Mycobacterium avium (MA) is divided into four subspecies based primarily on
host-range and consists of MA subsp. avium (birds), MA subsp. silvaticum (wood
pigeons), MA subsp. paratuberculosis (broad, poorly-defined host range), and the
recently described MA subsp. hominissuis (humans and swine). MA is ubiquitous in the
environment and evidence suggests water is a possible source of human exposure.
Routes of exposure to waterborne pathogens include ingestion, inhalation of water vapor,
and ingestion of produce irrigated or washed with contaminated water. The goal of this
study was to determine the specificity of MA subsp. hominissuis to humans by
subspeciating human clinical isolates of MA, and also to determine the proportion of MA
isolated from food and water that belongs to this subspecies. Understanding the
occurrence of MA subsp. hominissuis in water and food will aid the U.S. Environmental
Protection Agency in assessing the risk of human exposure to these sources.
Temperature growth range, IS1245 copy number, sequencing of the 16S-23S internal
transcribed spacer region or hsp65 gene and other methods have been used to
differentiate subspecies hominissuis from the other subspecies. MA subsp. hominissuis
can grow on Lownestein Jensen (LJ) medium supplemented with pyruvate at 45 Cº while
bird type isolates cannot (Mijs et al., 2002). Clinical and environmental MA isolates
from our collection were inoculated on this medium and allowed to incubate for up to 90
days. Ninety percent of the isolates grew at 25 Cº and 67% grew at 45 Cº. Isolates that
grew at 45 Cº also grew at 25 Cº. Nearly 100% of the human isolates included in this
study grew at 45 Cº, suggesting they are strains of MA subsp. hominissuis. In addition,
we performed PCR and sequence analysis on the 3' end of the hsp65 gene as a means to
confirm MA subsp. hominissuis identification. Human isolates had hsp65 sequences
closely related to hsp65 sequences from MA subspecies hominissuis described
previously. Moreover, several hsp65 sequences from environmental isolates were
identical or very closely related to those of hominissuis. None of the human or
environmental isolates in our study had hsp65 sequences that clustered with MA subsp.
avium or paratuberculosis. These data suggest that the type of MA that is specific to
humans may be more prevalent in the environment than was previously thought. Includes 6 references, table, figure.