David Baker - US grants

The proposed research involves the biosynthesis of the four nucleoside analogs, 2'-deoxycoformycin (NSC-218321, 2'-dCF), 2'-chlorodeoxycoformycin (2'-C1dCF), coformycin and 9-Beta-D-arabinofuranosyladenine (NSC-404241, ara-A) which are elaborated by Streptomyces antibioticus. The research goals of this application are: (i) to study the biosynthesis of 2'-dCF; (ii) to determine the mechanism by which adenosine is converted to ara-A by the partially purified adenosine-2'-epimerase isolated from S. antibioticus, and (iii) to perform molecular cloning of the genomic DNA for adenosine-2'epimerase. Information will be gained with respect to the biosynthetic interrelationship of these antiviral/antitumor nucleoside analogs. Mutants which have additional growth requirements as compared to the original S. antibioticus strain will be selected. Once the mutants are obtained, it will be possible to elucidate the biosynthetic interrelationship of the the nucleoside antibiotics with respect to requirements for amino acids, carbohydrates or nucleosides. Experiments are described to determine how the "extra" carbon from C-1 of D-ribose is inserted between N-1 and C-6 of the purine ring to form the 1,3-diazepine ring, how the reduction at C-2' of the D-ribosyl moiety of the 2'-dCF occurs and how the chlorine is inserted at C-2' to form 2'-C1dCF. 1H, 2H, 13C NMR spectroscopy, as well as MS techniques, will be used to determine product composition and labeling. The application of molecular cloning to the Streptomyces will make it possible to isolate and analyze the DNA and to study the regulation of adenosine-2'-epimerase. The introduction of the epimerase DNA into S. lividans followed by expression of this enzyme will provide us with sufficient epimerase to complete enzyme kinetic studies in the conversion of adenosine to ara-A.

Acetomycin (1) is a polyfunctionalized lactone that has demonstrated potent antitumor activity in cell culture assays against a variety of tumors including those in a human cloning system. The in vivo antitumor activity, however, is greatly diminished owing to a rapid deacetylation and degradation of the compound. Herein is proposed a study in which prodrugs of 1 and certain congeners of 1 will be synthesized and evaluated for antitumor activity. A total synthesis of acetomycin is proposed. Evaluation of all new compounds will be conducted in both in vitro and in vivo screens.

The Drug Synthesis and Chemistry Branch (DS&CB) of the Developmental Therapeutics Program (DTP) of the Division of Cancer Treatment (DCT) of the National Cancer Institute (NCI) is seeking contractors with chemical synthesis expertise to synthesis a variety of compounds for evaluation as potential anticancer agents. The objectives of this project are to: synthesize congeners of synthetic compounds with confirmed activity; design and synthesize "pro-drugs" and other compounds that possess elements of both congener and pro-drug; and synthesize compounds related to products of natural origin and other related heterocycles.

This proposal is a plan to design, synthesize and evaluate a number of inositol and phosphatidylinositol (PI) analogues as selective probes for key enzymes in the phosphoinositide pathway. The biological goal is to gain understanding, through the use of biochemical probes, into the regulation of cell growth, and, hopefully, cancer. The strategy is essentially threefold: (1) Design, synthesis, and evaluation of suitable fraudulent cyclitols that are substrates and/or inhibitors for phosphatidylinositol synthetase (PI synthetase). (2) Study, either through synthetic or biosynthetically produced fraudulent analogues, their effects on the other enzymes: phosphatidyl inositol kinase (PI kinase), phosphatidylinositol-4-phosphate kinase (PIP kinase) and phospholipase C. This component of the project allows for strategically modified PI's to be evaluated as rationally designed inhibitors. (3) Finally, a series of phosphonate analogues to be designed on the basis of the results of these enzyme studies are proposed as possible membrane-permeable PI analogues. Our primary interest in designing probes for these enzymes, which are believed to be key factors in the intracellular communication process, stems from the possibility that neoplastic growth may result, at least in part, from an altered response to growth factors, of which the enzymes of the PI pathway play a fundamental role. The discovery of hormone and growth factor induced PIP hydrolysis by phospholipase C suggests that these enzymes are viable new targets for drug design. As some specificity could result because cells differ in specific growth factors to which they respond, one might predict that it should be possible to target a narrower subset of proliferative cells than is possible with the usual chemotherapeutic agents which block ubiquitous and essential cellular processes.

antiAIDS agent; AIDS; drug design /synthesis /production; prodrugs;

The modification of antiviral drugs identified as effective against HTLV-III/LAV/HIV in vitro by the screening efforts of NIAID or National Cooperative Drug Discovery Groups for the Treatment of AIDS (NCDDG-AIDS) may be required to allow the delivery of drugs to the central nervous system (CNS). Thus, the objective of this proposed project is to ensure that efforts will be made now to develop the expertise necessary to deliver drugs to the CNS. NIAID, in collaboration with the United States Army Research Development Command, has established a rapid, in vitro screening program to evaluate the effectiveness of potential HTLV- III/LAV/HIV drugs. NIAID will undertake the lead role this year, in collaboration with NCI, in organizing scientists into groups focused on the discovery of novel drugs for the treatment of AIDS (NCDDG-AIDS). Through these efforts and other independent efforts, drugs which will prevent the replication of retroviruses will be identified and developed by the AIDS Program. Drugs which prevent HTLV-III/LAV/HIV replication may cross the blood brain barrier and achieve therapeutic levels efficiently, poorly or not at all. Recent reports have shown the ability to make dihydropyridine derivatives of nucleosides by the attachment of a chemical carrier through an ester linkage. These modified drugs (termed prodrugs) are greatly enhanced in their ability to cross the blood brain barrier. The successful development of improved methods for delivery and targeting of effective agents to the CNS will be especially beneficial to halt the progression of the disease, the spread of the infection and control a reservoir of the virus. The purpose of this solicitation is two-fold: first, to modify known antiretroviral drugs to increase their ability to cross the blood brain barrier; second, to encourage the development of innovative approaches for targeting drugs to the central nervous system.

MAO for the synthesis of drugs for DCT.

neoplasm /cancer pharmacology; antiAIDS agent; drug design /synthesis /production; drug screening /evaluation; antineoplastics; chemical synthesis; drug metabolism; peptides; analytical chemistry; heterocyclic compounds; nucleosides; ultraviolet spectrometry; nuclear magnetic resonance spectroscopy; infrared spectrometry;

9458178 Baker Little is known about how amino acid sequences specify the native states of proteins. The proposed research is a combined molecular biological and biophysical approach to this problem. Because the complexity the folding problem increases with chain length, the research is focused ont he shortest sequences known to fold into unique, stable structures without disulfide bonds. The goal is to identify the determinants of folding for two small - proteins which have similar secondary structural elements but different folding topologies. Both in vivo and in vitro selection methods will be used to identify 1) large sets of divergent sequences which adopt each folding topology 2) amino acid sequence changes which convert one topology into another. The 3-D structures adopted by particularly interesting sequences will be studied in detail using NMR and X-ray crystallography. Analysis of the resulting sequence- structure database and comparison with the sequences and structures of natural - proteins will seek to identify the sequence patterns which specify each of the two topologies. A detailed understanding of how amino acid sequence specifies tertiary structure for these particularly simple proteins should contribute to the understanding of the folding of more complex proteins. %%% Due to advances in technology and the initation of large scale DNA sequencing projects, DNA sequence information is being generated at an ever increasing rate. However, understanding the role of the newly sequenced genes is difficult from the DNA sequences alone; what one needs to know are the structures and functions of the proteins that the genes encode. The research project is directed at understanding the rules which connect amino acid sequences with protein structures for two particularly simple proteins. To the extent that the rules generalize, this knowledge should contribute to the understanding of the folding of larger and more complex proteins, and thus to th e interpretation of data generated in genome sequencing projects. ***

Protein folding requires the existence of pathways to the native state. Although the contribution of amino acid residues to the thermodynamic stability of proteins has been intensively studied, very little is known about how amino acid sequences specify folding pathways. The proposed research is a combined molecular biological and biophysical approach to this problem. Because the complexity of the folding problem increases with chain length, the research will focus on one of the shortest sequences known to fold into a unique, stable structure without disulfide bonds: the 56 residue IgG binding domain of Peptostreptococcal Protein L. Extremely heavy mutagenesis protein L followed by selection for IgG binding using the phage display technology will be used to generate a database of very divergent sequences which adopt the same fold. Analysis of features conserved in the database should identify residues and interactions important in specifying the folding pathway. Determination of the folding times of a divergent subset of the sequences will provide insight into how sequence controls the selection and rate of traversal of kinetic pathways. The folding pathways of the most slowly folding mutants will be mapped using NMR methods. The sequence, rate and structure database together with the biophysical data on the folding pathway will be used to guide and constrain the development of a quantitative theory for the folding of this small protein. A detailed understanding of how amino acid sequence specifies tertiary structure in this simplest possible case should contribute to the understanding of the folding of more complex proteins.

This proposal requests biophysical instrumentation and computing equipment to enable a broad interdisciplinary investigation of fundamental aspects of protein folding and protein-ligand interactions. The participating research groups are from the Biomolecular Structure Center and the departments of Biochemistry, Biological Structure, and Bioengineering at the University of Washington. The dramatic increase in the power of computers over the last five years has ushered in a truly exciting new era in structural molecular biology. There is now hope of understanding the multitude of complex interatomic interactions in proteins at the level of protein folding and proteinligand interactions. The development of adequate computational models, however, requires detailed biophysical characterization of relevant experimental systems. We propose to take a combined biophysical and computational approach to understanding the interactions which underlie the structure and function of proteins. Spectroscopic methods will be used to probe the kinetics and thermodynamics of protein folding and protein-ligand interactions in detail. These results will be used to develop computational models that have as their long range goals the understanding of the rules linking amino acid sequences, protein 3-dimensional structures, and protein-ligand interactions. The work described in this proposal has three main thrusts. The first part is focused on understanding the dependence of protein folding on amino acid sequence, using extremely simple model proteins. The second part is focused on the detailed understanding of a model proteinligand interaction: the binding of biotin to streptavidin. The third part seeks to extend knowledge gained in parts one and two to develop more general methods for experimentally and computationally charactelizing a wide valiety of protein ligand interactions, and for understanding the "restricted folding" of a short peptide which is enclosed by a larger multimeri c protein assembly. The links between three areas are not merely conceptual; the biophysical and computational methods required to address all three issues are highly overlapping. In each case the range of scientific questions currently accessible is limited by our current inability to experimentally determine such key kinetic parameters as on-rates and off-rates and to monitor protein folding reactions. We are similarly limited by current computational resources as to the size and complexity of model systems which may be simulated. The requested instrumentation therefore consists of (1) absorption and circular dichroism spectrometers with associated stopped-flow apparatus to follow the thermodynamics and kinetics of protein folding in solution, (2) surface plasmon resonance (SPR) apparatus which allows exquisitely sensitive measurement of the kinetics of protein-ligand binding, and (3) computer equipment to support the theoretical and predictive side of the investigation as well as visualization of the model systems under study.

The objective of this proposal is to use a combination of experimental and computational approaches to develop a physically plausible model for the folding of the src SH3 domain capable of accurately predicting the results of further experimental measurements. The rate limiting step in folding will be probed by analyzing the kinetic consequences of both single and multiple amino acid substitutions in combination with both all atom and simplified computational models. The starting point of the folding reaction, the denatured state, will be probed by studying peptide models, the kinetics of exchange from the native state, and by direct NMR characterization. The importance of hydrophobic-hydrophyllic patterning and the relationship between sequence conservation and kinetics will be investigated using phage display selection methods. The experimental data will be compared with the results of computational methods including a recently developed ab initio folding simulation method, a simple model which computes the rate and transition state structure from the sequence and native state using simple physical principles, and all atom MD simulations. The ultimate goal is a therapy for beta sheet proteins capable for predicting the folding rate, the structure in the transition state ensemble, and the residual structure in the denatured state from the sequence and structure of the native protein.

The relationship between amino acid sequences, protein structures, and folding kinetics will be probed using a combination of combinatorial library selection methods, biophysical characterization, and computational analysis. Folded proteins will be retrieved from high complexity libraries using phage display and in vitro selection methods in conjunction with a novel "loop entropy reduction" based selection for folding. The selection will be applied both to ensembles of random amino acid sequences constructed from synthetic cassettes, and to amino acid sequences constructed by splicing together approximately 20 residue segments of naturally occurring proteins. The structures of folded proteins retrieved from these libraries will be characterized to investigate how the set of naturally occurring proteins compares to the universe of protein structures accessible to polypeptide chains. In complementary experiments, rational computer based design methods will be used to design large ensembles of sequences likely to fold into specific novel topologies not found in nature, and properly folded proteins will be selected from libraries encoding these sequences using the loop entropy reduction and topology-specific binding selections. The formulation of the potential function and the sequence search strategy used in the library design are natural extensions of our relatively successful approach to ab initio protein tertiary structure prediction. The kinetics of folding of the novel proteins recovered in all of the above selections will be measured and compared to those of naturally occurring proteins with similar lengths and contract orders to determine the extent to which natural selection has operated on folding kinetics. These more general studies will be complemented by a continuation of focused studies on the folding mechanism of the 62 residue IgG binding domain of Protein L using NMR, kinetics and single molecule detection methods. These studies aim to characterize the distribution of structure in the denatured state under as close to physiological conditions as possible., to characterize the rate limiting step in folding in detail, and to identify the origins of the symmetry breaking during folding evident in earlier studies of Protein L folding. Taken together, these studies should illuminate some of the least well understood aspects of sequence/structure relationships in proteins, and contribute to the understanding of the evolutionary history of protein domains as well as to improvements in protein design and protein structure prediction.

structural biology; chemical structure; biomedical resource;

functional /structural genomics; biomedical resource;

Internet; structural biology; protein structure; model design /development; chemical models; computer program /software; computer system design /evaluation; biomedical resource;

protein structure; chemical models; meeting /conference /symposium; biomedical resource;

DESCRIPTION (provided by applicant): Attempts to identify the neural basis of addiction have demonstrated a critical role for glutamate neurotransmission, particularly in the nucleus accumbens, in cocaine-seeking behavior. The experiments in the present proposal will examine the contribution of a novel source of glutamate, specifically nonvesicular glutamate release from cystineglutamate antiporters, to the behavioral and neurochemical effects of cocaine. These studies will test the primary hypothesis that cocaine-induced pathogenic neuroplasticity includes adaptations in cystine-glutamate antiporters, and targeting these adaptations represents a novel approach in treating addiction. Experiments in the first aim will determine whether glutamate released from cystine-glutamate antiporters blocks cocaine reinstatement by stimulating group 2/3 metabotropic glutamate receptors. This could potentially block cocaine reinstatement by preventing cocaine-induced elevations in extracellular glutamate and dopamine, which have been shown by others to be critical for cocaine reinstatement. Toward this end, the capacity of the group 2/3 mGluR antagonist to block N-acetylcysteine regulation of cocaine-induced elevations in extracellular glutamate and reinstatement will be examined. Experiments in the second aim will examine whether cocaine-induced plasticity involving cystine-glutamate antiporters emerges during the course of self-administration or withdrawal and whether these adaptations are sensitive to differential cocaine intake. In addition, these experiments will examine whether cocaine intake and length of withdrawal produce parallel changes in cocaine reinstatement and cocaine-induced plasticity involving cystine-glutamate antiporters. Finally, the last set of experiments will utilize a more clinically relevant procedure to examine the putative anti-craving efficacy of the cysteine prodrug N-acetylcysteine. Specifically, these experiments will examine the capacity of chronic administration of N-acetylcysteine to reverse the neurochemical and behavioral effects of cocaine. It is the goal of this proposal to reveal cystine-glutamate antiporters as a novel target for potential pharmacotherapies for cocaine addiction. Moreover, these experiments also have the potential to illustrate that nonvesicular release of glutamate by cystine-glutamate antiporters is a fundamental component of glutamate neurotransmission in both the normal and diseased states, which would have far reaching implications given the number of disorders that involve glutamate.

[unreadable] DESCRIPTION (provided by applicant): Schizophrenia is a debilitating disorder that affects almost 1% of the world's population. Unfortunately, current antipsychotics induce severe side effects and are ineffective in treating a number of the symptoms of schizophrenia. The development of more effective pharmacotherapies will likely await a better understanding of the neurobiology of schizophrenia. The present experiments will examine the contribution of a novel source of glutamate, specifically nonvesicular glutamate release from cystine-glutamate antiporters, to the behavioral and neurochemical effects of phencyclidine (PCP), which represents one of the most commonly used animal models of schizophrenia. The experiments in the first aim will test the hypothesis that targeting the cystine-glutamate antiporter using the cysteine prodrug N-acetylcysteine represents a novel treatment strategy for schizophrenia. Specifically, these experiments will examine the impact of N-acetylcysteine treatment on locomotor activity, deficits in working memory and social withdrawal) reduced by acute and subchronic administration of PCP. These behaviors have been used previously to gain insight into the neurobiology of schizophrenia. Additional experiments in this aim will utilize in vivo microdialysis to assess the capacity of N-acetylcysteine administration to reverse the neurochemical effects of PCP in the medial prefrontal cortex. The experiments in the second aim will test the hypothesis that a dysregulation in the activity of cystine-glutamate antiporters contributes to the pathophysiology of PCP, and potentially schizophrenia. Specifically these experiments will examine whether acute or repeated administration of PCP alters extracellular or tissue levels of cystine and glutathione in the medial prefrontal cortex. Collectively, these experiments have the potential to identify the cystine-glutamate antiporter as a novel cellular mechanism in the pathophysiology of schizophrenia, as well as to demonstrate the potential antipsychotic properties of N-acetylcysteine. [unreadable] [unreadable]

chemical structure; biomedical resource;

[unreadable] DESCRIPTION (provided by applicant): Protein-protein interactions play critical roles in almost all aspects of cellular function and intercellular communication. The proposed research is directed at improving our understanding of the fundamental contributions to the free energy of protein-protein interactions, and using this understanding to redesign protein-protein interaction specificity and to predict the structures of protein-protein complexes from the structures of the unbound partners. The research will involve an integrated combination of computational and experimental approaches. Feedback from the design and prediction studies will guide the development of an improved physical model for the free energy of protein-protein interactions. The design and prediction methods should add significant value to the large number of monomeric protein structures being determined in current large scale structural genomics efforts by allowing them to be assembled into functionally important complexes, and will provide powerful tools for teasing apart complex interaction networks by reprogramming the specificities of individual pairs of interacting proteins. [unreadable] [unreadable]

DESCRIPTION (provided by applicant) Incorrect enzymatic transformation of DNA, may lead to DNA damage, somatic mutation, cancer, genetic disease, aging, and death. Numerous nucleases involved in DNA repair, restriction, and recombination ("RRR") have been widely used to carry out genetic manipulations in vivo and in vitro, to perform medical diagnostics, and as model systems to study enzymatic reactions. This large class of enzymes is therefore of central importance in medicine and biotechnology. Several RRR nucleases were found to exhibit different 3D folds, typically of the alpha-beta class. Our understanding of sequence-structure-function relationships in these enzymes is severely limited by the slow progress of the structure determination - for most of them the 3D structure remains unknown. This project has two goals: 1) to develop a computational method that generates an experimentally validated model of the protein 3D structure, and 2) to apply this method to determine protein folds of RRR nucleases. For members of all RRR nuclease families, large ensembles of models will be generated using protein fold-recognition methods and the de novo structure prediction algorithm ROSETTA. At least 10 candidates for different 3D folds will be selected and probed by cross-linking, chemical modification and mutagenesis. Best models will be identified based on their fit with the experimental data and further tested by additional experiments. The results will advance our understanding of the structural diversity of RRR nucleases and will provide a structural platform for further studies of the processes of DNA repair, restriction, and recombination. Likely candidates for novel folds will be identified, which could be targeted for structure determination by X-ray crystallography. The computer software and the research protocol developed during this study will yield a predictive method for protein structure modeling that will be broadly applicable to all proteins.

CRISP; Computer Retrieval of Information on Scientific Projects Database; Data Set; Dataset; Development; Disease regression; Funding; Grant; Institution; Investigators; Learning, Machine; Machine Learning; Measures; Methods; Modeling; NIH; National Institutes of Health; National Institutes of Health (U.S.); Performance; Regression; Research; Research Personnel; Research Resources; Researchers; Resources; Score; Source; Structure; Testing; United States National Institutes of Health; kernel methods; protein structure; statistical learning; support vector machine; vector

university

Biological nanomachines are the assemblies that carry out all the basic biological processes in a living organism. Electron cryo-microscopy (cryoEM) is the most appropriate structural tool to determine molecular structures of biological nanomachines that generally consist of multiple protein subunits and/or nucleic acids with a total mass greater than 0.5 million Daltons. The goal is to develop information discovery and integration methodologies for deriving atomic models of nanomachines. Such models will be derived from 3-dimensional (3-D) cryoEM mass density function (i.e. a volumetric density map) in conjunction with physics of protein folding and informatics data. This project is made possible by an integration of the expertise of five investigators in computer graphics, computational biophysics, structural informatics and cryoEM. The intellectual merit of this research is highlighted by the computational approaches of extracting structural information from low-resolution, complex cryoEM volume densities and integrating this information into classical protein structure modeling paradigms, such as comparative modeling and ab initio modeling, for understanding biological nanomachines. The three research goals involve information discovery, information integration and validation of the proposed algorithms. The proposed research will have significant impacts in three disparate disciplines: computer science, molecular modeling, and cryoEM. Furthermore, the team will disseminate their resulting tools freely to the academic community and will host a workshop towards the end of the project. To enhance the impact of their research, the investigators will integrate research with education at each member institution with an eye towards diversity. In particular, these investigators will develop a virtual didactic course in modeling of biological nanomachines for graduate and senior undergraduate students at the five participating institutions.

With this award from the Chemistry Research Instrumentation and Facilities: Multi User program (CRIF:MU), the Department of Chemistry at Youngstown State University (YSU) will acquire a 400 MHz NMR spectrometer. The NMR spectrometer will be utilized in research projects including 1) synthetic heterocyclic chemistry, 2) bioorganic chemistry, 3) flexible macrocycles, 4) polymer and inorganic chemistry, and 5) nanoscale organometallic materials. This proposal will potentially include a large number of diverse and underrepresented students from institutions collaborating with Youngstown State University: Muskingum College (50% student body drawn from low-income Appalachian families), Harold Washington College (65% enrollment is African-American or Hispanic), and Delta College (majority of enrollment consists of first generation college attendees from blue-collar families).

Nuclear Magnetic Resonance (NMR) spectroscopy is one of the most powerful tools available to chemists for the elucidation of the structure of molecules. It is used to identify unknown substances, to characterize specific arrangements of atoms within molecules, and to study the dynamics of interactions between molecules in solution. Access to state-of-the-art NMR spectrometers is essential to chemists who are carrying out frontier research. The results from these NMR studies will have an impact in synthetic organic chemistry and biochemistry.

We will use our computational design methodology, based on an explicit physical model of protein-DNA interfaces, to design novel homing endonuclease variants predicted to cleave specifically within sites in XSCID and other therapeutically important genes. Genes corresponding to the designed proteins will be synthesized, the in vitro and in vivo cleavage specificities determined in Component 2 - Monnat, Component 4 - Scharenberg, and Component 5 - Stoddard groups, improved variants obtained using molecular evolution by Component 4 - Scharenberg, and structures determined of promising designs in Component 5 - Stoddard. These data will be used to refine and improve our computational design methodology. Shortcomings of the physical model underlying our current approach include the limited treatments of backbone flexibility and of water-mediated hydrogen bonding interactions which can contribute significantly to the energetics of protein-DNA interactions, and we will use the experimental feedback to improve both aspects of our model;for example the crystal structures will guide our approach to modeling backbone flexibility. We will use the improved computational design methods to design a second round of endonucleases with therapeutically important cleavage specificities and these will be characterized as in the first round and the experimental feedback used to further refine the computational method. By cycling in this way between detailed computational modeling and in depth experimental characterization we aim to develop robust methods for creating tailored enzymatic reagents for targeted genetic therapies via gene correction.

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. Protein-protein interactions are a central part of most biological processes, and our ability to understand and engineer these fundamental contacts will enable us to rationally modify these processes, including those that are involved in cancer, infectious or neurodegenerative diseases. To generate de novo protein-protein interactions, we first focus on the design of a small number of very high affinity sidechain-target interactions, following up with the design of surrounding residues to keep the key interactions in place. Designed binders are expressed as large libraries and screened via yeast surface display and fluorescence activated cell sorting to isolate the cells displaying the highest affinity binders. After further characterization of identified binders, affinity maturation is performed to achieve high-specificity and affinity binders, and to highlight ways to further optimize our computational design approach. Current targets under investigation for the generation of novel binding partners are several virulence factors and cytokines.

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. Reliable structure-prediction methods for membrane proteins are important because the experimental determination of high-resolution membrane protein structures remains very difficult, especially for eukaryotic proteins. However, membrane proteins are typically longer than 200 aa and represent a formidable challenge for structure prediction. We have developed a method for predicting the structures of large membrane proteins by constraining helix-helix packing arrangements at particular positions predicted from sequence or identified by experiments. We tested the method on 12 membrane proteins of diverse topologies and functions with lengths ranging between 190 and 300 residues. Enforcing a single constraint during the folding simulations enriched the population of near-native models for 9 proteins. In 4 of the cases in which the constraint was predicted from the sequence, 1 of the 5 lowest energy models was superimposable within 4 A on the native structure. Near-native structures could also be selected for heme-binding and pore-forming domains from simulations in which pairs of conserved histidine-chelating hemes and one experimentally determined salt bridge were constrained, respectively. These results suggest that models within 4 A of the native structure can be achieved for complex membrane proteins if even limited information on residue-residue interactions can be obtained from protein structure databases or experiments.

Addiction to cocaine and other illicit drugs is estimated to cost our society $181 billion which equates to $603 per U.S. citizen. The cost of addiction can be dramatically lowered through the use of treatments;unfortunately, many drugs of abuse, including cocaine, lack a single approved pharmacotherapy. Addiction to psychomotor stimulants, such as cocaine, is marked by a transition in drug consumption from a casual, recreational style of use to a more compulsive, excessive pattern that arises as a result of drug-induced changes in brain functioning. In order to develop effective treatments, it will likely be necessary to identify and target altered brain functioning underlying addiction. Towards this end, drug-induced changes in glutamate release from cystine-glutamate antiporters have been linked to pathological alterations in neural transmission and normalizing cystine-glutamate exchange blocks compulsive drug-seeking in preclinical models. Further, small-scale clinical studies using acetylcysteine to target cystine-glutamate exchange have shown modest efficacy including reduced drug craving and cocaine use. The efficacy of N-acetyl cysteine is limited due to extensive metabolism in the liver and poor passive transport into the brain. As a result, the present proposal seeks to develop novel chemical entities that are more potent and effective in targeting cystine-glutamate exchange in the brain. Aim 1 will involve the design of 32-40 compounds. Aim 2 will utilize in vitro and in vivo screening techniques to determine which compounds are most effective and potent in targeting cystine-glutamate exchange. Specifically, we will use pure glial cortical cultures to determine the capacity of brain cells to utilize the novel ligands to target cystine-glutamate exchange. Next, we will screen the most promising compounds in vivo by assessing the capacity of these ligands to bypass hepatic metabolism, enter into the brain, and target cystine-glutamate antiporters. Aim 3 will determine the potency and efficacy of these novel compounds in blocking cocaineprimed, stress-primed, and cocaine-paired cue primed reinstatement of cocaine-seeking in preclinical models of compulsive drug seeking. Collectively, these experiments have the potential to identify cystine-glutamate antiporters as a novel target in the treatment of addiction and to generate a series of compounds that may ultimately be effective in treating cocaine addiction.

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. Structures of intact receptors with single-pass transmembrane domains are essential to understand how extracellular and cytoplasmic domains regulate association and signaling through transmembrane domains. A chemical and computational method to determine structures of the membrane regions of such receptors on the cell surface is developed here and validated with glycophorin A. An integrin heterodimer structure reveals association over most of the lengths of the alpha and beta transmembrane domains and shows that the principles governing association of hetero and homo transmembrane dimers differ. A turn at the Gly of the juxtamembrane GFFKR motif caps the alpha TM helix and brings the two Phe of GFFKR into the alpha/beta interface. A juxtamembrane Lys residue in beta also has an important role in the interface. The structure shows how transmembrane association/dissociation regulates integrin signaling. A joint ectodomain and membrane structure shows that substantial flexibility between the extracellular and TM domains is compatible with TM signaling.

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. Protein structure determination is a critical step in understanding protein behavior, and traditional methods require a comprehensive set of experimental data in order to generate a useful structure, a requirement that can be notably difficult to fulfill. Our goal is to reduce the amount of data needed to generate a reliable structure by using the Rosetta structure prediction algorithm to introduce structural information that can be derived from physical modeling as well as a statistical analysis of the PDB, information that the Rosetta protocol takes from its suite of score functions used to guide Monte Carlo fragment insertion as well as full atom energy minimization. This free modeling protocol can produce high accuracy models for a subset of structures using sequence alone, and by incorporating limited experimental constraints such as sparse NOE data, pseudo contact shifts, chemical crosslinking, and solvent accessibility footprinting into the energy functions we plan to build a general system for exploring the structural data described by a wider range of experiments.

DESCRIPTION (provided by applicant): A major current focus of modern structural biology is the determination of the structures of macromolecular assemblies and machines. Such systems are often not amenable to high-resolution x-ray crystallographic techniques. This proposal aims to develop powerful new methods which integrate NMR data, information from homologous structures, cryo electron microscopy data, low resolution x-ray crystallographic data, evolutionary covariance data, and other sources of information to generate models with atomic level accuracy for macromolecular assemblies and machines. With collaborators, the new methodology will be used to solve cutting-edge structural biology problems which cannot be solved by currently available methods. The new methodology will be incorporated into the freely available Rosetta software suite for use throughout the scientific community.

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. Primary support for the subproject and the subproject's principal investigator may have been provided by other sources, including other NIH sources. The Total Cost listed for the subproject likely represents the estimated amount of Center infrastructure utilized by the subproject, not direct funding provided by the NCRR grant to the subproject or subproject staff. We describe a general computational method for designing proteins that bind a surface patch of interest on a target macromolecule. Favorable interactions between disembodied amino acid residues and the target surface are identified and used to anchor de novo designed interfaces. The method was used to design proteins that bind a conserved surface patch on the stem of the influenza hemagglutinin (HA) from the 1918 H1N1 pandemic virus. After affinity maturation, two of the designed proteins, HB36 and HB80, bind H1 and H5 HAs with low nanomolar affinity. Further, HB80 inhibits the HA fusogenic conformational changes induced at low pH. The crystal structure of HB36 in complex with 1918/H1 HA revealed that the actual binding interface is nearly identical to that in the computational design model. Such designed binding proteins may be useful for both diagnostics and therapeutics.

The Chemical Catalysis Program supports Professors Thomas Spiro, David Baker, Daniel R. Gamelin, and Christine K. Luscombe, all of the University of Washington, who seek support for research on the creation of metallo-macrocycle-protein constructs as novel photocatalysts for H2 production from water. The central idea of this project is to combine photon capture and H2 production into a single chemical assembly. This proposal brings together the key design elements needed to produce an effective H2-evolving solar photocathode. Progress toward the goal of efficient H2 photoproduction will provide more in-depth understanding of photoelectrochemistry and catalysis. Both topics are of intense current interest, but are generally studied separately. Should the project succeed, it will bring both scientific and practical dividends. Water splitting is a major scientific focus for the future, and these results will garner wide interest.

With the support of the Chemical Catalysis Program, Professors Thomas Spiro, David Baker, Daniel R. Gamelin, and Christine K. Luscombe, all of the University of Washington (UW), will perform research on a project that may point the way toward efficient and inexpensive photocatalysts, which could play a significant role in the development of a solar-hydrogen economy. This work is motivated by the urgency of developing new and clean sources of energy. The capture and conversion of sunlight can meet a major fraction of future human energy needs; however, energy storage and conversion to transportation fuels pose major challenges. Broader impacts include communicating with the public through participation in the University of Washington's CLEAN ENERGY initiative, which maintains a website and runs conferences and workshops for energy stakeholders. The PIs will also participate in Seattle's Pacific Science Center outreach activities. In addition, the general public will be recruited to contribute to the protein design aspects of the project through the popular fold.it (http://fold.it) on-line computer game developed by the Baker group and collaborators. They will also participate in the successful high-school outreach program run by the NSF Center for Enabling New Technologies through Catalysis, housed at UW, which also provides us a mechanism for recruiting minority undergraduates to their laboratories.

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. Primary support for the subproject and the subproject's principal investigator may have been provided by other sources, including other NIH sources. The Total Cost listed for the subproject likely represents the estimated amount of Center infrastructure utilized by the subproject, not direct funding provided by the NCRR grant to the subproject or subproject staff. The de novo design of protein-protein interfaces is a stringent test of our understanding of the principles underlying protein-protein interactions and would enable unique approaches to biological and medical challenges. Here we describe a motif-based method to computationally design protein-protein complexes with native-like interface composition and interaction density. Using this method we designed a pair of proteins, Prb and Pdar, that heterodimerize with a Kd of 130 nM, 1000-fold tighter than any previously designed de novo protein-protein complex. Directed evolution identified two point mutations that improve affinity to 180 pM. Crystal structures of an affinity-matured complex reveal binding is entirely through the designed interface residues. Surprisingly, in the in vitro evolved complex one of the partners is rotated 180 degrees relative to the original design model, yet still maintains the central computationally designed hotspot interaction and preserves the character of many peripheral interactions. This work demonstrates that high-affinity protein interfaces can be created by designing complementary interaction surfaces on two noninteracting partners and underscores remaining challenges.

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. Primary support for the subproject and the subproject's principal investigator may have been provided by other sources, including other NIH sources. The Total Cost listed for the subproject likely represents the estimated amount of Center infrastructure utilized by the subproject, not direct funding provided by the NCRR grant to the subproject or subproject staff. Symmetric protein dimers, trimers, and higher-order cyclic oligomers play key roles in many biological processes. However, structural studies of oligomeric systems by solution NMR can be difficult due to slow tumbling of the system and the difficulty in identifying NOE interactions across protein interfaces. Here, we present an automated method (RosettaOligomers) for determining the solution structures of oligomeric systems using only chemical shifts, sparse NOEs, and domain orientation restraints from residual dipolar couplings (RDCs) without a need for a previously determined structure of the monomeric subunit. The method integrates previously developed Rosetta protocols for solving the structures of monomeric proteins using sparse NMR data and for predicting the structures of both nonintertwined and intertwined symmetric oligomers. We illustrated the performance of the method using a benchmark set of nine protein dimers, one trimer, and one tetramer with available experimental data and various interface topologies. The final converged structures are found to be in good agreement with both experimental data and previously published high-resolution structures. The new approach is more readily applicable to large oligomeric systems than conventional structure-determination protocols, which often require a large number of NOEs, and will likely become increasingly relevant as more high-molecular weight systems are studied by NMR.

DESCRIPTION (provided by applicant): Abnormal glutamate signaling within corticostriatal pathways has been linked to craving in humans and cocaine seeking in rats. Unfortunately, our limited understanding of glutamate has contributed to the lack of effective, well-tolerated treatments for many CNS diseases, including drug addiction. While glutamate is described as the primary excitatory neurotransmitter in the brain, it is unclear how the many components of this complex network of transporters and release mechanisms function in an integrated manner to regulate excitatory signaling. Due to a lack of available tools that selectively target these novel mechanisms, it has been difficult to convincingly demonstrate the importance of these novel mechanisms. One such component is system xc-, a source of nonvesicular glutamate release that is primarily expressed on astrocytes. It functions by exchanging extracellular cysteine for intracellular glutamate. System xc influences synaptic activity and plasticity through the release of glutamate and dopamine in multiple brain regions. Repeated cocaine produces a persistent reduction in system xc- activity, which appears to be necessary for glutamate-induced compulsive drug seeking. In contrast, manipulations that prevent or reverse cocaine-induced changes in system xc- activity normalize glutamate levels and blunt cocaine-induced reinstatement. In humans, N-acetylcysteine has shown promise in the treatment of drug addiction and related compulsive disorders. Studies such as these indicate that system xc- function may have profound implications in revealing the cellular basis of addiction, as well as the role of astrocytes in central nervous system activity - especially if it is determined that system xc- is the primary mechanism of action for N-acetylcysteine. Efforts to manipulate system xc- in rats typically involve the use of pharmacological tools that are associated with predictable pharmacological concerns. Increasing system xc activity by direct infusion of cystine into the brain or systemic administration of a cysteine prodrug (e.g., N acetylcysteine) are both effective since the rate of cysteine-glutamate exchange is a function of the relative extracellular/intracellular concentration gradients of its substrates. Mutations in the gene giving rise to xCT, the active subunit for system xc, is present in multiple mouse strains. However, essentially every study linking system xc to glutamate homeostasis or addiction has been conducted in rats or primates. The goal of this proposal is to use the novel Zinc Finger Nucleases (ZFN) approach to mutate the Slc7a11 gene encoding xCT in rat. After creating an xCT deficient rat model (aim 1), we will verify and characterize the general phenotype (aim 2) as well as addiction-specific phenotypes (aim 3). The development and application of these technologies to generate transgenic rat strains may result in a major paradigm shift in studying the neural basis of addiction by enabling more sophisticated and highly specific manipulations in a species that better models critical aspects of human addiction.

In this project funded by the Designing Materials to Revolutionize and Engineer our Future program of the Chemistry Division, Professors David Baker of the University of Washington and Todd Yeates of the University of California, Los Angeles will develop robust methods for the design of complex protein-based nanomaterials. The research project will integrate theory, computation, and experiment to describe the possible space of symmetric protein assemblies and develop methods for the rapid and reliable production of novel materials. An atlas of theoretical symmetric architectures will be delineated and integrated with cutting-edge protein structure modeling software to identify novel amino acid sequences predicted to self-assemble into precisely defined structures. The corresponding proteins will be produced experimentally and their structures determined at high resolution to provide feedback for the improvement of both the theoretical and computational aspects of the program. In this way, a general approach for patterning complex protein-based materials with sub-nanometer resolution will be developed that is expected to have a profound impact on the fields of molecular self-assembly and nanomaterials.

Biological systems use proteins to carry out complex tasks at the molecular level. Proteins are "oligomers": they are composed of multiple chemical subunits linked together in a long chain. The chain must fold into a specific, complex shape in order for the protein to become functional. Additionally, protein units can interact to form larger structures that perform an array of chemical and mechanical functions. This research project involves a joint experimental and computational effort to design proteins that can fold and assemble into preordained structures. Long-term outcomes of this basic research could include the development of new types of medicines, materials with unprecedented properties, and other useful chemical technologies.

This project, funded by the Systems and Synthetic Biology Program in MCB and the Biotechnology, Biochemical and Biomass Engineering Program in CBET, is part of a larger ERASynBio funded collaborative. The team of investigators will develop the rules that govern the design of complex nanometer scale structures made from nucleic acids and polypeptides, and use those rules to create new biological molecules that have never been seen before in nature. They will also develop the tools that will enable the assembly of these new materials that could potentially be used to control many aspects of cell function, or create new materials that could be used as sensors or in biomanufacturing.

Technical: Biological organisms are capable of producing chemicals, materials and molecular machines that far exceed our engineering capabilities. Underlying these abilities are the unique properties of proteins, exquisitely evolved for function, allowing precise positioning of atoms and chemistries. Designing novel proteins is difficult because of our still incomplete understanding of how proteins fold for a given primary amino-acid sequence. In this project, researchers will apply principles of synthetic biology to define and modularize building blocks that can be combined in rational ways to enable control of 3D positioning in designed macromolecular structure. Members of the consortium have advanced design and engineering principles for polypeptide- and DNA-based nanostructures and developed next-generation gene synthesis to facilitate high-throughput approaches. The team will build on these foundations to engineer bio-macromolecular assemblies with shapes and functions of unprecedented complexity. They will deliver an expanded toolbox of polypeptide building elements; rules, design principles and methods for constructing complex bionanostructures; and routes to nucleic acid/ polypeptide-hybrid platforms for the community of synthetic biology. The project will expand the limits of the designed polypeptide and nucleic acid/protein hybrid providing a platform to facilitate their use in a wide range of biomanufacturing applications.

DESCRIPTION (provided by applicant): An insidious aspect of addiction is that afflicted individuals are at risk of relapse even after extended periods of abstinence. Stressful life events are important contributors to relapse in recovering cocaine addicts, but the mechanisms by which they influence motivational systems are poorly understood. Studies suggest that stress may set the stage for relapse by increasing the sensitivity of brain reward circuits to drug-associated stimuli. This proposal seeks to elucidate the mechanisms by which stress, through increases in glucocorticoid hormones, influences relapse vulnerability. We have previously shown that treatment of rodents with stress levels of glucocorticoids does not lead to reinstatement of drug-seeking behavior, but potentiates reinstatement in response to a dose of cocaine that, by itself, is not sufficient to trigger relapse. In parallel to its behavioral effect, corticosterone pretreatment also potentiates the effects of low-dose cocaine on extracellular dopamine concentration in the nucleus accumbens, suggesting that glucocorticoids may potentiate drug seeking by enhancing dopaminergic neurotransmission in this critical reward-processing brain region. We are examining the role of organic cation transporter 3, a high-capacity dopamine transporter that is acutely and directly inhibited by glucocorticoids, in mediating the effects of glucocorticoids on dopaminergic neurotransmission, cocaine relapse, and motivated behavior in rodents. Because of a lack of pharmacologically specific inhibitors for OCT3, we are using two different genetic approaches to test the hypothesis that corticosterone potentiates cocaine-induced dopaminergic neurotransmission and drug-seeking behavior by inhibiting OCT3-mediated clearance of dopamine in the nucleus accumbens. In the first aim, we will determine the impact of corticosterone-induced inhibition of dopamine clearance in the nucleus accumbens on dopamine signaling and drug relapse by using in vivo microdialysis and fast-scan cyclic voltammetry to measure dopamine concentration and clearance in cocaine-seeking animals. In the second aim, we will determine the role of OCT3 in the behavioral and neurochemical effects of corticosterone by examining corticosterone effects on drug-seeking behavior and nucleus accumbens dopamine signaling in animals genetically modified to lack OCT3 expression either globally or specifically in the nucleus accumbens. In the third aim, we will test the hypothesis that corticosterone-induced decreases in dopamine clearance modulate reward sensitivity and natural reward processing. These findings will thoroughly characterize a novel mechanism by which stress hormones can rapidly regulate dopamine signaling and contribute to the impact of stress on drug intake and motivated behavior in general.

Rosetta is the best-overall-performance molecular modeling and design software toolkit, and is best-in-class both for predicting new protein structures and for designing new proteins for industrial and medical applications. Rosetta is currently licensed to enterprise and has gained some traction, but the top 3 identified pain points are: Difficult to use, offers almost no user support and requires immense amounts of computational resources. This project aims to build an enterprise version of the Rosetta molecular modeling and design software package, and will accelerate the development of small-molecule and biological drugs in Pharma and Biotech firms, and allow for the creation of entirely new biological drugs, diagnostics, and biological tools at those firms through computational protein design.

This proposal will solve these problems with a clean 3D Graphical User Interface (GUI), workflow pipelining, and a cloud compute engine. The proposed software, Cyrus, is a wrapper around Rosetta. It provides a GUI front-end, automation of standard user procedures and compute on the cloud. The GUI is all-new, and will be built inside a web browser for maximum compatibility and ease of interface with the cloud. Automation and standardization of procedures consists of benchmarking and scripting of existing protocols to remove human interaction steps, and incorporate expert know-how. The cloud Implementation of Cyrus will be built from scratch using cutting edge open source tools such as Redis and OpenStack to provide power (1000?s of CPUs) and ease of use (graphic interface with cloud resources for the end user).

With this award, the Chemistry of Life Processes Program in the Chemistry Division is funding Dr. David Baker for a project to engage the Foldit community in a collaborative citizen science effort to address the ongoing Ebola outbreak in West Africa. Foldit is a game-based protein design platform that engages gamers drawn from the community at large. This project involves collaboration among the Foldit game-based scientific community, The University of Washington's Institute for Protein Design (IPD), and The University of Washington's Center for Game Science, to improve and tailor the Foldit design platform in order to arrive at proteins that bind and neutralize the Ebola glycoprotein. The project will collectively harness the expertise of computer scientists and protein design experts in collaboration with the constantly growing citizen scientist community to arrive at totally novel functional proteins in an efficient and rapid manner. The proposed research will deliver critical insight into the structure and binding modes of potential anti-Ebola therapeutics; development of these anti-Ebola binders will be extremely valuable both for academic and industrial communities. The work will also significantly expand and broaden the citizen scientist community focused on discovering anti-viral therapeutics for Ebola. In addition to advancing development of anti-Ebola therapeutics, this project will continue to educate the public at large about Ebola, its epidemiology and current research efforts underway to combat this virus.

This project sets out to improve the Foldit game-based scientific environment in the direction of rapid and direct collaboration among the Foldit community, The University of Washington's Institute for Protein Design (IPD) and The University of Washington's Center for Game Science. The project will use Foldit puzzle challenges to facilitate the design of novel anti-viral proteins that bind to the surface glycoproteins of Ebolaviruses, and will experimentally test these binders for in vitro efficacy in collaborator laboratories. The approaches involve (a) customizing the Foldit protein design platform for Ebola studies, (b) engaging the Foldit citizen scientist community with anti-Ebola directed protein design challenges, (c) refining the anti-Ebola Foldit player designs, (d) preparing synthetic genes and manufacturing anti-Ebola proteins with the designed proteins and (e) testing the optimized anti-Ebola proteins prepared, in cell-based studies and animal models. The project will enlist the knowledge of a broad cadre of experts in the design of novel protein binders targeting Ebola and will utilize the Foldit platform to continually educate the public about project progress and more broadly about the virus itself with regular updates and feedback on the work.

P2 - ABSTRACT The misfolding and aggregation cascade of the A? polypeptide is widely accepted to be the root cause of Alzheimer's disease (AD). However, the development of agents that are able to alter the misfolding and aggregation cascade has proven difficult. Small molecules are generally unable to present sufficiently large interaction surfaces to compete with the protein-protein interactions that drive aggregation, while antibodies are difficult to direct against specific conformational or oligomeric states. An alternative is the rational, computer- aided design of proteins and peptides that are able to bind specifically to particular species in the A? misfolding and aggregation cascade, and which can alter this process in well-defined ways. We propose to develop agents able to alter A? misfolding and aggregation in three ways, each representing a different possible strategy for mitigating A? neurotoxicity. First, we will create small proteins that are able to bind specifically to monomeric A? that has not yet entered the aggregation cascade, sequestering it or directing it for clearance or degradation. Second, we will create peptides able to cap growing A? amyloid fibrils to block fibril elongation, shifting the A? population to soluble species that can be cleared by other mechanisms. And third, we will create fibril-coating proteins that are able to stabilize pre-formed A? amyloid fibrils and prevent the dissociation that can produce toxic, soluble, oligomeric species. The peptides and proteins produced will be useful experimental agents for testing different AD treatment strategies, based on several different hypotheses of A? toxicity, in model systems.

Proteins are large molecules that have complex three-dimensional (3D) structures that give them sophisticated cellular functions. To provide access to complex protein structures not found in nature, Professors Baker and King of the University of Washington and Professor Yeates of University of California Los Angeles collaborate to compute and precisely control the assembly of proteins into 3D architectures. The research team shares these capabilities with the engineering and scientific communities and constructs a Designed Protein Nanomaterial (DPN) database to enable others to rapidly build on these advances. The long-term goal is to allow predictable design of protein-based materials for use in a wide range of applications such as targeted drug delivery, functional materials for energy conversion, and other chemical technologies. The interdisciplinary project provides excellent research training opportunities to postdoctoral researchers and students.

This research integrates theory, computation and experiment to close the design cycle on creation of new protein-based nanomaterials that are not currently accessible using conventional protein engineering approaches. It develops strategies to prepare more robust and complex protein nanomaterials from modular building blocks, whose 3D assembly and disassembly can occur in response to changes in environmental conditions. The research team recruits citizen scientists to solve complex protein design problems using Foldit, and to provide innovative input to enhance functionalization of such materials.

PROJECT ABSTRACT The Zika virus is a mosquito-borne pathogen that has been implicated in a number of grave complications to fetuses carried by infected women. More recently, Zika has been linked to a spike in the number of cases of Guillain-Barre syndrome and microcephaly, a condition in which babies have severely hindered brain development and are born with smaller heads. Given this possible link to these serious birth defects, there is a pressing need for a way to definitively diagnose whether a woman has been exposed to the virus. Current methods to test for the presence of Zika miss the narrow time window before the virus clears the body after infection and serological tests are often hindered by cross-reactivity to other flaviviruses endemic in the same regions as Zika. The studies described here propose the development of an immunoassay targeting only those antibodies in the host?s polyclonal response that bind to those epitopes in the Zika coat protein that are not shared by other Flaviviruses. Such an assay could lead to more clinically specific immunoassays for the viruses in question. First we will develop novel computationally designed hyperstable viral epitope mimetics (VEMs) of the Zika envelope protein with structures that can specifically bind human anti-Zika immunoglobulins. Second these VEMs will be implemented in a rapid, low-cost paper-based sandwich immunoassay capable of detecting anti-Zika antibodies in human plasma at clinically relevant concentrations.

PROJECT SUMMARY Emerging peptide-, protein-, and nucleic acid-based therapeutics for the treatment of Alzheimer?s disease and other neurologic conditions are blocked from diffusing into the brain by the blood?brain barrier (BBB). The long-term goal of this project is to deliver large therapeutic cargo such as these into the brain using designed BBB-crossing drug-delivery vehicles. The ?overall objectives are to (i) leverage recent breakthroughs in computational peptide design to yield new knowledge about BBB permeability and (ii) to designed from scratch new proteins that ferry cargo into the brain by exploiting natural systems that the brain uses to receive nutrients and signals. The ?central hypothesis is that the systematic design of functional biomolecules will yield new insights and tools for improving the delivery of large biomolecule therapeutics into the brain. The ?specific aims are: 1) to systematically and rationally discover the physiochemical properties which confer BBB permeability to designed peptide macrocycles (a promising new class of therapeutics); 2) to computationally design small, hyperstable proteins which bind to receptors that naturally cycle between the blood- and brain-side of the BBB; and 3) to fuse the binding proteins generated in Aim 2 to various drug-binding/packaging proteins, thereby creating protein assemblies that ferry large therapeutics into the brain. This project is ?innovative because it proposes to resolve a long-standing barrier to the treatment of neurologic diseases (namely, the difficulty of delivering therapeutics into the brain) by designing from scratch new BBB-crossing drug delivery vehicles. The project is ?significant because it is expected to provide tools which will improve outcomes in a range of future clinical trials of therapeutics which require delivery into the brain.

PROJECT SUMMARY Although immunostimulatory cytokines can be used to combat cancer, their poor stability and high off-target toxicity has limited their application in the clinic. The ?long-term goal of this project is to produce targeted anti-cancer cytokine mimetics with reduced toxicity. The ?overall objective is to apply recent breakthroughs in ?de novo protein design to yield a new category of targeted, non-toxic, immunostimulatory proteins. The ?central hypothesis is that the beneficial stimulatory effects of natural cytokines can be engineered into ?de novo designed proteins which do not engage in off-target binding, thereby circumventing the dose-limiting toxicities seen in clinical applications of natural/reengineered cytokines. The ?specific aims are to: (1) use ?de novo protein design to generate hyperstable, bioactive mimetics of interleukin-2, -4, -12, -15, and -21 which function by engaging with (i.e. binding to) interleukin receptors ?in vivo?; (2) to split these mimics into inactive parts which can regain activity by reassembling ?in vivo?; and (3) to fuse each of these inactive parts to specific targeting domains, thereby yielding conditionally-active cytokine mimics that stimulate T-cells by reassembling only on the surface of a targeted cells (i.e. cancer cells displaying two specific surface biomarkers). ?As proof of principle?, the first such ?de novo designed cytokine mimetic has been produced, split, and shown to reduce tumors in mice without accompanying toxicity or immunogenicity. This research proposal is ?innovative because it seeks to resolve a long-standing barrier to cancer immunotherapy (namely, the off-target toxicity of cytokine-based therapeutics) by designing from scratch a new class of non-toxic cytokine mimics. The proposal is ?significant because it would be the first example of computational protein design yielding targeted, biosuperior cancer therapeutics. Ultimately, such molecules have the potential to treat a wide range of cancers, including malignant melanoma, renal cell carcinoma, and more.

The Protein Data Bank (PDB) is the single global archive of three-dimensional (3D) structures of large biological molecules. Despite a steady increase in its holdings, the growth of the PDB is far outstripped by the growth in the available protein sequence data. Resources like Genome3D (genome3d.eu), funded by the BBSRC, aim to fill the gap in structure coverage of the protein sequence space with reliable predictions of structures. These approaches largely model proteins that are closely related to a protein of known structure. The Rosetta method for predicting protein structures, a world-leading approach developed by the Baker lab in the USA, was recently enhanced with information derived from evolutionary analyses of protein sequence data, yielding reliable models even for cases where sequence identity between the model and the available experimental structures is very low. This project will integrate Rosetta models into Genome3D to expand the coverage of structural data for important organisms for health and food security. It will also enrich both the experimentally determined and computationally predicted structures with valuable functional annotations, such as information pertaining to surface interfaces, a key ingredient in understanding how proteins interact with each other and with other biological molecules. By focusing on proteins dissimilar to those with known structures, this portal will help fill the gaps in structure coverage of the protein sequence space and will make structure data much more readily available and accessible. Finally, novel visualization tools integrating the presentation of the predicted and experimentally determined structures will be developed, maintaining a clear distinction between what is predicted and what is experimentally determined.

The expanded set of 3D models derived from this project will in turn help to expand the coverage of sequence space even further, since these models can be used to guide the experimental determination of protein structures being obtained by powerful new structural biology techniques like cryo-Electron Microscopy (EM). This project will also endeavor, where possible, to improve the assembly of individual protein structures into macromolecular complexes which can be analyzed to determine their biological role. Scientists in both academia and industrial sectors will benefit from access to such an integrated portal, assisting them in designing new medicines, understanding the mechanism of disease, or in designing proteins with novel properties. Recent advances in Electron Microscopy allows near routine determination of structures of large molecular machines and is in need of a large repertoire of "building blocks" in interpreting the experimental results, a need which will be partially addressed by the new portal and its provision of expanded domain structure libraries. The portal will also have ways to access the assembled data programmatically, benefiting power users: software developers and maintainers of other resources.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

Abstract Project 4 will test the hypothesis that computationally designed protein ?minibinders? and logic-gated switches targeting cerebrospinal fluid (CSF) protein biomarkers and protein particles can serve as versatile capture and detection agents to vastly improve the molecular characterization of CSF samples. The availability of these reagents should support the overall U19 goal of improving our understanding of CSF biomarkers as direct measures of age-related cognitive decline and Alzheimer's Disease (AD) pathophysiology. We have integrated our work plan within the highly focused U19 Project: Next Generation Translational Proteomics for Alzheimer's and Related Dementias to test the above hypothesis. In collaboration with Projects 1?3, our main goal is to develop an optimized set of computationally designed minibinders (hyperstable binding proteins of length <65 aa) as CSF protein biomarker capture agents, and ultra-specific logic-gated reagents for detection of CSF particles that have two defined protein components. The research will iterate between computational design and experimental testing, with feedback at each stage from CSF assay experiments conducted in collaboration with Projects 1-3 which will guide improvement of the minibinder design methods for capture of specified CSF biomarker protein targets, and for ultra-specific logic- gated detection of CSF particles with two composite protein components. The outcomes will be (i) specific CSF biomarker capture and detection systems for AD and other age-related neurodegenerative disorders, and (ii) an integrated computational-experimental pipeline for rapid on demand engineering of new protein based diagnostic agents for neurodegenerative disorders in general. Therefore, Project 4 relies on a close collaboration with Projects 1-3 and Cores 1-4 within the highly interactive U19 program.

Project Summary/Abstract Dysregulation of glutamate signaling is a core component of the pathological basis of drug addiction involving cocaine and many other substances. Recent progress in neuroscience and other fields clearly establishes that the molecular and cellular basis of glutamate-encoded signaling is vastly more complex than previously recognized. In particular, it is becoming increasingly evident that astrocytes, which are among the most abundant cells in the human brain, release glutamate to produce complex regulation over neural circuit activity. This novel form of glutamate-encoded intercellular signaling may be critical to understanding the pathological glutamate produced by long-term drug use since astrocytic glutamate release mechanisms, such as system xc- (Sxc), are altered by cocaine. While these discoveries raise numerous questions that may be essential in understanding glutamate signaling in the human brain, this proposal will focus on the question, how do neurons regulate glutamate release from astrocytes to control neural network activity and behaviors relevant to cocaine addiction. We will test the hypothesis that this is achieved by the actions of the neuropeptide pituitary adenylate cyclase-activating polypeptide (PACAP), which we believe to be an unrecognized component of glutamate signaling in the nucleus accumbens (NAc). In support, we have found that A) PACAP mRNA is expressed in NAc afferents, B) PACAP receptors are expressed in NAc astrocytes and neurons projecting to the substantia nigra (SN) but not the ventral pallidum (VP), C) PACAP stimulates glutamate release from astrocytes by increasing the activity of system xc- (Sxc), D) PACAP application depresses eEPSCs in NAc medium spiny neurons (MSNs) projecting to the substantia nigra (SN) but not the ventral pallidum, and E) intra-NAc micro-infusion of PACAP blocks cocaine reinstatement. In this proposal, we will test the hypothesis that PACAP is a neuropeptide that regulates glutamate release from astrocytes and glutamate receptors on neurons to provide a novel form of pathway-specific regulation of NAc efferent pathways. In Aim 1, we will examine the molecular basis of PACAP-induced increases in Sxc activity and determine whether Sxc regulation is necessary for PACAP to depress eEPSCs in NAc-SN MSNs and block cocaine reinstatement. In this aim, we will also examine if PACAP increases glutamate from astrocytic release mechanisms other than Sxc. In Aim 2, we will examine the possibility that the form of astrocyte-neuron signaling triggered by PACAP require the regulation of neuronal glutamate receptors to produce the relevant changes in physiology that gate the output of NAc efferents and behavior. We will also explore whether PACAP alters presynaptic glutamate release. In Aim 3, we will investigate the role of endogenous PACAP to learn whether this neuropeptide is an unrecognized component of glutamate signaling in the NAc, and whether it is a key determinant of drug- seeking behavior.

PROJECT SUMMARY One of the most pressing public health priorities for the COVID-19 pandemic is the development of an effective and inexpensive therapeutic. The long-term goal of this proposal is to develop such COVID-19 treatments, as well as the methods needed to rapidly create such molecules as soon as any new pathogen is identified. The central hypothesis is that computational design can be used to quickly create proteins with potent antiviral activity and others that suppress ?cytokine storms? associated with advanced infection. Such countermeasures, if rapidly developed and deployed, could save millions of lives during an outbreak until vaccines are developed. The specific aims are to: 1) overcome current limitations in the discovery and development of protein therapeutics by creating methods for the de novo design of hyper-stable miniproteins that bind tightly to vulnerable binding sites on the SARS-CoV-2 Spike glycoprotein, including the receptor binding domain (RBD) of the ACE-2 cellular receptor and the fusion peptide region; 2) Enhance the avidity of such anti-Spike minibinders through genetic fusion of multiple copies, or through rational design of higher-order oligomers to create drug compounds that are less prone to viral mutagenic escape; 3) Apply the same minibinder design pipeline to create cytokine receptor antagonists of key cytokines IL-6 and IL-1? likely involved in acute respiratory distress syndrome (ADRS) associated with COVID-19 mortality; 4) Assess the efficacy of antiviral and anti-interleukin minibinders by several routes of delivery (intravenous, intranasal and subcutaneous) in rodent models of COVID-19 and assess immunogenicity in order to identify those designs best suited for further preclinical development. As proof of principle, the first anti-Spike minibinders have already been designed, were found to bind to SARS-CoV-2 Spike RBD, and were found to neutralize live virus with activities rivaling the most potent known antibodies. This proposal is innovative because it seeks to apply powerful emerging methods in the computational design of new protein therapeutics to the COVID-19 pandemic. The proposal is significant because it would be the first example of computational protein design yielding potent and entirely de novo antiviral and anti-inflammatory therapeutics for an active pandemic. Ultimately, rapid minibinder design methods have the potential to generate treatments for future pandemics, as well as for many other common and neglected diseases and conditions.

In this Molecular Foundation for Biotechnology (MFB) project, Professors David Baker and Frank DiMaio of the Department of Biochemistry at the University of Washington, and Barry Stoddard of Basic Sciences at the Fred Hutchinson Cancer Center together are developing new ways to model and design protein-DNA complexes using deep-learning (DL) methods. To do this, they will develop three DL-based models: (1) a model for prediction of protein-DNA complex structures from sequence, (2) a model for sequence design of protein-DNA complexes, and (3) a model for quality assessment of protein-DNA complex structure predictions and designs. The DL models developed in this project will be leveraged in a pipeline for design of sequence-specific DNA binding miniproteins capable of targeting specified sequences of dsDNA. The use of the novel DL-based models in this context will be useful for validating model accuracy and will have broad impact as a powerful tool for designing protein-DNA interfaces for biotechnology applications, such as the design of novel transcription factors, nucleic acid modifying enzymes, and gene correction reagents. This project lies at the interface of DL research, computational protein design, biochemistry, and structural biology and will provide multi-disciplinary training for undergraduates, graduate students, and postdocs involved in the project. The primary goals of their outreach and education programs are to attract young people to careers in STEM (science, technology, engineering and mathematics) and improve training in biochemistry and computational protein design. The outreach plan involves a multi-pronged effort focused on engaging undergraduates through individually mentored summer research and a cohort-based undergraduate research program that will run during the academic year. Both efforts will be focused on training undergraduates in contemporary methods in computational protein design and experimental methods for validating protein function, including the novel methods developed in this proposal. <br/><br/>This project seeks to develop a suite of machine learning/deep learning (ML/DL) techniques for modeling protein-DNA complexes. New tools capable of inferring protein-DNA complex structures, predicting the nucleotide specificity of DNA-binding proteins (DBPs), and evaluating accuracy of protein-DNA complex models would be invaluable in solving salient technological problems, such as developing novel transcription factors. Current approaches lack accuracy or are computationally intensive, primarily due to the difficulties in modeling indirect readout of DNA conformational flexibility, hydrogen bonding and electrostatic interactions, metal ion cofactors, and the highly solvated interfaces of protein-DNA complexes. The specific goals are to develop DL-based methods for (1) Inference of structure models of DNA and protein-DNA complexes from sequences and sequence alignments, based on the recently developed RoseTTAFold model, an ML framework for predicting protein structures; (2) A sequence prediction neural network for designing sequence specific DBPs and predicting their specificity given protein-DNA complex backbone information, and (3) An accuracy prediction model for evaluating structural models of protein-DNA complexes. The three DL methods developed in this project will be leveraged in the design of DBPs. Designed DBPs will be experimentally validated in a high-throughput pooled format using yeast display, cell sorting, and next-generation sequencing methods to approximate the binding affinity of pooled designs. Designs showing DNA binding activity in yeast display experiments will be further characterized for DNA binding affinity and specificity using in vitro biochemistry techniques and the design models will be confirmed with X-ray co-crystallization. Application of the ML models in this design context will provide validation of model accuracy and result in a powerful tool for designing protein-DNA interfaces for biotechnology applications, such as the design of novel transcription factors, nucleic acid modifying enzymes, and gene correction reagents. <br/><br/>This project is jointly supported by the Division of Chemistry (CHE), the Division of Information and Intelligent Systems (IIS), and the Division of Physics (PHY).<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.