Lawrence S. Shapiro, Phd - US grants

DESCRIPTION (provided by applicant): Cadherins are among the most important and widely distributed cell adhesion proteins, and their selective binding helps to define physical connections between the cells of vertebrates and invertebrates. Cadherin selectivity provides a critical mechanism to shape developing tissues. In earlier work, we defined the atomic-level molecular mechanisms of vertebrate classical cadherin adhesive interaction and junction formation. Here we propose specific aims in three main areas to build on this work: (1) we will quantitatively determine the affinities of interactions among the classical cadherins, and define cellular correlates of these molecular properties. (2) We will investigate structural and mechanistic aspects of adherens junctions, and (3) we will determine structures and binding mechanisms for Drosophila classical cadherins in order to develop a facile genetic system with which the molecular adhesive properties of cadherins can be tested for their effects on tissue development.

Studies making use of isolated proteins, ranging from production of antibodies to biochemical characterization of enzymes and animal injection studies of secreted signaling factors, are critical to understand their biological roles. To facilitate such studies, we plan to implement a Protein Expression Core Facility (the PE core). The PE core will produce and characterize purified proteins for members of the DERC. Specifically, the core will provide the following services: (1) Bacterial protein expression of recombinant proteins in E. coli. In general, only proteins with simple architecture will be produced in this system. (2) Mammalian expression using a cellsorting method to quickly enrich the fraction of highly expressing cells. (3) Purification using standard affinity, ion exchange, and gel filtration chromatography techniques. (4) Characterization of purity, appropriate folding, and oligomerization state, using standard biochemical methods and mass spectrometry.

ABSTRACT NOT PROVIDED

The Proteomics Shared Resource is a new Cancer Center SR that evolved from an established facility at CUMC with strong expertise in all aspects of sample preparation and mass spectrometric analysis that will be increasingly applied to cancer related research projects. The goal is to provide an essential battery of mass spectrometry-based proteomics tools to HICCC investigators. Specific services are to use state-ofthe- art experimental strategies for: ¿ Identification of proteins by in-gel digestion and mass spectrometry ¿ Identification of protein components within protein complexes ¿ Identification of phosphorylation sites and other covalent protein modifications Instrumentation in this facility includes two high-resolution quadrupole-TOF electrospray mass spectrometers equipped with nanoflow LC for LC-MS, and a MALDI-TOF mass spectrometer. A key activity will be education of researchers in the ways state-of-the-art proteomics tools can advance their research. The HICCC has made recent investments in the facility infrastructure, and has been publicizing the facility to HICCC members, leading to an increased focus on mass spectrometry for cancer research. However, currently the facility has minimal financial support for daily operating expenses from the HICCC and has raised funds for its continued operation by offering cost-effective mass spectrometry services to outside investigators. In this grant renewal we are proposing increased direct support for this Shared Resource. The facility will continue to have close intervhactions with investigators, advising them on experimental design, sample preparation, and data interpretation. Future goals include the implementation of quantitative proteomics techniques based on labeling methods coupled with LC-MS, and the expansion of the Resource to include a two-dimensional gel electrophoresis service. Gel-based proteomics will facilitate investigation of protein mixtures of greater complexity such as those found in tissues and biological fluids, and opens the possibility of finding novel diagnostic markers and therapeutic targets. During the last period of the CCSG (transitioning to Cancer Center management) 34% of the Columbia investigators using the facility were Center members with peer-reviewed funding, with those members representing over 30% of the utilization of the services. The proposed total operating budget of the facility is $380,006, of which we are requesting $84,858 from CCSG.

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. Our work focuses mainly on cadherin cell adhesion molecules. About 100 cadherins are found in the human genome. They are expressed on the surfaces of all cells of solid tissues, and binding between them holds cells together in multicellular structures such as tissue layers. The differential binding of cadherins plays an important role in the development of the multicellular structures of animals, yet their selectivity is little understood.

DESCRIPTION (provided by applicant): AMP-activated protein kinase (AMPK) coordinates metabolism with energy availability in eukaryotes by responding to changes in intracellular ATP and AMP levels. The kinase activity of AMPK is stimulated by AMP and inhibited by excess ATP, and it is thought that this unique regulatory behavior enables AMPK to act as a central cellular "fuel gauge". AMPK is thus the subject of intense interest as a target for therapeutics to treat metabolic disorders such as diabetes and obesity. AMPK is an abg heterotrimer that includes a subunit with serine/threonine kinase activity and an adenylate-binding regulatory region composed of elements from all three subunits. In preliminary data for this application we present crystal structures for AMP- and ATP-bound forms of the heterotrimeric adenylate sensor from the Schizosacharomyces pombe enzyme. This complex lacks the kinase catalytic domain, but reveals the conserved trimeric core architecture of AMPKs. ATP and AMP bind competitively to a single site within the g subunit, helping to explain their competing effects. Biophysical experiments show that the adenylate sensor complex binds the a subunit kinase domain in the presence of AMP but ATP binding prevents this association. These data help to provide an initial molecular understanding of AMPK regulation. A crystal structure of an AMPK-ADP complex, surprisingly, reveals a second binding site that can uniquely accommodate ADP. The overarching goal of this application is to gain an atomic-level understanding of AMPK regulation through the following specific aims: (1) characterizes the affinities of binding of various adenylate ligands, and use biophysical methods to determine how ligand binding affects interaction between the regulatory and kinase domains. Results from these studies will be correlated with kinase activity in various ligand-bound states. (2) To gain an understanding of the holoenzyme architecture, we will use site-directed mutagenesis to define the molecular regions responsible for nucleotide-dependent association between the kinase domain and regulatory adenylate sensor. Results from the proposed work will be critical for the rational development of AMPK-directed therapeutics. AMPK, a central regulator of cellular metabolism, is among the most attractive molecular targets for new therapeutics to treat diabetes, obesity, and other metabolic disorders. Prior studies have shown that activators of AMPK administered to diabetic animals can substantially ameliorate the physiological effects of diabetes. Despite the great promise of AMPK-directed therapeutics, little is known about the molecular mechanisms of regulation, and the design of appropriate small molecule drugs has been impeded by the lack of atomic-level information on the architecture of the enzyme. Our preliminary results and the further work proposed will provide high-resolution structural information on AMPK, and should directly enable the rational design of AMPK-directed therapeutics. PUBLIC HEALTH RELEVANCE: AMPK, a central regulator of cellular metabolism, is among the most attractive molecular targets for new therapeutics to treat diabetes, obesity, and other metabolic disorders. Prior studies have shown that activators of AMPK administered to diabetic animals can substantially ameliorate the physiological effects of diabetes. Despite the great promise of AMPK-directed therapeutics, little is known about the molecular mechanisms of regulation, and the design of appropriate small molecule drugs has been impeded by the lack of atomic-level information on the architecture of the enzyme. Our preliminary results and the further work proposed will provide high-resolution structural information on AMPK, and should directly enable the rational design of AMPK-directed therapeutics.

The overall objective of this proposal is to accelerate the acquisition of structural information about membrane proteins by applying a structural genomics approach informed by the collective experience of a team of expert investigators. We have established the New York Consortium on Membrane Protein Structure ��NYCOMPS��to work together toward this objective. NYCOMPS participates as a Specialized Center in Phase 2 of the Protein Structure Initiative ��PSI‐2��now. As constituted for PSI‐Biology, NYCOMPS will comprise 12 Principal Investigators at six institutions. Our pipeline for structure determination will select targets through a bioinformatics analysis of all known sequences, move on to recombinant DNA cloning, protein expression in bacteria or eukaryotic cells, and protein purification at moderately high throughput, and then continue on to determine structures by x‐ray crystallography. Our Protein Production Facility at the New York Structural Biology Center ��NYSBC��handles targets through purification at a mid‐scale level;and successful candidates are distributed to participant laboratories for scale‐up and crystallization. Functional analysis of structures will be perfomed both by computations and through routine experimental biochemistry. Targets will be identified through nominations from the biological community, including adjunct NYCOMPS members, and from NYCOMPS biological themes, which concern elucidation of the membrane protein universe and structural studies on energy homeostasis and metabolic disorders. A program in technology development will aim to improve pipeline efficiency and quality of results. The project will be managed to optimize output and to integrate effectively with the PSI‐Biology network and with other membrane protein structure efforts.

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 are working to structurally characterize the adhesive complexes of cell surface adhesion receptors, which help to assemble cells into the solid tissues of the body. Some adhesion proteins help to define the pattern of connectivity between neurons, and this help "wire" the nervous system.

DESCRIPTION (provided by applicant): Macromolecular complexes are of key importance in virtually all life processes. However, more powerful methods need to be developed to (1) validate potential interacting partners, (2) efficiently screen for complex formation, and (3) produce sufficient amounts of functional protein complexes for in depth functional and structural studies. To address these needs we will (1) develop new high-throughput fluorescence-based screens to assess protein-protein interactions; (2) develop rapid screening method to distinguish between transient and stably interacting partners. (3) Finally, since functional characterization and structure determination of protein complexes depends on the capability to produce them recombinantly in high yield, we will develop a multi-protein expression system, based on distinct fluorescent markers whose expression levels are correlated with each protein component. Together, these methods provide a high- throughput pipeline for validating, characterizing, and producing protein complexes for structural and functional studies. This platform, and the associated reagents developed under this proposal, will provide a much needed toolbox that will greatly aid any scientist studying eukaryotic protein complexes.

DESCRIPTION (provided by applicant): The overall objective of the proposed research is to determine the nature of trans (cell-to cell) and cis (same cell surface) interactions of protocadherin (Pcdh) cell surface proteins. The majority of mammalian Pcdhs are present in three large gene clusters, and the organization of these clusters leads to the generation of enormous single cell diversity. This diversity is thought to function as a molecular barcode for individual neurons, which allows cells to distinguish between self and non-self. In particular, the clustered Pcdhs have been shown to be required for dendritic self-avoidance and other aspects of neural circuit assembly in the mouse brain and spinal cord. This function is likely to require homophilic interactions between distinct Pcdh isoforms at the surface of opposing plasma membranes. Initially, a cell-cell aggregation assay will be used to carry out a comprehensive analysis of Pcdh isoform homophilic specificity. A selected subset of Pcdh Alpha, Beta and guama isoforms tagged with fluorescent labels will be cloned and transfected into mammalian cells in culture. Cells transfected with the same or different Pcdh isoforms will be mixed and cell aggregation quantitated by the size of the immunofluorescent cell aggregates by fluorescent microscopy. The trans-homophilic interaction domains will be identified by constructing and testing Pcdh isoforms missing specific extracellular domains. Once identified the interacting ectodomains will be subjected to biophysical analyses to determine the multimeric state and homophilic binding affinities. Characterization of cis Pcdh interactions will be accomplished through the transfection of cells with multiple Pcdh isoforms, followed by cell-cell aggregation assays. Domain deletion studies will be carried out to map the cis- interacting regions and these regions subjected to biophysical analyses. Once identified, trans dimeric homophilic domains will be expressed at high levels, and the dimeric complexes crystallized and subjected to three-dimensional atomic structure analyses. Structural hypothesis based on the three dimensional structure will be tested using site-directed mutagenesis and cell aggregation assays as a means a correlating structure and function. Finally, attempts will be made to produce and crystallize full-length Pcdhs and sub regions that encompass the cis interface and their structures determined by x-ray crystallography. If successful, the proposed studies will provide deep mechanistic insights into the role of the clustered Pcdhs in mediating self-avoidance and other aspects of neural circuit assembly. In addition, the studies may reveal the existence and nature of a Pcdh-mediated molecular recognition code.

Project Summary Desmosomes are intercellular junctions that impart strength to tissues by connecting the intermediate filament networks of adherent cells. Members of two cadherin subfamilies, the desmogleins and desmocollins, mediate desmosomal adhesion in the extracellular space. While electron microscopy analyses reveal a structure with apparently high order, the arrangement and interactions of desmocollins and desmogleins that underlie the extracellular architecture of desmosomes remains unknown. This lack of knowledge has persisted in part due to the inability to produce functional desmosomal cadherin ectodomains in recombinant expression systems. Here we present preliminary data in which we present a mammalian expression system in which all seven human desmosomal cadherin ectodomains are expressed functionally and at high levels; we have determined crystal structures of ectodomains from desmogleins 2 and 3 and desmocollins 1 and 2; we have biophysically characterized binding behavior of these proteins, revealing family-wise heterophilic specificity ? where desmogleins only bind to desmocollins, and vice versa. In this proposal we will (1) perform functional mutagenesis analyses of structurally identified putative cis and trans interface regions, with biophysical and structural readout; (2) perform cell-based mutagenesis experiments to correlate molecular structural features with desmosome morphology and function in transfected cells; and (3) produce Cryo-EM tomographic reconstructions of desmosome-like junctions that spontaneously polymerize from Dsg and Dsc extracellular domains between adherent liposome membranes. Overall, this work will provide an atomic-level understanding of desmosome extracellular architecture.

Pemphigus is a group of potentially life-threatening antibody-mediated autoimmune diseases of the skin and other stratified epithelia in which acantholysis ? the loss of cell adhesion ? causes skin blistering and erosions. Acantholysis in pemphigus is caused by autoantibodies directed against desmosome cell-adhesive junctions ? specifically against the transmembrane cadherin- family proteins that bind between cells to mediate adhesion in desmosomes. Several subtypes of pemphigus disease are known, including two major forms pemphigus vulgaris (PV) and pemphigus foliaceus (PF). Broadly, PV is characterized by acantholysis in the basal layers of mucosae (mucosal form) or mucosae and skin (muco-cutaneous form), while PF is characterized by acantholysis specifically in the subcorneal upper layers of the skin. Pathogenic pemphigus autoantibodies have been identified from patients with each form of the disease, but structural information on pemphigus autoantibodies is lacking. Thus, the precise epitopes targeted by pemphigus autoantibodies, and the antibody regions (paratopes) that mediate recognition, remain unknown. The overall goal of the research proposed here is to bring atomic-level definition to the study of pemphigus disease through the application of modern methods of structural biology. Atomic resolution co-crystal structures will unambiguously identify functional regions and define the precise molecular interactions mediating recognition between pemphigus autoantibodies and the cadherin cell-adhesion proteins they target. In addition, to determine how different pemphigus autoantibodies impair desmosome structure and cause blistering, we will analyze their effects on reconstituted desmosome junctions at high resolution using cryo-EM tomography. The research proposed here will produce an atomic-level understanding of the interaction of pemphigus autoantibodies with desmosomes, and is expected to transform our understanding of pemphigus disease.

Deafness is a major health problem. A major cause of deafness is defects in hair cells, the mechanosensory cells of the cochlea that convert sound induced vibrations into electrical signals to provide our sense of hearing. Mutations in the genes encoding protocadherin (PCDH15) and cadherin 23 (CDH23) cause hearing loss. Both genes are expressed in the hair bundles of the mechanosensory hair cells of the inner ear where they form heterophilic adhesion complexes that are important for hair bundle morphogenesis and mechanotransduction. Significantly, different mutation in both PCDH15 and CDH23 lead to different disease outcomes. While some mutations cause profound congenital deafness with retinal impairment (Usher Syndrome) others lead to recessive and progressive hearing loss without visual involvement. Gene-association studies also suggest a link of CDH23 polymorphisms with age- and noise-induce hearing loss. The mechanisms by which different mutations lead to distinct disease outcomes are poorly defined. We propose here to combine high-resolution structural studies with functional studies in hair cells to gain insights into the mechanisms by which PCDH15 and CDH23 regulate hair cell function and to define disease mechanisms. To achieve this goal, a laboratory with expertise in studying the biophysical and structural properties of cadherins and a laboratory dedicated to the study of auditory neuroscience have combined their efforts to achieve what either could not accomplish alone. Unlike previous studies that have focused on structural analysis of small monomeric fragments of CDH23 and PCDH15 expressed in bacteria, the team proposed to define the high- resolution structure of natively assembled PCDH15-CDH23 complexes using crystallography and cryo-EM. Structural data will be validated biochemically and by functional interrogation of mutant cadherins in the physiologically relevant mechanosensory hair cells paying attention to mutations associated with disease. We anticipate that our studies will provide the first high-resolution native structure of any protein complex important for mechanotransduction and provide mechanistic insights into its functional properties and pathophysiological mechanisms that are associated with different forms of hearing impairment.

Abstract A fundamental function of the kidney is the recovery of filtered water, electrolytes, and proteins in order to conserve valuable nutrients while discarding the final urine. The sequential recovery of electrolytes and water is well understood, however, less is known about the capture of proteins from the filtrate. Protein capture is mediated by two enormous proteins called megalin and cubilin. Despite the critical function of these two molecules, little is known about their molecular mechanisms, and fundamental questions about megalin and cubilin function remain unanswered: How does a single receptor recognize and bind so many different proteins? What is the receptor:ligand stoichiometry and affinity? Do different types of proteins bind to the same receptor molecule at the same time, or do ligands cooperate or compete with one another for binding? These questions have remained unanswered in part because of the large sizes of megalin and cubilin, 600kDa and 450kDa respectively, making their biochemical, molecular and structural analysis daunting. In a labor-intensive undertaking, Drs Shapiro and Brasch and Drs Barasch and Beenken have purified to homogeneity these massive proteins as well as a native megalin-cubilin-albumin complex. The isolated proteins demonstrated non- aggregated, well-behaved single particle behavior in electron microscopy experiments. 3D reconstructions from negative stain EM reveal a remarkable architecture, in which the domains of megalin fold to form a large globular structure in which deep crevices and holes of different sizes are formed by association of the numerous megalin sub-domains. These crevices and holes are large enough to dock different urinary ligands such as NGAL and albumin. Based on these observations, we propose that megalin may act as a ?sponge? with binding pockets complementary to different ligands. The megalin-cubilin-albumin complex appears larger and has distinct structural features. Beginning with these preliminary data, our goal is to define the structure of the megalin and megalin-cubilin protein-recycling receptors, and their complexes with filtered-protein ligands. We will first use single particle cryo-EM to assign the identities of receptor sub-domains visualized in 3D EM reconstructions, and use these assignments to identify ligand-interacting regions of the receptors. To achieve high resolution, some of these studies will be performed with smaller recombinant receptor fragments with structure determination by x-ray crystallography. We will assess the function of different receptor domains with mutagenesis and analysis of ligand binding by SPR, using both known megalin and cubilin ligands as well as novel candidates isolated from urine of humans with defined Donnai Barrow mutations. This work is the first to visualize full-length megalin and megalin-cubilin structures; we expect that our structures will connect the function of the giant recycling receptors to sequence and chemistry, and we expect this will be transformative for kidney biology.