the binding of the toxin to potassium channels that will be useful for improving affinity and selectivity properties of Kv1 channels.

High-risk long QT syndrome mutations in the Kv7.1 (KCNQ1) pore disrupt the molecular basis for rapid K(+) permeation

Type 1 long QT syndrome (LQT1) is caused by loss-of-function mutations in the KCNQ1 gene, which encodes the K(+) channel (Kv7.1) that underlies the slowly activating delayed rectifier K(+) current in the heart. Intragenic risk stratification suggests LQT1 mutations that disrupt conserved amino acid residues in the pore are an independent risk factor for LQT1-related cardiac events. The purpose of this study is to determine possible molecular mechanisms that underlie the loss of function for these high-risk mutations. Extensive genotype-phenotype analyses of LQT1 patients showed that T322M-, T322A-, or G325R-Kv7.1 confers a high risk for LQT1-related cardiac events. Heterologous expression of these mutations with KCNE1 revealed they generated nonfunctional channels and caused dominant negative suppression of WT-Kv7.1 current. Molecular dynamics simulations of analogous mutations in KcsA (T85M-, T85A-, and G88R-KcsA) demonstrated that they disrupted the symmetrical distribution of the carbonyl oxygen atoms in the selectivity filter, which upset the balance between the strong attractive and K(+)-K(+) repulsive forces required for rapid K(+) permeation. We conclude high-risk LQT1 mutations in the pore likely disrupt the architectural and physical properties of the K(+) channel selectivity filter.

Molecular dynamics simulations of membrane proteins

Membrane proteins control the traffic across cell membranes and thereby play an essential role in cell function from transport of various solutes to immune response via molecular recognition. Because it is very difficult to determine the structures of membrane proteins experimentally, computational methods have been increasingly used to study their structure and function. Here we focus on two classes of membrane proteins-ion channels and transporters-which are responsible for the generation of action potentials in nerves, muscles, and other excitable cells. We describe how computational methods have been used to construct models for these proteins and to study the transport mechanism. The main computational tool is the molecular dynamics (MD) simulation, which can be used for everything from refinement of protein structures to free energy calculations of transport processes. We illustrate with specific examples from gramicidin and potassium channels and aspartate transporters how the function of these membrane proteins can be investigated using MD simulations.

Affinity and selectivity of ShK toxin for the Kv1 potassium channels from free energy simulations

The voltage-gated potassium channel Kv1.3 is an attractive target for treatment of autoimmune diseases. ShK toxin from sea anemone is one of the most potent blockers of Kv1.3, and therefore ShK and its analogues have been proposed as therapeutic leads for such diseases. Increasing the selectivity of the proposed leads for Kv1.3 over other Kv1 channels is a major issue in this endeavor. Here we study binding of ShK toxin to Kv1 channels using free energy simulation methods. Homology models for Kv1.1 and Kv1.3 channels are constructed using the crystal structure of Kv1.2. The initial poses for the Kv1.x-ShK complexes are obtained using HADDOCK, which are then refined via molecular dynamics simulations. The binding mode in each complex is characterized by identifying the strongly interacting residues, which compare well with available mutagenesis studies. For each complex, the potential of mean force is calculated from umbrella sampling simulations, and the corresponding absolute binding free energy is determined. The computed binding free energies are in good agreement with the experimental data, which increases the confidence on the model complexes. The insights gained on Kv1.x-ShK binding modes will be valuable in the development of new ShK analogues with better selectivity properties.

Comparative study of the energetics of ion permeation in Kv1.2 and KcsA potassium channels

Biological ion channels rely on a multi-ion transport mechanism for fast yet selective permeation of ions. The crystal structure of the KcsA potassium channel provided the first microscopic picture of this process. A similar mechanism is assumed to operate in all potassium channels, but the validity of this assumption has not been well investigated. Here, we examine the energetics of ion permeation in Shaker Kv1.2 and KcsA channels, which exemplify the six-transmembrane voltage-gated and two-transmembrane inward-rectifier channels. We study the feasibility of binding a third ion to the filter and the concerted motion of ions in the channel by constructing the potential of mean force for K(+) ions in various configurations. For both channels, we find that a pair of K(+) ions can move almost freely within the filter, but a relatively large free-energy barrier hinders the K(+) ion from stepping outside the filter. We discuss the effect of the CMAP dihedral energy correction that was recently incorporated into the CHARMM force field on ion permeation dynamics.

Modeling GPCR active state conformations: the β(2)-adrenergic receptor

The recent publication of several G protein-coupled receptor (GPCR) structures has increased the information available for homology modeling inactive class A GPCRs. Moreover, the opsin crystal structure shows some active features. We have therefore combined information from these two sources to generate an extensively validated model of the active conformation of the β(2)-adrenergic receptor. Experimental information on fully active GPCRs from zinc binding studies, site-directed spin labeling, and other spectroscopic techniques has been used in molecular dynamics simulations. The observed conformational changes reside mainly in transmembrane helix 6 (TM6), with additional small but significant changes in TM5 and TM7. The active model has been validated by manual docking and is in agreement with a large amount of experimental work, including site-directed mutagenesis information. Virtual screening experiments show that the models are selective for β-adrenergic agonists over other GPCR ligands, for (R)- over (S)-β-hydroxy agonists and for β(2)-selective agonists over β(1)-selective agonists. The virtual screens reproduce interactions similar to those generated by manual docking. The C-terminal peptide from a model of the stimulatory G protein, readily docks into the active model in a similar manner to which the C-terminal peptide from transducin, docks into opsin, as shown in a recent opsin crystal structure. This GPCR-G protein model has been used to explain site-directed mutagenesis data on activation. The agreement with experiment suggests a robust model of an active state of the β(2)-adrenergic receptor has been produced. The methodology used here should be transferable to modeling the active state of other GPCRs.

Selective complexation of K+ and Na+ in simple polarizable ion-ligating systems

An influx of experimental and theoretical studies of ion transport protein structure has inspired efforts to understand underlying determinants of ionic selectivity. Design principles for selective ion binding can be effectively isolated and interrogated using simplified models composed of a single ion surrounded by a set of ion-ligating molecular species. While quantum mechanical treatments of such systems naturally incorporate electronic degrees of freedom, their computational overhead typically prohibits thorough dynamic sampling of configurational space and, thus, requires approximations when determining ion-selective free energy. As an alternative, we employ dynamical simulations with a polarizable force field to probe the structure and K(+)/Na(+) selectivity in simple models composed of one central K(+)/Na(+) ion surrounded by 0-8 identical model compounds: N-methylacetamide, formamide, or water. In the absence of external restraints, these models represent gas-phase clusters displaying relaxed coordination structures with low coordination number. Such systems display Na(+) selectivity when composed of more than ∼3 organic carbonyl-containing compounds and always display K(+) selectivity when composed of water molecules. Upon imposing restraints that solely enforce specific coordination numbers, we find all models are K(+)-selective when ∼7-8-fold ion coordination is achieved. However, when models composed of the organic compounds provide ∼4-6-fold coordination, they retain their Na(+) selectivity. From these trends, design principles emerge that are of basic importance in the behavior of K(+) channel selectivity filters and suggest a basis not only for K(+) selectivity but also for modulation of block and closure by smaller ions.