High-throughput, sensitive, and accurate multiplex PCR-microsphere flow cytometry system for large-scale comprehensive detection of respiratory viruses

Abstract

Human respiratory viruses are a diverse group of pathogens composed of hundreds of virus strains, and this presents a major challenge for diagnostic laboratories. To efficiently detect numerous viruses in a large epidemiologic study, we developed a fast, multitarget, sensitive, and specific assay named the Respiratory MultiCode-PLx Assay (RMA). The RMA utilizes improved multiplex PCR chemistry (EraGen MultiCode-PLx technology) coupled with high-throughput microsphere flow cytometry (Luminex). Eighteen sets of virus-specific multiplex PCR primers were developed based on the conserved sequences of all available respiratory-virus sequences for eight distinct groups: human rhinovirus (HRV), respiratory syncytial virus (RSV), parainfluenza virus (PIV), influenza virus (InfV), metapneumovirus, adenovirus (Ad), coronavirus, and enterovirus. Each primer set detected 20 cDNA copies of the intended target per sample and had no reaction with 60,000 copies of human genomic DNA. The accuracy and sensitivity of the RMA for detecting respiratory viruses in human samples were tested with two sets of clinical specimens. First, 101 nasal-wash specimens that were positive for HRV, RSV, InfV, PIV, or Ad by traditional techniques were reanalyzed by RMA, and all target viruses were detected with an overall sensitivity of 94% and specificity of 99%. Second, 103 nasal-wash samples from 5-year-old children with asthma and respiratory symptoms were analyzed; RMA detected viruses in 74 specimens (71.8%) compared to only 24 (23.3%) by traditional culture and immunofluorescent-staining techniques. These results show that RMA is an accurate, sensitive, and practical test for respiratory-virus infections.

Figures

FIG. 1.

The RMA consists of five main steps that all occur in the same microwell. (Step 1) Target viral cDNA is PCR amplified with a pair of virus-specific primers (1 h). One of the primers contains an iC at its 5′ terminus. (Step 2) The PCR product is labeled with a virus-specific EraCode tag (an eight-base oligonucleotide composed of both natural bases and iG) and an iG-biotin (biotinylated 2′-deoxy-iG triphosphate) with a TSE primer (30 min). Both iG and iC are recognized by natural polymerase. iG and iC pair with each other, but not with natural C and G. (Step 3) The tagged TSE product is captured by a color-addressed microsphere through the hybridization of its tag to its precise complementary oligonucleotide conjugated to the surface the microsphere (10 min at room temperature). (Step 4) The captured product is labeled with fluorescent SAPE (10 min). (Step 5) Fluorescent signal associated with each microsphere is measured in a Luminex LabMap 100 cytometer (1 h for a 96-well plate).

FIG. 2.

Detection of target viruses by the RMA is highly sensitive and specific. Each well, containing 20 copies of cDNA of the target genomic regions (Table 2) of the indicated target virus, panel A (RSV A and B; PIV1, -2, -3, -4a, and -4b; InfV A and B; MPV; and HRV1A) and panel B (Ad3 [group B], Ad1 [group C], Ad4 [group E], CoV 229E, CoV NL63, CoV OC43, EV68, EV69, EV70, EV71, CA22 [coxsackievirus A22], and CB1 [coxsackievirus B1]), was analyzed with the RMA. Human genomic DNA (60,000 copies per well) was used as the negative control (NC). Each bar represents the average MFI of triplicate samples. The error bars indicate standard deviations.

FIG. 3.

Improved RNA extraction and cDNA synthesis conditions for the RMA. Samples containing the indicated numbers of HRV16 virions were processed by both the old and new protocols. Each cDNA preparation was subjected to RMA, and each bar represents the average MFI of triplicate samples. The error bars indicate standard deviations.