Comparison of four rapid DNA extraction techniques for conventional polymerase chain reaction testing of three Mycobacterium spp. that affect birds

Mycobacteriosis is an avian disease that is most commonly caused by Mycobacterium avium or Mycobacterium genavense. In order to optimize molecular laboratory tests for diagnosing mycobacteriosis in birds, we compared four methods of rapid DNA extraction with isolates of M. avium, M. genavense, and Mycobacterium fortuitum. DNA extraction methods included enzymatic lysis, boiling for 30 min followed by enzymatic lysis, four cycles of freezing and thawing followed by enzymatic lysis, and bead beating followed by enzymatic lysis. The DNA yield and purity for the four methods were evaluated by spectrophotometry and compared. The bead beating with enzymatic lysis technique yielded significantly purer and higher concentrations of extracted DNA compared with other DNA extraction methods. All four methods yielded extraction products for all three organisms that were successfully amplified by polymerase chain reaction (PCR) for a fragment in the 65-kD heat shock protein gene. Subjectively, the PCR amplification products were most abundant for samples extracted by bead beating with enzymatic lysis.     

Estimation of the spread of pathogenic mycobacteria in organic broiler farms by the polymerase chain reaction

Organic poultry breeding allows for increased exposure of birds to soil, faeces, and wildlife, which have been associated with the transmission of mycobacterial infections. Therefore the aim of this study was to investigate the spread of the major pathogenic mycobacteria in organically reared broilers in Greece using a diagnostic algorithm that relied on a combination of the polymerase chain reaction (PCR) and the restriction fragment length polymorphism analysis (RFLP). Liver, spleen and gonads from 81 to 150 days old broilers were aseptically collected post-mortem. 500 broilers from a population of 35,370, reared in the 25 registered as organic farms in Greece for the 2005 were used. DNA was isolated and incorporated to PCR targeted to 16S-rRNA gene (for Mycobacterium spp.), IS6110 (for Mycobacterium tuberculosis complex-MTBc), IS1245 (for Mycobacterium avium complex-MAC), IS901 (for M. avium subsp. avium-MAA) and hsp65 (for Mycobacterium genavense, by PCR-RFLP). The mean prevalence of mycobacteria detected by PCR with a 95% confidence interval was estimated to 4.4-8.8%. The relevant percentage with regard to the mycobacterial species that were included in this study was 0.17-2.03% for MAC, 2.11-3.39% for MTBc and 0.66-3.08% for mycobacteria not belonging to any of the above groups. None of the mycobacteria detected were identified as MAA or M. genavense. Considering that avian tuberculosis has been eradicated from conventional farms, the level and the pattern of positivity recorded here, indicates that our results may be associated with the specific conditions that apply to organic breeding.     

Outbreak of Mycobacterium avium subsp. avium infection in one flock of domestic pigeons

An outbreak of Mycobacterium avium subsp. avium infection was diagnosed in one breed of domestic pigeons (Columba livia f. domestica) in the Czech Republic. Nodular granulomatous lesions were found in 42 (9.7%) pigeons of the 435 examined; histopathologic examination of livers with gross lesions of mycobacteriosis from 15 randomly selected pigeons revealed granulomatous inflammation typical for avian mycobacteriosis in all samples. Direct Ziehl-Neelsen (ZN) microscopy and conventional culture were performed for a total of 117 liver samples (42 pigeons with nodular lesions, 55 randomly selected pigeons without nodular lesions, and 20 randomly selected squabs). Acid-fast bacilli were observed in 19 (16.2%), and conventional culture yielded growth of M. a. avium in 40 (34.2%) liver samples. A triplex quantitative real-time PCR assay based on the IS901 detection system was performed successfully in 115 liver samples and revealed M. a. avium in 63 (54.8%) of them. Mycobacterium a. avium was also detected in two squabs. Eight domestic rabbits (Oryctolagus cuniculus f. domestica) living in the breeding facility were also examined. Pyogranulomatous lesions were only found in one adult male rabbit. At necropsy, both direct ZN microscopy and culture gave negative results for mycobacteria in all examined rabbit tissues. Mycobacterium a. avium was diagnosed in a liver sample of one juvenile rabbit using triplex qPCR, suggesting that M. a. avium infection can occur as early as juvenile animals.     

Avian mycobacteriosis in companion birds: 20-year survey

The causative agents of avian mycobacteriosis in pet birds are rarely identified. The aim of this study is to add information about the etiology of avian mycobacteriosis. The identification of mycobacterium species in 27 cases of avian mycobacteriosis in pet birds was investigated by polymerase chain reaction (PCR) and sequencing of a rRNA hypervariable region. Avian mycobacteriosis appeared to be an infrequent diagnosis. Interestingly, a few cases of avian mycobacteriosis were recorded in very young birds. The most commonly affected species were the canary (Serinus canarius), the Eurasian goldfinch (Carduelis carduelis) and the red siskin (Spinus cucullatus). All but one bird were infected with Mycobacterium genavense. Mycobacterium avium was identified only in one case.     

Molecular epidemiology of Mycobacterium avium subsp. avium and Mycobacterium intracellulare in captive birds

Mycobacterium avium subsp. avium and Mycobacterium intracellulare are primary causes of mycobacteriosis in captive birds throughout the world, but little is known about how they are transmitted. To define the local epidemiology of infection, we strain-typed 70 M. avium subsp. avium and 15 M. intracellulare culture isolates obtained over a 4-year period from captive birds. Typing was performed using randomly amplified polymorphic DNA (RAPD) PCR, amplified fragment length polymorphic (AFLP) fragment analyses, and for a subset of isolates, DNA sequencing of a segment of the 16S-23S rRNA internal transcribed spacer region. Six strain clusters comprising 43 M. avium subsp. avium, isolates were identified; 42 isolates had unique typing patterns, including all M. intracellulare isolates. Phylo-geographical analyses using RAPD and AFLP fingerprints and animal confinement histories showed no correlation between housing of infected birds and mycobacterial strain-type, except for two animals. The diversity of M. avium subsp. avium and M. intracellulare isolates and minimal evidence for bird-to-bird transmission suggest that environmental reservoirs may be important sources of infection in captivity.     

HPLC does not differentiate Mycobacterium paratuberculosis from Mycobacterium avium

HPLC, which is gaining its place as identification tool in mycobacteriology laboratories, has been proposed to distinguish Mycobacterium paratuberculosis from Mycobacterium avium. We had reported no significant difference between M. avium and M. paratuberculosis reference strain ATCC 19698. Because of the advantages offered by such a method, we enlarged our observations to include more isolates of M. paratuberculosis. Within the double cluster of peaks obtained by both M. avium and M. paratuberculosis, we could not find a consistent difference typical of M. paratuberculosis. Therefore, the present study confirmed that M. avium and M. paratuberculosis could not be distinguished by HPLC, raising doubts of a straightforward use of HPLC to identify M. paratuberculosis.     

Granulomatous dermatitis caused by Mycobacterium genavense in two psittacine birds

Two cases of cutaneous mycobacteriosis in psittacine birds showing featherless, non-painful, non-pruritic nodules are described. Histopathological studies of skin biopsies from both cases demonstrated the presence of a diffuse granulomatous dermatitis with acid- fast organism s inside macrophages, which led to the diagnosis of cutaneous mycobacteriosis. In one case, generalization of the process to internal organs (intestinal and hepatic serosae) was observed. Mycobacterial organisms could not be cultured using conventional isolation media (Coletsos and Löwenstein–Jensen), but polymerase chain reaction (PCR) technique performed on pathological samples from both birds revealed the presence of Mycobacterium genavense . It is thus proposed that cutaneous mycobacteriosis infections, in particular those caused by M . genavense , should be included in the differential diagnosis of skin nodular processes in psittacine birds. The usefulness of PCR techniques for aetiological diagnosis of mycobacterial infections is emphasized.     

High genetic relatedness among Mycobacterium avium strains isolated from pigs and humans revealed by comparative IS1245 RFLP analysis

Members of the Mycobacterium avium complex cause pig mycobacteriosis and opportunistic human infections. Infections due to environmental mycobacteria are increasing in both industrial and developing countries. Mycobacterium-infected pig carcasses can pass for human consumption due to the poor specificity of meat control by visual detection at the slaughter houses. The genetic relatedness of porcine and human MAC isolates in Finland has been unknown. M. avium isolates isolated from pig organs (n=16) and clinical samples (n=13) were compared by IS1245 RFLP analysis to evaluate the similarity of the isolates obtained from human and porcine samples. Nearly identical multicopy M. avium subsp. hominissuis IS1245 RFLP fingerprints were obtained for isolates of porcine and human origin. IS1245 RFLP patterns of 38% of the porcine and human M. a. hominissuis isolates were >90% similar. The RFLP patterns of two porcine and two human isolates showed >95% similarity. The high similarity of the IS1245 RFLP patterns of the human and porcine M. a. hominissuis isolates indicates close genetic relatedness, suggesting that M. a. hominissuis is transmitted between pigs and humans, or that pigs and humans share common environmental sources of infection. Porcine and human isolates with RFLP patterns differing by only one or two bands were found, which shows that the same M. a. hominissuis strains may infect both humans and pigs.     

Diagnosis of Mycobacterium bovis infection in calves sensitized by mycobacteria of the avium/intracellulare group

Because of the frequent exposure of cattle to mycobacteria of the avium/intracellulare group, an investigation was carried out into the possible repercussions thereof on the diagnosis of bovine tuberculosis. Three calves from a bovine tuberculosis-free herd, scored avian reactors in the gamma-interferon assay for bovine tuberculosis, were sedated and inoculated endotracheally with a virulent Mycobacterium bovis strain. Then, three other avian reactors were housed with the above donor calves. Mycobacterium bovis was isolated from the nasal swabs of the three endotracheally infected, donor calves. On these samples, TB complex-specific polymerase chain reaction (PCR) tests for IS6110 were also positive, albeit with a different time kinetics. The three contact-infected calves showed clear immunological signs of infection; however, their nasal swabs were always PCR-negative and only Mycobacterium avium was isolated. In the endotracheally infected donor calves there was a rise of the gamma-interferon responses to avian and bovine purified protein derivative (PPD) tuberculins, which reached the same stable plateau levels over the whole experiment. The above effect was also observed in the contact-infected calves, even though the response to avian PPD tuberculin always remained at a higher level. By using conventional bovine and avian PPD tuberculins, the comparative intradermal test was generally positive in endotracheally infected, as opposed to contact-infected calves; a positive intradermal test for M. bovis was obtained in two contact-infected calves by different bovine PPD tuberculins based on M. bovis bacillus Calmette-Guerin (BCG) secreted or somatic antigens. It was concluded that M. bovis infection may be concealed for some time in cattle sensitized by mycobacteria of the avium/intracellulare group and that different diagnostic procedures should be adopted for such animals.     

Ante mortem diagnosis of mycobacterial infection by liver biopsy in a budgerigar (Melopsittacus undulatus)

Avian mycobacteriosis is an important disease which affects exotic, wild and domestic birds. The disease is most commonly caused by the ubiquitous soil and water organisms Mycobacterium avium sp. avium and Mycobacterium genavense. Mycobacteriosis should be considered in the differential diagnosis of any pet bird with chronic disease characterized by weight loss, an inflammatory leukogram and abdominal enlargement.Ante mortem diagnosis of avian mycobacteriosis in birds remains difficult. The present case report describes that even in very small species such as a budgerigar the investigating of a biopsy sample is an efficient way to rule out or confirm the disease in cases where other methods fail to be conclusive.     

Study of Mycobacterium avium complex strains isolated from cattle in the Czech Republic between 1996 and 2000

This study surveys 2,593,348 cattle slaughtered between 1996 and 2000, and further investigates 571 (0.02%) animals found to have tuberculous lesions. Culture of 346 randomly selected tissue samples from animals younger (n = 215) and older (n = 131) than 2 years, isolated mycobacteria from 91 animals (26.3%). These included 74 Mycobacterium avium subsp. avium isolates of IS901+ and IS1245+ genotype and serotype 2, 13M. avium subsp. hominissuis isolates of IS901- and IS1245+ genotype and serotypes 8 (n = 7) and 4 (n = 6), two M. chelonae, one M. avium subsp. paratuberculosis (RFLP type B-C1), and one M. terrae. Culture of mesenteric lymph node samples obtained 66 isolates of M. avium complex (MAC) and four isolates of other mycobacterial species. M. bovis was significantly absent from all samples. Mycobacteria were more frequently (P = 0.01) isolated from tissues of animals under 2 years (34.4%) than animals over 2 years (13.0%). IS901 and IS1245 RFLP methods were used to type 17 randomly selected MAC isolates, virulent after intramuscular inoculation of pullets, from 17 different cattle herds. These revealed 11 distinct IS901 RFLP types and three IS1245 RFLP profiles. Polyclonal infection of individual animals was detected by IS901/IS1245 typing in 2 of the 17 selected isolates.     

IS900/ERIC-PCR as a tool to distinguish Mycobacterium avium subsp. paratuberculosis from closely related mycobacteria

There is an increasing demand for fast and reliable methods to distinguish Mycobacterium avium subsp. paratuberculosis (M. paratuberculosis) from closely related mycobacteria and also a need for rapid strain specific typing of clinical isolates for epidemiological reasons. In the present study, the potential of rep-PCR as a fingerprinting method for M. paratuberculosis was assessed and compared to conventional RFLP. A PCR assay was designed and optimised to obtain reproducible fingerprints of mycobacterial DNA with primers targeting the enterobacterial intergenic consensus (ERIC) sequence and the M. paratuberculosis specific insertion sequence IS900. Reproducible fingerprints were obtained with 60 strains of M. paratuberculosis, 16 strains of M. avium subsp. avium, 3 strains of M. intracellulare, and 11 other mycobacterial strains. A species-specific band pattern that was clearly distinguishable from that of other mycobacteria was obtained with M. paratuberculosis. The rep-PCR did not detect any differences among M. paratuberculosis strains of different RFLP types, and was therefore not considered as an alternative fingerprinting method. However, the species-specific band pattern make IS900/ERIC-PCR a suitable alternative for distinguishing M. paratuberculosis from other mycobacteria, especially in cases of IS900 PCR positive mycobacteria. The fingerprinting method reported was fast and easy to perform, and produced highly reproducible results.     

Disseminated Mycobacterium genavense infection in a FIV-positive cat

An 8-year-old FIV-positive Australian cat was presented with coughing, periocular alopecia, pyrexia and inappetence. Skin scrapings demonstrated Demodex cati mites. Antibiotics were administered and it was treated successfully for periocular demodectic mange, but the cat continued to exhibit respiratory signs and lose weight. Further investigation revealed an ascarid infection and active chronic inflammation of undetected cause affecting the lower airways. Repetitive treatment with pyrantel failed to eradicate the ascarid infection. The cat became cachectic and developed moist ulcerative dermatitis of the neck, severe non-regenerative anaemia, leucopenia and thrombocytopenia. Necropsy and histopathology revealed mycobacteriosis affecting skin, lungs, spleen, lymph nodes, liver and kidney. Attempted culture of frozen tissues at a mycobacteria reference laboratory was unsuccessful. Paraffin-embedded, formalin-fixed tissue was retrieved and examined using PCR to amplify part of the 16S rRNA gene. A diagnosis of disseminated Mycobacterium genavense infection was made based on the presence of acid fast bacteria in many tissues and partial sequence of the 16S rRNA gene. Although M genavense has been identified previously as a cause of disseminated disease in AIDS patients, this is the first report of infection in a cat. It was suspected that the demodecosis, recurrent ascarid infections and disseminated M genavense infection resulted from an immune deficiency syndrome consequent to longstanding FIV infection.     

Phenotypic and genotypic characteristics of Mycobacterium isolates from fighting fish Betta spp. in Malaysia

Mycobacteriosis due to mycobacteria is one of the most common bacterial diseases in ornamental fish. We describe here the phenotypic and genotypic characteristics of Mycobacterium isolates from fighting fish Betta spp. using ATCC Mycobacterium marinum, Mycobacterium fortuitum and Mycobacterium chelonae as references. A total of four isolates (M1, M2, M3, M4) were obtained from four out of 106 fish samples using selective agar, and identified to Mycobacterium genus using acid-fast staining and 16s rRNA gene-based genus specific polymerase chain reaction. DNA sequencing and NCBI-BLAST analysis further identified isolate M1 as M. marinum and isolates M2, M3, M4 as M. fortuitum. Morphological, physiological and biochemical tests were carried out for phenotypic characterizations. Universal M13 and wild-type phage M13 RAPD dendogram was generated to illustrate the genetic relationship of the isolates and reference strains.     

Polymerase chain reaction identification of Mycobacterium avium in formalin-fixed, paraffin-embedded animal tissues.

A PCR procedure previously developed for identification of Mycobacterium bovis in formalin-fixed tissues was used to identify mycobacteria of the M. avium complex. Tissues were examined from 100 culture-positive cases of M. avium complex infection, including 86 in which the subspecies was not identified and 14 that had been identified as M. avium subsp. paratuberculosis. Each sample was tested with 5 primer sets, 16S ribosomal RNA (rRNA), IS900, IS901, IS1245, and a heat shock protein (hspX), that detect 1 or both M. avium subspecies. The success rate of PCR detection varied with the primers used and the animal species tested. Among the 86 cases with no M. avium subspecies designation, primers for the 16S rRNA gene were clearly the most efficient because they produced amplicons from all samples that reacted with any other primer set. The overall detection rate in this group of samples was 71%: highest in avian tissues (89%) followed by swine (72%) and ruminants (57%) None of the avian or swine tissues reacted with primers for IS900 or hspX, which identify M. a. paratuberculosis. In contrast, 7 of the 12 ruminant samples that were 16S rRNA positive reacted with 1 or both of these primers. All of the 14 cases shown by culture to be M. a. paratuberculosis infections were positive with IS900 primers, whereas only 11 were positive for 16S rRNA. These results indicate that 16S rRNA primers are the most useful for PCR identification of M. avium in formalin-fixed tissues of nonruminant species. However, IS900 primers should also be used when ruminant tissues are examined because these primers provide the greatest sensitivity for detection of M. a. paratuberculosis infections.     

Isolation of Mycobacterium bovis and other mycobacterial species from ferrets and stoats

As part of wildlife surveillance for bovine tuberculosis, pooled lymph nodes from 21,481 ferrets, 1056 stoats and 83 weasels were cultured for mycobacteria. A total of 268 isolates of Mycobacterium bovis were obtained from ferrets, 2 from stoats and none from weasels, demonstrating the presence of a wildlife reservoir of infection in ferrets. DNA typing by restriction endonuclease analysis (REA) of 48 selected isolates of M. bovis revealed 23 REA types. Twenty-one of these types had previously been isolated from cattle and farmed deer, demonstrating a complex cycle of infection involving wildlife and domestic animals. Apart from M. bovis, a further 208 mycobacterial isolates were obtained, the majority of which (178) were members of the M. avium complex. Speciation of the remaining 30 mycobacterial isolates by DNA sequencing of the 16s rRNA gene, identified half the isolates as M. triplex. Other species identified included M. fortuitum, M. florentinum, M. interjectum, M. intracellulare, M. holsaticum, and M. septicum/M. peregrinum.     

Diagnosis of avian mycobacteriosis: comparison of culture,acid-fast stains,and polymerase chain reaction for the identification of Mycobacterium avium in experimentally inoculated Japanese quail (Coturnix coturnix japonica)

In this study we compared culture, acid-fast stains, and polymerase chain reaction (PCR) for the detection of acid-fast organisms in fecal and tissue samples from Japanese quail (Coturnix coturnix japonica) that were experimentally inoculated intravenously with Mycobacterium avium. For culture, three different culture media (modified Herrold egg yolk with mycobactin; Lowenstein-Jensen [L-J]; and L-J with cyclohexamide, naladixic acid, and lincomycin) were tested to determine which medium had the greatest success in isolating mycobacteria. Acid-fast staining methods included Zichl-Neelsen (Z-N) and Truant. The PCR assay detected mycobacterial DNA with primers specific for the 65-kD heat shock protein gene. Culture was considered the "gold standard." Compared with other culture media, L-J yielded more positive cultures and greater numbers of colonies on positive tubes, and incubation times were shorter. Mycobacterium avium was isolated from all of the harvested tissue samples (liver, spleen, and intestine) of inoculated birds. Mycobacteria were isolated from 53% (69/130) of fecal samples from inoculated birds. As the disease advanced, fecal culture was positive on more culture days, indicating that the culture-positive rate was higher later in the course of the disease. Compared with culture, all of the laboratory methods had 100% specificity for the tissue samples. Sensitivities for the tissue samples were 82.6% (Z-N), 95.7% (Truant), and 100% (PCR). For the fecal samples, the specificity was >95% for all methods. Sensitivities compared with fecal culture were 7.2% (Z-N), 30.4% (Truant), and 20.3% (PCR). Tissue and fecal samples from the two control birds were negative for acid-fast organisms by any method. These results were comparable with clinical cases of avian mycobacteriosis where culture and PCR of tissue samples seem to be the most sensitive and specific laboratory tests and evaluation of fecal samples still remains challenging. On the basis of the results of this study, identification of mycobacteria in fecal samples from Japanese quail can be optimized by repeated cultures and Truant acid-fast staining of fecal smears.     

Isolation and characterisation of Mycobacterium avium and Rhodococcus equi from granulomatous lesions of swine lymph nodes in Slovenia

Granulomatous lesions in bovine and especially swine lymph nodes are still frequently observed during routine veterinary meat inspections even though Mycobacterium bovis infections are no longer detected in domestic animals in Slovenia. Different lymph nodes of pigs (n = 260) were investigated using classical bacteriological and molecular methods. Mycobacterium avium alone was isolated in 47.3% of pigs and in mixed infection with Rhodococcus equi in 3.9% of pigs. R. equi alone was isolated in 27.3% and in mixed infection with mycobacteria other than M. avium in 1.5% of pigs. A total of 133 M. avium isolates were typed using the IS1245, IS901 and FR300 PCR. Almost two thirds (60.9%) of isolates belonged to M. avium hominissuis (IS901-, IS1245+ genotype), 33.8% of isolates belonged to M. avium avium (IS901+, IS1245+ genotype) and 5.3% of isolates remained non-typed. Fifty out of 85 R. equi isolates were tested for the virulence-associated antigens (VapA and VapB). Nearly two thirds (60.0%) were positive for VapB while all the other isolates were VapA- and VapB-negative.     

Phenotypic characteristics and virulence genotypes of Trueperella (Arcanobacterium) pyogenes strains isolated from European bison (Bison bonasus)

In a breeding and fattening pig farm an increasing number of cases of abortion and generalized mycobacteriosis at slaughter occurred. Pathological findings compatible with mycobacteriosis, acid-fast organisms in tissues, and isolation of mycobacteria from tissue samples including fetuses, lungs and reproductive organs from sows, genital swabs, mesenteric lymph nodes, and from a sperm sample revealed the cause of the disease. Bacterial cultures were identified as Mycobacterium avium subsp. hominissuis using IS901-/IS1245-specific PCR. Genotyping of selected isolates from animals as well as from their environment by MIRU-VNTR analysis showed that the herd was infected with one single outbreak strain. The same genotype was also isolated from pigs of two other farms which showed comparable symptoms and were in direct contact with the index farm as well as from their environment. Immunological host responses detected by tuberculin skin test and ELISA gave positive results at herd level only. Despite the detection of other potential pathogens mycobacteria were regarded as the causative agent of the reproductive disorders. To our knowledge this is the first report of an epidemic mycobacterial infection in a pig holding associated with reproductive disorders, which could be attributed to one single virulent strain, and the first report of detection of M. avium subsp. hominissuis in pig sperm.     

Earthworms (Oligochaeta,Lumbricidae) and mycobacteria

The objective of the study was to define the role of earthworms in the survival of mycobacteria in animal populations. In 13 sampling sites mycobacteria were detected in 53 (5.5%) samples of faeces and parenchymatous tissues from animals, in 25 (7.3%) environmental and in nine (8.2%) earthworm samples. In cattle and goat farms affected by Mycobacterium avium subsp. paratuberculosis (M. paratuberculosis) of IS900 restriction fragment length polymorphism (RFLP) type B-C1 was isolated from 37 (4.6%) faecal samples, three (1.4%) environmental and one (3.1%) earthworm sample. Investigations of aviaries affected by avian tuberculosis detected M. avium of genotype IS901+ and IS1245+ in six (7.9%) bird's faecal and in four (4.4%) environmental samples. M. avium (genotype IS901- and IS1245+) was detected in four (4.4%) and M. abscessus in one (1.1%) environmental sample. M. avium of genotype IS901- and IS1245+ and M. gastri were isolated from three (6.4%) earthworm samples. In pig farm with mycobacteriosis M. avium of genotype IS901- and IS1245+ was detected in five (20.0%) faecal samples from pigs and in four (12.9%) environmental samples. M. scrofulaceum was isolated in one (4.6%) sample of Lumbricus rubellus. In laboratory experiments identical RFLP types of M. paratuberculosis were isolated from bodies and faeces of earthworms 1-2 days after the last contact with the faeces contaminated with the same RFLP type of M. paratuberculosis. The results suggest that earthworms may become vectors of mycobacteria.