Mycobacterial aetiology of Crohn's disease: serologic study using common mycobacterial antigens and a species-specific glycolipid antigen from Mycobacterium paratuberculosis.

Crohn's disease is a granulomatous form of enteritis superficially similar to Johne's disease (paratuberculosis) of ruminants. Recently, a Mycobacterium sp closely related to Mycobacterium paratuberculosis was cultured from tissues of patients with Crohn's disease suggesting that M paratuberculosis may be the aetiologic agent in some cases. In addition, greater seroreactivity to M paratuberculosis has been reported in patients with Crohn's disease. In the present study, we have evaluated the serum antibody response to disrupted M paratuberculosis using ELISA and serum specimens from 33 people with Crohn's disease, 21 with ulcerative colitis, and 12 non-inflammatory bowel disease controls. We failed to find a consistent IgG, IgM, or IgA antibody response to Mycobacterium paratuberculosis. The results indicate that, as in bovine paratuberculosis, serum seroreactivity is not a reliable tool for examining the relationship between human intestinal disease and mycobacteria.     

Lack of support for a common etiology in Johne's disease of animals and Crohn's disease in humans.

The superficial similarity of Johne's disease to Crohn's disease led to the hypothesis that, like the former. Crohn's disease was caused by Mycobacterium paratuberculosis. Detailed pathologic comparisons, however, reveal little similarity between these two entities, including the lack of important extraintestinal manifestations. Attempts to recover M. paratuberculosis by culture have only rarely succeeded and the significance of spheroplasts that appear more frequently on culture is seriously in question. Five immunocytochemistry studies have failed to find mycobacterial antigens in diseased tissues and the five most recent polymerase chain reaction (PCR) attempts to find genomic evidence of M. paratuberculosis were uniformly negative. Numerous serologic studies failed to demonstrate antibody to M. paratuberculosis and attempts to show cell-mediated immunity were also unrewarding. Inoculation of numerous experimental animals with Crohn's disease tissue has failed to induce Johne's disease, and inoculation of various animal species with M. paratuberculosis has equally failed to result in Crohn's disease. Controlled studies of the treatment of Crohn's disease with antimycobacterial agents have generally resulted in no improvement, and most studies that have shown a positive response are either uncontrolled or include broad-spectrum antibiotics that may be acting on pathogens other than mycobacteria. Finally, although Johne's disease is common in farm animals, and infected animals shed M. paratuberculosis in large numbers, no record of zoonotic transmission has been recorded.     

Treatment of severe Crohn''s disease using antimycobacterial triple therapy — approaching a cure?

BACKGROUND: Mycobacterium avium subspecies paratuberculosis is probably the best candidate for a microbial cause of Crohn's disease although arguments to the contrary can be equally convincing. Growing evidence suggests that prolonged antimycobacterial combination therapy can improve Crohn's disease in some patients. AIM: To report long-term observations in patients with severe Crohn's disease treated with triple macrolide-based antimycobacterial therapy. PATIENTS: A series of 12 patients (7 male, 5 female; aged 15-42 years) with severe, obstructive or penetrating Crohn's disease were recruited. METHODS: Patients failing maximal therapy were commenced prospectively on a combination of rifabutin (450 mg/d), clarithromycin (750 mg/d) and clofazimine (2 mg/kg/d). Progress was monitored through colonoscopy, histology, clinical response and Harvey-Bradshaw activity index. RESULTS: Follow-up data were available for up to 54 months of therapy Six out of 12 patients experienced a full response to the antiMycobacterium avium subspecies paratuberculosis combination achieving complete clinical, colonoscopic and histologic remission of Crohn's disease. Four of these patients were able to cease treatment after 24-46 months, 3 of whom remained in total remission without treatment for up to 26 months and one patient relapsed after six months off treatment. A partial response to the anti-Mycobacterium avium subspecies paratuberculosis combination was seen in 2 patients showing complete clinical remission with mild histologic inflammation. Return to normal of terminal ileal strictures occurred in 5 patients. Harvey-Bradshaw activity index in patients showing a full or partial response to therapy fell from an initial 13.4 +/- 1. 91 to 0. 5 +/- 0. 47 [n = 8, p < 0. 001) after 52-54 months. CONCLUSIONS: Reversal of severe Crohn's disease has been achieved in 6/12 patients using prolonged combination anti-Mycobacterium avium subspecies paratuberculosis therapy alone. Three patients remain in long-term remission with no detectable Crohn's disease off all therapy These results support a causal role for Mycobacterium avium subspecies paratuberculosis in Crohn's disease while also suggesting that a cure may become possible.     

Mycobacterium paratuberculosis DNA not detected in Crohn's disease tissue by fluorescent polymerase chain reaction.

The role of mycobacteria in the aetiology of Crohn's disease has been a contentious subject for many years. Mycobacterium paratuberculosis is known to cause a chronic granulomatous enteritis in animals (Johne's disease) and has been implicated as a possible infectious cause of Crohn's disease. However this fastidious organism is only rarely detected by conventional microbiological techniques. This study used oligonucleotide primers to the species-specific M paratuberculosis IS900 DNA insertion element and the polymerase chain reaction to amplify any M paratuberculosis DNA from intestinal tissue DNA extracts. One oligonucleotide primer was fluorochrome-labelled and the presence of fluorescent amplified product was determined using an automated DNA sequencer with a computerised gel-scanning laser. This method was shown capable of detecting 1-2 mycobacterial genomes. Intestinal tissue samples were obtained from 68 patients with histologically confirmed Crohn's disease, 49 patients with histologically confirmed ulcerative colitis, and 26 non-inflammatory bowel disease controls. In no case was M paratuberculosis detected in any of the inflammatory bowel disease tissue samples and only one non-inflammatory bowel disease case was positive. These results do not support the hypothesis that M paratuberculosis has an aetiological role in Crohn's disease.     

Molecular evidence for Mycobacterium avium subspecies paratuberculosis (MAP) in Crohn''s disease correlates with enhanced TNF-α secretion

BACKGROUND: Support for a role of Mycobacterium avium subspecies paratuberculosis in Crohn's disease is largely based on epidemiological evidence, as no data on mechanisms linking the presence of M. avium subspecies paratuberculosis with gut damage is available. AIMS: To determine whether the presence of M. avium subspecies paratuberculosis contributes to the pathogenesis of Crohn's disease by promoting cytokine secretion within gut mucosa. PATIENTS AND METHODS: A total of 235 subjects were recruited: 63 with Crohn's disease, 53 with ulcerative colitis, 45 with irritable bowel syndrome and 74 normal controls. M. avium subspecies paratuberculosis status was defined by nested PCR using IS900 sequence. Gut mucosal organ cultures were established to detect cytokine secretion patterns. RESULTS: Significantly higher tumour necrosis factor-alpha concentrations were found in culture supernatants for Crohn's disease compared to ulcerative colitis (p<0.05), irritable bowel syndrome (p<0.01) and controls (p<0.0001). When tumour necrosis factor-alpha levels were correlated with the presence of M. avium subspecies paratuberculosis, significantly greater concentrations were only found in M. avium subspecies paratuberculosis-positive Crohn's disease patients (p<0.05). Tumour necrosis factor-alpha levels in M. avium subspecies paratuberculosis-positive Crohn's disease were significantly higher than in M. avium subspecies paratuberculosis-positive ulcerative colitis (p<0.01), M. avium subspecies paratuberculosis-positive irritable bowel syndrome (p<0.05) and M. avium subspecies paratuberculosis-positive controls (p<0.01) and all M. avium subspecies paratuberculosis-negative specimens. CONCLUSIONS: The data link M. avium subspecies paratuberculosis with a pathogenic mechanism in Crohn's disease and is consistent with abnormal macrophage handling of M. avium subspecies paratuberculosis.     

The role of bacteria and pattern-recognition receptors in Crohn's disease

Crohn's disease is widely regarded as a multifactorial disease, and evidence from human and animal studies suggests that bacteria have an instrumental role in its pathogenesis. Comparison of the intestinal microbiota of patients with Crohn's disease to that of healthy controls has revealed compositional changes. In most studies these changes are characterized by an increase in the abundance of Bacteroidetes and Proteobacteria and a decrease in that of Firmicutes. In addition, a number of specific mucosa-associated bacteria have been postulated to have a role in Crohn's disease, including Mycobacterium avium subspecies paratuberculosis, adherent and invasive Escherichia coli, Campylobacter and Helicobacter species. The association between mutations in pattern-recognition receptors (Toll-like receptors and Nod-like receptors) and autophagy proteins and Crohn's disease provides further evidence to suggest that defective sensing and killing of bacteria may drive the onset of disease. In this Review, we present recent advances in understanding the role of bacteria and the contribution of pattern-recognition receptors and autophagy in the pathogenesis of Crohn's disease.     

克隆产现与副结核杆菌

分支杆菌、尤其是副结核杆菌长期被疑为克隆病的致病菌。采用聚合酶链反应技术对手术及内镜活检的７４例石蜡包埋组织中的副结核杆菌ＤＮＡ进行检测。扩增的靶ＤＮＡ为副结核杆菌染色体特异重复插入序列ＩＳ９００１上４００ｂｐ的片段。其产笺特异性通过生物素杆主民的副结核杆菌全染色体探针Ｓｏｕｔｈｅｒｎ杂交证实。     

Specific detection of Mycobacterium paratuberculosis DNA associated with granulomatous tissue in Crohn's disease.

The role of mycobacteria, specifically Mycobacterium paratuberculosis, in Crohn's disease has aroused considerable controversy for many years. Using the ultra sensitive polymerase chain reaction some studies have reported detection of M paratuberculosis DNA in as many as 65% of Crohn's disease patients but also in patients without disease. Other studies have been negative for both groups. We therefore designed a double blind control trial to investigate the presence of mycobacterial DNA in age, sex, and tissue matched paraffin wax embedded tissues from 31 Crohn's disease tissues, 20 diseased gut control tissues, and 10 ulcerative colitis tissues. The specimens were coded and analysed blind with three separate polymerase chain reactions (PCR) based on DNA sequences specific for M paratuberculosis (IS900), M avium (RFLP type A/1) (IS901), and the Mycobacterium genus (65 kDa gene, TB600). The number of granulomata and presence of acid fast bacilli in each Crohn's disease tissue was also investigated. The sensitivity of the system was determined using similarly prepared gut tissue from an animal infected with M paratuberculosis. Four of 31 Crohn's disease tissues and none of the 30 control and ulcerative colitis derived tissues amplified M paratuberculosis DNA. Crohn's disease tissues containing granulomata were significantly more likely to amplify M paratuberculosis specific DNA on PCR than the non-Crohn's disease tissues (p = 0.02). All the positive Crohn's disease tissues contained granulomata, and none contained acid fast bacilli. Equivalent numbers of Crohn's and non-Crohn's disease tissues amplified the region of the 65 kD gene on PCR for non-specific mycobacterial DNA (11/31 and 9/30 respectively). No sections produced an amplified product with the IS901 PCR. These results suggest that few Crohn's disease gut biopsy sections contain M paratuberculosis DNA in association with granulomata. The absence of such DNA in any control and ulcerative colitic tissue strengthens the case for it having a specific association, which may be pathogenic, with Crohn's disease in this minority of patients.     

The evidence for Mycobacterium paratuberculosis in Crohn's disease.

PURPOSE OF REVIEW: Though long hypothesized, the putative link between Mycobacterium avium paratuberculosis and Crohn's disease remains neither confirmed nor refuted. This article reviews published contributions that directly or indirectly address this question. RECENT FINDINGS: Epidemiologic studies, looking for M. avium paratuberculosis DNA in Crohn's tissue, show a strong association between the agent and this disease. Supporting data, however, are presently inconclusive on a causal role. Genetic studies provide indirect support for a role of mycobacteria in Crohn's disease, by identifying susceptibility genes that encode proteins implicated in innate immunity to intracellular bacteria. Clinical trial data support at least a short-term benefit for antimycobacterial therapy in Crohn's disease, but the microbial specificity of this response is presently unknown. SUMMARY: There appears to be a strong association between M. avium paratuberculosis and Crohn's disease, but the causality of this association is unknown. Consequently, the therapeutic implications of this association require further study. A number of critical questions about the biology of M. avium paratuberculosis remain unanswered. Data from studies of this organism, and its interaction with the immune system, can help address proposed reasons for or against a role of M. avium paratuberculosis in the etiology of Crohn's disease.     

Antibodies to Mycobacterium paratuberculosis and nine species of environmental mycobacteria in Crohn's disease and control subjects.

Cultural and serological studies have provided limited, often conflicting, evidence of a role for mycobacteria in the pathogenesis of Crohn's disease. Interest has focussed on Mycobacterium paratuberculosis, previously considered to be common in the environment with no major role as a human pathogen. Whether a specific serum antibody response to mycobacteria occurs in Crohn's disease or ulcerative colitis was investigated. Sera from patients with Crohn's disease (n = 38), ulcerative colitis (n = 15), and a healthy control population (n = 30) were assayed in an enzyme linked immunosorbent assay (ELISA) using eight filtered sonicate mycobacterial preparations and a purified protein derivative made from the bovine tubercle bacillus. In addition, IgG, IgM, and IgA levels to M paratuberculosis were determined in sera from patients with active (n = 24) or inactive (n = 29) Crohn's disease and the control populations. There was strong evidence of contact with environmental mycobacteria in all patients and control populations, with the greatest responses to preparations of M avium, M tuberculosis, and M kansasii. A large proportion of patients with Crohn's disease had antibodies that bound most antigens tested but there were no statistical differences between these values and those of the control population. Similarly, there were no differences in antibody levels to M paratuberculosis in patient and control groups. Although a subset of patients with active Crohn's disease (25%) had IgG concentrations that exceeded the control mean by more than 2 SD, this phenomenon may not be specific to Crohn's disease: 20% of a small group of patients with coeliac disease had similarly raised IgG levels to M paratuberculosis. These findings do not provide serological evidence of a role for this organism in the pathogenesis of Crohn's disease.     

Is Crohn's disease caused by a mycobacterium? Comparisons with leprosy,tuberculosis, and Johne's disease

Although Crohn's disease is considered to be autoimmune in origin, there is increasing evidence that it may have an infectious cause. The most plausible candidate is Mycobacterium avium subspecies paratuberculosis (MAP). Intriguingly, Koch's postulates may have been fulfilled for MAP and Crohn's disease, even though they still have not been met for Mycobacterium leprae and leprosy. In animals MAP causes Johne's disease, a chronic wasting intestinal diarrhoeal disease evocative of Crohn's disease. Johne's disease occurs in wild and domesticated animals, including dairy herds. Viable MAP is found in human and cow milk, and is not reliably killed by standard pasteurisation. MAP is ubiquitous in the environment including in potable water. Since cell-wall-deficient MAP usually cannot be identified by Ziehl-Neelsen staining, identification of MAP in human beings requires culture or detection of MAP DNA or RNA. If infectious in origin, Crohn's disease should be curable with appropriate antibiotics. Many studies that argue against a causative role for MAP in Crohn's disease have used antibiotics that are inactive against MAP. However, trials that include macrolide antibiotics indicate that a cure for Crohn's disease is possible. The necessary length of therapy remains to be determined. Mycobacterial diseases have protean clinical manifestations, as does Crohn's disease. The necessity of stratifying Crohn's disease into two clinical manifestations (perforating and non-perforating) when interpreting the results of antibiotic therapy is discussed. Rational studies to evaluate appropriate therapies to cure Crohn's disease are proposed.     

T-cellular immune reactions (in macrophage inhibition factor assay) against Mycobacterium paratuberculosis, Mycobacterium kansasii, Mycobacterium tuberculosis, Mycobacterium avium in patients with chronic inflammatory bowel disease.

A mycobacterial aetiology has been suggested for Crohn's disease. A slow growing mycobacterium, biochemically and genetically identical to M paratuberculosis, the causative agent of enteritis in ruminants (Johne's disease), has been isolated from gut specimens of patients affected by Crohn's disease. If M paratuberculosis or other mycobacteria play a role in the pathogenesis of Crohn's disease, then patients may have been sensitised to these mycobacteria or show an anergy immune reaction. We therefore investigated the T-cell mediated immune response to sonicates of M paratuberculosis, M kansasii, M avium, and M tuberculosis in 35 patients with Crohn's disease, 28 with ulcerative colitis, and 25 controls using a macrophage inhibition factor assay on peripheral blood lymphocytes. Two types of reaction patterns were identified--that is, 'responders' (subjects with a macrophage inhibition factor assay in which a dose response relation was present and a percentage of inhibition exceeding 20%), and 'non-responders'. There was no significant difference in the prevalence of responders (59%-80%) and non-responders (20%-41%) to these mycobacteria between the group of Crohn's disease, ulcerative colitis, and control group. We found also that a large proportion of controls showed T-cell immunisation to the mycobacteria which supports the contention that the antigens are practically commensal. Our results do not support the proposed involvement of mycobacteria in the pathogenesis of Crohn's disease.     

Mycobacterium paratuberculosis DNA in Crohn''s disease tissue.

Crohn's disease has long been suspected of having a mycobacterial cause. Mycobacterium paratuberculosis is a known cause of chronic enteritis in animals, including primates, but may be very difficult to detect by culture. IS900 is a multicopy genomic DNA insertion element highly specific for M paratuberculosis. A polymerase chain reaction (PCR) based on the 5' region of IS900 and capable of the specific detection of a single M paratuberculosis genome was developed. This was applied to DNA extracts of full thickness samples of intestine removed at surgery from 40 patients with Crohn's disease, 23 patients with ulcerative colitis, and 40 control patients without inflammatory bowel disease. Stringent precautions were taken that excluded contamination artefact. M paratuberculosis was identified in 26 of 40 (65%) Crohn's disease, in 1 of 23 (4.3%) ulcerative colitis, and in 5 of 40 (12.5%) control tissues. Positive samples from Crohn's disease were from both the small intestine and colon, those from control tissues were from the colon those from control tissues were from the colon only. All PCR internal control reactions were negative. The presence of M paratuberculosis in a small proportion of apparently normal colonic samples is consistent with a previously unsuspected alimentary prevalence in humans. The presence in two thirds of Crohn's disease tissues but in less than 5% of ulcerative colitis tissues is consistent with an aetiological role for M paratuberculosis in Crohn's disease.     

Detection and Isolation of Mycobacterium avium subspecies paratuberculosis from intestinal mucosal biopsies of patients with and without Crohn's disease in Sardinia

OBJECTIVES: Sardinia is an island community of 1.6 million people. There are also about 3.5 million sheep and one hundred thousand cattle in which Johne's disease and Mycobacterium avium subspecies paratuberculosis infection are endemic. The present study was designed to determine what proportion of people in Sardinia attending for ileocolonoscopy with or without Crohn's disease were infected with this pathogen. METHODS: Mycobacterium avium subspecies paratuberculosis was detected by IS900 PCR on DNA extracts of fresh intestinal mucosal biopsies as well as by isolation in culture using supplemented MGIT media followed by PCR with amplicon sequencing. RESULTS: Twenty five patients (83.3%) with Crohn's disease and 3 control patients (10.3%) were IS900 PCR positive (p = 0.000001; Odds ratio 43.3). Mycobacterium avium subspecies paratuberculosis grew in cultures from 19 Crohn's patients (63.3%) and from 3 control patients (10.3%) (p = 0.00001; Odds ratio 14.9). All patients positive by culture had previously been positive by PCR. Mycobacterium avium subspecies paratuberculosis first appeared in the liquid cultures in a Ziehl Neelsen (ZN) staining negative form and partially reverted through a rhodamine-auramine positive staining form to the classical ZN positive form. This resulted in a stable mixed culture of all 3 forms illustrating the phenotypic versatility of these complex chronic enteric pathogens. CONCLUSIONS: Mycobacterium avium subspecies paratuberculosis was detected in the majority of Sardinian Crohn's disease patients. The finding of the organism colonizing a proportion of people without Crohn's disease is consistent with what occurs in other conditions caused by a primary bacterial pathogen in susceptible hosts.     

Agglutinins to bacteria in Crohn's disease.

Sera from patients with Crohn's disease were tested for antibodies against organisms which are thought to cause inflammatory bowel disease in animals, or have been implicated in human Crohn's disease. Control sera were collected from healthy individuals and patients with ulcerative colitis. Sera from Crohn's disease and controls failed to agglutinate Clostridium colinum or Campylobacter sputorum subsp. mucosalis and two strains of Mycobacterium paratuberculosis (M26 and M27). Most of the sera agglutinated a Citrobacter freundii variant, Mycobacterium paratuberculosis (M28) and Mycobacterium avium (M41) but Crohn's disease sera did not differ from controls. A complement fixation test against Chlamydia gave more positive reactions in patients with Crohn's disease and colitis than in healthy controls. There was a clear difference between the sera from patients with Crohn's disease and other sera, including ulcerative colitis, in agglutination tests with the commensal coccoid rods of the genera Eubacterium and Peptostreptococcus; in these tests 54% of sera from Crohn's disease were positive compared with 11% in ulcerative colitis and none of the sera from healthy controls. All the results were essentially negative with the exception of those from Eubacterium and Peptostreptococcus and these bacteria merit investigation.     

COMBINATION ANTI-PARATUBERCULOSIS THERAPY WITH CLARITHROMYCIN, RIFABUTIN and CLOFAZIMINE IN THE TREATMENT OF CROHN'S DISEASE. A PRELIMINARY REPORT ON THE AUSTRALIAN MULTICENTRE TRIAL

Mycobacterium paratuberculosis has been proposed as the cause of Crohn's disease, although this remains controversial. Observational studies of antibiotics with activity against this organism have led to the establishment of an Australian multicentre placebo-controlled trial of clarithromycin, rifabutin and clofazimine in patients with active Crohn's disease. All subjects receive treatment with a tapering regimen of prednisolone for the initial 16 weeks, combined with either antibiotics or matching placebos. If remission is achieved at the end of this period and prednisolone has been ceased they continue the trial medications for 2 years. The primary outcomes – remission rates at 1, 2 and 3 years – aim to determine whether these antibiotics are of long-term benefit in Crohn's disease. Secondary outcomes include rate of remission at 4 months, safety and quality of life.
As of May 2001, 171 of the required 212 subjects have been enrolled. Eighty-seven of 146 potential subjects have reached 16 weeks in remission (60%). Eighty of these are ongoing (92%). Only five withdrawals from the trial have occurred for adverse events thought probably related to trial medications – four subjects with raised LFTs and one who developed a rash. Other withdrawals have been mainly due to lack of response or worsening of Crohn's disease (37), hypomania from steroids (one patient) and pregnancy (one patient on placebo). There have been 11 withdrawals for protocol violations. An Independent Data Monitoring Committee has analysed progress of the trial and unanimously recommended that it continue.
This study is the largest and longest suitably powered randomised controlled trial of antibiotics undertaken in Crohn's disease. It will determine whether this antibiotic combination alters the natural history of this disorder.
This study is being supported by Pharmacia Australia Pty Limited.     

Mycobacteria in Crohn's disease: a persistent hypothesis

Efforts to explore a mycobacterial origin for Crohn's disease typically have involved an epidemiological approach, searching for Mycobacterium avium subsp. paratuberculosis in patient tissue. An alternative approach involves consideration of genetic and experimental data regarding host resistance to mycobacteria. From human and mycobacterial genetics, it is known that mycobacterial diseases depend on both pathogen and host factors and that tuberculosis and leprosy are effectively genetic diseases. The discovery of a number of Crohn's susceptibility genes, including NOD2/CARD15, demonstrates that Crohn's also is a complex genetic disease. Mutations in NOD2/CARD15 do not necessarily lead to Crohn's disease, so other mitigating factors, genetic and/or environmental, probably are required to produce illness. Recent work has shown that NOD2/CARD15 serves a role in bacterial sensing and activation of innate immune responses, providing a link between Crohn's genetics and an environmental factor, potentially a bacterial trigger. In this review, we discuss the current understanding of mycobacterial and Crohn's genetic susceptibility and review the evidence that NOD2/CARD15 may mediate host resistance to mycobacterial infection.     

The mycobacteria story in Crohn's disease

The heterogeneity of Crohn's disease suggests that it would be unwise to dismiss an infectious contribution to the pathogenesis in a subset of patients. The most enduring infectious candidate has been Mycobacterium paratuberculosis, which appears to be widespread in nature and appears to have the potential to infect humans. However, there are many counterarguments to the notion that MAP causes Crohn's disease, and numerous observations are seemingly at variance with this concept.     

Polymerase chain reaction detection of Mycobacterium paratuberculosis and Mycobacterium avium subsp silvaticum in long term cultures from Crohn''s disease and control tissues.

Thirty one cultures were established in MG3 medium from the intestinal tissues of 29 patients, including 18 with Crohn's disease, five with ulcerative colitis, and six non-inflammatory bowel disease controls. All cultures grew either acid fast bacilli or uncharacterized spheroplasts. Pellets from these cultures were coded and assayed blind for M paratuberculosis and M avium subsp silvaticum using IS900- and IS902-PCR (polymerase chain reaction) assays, respectively. IS900 and IS902 are multicopy DNA insertion elements specific for these two organisms. Six Crohn's disease cultures and a single non-inflammatory bowel disease control were positive for M paratuberculosis. A further six cultures were positive for M avium subsp silvaticum, of which two each were from Crohn's disease, ulcerative colitis, and non-inflammatory bowel disease controls. The intensity of the IS900-PCR signals indicated very low numbers of M paratuberculosis organisms and bore no relation to visible spheroplastic or bacillary mycobacterial growth. The results suggest that M paratuberculosis isolated from man exists in a form which hardly replicates if at all when cultured in MG3 medium in vitro, and are consistent with the involvement of this known animal enteric pathogen in a proportion of chronic enteritis in man.