Paratuberculosis, popularly called as Johne's (pronounced "yo-nees") disease, is an infectious disease caused by Mycobacterium avium subspecies paratuberculosis (MAP). It is primarily a disease of domestic and wild ruminants and causes heavy economic losses to the dairy industry. Recent studies have shown that the disease is not restricted to ruminants only and the host range is expanding with reports of natural occurrence of the disease in rabbits, carnivores, and carrion eating birds (Greig et al., 1997; Beard et al., 2001). The zoonotic potential of the organism has been considered because of its implicated association with Crohn's disease (Thompson, 1994, Greenstein, 2003). In India, Johne's disease has been occurring since early 1930s in cattle sheep and goats in almost every state (Tripathi et al., 2002). This happened because of poor knowledge of the complex biology of the organism and lack of adequate and sustained efforts to restrict the spread of infection.

A considerable gap of knowledge still remains in our understanding of pathogenesis, host immune response, bacterial structure and epidemiology that challenge this perception (Stevenson and Sharp, 1997). Early pathologic events particularly interaction between the Mycobacterium and the host are poorly understood. Animals are usually infected during early phase of life through faecal oral route (Chiodini et al., 1984). The mycobactria invade the M cells of follicle associated epithelium (FAE) and subsequently engulfed by intestinal macrophages where they remain for several months to years before replicating extensively and developing clinical disease (Momotani et al., 1988). The existing knowledge on pathobiology of the MAP infection, the immune spectrum and molecular genetics of the organism have resulted into development of a number of diagnostic assays that can be judiciously used for control of the disease. Further, improvements in the diagnostics and vaccination against Johne's disease appear plausible due to complete sequencing of the MAP genome (Chacon et al., 2004).

This manuscript deals with the basic complexity of the bacterium (MAP), infection process and its epidemiology, prevalence in India, diagnosis, control and impact on the public health of the disease.

The characteristic of MAP and the slow progressive infection it causes has significant impact on the diagnosis and control of the disease. The acid-fast and slow growing Mycobacterium has a unique morphological structure and composition of the cell envelope that offer significant biochemical and immunological resistance during intracellular life in macrophages. The outer membrane contains lipids such as phospholipids, glycolipids, peptidolipid and mycosides. The innermost wall is composed of peptidoglycans linked to arabinogalactan polysaccharides, which are esterified with high molecular weight mycotic acid (Minnikin, 1991).

A prominent molecule with profound impact on pathogenesis and diagnosis is lipoarabinomannan (LAM), a polysaccharide that is now recognized as a multiglycosylated form of mannosyl phosphetidyl-inositol (Brennan and Nikaido, 1995). LAM is highly immuno-reactive and has a role in the virulence of MAP as reported for Mycobacterium tuberculosis (Roach et al., 1993). Mycobacterial cell wall envelope particularly of MAP allows the bacterium to resist against heat, disinfectants and hydrophilic antibiotics to which other simple bacteria normally succumb. This suggests the ability of the organism to persist in the environment for a long time, a characteristic integral to the epidemiology and control of the infection. MAP is more themo-resistant than any other mycobacteria and may even be present in the pasteurized milk (Anon, 2000, Manning and Collins, 2001), a finding important from public health point of view if zoonotic value of MAP is finally accepted. Biochemical tests applied to distinguish among other species of mycobacteria are not used to identify MAP. The tests are difficult to perform due to extremely slow growth rate of organisms and variation in the test results among strains of organisms. MAP has tendency to form clumps, which distinguishes this organism phenotypically from Mycobacterium avium

Genetics of MAP has considerable impact on the diagnosis and control of Johne's disease. It exhibits more than 95% DNA homology with M. avium subsp. avium and the 16S rRNA sequences of these two organisms are identical (Van der Giessen et al., 1992). This has necessitated for reclassification and that is why MAP has been kept in M. avium group. Despite these, considerable phenotypic differences exist with regard to the growth rate, pathogenicity and in vitro mycobactin dependency. An insertion sequence IS900 was considered the sole genetic element in MAP that distinguishes it from M. avium, but recent isolation of mycobacterial species from ruminants containing IS900 like sequence, (71–79% homology with IS900 of MAP) has cautioned about PCR false positivity unless confirmed by RFLP (Cousins et al., 1999). This sequence has been extensively used for the molecular diagnosis of paratuberculosis infection. Inaddition, a few more genes such as ISMav2 (Strommenger et al., 2001) and f57 are specific to MAP and have diagnostic potentials.The close relatedness of MAP with M. avium has considerable diagnostic implication due to the presence of large number of common antigens (Mutharia et al., 1997). This property of MAP reduces the specificity of immunologically based tests for the diagnosis of MAP infection.

The MAP genome has recently been sequenced at the University of Minnesota, St. Paul, in collaboaration with National Animal Disease Centre, Ames, Iowa. The size of the complete genome is 4.83 Mb (Chacon et al., 2004). Use of comparative genomics has led to the discovery of 21 sequences specific to MAP ranging in size from 7 to 91 kDa (Bannantine et al., 2004). Five of these proteins were immunogenic in mice and rabbits, and they all reacted with sera from cattle with clinical disease. This appears promising for the first time to find antigens specific to MAP that could be used in serological tests for MAP detection in animals and human hosts.

The ability to survive and multiply in the antibacterial environment of macrophages is key to the pathogenesis of paratuberculosis, which is not adequately explained. The two properties of MAP; the chemically unique cell wall and the factors produced by the mycobacteria play major role in the persistent of the infection. After ingestion through contaminated feed and water, the organism invades the intestinal mucosa through M cells and reaches to lymphoid tissues. The Mycobacterium is then taken up by the macrophages, which become activated and elicit T cell response (Momotani et al., 1988). A variety of cytokines initially secreted by macrophages takes part in the inflammating process. Two subpopulations of helper T cells play major role in the resistance and progression of MAP infection. Under the Th 1 response, animal shows strong cell-mediated immunity and produces a tuberculoid type of pathology. Tuberculoid lesion (also celled as paucibacillary) is characterised by intestinal tissue infiltration with large number of lymphoid cells and few macrophages containing few or no AFB. The earliest granuloma even before tuberculoid lesion can be found in lymphoid tissues of ileum and ileocaecal valve (Kurade et al., 1999). Th-1 cytokines, mainly IFN-γ, IL-2 and TNF-α are believed to orchestrate the CMI functions necessary to contain the intracellular infection. This occurs in subclinical infection, which last for several months to years. Organisms in this type of animals are infrequently demonstrated in tissues and excretion in faeces is either absent or intermittent.

Animals develop nonspecific clinical symptoms such as weight loss and diarrhoea with the occurrence of lepromatous pathology under the influence of Th-2 cytokines (IL-4, IL-5, IL-6, IL-10). However, trigger for transition of Th-1 to Th-2 response is not precisely known (Koets et al., 1990; Stabel, 1990). Th-2 cytokines stimulate humoral immune response, which is not protective and hence progression of MAP infection and pathology go unabated. This results in diffuse infiltration of intestine with macrophage making them nonfunctional, leading to malabsorption and protein losing enteropathy (Patterson et al., 1967). Concomitantly, elevated TNF-α may contribute to emaciation through stimulation of tissue catabolism (Adams et al., 1996). In clinical cases, organisms may be disseminated within and beyond gastrointestinal tract as evidenced by presence of lesions in liver, kidneys, mammary glands etc.

It is evident from the pathobiology of MAP infection that diagnosis of paratuberculosis in early phase of infection is difficult. A number of diagnostic methods are available, whose sensitivity and specificity vary with the infection prevalence. There is common misunderstanding in minds of most people that diagnostic methods are not available for Johne's disease and that has led to a compromising attitude towards this disease notwithstanding its considerable economic impact. As a result infection has spread to several farms, where disease was absent. Broadly all the diagnostic tests fall in two categories; the one, which is direct and detect organisms or its genomes, the other indirect that detect either cell-or humoral-mediated immune response elicited by the host. Detection of organisms in the clinical samples by culturing with characteristic phenotype is considered a gold standard in paratuberculosis diagnosis and research. It is highly specific but slow method requiring 12–16 weeks to obtain a positive result. An automated radiometric culture method developed recently decreases the incubation period considerably with higher analytical diagnostic sensitivity (Whittington et al., 1998; Manning and Collins, 2001). The sensitivity of the culture method depends on the phase of infection and type of clinical sample used (Tripathi et al., 2002; Sivakumar et al., 2005).

At least three CMI based assays viz., the skin test, lymphocytic stimulation test (LST), and gamma interferon (IFN-γ) test exist. Skin test for delayed hypersensitivity has not been successful in controlling paratuberculosis because of nonspecficity. LST as a measure of in vitro cell mediated response, though sensitive, has been limited to laboratory only because of procedural complexity and variation in results (Milner et al., 1981; Storset et al., 2001; Kurade et al., 2004). The most sensitive and specific measure of CMI is the in vitro assay of IFN-γ following stimulation of peripheral blood T lymphocytes with johnin or avian PPD. A sandwich ELISA for the IFN-γ assays was developed at CSIRO, Australia. Skin test is being replaced with this ELISA. It detects animals before testing positive for serum antibody and before becoming consistently fecal culture positive. However, this test has not been evaluated for its efficacy in Indian cattle.

Antibodies to MAP can be detected by a variety of test including agar gel immunodiffusion (AGID), complement fixation test (CFT) and absorbed ELISA. These tests detect infection during later part of the infection, albeit before appearance of the clinical symptoms. Absorbed ELISA is most widely used and accepted by OIE (Milner et al., 1990; Bech-Nielsen et al., 1992; Vannuffel et al., 1994 ; Rajukumar et al., 2001; Sivakumar, 2003). The assay has overall sensitivity of 40–70% with specificity more than 98% (Collins, 1996). AGID is less sensitive but 100% specific. AGID and ELISA have been used in surveillance programme of paratuberculosis in cattle and sheep. Another ELISA using LAM antigen has been reported to be more sensitive but has not been extensively evaluated (Jark et al., 1997). A number of purified and recombinant antigens have been reported from MAP but their field evaluations have hardly been carried out (Stevenson and Sharp, 1997).

An insertion sequence, IS900, found in multiple copies and unique to MAP has been extensively exploited for the specific diagnosis of paratuberculosis (Green et al., 1989; Vary et al., 1990). IS900 based PCR has been used for identification of mycobacterial isolates and for the diagnosis of Johne's disease in clinical (blood and faecal) and necropsy samples (Stevenson and Sharp, 1997; Sivakumar et al., 2005). PCR and the DNA probe (IS900 based) have not been used for any control programme because of its less sensitivity. Immunomagnetic separation of MAP from the clinical samples followed by IS900 PCR has been developed and found to have increased sensitivity in comparison to conventional PCR (Grant et al., 1998).

Application of diagnostic methods in paratuberculosis depends on the objective of the diagnosis, status of the disease on the farm, location and distance of the farm from the laboratory, cost benefit ratio and involvement of the governmental agencies in the control programme. Thus selection of the tests varies with the situations. AGID and ELISA can achieve confirmation of the clinical diagnosis. However, latter can give false positive results at times. If clinical diagnosis is being confirmed for the first time, faecal samples must be submitted from ELISA positive animals for confirmation by culturing. In such situation more than one test can be adopted to confirm the diagnosis. Greater reliability can be placed on any of the methods on an animal or herd if Johne's disease has been confirmed there (Collins, 1996). In possible circumstances, disease diagnosis must be confirmed by histopathology with demonstration of acid-fast bacilli and tissue PCR targeting at least two genes known to be specific to MAP.

More than 85% of the ruminant population in India lives in villages in small groups and in most instances many species are reared together. Hence, India has a unique animal husbandry structure that is quite different from several other developed countries of the world. Animals also share common grazing pastures in villages. Therefore, each village can be considered a quasi-organised dairy unit, which is distinctly different from organised dairy farm in many respects. The major factors that increase the incidence of the disease at the farm is crowding, more chances of contamination of feed, water and milk by bacteria, and use of common equipments for dairy operations. In villages, most of these factors do not operate by virtue of animal husbandry practiced. It is considered that transmissionof the paratuberculous infection to small dairy farms in urban and suburban areas, and in villages is largely due to introduction of animals from Government farms, where disease is quite prevalent. This scenario of animal husbandry in India logically allow us to suggest that Johne's disease can be controlled better here than in the European countries where most animals are restricted to organised farms. But the demerits of the mixed animal husbandry in rural areas could lead to interspecies transmission of MAP infection. The economic losses due to this terminally fatal disease are significant and it should be borne in the mind that once the disease is established at the farm, the eradication cost is exorbitantly high (see biology of MAP).

Though pathogenesis and epidemiology of paratuberculosis is poorly understood, control and prevention strategies are well described (Kennedy et al., 2002). The basic principles of controlling the disease revolve around three factors; the etiological agent, the host range, and the environment. Petty good information is available on these factors to institute an effective control programme. The magnitude and prevalence of the disease at particular farm and in that region along with the animal husbandry practices followed also play roles in developing strategies for controlling and eradicating paratuberculosis.

The control programme should basically aim to protect the uninfected farms and to control and prevent infections at the infected farms. Non-infected farms are usually infected through introduction of new animals of unknown status. Transfer of animals from one government farm to other is quite common and thus over time several farms due to this practice have been infected increasing the herd/flock prevalence rate in the country. Not only this, the infection has spread to private small dairies in urban, suburban and rural areas due to introduction of animals purchased through auction from governmental farms. This practice needs to be avoided to check spread of the infection. In India, only johnin skin test is available at the field level. The screening of farms may continue using this test until more specific ELISA becomes available for field use. Absorbed ELISA is a cost effective method, which can be used alternatively with the faecal culture for surveillance of the disease. A farm testing negative in three consecutive tests may be declared free of Johne's disease and can be certified to become a source of replacement stock for other farms (Collins, 1996). Most important source of infection at the farm is clinically infected animals and symptomatic carriers, which should be removed by "test and cull" policy. Simultaneously, to break the infection cycle, the young ones are separated from adults.

Farm specific control programme can be developed keeping certain factor in mind such as infection prevalence, interest of the owner, laboratory facilities etc. Some progressive farmers may be enthusiastic to institute control programme on their own. But government must be involved in any control programme to succeed and to last long. Persistence of organisms in the environment including soil, water and faeces and subclinically infected animals are the greatest challenge to any control and eradication programme. Destocking of the infected premises at least for one year is preferred but it is not always possible to make alternative arrangement.

Praratuberculosis is an officially notifiable disease in nine countries. Majority of the countries do not have any currently government sponsored operating control programmes. The reported official disease control measures include test and cull, restriction of movement from infected herd/flock, management of newborn animals, restricted contact between herds and accreditation of negative herds. Norway and Sweden used stamping out by depopulation in the infected herds (Kennedy et al., 2002).

A variety of vaccines are available for paratuberculosis but its use as a method of control has been largely controversial. A heat killed and a live vaccine is generally used. In USA, heat killed vaccine is used only on small scale and its application is decided on case-by-case basis. Netherlands also uses heat-killed vaccine. In some other Euoropean countries live attenuated vaccine is used. These vaccine results in substantial reduction of incidence of clinical disease, number of organisms shed in the faeces and economic losses due to culling of animals. The vaccine, however, does not eliminate or prevent infection because it does not offer an absolute resistance to MAP. The other disadvantage of vaccination is that it interferes with serological testing for tuberculosis and paratuberculosis. Vaccination can be recommended at farms with high prevalence of infection. Several laboratories are currently working on recombinant proteins specific for MAP, which could be used for vaccine purpose. Early secreted proteins from other mycobacteria (M.tuberculosis) have been reported to be protective (Andersen, 1994) and it would be worth to look for similar proteins from MAP for their potential use as subunit vaccine.

There has been considerable research and discussion on association of MAP with Crohn's disease (CD) in humans ever since the disease was first identified in 1913 (Thompson, 1994; Greenstein, 2003; Chacon et al., 2004). It is mostly due to similarities between Crohn's disease and Johne's disease on microbiological, immunological and pathological grounds. Both the diseases have similar chronic granulomatous lesions mostly concentrated in the terminal part of ileum. Crohn's disease is histologically characterised by non-caseating granuloma consisting of loose aggregates of epithelioid cells, Langhans giant cells, and lymphocytes, affecting all layers from mucosa to serosa (Wakefield et al., 1991). Mesenteric lymph nodes and lymphatics are often affected. Ulceration and necrosis are also seen. Intestinal tuberculosis resembles CD in respect of intestinal ulceration and fibrosis, but the presence of tubercles with caseation and demonstration of M. tuberculosis usually distinguish tuberculous cases from CD (Clarke, 1997).

Certain facts that argued against the involvement of MAP in the etiology of Crohn's disease were the isolation of several species of mycobacteria including MAP and failure of the special stains to demonstrate organisms in the infected tissues (Collins, 2003). Other pathological features such as fibrosis, fissures, fistulas, abscesses, bowel loop adhesions, blood in stools and fibrous thickening of mesentry observed in Crohn's disease are not found in Johne's disease (Clarke, 1997).

M.a. paratuberculosis occurs in two forms; ‘bacillary’ and the ‘spheroplast’. Many paratuberculous bacteria of the bacillary form may be required to cause clinical disease, whereas only a few paratuberculosis bacteria of spheroplast form will cause the disease (Graham et al., 1987; Gitnick et al., 1989). The difference between diseases caused by two forms of bacteria is due to immune reaction of the infected host. Cell wall deficient spheroplast like organisms have been detected by electron microscopy in paucibacillary lesions of cattle with subclinical paratuberculosis (Condron et al., 1994). In addition, clinical cases of paratuberculosis without demonstrable AFB but with tuberculoid lesions and immunological evidence of the disease have been reported in sheep (Clarke and Little, 1996; Perez et al., 1996). Further, an isolate with characteristics of MAP from a clinical case of CD produced typical paratuberculous lesions in goats experimentally (Van Kruiningen et al., 1986).

A genetic material, the IS900 insertion sequence that is unique to MAP is found in both the bacillary and spheroplast forms. In more than 70 per cent of the intestinal tissues of the Crohn's disease, IS900 sequence has been demonstrated (Moss et al., 1992; Fidler et al., 1994; Suenega et al., 1995; Collins et al., 2000). Recently, a report of cervical lymphadenitis followed five years later by terminal ileitis similar to Crohn's disease in a young boy has provided circumstantial evidence of its association with paratuberculosis. Due to advent of in-situ hybridisation staining technique, it was possible to demonstrate cell wall deficient MAP in 35 of 48 (73%) of Crohn's patients (Sechi et al., 2001. More recently, Bull et al. (2003) reported isolation of MAP from 14 of 33 (42%) Crohn's patients compared with 3 of 33 (9%) of non-inflammatory bowel disease controls. With the help of IS900 probe, genetic materials are being demonstrated more consistently in the biopsy materials of bowel than the living organisms in the culture. A commercial ELISA for bovine adapted to detection of antibody to MAP in Crohn's patients revealed statistically significant number of positive cases in comparison to the control human subjects (Collins et al., 2000).

The possible source of infection to humans is mainly the food chain. These include milk, milk products, meat and contaminated water (Grant et al., 1998, 1999; Millar et al., 1996). Milk is the major source of nutrition for children, and if pathogenesis of Crohn's disease resembles paratuberculosis, children are more susceptible to infection than the adults. Also complete killing of the MAP by pasteurisation has been questioned in several studies (Sung and Collins, 1998; Grant et al., 1999). In the United Kingdom and Ireland, milk from JD infected cows is not fed to calves and such milk is banned for human consumption. Milk products are also not made from the infected milk. Animal products made from cattle culled due to clinical signs of JD may be infected. Water is a common source of M.a. avium and MAP has been detected in domestic water on at least one occasion (Mishina et al., 1996). While still more research is required to confirm the zoonotic status of MAP, all precautions from veterinary-public health point of view must be taken to prevent the transmission of infection to humans before it establishes firmly in the human population too like animals.

Due to complex pathogenesis and epidemiology of MAP infection, diagnosis and control is difficult. Ecology and characteristic of the bacterium is such that normal microbiological procedures and interventions are ineffective against MAP. Greater appreciation of the mechanism of persistence and spread of MAP in animal population, spectrum of clinical, immunological and pathological features and performance of different diagnostic assays in different stages of infection will lead to the development of diagnostic and control strategies. A number of diagnostic tests are available that can be used judiciously for diagnosis. Due to expansion of the host range, strain differentiation will be another important issue to be addressed for epidemiological studies. This has now become plausible with the advent of molecular techniques such as RFLP, PFGE, RAPD and PCR. In Indian situation a low cost field adaptable test should be developed to suit veterinarians. PCR and DNA probes need further improvements, as these tests are yet to give desired efficacy particularly because of inadequate DNA extraction procedures and presence of PCR inhibitors in the samples. In view of the pathobiology of MAP infection, a definitive test may not be immediately available in the near future. But discovery of 21 species specific genes in MAP by comparative genomics in recently completely sequenced MAP genome has raised hope for development of new diagnostic assays endowed with desired efficacy. It is, therefore, imperative to accept reality of paratuberculosis and ensure fullest use of available technology in the diagnosis, control and prevention rather than waiting for technological revolution leading to development of a perfect test.

References