John Leong - US grants

9413346 Hills This award will establish a high speed, wireless data communications infrastructure on the campus of Carnegie Mellon University. The research efforts supported by this new infrastructure are devoted to making ubiquitious wireless data communications service available to small mobile computers. The research activities to be supported by the new infrastructure are the CODA project, which is concerned with shared data access over wireless links; the development of a series of wearable computers; development of the Mobile Host Routing Protocol, which provides network layer addressing service to mobile terminals; and a computer model, which is being used to explore the engineering and economic aspects of wireless networks. The applications being developed are: the FRIEND system, which supports emergency management, the Health Link system, which supports home health care; and Wireless Andrew, a proposed application which will make available on wireless terminals the full functionality of "Andrew," Carnegie Mellon's campus computing network. ***

DESCRIPTION (adapted from the application) Enterohemorrhagic E. coli (EHEC) is an important cause of diarrheal disease and renal failure in the U.S. EHEC colonizes the mucosa of the large bowel, and the extraintestinal manifestations of EHEC infection result from the absorption across the epithelium of Shiga-like toxin (Stx) produced by intestinal EHEC. During attachment to colonic epithelium, the bacterium disrupts the host cell cytoskeleton and forms a highly organized cytoskeletal structure underneath the bound bacterium, termed an attaching and effacing (AE) lesion. Intimin, a bacterial outer membrane protein that mediates tight host cell attachment, is required for AE lesion formation and full virulence. Experimental infection with EHEC expressing intimin from enteropathogenic E. coli (EPEC), a pathogen that infects a different intestinal site, suggested that intimin influences tissue tropism. We postulate that: (1) intimin plays a central role in determining the site of colonization; and that (2) intimin-mediated cytoskeletal disruption of intestinal epithelial cells facilitates delivery of Stx to extraintestinal sites. To characterize the features of intimin that influence tissue tropism and promote Stx translocation, the following questions will be addressed: 1. Does an EHEC strain that expresses EPEC intimin under appropriate regulatory controls demonstrate altered tissue tropism? The EHEC chromosomal eae coding sequence will be specifically replaced by the EPEC eae coding sequence, and the effect of this alteration on tissue tropism will be assessed. 2. What domain of intimin influences the site of intestinal colonization? The region of EPEC intimin responsible for the differences in tissue tropism that we expect to find in Aim 1 will be identified by analyzing strains that express hybrid EHEC/EPEC intimin proteins. 3. Does the ability to generate robust AE lesions correlate with mucosal damage and/or toxin translocation? Isogenic EHEC strains differing in their ability to generate AE lesions will be characterized for differences in mucosal damage and in translocation of toxin across intestinal epithelium. By developing a detailed understanding of the role of intimin in the pathogenesis of EHEC, the proposed experiments may lead to therapeutic strategies designed to prevent colonization at one of the earliest steps or to minimize the intestinal absorption of toxin after infection has been established.

Enterohemorrhagic E. coli (EHEC) has emerged as an important agent of diarrheal disease and the leading cause of pediatric renal failure in the U.S. Intimate attachment to host cells is an essential step during intestinal colonization by EHEC. After initial host cell attachment, the bacterium injects into the host cell a number of molecules that trigger signaling pathways and result in the disruption of the eukaryotic cytoskeleton. Among the injected proteins is Tir, a protein that becomes localized in the host cell membrane and acts as a receptor for the bacterial outer membrane protein intimin. Intimin, encoded by the eae gene, is required for the formation of a highly organized for the formation of a highly organized cytoskeletal structure containing filamentous actin directly beneath the bound bacterium that lifts the bacterium above the plane of the host cell membrane on a "pedestal". Deletion mutants of eae, which cannot induce the formation of this pedestal, are deficient for intestinal colonization. Thus, we postulate that Tir-intimin interaction is an essential early event in the development of disease caused by EHEC. We have identified regions of intimin and Tir that interact with each other , and have shown that the Tir-binding region of intimin is sufficient to induce actin condensation after pre-infection of host cells with E. coli. A detailed understanding of Tir-intimin binding, as well as of the molecular signals immediately downstream of this interaction, are required to gain insight into how EHEC colonizes the intestine and promotes damage. Thus, the following questions will be addressed: 1. What is the topological map of Tir in the eukaryotic membrane? 2. Is Tir binding by intimin sufficient to trigger actin condensation on preinfected cells? Latex beads that artificially bind TIR will be tested for the ability to induce actin condensation on preinfected eukaryotic cells. 3. How does intimin and Tir recognize each other? Genetic and biochemical approaches, including crystallographic studies, will be pursued to understand the molecular basis for this interaction. 4. Is Tir-intimin interaction essential to promote intestinal colonization? Point mutations in eae and tir that disrupt or restore Tir-intimin binding will be tested for their effect on colonization in an animal model for EHEC infection. 5. What mammalian cell factors interact with the cytoplasmic region(s) of Tir? Mammalian cell factors that directly receive from Tir the biochemical signal for actin filament formation will be identified. The proposed experiments may provide novel targets for therapeutic intervention during EHEC infection, as well as provide insight into the general cellular mechanisms by which actin assembly controlled.

DESCRIPTION (Adapted from the Applicant's Abstract): Borrelia burgdorferi is the causative agent of Lyme disease, and B. hermsii and B. turicatae are causative agents of tick-borne relapsing fever. Pathogen-host cell interactions are thought to be critical determinants of the site and severity of infection, and Dr. Leong's group has focused on Borreliae recognition of two classes of host cell molecules: (1) glycosaminoglycans (GAGs); and (2) integrins and their associated proteins. For B. burgdorferi, they have found that differences in GAG recognition were associated with differences in host cell type-specific binding, and identified a surface protein, Bgp, that may be the major B. burgdorferi GAG receptor. This bacterium also recognizes the activation-dependent platelet integrin alphaIIbbeta3 and thereby selectively binds to activated (vs. resting) platelets. This integrin-binding activity is predicted to target the Lyme disease spirochete to the vessel wall at sites of platelet adherence, and could explain a salient feature of Lyme disease: vascular pathology of the arterial circulation. In Dr. Leong's studies of relapsing fever spirochetes, high-level GAG-binding correlated with high-level growth in the bloodstream, and a variable major protein, VspB, promoted attachment to GAGs. Additionally, in contrast to B. burgdorferi, B. hermsii bound and activated resting platelets. The platelet activation activity is apparently mediated by the integrin-associated platelet-signaling molecule CD9. Dr. Leong speculates that prior to the development of an antibody response, attachment of relapsing fever spirochetes to the vessel wall, either directly via GAGs or indirectly, via activated and adherent platelets, could diminish the clearance of bacteria from the bloodstream by the reticuloendothelial system. Continued replication by these adherent bacteria would result in high level bacterial seeding of the bloodstream. Interaction of spirochetes with platelets could also contribute to thrombocytopenia, a common manifestations of relapsing fever.

DESCRIPTION (provided by applicant): Enterohemorrhagic and enteropathogenic E. coli (EHEC and EPEC, respectively) are important agents of diarrheal disease that induce actin pedestals on intestinal epithelial cells beneath sites of bacterial attachment. They do so by injecting the host cell with proteins that ultimately activate a host cell regulator of actin assembly known as N-WASP. One critical effector for both pathogens is Tir, a bacterial protein that is inserted into the host cell membrane and acts as a receptor for the bacterial outer membrane protein intimin. Despite these similarities, pedestal formation by EHEC and EPEC involve fundamentally different mechanisms. For EPEC, Tir is the only bacterial protein delivered to host cells that is required to induce formation of pedestals. Clustering of this protein in the host cell membrane promotes binding to Nck, a host adaptor protein that in turn activates N-WASP. In contrast, the Tir of EHEC neither binds to nor requires Nck for pedestal formation, and is not the only translocated bacterial protein required for pedestal formation by EHEC. Instead, EHEC requires a second translocated bacterial protein, termed EspFU, recently identified by our laboratory. Our current data support a model in which EspFU interacts directly with N-WASP to promote the formation of Tir/N-WASP complexes. However, unlike Nck, EspFU does not appear to interact directly with Tir. This observation implies that an unidentified host factor is likely to be required for the ultimate formation of Tir/N-WASP complexes. We propose to test and refine this model by: (1) delineating the elements of the cytoplasmic domain of EHEC Tir essential for actin assembly; (2) defining the essential interactions between Tir, EspFU and N-WASP, and identifying the putative host protein that is required for Tir-EspFU interaction, should we confirm its existence; (3) recapitulating Tir/EspFU-mediated actin assembly in cell-free extracts; (4) assessing the role of actin pedestal formation in intestinal colonization and the induction of tissue damage during EHEC infection.

Borrelia burgdorferi, the agent of Lyme disease, causes a chronic multisystemic illness, and its interaction with several components of extracellular matrix (ECM), such as fibronectin (Fn), the proteoglycan decorin, and glycosaminoglycans (GAGs), is thought to promote infection of diverse tissues. Several B. burgdorferi molecules that may promote this binding have been identified biochemically, including the GAG-binding protein Bgp, the decorin binding proteins DbpA and DbpB, and the fibronectin binding protein BBK32. We found that B. burgdorferi harboring an insertion in bgp remained infectious in mice, indicating that Bgp is not required for colonization. Consistent with this, B. burgdorferi binds more efficiently to GAGs upon adaptation to the host environment, but without demonstrable induction of bgp, suggesting that other adhesins may contribute to GAG binding. In fact, by expressing DbpA, DbpB or BBK32 on the surface of a high-passage, otherwise nonadherent B. burgdorferi strain, we demonstrated that in addition to their known ECM targets, all three are capable of promoting bacterial attachment to GAGs. The GAG-binding activity of DbpA is subject to allelic variation, and the GAG- and fibronectin-binding activities of BBK32 are apparently separable. We have recently generated targeted mutations of bgp, dbpA/dbpB and bbk32 in infectious B. burgdorferi strain backgrounds. To identify which components of ECM are physiologic receptors of B. burgdorferi adhesion, and whether their identity varies with different target tissues, we will generate derivatives of BBK32 that have lost Fn- and/or GAG-binding activity, and derivatives of DbpA and B that have lost decorin- and/or GAG- binding activity. Mice will be infected with B. burgdorferi bbk32 or dbpA/B mutants to assess the roles of these adhesins during infection. If bbk32, dbpA, /and/or dbpB are required for colonization of one or more tissues, variants of these genes that result in selective loss of GAG-, decorin- and/or Fn-binding activity will be tested for their ability to complement the colonization defect. By developing detailed knowledge of the interactions that are critical to colonization and disease of the Lyme disease spirochete, these studies may lead to novel therapeutic strategies aimed at preventing colonization by this important pathogen. Such studies may also shed light on general principles that govern tissue-specific infection by bacterial pathogens.

Project Summary New drug delivery platforms are vitally needed for the targeted delivery of high-specificity therapeutics to sites of disease in order to maximize therapeutic efficacy while limiting off-target side effects. The majority of efforts currently underway for the development of such targeted drug delivery systems are focused on the development of synthetic nanoparticles, materials which are costly to produce, store, and distribute. Here, we propose to develop cost-effective, self-replicating and flexible, programmable designer probiotics for the targeted delivery of therapeutics directly to sites of disease. We propose to utilize a synthetic biology approach to genetically engineer a widely and safely administered probiotic, Escherichia coli Nissle 1917, to express a nanomachine that can secrete therapeutic payloads into the intestinal milieu. These designer bacteria will be equipped with a type 3 secretion system modified to secrete proteins into the intestinal lumen rather than their innate target, the cytosol of mammalian cells. As proof of concept and towards the development of these designer probiotics as therapeutics, we will engineer these designer probiotics to secrete a new class of well- documented therapeutic biomolecules of exquisite specificity, single domain antibodies, also referred to as VHH. We will focus on the delivery of VHH multimers that exhibit profound neutralizing activity, VHH-based neutralizing agents (VNAs), which inhibit the activity of essential bacterial toxins or proinflammatory cytokines. Furthermore, we will investigate the potential of these strains as novel therapeutic paradigms for the treatment of both intestinal infections and inflammation disorders. Specifically, we will investigate the efficacy of these strains in the prevention of treatment of Clostridium difficile infections (CDI), hemolytic uremic syndrome (HUS) and inflammatory bowel disease (IBD). We appreciate that there might be some concern regarding the administration of genetically modified bacteria as therapeutics. However, as we enter the `era of the microbiome,' it seems extremely likely that such interventions are to become an integral component of the armamentarium utilized to treat infections and inflammatory disorders, particularly those rooted in the gastrointestinal tract, especially because of the high likelihood that such agents can overcome many issues associated with the wide-spread usages of antibiotics and systemic immunosuppressive agents. While efforts here are specifically devoted towards the development of these designer probiotics for the treatment of CDI, HUS, and IBD, once established as a therapeutic paradigm, this designer probiotic platform can be extended to treat a variety of intestinal based diseases. For example, the designer probiotics could be programmed to deliver a variety of protein-based therapeutic payloads, including cytokines, such as IL-10, that suppress intestinal inflammation or VNAs designed to target essential exposed virulence proteins of enteric bacterial pathogens, e.g., adhesins or essential components of virulence factor delivery systems.