Clinical symptoms of Crohn's disease, a chronic human enteritis of undefined etiology, mimic those of Johne's disease of ruminants, which is induced by Mycobacterium paratuberculosis, hence the hypothesis of a common mycobacterial etiology. Recently, a species-specific polypeptide (a362) of M. paratuberculosis was obtained by genetic engineering technology from an immunodominant protein of the M. paratuberculosis A36 complex. Sensitivity and specificity of a362 prompted its use in an enzyme-linked immunosorbent assay (ELISA) for paratuberculosis. This assay correctly diagnosed the tested blood samples from infected cattle¹. In the present study, such an a362-based ELISA was used to compare the serological response of Crohn's patients with healthy controls. The a362 polypeptide was used for serological analysis of 50 healthy subjects and 73 Crohn's patients. Healthy subjects were blood donors from the Belgian Red Cross Organization and Crohn's patients were diagnosed at the Saint Luc University Clinics by conventional clinical, radiological, endoscopic and histological criteria. Statistical comparison of ELISA-data from the two populations was made using the nonparametric Mann-Whitney U test. Crohn's patients had significantly higher IgG (P < 0.05) and IgA (P < 0.001) titers than controls. Anti-a362 IgA titers were then analyzed using a mixture population modeling². A monomodal Gaussian distribution of control values, compatible with a single population, was observed; whereas the bimodal distribution of Crohn's patients sera supported the occurrence of two populations. In 64% of Crohn's patients, the anti-a362 IgA level was comparable to that of controls whereas titers were significantly higher in 36% of the patients. Several hypothesis can be proposed to explain the occurrence of two Crohn's subpopulations. In the case of patients with high anti-a362 IgA titers, M. paratuberculosis may be the etiological agent. In other cases, mycobacteria, such as M. kansasii or M. tuberculosis, might be responsible for some of the 64% Crohn's cases diagnosed negative by the a362 test, which is species-specific. On the other hand, the level of anti-M. paratuberculosis is possibly low in some steps of the disease, as it is the case for tuberculoid leprosy or primary tuberculosis. Finally, some Crohn's disease patients may perhaps recognize mycobacterial B-cell epitopes different from those present in the a362 polypeptide, on which our assay is based. The use of the species-specific a362 assay would lower the high background level of antibody directed against mycobacteria in Crohn's patients (due to the presence in the intestinal tract of different mycobacteria) and restrict the response to specific target. Moreover, IgA should be preferentially involved in such enteropathy as Crohn's disease, and differences with respect to controls are expected to be more evident, whenever a study is restricted to a specific immunoglobulin. ¹ J. Clin, Microbiol. 1993, 31:947-954. ² J. Virol. Meth. 1990, 27:135-144.