Evaluation of an enzyme-linked immunosorbent assay for the detection of ectromelia (mousepox) antibody

Ectromelia virus, an orthopoxvirus that can cause extensive morbidity and mortality (mousepox) in colonized mice, has been epizootically responsible for serious disruption of biomedical research since 1930. The lack of a sensitive and specific serological assay for infection with this virus became apparent during outbreaks of mousepox at the National Institutes of Health, Bethesda, Md., and other biomedical research institutions in 1979 and 1980. To fill this need, we evaluated an enzyme-linked immunosorbent assay. Sucrose gradient-purified ectromelia and vaccinia viruses were compared as antigens in tests on approximately 1,000 mouse sera from experimentally infected mice and conventional colonies of uninfected mice. A statistical analysis based on the frequency distribution of the absorbance values for 152 mouse sera (free of ectromelia antibody) gave 0.22 as a value to differentiate ectromelia-positive sera from ectromelia-negative sera. When enzyme-linked immunosorbent assay results were compared with those obtained by an indirect immunofluorescence assay, the former was found to be at least 10-fold more sensitive. With the procedures employed, including the use of purified vaccinia virions as antigen, the enzyme-linked immunosorbent assay proved to be highly sensitive and specific for detecting antibodies to ectromelia and vaccinia viruses. False-positive results were not encountered. False-negative results were observed in 3% of 108 separate tests of a known positive serum. Although data indicated that ectromelia antibody can be differentiated from vaccinia antibody with homologous and heterologous antigen, this procedure probably cannot be generally used because of unavailability of ectromelia antigens.