Skip to Main Content
From Computational Biophysics to Systems Biology (CBSB2015)
May 17-19, 2015  Oklahoma City, Oklahoma
OUHSC link

OU link

OU link

UCO link

OMRF link

  • Andrew Adams (Oklahoma State University)
  • Lucas Adams(Oklahoma Babtist University), Poster
  • Darin Akins (University of Oklahoma Health Sciences Research Center
  • Erik Alred(University of Oklahoma), Poster
  • Pradipta Bandyopadhyay (Jawaharlal Nehru University), Invited Talk
  • Anirban Banerji(The Research Institute at Nationwide Children Hospital), Contributed Talk
  • Workalemahu Berhanu(University of Oklahoma), Poster
  • Nathan Bernhardt(University of Oklahoma), Poster
  • Robert B. Best (National Institutes of Health), Invited Talk
  • Manikanthan Bhavaraju(University of Oklahoma), Poster
  • Rebba Boswell-Casteel (University of Oklahoma Health Science Center), Poster
  • Christina Bourne (University of Oklahoma)
  • Christina Bruxvoor(University of Oklahoma Health Sciences Center), Poster
  • Carlos J. Camacho (University of Pittsburgh), Invited Talk
  • Ludmila Chandler (Southeastern Oklahoma State University)
  • Yi Cao (Nanjing University)
  • Gabriel Cook (Oklahoma State University), Poster
  • Lilian Chooback(University of Central Oklahoma), Poster
  • April Clevenger (University of Oklahoma), Poster
  • Kadir Demir(Bulent Ecevit University), Poster
  • Ken A Dill (Stony Brook University) Keynote
  • Mai Do (University of Central Oklahoma)
  • Kelvin Droegemeier (University of Oklahoma)
  • Ron Elber (University of Texas, Austin), Keynote
  • Christopher Fennel Oklahoma State University, Invited Talk , Poster
  • Dane Fleming (University of Central Oklahoma)
  • Clay Foster (University of Oklahoma)
  • Alamayehu Gorfe (University of Texas Health Sciences Center at Houston), Invited Talk
  • Roxanna Grove (University of Central Oklahoma)
  • Kapli Gumte (Oklahoma State University)
  • Ron Halterman (University of Oklahoma)
  • Ulrich H.E. Hansmann (University of Oklahoma)
  • Tingting Hao(Clemson University), Contributed Talk
  • Franklin A. Hayes (University of Oklahomai Helath Sciences Center)
  • Grace Harrel (Oklahoma State University)
  • Viridiana Herrera (University of Oklahoma), Poster
  • Abu Gafar Hossion(University of Kansas), Contributed Talk
  • Zhe Jia(Clemson University), Poster
  • Steven Karpowicz (University of Central Oklahoma
  • Daisuke Kihara (Purdue University), Invited Talk
  • Sang Beom Kim (Princeton University), Outstanding Young Researcher Award Talk
  • Maksim Kouza(University of Warsaw), Contributed Talk
  • Maria Kurnikova (Carnegie Mellon University), Invited Talk
  • Emily Kurdzo (University of Oklahoma Health Sciences Center)
  • Sukyoung Kwak (University of Central Oklahoma(
  • S. Lakshmivarahan (University of Oklahoma)
  • Suzanne Lapolla (University of Oklahoma Health Science Center)
  • Kai Leong(University of Arkansas, Fayetteville), Contributed Talk
  • Wenfei Li(Nanjing University), Poster
  • Ruibin Liang (University of Chicago), Outstanding Young Researcher Award Talk
  • Chengyi Liao (University of Vermont), Outstanding Young Researcher Award Talk
  • Jialing Lin (University of Central Oklahoma
  • Xubo Lin(The University of Texas Health Science Center at Houston), Contributed Talk
  • Yinling Liu(Clemson University), Contributed Talk
  • Maggie Martin( University of Oklahoma), Poster
  • Dale Merz (University of Cincinnati), Outstanding Young Researcher Award Talk
  • Fares Najar (University of Oklahoma)
  • Michael Nguyen (University of Oklahoma
  • Ruth Nussinov (National Institutes of Health) Keynote
  • Sherin Paranahewage(Oklahoma State University), Contributed Talk
  • Nancy Paiva (Southeastern Oklahoma State University)
  • Zoya Petrushenko (University of Oklahoma), Poster
  • Malachi Phillips(University of Oklahoma), Poster
  • Minu Pilvankar (Oklahoma State University)
  • Augie Pioszak (University of Oklahoma Health Sciences Center)
  • Meng Qin (Nanjing University)
  • Pallavi Rajaputra (University of Oklahoma Health Science)
  • Xiavan Roopnarinesingh (Oklahoma Medical Research Facility)
  • Rupa Sarkar (University of Oklahoma)
  • Serzhan Sakipov(Carnegie Mellon University), Poster
  • Susan Schroeder (University of Oklahoma)
  • Yu-Hsuan Shih(University of Cincinnati), Contributed Talk
  • Yuko Tsutsui (University of Oklahomai Helath Sciences Center)
  • Sahin Uyaver(Istanbul Ticaret University), Poster
  • Melville Vaughan(University of Central Oklahoma), Poster
  • Ashlee Ford Versypt(Oklahoma State University), Poster
  • Thomas Vogel (Los Alamos National Laboratory), Outstanding Young Researcher Award Talk
  • Feng Wang (University of Arkansas), Invited Talk
  • Wei Wang (Nanjing University), Invited Talk
  • Jessica Webb (University of Central Oklahoma)
  • John White (University of Oklahoma)
  • Jonathan Wren(Oklahoma Medical Research Foundation), Poster
  • Hao Wu (Oklahoma Medical Research Foundation)
  • Gang Xu(University of Central Oklahoma), Poster
  • Sichun Yang (Case Western University), Invited Talk
  • Fatih Yasar(Hacettepe University), Poster
  • Hang Zhao (University of Oklahoma), Poster
  • Haiqing Zhao (University of Maryland), Contributed Talk
  • Yu Zhao (University of Oklahoma)



  • Harnessing confusing combinatorial information within MD simulations for modeling proteins.
    Ken A. Dill
    Stony Brook University

    A standard approach in integrative structural biology is to combine atomistic MD simulations, which accounts for energetics, with external experimental or restraint information, which gives guidance from structural information. However, this approach doesn't work if the information is not correct, accurate and specific. We are developing a statmech approach, called MELD, for harnessing troublesome information within MD simulations. For example, by telling an MD simulation to find structures that have both low free energies while also having good hydrophobic cores and good secondary structures, we can fold small proteins close to their native structures and aid in structure refinement.
    Back to the top of the page

  • Computer simulations of protein function: Challenges and solutions`
    Ron Elber
    Department of Chemistry and Biochemistry, University of Texas, Austin

    Biological processes are controlled by kinetics. The next biochemical reaction that will happen in a living system is determined by the speed of the reaction and not by the relative stability of the products. It is therefore crucial to understand kinetics and time scales of biological processes in order to model living systems. In my talk I will describe a simulation technology, Milestoning, developed in my group that makes it possible to investigate biochemical processes at the atomic level and extended temporal scales. As examples, I will discuss functional conformational transitions in the proteins myosin and HIV reverse transcriptase. I will also examine mechanical stability of helices and coiled coil constructs.
    Back to the top of the page

  • K-Ras Ensembles, Biology and Signaling: Questions and Observations
    Ruth Nussinov
    Cancer and Inflammation Program National Cancer Institute and Medical School, Tel Aviv University

    Ras proteins are small GTPases that act as signal transducers between cell surface receptors and several intracellular signaling cascades. KRas4B is among the frequently mutated oncogenes in human tumors. Ras proteins consist of highly homologous catalytic domains, and flexible C-terminal hypervariable regions (HVRs) that differ significantly across Ras isoforms. We have been focusing on key mechanistic questions in Ras biology from the structural standpoint. These include whether Ras forms dimers, and if so what is their structural landscape; how do Ras dimers activate Raf, a key Ras effector in a major signaling pathway; how calmodulin temporally inhibit Raf signaling, and the potential role of the hypervariable region and its membrane anchoring regulation. We believe that structural biology, computations and experiment, are uniquely able to tackle these fascinating questions
    Back to the top of the page

Invited Talks

  • Computational modeling of cytoplasm of a bacterial cell: diffusion and hydrodynamics
    Pradipta Bandyopadhyay
    School of Computational and Integrative Sciences, Jawaharlal Nehru University

    It is known that the kinetics and thermodynamics of processes occurring in living systems can be significantly affected by the crowded nature of the biological cell. Indeed several in-vivo experiments have quantified the effect of molecular crowding. Computational modeling of a cell (or a part of it) is challenging because of the heterogeneous and complex environment inside a cell. In the current work, we have modeled the cytoplasm of an E. coli cell with a coarse grained description. Our model differs from the previous works is that shape of proteins is taken approximately and hydrodynamics interaction is modeled in a mean field manner. We have performed extensive Brownian dynamics simulation of hundreds of proteins of the model cytoplasm. Our calculated diffusion coefficient values for several proteins match closely with experimentally known values. In the talk the details of implementation of our Brownian dynamics simulation and the analysis will be described. We have also developed an analytical model to describe subdiffusion in crowded environment, which will be described as well.
    (1) A New Coarse-Grained Model for E. coli Cytoplasm: Accurate Calculation of the Diffusion Coefficient of Proteins and Observation of Anomalous Diffusion.
    Sabeeha Hasnain, Christopher L. McClendon, Monica T. Hsu, Matthew P. Jacobson, Pradipta Bandyopadhyay, PLOS ONE, 2014, DOI: 10.1371/journal.pone.0106466
    (2) An analytical correlated random walk model and its application to understand subdiffusion in crowded environment, Sabeeha Hasnain, Pradipta Bandyopadhyay (In preparation).
    Back to the top of the page

  • Accurate atomistic simulations of intrinsically disordered proteins
    Robert B. Best
    National Institutes of Health

    Intrinsically disordered proteins (IDPs), which lack structure in isolation, are increasingly recognized to play important roles in biology. While atomistic models with explicit solvent are the most accurate applicable biomolecular simulation method, their application to intrinsically disordered proteins has been challenging due to obvious shortcomings in the energy functions. I will show how relatively modest changes to the force fields can bring about substantial improvements in treating secondary structure propensity and overall dimensions of disordered proteins, and will present some applications to studying structure and dynamics in IDPs and unfolded globular proteins.
    Back to the top of the page

  • Protein non-folding as a regulatory phenomenon
    Carlos J. Camacho
    Department of Computational and Systems Biology, University of Pittsburgh

    A large number of proteins are sufficiently unstable that their full 3D structure cannot be resolved. The origins of this intrinsic disorder are not well understood, but its ubiquitous presence undercuts the principle that a protein’s structure determines its function. My seminar will present the theory of disorder in proteins, making quantitative predictions regarding its role in protein structure and function. Predictions are validated based on genome-wide surveys of both the amount of disorder for different functional categories and binding affinities for both prokaryotic and eukaryotic genomes. Without assuming any a priori structure–function relationship, we demonstrate that enzymes and low-affinity binding proteins prefer ordered structures, whereas only high-affinity binding proteins (found mostly in eukaryotes) can tolerate disorder. Relevant to regulatory networks, I will show how the increasing amount of intrinsic disorder in eukaryotes is essential to maximize the specificity of promiscuous regulatory interactions. Implications of intrinsic disorder in the evolution of regulatory networks and protein folding will also be discussed.
    Back to the top of the page

  • Modeling protein diffusion in the extracellular matrix
    Christopher Fennel
    Oklahoma State University

    Intercellular signaling is a critical component of organized tissue function in biological systems. For example, in the case of tissue damage or immune response to foreign bodies, there is a localized upregulation of signaling protein production forming a protein gradient which white blood cells follow to accumulate at the target site. Errors in this signaling system can lead to excessive tissue inflammation, such as that seen in sudden allergic reactions or longer-time health concerns like atherosclerosis. We explore here the microscopic interface of a portion of this signaling system through all-atom molecular dynamics simulations of extracellular diffusion of monocyte chemotactic protein 1 (MCP1). We perform a systematic investigation of MCP1 diffusion and adhesion in aqueous environments with varied collagen concentration, and we see quantitative agreement with diffusion in experimental tissue models. Through detailed analysis of protein conformation distributions, we highlight the primary microscopic interactions governing flow of MCP1 and its potential buildup in extracellular environments.
    Back to the top of the page

  • Membrane binding, conformational plasticity and druggability of Ras proteins: A comprehensive computational analysis
    Alemayehu Gorfe
    University of Texas Medical School at Houston

    Despite decades of studies and the well-established roles of human Ras proteins in health and disease, much remains to be learned about how they work at the molecular level. Major challenges associated with the highly dynamic nature of these proteins and the resolution limits of experimental techniques have hampered our ability to address a number of key issues. These include, at the fundamental level, the sources of signaling specificity and non-overlapping membrane organization of Ras homologous, as well as the structural determinants for the distinct role of specific point mutations in different cancer types. At a more practical level, we lacked the means by which to challenge the <93>invincibility<94> of Ras as an anti-cancer drug target. Considerable progress has been made over the last several years toward addressing each of these issues. Molecular simulations and other computational structural biology techniques have made significant contributions to this progress. Our laboratory studies Ras proteins using multi-scale molecular dynamics simulations in solution and membrane environments coupled with a variety of novel concepts, analysis tools and collaborative experiments. The current presentation will focus on the clinically most relevant isoform of Ras proteins: K-Ras. We will discuss our recent efforts toward elucidating the mechanisms of membrane binding and oligomeration of K-Ras as well as the impact of selected point mutations on its dynamics. We will highlight how some of the lessons from Ras can be generalized to other lipid-modified signaling proteins.
    Back to the top of the page

  • Detecting local residue environment similarity for recognizing near-native structure models
    Daisuke Kihara
    Purdue University

    One of the key steps in protein structure prediction/modeling is to recognize near-native models from a pool of alternative models. We developed a new representation of local amino acid environments in protein structures called the Side-chain Depth Environment (SDE) (Kim & Kihara, Proteins 2014). An SDE defines a local structural environment of a residue considering the coordinates and the depth of amino acids that locate in the vicinity of the side-chain centroid of the residue, and thus captures multi-residue interactions. When benchmarked on commonly used computational model datasets, the residue environment score compared favorably with the other existing scoring functions in selecting native and near-native models. SDE was also used in the recent CASP11 protein structure prediction assessment and led our group to the best performance among all the participants in the free modeling category.
    Back to the top of the page

  • Modeling of pH - driven protein unfolding by a combination of Molecular Dynamics techniques
    Maria Kurnikova
    Carnegie Mellon University

    Diphtheria toxin is a multi-domain protein that invades cells using their endocytosis mechanism, in which an endosome, a lipid bilayer vesicle encapsulates a foreign object. The pH inside endosome is then lowered. A translocation domain (T-domain) of the diphtheria toxin undergoes a conformational rearrangement at low pH, followed by membrane insertion, and subsequent translocation of its N-terminus with the attached catalytic domain into the cell. In solution at neutral pH a stand-alone T-domain peptide assumes globular form; acidification of solution triggers its conformational rearrangement and subsequent lipid membrane binding and insertion. The details of this conformational restructuring of the protein are not known. Recently we characterized a pH-dependent partial unfolding of T-domain by using microsecond long MD simulations and Thermodynamic Integration techniques (1). However, simulating statistical ensembles of unfolded conformation of large proteins is still a prohibitively expensive task. We will describe a novel Direct Intra Solute Electrostatic Interactions accelerated Molecular Dynamics (DISEI-aMD) approach to facilitate generating of a partially unfolded protein ensembles at a reasonable cost (2). This method aims to reduce energy barriers within rapidly changing electrostatic interactions between solute atoms at short range distances. It also results in improved reconstruction quality of the original statistical ensemble of the system. We focus on the study of conformational changes of a low pH T-domain model in explicit solvent using the DISEI-aMD. Based on the simulations of the low pH T-domain model we show that the proposed sampling method accelerates conformational rearrangement significantly faster than multiple standard aMD simulations and microsecond long conventional MD simulations. An approach to estimate an unfolded state apparent pKa will also be presented using a clustering technique and a Free Energy Perturbation method.
    1)pH-Triggered Conformational Switching of the Diphtheria Toxin T-Domain: The Roles of N-Terminal Histidines, I.V. Kurnikov, A. Kyrychenko, J.C. Flores-Canales, M.V. Rodnin, N. Simakov, M. Vargas-Uribe, Y.O. Posokhov, M. Kurnikova, and A.S. Ladokhin J. Mol. Biol. 425, 2752 (2013)
    2)Targeting Electrostatic Interactions in Accelerated Molecular Dynamics with Application to Protein Partial Unfolding, J.C. Flores-Canales and M. Kurnikova JCTC (2015)
    Back to the top of the page

  • Simulations on metal ion coupled protein folding and allosteric motion
    Wei Wang
    National Laboratory of Solid State Microstructure and Department of Physics, Collaborative Innovation Center of Advanced Microstructures, Nanjing University,

    Binding of metal ions can reshape the energy landscapes of proteins, thereby modifying the folding and allosteric motions. For example, such binding may make the intrinsically disordered proteins have funneled energy landscapes, consequently, ensures their spontaneous folding. In addition, the binding may activate certain biological processes by inducing related conformational changes of regulation proteins. However, how the local interactions involving the metal ion binding can induce the global conformational motions of proteins remains elusive. Investigating such question requires multiple models with different details, including quantum mechanics, atomistic models, and coarse grained models. In our recent work, we have been developing such multiscale methods which can reasonably model the metal ion binding induced charge transfer, protonation/deprotonation, and large conformational motions of proteins. With such multiscale model, we elucidated the zinc-binding induced folding mechanism of classical zinc finger and the calcium-binding induced dynamic symmetry breaking in the allosteric motions of calmodulin. In addition, we studied the coupling of folding, calcium binding and allosteric motions of calmodulin domains. In this talk, I will introduce the above progresses on the metal ion coupled protein folding and allosteric motions.
    [1] Wenfei Li, Jian Zhang, Jun Wang, and Wei Wang, Metal-coupled folding of Cys2His2 zinc-finger, J. Am. Chem. Soc. 130, 892(2008).
    [2] Wenfei Li, Wei Wang, and Shoji Takada, Energy landscape views for interplay among folding, binding, and allostery of calmodulin domains, PNAS, 111, 10550 (2014).
    [3] Cheng Tan, Wenfei Li, Wei Wang, and D. Thirumalai, Water mediated Calcium coordination and asymmetric allosteric coupling in paired Calmodulin EF-hands, To be submitted.
    [4] Wenfei Li, Jun Wang, Jian Zhang, and Wei Wang, Molecular simulations of metal-coupled protein folding, Curr. Opin. in Struct. Biol.,30,25(2015).
    Back to the top of the page

  • First Principle Solvation Free Energy of Monovalent Ions from Adaptive Force Matching
    Feng Wang
    University of Arkansas

    The interplay of protein hydration and solvation of Ions plays an important role in the stability of proteins. The intriguing correlation is manifested as the famous Hofmeister series. We parameterized pair-wise additive potentials for ion-water interactions using the adaptive force matching protocol developed in the group. By only fitting to electronic structure forces calculated with the Møller-Plesset perturbation theory (MP2) at the second order, the ion-water potential developed correctly predict the solvation free energy of various salts with an error of less than 5% when compared to experimental measurements. The solvation structure of ions is in good agreement with prior simulations based on density function theory (DFT) and QM/MM MP2. The simulations reveal new insight regarding the solvation structure of ions.
    Back to the top of the page

  • New Fold of Estrogen Receptor: Examining Cross-talk between Domains
    Sichun Yang
    Case Western Reserve University

    Estrogen receptor alpha (ER) regulates allosteric signal transduction between hormone and transcriptional DNA binding. Despite the ER being a key target of breast cancer therapeutics, however, little is known about signaling pathways and how its hormone/DNA-binding domains communicate is still elusive. Here, we discover that a novel fold is adopted by ER, completely different from any available fold existed for hormone receptors, and identify that a set of amino acids (mostly hydrophobic) are involved at the interfaces responsible for domain cross-talk. We further derive structure ensembles of this multidomain ER fold by using integrative approaches of modeling on biophysical SAXS and synchrotron radiation measurements, and by utilizing large-scale biomolecular simulations that fully examine the energy landscape of protein-protein interactions occurred in the ER. Discovery of the hidden cross-talk between domains establishes a new pathway of ER allostery for elucidating the notorious resistance of antiestrogen therapy.
    Back to the top of the page

Outstanding Young Researcher Award Talks

  • Mechanical effects of nucleotide binding in Hsp70 from pulling experiments and simulations
    Dale Merz
    University of Cincinnati

    The regulation of protein function through ligand-induced conformational changes is crucial for many signal transduction processes. Heat shock protein (70 kDa) is one example of a protein undergoing conformational changes upon ligand hydrolysis. It consists of a nucleotide binding domain (NBD) covalently linked to a substrate binding domain (SBD) via an interdomain linker [1]. We studied the nucleotide binding domain (NBD) of Hsp70 using a combination of coarse-grained pulling molecular dynamics simulations [2] along with corresponding single molecule experiments to uncover the changes in mechanical stability between the two lobes of the NBD during binding, hydrolysis, and exchange of ATP [3]. We found that nucleotide binding and hydrolysis played a significant part in regulating the mechanical balance between the two lobes of the NBD. ADP binds strongly to lobe II increasing its stability relative to lobe I, which is the more stable lobe in the absence of a nucleotide. Extraction of the C-Terminal helices occurs first independent of the nucleotide status. The helix extraction and shift in the mechanical hierarchy between the lobes of the NBD offer insight into its signal transduction mechanism with the substrate binding domain and can be used for a general interpretation of signal transduction within the actin/sugar kinase superfamily.
    [1] Swain, J. F. J., Dinler, G., Sivendran, R., Montgomery, D. L., Stotz, M., & Gierasch, L. M. (2007). Hsp70 chaperone ligands control domain association via an allosteric mechanism mediated by the interdomain linker. Molecular Cell, 26, 27–39.
    >2] Zhmurov, A., Dima, R. I., Kholodov, Y., & Barsegov, V. (2010). Sop-GPU: accelerating biomolecular simulations in the centisecond timescale using graphics processors. Proteins, 78, 2984–99.
    [3] Daniela Bauer, Dale R. Merz Jr., Benjamin Pelz, Kelly E. Theisen, Gail Yacyshyn, Dejana Mokranjac, Ruxandra I. Dima, Matthias Rief, Gabriel Zoldak. Submitted to PNAS (March 2015).
    Back to the top of the page

  • Microsecond MD simulations to reveal the dynamics and mechanism of a Class B GPCR
    Chenyi Liao
    University of Vermont

    We have studied a class B G-protein coupled receptor (GPCR), which is crucial for transducing signal in nervous system and thus considered a potential drug target for psychiatric treatments. Given the lack of knowledge about the three-dimensional structure, it is a challenging task to understand the activation pathway and the regulation mechanism of this GPCR. Although its antagonism is believed to provide promising treatment to chronic stress, rational design of small-molecule antagonists has not been successfully achieved. To gain structural and mechanistic insight at the molecular level, we created reliable GPCR homology models, and carried out microsecond-long simulations. We have observed high stability of the transmembrane domain and N-terminal extracellular domain in our simulations, while the linker connecting these domains is fairly flexible — leading to multiple conformational states of our GPCR protein. In particular, two major states, for the first time, have been identified, which are likely related to ligand binding and unbinding behaviors. Further, a careful analysis based on the community network has revealed critical information about the protein dynamics, providing hints for us to propose the probable activation mechanism. Moreover, our water density analysis helps us to precisely locate the potential ligand-binding site, which will guide rational design of small-molecule antagonists. In summary, our long simulation studies reveal the detailed dynamics to shed light on rational antagonist design to modulate the targeted class-B GPCR receptor.
    Back to the top of the page

  • Systematic Characterization of Protein Folding Pathways using Diffusion Maps and Molecular Simulation
    Sang Beom Kim
    Princeton University

    Understanding the mechanisms by which proteins fold from disordered amino-acid chains to unique secondary/tertiary structures remains an active area of research. In complement to experimental findings, molecular simulations provide atomistic details of protein folding mechanisms and dynamics. We present the first study of the explicit characterization of folding pathways of Trp-cage miniprotein that utilizes the combination of molecular dynamics simulation and the nonlinear dimensionality reduction technique known as diffusion maps. Conventional order parameters, such as root-mean-square deviation and radius of gyration, are commonly used to provide structural information but fail to effectively capture the underlying dynamics of the protein folding process. It is therefore advantageous to adopt a method that can systematically analyze simulation data to extract relevant structural as well as dynamical information. Diffusion maps automatically embeds the high-dimensional folding trajectories into a lower-dimensional space from which one can more easily visualize folding pathways. The eigenvectors that parameterize the low-dimensional space, furthermore, are determined systematically, rather than chosen heuristically, as is done with phenomenological order parameters. We show that two distinct folding pathways and several important intermediate structures can be characterized with a two-dimensional diffusion maps embedding, while the conventional phenomenological observables can only capture such mechanistic details only imperfectly. Importantly, the identified pathways and intermediates are consistent with previous experimental and simulation studies. This demonstrates that diffusion maps technique can be employed as an effective way of analyzing and constructing protein folding pathways from molecular simulations, with no need for a prior knowledge of the order parameters to characterize the folding dynamics.
    Back to the top of the page

  • Multiscale simulation reveals a multifaceted mechanism of proton permeation through the influenza A M2 proton channel
    Ruibin Liang
    University of Chicago

    The influenza A virus M2 channel (AM2) is crucial in the viral life cycle. Despite many previous experimental and computational studies, the mechanism of the activating process in which proton permeation acidifies the virion to release the viral RNA and core proteins is not well understood. Herein the AM2 proton permeation process has been systematically characterized using multiscale computer simulations, including quantum, classical, and reactive molecular dynamics methods. We report, to our knowledge, the first complete free-energy profiles for proton transport through the entire AM2 transmembrane domain at various pH values, including explicit treatment of excess proton charge delocalization and shuttling through the His37 tetrad. The free-energy profiles reveal that the excess proton must overcome a large free-energy barrier to diffuse to the His37 tetrad, where it is stabilized in a deep minimum reflecting the delocalization of the excess charge among the histidines and the cost of shuttling the proton past them. At lower pH values the His37 tetrad has a larger total charge that increases the channel width, hydration, and solvent dynamics, in agreement with recent 2D-IR spectroscopic studies. The proton transport barrier becomes smaller, despite the increased charge repulsion, due to backbone expansion and the more dynamic pore water molecules. The calculated conductances are in quantitative agreement with recent experimental measurements. In addition, the free-energy profiles and conductances for proton transport in several mutants provide insights for explaining our findings and those of previous experimental mutagenesis studies.
    Back to the top of the page

  • Accelerating the convergence of replica exchange simulations using Gibbs sampling and adaptive temperature sets
    Thomas Vogel
    Los Alamos National Laboratory

    We present a novel replica-exchange scheme in which an individual replica can sample from states encountered by other replicas at any previous time by way of a global configuration database, enabling the fast propagation of relevant states through the whole ensemble of replicas. This mechanism depends on the knowledge of global thermodynamic functions which are measured during the simulation and not coupled to the heat bath temperatures driving the individual simulations. Therefore, this setup also allows for a continuous adaptation of the replica-exchange temperature set. The method is particularly useful for the fast and reliable estimation of the microcanonical temperature T(U) or, equivalently, of the density of states g(U) over a wide range of energies and can be applied in many areas, including biophysics and materials science, for example.
    Back to the top of the page

Contributed Talks

  • Finding needle in a haystack! Fishing out the druggable proteins from the proteome with a structure-based computational framework.
    Anirban Banerji
    The Research Institute at Nationwide Children Hospital

    The fact that even after decades of intense research about 60% of drug discovery projects fail because the target proteins are found to be non-druggable[1], demonstrates clearly that the contemporary approaches to ascertain druggability do not match the complexity of the problem. We calculated the self-similar symmetries of all 210 pairwise residual interactions in proteins, hypothesizing that the ability of a protein structure to be modulated by drug molecule's binding is related to a protein structure-wide metric based on its residual interactions that may connect local reality of protein space with its global property. Furthermore, because proteins are not classical(Debye) solids but are fractal objects, the self-similar symmetries of residual interactions were calculated by measuring their correlation dimension. Investigations, applying this novel measure on the benchmark druggable and non-druggable dataset defined by[2], revealed 15 signatures of self-similar residual interaction symmetries, which could clearly distinguish between druggable and non-druggable proteins. Such discriminatory power is noteworthy because the hyperplane to segregate the druggable proteins from the non-druggable ones is, going by the present state of knowledge, blurred. Upon applying the measure on another benchmark dataset[3] enlisting diverse druggable targets identified not only through the classical pharmacological means but also by the contemporary genomics and molecular biology based methods, equally strong discriminatory characteristics to separate druggable targets were recorded. Finally, upon applying the same measure on all the known targets against the disease Tuberculosis enlisted in "DrugBank", the signature patterns in the correlation dimension of the pairwise-residual interactions recorded from training set and test set were found to be present in them with 75% confidence interval. An extensive and systematic investigation is now being conducted on the "DrugBank" enlisted targets for other diseases.
    [1] T. Liu, R. B. Altman, CPT Pharmacometrics Syst Pharmacol.,3:e93, 2014
    [2] A. Krasowski, D. Muthas, A. Sarkar, S. Schmitt, R. Brenk, J Chem Inf Model., 51(11):2829-2842, 2011.
    [3] J. P. Overington, B. Al-Lazikani, A. L. Hopkins, Nat. Rev. Drug Discov. 5(12):993-996, 2006.
    Back to the top of the page

  • Synthesis and cytotoxic evaluation of analogues of the tubulin-binding agent soblidotin
    Abu Gafar Hossion
    The University of Kansas

    To effectively kill targeted cells, anticancer antibody-drug conjugates generally incorporate highly toxic small molecules. Small molecules with sufficient potency for these applications include analogues of the natural product dolastatin 10, such as soblidotin. This compound selectively disrupts tumor vasculature and potently inhibits the polymerization of tubulin. Only a few drug classes are considered sufficiently toxic to generate effective antibody drug conjugates. One such class is the pentapeptide soblidotin. In an effort to create more effective antibody-drug conjugates, we synthesized soblidotin and derivatives, and we are investigating the biological properties of these compounds. Many of these agents are highly cytotoxic against human breast cancer (SKBR3), prostate cancer (PC3) and leukemia (Jurkat) cell lines with IC50 values ranging 0.04 nM to 100 nM. Among the investigated soblidotin derivatives, the new compound (soblidotin OEt ester) showed ca. 5 times better cytotoxic effect than soblidotin towards Jurkat cell line. To prepare soblidotin and analogues, we prepared the dolaisoleucine and dolaproline amino acids on gram scales by modifications of existing methods. The synthesis and properties of these compounds will be described
    Back to the top of the page

  • Computational studies of collaboration between E. coli ClpB and cofactors DnaK/GrpE during protein disaggregation
    Yu-Hsuan Shih
    University of Cincinnati

    The ClpB chaperone, which belongs to first class of ATPases Associated with diverse cellular Activities (AAA+) superfamily, plays an essential role for protein disaggregation. Co-factors, such as DnaK/GrpE, collaborate with ClpB to suppress aggregates under stress conditions. To understand the molecular mechanism of ClpB/DnaK/GrpE (BKE) collaboration, we utilize computational modeling and protein-protein docking approaches to investigate binding contacts at the interfaces. To this end, we construct a hexameric model of ClpB by using constraints derived from an asymmetric cryoEM map and crystal structure of monomeric fragments1. Our results, which indicate that GrpE and ClpB compete for DnaK binding, are consistent with in vivo and in vitro biochemical studies. On-going biochemical studies confirm the existence of hot spots on M-domains and Nucleotide binding domain 1 (NBD1), which were predicted by our modeling approach. Interestingly, our docking studies support multivalent binding of DnaK to ClpB assembly, suggesting potential functional regions and residues involved in protein disaggregation. Based on these results, we propose that DnaK may be promoting an active ClpB conformation or helping to stabilize ClpB hot spots via ATP hydrolysis.
    Doyle, S. M.; Shastry, S.; Kravats, A. N.; Shih, Y.-H.; Miot, M.; Hoskins, J. R.; Stan, G.; Wickner, S., Interplay between E. coli DnaK, ClpB and GrpE during Protein Disaggregation. Journal of Molecular Biology 2015, 427 (2), 312-327
    Back to the top of the page

  • Effects of Peptide Concentration, Bilayer Composition and Nanoparticles on the Dynamics and Stability of H-Ras Peptide Nanoclusters
    Xubo Lin
    The University of Texas Health Science Center at Houston

    The dynamics and stability of nanoclusters of lipid-anchored Ras proteins play an important role in Ras signaling. However, how changes in protein concentration and membrane domain stability might affect the stability of Ras nanoclusters is far from clear. Here we used coarse-grained molecular dynamics simulations to examine the effects of peptide concentration, cholesterol concentration and nanoparticles (C60) on the dynamics and stability of peptides representing the H-Ras lipid anchor that form dynamic clusters in a two-domain lipid bilayer. Combined with our previous studies [Janosi L. et al. Proc. Natl. Acad. Sci. U.S.A. 2012, 109, 8097; Li Z. et al. J. Am. Chem. Soc. 2012, 134, 17278], the current simulations indicate a reversible effect of peptide/cholesterol concentrations on the dynamics and stability of H-Ras nanoclusters, and suggest a correlation between the stabilities of lipid domains and peptide clusters. We further show that C60 nanoparticles penetrate into the bilayer core and localize in the liquid-disordered domain; they destabilize the domain boundary and thereby the H-Ras clusters. Taken together, these results suggest that the stability of the boundary-prone H-Ras nanoclusters is largely determined by the stability of the domain boundary
    Back to the top of the page

  • Molecular Dynamics Investigation of surface tension and internal pressure of nanodroplet
    Kai Leong
    University of Arkansas, Fayetteville

    Understanding the physical chemistry of nanodroplets is important for many biological and industrial processes. For example, protein molecules are solvated in nanometer-scale droplets for free electron x-ray laser based structure determination. In this work, the internal pressure of nanodroplets from 4 to 10 nm in diameter was measured with a finite difference method at different probing depth. Our preliminary results indicate a much larger effective surface tension when the bubble is on the sub-10 nm scale. The Young-Laplace equation breaks down for very small bubbles and the internal pressure of the bubbles are not inversely proportional to the bubble radius.
    Back to the top of the page

  • Preformed template fluctuations promote fibril formation: Insights from lattice and all-atom models
    Maksim Kouza
    University of Warsaw

    Fibril formation resulting from protein misfoding and aggregation is a hallmark of several neurodegenerative diseases such as Alzheimer's and Parkinson's diseases. Despite the fact that the fibril formation process is very slow and thus poses a significant challenge for theoretical and experimental studies, a number of alternative pictures of molecular mechanisms of amyloid fibril formation have been recently proposed. What seems to be common for the majority of the proposed models is that fibril elongation involves the formation of pre-nucleus seeds prior to the creation of a critical nucleus. Once the size of the pre-nucleus seed reaches the critical nucleus, its thermal fluctuations are expected to be small and the resulting nucleus provides a template for sequential (one-by-one) accomodation of added monomers. The effect of template fluctuations on fibril formation rates has not been explored either experimentally or theoretically so far. In this paper we make the first attempt at solving this problem by two sets of simulations. To mimic small template fluctuations, in one set, monomers of the preformed template are kept fixed, while in the other set they are allowed to fluctuate. The kinetics of ddition of a new peptide onto the template is explored using all-atom simulations with explicit water and the GROMOS96 43a1 force field and simple lattice models. Our result demonstrates that preformed template fluctuations can modulate protein aggregation rates and pathways. The association of a nascent monomer with the template obeys the kinetics partitioning mechanism where the intermediate state occurs in a fraction of routes to the protofibril. It was shown that template immobility greatly increases the time of incorporating a new peptide into the preformed template compared to the fluctuating template case. This observation has also been confirmed by simulation using lattice models and may be invoked to understand the role of template fluctuations in slowing down fibril el! ongation in vivo.
    Back to the top of the page

  • Alternative approaches to electrostatics accumulation in free energy calculations
    Sherin Paranahewage
    Oklahoma State University

    We are interested in using linearly scaling real-space only electrostatics summation algorithms to assess their suitability for application in calculations of small molecule transfer between air and water. To this end, we calculated hydration free energies using several electrostatics summation algorithms for a diverse set of small molecules and peptide sequences. We monitored computational performance and calculation accuracy for these algorithms, both as a function of increasing cutoff radii. We have identified specific electrostatic summations that exhibit enhanced convergence, and performance tests clearly show the benefits of these approaches over conventional methods for handling long-ranged electrostatic interactions. Decomposition of the free energies into polar and nonpolar components shows that electrostatic convergence is often dependent on the unique structure of individual molecules.
    Back to the top of the page

  • Investigating the domain motions of NaAtm1 family ATP Binding Cassette Exporter
    Yinling Liu
    Clemson University

    The NaAtm1 is a member of the adenosine triphosphate (ATP)-binding cassette (ABC) exporters that use ATP as an energy source in the export of substrates across the membrane in transition metal homeostasis and detoxification processes. NaAtm1 ABC exporter feature specific mutations in both the substrate and ATP binding sites, altering ATPase activity compared to the non-mutated site in apo and thiol- containing glutathione GSH-bound systems. In this study, we first use molecular dynamics (MD) simulations and a molecular mechanics-Generalized Born surface area (MM/GBSA) approach to explore the atomistic basis of substrate specificity in the binding sites and the origins of driving forces for biological function during the substrate translocation cycle. A total simulation time of ca. 5 microseconds was performed. Our results showed that several specific locations in the large drug binding pocket are identified as preferred binding locations. These observations are consistent with the hypothesis that the entire internal cavity plays a role in substrates recognition, binding and transportation. We found that the NaAtm1 ABC exporter undergoes large conformational changes, which lead to the highly conserved nucleotide binding domains (NBDs) coming closer together, in agreement with previous observations for other ABC transporter proteins. Principal component analysis (PCA) and anisotropic network model (ANM) studies suggest that the approach of the NBDs was asymmetrical and some mutants may work by keeping the NBDs apart thus preventing ATP hydrolysis, resulting in a reduced ATPase activity. In addition, key segments from the transmembrane domains (TMDs) and the NBDs are identified to participate in the interaction network that contributes to the conformational transition. Key residue pairs in the coupling helices (CHs), which are located at the TMD-NBD interface with high covariance in motion leading to the closure of the NBDs are also identified. These findings from MD simulations offer significant insights into the structure-function relationship of ABC transporters
    Back to the top of the page

  • Perturbation of dynamic properties through mutations in plasmodium falciparum dihydrofolate reductase
    Tingting Han
    Clemson University

    Dihydrofolate reductase-thymidylate synthase (DHFR-TS) in plasmodium falciparum (pf) is a bifunctional protein, and the pfDHFR domain plays an essential role in the folate pathway, reducing dihydrofolate (DHF) to tetrahydrofolate (THF), which is crucial in the production of DNA, RNA, and protein. Therefore, Dihydrofolate reductase (DHFR) usually acts as the target of the antifolate drugs. Pyrimethamine (Pyr) is one of antimalarial drugs. However, during the course of Pyr treatment, mutations have emerged, such as S108N, C59R/S108N, N51I/C59R/S108N/I164L/, which have led to antimalarial resistance. To gain more insight into the drug resistance mechanisms, we applied molecular dynamics simulation to pfDHFR-TS and three popular mutants, which are complexed with Pyr and NADPH. The results indicate that mutations have caused significant conformational changes in and out of protein binding pocket. Molecular mechanics/Generalized Born surface area (MM-GBSA) calculations are then applied to evaluate the binding free energy between Pyr and protein. The results are consistent with experimental observations that indicate the quadruple mutant is most resistant to Pyr relative to the wild type, single mutant and double mutant protein mentioned above. The effect of mutations on collective motions of pfDHFR-TS is also explored through principal component analysis and cross-correlation. MD simulations of pfDHFR-TS in the absence of Pyr are then performed and, through comparison to our studies of Pyr complexed with pfDHFR-TS, a model describing how Pyr binding perturbs protein conformational dynamics emerged.
    Back to the top of the page

  • The Binding Landscape of CENP-A/H4 Dimer Reveals Thermodynamic Distinctions from H3/H4 Dimer
    Haiqing Zhao
    University of Maryland

    Centromere protein A (CENP-A) is a centromere-specific variant of histone H3 and shares ~50% amino acid identity with canonical H3 protein. CENP-A is required to package the centromere, and to separate sister chromatids during mitosis. Despite their discrete functions, previous reported co-crystal structures reveal surprising similarities exists between CENP-A/H4 and H3/H4 dimers. In this work, to distinguish features of CENP-A/H4, which might be unique, we use molecular dynamics simulations to map the binding free energy landscape for CENP-A/H4 and H3/H4 dimers. The Associated memory, Water mediated, Structure and Energy Model (AWSEM) and umbrella sampling were applied for each simulation to obtain two-dimensional free energy profiles of monomeric protein association and folding. Interestingly, our calculations revealed significant thermodynamic distinctions between the dimerization profiles of CENP-A/H4 and of H3/H4 pairs. Furthermore, the free energy landscape of CENP-A/H4 dimer is significantly remodeled in the presence of its cognate chaperone HJURP. These results are in general agreement with the available experimental data and provide new thermodynamic insights into the mechanisms underpinning chaperone-mediated histone variant CENP-A nucleosomes assembly in vivo.
    Back to the top of the page


  • Stability of the self-assembly of computer-aided design structure
    Kadir Demir
    Bulent Ecevit University

    In order to understand the mechanism of folding and association of a special tetramer structure, we have perfomed MD simulation to an asymmetric tetramer unit using antiparallel dimers whose structures were solved using by x-ray crystallography (PDB accession code 3S0R) . The stability of the system was checked for several temperatures from 300 K to 325 as well as 500K. All simulations are performed with the latest version of the GROMACS molecular modeling package.
    Back to the top of the page

  • Hybrid MC/MD simulation of the Trp-cage miniprotein in implicit Solvent
    Fatih Yasar
    Hacettepe Üniversitesi

    Traditional Monte Carlo (MC) or Molecular Dynamics methods are not well suited for obtaining a true sample of the complete conformational space of realistic protein and peptides. The main factor that create this difficulty is the very rugged shape of potential energy surface of protein and npeptides which usually causes conventional simulation methods to be come trapped in the valley of a particular energy minimum at relatively low temperature. This problem can be overcome by a nnovel technique. In present work, it will created a hybrid form of MC and MD methods and nperform to the Trp-cage miniprotein in implicit solvent
    Back to the top of the page

  • Mixed Model Replica Exchange to Study Switching Proteins
    Nathan Bernhardt
    University of Oklahoma

    The energy landscape of proteins has long since been a topic of great debate. For many years now it has been accepted that the vast majority of proteins have a funnel shaped energy landscape. This was originally suggested as a way to explain the short folding time of proteins. However, this may not always be the case and for some proteins, switching proteins, two funnels in the energy landscape may exist. To better understand such proteins and their folding mechanism we propose to use a modified version of Replica Exchange Molecular Dynamics, implemented into GROMACS 4.6.5, to sample the energy landscape of the GA and GB subdomains of protein G as well as their mutated counterparts GA98 and GB98. Here, we implement a mixed model, a Go model provide by the SMOG website alongside a physical model handled by the AMBER 99sb forcefield, where we couple the 2 models to track one another through a harmonic energy potential. On one side of the replicas, we bias our Go model toward the GA fold and on the other the GB fold. We hope that such sampling will reveal how populated GA is relative to GB when either a GA protein sequence is used or a GB protein sequence. Such information should reveal interactions important to selection of one fold over the other. Additionally, this sampling should help to find intermediate structures in the switching pathway
    Back to the top of the page

  • Effect of single point mutations in a form of systemic amyloidosis
    Manikanthan Bhavaraju
    University of Oklahoma

    Amyloid deposits of light chain proteins are associated with the most common form of systemic amyloidosis. We have studied the effects of single point mutations on amyloid formation of these proteins using explicit solvent model molecular dynamics simulations. For this purpose, we compare the stability of the wild type immunoglobulin light chain protein REI in its native and amyloid forms with that of four mutants: R61N, G68D, D82I, and A84T. We show that the experimentally observed differences in the propensity for amyloid formation result from two effects. First, the mutant dimers have a lower stability than the wild type dimer due to increase exposure of hydrophobic residues. The second effect is a shift in equilibrium between monomers with amyloid-like structure and such with native structures. Our results suggest that for the development of drugs targeting light-chain associated systemic amyloidosis one should look for components that either stabilize the dimer by binding to the dimer interface, or reduce the probability for the amyloid form in the monomers.
    Back to the top of the page

  • Statistical inefficiency of MM/GBSA in protein-protein interactions
    Zhe Jia
    Clemson University

    The Molecular Mechanics/Generalized Born Surface Area (MM/GBSA) method is often used to estimate free energies of macromolecular systems with the aim of achieving a statistical precision of ~1kJ/mol. With a commonly used trajectory snapshot interval of 5ps, 1-10ns simulations usually give satisfactory results for relative binding energies between most proteins and rigid ligands. To systematically evaluate the statistical performance of this method in evaluating protein-ligand binding, we studied the binding of 4 protein-ligand systems and 9 protein folding systems that vary in size, shape, interaction surface area and interaction landscape. In a protein-protein binding system involving a large flexible interaction surface area (~2400ล2), we found the autocorrelation time between MM/GBSA energies is 500-1000ps. Coulombic energies and solvation energies show stronger autocorrelations than VDW energies, due to the slower decay of autocorrelation resulting from the lower degree in the Coulombic and continuum polar solvation energy terms. In considering the full MM/GBSA energy, the autocorrelation of energies has not fully decayed when using the commonly used 5ps sampling interval for flexible protein-protein binding. Considering the standard deviation of 20-100kJ/mol for most cases, 200ns-10μs of total simulation time is required to reach a precision of 1kJ/mol for flexible protein-protein interaction energies. To obtain statistically valid results, the cost of MM/GBSA method changes significantly with the flexibility of macromolecules.
    Back to the top of the page

  • On the lack of polymorphism in ABeta-peptide aggregates derived from patient brains
    Erik Alred
    iUniversity of Oklahoma

    The amyloid beta oligomers and fibrils that are found in neural tissues of patients suffering from Alzheimer's disease may either cause or contribute to the pathology of the disease. In vitro, these Aβ-aggregates are characterized by structural polymorphism. However, recent solid state NMR data of fibrils acquired post mortem from the brains of two Alzheimers patients indicate presence of only a single, patient-specific structure. Using enhanced molecular dynamic simulations we investigate the factors that modulate the stability of Aβ-fibrils. We find characteristic differences in molecular flexibility, dynamics of interactions, and structural behavior between the brain-derived Aβ-fibril structure and in vitro models. These differences may help to explain the lack of polymorphism in fibrils collected from patient brains, and have to be taken into account when designing aggregation inhibitors and imaging agents for Alzheimer's disease.
    Back to the top of the page

  • Tuning force fields using liquid properties to sharpen free energy calculations
    Christopher Fennel
    Oklahoma State University

    We present a simple optimization strategy for incorporating experimental dielectric response information of neat liquids in classical molecular models of alcohol. Using this strategy, we determine simple and transferable hydroxyl modulation rules that, when applied to an existing molecular parameter set, result in a newly dielectric corrected (DC) parameter set. We applied these rules to the general Amber force field (GAFF) to form an initial set of GAFF-DC parameters, and we found this to lead to significant improvement in both the calculated dielectric constant and hydration free energy values for a wide variety of small molecule alcohol models. Tests of the GAFF-DC parameters in the SAMPL4 blind prediction event for hydration show these changes improve agreement with experiment. Surprisingly, these simple modifications also outperform detailed quantum mechanical electric field calculations using a self-consistent reaction field environment coupling term. This work provides a potential benchmark for future developments in methods for representing condensed-phase environments in electronic structure calculations.
    Back to the top of the page

  • Systems Biomedicine & Pharmaceutics
    Ashlee Ford Versypt
    Oklahoma State University

    The Systems Biomedicine & Pharmaceutics Laboratory at Oklahoma State University directed by Dr. Ashlee Ford Versypt is focused on research at the intersection of chemical engineering, computational science and engineering, applied mathematics, biomedical science, and pharmaceutical science. Many applications with great societal impact lie within the intersection of these disciplines, such as advanced pharmaceuticals and medical technologies. The group specializes in developing computational models for predicting dynamic and spatially distributed human physiological systems and pharmaceutical drug delivery devices. The primary methodologies are partial and ordinary differential equation modeling of chemical reactions and mass transport. Active areas of research highlighted in the poster presentation are controlled release drug delivery, the onset of diabetic nephropathy leading to chronic kidney disease, and the role of the extracellular matrix in cancer metastasis.
    Back to the top of the page

  • Tension-maintaining tissue model reveals novel epidermal morphogenesis
    Melville Vaughan
    University of Central Oklahoma

    Myofibroblasts are contractile, collagen-producing cells of wound healing and various pathological contractures and fibroses. Our goal is to study epithelial cell interactions with myofibroblasts in vitro. Traditional models lack tension and are inappropriate for these studies. We developed a tension-maintaining tissue model and found it to provide the necessary microenvironment for myofibroblasts and epithelial cells. This presentation will detail the methods used to generate and culture skin cells using this model, and provide novel observations of how h-ras-overexpressing keratinocytes are affected when cultured in this model as compared to the traditional model.
    Back to the top of the page

  • The stability of the Osaka mutant structure by replica exchange and conventional MD simulation and its comparison with wild type.
    Workalemahu Berhanu
    University of Oklahoma

    An amyloid-beta (Aβ) aggregate is a hallmark of Alzheimer’s disease. Rare familial forms have been identified that involve a single amino acid mutations and that lead to early onset and increased severity of the disease. Experimental attempts to chemically analyze the structure of Aβ and familial from have yet to reveal difference in the mechanism of fibril formation. The Osaka mutation (Aβ1-40E22Δ) quickly forming early-stage fibrils, is more toxic than wild-type Aβ1-40, preformed Aβ1-40E22Δ efficiently cross seeded WT monomer. To better understand the fibril-forming mechanism of the Osaka mutant peptide, we ran several long all-atom explicit water molecular dynamics simulations of the mutant and WT structures starting from NMR experimental structures. We also performed REMD simulation in explicit solvent structural stability of Osaka fibril and Modeling the addition of Aβ1-40 WT monomer to nucleation unit of preformed Osaka fibril. Our simulations show that Osaka mutant model is energetically the most stable configuration; has a wider water channel and water content compared to wild type that could explain its more toxicity and higher aggregation rate. The simulation also indicates the Osaka mutant model is more rigid due to several stabilizing factors including hydrophobic packing and a dynamic network of salt bridge networks
    Back to the top of the page

  • Binding of ACE-inhibitors to In Vitro and Patient-derived Amyloid-β Fibril Models
    Malachi Phillips
    University of Oklahoma

    As of today, no drugs exist that can prevent or reverse Alzheimer’s disease (AD), a neurodegenerative disease associated with the presence of plaque in the brain that is composed of β-amyloid (Aβ) peptides. However, recent studies suggest that angiotensin-converting enzyme (ACE) inhibitors, a set of drugs commonly used to treat hypertension, can prevent amyloid formation in vitro. In the present study, we investigate through computer simulations the binding of ACE inhibitors with patient-derived Aβ fibrils, and compare it with that of ACE inhibitors with in vitro generated fibrils. The binding affinities of the ACE inhibitors are compared with that of Congo Red, a dye that is commonly used to identify amyloid structures and that is known to be a weak inhibitor of Aβ. We find that ACE inhibitors have a lower binding affinity to the patient-derived fibrils than to in vitro generated ones. For patient-derived fibrils their binding affinity is even lower than that of Congo Red. These observations raise doubts on the ability of these drugs to inhibit fibril formation in Alzheimer patients.
    Back to the top of the page

  • Crystal Structure Study of Dihydrodipicolinate Synthase
    Lilian Chooback
    University of Central Oklahoma

    Dihydrodipicolinate synthase (DHDPS) catalyzes the first committed step in the L-lysine biosynthetic pathway in bacteria and plants and is a potential target for development of new antibacterial compounds. The enzyme catalyzes the formation of 2,3-dihydrodipicolinate from pyruvate and L-aspartate--semialdehyde. L-lysine is an allosteric inhibitor of the enzyme. Several crystal structures have been solved of the enzyme in complex with pyuvate, pyruvate and lysine, and acetopyruvate and lysine. Acetopyruvate (ACP), an analog of pyruvate, is a slow-binding inhibitor of DHDPS. The structure of the E:pyruvate complex indicates pyruvate is covalently bound to K131 at the active site. The -carboxyl group of pyruvate is bound to enzyme via interactions with the hydroxyl group of T45 and the backbone nitrogens of T45 and T44. The position and interactions with enzyme groups is only moderately altered in the E:pyr:lys complex compared to the E:pyr structure. ACP is also found to be covalently bound to K161. However, the -carboxyl group of ACP is rotated about 60o compared to the position of the -carboxyl of pyruvate and makes new contacts with the side chain hydroxyl group of T44 and the phenolic group of Y133. The aceto-oxygen of ACP forms a hydrogen bond with T45. The allosteric lysine binding site is located at the interface of two enzyme monomers and binds two lysine molecules at the site. One of the two bound lysine molecules is located in a similar position in both the E:pyr:lys and E:ACP:lys, but the second bound lysine is located in a fairly different position and different interactions with enzyme groups. In addition, in the E:ACP:lys complex the allosteric site is open, but is closed in the E:pyr:lys structure due primarily to movement of H53 and D84 to cover the entrance to the site.
    Back to the top of the page

  • The Final Third: Discovering function for the remaining unknown human genes
    Jonathan Wren
    Oklahoma Medical Research Foundation

    Motivation: Approximately 8,200 (33%) of the 25,000 cataloged Human genes have no publications documenting their function and information on the rest is highly skewed (Figure 1). Even less is known about non-coding RNAs, of which there are about 35,000, and 95% of them are uncharacterized. To better understand human biology, including the contributions of transcripts (protein-coding or non-coding) to aging, we have developed a method to infer function and potential relevance to aging using transcriptional network analysis.
    Results: A program called GAMMA (Global Microarray Meta-Analysis) was written to process over 80,000 GEO human microarray datasets to identify genes with recurrent, reproducible patterns of co-expression across different conditions. Computational benchmarking on genes with known function suggest the approach is accurate, and in 46 out of 54 in-vitro experiments (85%), we have observed the predicted phenotype when knocking down the uncharacterized genes using RNAi.
    Conclusions: We can predict which of the remaining uncharacterized or poorly characterized genes may be relevant to aging-related processes by identifying groups of known aging-related genes they are highly correlated with across heterogeneous experiments.
    Back to the top of the page

  • Accurate pKa Calculation of Titrable Groups in Ensembles of Protein Conformations with Application to pH Dependent Unfolding
    Serzhan Sakipov
    Carnegie Mellon University

    pH-dependent protein conformational change or structure unfolding occur in a variety of mechanisms of protein functioning. We have recently described such pH-dependent structure transformation of a water-soluble form of the Diphtheria Toxin Translocation (T) domain [1]. Acidification of the environment solution leads to the diphtheria toxin conformational reorganization and insertion into a cellular membrane. Partial unfolding and refolding of the T-domain in water with protonated histidine residues was observed in microsecond-long molecular dynamics simulations and supported by experiments [1]. Whether a specific protonatable residue drives modification of the structure depends on the difference in pKa values of such group in different conformations of a protein (for example, in a folded and unfolded state). Even in a stable protein that remains in a single conformation, fluctuations of the structure near equilibrium may affect a pKa. In order to estimate pKa values of the residues in different protein conformations it is necessary to account for structural fluctuations. In this work, Free Energy Perturbation (FEP) method using AMBER with the AMBER99SB force-field was implemented and used for calculating free energy change due to protonation of the titrable groups for proteins in different conformations. Accurate calculations of the method convergence and selection of the reference compounds will also be presented. We calculated the pKas of the histidine groups for different conformations that occur during the T-domain unfolding in MD simulations. We will also present calculated pKas of the histidines in several stable proteins and compare our results with experimental data.
    1. Kurnikov, I. V., Kyrychenko, A., Flores-Canales, J., et al. (2013) pH-Triggered Conformational Switching of the Diphtheria Toxin T-Domain: The Roles of N-Terminal Histidines. Journal of Molecular Biology 425, 2752-2764
    Back to the top of the page

  • Molecular simulations of S100A12 dimerization coupled with calcium binding and allosteric motions
    Wenfei li
    Nanjing University

    Metal ions introduce another degree of freedom to tailor the protein landscape, thus altering their folding, conformational motions, and aggregations. In this work, within the framework of energy landscape perspective, we studied the metal ion coupled dimerization of a calcium binding protein S100A12 by performing coarse grained molecular dynamics simulations. The simulations revealed a tight interplay among the calcium binding, allosteric motions, monomer folding and dimerization. Particularly, the simulations demonstrated the calcium-coordination induced frustrations and the importance of local unfolding during the dimerization of S100A12
    Back to the top of the page

  • Development and application of a genetic algorithm for predictive modeling of a 5-year mortality using questionnaire data
    Lucas Adams
    Oklahoma Baptist University

    Background: As many pattern recognition methods cannot perform optimally with large amounts of irrelevant features, variable selection techniques are often necessary to analyze large datasets. Among such techniques are genetic algorithms (GAs), which use “natural selection” of random variable clusters to find the best predictive variables to achieve the best predictive model. In this study, GAs were implemented to select candidate variables from the National Health and Nutrition Examination Survey (NHANES) questionnaire data for predicting an individual’s 5-year mortality.
    Materials and methods: The NHANES questionnaire initially included 5444 individuals who answered 2058 questions (variables). Variables with >30% missing information, zero variance, were removed from the analysis, as were groups of highly collinear variables. All data were transformed to an integer scale with uniform directionality, with higher numbers indicating poorer health. For variables where directionality was not explicit, the order was inferred from factors known to contribute to mortality. Variables that could not be transformed in this manner were removed, generating 123 variables to be used in our final analysis. Missing information for the remaining variables was imputed using Amelia II as required for implementation of the GA using GALGO software in R. The data were then randomly divided into 70% training (3810 individuals (288 cases/3522 controls)) and 30% testing (1634 individuals (124 cases/1510 controls)) sets. Ten iterations of GA were run using the training set. The variables optimally selected from this procedure were then used to build predictive models using various machine-learning techniques including LASSO, ridge regression (RR), artificial neural network (ANN), boosting, support vector mac! hine (SVM), recursive partitioning and regression trees (RPART), and partial least squares discriminant analysis (PLS-DA). The performance of these algorithms was then compared in order to determine the set of variables demonstrating the most accurate prediction.
    Results: GA was able to select the top 24 questions from an initial 123. Questions related to stroke, emphysema, and general health problem requiring the use of a special equipment were determined to be the most frequent variables selected by GA. Utilization of these top 24 variables for predictive modeling using 5-year mortality as outcomes shows that RR had the best performance (AUC=0.740) while SVM had the poorest (AUC=0.590). This study shows how GA in conjunction with various machine-learning techniques can be utilized for selecting optimum questions predictive of a binary outcome of interest.
    Back to the top of the page

  • Simulation of a weak polyampholyte
    Sahin Uyaver
    Istanbul Ticaret University

    Using grand canonical Monte Carlo simulations of a flexible polyelectrolyte where the charges are in contact with a reservoir of constant chemical potential given by the solution pH, we study the behavior of weak polyelectrolytes in poor and good solvent conditions for polymer backbone. We address the titration behavior and conformational properties of a flexible diblock polyampholyte chain formed of two oppositely charged weak polyelectrolyte blocks, each containing equal number of identical monomers. The change of solution pH induces charge asymmetry in a diblock polyamrpholyte. For diblock polyampholyte chains in poor solvents, we demonstrate that a discontinuous transition between extended (tadpole) and collapsed (globular) conformational states is attainable by varying the solution pH. The double-minima structure in the probability distribution of the free energy provides direct evidence for the first-order like nature of this transition. At the isoelectric point electrostatically driven coil-globule transition of diblock polyampholytes in good solvents is found to consist of different regimes identified with increasing electrostatic interaction strength. At pH values above or below the isoelectric point diblock chains are found to have polyelectrolyte-like behavior due to repulsion between uncompensated charges along the chain.
    Back to the top of the page

  • How BCL-XL Interacts With tBID At Mitochondria To Regulate Apoptosis
    Christina Bruxvoort
    University of Oklahoma Health Sciences Center

    Introduction: Apoptosis is a controlled way for unwanted and damaged cells to perish during normal growth and development in any healthy multicellular organism. Either insufficient or excessive apoptosis can cause diseases like cancer, autoimmunity, and neurodegeneration. BCL-XL and BID are two BCL-2 family members that play antagonistic roles in the intrinsic pathway of apoptosis. Principally, t-BID activates BAX at the mitochondrial outer membrane (MOM) eventually permeabilizing the mitochondria and killing the cell. However, BCL-XL sequesters tBID to prevent it from activing BAX. How BCL-XL and tBid interact at the MOM is largely unknown.
    Methods: Molecular modeling programs were used to generate an in silico structural model to predict key interactions within the BCL-XL/tBID complex. To test this model, disulfide crosslinking of single-cysteine mutant pairs of BCL-XL and tBID followed by BCL-XL and tBID-specific immunoprecipitation were used to capture the potential molecular interactions between full-length BCL-XL and tBID at the MOM. Mutagenesis of the identified interface residues was performed to assess their role in mediating the BCL-XL/tBID interaction and in the BCL-XL inhibition of tBID.
    Results: The computational structural model predicted a BH3-in-groove dimer interface in the BCL-XL/tBID complex. Positive disulfide crosslinking data were obtained from six single-cysteine BCL-XL/tBID mutant pairs, demonstrating the existence of this interface in the MOM-bound BCL-XL/tBID complex and was confirmed with immunoprecipitation.
    Conclusion: The BH3-in-groove interaction mediates the formation of BCL-XL/tBID dimer at the MOM. The mutagenesis study is underway to test the functional significance of this interaction. The knowledge from this study is expected to provide clues for developing modulators of this potentially critical apoptotic interaction that is commonly dysfunctional in cancer, autoimmune disorders, and neurodegenerative diseases
    Back to the top of the page

  • Probing the shear resistance to inter-doublet sliding of the flagellar axoneme
    Gang Xu
    University of Central Oklahoma

    Cilia and flagella are thin subcellular organelles that bend actively to propel cells or move fluid in airways and other passages. The ciliary axoneme (cytoskeleton) consists of nine interconnecting outer microtubule doublets surrounding a central pair of singlet microtubules. Large bending oscillations of the axoneme involve relative sliding of the outer doublets driven by the motor protein dyneins. The genetics and biology of the ciliary function are under extensive study, but the interplay between mechanics and biochemistry of the axoneme remains unclear. In this study, we used the biflagellate alga Chlamydomonas as the model system to probe the shear resistance to inter-doublet sliding in the highly conserved molecular structure of the flagellar axoneme. Controlled bending of individual flagella was induced by either the optical tweezers or microneedles. Piconewton forces and nanometer displacements were measured in order to estimate the elastic shear stiffness and the flexural rigidity of the axoneme based on the beam theory. The quantitative understanding of axonemal biophysics will help illuminate the roles of certain genes and molecular structures in the normal and abnormal axoneme.
    Back to the top of the page

  • Structural Predictions of the Prion N-Terminal Domain
    Maggie Martin
    University of Oklahoma

    Prions are proteins that are found in the neural tissue of most mammalian and some avian species. These proteins can become infectious through the misfolding of a helical domain that in turn initiates the misfolding of other prion proteins in a self-iterative way. The infectious prions aggregate into forms similar to that observed in amyloid diseases such as Alzheimer's. Experimental studies of prion structures are difficult because the first 91 residues of the N-terminal domain are disordered. Alternative means of observing the unstructured region are computational studies. As a starting point, we have compared the structures proposed by various protein-structure-prediction programs such as ITASSER, SwissProt, and Modeller. These programs are based on known structural data and can provide a reasonable guess of the unstructured region of the prion protein. After prediction of the protein structure, we preform all-atom molecular dynamic simulations in order to compare the properties of the various models. Using NMR data for the helical C-terminal domain as a reference, we hope to discern a starting point for future molecular dynamic studies that would model the transition to the prion disease state. Hence, the data from this and future studies provide a mechanism for the conversion of prions to the infectious structure, and for aggregation of this structure into higher-order aggregates. Such information may open new avenues for structure-based design of drugs targeting prion diseases.
    Back to the top of the page

  • Phenotypes associated with mutation in genes of Pseudomonas aeruginosa condensins
    Hang Zhao
    University of Oklahoma

    Pseudomonas aeruginosa is an opportunistic human pathogen, which presents a serious threat to patients. Chromatin structure is a major factor that governs bacterial adaptation to environmental changes and thereby increases fitness of bacteria. Unlike the usual laboratory strains, P. aeruginosa encodes several specialized condensins which appear to be differentially expressed during cell growth. In PAO1, two family of condensins were identified, which might serve as different functions. Both SMC and MksB contribute to faithful chromosome partitioning. The copy number of them varied, with about 100 fold higher of MksB than SMC. Increased frequencies of anucleate cells were observed, and decreased virulenceand Bacterial load in the lung tissues were found in SMC/MksB mutants
    Back to the top of the page

  • Characterization of Pseudomonal condensins MksBEF1 and MksBEFG2
    Viridiana Herrera
    University of Oklahoma

    Chromosomes of all living organisms are folded into highly compact and organized structures; this is required both to fit all genetic material within the cell and to support cellular functions. A widely accepted view depicts Structural Maintenance of the Chromosome proteins (SMCs) as the key component of a protein scaffold that subdivides the chromosome into a series of giant loops. While a single condensin complex governs chromosome organization of some bacteria such as E. coli and B. subtilis, other prokaryotes such as Pseudomonas aeruginosa, require up to three distinct SMCs. The presence of multiple condensins in a single cell suggests that chromosome organization is more complex than previously thought. To shed light on the mechanism of action of bacterial condensins, we seek to characterize the interaction of different Pseudomonal condensins MksBEF1 and MksBEF2 with DNA.
    Back to the top of the page

  • Functional Characterization of a Purified Equilibrative Nucleoside Transporter
    Rebba C. Boswell-Casteel
    University of Oklahoma Health Science Center

    Equilibrative nucleoside transporters (ENTs) are major pharmaceutical targets responsible for modulating the efficacy of more than 30 FDA/EMA approved drugs that treat an expansive range of disease states (e.g., pancreatic cancer, acute myeloid leukemia, non-Hodgkin lymphoma). In fact, expression levels of human ENT1 have been linked to prolonged survival for pancreatic cancer patients receiving gemcitabine treatment – a nucleoside analog transported by ENTs. However, the molecular mechanism and chemical determinants of ENT-mediated substrate transport, and the atomic resolution topology of ENTs, remains a mystery. The current studies are focused on defining the molecular basis for how therapeutics interact with this class of integral membrane proteins. Function Unknown Number 26 (FUN26) is a yeast ortholog of the human equilibrative nucleoside transporter (ENT) family. FUN26 was expressed and purified to homogeneity and incorporated into proteoliposomes for functional analysis. A fundamental element of defining mechanism is the identification of mutations that significantly alter protein function. Gain-of-function mutations have been identified that significantly alter ENT substrate specificity by allowing the transport of nucleotides. An ab initio structural model was generated and suggests the mutations play a role in substrate binding and conformational switching/gating of the protein. Additionally, FUN26 has positional sensitivities for nucleoside substrates, which appear to alter substrate selectivity. Defining the molecular transport mechanism and the chemical properties that govern substrate transport will facilitate the development and tuning of novel therapeutics that selectively utilize isoform-specific transporter functions of ENTs to gain access to intracellular targets. Therefore, targeting the transport mechanism of ENTs through rational drug design will benefit multiple disease states requiring pharmaceutical intervention.
    This work is supported by the NIH COBRE award (P20GM103639), OCAST (HR11-046), and AHA predoctoral fellowship (13PRE17040024).
    Back to the top of the page

  • The role of condensin MksBEF in the physiology of Pseudomonas aerugionsa
    April Clevenger
    iUniversity of Oklahoma

    Pseudomonas aerugionsa is a highly adaptive, robust human pathogen. The PA01 strain encodes two condensins, SMC-ScpAB and MksBEF. Our overall objective was to determine if MksBEF plays a physiological role in Pseudomonas aeruginosa. Specifically, we analyzed the contribution of MksBEF to biofilm formation. In order to determine this, we tested the effects of PA01 condensin deletions, mksB ATPase point mutants, and complementation constructs on biofilm formation relative to PA01 WT. We have found that both deletion of MksB and point mutations of MksB in the ATPase region results in reduced biofilm formation. Complementation of this phenotype was initially difficult to demonstrate owing presumably to complex epigenetic behavior in Pseudomonas aeruginosa. To overcome this problem, a technique using the Degron system was developed. Current experiments involve using a Tn5 transposon to generate stable libraries of mutants in PA01 strains (∆MksB1_SMC’ and ∆SMC_MksB1’). These libraries are then replica plated on +/- IPTG plates in order to screen for synthetically lethal mutants that correlate with conditional MksB and SMC strains. Transposon insertions will then be identified using ligation mediated pcr and sequencing. The identification of synthetically lethal mutants will help elucidate pathways in PA01 that compensate for condensin defects.
    Back to the top of the page

  • Condensin dependent potentiation of antibiotics
    Zoya Petrushenko
    University of Oklahoma

    The emergence and spread of antibiotic resistant pathogenic bacteria presents an increasingly acute threat to public health. In this light, identification of new potential anti-bacterial drug targets remains of high importance. This proposal focuses on bacterial condensins, which play a central role in chromosome organization and segregation. Correct chromosome packing is essential for faithful completion of replication and segregation of genetic information and, thus, is a fundamental aspect of cell metabolism. As a potential drug target, condensins remain underexplored. In bacteria, inactivation of condensins markedly reduces cellular fitness and dramatically potentiates activity of some antibiotics currently used in clinics. This proposal seeks to explore the mechanism of drug potentiation by condensins with the emphasis on identification of metabolic pathways that could contribute to this phenomenon and development and validation of MukBEF as a potential drug target. We expect that these data will help develop new potential drug targets that could expand the efficacy of some antibiotics currently used in clinics.\r\n PROJECT PROGRESS\r\nTo study condensin dependent potentiation of antibiotics we created several bacterial strains with deletion of condensin and tolC (a part of E.coli multidrug efflux pump) genes. 鎠mukB鎠tolC E.coli mutants were hypersensitive to antibiotics novobiocin and fluoroquinolones, but not to several other drugs. Constructed strains were used for development of semi-robotic drug potentiation assay for screening of NCI diversity library. The library screening revealed several compounds, which were potentiated by novobiocin but not erythromycin. We found defects in nucleoid segregation (par phenotype) and cell morphology of E.coli, grown in liquid medium, containing hit compounds “H1” or “H2”. In both cases there was dramatic increase of anucleated and elongated cells. To study effect of the drugs on biochemical activities of Mu! kB we created plasmids with mukB-9his-strepII and mukB-9his-strepII-MBP TAP tags fusion, which were used for protein purification. To distinguish which cellular pathways are effected by the drugs and to characterize genetic interactions of condensins we constructed plasmids, carrying novobiocin and norfloxacin resistant mutants of E.coli type-2 DNA topoisomerases, DNA gyrase and topo IV. This constructs will be used for replacement of endogenous E.coli topoisomerase genes with drug resistant variants
    Back to the top of the page

  • Structure, Dynamic and Interaction Studies of Membrane Glycoproteins by Nuclear Magnetic Resonance Spectroscopy
    Gabriel Cook
    Oklahoma State University

    a A large number of important glycoproteins are integral membrane proteins. They can be found in the lipid bilayers that make up the plasma membrane and the membranes of organelles. Nuclear magnetic resonance spectroscopy has been shown recently to be the ideal method for studying large membrane proteins in a native-like lipid environment. A combination of solution and solid-state NMR will be employed to study the effects of glycosylation on structure, dynamics and function of these important proteins. We aim to prepare glycosylated membrane proteins in lipid environments that resemble biological membranes that are suitable for NMR. These studies will be the groundwork for future studies exploring protein-protein interactions of membrane glycoproteins
    Back to the top of the page