The innate immune system recognizes invading microbes by means of pattern recognition receptors (PRRs) that bind unique products of microbial metabolism (pathogen-associated molecular patterns, or PAMPs). Examples of PAMPs recognized by PRRs such as peptidoglycan recognition proteins (PGRPs), nucleotide-binding oligomerization domain-containing proteins (NODs), and the extracellular matrix protein mindin include LPS and peptidoglycan (PGN). Although much is known about the function of PRRs in host defense, far less is known about how they recognize PAMPs. Accordingly, we determined the crystal structure of human PGRP-Ia in complex with a muramyl tripeptide (MTP) representing the core of Lys-type PGNs. The peptide stem of MTP is buried at the deep end of a long binding groove, with N-acetylmuramic acid in the middle of the groove, whose shallow end could accommodate a linked N-acetylglucosamine. Most interactions are with the peptide, rather than glycan, suggesting that PGRPs mainly recognize the variable peptide stems of PGNs. This work provides a starting point for detailed structure-function analysis of PRR-PAMP interactions. In addition to PGRPs, we will focus on two other key PRRs: NODs, which mediate intracellular detection of PGNs, and mindin, which interacts with LPS and is essential for initiation of immune responses. The specific objectives of the proposed studies are: 1. Determination of the structures of selected human and Drosophila PGRPs in complex with both Lys- and Dap-type PGN analogs in order to explain the ability of PGRPs to discriminate between Gram-positive and Gram-negative bacteria. Structures will be correlated with the affinity of PGRP-PGN interactions measured by ITC or SPR. 2. Determination of the structure of Drosophila PGRP-LC, a cell surface receptor required for Imd pathway activation, PGRP-LC exists as three splice isoforms (LCa, LCx, LCy) that form homo- or heterodimers with distinct PGN specificities. To understand ligand recognition by this receptor, we will examine dimerization of PGRP-LC isoforms in solution, followed by structure determination of homo- and heterodimers. 3. Determination of the basis for PGN recognition by NODI and NOD2 in order to define the relationship between their PGN binding mode and that of PGRPs. We will engineer recombinant forms of NOD1 and NOD2 and determine their structures bound to PGN ligands. 4. Identification of LPS determinants recognized by mindin. We have expressed recombinant mindin for use in ITC and SPR binding experiments to synthetic LPS oligosaccharides and lipid A derivatives, and for crystallization of specific mindin-ligand complexes.
