Two experiments were conducted to evaluate the feeding of direct fed microbials (DFMs) on fecal shedding of O157:H7 in naturally infected cattle (experiment I) and on in the feces and peripheral lymph nodes (PLNs) of experimentally infected cattle (experiment II). Thirty cattle, 10 per treatment, were used in each experiment. Treatments in experiment I consisted of a control (lactose carrier only); DFM1, a 1:1 ratio of and ; and DFM2, a 1:1 ratio of and . In Experiment II, DFM1 was replaced with DFM3, a 1:2 ratio of and other strains. Additives were mixed in water and applied as a top-dressing to each pen's daily ration for 50 days. Approximately half-way through each experiment, the DFM concentration was doubled for the remainder of the study. Fecal samples were collected throughout experiment I and cultured for O157:H7. Cattle in experiment II were inoculated intradermally with Montevideo on days 32, 37, and 42 and then necropsied on days 49 and 50 (five cattle per treatment on each day). Innate immune function was assessed on days 29, 49, and 50. In experiment I, fecal concentration and prevalence of O157:H7 were not different ( > 0.10) nor was there an effect ( = 0.95) on the percentage of super shedders (cattle shedding ≥3.0 log CFU/g of feces). In experiment II, no treatment differences ( > 0.05) were observed for in the PLNs except for the inguinal nodes, which had a significantly lower prevalence in DFM-supplemented cattle than in the controls. Immune function, as measured by monocyte nitric oxide production and neutrophil oxidative burst, was decreased ( < 0.05) in the DFM treatment groups. Although results of this research indicate little to no effect of these DFMs on O157:H7 or in cattle, an increase in the duration of administration to that similar to what is used for commercial cattle might elicit treatment differences.