Numerous reports have described diagnostic methods based on the polymerase chain reaction (PCR) used to detect Mycobacterium avium subsp. paratuberculosis, the causative agent of Johne's disease. The result of conventional PCR tests has been only qualitative, either positive or negative; it does not present any quantitative information about the number of the agents in the specimen. A quantitative PCR method (IS900 TaqMan) was developed to measure the number of M. a. paratuberculosis organisms present in field and clinical samples. The sensitivity of IS900 TaqMan was 1 colony-forming unit (CFU) for M. a. paratuberculosis ATCC 19698. The specificity of the method was determined by testing 14 mycobacterial species (M. abscessus, M. asiaticum, M. avium subsp. avium, M. bovis, M. fortuitum subsp. fortuitum, M. intracellulare, M. kansasii, M. marinum, M. phlei, M. scrofulaceum, M. simiae, M. smegmatis, M. terrae, and M. ulcerans) and 9 nonmycobacterial species (Borrelia burgdorferi, Chlamydia psittaci, Ehrlichia canis, E. equi, E. risticii, Escherichia coli, E. coli O157:H7, Streptococcus equi, and S. zooepidemicus). Even at high cell numbers (10(5) CFU/reaction), most of the organisms tested negative for the IS900 insertion element except M. marinum and M. scrofulaceum. This finding for M. scrofulaceum was consistent with previous reports that several M. scrofulaceum-like isolates were positive for IS900. Those isolates had 71-79% homology with M. a. paratuberculosis in the region of IS900. When used in conjunction with the new liquid medium-based ESP culture system II for bovine clinical fecal samples, IS900 TaqMan confirmed that the ESP II-positive samples contained 10(5)-10(6) CFU/ml of M. a. paratuberculosis. All of the 222 ESP II-positive and acid-fast bacilli-positive samples tested in this study were positive by IS900 TaqMan. IS900 TaqMan was also useful in the study of growth characteristics of 3 groups of M. a. paratuberculosis strains in bovine fecal samples from 3 shedding levels (heavy, medium, and low) based on cell numbers measured by Herrold egg yolk (HEY) agar culture. When cultured in ESP medium, M. a. paratuberculosis reached 10(5)-10(6) CFU/ml within 2 weeks for heavy shedders, 3-4 weeks for medium, and 6-8 weeks for low shedders. No significant growth was observed after up to 5 weeks of incubation for some of low shedders. No or extremely slow growth characteristic of low shedders might be a possible explanation for frequent false-negative results by HEY. The detection time was dependent on the inoculum size and the growth rate of M. a. paratuberculosis. Generation times were inversely proportional to the shedding level: 1-2 days for medium and heavy shedders and >4 days for low shedders. IS900 TaqMan could be a useful tool for determining viable cell counts by measuring changes in cell numbers over the incubation period.