Sequence polymorphism occurs throughout the genomes of many viruses, and any one viral species can comprise multiple genotypes exhibiting considerable sequence variation. Furthermore, within any single patient population or between different populations different strains of a viral species may circulate. If polymorphism occurs in the primer or probe target sites on the viral genome, then the PCR assay may have a lower sensitivity in its ability to detect these strains or at worse fail to detect the virus strain altogether.
Recently, we compared a 5' nuclease assay targeting the parainfluenza 3 nucleocapsid protein gene (WhSl-para3-5N; unpublished data) with a previously described 5' nuclease assay targeting the parainfluenza 3 hemag-glutinin-neuraminidase gene (HN-para3-5N; Watzinger et al., 2004) for the detection of parainfluenza 3 in nasopharyngeal aspirate specimens collected from a local patient population. During the course of the study, 34 positive samples were identified with both assays detecting 33 of these. The cycle threshold (Ct) values for these 33 positive samples were comparable in both assays, with an average difference of only 1.1 cycles. However, one sample was negative by the WhSl-para3-5N assay but was positive by the HN-para3-5N assay, providing a Ct value of 25. Such a low Ct value indicates that the viral load was relatively high in this specimen and therefore the false-negative result could not be explained by the differences in the detection limit of each assay. The presence of parainfluenza 3 in this specimen was confirmed by a direct fluorescence assay (DFA), yet the specimen consistently provided negative results in the WhSl-para3-5N assay. Sequencing of the viral genome that was targeted by the TaqMan® probe showed four mismatches. These were almost certainly responsible for the false-negative result (see Troubleshooting guide 14.1).
The WhSl-para3-5N assay was designed using the parainfluenza 3 sequences for the nucleocapsid protein gene currently available on the Genbank sequence database. Based on this information, the primer and probe targets appeared to be well conserved and certainly the mismatches identified above were not represented in Genbank sequences. Quite simply, this demonstrates that the use of sequence information from public databases has limitations for viral assay design. Furthermore, the above results highlight that sequence polymorphism between viral strains circulating in a population can lead to false-negative results in PCR assays, and that these results may occur irrespective of viral load. Overall, our experience indicates that sequence variation is a more widespread problem for the design and performance of viral PCR assays. However, the occurrence of false-negative results due to sequence variation can usually be identified by the use of a second assay and confirmed by gene sequencing.
Was this article helpful?