Search

IS900 PCR for Mycobacterium paratuberculosis was applied to cream, whey, and pellet fractions of centrifuged whole cows' milk. The test and simultaneous control reactions gave correct results for spiked milk and for native milk samples obtained directly from M. paratuberculosis-free, subclinically infected, and clinically infected cows. The test was then applied to units of whole pasteurized cows' milk widely obtained from retail outlets throughout central and southern England from September 1991 to March 1993. With peak periods in January to March and in September to November, when up to 25% of units were affected, an overall 22 of 312 samples (7%) tested positive for M. paratuberculosis. In 18 of the 22 positive samples (81%), the PCR signal segregated to the cream or pellet fractions or both, consistent with the presence of intact mycobacteria. Nine of 18 PCR-positive milk samples (50%) and 6 of 36 PCR-negative milk samples (16%) yielded long-term liquid cultures which tested positive for M. paratuberculosis after 13 to 40 months of incubation, despite overgrowth by other organisms. Taken together with data on the prevalence of M. paratuberculosis infection in herds in the United Kingdom, the known secretion of M. paratuberculosis in milk from subclinically infected animals, and the inability of laboratory conditions simulating pasteurization to ensure the killing of all these slowly growing or unculturable organisms, there is a high risk, particularly at peak times, that residual M. paratuberculosis will be present in retail pasteurized cows' milk in England.