Assay optimization for molecular detection of Zika virus

To examine the diagnostic performance of real-time reverse transcription (RT)-polymerase chain reaction (PCR) assays for Zika virus detection.

We compared seven published real-time RT-PCR assays and two new assays that we have developed. To determine the analytical sensitivity of each assay, we constructed a synthetic universal control ribonucleic acid (uncRNA) containing all of the assays' target regions on one RNA strand and spiked human blood or urine with known quantities of African or Asian Zika virus strains. Viral loads in 33 samples from Zika virus-infected patients were determined by using one of the new assays.

Oligonucleotides of the published real-time RT-PCR assays, showed up to 10 potential mismatches with the Asian lineage causing the current outbreak, compared with 0 to 4 mismatches for the new assays. The 95% lower detection limit of the seven most sensitive assays ranged from 2.1 to 12.1 uncRNA copies/reaction. Two assays had lower sensitivities of 17.0 and 1373.3 uncRNA copies/reaction and showed a similar sensitivity when using spiked samples. The mean viral loads in samples from Zika virus-infected patients were 5 × 10⁴ RNA copies/mL of blood and 2 × 10⁴ RNA copies/mL of urine.

We provide reagents and updated protocols for Zika virus detection suitable for the current outbreak strains. Some published assays might be unsuitable for Zika virus detection, due to the limited sensitivity and potential incompatibility with some strains. Viral concentrations in the clinical samples were close to the technical detection limit, suggesting that the use of insensitive assays will cause false-negative results.