- Journal List
- J Clin Microbiol
- v.36(11); 1998 Nov
- PMC105291
As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health.
Learn more: PMC Disclaimer | PMC Copyright Notice
J Clin Microbiol. 1998 Nov; 36(11): 3149–3154.
PMCID: PMC105291
PMID: 9774555
Carla Osiowy*
Author information Article notes Copyright and License information PMC Disclaimer
Abstract
Diagnosis of respiratory virus infections currently involves detection by isolation or antigen detection, which usually identifies only a single suspected agent. To permit identification of more than one respiratory virus in clinical specimens, a rapid detection method involving a single-step, multiplex reverse transcription-PCR (RT-PCR) assay was developed. The assay included five primer sets that amplified the RNA of respiratory syncytial virus subtypes A and B, parainfluenza virus types 1, 2, and 3, and adenovirus types 1 to 7. Initially the assay was tested on tissue culture-grown virus and was found to be specific for all 12 prototype viruses tested, with no interassay cross amplification or amplification of other respiratory viruses. Assay sensitivity allowed a detection range of 0.2 50% tissue culture infectious dose (TCID50) for adenovirus to 250 TCID50 for parainfluenza virus type 1. The multiplex RT-PCR assay was also able to directly detect viruses in respiratory specimens, with virus being detected in 41 of 112 samples as compared to 34 of 112 samples detected by direct immunofluorescence or antigen detection following specimen culture. This suggests that the multiplex RT-PCR assay can be used as a rapid and sensitive diagnostic method for major respiratory viruses.
Respiratory infections caused by respiratory syncytial virus (RSV), parainfluenza virus (PIV), or adenovirus may result in severe lower respiratory tract disease requiring hospitalization (12). Indeed, the leading cause of severe lower respiratory tract infection in infants and young children is RSV (18), followed by PIVs (3). Adenoviruses also contribute significantly to endemic and epidemic respiratory disease (23), with an estimated 10% of all cases of childhood pneumonia due to adenovirus infection (17). These viruses, particularly RSV, also contribute to considerable morbidity in the adult population and in immunocompromised individuals (4, 14). For these reasons, a rapid, sensitive, and specific diagnostic tool is important for management of patients presenting with a respiratory infection (1, 26).
Direct antigen testing provides rapid results; however, it often lacks sensitivity and thus requires confirmation by virus isolation or indirect antigen testing following specimen culture (7). As well, specimen integrity and the number of intact cells present in the specimen are crucial for a reliable direct immunofluorescence assay (DIF) (22). In the case of RSV, direct antigen tests are often found to lack sensitivity for specimens obtained from older children and adults (6). Direct antigen tests may also fail to detect emerging variants having altered amino acid sequences on envelope or outer capsid proteins (24).
In the current study, a new molecular diagnostic technique was developed to permit the rapid and sensitive detection of the three major groups of respiratory viruses involved in lower respiratory tract infection and hospitalization. The reverse transcription-PCR (RT-PCR) assay developed in the present study is similar to previously published methods for the detection of multiple respiratory viruses (5, 8, 9, 25); however, the present assay permits detection of RSV, PIV type 1 (PIV1), PIV2, PIV3, and respiratory tract-associated adenoviruses, in a single-step multiplex RT-PCR. The various virus types are differentiated by their unique amplicon sizes following separation of PCR products on agarose gels. Subtype characterization of PCR products by hybridization with subtype-specific probes, restriction digestion, or sequencing is also possible. In order to determine the utility of such an assay, the multiplex RT-PCR assay was used to detect respiratory viruses in clinical specimens and compared to detection by DIF or indirect immunofluorescence assay (IIF) following specimen culture. Results suggest that the multiplex RT-PCR assay is a rapid, specific, and sensitive method of testing for multiple respiratory viruses in individual clinical specimens.
(This material was presented in part at the Annual Meeting of the American Society for Virology, Vancouver, British Columbia, Canada, July 1998.)
MATERIALS AND METHODS
Virus culture and respiratory specimens.
Respiratory virus strains (RSV subtype A [RSV-A] [Long strain], ATCC VR-26; RSV-B [strain 9320], ATCC VR-955; PIV1, ATCC VR-94; PIV2, ATCC VR-92; PIV3, ATCC VR-93; adenovirus type 1, ATCC VR-1; adenovirus type 2, ATCC VR-846; adenovirus type 3, ATCC VR-3; adenovirus type 4, ATCC VR-4; adenovirus type 5, ATCC VR-5; adenovirus type 6, ATCC VR-6; adenovirus type 7, ATCC VR-7) were obtained from the American Type Culture Collection for development of the multiplex RT-PCR assay. RSV strains and all adenovirus subtypes were cultured in HEp-2 cells (ATCC CCL-23), while PIV subtypes were cultured in LLC-MK2 cells (ATCC CCL-7). Cells were maintained in minimal essential medium supplemented with 10% fetal calf serum and antibiotics (penicillin G at 100 U/ml and streptomycin at 100 μg/ml).
Original respiratory specimens submitted to (i) the Children’s Hospital of Eastern Ontario and (ii) the Virology Laboratory, Laboratory Services Branch of the Ontario Ministry of Health, from February 1996 to February 1998 for virus isolation and antigen detection were requested for detection of virus nucleic acid by RT-PCR. A total of 112 respiratory specimens (84 nasopharyngeal swabs, 21 nasopharyngeal aspirates, 4 throat swabs, 2 nasal swabs, and 1 bronchoalveolar lavage sample) were sent to the Laboratory Center for Disease Control (Ottawa, Ontario, Canada), on ice, and were immediately stored at −70°C. Most specimens were stored at the participating laboratories at 4°C for various periods prior to their being received (range, 1 to 21 days), with the majority being stored for approximately 1 week. Specimens were normally processed for RT-PCR within a week of reception; however, 25 of the 112 specimens were stored for approximately 2 years at −70°C prior to their being processed. All specimens were coded by the participating laboratories to prevent investigator bias, but were known to contain a variety of specimens, either negative or positive for various respiratory viruses as determined by antigen detection and specimen culture. Participating laboratories normally performed DIF or enzyme immunoassay for various respiratory viruses, including RSV, PIV1 to -3, and adenovirus, upon the receipt of a specimen. If the specimen was virus positive by DIF, no further testing was carried out and the result was reported. Specimens negative by DIF were cultured on human fetal lung and rhesus monkey kidney cells. At 24 and 48 h postinoculation, hemadsorption and immunofluorescence testing were performed on cultured specimens. If the result was still negative, cultures were continued for 8 days and testing was repeated.
Nucleic acid extraction.
RNAs from infected cultured cells and respiratory specimens were extracted with Trizol LS reagent (Gibco Laboratories, Burlington, Ontario, Canada) according to the manufacturer’s suggested method. Approximately 2 × 106 infected cells or 100 μl of respiratory specimen was extracted, and the final RNA pellet was resuspended in 5 μl of diethylpyrocarbonate (DEPC)-treated water. RNA extracts were placed on ice and used immediately for RT-PCR.
Multiplex RT-PCR.
Five sets of oligonucleotide primers were designed for RT and amplification of (i) the nucleocapsid gene of RSV, (ii) the nucleocapsid gene of PIV1, (iii) the nucleocapsid gene of PIV2, (iv) the nucleocapsid gene of PIV3, and (v) the hexon genes of adenovirus types 1 to 7, according to nucleotide sequences available from GenBank (see Table Table1).1). Primers for RSV were designed visually from a highly conserved area of a nucleotide sequence alignment of RSV-A and -B nucleocapsid genes. Adenovirus primers were designed visually from a highly conserved area of nucleotide sequence alignment of the hexon genes from adenovirus types 2, 3, 4, 5, and 7. No published sequences were available for adenovirus types 1 and 6 hexon genes at the time the primers were designed. All primer sets were designed to have similar melting temperatures (range, 65 to 70°C) and a higher G+C content to allow a higher annealing temperature to be used during amplification. Primer sequences were analyzed for suitability by using PC/GENE sequence analysis software (release 6.8; Intelligenetics, Inc.).
TABLE 1
Sequences of oligonucleotide primers and probes used for detection of viruses in this study
Primer or probe | Genea | Positionb | Sequencec |
---|---|---|---|
Primers | |||
RSVN3 | N | 426–451 | GGGAGAGGTGGCTCCAGAATACAGGC |
RSVN5 | N | 748–773 | AGCATCACTTGCCCTGAACCATAGGC |
PIV1PR3 | NP | 64–89 | TCTGGCGGAGGAGCAATTATACCTGG |
PIV1PR5 | NP | 122–147 | ATCTGCATCATCTGTCACACTCGGGC |
PIV2PR3 | NP | 360–385 | AACTATGTCCAGAGGAGAGGTGCTGG |
PIV2PR5 | NP | 498–523 | CCATGCCTGCATAAGCACACTGTAGC |
PIV3PR3 | NP | 416–441 | ACCAGGAAACTATGCTGCAGAACGGC |
PIV3PR5 | NP | 624–649 | GATCCACTGTGTCACCGCTCAATACC |
ADHEX3 | Hexon | 154–179d | CCTACGCACGATGTGACCACAGACCG |
ADHEX5 | Hexon | 343–368d | GTGTTGTAGGCAGTGCCGGAGTAGGG |
Probes | |||
rsva | 588–612 | TTACAAAGGCTTACTACCCAAGGAC | |
rsvb | 588–612 | CTACAAGGGCCTCATACCAAAGGAT | |
piv2 | 395–419 | CATTAGCTGAGGACATTCCTGATAC | |
piv3 | 482–506 | GGTATCCATCATGTTTAGGAGCTCT | |
adno | 295–321d | ACGTACTTTGACATCCGGGGCGTGC |
Open in a separate window
aN, nucleocapsid gene of RSV; NP, nucleocapsid gene of PIV.
bRelative to the translation start site.
cSequences are shown 5′ to 3′.
dPositions shown are according to the adenovirus type 5 hexon gene sequence.
Both steps of the multiplex RT-PCR were performed in the same reaction tube with a single reaction buffer from a formulation previously published (10). Each reaction tube contained the following in a final volume of 50 μl: 0.2 mM concentrations (each) of dATP, dCTP, dGTP, and dTTP (Boehringer Mannheim, Laval, Quebec, Canada); 0.2 μM concentrations (each) of RSV-specific primers (RSVN3 and RSVN5) and PIV1-specific primers (PIV1PR3 and PIV1PR5); 0.1 μM concentrations (each) of PIV2-, PIV3-, and adenovirus-specific primers (PIV2PR3 and PIV2PR5, PIV3PR3 and PIV3PR5, and ADHEX3 and ADHEX5, respectively); 8 U of RNase inhibitor (Boehringer Mannheim); reaction buffer (50 mM KCl, 1.5 mM MgCl2, 0.1% Triton X-100, 10 mM Tris [pH 9.0]); 10 U of Expand reverse transcriptase (Boehringer Mannheim); and 0.5 U of ID-PROOF DNA polymerase (ID Labs Biotechnology, London, Ontario, Canada). To minimize pipetting variables and reduce nonspecific priming, separate master premixes were made and mixed just prior to RT-PCR. RT-PCR was performed in a Perkin-Elmer (Norwalk, Conn.) GeneAmp 9600 thermocycler under the following conditions: 42°C for 30 min and 94°C for 2 min, 10 cycles of 94°C for 40 s, 68°C for 30 s, and 72°C for 45 s and 25 cycles of 94°C for 40 s, 68°C for 30 s, and 72°C for 45 s (with a 5-s primer extension added per cycle). A final primer extension step of 5 min at 72°C completed the thermocycling program.
Positive controls for the multiplex RT-PCR assay included RNA extracted from RSV-A-, adenovirus type 5-, or PIV3-infected cells, as well as cloned PCR products. All amplification products, resulting from RNA extracted from cells infected with one of the 12 prototype viruses, were cloned in the pGEM-T cloning vector (Promega Corporation, Madison, Wis.), and 10 to 100 plasmid copies of the various clones were used randomly as an additional positive control. Specific cloned products (at approximately 100 copies) were also used to spike RNA extracts to test for PCR inhibitors.
In certain cases, confirmatory tests were run to verify the initial results. These tests included nested PCR and RT-PCR of specimen extracts with primers hybridizing a region of the viral genome different from that hybridized by the five multiplex primer sets. Nested PCR was performed following multiplex RT-PCR on specimens suspected of having RSV or adenovirus amplicons. For each nested PCR, a 0.1 μM concentration of each primer was used (for RSV, RSVNST3 [5′ CTGGTAGAAGATTGTGC 3′] and RSVNST5 [5′ ACTAAGTTAGCAGCAGG 3′]; for adenovirus, ADNST3 [5′ TAGGACCTCTGTCAAGC 3′] and ADNST5 [5′ TACTCGTACAAAGCTCG 3′]), 2 μl of the first-run RT-PCR product was added, and an annealing temperature of 50°C was used. RT-PCR of specimen extracts was also performed with previously published primer sequences for the PIV3 F gene (13) and the adenovirus hexon gene (11).
To prevent possible PCR contamination and false-positive results, many precautions were taken as detailed previously (19). Negative controls included template-free reaction tubes as well as RNA extracted from uninfected HEp-2 and LLC-MK2 cells. RNAs extracted from respiratory specimens known to be virus negative or containing influenza virus A or B were also included as negative controls.
Detection of amplified products.
Amplified products were electrophoretically separated on 3% NuSieve agarose (FMC Bioproducts, Guelph, Ontario, Canada) gels, for purposes of differentiating virus-specific bands, or 1% agarose (Gibco Laboratories) gels, for Southern blotting. The gels were then stained with ethidium bromide and visualized under UV light.
Amplicon DNA transferred to nylon membranes was detected nonradioactively with oligonucleotide probes 3′-end-labeled with fluorescein-11-dUTP (ECL 3′ oligolabeling and detection system; Amersham Life Science, Oakville, Ontario, Canada). A fluorescein-11-dUTP-labeled DNA marker (Lambda DNA digested with EcoT14I; Amersham Life Science) was included on all 1% gels to assist in the determination of amplicon size. Probes were designed internal to the amplicon sequence, and in the case of RSV, subtype-specific probes were prepared (see Table Table1).1). Probes were added to hybridization solutions at 5 ng/ml, and all probes were hybridized at 50°C. RSV-A amplicons were specifically detected with the rsva probe, while RSV-B amplicons were specifically detected with the rsvb probe. PIV2 and PIV3 amplicons were specifically detected with piv2 and piv3 probes, respectively, and adenovirus amplicons were specifically detected with the adno probe. Labeling, hybridization, and detection were performed according to the manufacturer’s protocol. In certain cases following detection, blots were stripped of the original probe, according to the manufacturer’s protocol, and reprobed with a different virus-specific probe.
RESULTS
Antigen detection in respiratory specimens.
A total of 112 respiratory specimens were received from participating laboratories for the detection of viral nucleic acid by multiplex RT-PCR. It was observed, following breaking of the specimen code, that most RSV-positive specimens were most often reported positive following direct antigen testing, such as DIF or enzyme immunoassay. Respiratory viruses other than RSV or specimens having no detectable viruses were most often reported following specimen culture and indirect antigen testing. Of the 112 specimens assayed by antigen testing and culture, 29 were positive for RSV, 3 were positive for PIV3, 2 were positive for adenovirus, 4 were positive for herpes simplex virus type 1, 19 were positive for influenza virus type A, 3 were positive for influenza virus type B, and 3 were positive for rhinovirus, and no virus was detected for 50 specimens. In one specimen, both influenza virus type A and RSV were detected.
Multiplex RT-PCR of five respiratory virus types.
Oligonucleotide primers for multiplex RT-PCR (Table (Table1)1) were designed to permit high annealing temperatures for increased specificity and to allow the use of Tth DNA polymerase in both steps of RT-PCR (16). The high temperatures, at which Tth DNA polymerase exhibits stable RT activity, help decrease secondary structures present in the RNA template. Alignment of all strain sequences derived from GenBank searches for the viruses used in this study was used to produce consensus primer sequences for each virus type. The RSV primer set amplified a 348-bp region of the nucleocapsid gene of both RSV-A and -B (Fig. (Fig.1A),1A), which could be differentiated by subtype-specific probes (Fig. (Fig.1B).1B). Primer dimers are visible as the lower of the two bands observed in Fig. Fig.1A,1A, lanes 2 and 3. When a higher percentage of agarose is used for electrophoresis, primer dimers are much fainter and do not interfere with migration of virus-specific bands (see Fig. Fig.2,2, lane 3, for example). PIV primer sets amplified an 84-, 164-, and 234-bp region of the nucleocapsid genes of PIV1, PIV2, and PIV3, respectively, while the adenovirus primer set amplified a 215-bp region of the hexon gene of adenovirus types 1 to 7 (data not shown). Agarose gel bands of the correct size were verified by Southern blotting (for RSV, PIV2, PIV3, and adenoviruses) and restriction digestion with MboII (for PIV1) and AgeI, MboII, HaeII, and ApaI (for differentiation of individual adenoviruses) (data not shown).
FIG. 1
Multiplex RT-PCR detection of RSV-A and RSV-B. (A) RNA extracted from RSV-A- or RSV-B-infected HEp-2 cells was reverse transcribed and amplified under the conditions described in Materials and Methods. PCR products were separated on a 1% agarose gel which was stained with ethidium bromide and visualized under UV light. Lanes: 1, fluorescein-11-dUTP-labeled base pair marker; 2, RSV-A; 3, RSV-B. (B) Southern hybridization of the same gel in panel A hybridized to RSV subtype-specific oligonucleotide probes 3′ end labeled with fluorescein-11-dUTP. The blot was stripped and reprobed with the second probe according to procedures described in Materials and Methods. Blot 1, probed with rsva; blot 2, probed with rsvb.
FIG. 2
Single-tube, multiplex RT-PCR detection of five respiratory virus types. RNA extracted from RSV-A-, PIV3-, adenovirus type 5-, PIV2-, and PIV1-infected cells was pooled and reverse transcribed and amplified in a single tube under conditions described in Materials and Methods. RNA extracted from uninfected HEp-2 and LLC-MK2 cells was pooled and used as a negative control. PCR products were separated on a 3% NuSieve agarose gel that was stained with ethidium bromide and visualized under UV light. Lanes: 1, Template-free negative control; 2, pooled, uninfected-cell RNA; 3, pooled, infected-cell RNA; 4, 100-bp ladder DNA marker (Boehringer Mannheim). Expected sizes for the five virus type amplicons are as follows: for RSV, 348 bp; for PIV3, 234 bp; for adenovirus, 215 bp; for PIV2, 164 bp; and for PIV1, 84 bp.
Single-tube, multiplex RT-PCR was developed through initial optimization of each individual component of the process. First, RNA from all prototype viruses was amplified separately to ensure buffer and temperature conditions were appropriate. An annealing temperature of 68°C was determined to be optimal; however, an annealing temperature as high as 70°C was observed to permit amplification. Second, single-tube RT-PCR conditions were optimized. This involved experimenting with several one-step procedures including EZ rTth polymerase (Perkin-Elmer) and Titan RT-PCR (Boehringer Mannheim). The buffered two-enzyme approach used in the present study was the most cost-effective and reproducible assay system investigated. The third optimization step involved multiplex conditions. Primer concentrations in each reaction mixture were manipulated such that amplification of equal amounts of RNA provided similar band intensities. Figure Figure22 demonstrates that all five amplified products are present and are easily distinguishable following single-tube multiplex RT-PCR of RSV-A, PIV3, adenovirus type 5, PIV2, and PIV1. A higher-molecular-weight product which migrates close to the RSV-specific band is visible in the pooled, uninfected-cell RNA lane (lane 2). This band was determined to result from nonspecific amplification of the cellular nucleic acid, as the product migrates faster than the RSV amplicon and is not reproducibly observed in amplification products from uninfected cell RNA.
Specificity and sensitivity of multiplex RT-PCR.
The specificity of the five multiplex primer sets was tested by amplification of RNA from influenza virus types A and B, measles virus, herpes simplex virus type 1, and rhinovirus. No specific amplification was obtained, nor was any interassay cross amplification observed when the five multiplex primer sets were used in combination with any of these viruses (data not shown). Assay sensitivity was determined by amplification of extracted RNA from duplicate serial 10-fold dilutions of each virus type 50% tissue culture infectious dose (TCID50) stock (stock titer: RSV-A, 108.9 TCID50/ml; adenovirus type 5, 104.4 TCID50/ml; PIV1, 103.4 TCID50/ml; PIV2, 104.9 TCID50/ml, PIV3, 106.7 TCID50/ml). Amplified products detected by agarose gel electrophoresis were observed at various dilutions for each virus type and corresponded to a calculated minimal amount of detectable virus RNA of 5 TCID50 for RSV, 0.2 TCID50 for adenovirus, 250 TCID50 for PIV1, 10 TCID50 for PIV2, and 30 TCID50 for PIV3.
RNA detection in respiratory specimens.
Respiratory specimens were received from both participating laboratories under code for the purposes of detecting respiratory viruses by RT-PCR. All specimens were processed, and positive results were confirmed by duplicate extraction or Southern hybridization prior to the code’s being broken. For the five respiratory virus types assayed in the present study, 30% of specimens (34 of 112) were positive by DIF-IIF, while 37% of specimens (41 of 112) were positive by multiplex RT-PCR. RSV was detected most often by both methods (25%; 28 of 112), while no PIV1 or PIV2 was detected by either method. Table Table22 shows the comparison between the two methods for virus detection in respiratory specimens. Multiplex RT-PCR correlated very closely to direct antigen detection of RSV, with only one specimen being detected by RT-PCR but not DIF, and one specimen being detected by DIF but not RT-PCR. PIV3 RNA was detected in five specimens, three of which were also positive by DIF-IIF.
TABLE 2
DIF-IIF and multiplex RT-PCR detection of respiratory viruses in respiratory specimens
Results (RT-PCR/DIF-IIF)a | No. of specimens with indicated results | ||||
---|---|---|---|---|---|
RSV | PIV1 | PIV2 | PIV3 | Adenovirus | |
+/+ | 28 | 0 | 0 | 3 | 0 |
+/− | 1 | 0 | 0 | 2 | 7 |
−/− | 82 | 0 | 0 | 107 | 103 |
−/+ | 1 | 0 | 0 | 0 | 2 |
Open in a separate window
aThe four possible combinations of RT-PCR and DIF-IIF results. +, positive; −, negative.
Comparison of multiplex RT-PCR detection results to detection by DIF-IIF as a “gold standard” gave a sensitivity value for RT-PCR of 91%, a specificity value of 87%, and positive and negative predictive values of 76 and 96%, respectively. Individually, respective sensitivity and specificity values were as follows: for RSV RT-PCR detection, 97 and 99%; for PIV3 RT-PCR detection, 100 and 98%; and for adenovirus RT-PCR detection, 0 and 94%. Interestingly, the only two specimens positive for adenovirus by DIF-IIF were consistently negative by RT-PCR and Southern hybridization; however, seven other specimens having a negative result by DIF-IIF were positive by RT-PCR for adenovirus. All RT-PCR results that were discrepant from DIF-IIF results were rigorously verified by replication, Southern hybridization, nested PCR (for RSV and adenovirus), and PCR with a different primer set (PIV3 and adenovirus). As well, all negative extraction and RT-PCR controls remained negative throughout the study. In all cases of verification, the results confirmed the initial RT-PCR result, therefore suggesting that positive RT-PCR results discrepant from DIF-IIF were true positives not detected by antigen detection or culture.
A representative group of respiratory specimens, as detected by agarose gel electrophoresis and Southern hybridization following multiplex RT-PCR, is shown in Fig. Fig.3.3. The results shown in Fig. Fig.3A3A and B include specimens virus negative by both antigen detection and RT-PCR (lanes 1 and 5), an antigen-negative but RT-PCR adenovirus-positive specimen (lane 3), virus-positive specimens detected by both antigen detection and RT-PCR (lanes 2, 4, and 6), and a dually virus-positive specimen (lane 6).
FIG. 3
Multiplex RT-PCR detection of respiratory viruses in respiratory specimens. (A) Respiratory specimens were processed for detection of viral RNA by multiplex RT-PCR as described in Materials and Methods. PCR products were separated on a 1% agarose gel which was stained with ethidium bromide and visualized under UV light. The results from a representative group of respiratory specimens are shown. The antigen detection result and, in parentheses, the RT-PCR result for each specimen were as follows: lane 1, virus negative (virus negative); lane 2, RSV (RSV); lane 3, virus negative (adenovirus); lane 4, RSV (RSV); lane 5, virus negative (virus negative); and lane 6, PIV3 (PIV3 and adenovirus). (B) Southern hybridization of the same gel in panel A hybridized to oligonucleotide probes 3′ end labeled with fluorescein-11-dUTP. The blot was stripped and reprobed with each individual probe according to procedures described in Materials and Methods. The sizes of a fluorescein-labeled DNA marker (Lambda DNA digested with EcoT14I; Amersham Life Science) run on the transferred gel are shown to the right of each reprobed blot. (C) Gel (3% NuSieve agarose) of multiplex RT-PCR products from the PIV3 antigen-positive respiratory specimen shown in lane 6 of panel A. Lanes: 1, 100-bp ladder DNA marker; 2, respiratory specimen positive for PIV3 by antigen detection (PIV3-specific amplicon band, 234 bp; adenovirus-specific amplicon band, 215 bp).
Multiple respiratory viruses were observed in four specimens: (i) PIV3 (detected by DIF-IIF and RT-PCR) and adenovirus, (ii) influenza virus type A and adenovirus, (iii) influenza virus type A and RSV (detected by DIF-IIF and RT-PCR), and (iv) rhinovirus and adenovirus. The PIV3-adenovirus dual-infection specimen is included in Fig. Fig.3A3A and B (lane 6), with the PCR products from this extract also shown separated on a 3% NuSieve agarose gel (Fig. (Fig.3C)3C) to illustrate the presence of both bands.
DISCUSSION
Current rapid methods for detection of respiratory viruses mainly involve antigen detection. Direct antigen testing of specimens can provide a result within hours of receipt with variable levels of sensitivity (6, 22). However, it is still recommended that rapid antigen testing be confirmed by culture and indirect antigen detection in order to increase the level of sensitivity (15), thus prolonging the diagnostic procedure from approximately 1 to 7 days. This type of testing is often further limited by assaying for only a single infectious agent, therefore preventing the detection of multiple respiratory infections or agents for which testing was not requested. Newer molecular diagnostic techniques have been developed which have increased sensitivity over rapid antigen detection or culture (7, 11, 20); however, they are still limited by detection of a single respiratory virus. The present study aimed to develop a rapid detection method that permitted identification of more than one respiratory virus in a single specimen. The method developed in this study is the first to include a five-primer multiplex set for one-step RT-PCR amplification within a single tube. The multiplex RT-PCR was capable of detecting all five major respiratory virus types in a single pooled extract, without requiring further hybridization steps for identification. Further characterization and subtyping are then possible by hybridization or restriction digestion of amplified products.
Despite the presence of five primer sets within the reaction mixture, the RT-PCR assay was able to sensitively and specifically detect all virus types tested. Only PIV1 had a relatively higher limit of detection (250 TCID50). It is possible that PIV1 sensitivity was decreased in the context of the multiplex, where a much higher limit of detection may have been possible by amplification with PIV1 primers alone. Sensitivity limits by agarose gel detection for the other viruses were similar to or better than limits previously published for RT-PCR detection of RSV (5, 20), adenovirus (2, 17), and PIV3 (21). The multiplex primers were also found to be specific for the virus types tested, as no interassay cross amplification or amplification of other respiratory viruses was observed. The use of individual, specific primer sets for highly conserved areas of each viral genome as well as a high annealing temperature likely contributed to the high sensitivity and specificity observed (7). In the current study, Southern hybridization increased sensitivity limits slightly over agarose gel detection; however, it was primarily used to confirm the identity of amplified bands.
RT-PCR detection of RSV in respiratory specimens closely paralleled direct antigen detection, while detection of PIV3 and detection of adenovirus were much more dissimilar from direct antigen detection. Although the sensitivities and specificities of the immunofluorescence detection methods used by the participating laboratories were not known, it is likely that RSV antigen detection was the most sensitive and specific. The one specimen positive by DIF but not PCR was initially very faintly RT-PCR positive, but this result could neither be replicated nor be further verified by nested PCR. This suggests that the level of RSV in the specimen was close to the limit of detection, and further freeze-thaw treatment of the specimen may have affected RNA integrity. Adenovirus results by the two detection methods were the most disparate, with two positive results by DIF-IIF which were not detected by RT-PCR and seven other specimens positive for adenovirus by RT-PCR but not by DIF-IIF. The inability to detect adenovirus RNA in the two IIF-positive samples may have been due to a loss of nucleic acid integrity, as the specimens were several years old. PCR inhibitors may also have prevented detection; however, upon spiking of the specimen extracts with cloned amplicon, successful amplification was observed. Except for the three false-negative results by RT-PCR (RSV and two adenovirus specimens), detection rates were very high, especially considering that most specimens were stored for several days or weeks at 4°C, or for several years at −70°C, prior to being processed. False-positive results were guarded against by using numerous negative controls in each RT-PCR run (all of which remained negative) as well as verifying all positive results differing from DIF-IIF results by replication, nested PCR, and RT-PCR with primers hybridizing to a different part of the viral genome. The initial RT-PCR result of the seven suspected adenovirus specimens was confirmed following these supplementary investigations, suggesting that these RT-PCR-identified adenovirus infections were true positives. In this regard, the multiplex RT-PCR provided a much higher sensitivity for detection of adenovirus in respiratory specimens than did detection by DIF-IIF.
Comparison of multiplex RT-PCR with DIF-IIF detection resulted in high respective sensitivity and specificity values for RSV (97 and 99%) and PIV3 (100 and 98%). Sensitivity for adenovirus was 0%, likely for those reasons mentioned above, while specificity was 94%. In the present clinical study, none of the specimens were positive for PIV1 or PIV2 by either RT-PCR or DIF-IIF. Further RT-PCR experiments with respiratory specimens positive for PIV1 and PIV2 by antigen detection should verify the method’s ability to detect these particular viruses in respiratory specimens. Measured against DIF-IIF, the multiplex RT-PCR assay for all viruses detected had a sensitivity of 91% and specificity of 87%. However, as explained earlier, positive RT-PCR results were rigorously verified and therefore suggest that the multiplex RT-PCR has increased sensitivity for detection of respiratory viruses, particularly adenovirus. As well, four multiple respiratory infections were detected in the present study, further illustrating the potential usefulness of the multiplex RT-PCR assay.
The multiplex RT-PCR could be beneficial as a respiratory diagnostic service, particularly in conjunction with an influenza detection and typing RT-PCR assay (27); however, the cost-effectiveness of such a service needs to be demonstrated in comparison to rapid antigen testing (9). The benefits of the multiplex RT-PCR in relation to DIF-IIF are its speed, sensitivity, specificity, low volume of specimen required for testing (100 μl), ability to detect viruses inactivated during collection, and, most importantly, ability to assay for more than one respiratory virus in a single specimen. The last of these factors is important if limited specimen is available, which may limit the number of viruses assayed for by DIF-IIF. The multiplex RT-PCR assay could also be coupled with detection by microplate hybridization for confirmation of results, for subtype differentiation, or to possibly increase sensitivity and specificity. Sensitive, rapid testing for respiratory viruses is crucial in the clinical setting to reduce the potential for nosocomial transmission to high-risk patients, to limit unnecessary antibiotic use, and to direct therapy following a specific diagnosis (1, 26). In this regard, the multiplex RT-PCR assay provides a rapid and highly sensitive means of detection of five of the major respiratory pathogens implicated in lower respiratory tract infections and hospitalizations.
ACKNOWLEDGMENTS
I acknowledge the Virology Laboratory, Laboratory Services Branch of the Ontario Ministry of Health, and Rose Milk and Lee Sullivan of the Children’s Hospital of Eastern Ontario for providing respiratory specimens and detection results for use in this study. I also acknowledge the DNA Core Facility, Laboratory Center for Disease Control, for oligonucleotide primer and probe synthesis. I am grateful to Shimian Zou and Michael Drebot for valuable suggestions and comments made during manuscript preparation.
I am supported by a Visiting Fellowship in Canadian Government Laboratories.
REFERENCES
1. Adcock P M, Stout G G, Hauck M A, Marshall G S. Effect of rapid viral diagnosis on the management of children hospitalized with lower respiratory tract infection. Pediatr Infect Dis J. 1997;16:842–846. [PubMed] [Google Scholar]
2. Allard A, Girones R, Juto P, Wadell G. Polymerase chain reaction for detection of adenoviruses in stool samples. J Clin Microbiol. 1990;28:2659–2667. [PMC free article] [PubMed] [Google Scholar]
3. Belshe R B, Newman F K, Ray R. Parainfluenza virus vaccines. In: Kiyano H, Ogra P L, McGhee J R, editors. Mucosal vaccines. London, United Kingdom: Academic Press, Inc.; 1996. pp. 311–323. [Google Scholar]
4. Dowell S F, Anderson L J, Gary H E, Jr, Erdman D D, Plouffe J F, File T M, Jr, Marston B J, Breiman R F. Respiratory syncytial virus is an important cause of community-acquired lower respiratory infection among hospitalized adults. J Infect Dis. 1996;174:456–462. [PubMed] [Google Scholar]
5. Eugene-Ruellan G, Freymuth F, Bahloul C, Badrane H, Vabret A, Tordo N. Detection of respiratory syncytial virus A and B and parainfluenzavirus 3 sequences in respiratory tracts of infants by a single PCR with primers targeted to the l-polymerase gene and differential hybridization. J Clin Microbiol. 1998;36:796–801. [PMC free article] [PubMed] [Google Scholar]
6. Falsey A R, McCann R M, Hall W J, Criddle M M. Evaluation of four methods for the diagnosis of respiratory syncytial virus infection in older adults. J Am Geriatr Soc. 1996;44:71–73. [PubMed] [Google Scholar]
7. Fan J, Henrickson K J. Rapid diagnosis of human parainfluenza virus type 1 infection by quantitative reverse transcription-PCR-enzyme hybridization assay. J Clin Microbiol. 1996;34:1914–1917. [PMC free article] [PubMed] [Google Scholar]
8. Freymuth F, Vabret A, Galateau-Salle F, Ferey J, Eugene G, Petitjean J, Gennetay E, Brouard J, Jokik M, Duhamel J, Guillois B. Detection of respiratory syncytial virus, parainfluenzavirus 3, adenovirus and rhinovirus sequences in respiratory tract of infants by polymerase chain reaction and hybridization. Clin Diagn Virol. 1997;8:31–40. [PubMed] [Google Scholar]
9. Gilbert L L, Dakhama A, Bone B M, Thomas E E, Hegele R G. Diagnosis of viral respiratory tract infections in children by using a reverse transcription-PCR panel. J Clin Microbiol. 1996;34:140–143. [PMC free article] [PubMed] [Google Scholar]
10. Hamel A L, Wasylyshen M D, Nayar G P S. Rapid detection of bovine viral diarrhea virus by using RNA extracted directly from assorted specimens and a one-tube reverse transcription PCR assay. J Clin Microbiol. 1995;33:287–291. [PMC free article] [PubMed] [Google Scholar]
11. Hierholzer J D, Halonen P E, Dahlen P O, Bingham P G, McDonough M M. Detection of adenovirus in clinical specimens by polymerase chain reaction and liquid-phase hybridization quantitated by time-resolved fluorometry. J Clin Microbiol. 1993;31:1886–1891. [PMC free article] [PubMed] [Google Scholar]
12. Johnston S, Holgate S. Epidemiology of viral respiratory tract infections. In: Myint S, Taylor-Robinson D, editors. Viral and other infections of the human respiratory tract. London, United Kingdom: Chapman and Hall; 1996. pp. 1–38. [Google Scholar]
13. Karron R A, O’Brien K L, Froehlich J L, Brown V A. Molecular epidemiology of a parainfluenza type 3 virus outbreak on a pediatric ward. J Infect Dis. 1993;167:1441–1445. [PubMed] [Google Scholar]
14. Larson E. The clinical spectrum of disease in adults. In: Myint S, Taylor-Robinson D, editors. Viral and other infections of the human respiratory tract. London, United Kingdom: Chapman and Hall; 1996. pp. 39–45. [Google Scholar]
15. Leland D S, Emanuel D. Laboratory diagnosis of viral infections of the lung. Semin Resp Infect. 1995;10:189–198. [PubMed] [Google Scholar]
16. Marquardt O, Straub O C, Ahl R, Haas B. Detection of foot-and-mouth disease virus in nasal swabs of asymptomatic cattle by RT-PCR within 24 hours. J Virol Methods. 1995;53:255–261. [PubMed] [Google Scholar]
17. Morris D J, Cooper R J, Barr T, Bailey A S. Polymerase chain reaction for rapid diagnosis of respiratory adenovirus infection. J Infect. 1996;32:113–117. [PubMed] [Google Scholar]
18. Murry A R, Dowell S F. Respiratory syncytial virus: not just for kids. Hosp Pract. 1997;32:87–98. [PubMed] [Google Scholar]
19. Osiowy C, Prud’homme I, Monette M, Zou S. Detection of human herpesvirus 6 DNA in serum by a microplate PCR-hybridization assay. J Clin Microbiol. 1998;36:68–72. [PMC free article] [PubMed] [Google Scholar]
20. Paton A W, Paton J C, Lawrence A J, Goldwater P N, Harris R J. Rapid detection of respiratory syncytial virus in nasopharyngeal aspirates by reverse transcription and polymerase chain reaction amplification. J Clin Microbiol. 1992;30:901–904. [PMC free article] [PubMed] [Google Scholar]
21. Ralston S H, Digiovine F S, Gallacher S J, Boyle I T, Duff G W. Failure to detect paramyxovirus sequences in Paget’s disease of bone using the polymerase chain reaction. J Bone Miner Res. 1991;6:1243–1248. [PubMed] [Google Scholar]
22. Reina J, Ros M J, Del Valle J M, Blanco I, Munar M. Evaluation of direct immunofluorescence, dot-blot enzyme immunoassay, and shell-vial culture for detection of respiratory syncytial virus in patients with bronchiolitis. Eur J Clin Microbiol Infect Dis. 1995;14:1018–1020. [PubMed] [Google Scholar]
23. Singh-Naz N, Rodriguez W. Adenoviral infections in children. Adv Pediatr Infect Dis. 1996;11:365–388. [PubMed] [Google Scholar]
24. Swierkosz E M, Erdman D D, Bonnot T, Schneiderheinze C, Waner J L. Isolation and characterization of a naturally occurring parainfluenza 3 virus variant. J Clin Microbiol. 1995;33:1839–1841. [PMC free article] [PubMed] [Google Scholar]
25. Valassina M, Cuppone A M, Cusi M G, Valensin P E. Rapid detection of different RNA respiratory virus species by multiplex RT-PCR: application to clinical specimens. Clin Diagn Virol. 1997;8:227–232. [PubMed] [Google Scholar]
26. Woo P C Y, Chiu S S, Seto W, Peiris M. Cost-effectiveness of rapid diagnosis of viral respiratory tract infections in pediatric patients. J Clin Microbiol. 1997;35:1579–1581. [PMC free article] [PubMed] [Google Scholar]
27. Zou S. A practical approach to genetic screening for influenza virus variants. J Clin Microbiol. 1997;35:2623–2627. [PMC free article] [PubMed] [Google Scholar]
Articles from Journal of Clinical Microbiology are provided here courtesy of American Society for Microbiology (ASM)