ABSTRACT
Enzyme-linked immunosorbent assay (ELISA) for human granulocytic ehrlichiosis (HGE) using two different recombinant P44 proteins (rP44 and rP44-2hv) of the HGE agent as antigens was evaluated. Sera from a total of 72 healthy humans both from regions where HGE is nonendemic and regions where HGE is endemic were used as negative controls to determine the cutoff value for ELISA. Sera from a total of 14 patients (nine from whom the HGE agent was isolated and five who were HGE-PCR positive) were used as positive controls. One hundred nine sera from 72 patients in an area where HGE is endemic who were suspected of having HGE were examined by ELISA and indirect immunofluorescence assay (IFA). All IFA-negative sera were negative by both ELISAs. Of 39 sera that were IFA positive, 35 and 27 were positive by ELISA using rP44 and rP44-2hv, respectively, indicating that the use of rP44 is more sensitive. Western blot analysis of the four rP44-ELISA-negative IFA-positive sera using whole HGE agent as antigen suggests that these four sera were false IFA positive. There was no difference in results with or without the preabsorption of sera with Escherichia coli or with or without the cleavage of the fused protein derived from the vector. There was a significant positive correlation between IFA titers and optical densities of ELISAs. Four Ehrlichia chaffeensis-positive and 10 Borrelia burgdorferi-positive sera were negative by ELISA. However, twoBabesia microti-positive sera showed strong cross-reactivity to the fused vector protein, which was eliminated after cleavage of the protein. Thus, ELISA using rP44 nonfusion protein would provide a simple, specific, and objective HGE serologic test which can be easily automated.
Comparison of Commercial Enzyme-Linked Immunosorbent Assay and Immunofluorescence Assay for Diagnosis of Acute Rickettsia typhi Infections
