Shedding and reversion of oral polio vaccine type 3 in Mexican vaccinees: comparison of mutant analysis by PCR and enzyme cleavage to a real-time PCR assay

Abstract

A uracil-to-cytosine mutation at nucleotide position 472 of oral poliovirus vaccine type 3 (OPV3) contributes to the development of vaccine-associated paralytic poliomyelitis (VAPP). To analyze OPV3 shedding patterns, we previously used the multistep method of mutant analysis by PCR and enzyme cleavage (MAPREC). This involves conventional reverse transcription-PCR to detect OPV3, followed by a restriction digest to quantify position 472 reversion. Real-time PCR detects and quantifies nucleic acid as PCR occurs and avoids postreaction processing. The goal of this study was to compare a real-time PCR method to MAPREC. Seventy-three stool samples from Mexican OPV recipients underwent the reverse transcription-PCR step of MAPREC and real-time PCR. Real-time PCR identified 23% more OPV3-positive samples than conventional reverse transcription-PCR. When reversion was compared, the revertant proportion (RP), defined as the percentage of revertants in a sample, differed by < or =10% in 21/25 (84%) samples. The four samples differing by >10% were obtained within 5 days of OPV administration. The real-time PCR assay identified samples with an RP of > or =85% with 94% sensitivity and 86% specificity compared to MAPREC. The mean difference in RP between the two methods was 3.6% (95% confidence interval, -0.3 to 7.5%). Real-time PCR methods reliably detect OPV3, and reversion estimates correlate more consistently with MAPREC when OPV3 reversion rates are high. Detecting VAPP-related mutations by real-time PCR is rapid and efficient and can be useful in monitoring ongoing global polio eradication efforts.

FIG. 1.

Efficiencies of the real-time PCR assays for revertant and nonrevertant strains. Revertant (b) and nonrevertant (a) plasmid control constructs were diluted serially by a factor of 10, and C_T values obtained from each dilution were plotted. Both slopes correspond to efficiencies of 90 to 100%.

FIG. 2.

Discrimination of the real-time assays. C_T values for revertant and nonrevertant templates differed by 3 to 6 in the assay for nonrevertants (a) and by 5 to 6 in the assay for revertants (b).

FIG. 3.

Comparison of RPs obtained by real-time assays with known RPs in prepared construct mixtures. Mixtures of revertant and nonrevertant constructs underwent several runs in real-time PCR assays.

FIG. 4.

Comparison of real-time PCR to MAPREC in clinical samples. MAPREC generally predicted a higher RP, but a closer correlation between the two methods was seen when the RP was high.