[Immunochemical analysis of the immune response in late manifestations of Lyme borreliosis].

Compared to American strains, European Borrelia burgdorferi strains revealed considerable heterogeneity of major proteins. Four strains isolated from ticks, human skin and human CSF were selected from our 23 Borrelia burgdorferi isolates. These strains and the American type strain B31 were characterized by SDS-PAGE (Coomassie Blue staining) and Western blots (using rabbit immune sera against two of the strains and two monoclonal antibodies (H5323 and H3TS) against a major outer surface protein (OspA]. The strains showed considerable differences in SDS-PAGE pattern. Corroborating the results of a previous study, we could demonstrate that the OspA (31/32K) can change from a minor to a major protein and in reverse the pC (21/22K) from a major to a minor protein during subculturing. Moreover, European strains can antigenically differ in OspA, pC and also in a further low molecular weight protein of 17/18K. To examine whether the antigenic heterogeneity of European isolates is reflected in the immune response of European patients we examined sera from patients with late manifestations of Lyme Borreliosis by Western blot using the five strains as antigens. Sera from seven patients with acrodermatitis chronica atrophicans (ACA) showed a surprisingly strong reactivity with the skin isolate. All sera had antibodies against the 17/18K protein of the skin isolate, but none was reactive with the analogous 17/18K of the other strains. On the other hand a comparable predominance of one strain was not found testing sera from patients with Lyme arthritis. One patient even had antibodies against OspA and OspB proteins of strain B31. Contrary to findings in American Lyme Disease antibodies against the OspA were rarely observed in the sera of our patients (only one patient had such antibodies) although we tested the patients sera with five different strains. Only two patients had stronger reactions with the skin isolate. These findings suggest that ACA is caused by antigenically closely related Borreliae. This could explain the finding that ACA is rarely observed in the US (US strains are antigenically closely related to strain B31). Our findings in patients with Lyme Arthritis--on the other hand--suggest that "different serotypes" can cause Lyme Arthritis. This does not exclude the possibility that Borrelia proteins are an important factor in the pathogenesis of Lyme arthritis. Finally the differences in reactivity of sera with different strains in the Western blot led us to examine whether such differences are also found in serodiagnostic tests using different strains as antigens.(ABSTRACT TRUNCATED AT 400 WORDS)