Identification of genes responsible for the pathogenesis of MAP is important in understanding how this bacterium causes disease. A transposon mutagenesis system was developed to randomly deliver a transposable element (Tn5367) into the genome of the virulent wild type strain K-10. We obtained 13,500 independent mutants necessary for a 95% theoretical coverage of the genome (Harris and Barletta, Clin. Microbiol. Rev., 2001). In this study, we focused on further in vitro and in vivo characterization of mutants 1F3, 40A9 and 4H2. Strain 1F3 displays a haloless phenotype on siderophore-detection medium (possibly indicating no siderophore production and/or secretion); strain 40A9 is a random transposon mutant that grows faster on D-cycloserine; and strain 4H2 is a colony morphology mutant. We used these mutants in baby-goat and rodent infection experiments and in vitro cell culture assays. We focused on the initial critical steps in the pathogenesis: gut invasion and colonization in vivo, and phagocytosis and survival of MAP in macrophages. In vivo, no reduction in colonization was found, all baby-goats and mice, regardless of bacterial strain used for inoculation, were positive by PCR and histologically by ZN staining of ileum, jejunum and mesenteric lymph nodes six months after inoculation. All three mutants maintained their ability to colonize the small intestine in both baby goats and mice. In vitro, we showed that MAP induces nitric oxide synthetase (iNOS) gene expression and NO production using monocyte-derived bovine macrophages, mouse peritoneal macrophages, and in the mouse macrophage-like cell line RAW264.7. Initial findings showed that all mutant strains induced less NO production relative to K-10. Macrophage infection and intracellular survival were also reduced for the mutant strains in comparison to K-10. Experiments to assess the overall virulence of MAP strains in goats and mice are in progress.