Comparative detection of rabies RNA by NASBA, real-time PCR and conventional PCR.

Five methods for the RNA detection of rabies virus were directly compared in this study. These included conventional nucleic acid sequence-based amplification with electrochemiluminescence (NASBA-ECL) assay, reverse transcription (RT)-heminested (hn) polymerase chain reaction (PCR) and TaqMan real-time RT-PCR using protocols as described previously. The first two methods have been routinely utilised for ante-mortem diagnosis of human rabies in Thailand and other rabies-endemic Asian and African countries. In addition, two real-time NASBA assays based on the use of a NucliSens EasyQ analyser (NASBA-Beacon-EQ) and LightCycler real-time PCR machine (NASBA-Beacon-LC) were studied in parallel. All methods target the N gene, whereas the L gene is used for RT-hnPCR. Using serial dilutions of purified RNA from rabies-infected dog brain tissue to assess sensitivity, all five methods had comparable degrees of sensitivities of detection. However, both real-time NASBA assays had slightly lower sensitivities by 10-fold than the other three assays. This finding was also true (except for TaqMan real-time RT-PCR due to a mismatch between the target and probe sequences) when laboratory-adapted (challenge virus standard-11) virus was used in the assays. Testing on previously NASBA-ECL positive clinical samples from 10 rabies patients (saliva [6] and brain [4]) and 10 rabies-infected dog brain tissues, similar results were obtained among the five methods; real-time NASBA assays yielded false-negative results on 2 saliva samples. None of the assays showed positive results on cerebrospinal fluid specimens of 10 patients without rabies encephalitis. Due to the unavailability of the NASBA-ECL assay, the results show that TaqMan real-time RT-PCR and RT-hnPCR can be useful for ante- and post-mortem diagnosis of rabies.