What is Mycobacterium avium subsp. paratuberculosis?

We read with interest the article by Beumer et al. (1) on the detection of Mycobacterium avium subsp. paratuberculosis in 81% of drinking water and biofilm samples by PCR. Although these findings could have a significant influence on our understanding and the potential public impact of M. avium subsp. paratuberculosis as an agent of disease, much of the information on M. avium subsp. paratuberculosis based on genomic sequence detection and the environment tends to raise more questions than answers and fuels an ever-increasing mass of conflicting data. It must first be questioned whether the detection of a specific sequence in bulk-extracted DNA actually represents a microbe, an in situ microbe, and a member of the microbial community without establishing a direct link between the molecular sequence and an actual in situ active microbe. It may be a leap of faith to assume that it does. Perhaps it would be more accurate to express such studies as detection of a genomic sequence rather than detection of an actual microbial species? Although IS900 has proved to be species specific and reliable for the detection and confirmation of M. avium subsp. paratuberculosis infection in animals with Johne’s disease (ruminant paratuberculosis) (2, 4), it may again be a leap of faith to extrapolate that specificity to other biological systems such as the environment and human tissues. In today’s era of gene sequencing and genomic databases, we tend to assign sequence specificity based on information contained within those databases without appreciating that they reflect only a small fraction of known bacteria and even a smaller percentage of the genomic diversity of wild-type strains. Within the human gut microbiome, for example, only 20% of ribosomal gene sequences align with known cultivable species of bacteria, suggesting that 80% of all bacteria within the human gut are unknown and/or uncultivable (7). We should expect similar findings in other ecosystems and microbial communities. Beumer et al. (1) have recognized and shown that other mycobacterial species that contain IS900 and/or IS900-like sequences have been identified (although not in genomic databases) and that some strains of M. avium subsp. paratuberculosis may not contain IS900. We agree with the authors’ premise that bona fide strains of M. avium subsp. paratuberculosis are IS900 and 251F positive, as documented in their specificity studies (1), those of others (8), and our own. But what do we make of those strains that are IS900 negative but 251F positive and those that are IS900 positive but 251F negative? Are they or are they not M. avium subsp. paratuberculosis? Historically, the identification of M. avium subsp. paratuberculosis has not been stringent and has been based solely on acid fastness, slow growth, and mycobactin dependency (4). With the advent of liquid culture systems and molecular biological methods, identification has been diminished further to simply acid fastness and IS900 positivity (2). Combined with the knowledge that “species-specific” insertion sequences have been shown to cross species barriers (5), the sole reliance on IS900 as a means of identifying M. avium subsp. paratuberculosis would also seem to be a misguided leap of faith when applied to remote ecosystems. Accepting that the detection of IS900 is prima facie evidence of M. avium subsp. paratuberculosis and that M. avium subsp. paratuberculosis is widely distributed in the environment, the data still fail to hold water. If M. avium subsp. paratuberculosis was as widespread as suggested by Beumer et al. and others (6) and the primary biomass of M. avium subsp. paratuberculosis was the environment, why are not all domestic and wild ruminants infected? Was it not these precise inconsistencies that led to the realization that M. avium subsp. avium (the causative agent of avian tuberculosis) was not an environmental organism and to the designation of M. avium subsp. hominissuis as the only bona fide environmental M. avium strain (9)? Is the overreliance on IS900 as the sole basis for the identification of M. avium subsp. paratuberculosis going to result in a state of confusion similar to that created by strain 18 (3)?