Mycobacterium avium Subspecies paratuberculosis Bacteremia in Type 1 Diabetes Mellitus: An Infectious Trigger?

To the Editor—Mycobacterium avium subspecies paratuberculosis (MAP) is the established cause of paratuberculosis in ruminants (i.e., Johne disease). The bacterium is shed in the milk of infected cows and survives pasteurization. Recently, an association between MAP and Crohn disease has been suggested, wherein MAP has been found to persist in a cell wall—deficient form, escaping clearance by the host immune system [1, 2]. Type 1 (insulin‐dependent) diabetes mellitus (T1DM) reflects the interactions of polygenic traits with ill‐defined environmental factors, and it is unknown what initiates and maintains autoimmunity to self‐antigens expressed in the pancreatic islets of Langerhans [3]. Consumption of cow's milk early in life has been a recognized risk factor in the development of this disease, and environmental microorganisms are thought to trigger autoimmune responses in genetically susceptible individuals [4]. MAP bacilli have recently been hypothesized to trigger molecular mimicry and killing of pancreatic islet cells by the immune system. We attempted to test the association of MAP with T1DM in an area of endemicity, such as Sardinia, by testing patients with T1DM for the presence of MAP bacilli in peripheral blood. A total of 96 participants, composed of 46 patients with T1DM and 50 healthy control subjects, were tested for the presence of MAP‐specific IS900 signature (figure 1) using total DNA extracted from PBMCs. Informed consent from patients, as well as other necessary clearances, were procured before blood samples were obtained. PCR was performed to detect MAP DNA, as described elsewhere [2]. IS900 fragment identity was confirmed by sequencing (GenBank accession group 1517012) and by BLAST analysis against sequences in the National Center for Biotechnology Information database [5]. Representative PCR results showing amplification (or otherwise) of 298—bp genomic fragment corresponding to IS900 in diabetic patients (lanes 1–8). Lanes 9 and 10, Control subjects without type 1 diabetes mellitus. Lanes 11 and 12, Mycobacterium avium subspecies paratuberculosis— positive and M. avium subspecies paratuberculosis—negative control subjects, respectively. M, molecular marker (100-bp ladder). A total of 29 (63%) of 46 blood samples from diabetic patients were found to be positive for MAP, whereas only 8 (16%) of the 50 samples from healthy control subjects were positive for MAP. Although a majority of diabetic patients with positive PCR results had a family history of diabetes or other genetic and/or autoimmune disorders, 14 diabetic individuals with positive PCR results did not have any history of diabetes or other autoimmune diseases in their families. Although the differences in the outcome of PCR positivity were statistically validated and found to be significant (c2=10.07; P⩽.01), it is our understanding that MAP‐positive control subjects might possibly be representing either the genetic resistance or the asymptomatic forms of variable duration that generally precede the clinical presentation of T1DM. Genetic evidence suggests that there are specific states of immune dysfunction that promote both T1DM and mycobacterial infection [6, 7]. Deficiency of vitamin D has been implicated as a risk factor for T1DM. Interestingly, vitamin D is also implicated in limiting mycobacterial infection by upregulating expression of an antimicrobial peptide [8]. The island of Sardinia has a high incidence of Crohn disease and other autoimmune diseases, such as T1DM, with a very high prevalence of MAP in Sardinian patients with Crohn disease [2, 3]. Because MAP is present in almost one‐half of the sheep herds tested in Sardinia, it is probably endemically contaminating water, milk, and animal feed [9]. Finding evidence of MAP bacteremia in patients with T1DM is a novel observation that might provide an important foundation in establishing an infectious etiology for T1DM. These results also might possibly have implications for countries that have the greatest livestock populations and high incidence of MAP concurrent with the highest numbers of patients with T1DM. Although it is tempting to suggest that MAP could be a potential cause of autoimmune responses associated with T1DM in Sardinia, we believe that a large‐scale study involving patients from different genetic and geographic backgrounds might further validate our findings. Financial support. The Italian Ministero Italiano Università E Ricerca (2005064503_003) and the University of Sassari FAR 2006. Potential conflicts of interest. All authors: no conflicts.