Alexander Gow - US grants

Wayne State University School of Medicine, Detroit, Michigan.

DESCRIPTION (provided by applicant): The long term goals of this project are to define the functions of tight junctions in Stria vascularis of the inner ear and to determine if the diversity of Claudin family members that comprise fight junctions located around Scala media have overlapping or non-overlapping properties. The knowledge gained from these studies will provide deep insights into the roles of the stria vascularis in hearing and potassium recycling to the endolymph. The specific aims are: 1) to test the hypothesis that inner ear pathology in Claudin//-null mice stems from morphological defects. Claudin 11 is expressed in the developing vestibulocochlear apparatus of embryos from E13.5 as well as in adults and evidence of gross structural changes during development and postnatally that could account for the pathology in the knockout mice will be sought. 2) to test the hypothesis that Claudin 11-null mice exhibit early hearing defects. Young adult knockout mice exhibit elevated auditory brainstem response (ABR) thresholds which indicate hearing loss, Further testing of ABRs as well as measurements of endocochlear potentials and DPOAEs at different ages will reveal if this phenotype is progressive or present at the time mice normally start to hear. 3) to optimize a transgene cassette for the expression of heterologous genes in basal cells of Stria vascularis. Basal cell-specific enhancer elements already in-hand will be located within the Claudin 11 gene and used to drive expression of a beta-galactosidase reporter gene in the cochlea, Reporter expression will be characterized with a view to expressing several claudin cDNAs in the cochlea in attempts to rescue hearing loss in Claudin 11-null mice.

The need for higher level processing of Life Sciences resources using a database-like system has increased in the post genomic era, as researchers move from sequence-based to whole genome-based analysis. High throughput and low-cost data processing and gathering techniques are helping researchers to publish an abundance of interesting data from across the globe at an exponential rate. Paradoxically, researchers are now faced with the problem of efficiently and effectively locating and using resources of interest from the vast amount of distributed data repositories. So, the need for generic system and tool support for higher level complex query processing, interpretation of data, and workflow management in a declarative fashion has become critical. The overall goal of this project is to develop a platform, which we will name LifeDB, for ad hoc integration of Life Sciences resources based on a declarative workflow query language called BioFlow. To a large extent, the system will support command line as well graphical interfaces for ad hoc application development and workflow query processing involving distributed resources without any need for traditional programming. The team is developing a visual interface, called WebFusion, for the automatic integration and workflow query processing using biological data as a manual and initial solution, a formal syntax and semantics for BioFlow, a declarative language, to support on the fly integration and workflow query and, new methods for low level data integration, wrapper, and schema mapping service requests toward the implementation of LifeDB system. The approach will change the way biologists currently query databases. Successful implementation and future enhancements of the system will obviate the need for biologists to focus on the "how to" of their application and more on the "what to" of their research. Students are actively engaged throughout the research.

The long term goals of this research are to define and characterize molecular mechanisms by which mutations in proteins cause neurodegenerative disease and to identify pharmacological agents that ameliorate disease symptoms stemming from the accumulation of unfolded protein intermediates in neural cells. The accumulation of unfolded proteins in intracellular compartments is thought to underlie pathogenesis for a number of neurodegenerative diseases, including ALS, Parkinson, Huntington and Alzheimer diseases. More recently, the unfolded protein response has also been implicated in the pathophysiology of metabolic syndrome (diabetes/obesity/insulin resistance/cardiovascular disease), which significantly increases the impact of our work because unfolded diseases may be relevant to more than 6% of Americans. Two key components of our research have been the availability of well-defined mouse models for genetic manipulation, such as gene ablation or the introduction of heterologous transgenes, and the willingness of patients with unfolded protein disease to be involved in clinical studies. In this application, we will continue our work with patients to measure several clinical metrics from MRI and correlate these with disease severity. In addition, we will generate a novel mouse model on unfolded protein disease that will allow us to extend our understanding of the pathophysiology of unfolded protein disease. Specific Aim#1: we will develop a novel in vivo mouse model in which the unfolded protein response can be followed in real time, in individual cells, in vivo. This model will allow us to identify novel gene targets of UPR-induced transcription factors. Currently, several targets have been identified from in vitro cells in culture;however, we find that the data from in vivo experiments often contradict in vitro studies. Our paradigm will enable us to characterize the temporal sequence of events that generate the pathophysiology of neurodegeneration associated with protein trafficking defects in oligodendrocytes.

The long term goals of our research are to understand how metabolic stress perturbs cellular homeostasis and to identify drugs that restore homeostasis by ameliorating this stress. The molecular pathogenesis of oligodendrocyte disorders, from rare diseases like Pelizaeus-Merzbacher disease (PMD) to common diseases like multiple sclerosis, involves a metabolic stress pathway known as the unfolded protein response (UPR). We will study oligodendrocytes as the model cell and use naturally-occurring missense mutations in the proteolipid protein (Plp1) gene to induce the UPR in vivo (i.e. mouse models) and in vitro (transfected cells). To gain insights into this metabolic stress, we will manipulate UPR signaling to alter the pathophysiology associated with mutant Plp1 expression. To do this, we will examine the function of the decoy kinase, Trb3, which is the central regulator of the UPR and Akt signaling pathways. Our work indicates that these pathways are involved in the pathogenesis of PMD and that overexpressing Trb3 in oligodendrocytes ameliorates the clinical phenotype of rsh mice. Our overall hypothesis is that Trb3 functions in UPR signaling to reduce the negative effects of metabolic stress on oligodendrocytes. In Specific Aim#1, the level of Trb3 overexpression in rsh oligodendrocytes that is necessary to ameliorate the phenotype will be determined. We have generated six transgenic lines expressing Trb3 at different levels and find that the highest expressor reduces the rsh phenotype and virtually eliminates UPR activation by three months of age. In the absence of Trb3 overexpression, the UPR remains unabated at this age. We will use electrophysiology, behavior, ultrastructure and expression analyses to determine a dose-response curve for Trb3 function as well as a timeline for disease amelioration. In Specific Aim#2, we will explore the interaction of Trb3 with the ATF4 protein in the UPR as well as Akt signaling. Published studies indicate that Trb3 interacts with ATF4 and Akt. Our data shows that deleting the Akt1 gene in oligodendrocytes reduces the rsh phenotype. And we will determine if deleting Atf4 in oligodendrocytes also reduces the rsh phenotype.

The long term goals of our research are to characterize the CNS myelin radial component, which is a multilamellar super-assembly of claudin 11 tight junctions (TJs) in compact regions of myelin sheaths. To do this, we will identify oligodendrocyte proteins that associate with claudin 11, which is the major membrane protein and is necessary for radial component assembly. We have defined the electrophysiology of the radial component in vivo, by mathematical modeling, and we have shown its physiological properties by X-ray and neutron diffraction. Most recently, we demonstrated that absence of the radial component leads to behavioral phenotypes in amygdala and auditory brainstem circuits. Currently, claudin 11 is the only canonical tight junction protein known to localize to the radial component. Moreover, no other canonical tight junction proteins are present in the radial component, indicating that this structure is almost completely novel in terms of its protein composition because loss of claudin 11 abolishes formation of the radial component. To identify and characterize radial component associated proteins, we will use BioID technology, which has been used in several in vitro systems to identify unknown proteins in macromolecular complexes. We have taken this technology a step further by generating transgenic mice that harness this technology in vivo. This step is necessary because current cell culture systems are not sufficiently robust to enable CNS myelinogenesis in vitro. In the preliminary data, we show that we have generated a mouse with BioID technology incorporated into the Claudin 11 gene. We show that this technology is expressed by oligodendrocytes and present in purified myelin membranes from spinal cord. Finally, we show that BioID is functional under appropriate conditions and we observe a number of unknown BioID-biotinylated proteins (candidate radial component proteins) ranging from 20-200 kDa. In Aim#1, we will purify these radial component proteins for mass spectrometry using 2 complimentary experimental approaches, BioID labeling and immunoprecipitation with claudin 11 antibodies. In addition, we will test that the candidate proteins identified actually colocalize with claudin 11 in tissue sections from mouse brain and primary oligodendrocyte cultures. We will also immunoprecipitate these proteins to show they pulldown claudin 11. In Aim#2, we will clone cDNAs encoding the candidate proteins and express them in oligodendrocyte cell lines to determine if they colocalize with claudin 11. We will also immunoprecipitate these proteins to show they pulldown claudin 11.

ABSTRACT The long term goal of our research is to characterize the molecular composition of the basal cell apical junction complex. Previously, we demonstrated that claudin 11 tight junctions in basal cells are necessary for generating the endocochlear potential and for intercalating connexin 26 gap junction domains, which recycle K+ into the endolymph. In this project, we will identify membrane-associated cytoplasmic proteins that recruit, coordinate the assembly and maintain the apical junctional complex. In the preliminary data, we show in vivo data from a knockin mouse we generated (BirA-cldn11), in which BioID technology has been incorporated into the Claudin 11 gene using homologous recombination. This mouse has no phenotype, and hearing thresholds and auditory brainstem responses are normal; thus, we can conclude that the E.coli BirA biotin ligase-claudin 11 fusion protein is not toxic. We show that the fusion protein is expressed by cochlea basal cells and colocalizes with streptavidin labeling of the stria vascularis under high biotin conditions but not in controls. We show by western blotting that several cochlear proteins (Mr 23-35kDa), one of which is likely claudin 11, are biotinylated in the presence of high biotin but not in controls. Finally, we show mass spectrometry data for a C-terminal tryptic peptide of claudin 11, containing a biotin moiety attached to an internal lysine residue. This peptide was purified from a BirA-cldn11 mouse under high biotin conditions. We have not isolated such a peptide from mice under low biotin conditions. In Aim 1a, we will purify cochlear proteins from BirA-cldn11 mice for mass spectrometry. We will identify an clone the cDNAs for these proteins for expression analysis in transfected cells in culture. In Aim 1b, we will test whether the candidate proteins identified actually colocalize with claudin 11 and connexin 26 in tissue sections from mouse inner ear. We will also generate stably transfected MDCK cells expressing these proteins and claudin 11 or connexin 26 and we will co-immunoprecipitate these candidates to determine if they pulldown claudin 11 and connexin 26 (and vice versa). In Aim 1c, we will use stably transfected MDCK cells expressing BirA-cldn11 to determine if we can purify and identify biotinylated candidate proteins by mass spectroscopy. Together, the experiments in Aims 1a-c comprise a detailed multi-step analysis of apical junctional complex proteins in basal cells that will likely provide insight into the assembly and function of this important complex that is required for normal hearing.