Antibody responses to MAP 1B and other Cowdria ruminantium antigens are down regulated in cattle challenged with tick-transmitted heartwater.

Serological diagnosis of heartwater or Cowdria ruminantium infection has been hampered by severe cross-reactions with antibody responses to related ehrlichial agents. A MAP 1B indirect enzyme-linked immunosorbent assay that has an improved specificity and sensitivity for detection of immunoglobulin G (IgG) antibodies has been developed to overcome this constraint (A. H. M. van Vliet, B. A. M. Van der Zeijst, E. Camus, S. M. Mahan, D. Martinez, and F. Jongejan, J. Clin. Microbiol. 33:2405-2410, 1995). When sera were tested from cattle in areas of endemic heartwater infection in Zimbabwe, only 33% of the samples tested positive in this assay despite a high infection pressure (S. M. Mahan, S. M. Samu, T. F. Peter, and F. Jongejan, Ann. N.Y. Acad. Sci 849:85-87, 1998). To determine underlying causes for this observation, the kinetics of MAP 1B-specific IgG antibodies in cattle after tick-transmitted C. ruminantium infection and following recovery were investigated. Sera collected weekly over a period of 52 weeks from 37 cattle, which were naturally or experimentally infected with C. ruminantium via Amblyomma hebraeum ticks, were analyzed. MAP 1B-specific IgG antibody responses developed with similar kinetics in both field- and laboratory-infected cattle. IgG levels peaked at 4 to 9 weeks after tick infestation and declined to baseline levels between 14 and 33 weeks, despite repeated exposure to infected ticks and the establishment of a carrier state as demonstrated by PCR and xenodiagnosis. Some of the serum samples from laboratory, and field-infected cattle were also analyzed by immunoblotting and an indirect fluorescent-antibody test (IFAT) to determine whether this observed seroreversion was specific to the MAP 1B antigen. Reciprocal IFAT and immunoblot MAP 1-specific antibody titres peaked at 5 to 9 weeks after tick infestation but also declined between 30 and 45 weeks. This suggests that MAP 1B-specific IgG antibody responses and antibody responses to other C. ruminantium antigens are down regulated in cattle despite repeated exposure to C. ruminantium via ticks. Significantly, serological responses to the MAP 1B antigen may not be a reliable indicator of C. ruminantium exposure in cattle in areas of endemic heartwater infection.

Population-based evaluation of the Ehrlichia ruminantium MAP 1B indirect ELISA.

The indirect MAP 1B ELISA based on the recombinant MAP 1B fragment of the immunodominant MAP I protein of Ehrlichia ruminantium is considered to be the most sensitive and specific assay for the serodiagnosis of heartwater. In this study, we evaluated its reliability in detecting exposure to E. ruminantium in field populations of domestic ruminants in Zimbabwe. Cattle and goat herds in endemically stable areas with high infection pressure and an expected close to 100% prevalence of E. ruminantium exposure were sampled. Bovine sera (858) and caprine sera (706) collected at seven locations representative of the two main production systems (communal lands and large scale commercial farms) in the two main agroecological zones of Zimbabwe (highveld and lowveld) were analysed. The prevalence of MAP 1B-specific antibodies in goats was similar and high, ranging from 67 to 100%, at all except one site (43%). Age-specific differences in goats (1, 2, 3, 4, 5 years) were not observed. In contrast, MAP 1B seroprevalence in cattle was significantly lower (P < 0.001), ranging from 46 to 61% in the lowveld communal area and from 24 to 33% in the remaining areas (P < 0.001). Age-specific differences in seroprevalence (1, 2, 3, 4, 5-7 + years) were similarly not evident in cattle (P < 0.15). Hence, the indirect MAP 1B ELISA may be an unreliable indicator of past exposure to heartwater in field-infected cattle in Zimbabwe. Although the reasons for this low response in field cattle are not fully understood, this study illustrates the need for field validation of the performance of new diagnostic tests prior to their use for epidemiological purposes.

Potential value of major antigenic protein 2 for serological diagnosis of heartwater and related ehrlichial infections.

Cowdria ruminantium is the etiologic agent of heartwater, a disease causing major economic loss in ruminants in sub-Saharan Africa and the Caribbean. Development of a serodiagnostic test is essential for determining the carrier status of animals from regions where heartwater is endemic, but most available tests give false-positive reactions with sera against related Erhlichia species. Current approaches rely on molecular methods to define proteins and epitopes that may allow specific diagnosis. Two major antigenic proteins (MAPs), MAP1 and MAP2, have been examined for their use as antigens in the serodiagnosis of heartwater. The objectives of this study were (i) to determine if MAP2 is conserved among five geographically divergent strains of C. ruminantium and (ii) to determine if MAP2 homologs are present in Ehrlichia canis, the causative agent of canine ehrlichiosis, and Ehrlichia chaffeensis, the organism responsible for human monocytic ehrlichiosis. These two agents are closely related to C. ruminantium. The map2 gene from four strains of C. ruminantium was cloned, sequenced, and compared with the previously reported map2 gene from the Crystal Springs strain. Only 10 nucleic acid differences between the strains were identified, and they translate to only 3 amino acid changes, indicating that MAP2 is highly conserved. Genes encoding MAP2 homologs from E. canis and E. chaffeensis also were cloned and sequenced. Amino acid analysis of MAP2 homologs of E. chaffeensis and E. canis with MAP2 of C. ruminantium revealed 83.4 and 84.4% identities, respectively. Further analysis of MAP2 and its homologs revealed that the whole protein lacks specificity for heartwater diagnosis. The development of epitope-specific assays using this sequence information may produce diagnostic tests suitable for C. ruminantium and also other related rickettsiae.

Evaluation of the MAP-1B ELISA for cowdriosis with field sera from livestock in Zimbabwe.

The Map 1B (Senegal) antigen-based indirect ELISA was evaluated in Zimbabwe with field sera from heartwater-free and heartwater-endemic areas. Of 205 sheep sera samples from a heartwater-free area, 34 were negative and 171 were positive by immunoblotting. These 171 samples were classified as false positives. Of the same 205 samples, 199 were negative and only 6 were positive by the MAP 1B ELISA. Of 72 cattle samples tested from a similar area, 71 were negative, with only 1 sample positive by MAP 1B ELISA. By immunoblotting, 43 of 72 cattle sera were negative and 29 were positive. Of the 46 goat samples tested from a heartwater-free area, only 2 were positive by the MAP 1B ELISA. Based on these results, the MAP 1B ELISA was more specific for heartwater than the immunoblotting assay. Of 96 and 282 cattle sera analyzed from heartwater-endemic farms in the lowveld and highveld of Zimbabwe, respectively, approximately 33% were positive by the MAP 1B ELISA. Goat sera from the same farms had a higher sero-prevalence (> 90%). The implications of these results for serodiagnosis of heartwater using the MAP 1B ELISA will be discussed.

Laboratory reared Amblyomma hebraeum and Amblyomma variegatum ticks differ in their susceptibility to infection with Cowdria ruminantium.

The susceptibility of laboratory reared Zimbabwean Amblyomma hebraeum and A. variegatum ticks to infection with geographically distinct Cowdria ruminantium strains was investigated by feeding both species simultaneously on individual sheep infected with one of the four strains (Crystal Springs [Zimbabwe], Ball 3 [South Africa], Gardel [Guadeloupe] and Nigeria [Nigeria]). A. hebraeum ticks demonstrated a high susceptibility to infection with all four C. ruminantium strains. In comparison, A. variegatum were less susceptible to infection with the Crystal Springs and Ball 3 strains (P < 0.001), but showed a similar susceptibility to the Gardel and Nigeria strains. The differences in susceptibility of A. variegatum to infection with the four strains of C. ruminantium correlated with the origin of these strains. The consistently higher susceptibility of A. hebraeum ticks to infection with geographically different C. ruminantium strains may be one explanation for the observation that heartwater is a more serious problem where A. hebraeum is the vector of the disease.

Size variation of the major immunodominant protein of Cowdria ruminantium.

An immunodominant response is made to a polypeptide of approximately 32 kDa in animals infected with the rickettsial pathogen Cowdria ruminantium. We show here using cultured strains of the rickettsia from different geographical areas that the apparent size of this polypeptide varies with strain origin. Changes in the primary structure between strains should be considered in the design of vaccines and diagnostic tests based on this antigen.

Molecular cloning of a gene encoding the immunogenic 21 kDa protein of Cowdria ruminantium.

Major immunogenic polypeptides (21, 32, 40, 46, 58, 85 and 160 kDa) of Cowdria ruminantium were identified by immunoprecipitation and immunoblotting. A pUC13 library of C. ruminantium genomic DNA was screened with hyperimmune sheep serum to identify Escherichia coli colonies which expressed genes encoding these immunogenic proteins. A recombinant E. coli colony, F5.2, was identified containing plasmid insert DNA of 2773 bp. The cloned DNA insert contained two long open reading frames (ORFs) of 627 bp (complete) and 831 bp (incomplete), both potentially encoding proteins containing an N-terminal signal peptide. Deletion experiments suggested that the hyperimmune sheep serum recognized a protein that was encoded by the 627 bp ORF. The 627 bp ORF was amplified by polymerase chain reaction (PCR), subcloned and expressed to a high level in E. coli. A sheep antiserum made to the expressed recombinant fusion protein recognized a 21 kDa protein of all strains of C. ruminantium tested, confirming that the 627 bp ORF encodes a native 21 kDa protein in C. ruminantium. Similarly, the recombinant protein was recognized by all sera tested from heartwater-infected animals. The antigenic conservation of the 21 kDa protein and its immunogenic nature are reasons for further testing of this recombinant protein in subunit diagnostic tests.

Reactivity of sera collected from dogs in Mutare, Zimbabwe, to antigens of Ehrlichia canis and Cowdria ruminantium.

Six of 20 sera from apparently healthy dogs in Mutare, Zimbabwe, contained antibodies which were reactive with Cowdria ruminantium and Ehrlichia canis in indirect fluorescent antibody tests at similar titres. In Western blots these sera recognised the immunodominant antigen bands of the two organisms. Sera with high titres (> 1/80) recognised additional antigen bands of each organism. It was not possible to determine whether these dogs had been exposed to C ruminantium, E canis or other Ehrlichia species. In areas where these organisms coexist serological tests for canine ehrlichiosis should be interpreted with caution.

Serological evidence for antigenic relationships between Ehrlichia canis and Cowdria ruminantium.

The sera from dogs experimentally infected with Ehrlichia canis reacted at very similar titres in indirect fluorescent antibody (IFA) tests with E canis and Cowdria ruminantium antigen. In Western blots these sera reacted strongly with the immunodominant antigens of E canis (27 kDa) and C ruminantium (25 and 32 kDa). The dogs experimentally infected with C ruminantium showed no clinical evidence of infection, though the presence of organisms in the blood was demonstrable, for up to two weeks after inoculation, by a polymerase chain reaction. The sera from these dogs were also reactive with both organisms in the IFA tests, but the antibody titres were four- to fivefold higher with C ruminantium than with E canis antigen. In Western blots, these sera reacted strongly with the 25 and 32 kDa antigens of C ruminantium but weakly with the 27 kDa antigen of E canis. These results provide strong evidence that the immunodominant antigens of E canis and C ruminantium contain cross-reacting epitopes.

An immunoblotting diagnostic assay for heartwater based on the immunodominant 32-kilodalton protein of Cowdria ruminantium detects false positives in field sera.

Heartwater, a major constraint to improved livestock production in Zimbabwe, threatens to invade areas which have been previously unaffected. To monitor its spread in Zimbabwe, an immunoblotting diagnostic assay based on the responses of animals to the immunodominant, conserved 32-kDa protein of Cowdria ruminantium was evaluated. In this assay, no false reactions were detected with sera known to be positive and negative, but sera from some cattle, sheep, and goats from heartwater-free areas of Zimbabwe reacted strongly with the 32-kDa protein, suggesting that either these animals had previous exposure to heartwater or they were false positives. To investigate the possibility of previous exposure to heartwater, 11 immunoblot-positive and 6 immunoblot-negative sheep from heartwater-free areas of Zimbabwe were compared regarding their susceptibilities to challenge with C. ruminantium. Prior to challenge, C. ruminantium could not be detected in any sheep by transmission to Amblyomma hebraeum ticks or by the polymerase chain reaction (PCR) conducted with plasma samples. All sheep were equally susceptible to the challenge, and infection was confirmed by brain biopsy, necropsy, PCR, and transmission of C. ruminantium to ticks. Our data suggest that the immunoblot-positive reactions of sera from heartwater-free areas were due not to previous C. ruminantium infection but rather to antigenic cross-reactivity between C. ruminantium and another agent(s) such as Ehrlichia species. In conclusion, the immunodominant 32-kDa protein is not antigenically specific to C. ruminantium and its use in serological diagnosis of heartwater requires reevaluation.

Development and persistence of Cowdria ruminantium specific antibodies following experimental infection of cattle, as detected by the indirect fluorescent antibody test.

Different breeds of cattle were experimentally infected with Palm River, a Zimbabwean isolate, or Ball-3, a South African isolate of Cowdria ruminantium, derived from tissue culture or tick or blood stabilates. C. ruminantium specific antibody responses were detected by an indirect fluorescent antibody test (IFAT) using C. ruminantium-infected bovine aortic endothelial (BAE) cell cultures as antigen. The first detection of antibodies to C. ruminantium generally coincided with the peak of the febrile reaction and the antibodies remained detectable for a period of 8-30 weeks in the Palm River infected group and 18-30 weeks in the Ball-3 infected group. Peak reciprocal antibody titres in both groups ranged from 64 to 2048 between 3 and 6 weeks post-infection. No apparent serological differences were observed among the various C. ruminantium isolates when tested in homologous and heterologous IFATs. Post-infection sera to Anaplasma marginale, Theileria parva parva, Babesia bigemina and Rickettsia conorii did not exhibit reactivity with the C. ruminantium antigen. These results indicate the possible use of C. ruminantium-infected cultures as antigen in IFATs to detect similar C. ruminantium-specific antibody responses in the field in clinically sick, recovered and carrier animals.