Enteropathogenic E. coli (EPEC), a leading cause of infantile diarrhea, adhere to cultured epithelial cells forming microcolonies in a pattern termed localized adherence. We have defined two classes of transposon mutants deficient in localized adherence and have identified the loci interrupted in both sets of mutants. One class of mutants, with transposon insertions on the large EPBC plasmid, have led us to the identification of a 10.8 kilobase type-IV fimbrial gene cluster that is required for the production of a Bundle-Forming Pilus (BFP). We have identified 12 open reading frames in this gene cluster and characterized two genes. One, bfpA, encodes bundlin, the major structural subunit of the BFP. The other, bfpP, encodes a prepilin peptidase necessary for proteolytic cleavage of pre-bundlin to produce the mature protein. In addition, we have found that mutants of the second category deficient in localized adherence have mutations in a chromosomal gene, (LsbA, which encodes a periplasmic disulfide bond oxidizing enzyme. We propose to continue these studies to test the following four hypotheses: (1) Bundlin, the major structural subunit of BFP, is the bacterial ligand responsible for the localized adherence phenotype of EPEC. Bundlin mediates both binding of the bacteria to each other and to epithelial cells. (2) The DsbA enzyme catalyzes the oxidation of cysteine residues in bundlin necessary for adherence. (3) The bfp gene cluster encodes the proteins necessary for assembly of the pilus on the surface of the bacterium. The proteins comprise a secretory apparatus that exports bundlin through the outer membrane and directs the incorporation of bundlin subunits into the fimbrial structure. Included in this gene cluster are the bfpP gene, which we have shown to encode a prepilin peptidase, and two loci that contain putative ATP-binding sites. We hypothesize that all of the genes in the cluster are necessary for the various steps in the pilus assembly process. And, (4) additional loci on the large FPEC plasmid represent genes required for fimbrial production. Given the wide distribution of type-IV fimbriae among important Gram negative pathogens (Pseudomonas aeruginosa, Neisseria gonorrhoeae, Vibrio cholerae, etc.) and the relative ease of genetic manipulations in E. coli, we anticipate that the results of these studies will have important implications for our understanding of and ultimately interference with colonization by pathogenic bacteria.
Showing the most recent 10 out of 38 publications
