From Sardinia to Japan: update on the role of MAP in multiple sclerosis

Multiple sclerosis (MS) is an immune-mediated disease of the central nervous system (CNS) characterized by demyelination and axonal degeneration [1]. It is widely believed that a combination of host genetic factors and exposure to environmental triggers contribute to disease induction or evolution [1]. Several species of mycobacteria, most of which are obligate intracellular pathogens, have been associated with many immune-mediated disorders including MS [1,2]. Multiple studies have suggested a possible causal relationship between Mycobacterium avium subsp. paratuberculosis (MAP) and MS [1–3]. MAP is an obligate intracellular bacterium belonging to Mycobacterium avium complex, it is the causative agent of Johne’s disease (paratuberculosis) in ruminants, and it has also been associated with several human autoimmune disorders such as Crohn’s disease, Type 1 diabetes and MS [2].

Initial evidence from Sardinia

The association between MAP and MS was proposed for the first time in 2010 in MS patients from Sardinia, a Mediterranean Italian island characterized by a high incidence of the disease (330 per 100,000 people), and by the endemic presence of MAP (60% of sheep herds are infected by the bacilli) [4]. Briefly, the association between MAP and MS was initially based on two key findings: the detection of MAP DNA in the peripheral blood of 25% of MS patients compared with 4% of healthy subjects, and the discovery of several MAP-specific antigens that share sequence homology with MS-related human proteins which are capable of eliciting a significantly stronger humoral and cell-mediated response in both sera and cerebrospinal fluid (CSF) of MS patients compared with controls [4–6]. For this reason, molecular mimicry was the main hypothesis as to how MAP antigens initiate or accelerate the autoimmune responses, which led to specific CNS damage [4], suggesting that autoimmunity is triggered by a combination of genetic predisposition and environmental factors such as MAP and Epstein–Barr virus [7]. Indeed, competitive assays have shown that peptides from different pathogens can be cross-recognized by antibodies targeting self-epitopes, indicating that Epstein–Barr virus and MAP might cause autoimmunity through a common target [6–8]. In previous work, Cossu et al. reviewed the association with MAP and MS in Sardinia and a hypothetic course of action of the bacteria [3].

Evidence from Japan

Globally, the incidence and prevalence of MS vary substantially, and the mechanisms underlying these conundrums in geographical incidence of MS are unknown. MS is most common in the western world and in contrast, is uncommon in Asia [9]. However, epidemiological studies have revealed the greatest increase in newly diagnosed MS patients in Japan (crude incidence rate is 15 per 100,000 people) [10]. Genetic changes cannot solely explain such rapid rates of increased incidence, so is it possible that the presence of some exclusive environmental factors may modify the prevalence of the disease in Japan. Momotani reported in 2012 that the prevalence of bovine paratuberculosis in Japan was only 2%, in contrast to the highest incidence in western countries [11]. However, keeping in view the increasing prevalence of MAP, no detailed recent studies on prevalence of paratuberculosis have been attempted in Japan.

Otsubo et al. measured the serum levels of total IgG and subclasses targeting MAP surface antigens in 136 Japanese healthy subjects, detecting IgG1 and IgG4 antibodies against MAP antigen in 14% of the population [12]. The results obtained represented the first piece of evidence implying that the Japanese population was exposed to MAP antigens. Since fecal–oral transmission is considered the first route of transmission from animal to human, the exposure to heath-killed MAP antigenic components was likely due to the consumption of dairy products imported from western countries [12]. 1 year later, in an international collaboration between Juntendo University and Sassari University, Cossu et al. performed an analysis of the humoral response mounted against host-mimetic homologous peptides of MAP and MS in Japanese patients with inflammatory demyelinating diseases of the CNS [13]. A strong, statistically significant humoral response was detected in the sera of relapsing-remitting MS patients only against MAP_2694_295–303 peptide. The latter is a surface-exposed epitope of an MAP-specific transmembrane protein and displays sequence homology with the C region of the TCR γ chain expressed by γδ T cells [4,14]. Kumar et al., using a molecular modeling approach, observed that MAP_2694_295–303 peptide displays a preferential binding to the HLA proteins associated with MS (*1501, *0301), showing a higher probability of T-cell activation which could be involved in the development of autoimmune diseases [15]. The prevalence of antibodies against MAP_2694_295–303 peptide in MS Japanese subjects (30%) [14] was essentially similar to that in Italian study (35%) [5], highlighting the importance of MAP_2694_295–303 as an immunogenic peptide that likely plays a role in determining immunogenicity in patients with a genetic susceptibility to the mycobacterium. Of note, HLA DPB1*0301, and in particular HLA DRB1*0405 alleles, are associated with a greater risk of developing MS in a few countries such as Sardinia and Japan [13]. A genetic study conducted by Frau et al. revealed that Sardinian MS patients showed a presence of *03:01–*02:01 and *04:05–*03:01 haplotypes in 36 and 6% of the MAP_2694-positive subjects, respectively [16]. On the other hand, the discrepancies in antibody determination between the Japanese and Sardinian populations suggested that MS patients from different ethnic backgrounds have an altered immune response to MAP antigens. The genetic background, the differing assessment of human exposure to the mycobacterium and the different coverage rates of Bacillus Calmette-Guérin (BCG) vaccination between populations are all important determinants of susceptibility to the disease. Regarding BCG, clinical trials have shown a vaccine-associated reduction in symptom severity among patients with MS and clinical isolated syndrome [17], however, the mode of action has not yet fully elucidated.

In a recent retrospective study, a total of 97 Japanese patients affected by inflammatory demyelinating disorders of the CNS and all BCG vaccinated, were tested for the detection of total IgG against MAP and BCG surface antigens by indirect ELISA [17]. The study provided more evidence supporting the causative role of MAP, while BCG seemed to be inversely related to the risk of developing MS.

Yokoyama et al. detected higher serum levels of anti-MAP_2694_295–303 antibodies in paired CSF and serum samples from Japanese patients with MS when compared with patients with neuromyelitis optica spectrum disorder and controls [18]. Moreover, a group of patients with MS showed intrathecally restricted production of antibodies against MAP_2694_295–303 peptide. Women appeared to mount a stronger humoral response to mycobacterial antigens than men, suggesting sex-related differences in the immune response to MAP [18]. In the same work, a strong humoral response against MAP pentapeptide (MAP_5p) was also detected in both sera and CSF of patients with MS. MAP_5p is a cell-wall antigenic component able to discriminate MAP from other nontuberculosis pathogenic mycobacteria [18].

Evidence from animal studies

Not only human studies, but also new animal experiments, have revealed the adjuvant property of MAP in experimental autoimmune encephalomyelitis (EAE), a murine model of MS characterized by perivascular CD4⁺ T-cell and mononuclear cell inflammation and subsequent primary demyelination of axonal tracks in the CNS [19]. Groups of C57BL/6 mice were actively immunized with MOG_35–55 peptide emulsified in incomplete Freund’s adjuvant modified to contain heat-killed MAP. These groups demonstrated an early disease onset and more severe clinical scores in comparison with classical immunization with complete Freund’s adjuvant containing Mycobacterium tuberculosis [19]. These findings indicated the existence of species-specific antigens that recognize distinct pattern recognition receptors. Several antigenic components present in the complex mycobacterial cell wall are responsible for Toll-like receptor (TLR) activation, which plays an important role in orchestrating proinflammatory and innate immune responses. TLRs are a major driving force for the immune response necessary to develop EAE [19], and it has been shown that MAP components stimulate TLR2, TLR4 and TLR9 on immune cells [20]. At present, the interaction of MAP with the innate immune system, as well the role of the receptors involved in its recognition is not yet fully elucidated.

Ongoing studies are analyzing the efficacy of heath-killed MAP immunization through the oral route, compared with live BCG immunization via intradermal vaccination in the development of EAE, as well the in vivo effect of MAP by using spontaneous EAE murine TCR-transgenic models (specific for MOG_35–55 peptide), in order to simulate the clinical and pathological complexity of human MS.

Conclusion

Although these experimental observations from clinical and animal studies have confirmed the association and provided new data to the puzzle of MAP or BCG and MS, it is still unknown whether a causal link exists. More research is needed to further explore specific aspects of host–pathogen interplay in the physiopathology and treatment of immuno-mediated disorders, combining data obtained from animal models and epidemiological studies.