The vertebrate immune response to infection begins with the recognition by the innate immune system of conserved molecular signatures of pathogens, known as PAMPs (Pathogen Associated Molecular Patterns), provoking an immediate and often massive inflammatory response. The innate response holds the pathogen in check, but also plays a crucial role in the generation of acquired immunity. The recognition of PAMPs by the innate system is mediated by a number of receptors, of which the Toll-like Receptors (TLRs) play a prominent role. Unlike the antigen receptors of acquired immunity, the TLRs are encoded by a limited number of germline genes, ten in humans; however, in spite of their small numbers, the TLRs recognize a remarkably wide variety of PAMPs including glycolipids, proteins, and nucleic acids. We have been investigating several aspects of TLR structure and function: 1. The molecular basis for the recognition of a wide array of PAMPs by TLRs is a main interest of my laboratory. In collaboration with Dr. David Davies (LMB, NIDDK), we have recently succeeded in expressing, crystallizing, and determining the molecular structure of the TLR3 extracellular domain (ECD). The structure consists of a solenoid of 23 turns, bent into a horseshoe shape, with a large beta-sheet on the concave surface. The molecules is heavily glycosylated, except that one lateral face of the horseshoe is totally devoid of glycan. Although we have not yet obtained a high resolution structure of a TLR3-ligand (dsRNA) complex, we have located the dsRNA binding site by mutational analysis. The ligand binds on the glycan-free lateral face of the TLR3 molecule to histidine and asparagine residues, near the C-terminal end . We have now completed an investigation of the interaction of dsRNA oligos with TLR3-ECD protein. We have found that purified TLR3-ECD binds dsRNA specifically via a defined ligand-binding site and with an affinity that increases with buffer acidification and ligand size. TLR3-ECD is monomeric in solution, but it forms dimers when bound to dsRNA. These dimers are stabilized by cooperative interactions between the two TLR3-ECDs in a pair, and multiple TLR3ecd dimers bind to long dsRNAs. The smallest oligonucleotides that form stable complexes with TLR3ecd (40-50 bp) each bind one TLR3ecd dimer, and these are also the smallest dsRNAs that activate TLR3 in cells. Thus, we have demonstrated that the TLR3 signaling unit is a ligand-bound dimer of TLR3 molecules. Looking past TLR3, we plan to express and examine ECDs from other TLR paralogs, to see how they differ in structure and ligand binding function from TLR3. 2. Nucleic acid PAMPs such as dsRNA, ssRNA, and CpG DNA, ligands for TLRs 3, 7, 8, and 9, are normally sequestered within microorganisms and become available to interact with TLRs only after the pathogen is endocytosed and lysed intracellularly. By contrast, TLRs 1, 2, 4, 5, 6, and 10 interact with PAMPs that are normally present in the medium, and these TLRs are, as expected, located on the cell surface. Therefore, correct cellular localization is essential for TLR function. We have studied the localization of TLR9, and motifs within the TLR9 molecule that mediate this localization. We have found that TLR9 is located, prior to stimulation, in the ER, and that it interacts with CpG DNA in early endosomes. Both the cytoplasmic and extracellular domains contain internalization signals, and we have also located, within the cytoplasmic domain of TLR9 two regions that control intracellular localization. We are now initiating studies on the intracellular location of TLR3. In the case of TLR3, we have already established that the activation signal varies depending upon the cell type, and we are testing the hypothesis that signaling is influenced by intracellular localization, in particular the pH of the intracellular vesicle that contains TLR3. 3. TLRs play a pivotal role in acquired immunity by triggering the maturation of DC to competent APC, capable of priming naive T cells. Based on our observation that DC also express histamine receptors, we hypothesized that histamine would have an effect on the maturation process. In testing this hypothesis we found that histamine profoundly alters the cytokines released by DC during TLR induced maturation, and as a result, histamine exposure causes DC to polarize naive T cells toward a Th2 phenotype. Mast cells are the major source of histamine, and they are often located in close proximity to DC. We therefore hypothesized that mast cell degranulation at a site of immunization would alter the nature of the immune response, by acting on neighboring DC. By using a mouse model with adoptively transferred transgenic T cells, we have demonstrated an effect of mast cell degranulation on T cell polarization in vivo. By using mice that are deficient in mast cells, we have now shown that mast cells affect the Th1/Th2 balance in mice, and we are now asking whether this has an effect upon the type of antibody produced
Showing the most recent 10 out of 15 publications
