Molecular Basis of Peptidoglycan Recognition by a Bactericidal Gut Lectin

The mammalian gut is densely populated by varied microbial species. This relationship is mutually beneficial as long as bacteria remain corralled in the gut lumen. The epithelium is protected by the secretion of antimicrobial proteins by specialized epithelial cells in the intestinal crypts. This molecular arsenal includes the RegIII family. RegIII proteins are novel in that they are C-type lectins that directly kill Gram-positive bacteria and thus play a vital role in antimicrobial protection of the mammalian gut. RegIII proteins bind their bacterial targets via interactions with cell wall peptidoglycan, but lack the canonical sequences that support calcium-dependent carbohydrate binding in other C-type lectins. Given these novel functions and the lack of structural clues, nothing was known about the molecular mechanisms by which RegIII family members recognize and bind to peptidoglycan. Furthermore, the question of how RegIII proteins specifically recognize target microbes in the presence of soluble peptidoglycan shed by bacteria in vivo still remained. In this dissertation, I have used NMR spectroscopy as an unbiased approach to study the molecular basis for peptidoglycan recognition by HIP/PAP, a human RegIII lectin. I have shown that HIP/PAP recognizes the peptidoglycan carbohydrate backbone, showing that ligand recognition by RegIII family members is unique compared to other peptidoglycan recognition proteins. This work also shows that HIP/PAP recognizes peptidoglycan in a calciumindependent manner via a conserved ‘EPN’ motif that is critical for bacterial killing. While EPN sequences govern calcium-dependent carbohydrate recognition in other C-type lectins, the unusual location and calcium-independent functionality of the HIP/PAP EPN motif suggest that this sequence is a versatile functional module that can support both calcium-dependent and calciumindependent carbohydrate binding. Further, these studies show that HIP/PAP binding affinity for carbohydrate ligands depends on carbohydrate chain length, supporting a binding model in which HIP/PAP molecules "bind and jump" along the extended polysaccharide chains of peptidoglycan, reducing dissociation rates and increasing binding affinity. I propose that dynamic recognition of highly multivalent carbohydrate epitopes in native peptidoglycan is an essential mechanism governing high affinity interactions between HIP/PAP and the bacterial cell wall.