Background
Johne's disease caused by
Mycobacterium avium subsp.
paratuberculosis (MAP), is one of the most widespread and economically important disease of ruminants. It is a chronic granulomatous enteritis affecting primarily ruminants and many other species [
1], which is characterised by persistent diarrhoea, weight loss and a protein enteropathy, followed eventually by death [
2]. Most cattle are infected early in life by the ingestion of faeces, milk or MAP contaminated water. The relatively long incubation period is characterized by the excretion of MAP in faeces for months and years before clinical symptoms develop [
3]. The exposure to contaminated faeces constitutes one of the main risk factors for MAP transmission within the herd.
Johne's disease causes worldwide economic losses to farmers and dairy industries in terms of milk and meat production. It is considered a serious disease for dairy cattle as there is no effective treatment and it's control is difficult due to the long latent period. In dairy herds, losses are associated with reduced milk yield and weight gain, lower reproductive efficiency, premature culling and reduced values of culled cattle [
4]. The effect of paratuberculosis on dairy operations in the USA was estimated at around $200 to $250 million a year [
5].
Paratuberculosis has been a scheduled and notifiable disease in Ireland since 1955. Prior to the 1990 s, the low number of notified clinical cases (only 92 diagnosed between 1932 to 1992, mostly in imported animals) indicated that MAP wasn't widely established in the country. The importation of 85,000 cattle from continental Europe between 1992 and 2004 (Central Statistics Office, personal communication) as a consequence of the opening of the single European market in 1992 coincided with an increase in the prevalence of MAP infection in Ireland. Recent studies have shown that the risk of MAP exposure and transmission increases annually [
6]. Between 1995 and 2002, 232 animals infected with MAP among 106 Irish herds were reported [
7]. Also, in 2005, the seroprevalence of infected herds in Ireland was found to be 21.4% [
8]. Currently, the prevalence of paratuberculosis among herds in Ireland is lower than that reported in many countries in Europe (Denmark 55%, France 68% and Netherlands 54%) [
9] probably because of the late introduction and establishment of JD in Ireland. However it is likely that the prevalence in Ireland will continue to rise to match rates seen elsewhere, unless appropriate preventive and control measures are taken.
Similarities between Johne's disease in ruminants and Crohn's Disease (CD) in humans [
10] as well as studies which identified MAP in CD's patient [
11] have led to speculation that MAP may be a aetiological agent in Crohn's disease. Many reports showed the pathogenesis of CD is complex and multi-factorial with genetic and environmental contributions [
12,
13]. The causative link between MAP and CD is still controversial and the zoonotic potential of MAP remains a subject of debate [
14]. The identification of viable MAP in pasteurized milk [
15,
16] and meat [
17] should be regarded as a significant issue in terms of bio-security.
Detecting the presence of MAP is difficult because of the slow growth and the lack of sensitive tests to identify subclinically infected cattle. Specific and sensitive diagnostic tools as well as a better understanding of the pathogenesis of JD are needed to develop control programs to eradicate the disease.
This study was carried out on bovine faecal samples from 7 Irish herds which either had a history of or a likely exposure to MAP within 2 years of this study. The aim of this project was to evaluate a diagnostic strategy for MAP in targeted Irish herds. Specifically, we set out to compare a conventional culturing method with PCR, to evaluate the reliability of 2 different molecular targets and finally, to assess the improved sensitivity of a nested PCR assay.
Discussion
Due to the late introduction and establishment of JD in Ireland, there are relatively few publications on Paratuberculosis in this country. This study set out to evaluate a diagnostic strategy for MAP in selected Irish herds by comparing culture and molecular assays. This has relevance in terms of implementing future screening strategies for vets in Ireland and elsewhere. It is important to state that due to the logistical problems associated with obtaining samples, and the difficulty involved in enlisting willing farmers to participate, it was impossible to standardise the testing among herds in terms of obtaining equal samples from each herd. However, given this shortcoming, some important observations and conclusions can be made at the individual animal level and within each herd.
In general, we found a poor correlation between culture results from individual animals (7.9%) and the corresponding first round PCR results (36%). This may be due to the degree to which an animal was shedding the organism, the current disease status of the animal at the time of sampling, or the significant loss of viable cells during the harsh de-contamination step. Collectively however, it was interesting to note that all herds that tested positive by culture were also PCR positive (H1, H2, H3 and H4) whereas the other 3 herds were negative using both methods. This is significant in terms of how PCR may be used to manage and detect Johne's disease rapidly within a herd, i.e a quick and relatively cheap PCR test from a representative number of animals could provide an early indication of the disease status of the overall herd.
Two herds were sampled twice in this study (H1 and H2) and both produced interesting results. For example, in the herd H1, the percentage of MAP culture-positive results varied from 75% in 2006 to 30% in 2007 (although the sample size was not comparable). In terms of Herd H2, the difference between culture and PCR results in 2007 and the same herd one year later was striking (n = 44 and n = 0). This may be explained by the culling of infected animals and the possible introduction of "JD free" cattle in 2008, as seen previously by Richardson et al. (2009)[
25]. It may also be explained by the selection of an alternative cohort of animals by the vets during the second visit. Due to the sensitivities involved in dealing with farmers involved in this voluntary study, limited information was made available to us regarding the degree to which the herd had been re-populated during the intervening year.
There is a significant lack of comparable studies in Ireland, and although the purpose of this project was not to draw conclusions regarding the prevalence of MAP in Ireland, it is interesting to note that our % recovery is almost double that seen by a similar study in 2002 (O'Doherty et al. in 2002[
26]. Using a similar experimental strategy this group tested 221 animals from 16 herds suspected of paratuberculosis infection and found a prevalence rate of 4.1%.
As seen in other studies, the difference between the culture results in two dairy herds (H2, 13% and H3, 14%) and a beef herd (H4, 3%) was significant. The common practice in Ireland for dairy herd managers to feed pooled colostrums and milk to calves relatively increases the risk of transmission through contaminated milk which is a significant risk factor [
27].
In terms of PCR and the reliability of molecular testing, a complete correlation was observed between the results of the IS900 and ISMAP02 targets. This was surprising but not unusual as observed in a similar study [
23].
The nested ISMAP02 PCR detected 29 further positive samples increasing the sensitivity of the assay by 10%. However, among these additional samples, when re-examined, 8 from herd H4 were found negative by 2 other specific nested PCRs (IS900 and F57) based on independent triplicate results. The presence of these false positive test results may come from a single assay problem for H4, or may be representative of a larger reliability issue with nested PCR in this context. As well as the 8 putative false negatives, 10 of the remaining extra positives came from herds H2 (2008), H5 and H7 which were negative by both culture and first round PCR (Table
2). This may of course represent an excellent added sensitivity afforded by nested PCR for these 10 samples, or it could be indicative of potential problems with this extra test. We feel that the need for increased sensitivity must therefore be balanced with the risk of producing false positives among the results.
The presence of environmental mycobacteria (
Mycobacterium non-chromogenicum and
Mycobacterium terrae) within certain herds is another important finding although how it impacts on an animal's susceptibility to MAP infection is hard to assess (if at all). These mycobacteria belong to the M. terrae complex and were originally classified as non-pathogenic saprophytes, but their pathogenic status is changing. They have been isolated from soil and are usually present in clinical samples as an environmental contaminant.
Mycobacterium nonchromogenicum has also been isolated in cattle infected by bovine tuberculosis [
28]. Interestingly, the 18 samples from which
Mycobacterium non-chromogenicum and
Mycobacterium terrae have been isolated were PCR negative for MAP DNA which indicates that these two insertion sequence IS900 and ISMAP02 are not present in these mycobacteria.
Infected animals shed MAP in milk but also in the environment where MAP persists and survives in the soil, water and sediment [
29]. To minimise exposure of the human population as described previously by Hermon Taylor (2010), there is a need to understand all the potential reservoirs and all the transmission routes for MAP. If this pathogen is proven to be zoonotic, the implication for the dairy industry worldwide would be enormous [
30] especially in countries like Ireland where exported milk and dairy products are significant for the national economy. It follows therefore that all countries with a paratuberculosis related problem should be working towards a rapid, standardised and reliable indicator of infection. Based on our experience, PCR should play a significant role in this.
Conclusion
In this study culturing of MAP was characterised by poor recovery (7.9%) but high specificity. In contrast, the PCR produced faster results with improved sensitivity (36%). The correlation between the culture and PCR result was poor at the individual animal level but in complete agreement at the herd level. Based on our experience, the choice of test employed by veterinarians should be guided by their objectives; a PCR test will produce a rapid result which may act as an "early warning" for the disease status of the herd especially if multiple animals are tested concurrently and periodically. As seen with other studies, IS900 and ISMAP02 are equally reliable as DNA targets for MAP, and the added sensitivity afforded by a nested PCR must be balanced with the potential for introducing false positive results. Despite the obvious advantages of introducing routine molecular testing for potentially infected animals, the culturing method (liquid or solid based) still remains the gold standard despite its logistical and practical limitations.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
PED carried out the culturing, molecular analysis and drafting of the manuscript. WC and JB were the veterinarians responsible for co-ordinating the sampling, collection and transport of the faecal samples. AC and JO'M are the principal investigators and grant awardees who designed and managed the study, as well as editing the manuscript.
All authors read and approved the final manuscript.