Specificity of dot hybridization assay in the presence of rRNA for detection of rotaviruses in clinical specimens

Abstract

Nucleic acid hybridization is used to identify viral genomic sequences in clinical and environmental samples. However, RNA virus genomes have been reported to hybridize to mammalian rRNA from uninfected cells under stringent conditions, and caution has therefore been advised in the use of nucleic acid probes for detection of RNA viruses. To evaluate the effect of rRNA on a diagnostic assay for an RNA virus, we tested the specificity of a rotavirus dot hybridization assay with clinical specimens which contained eucaryotic rRNA. The cDNA probe used in this assay contained sequences complementary to all 11 rotavirus genes. Preliminary experiments indicated that hybridization between rRNA and the cDNA probe occurred only with greater than 50 ng of rRNA, and this interaction was easily distinguished from the hybridization of the rotavirus probe with homologous or heterologous strains of the same rotavirus group. When 95 clinical specimens were tested, the rotavirus dot hybridization assay had a specificity of 98.8%. The predictive value of a negative test was 94.2%, although nearly all of the specimens contained rRNA and also reacted with an rRNA probe. Although the specificity of all dot hybridization assays should be individually evaluated, we conclude that our dot hybridization assay for rotaviruses is highly specific even in the presence of quantities of rRNA that may be anticipated in extracts of fecal specimens.