This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lai, K. K.-Y.
Right arrow Articles by Jerome, K. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lai, K. K.-Y.
Right arrow Articles by Jerome, K. R.

 Previous Article  |  Next Article 

Journal of Clinical Microbiology, July 2003, p. 3133-3141, Vol. 41, No. 7
0095-1137/03/$08.00+0     DOI: 10.1128/JCM.41.7.3133-3141.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.

Evaluation of Real-Time PCR versus PCR with Liquid-Phase Hybridization for Detection of Enterovirus RNA in Cerebrospinal Fluid

K. Kay-Yin Lai,1 Linda Cook,1 Sharon Wendt,1 Lawrence Corey,1,2,3 and Keith R. Jerome1,3*

Departments of Laboratory Medicine,1 Medicine, University of Washington Medical Center, Seattle, Washington 98195,2 Program in Infectious Diseases, Fred Hutchinson Cancer Research Center, Seattle, Washington 981093

Received 27 January 2003/ Returned for modification 17 March 2003/ Accepted 28 April 2003


arrow
ABSTRACT
 
A LightCycler and two TaqMan real-time PCR assays were evaluated against an older PCR with liquid-phase hybridization method for the detection of enterovirus RNA in 74 patient samples. The two-step LightCycler and the two-step TaqMan formats correlated well with each other (r2 = 0.90) and were equally sensitive compared to the liquid-phase hybridization method, whereas the one-step recombinant Tth DNA polymerase format was rather insensitive, detecting enterovirus RNA in only about one-half of those patient samples previously positive by liquid-phase hybridization. The two-step TaqMan method was optimized utilizing 10 µl of cDNA and demonstrated the highest degree of analytical sensitivity among the methods evaluated in our study, being able to reproducibly quantify down to 510 copies of enteroviral RNA/ml of cerebrospinal fluid. This new assay can be performed in 4 h, is much less labor intensive, and showed less cross-reactivity with rhinovirus than the liquid-phase hybridization assay. Thus, the two-step TaqMan assay should prove useful in the diagnosis of enteroviral meningitis versus bacterial meningitis, thereby resulting in timely and appropriate clinical management that can amount to significant cost savings to the patient and health care system.


arrow
INTRODUCTION
 
Human enteroviruses are members of the family Picornaviridae, genus Enterovirus, and consist of more than 60 distinct serotypes. Most enteroviruses can be classified into one of five major groups, including coxsackieviruses A and B, echoviruses, enteroviruses, and polioviruses. Nonpolio enteroviruses represent the most common etiology of meningitis in the United States (22), accounting for more than 80% of all cases. Children, neonates, and immunocompromised individuals are particularly susceptible to enterovirus infection of the central nervous system (2, 16, 22). Though usually benign in nature, the clinical characteristics of enteroviral meningitis can be very similar to those of potentially life-threatening bacterial meningitis. Since bacterial meningitis requires hospital admission and aggressive treatment, the distinction of these entities is critical. Traditional culture-based methods for the detection of enterovirus in cerebrospinal fluid (CSF) are laborious, time-consuming, and often insensitive. Thus, molecular assays are attractive in that they can allow rapid and accurate diagnosis, resulting in more appropriate and cost-effective treatment (20, 28).

Real-time PCR technology has provided an excellent platform for diagnostic PCR assays. Real-time PCR is quantitative and has low interassay and intra-assay variability (1, 14, 15, 27). Recently, several groups have described real-time PCR methods for the detection of enterovirus in CSF (7, 17-19, 29-31). Based on bioinformatics analysis of published enterovirus sequences, we chose primers and probes to allow sensitive detection of these viruses. In this study, we evaluated three real-time PCR methods using these primers and probes on the TaqMan and LightCycler platforms. Our optimized method is more sensitive and reproducible than the earlier liquid hybridization assay and compares favorably to previously published real-time PCR methods for detection of enterovirus.


arrow
MATERIALS AND METHODS
 
Patient and culture samples. Samples from a total of 74 patients were submitted to the clinical laboratory for enterovirus testing during the period June 2001 through December 2002. Samples were stored at -70°C. A total of 38 enterovirus-positive patient-derived samples were available. An additional 36 enterovirus-negative patient samples from within the same time frame were used as controls. Sixty-three of the samples were CSF, 5 were from viral culture, and six were from other sites or not clearly identified. Experiments were also performed using culture fluids obtained from American Type Culture Collection (Manassas, Va.) stock enterovirus or rhinovirus strains.

Sample preparation. Extraction of viral RNA was performed using QIAamp viral RNA minikit (Qiagen, Valencia, Calif.) according to the package insert instructions. A total of 140 µl of sample was extracted into a final volume of 60 µl. This purified viral RNA was then used directly in the one-step recombinant Tth (rTth) reaction or used to produce cDNA. For production of cDNA, 10 µl of extracted RNA was incubated at 70°C for 5 min and quick-chilled on ice, and then reverse transcription (RT) mix was added (final concentrations of 33.2 µM enterovirus reverse primer, 3 mM MgCl2, 0.6 mM dithiothreitol, 4 mM deoxynucleoside triphosphates, 374 U of RNase inhibitor/ml, and 5.2 U of Moloney murine leukemia virus [M-MLV] reverse transcriptase/ml). The mixture was then incubated at 37°C for 1 h. Finally, the samples were heated at 95°C for 5 min and placed on ice or frozen at -20°C prior to use in the second step PCR. All samples were assayed in duplicate.

Enterovirus-specific primers and probes. Primers and probes were selected after a careful review of the literature. Essentially all published methods amplify a consensus region of the enterovirus 5' untranslated region and use primers and probes with very similar sequences (Fig. 1). Our liquid-phase hybridization method and all three of our real-time PCR assays used the same 3' downstream primer originally described by Rotbart in 1990 (21). Our liquid-phase hybridization method utilized the 5' upstream primer previously described by Rotbart in 1990 and a slightly modified probe based on that reported in the same study (21). For our real-time assays, we selected the 5' primer and probe derived from the work of Verstrepen (30), since their Tms most closely matched suggested TaqMan primer and probe conditions. The TaqMan probe was labeled with FAM at the 5' end and TAMRA at the 3' end. For rhinovirus detection, the liquid-phase hybridization probe was 5'-TAG TTG GTC CCA TCC CG-3'. For the one- and two-step TaqMan assays, the forward enterovirus primer concentration was 300 nM, the reverse primer concentration was 900 nM, and the probe concentration was 150 nM for the one-step assay and 100 nM for the two-step assay. For the LightCycler method, the forward primer concentration was 150 nM, the reverse primer concentration was 450 nM, and the probe concentration was 150 nM.




View larger version (144K):
[in this window]
[in a new window]
  		FIG. 1. Enterovirus sequence alignments for 5' primer, 3' primer, and probe regions. The 14 most clinically important enterovirus serotypes were determined, and sequences from the primer and probe binding sites were aligned using GeneDoc version 2.6.2. Shown for each region is the consensus sequence, any serotypes varying from the consensus sequence, and the alignment and sequences of primers and probes previously described in the literature. , identical base; {blacksquare}, not available; *, base deletion; X, insertion of A prior to this base. Superscript numbers are reference numbers. Superscript letters: a, numbering based on Cox A16 sequence, GenBank accession no. 458298; b, consensus sequence was present in Echo 30 (accession no. 14139951 and 17224872), Echo 7 (1199448, 15788480, and 1199449), Cox B2 (7263148 and 7263147), Echo 6 (17530514 and 15788479), Cox B1 (15788477 and 323417), Echo 25 (1154653 and 4539509), Cox A9 (12802345 and 221214), Echo 16 (15788484), Echo 18 (15788486), Cox B3 (17148519, 10863164, 5833884, and 5833880), Echo 11 (17530521, 17530520, and 17530519), enterovirus 71 (4753701, 4753698, 4753699, 4753694, and 18158530), and Cox B4 (61031 and 914055); c, consensus sequence was present in Echo 30 (14139951, 14139942, and 17224872), Cox B1 (914052), Echo 7 91199448), Cox B2 (7263148 and 7263147), Echo 6 (17530514 and 509558), Echo 11 (17530521, 17530520, and 17530519), enterovirus 71 (18158530), Echo 9 (769799 and 509559), Cox B2 (16555707, 16555709, and 16555708), and Echo 25 (4572558). d, consensus sequence was present in Echo 30 (14139951), Cox B2 (7263148 and 7263147), Echo 6 (17530514 and 509558), Echo 25 (1154653 and 4572558), Cox A9 (221214), Cox B1 (323417), Cox B3 (10863164, 5833884, and 5833880), Echo 7 (1199449), Echo 11 (17530520 and 17530519), Echo 9 (769799 and 509559), and Cox B4 (61031 and 914055).

Liquid-phase hybridization assay. For DNA amplification, 10 µl of cDNA was added to 90 µl of master mix (final concentrations of 16.6 µM reverse primer, 16.6 µM forward primer, 1x PCR buffer, 214.6 µM dATP, dCTP, and dGTP, 429.3 µM dUTP, 22 U of AmpliTaq/ml, including 10 µl of glycerol). Mixes were amplified using a Perkin-Elmer 9600 thermal cycler as follows: 2 min at 94°C; 35 cycles of 30 s at 96°C, 30 s at 54°C, and 30 s at 72°C; and then 5 min at 72°C for final terminal extension. Following PCR amplification, liquid-phase hybridization was done by adding 3.9 µl of amplified product to 10 µl of hybridization mix (final concentrations of 200 µM NaCl, 100 µM [each] deoxynucleoside triphosphates [A, T, G, and C], 0.02% bromophenol blue in formamide, 0.01 µM formamide, and 106 cpm of 32P-labeled virus-specific probe ranging from 3 to 100 nM and labeled using T4 polynucleotide kinase and [{gamma}-32P]ATP). The samples were then placed in the Perkin-Elmer 9600 thermal cycler, heated to 97°C for 5 min, and then slowly cooled to 20°C over 15 min. Radiolabeled hybridized amplicon product was then detected by loading 10 µl of each sample on 6% acrylamide gels (Invitrogen, Carlsbad, Calif.). Gel electrophoresis was done at 180 V for 20 to 30 min. The gel was then placed on blotting paper and dried on a gel dryer. Kodak X-Omat film was then exposed overnight at room temperature or at -70°C for 2 h.

One-step RT-PCR assay. The one-step RT-PCR test was performed using the EZ RT-PCR kit (no. 403028; Applied Biosystems, Foster City, Calif.) that utilizes rTth DNA polymerase. The master mix was prepared according to the package insert instructions, and then 3 µl of purified RNA from each sample was mixed with 47 µl of master mix. Samples were analyzed on the ABI 7700 sequence detection system (Applied Biosystems). Cycling parameters were 2 min at 50°C, 30 min at 60°C, 5 min at 95°C, and then 45 cycles of 20 s at 95°C followed by 1 min at 62°C.

Second-step PCR with ABI 7700. The second-step PCR was performed on the ABI 7700. Either 3 µl or 10 µl of cDNA was added to an in-house prepared master mix (12). Cycling parameters were 2 min at 50°C, 2 min at 95°C, then 45 cycles of 20 s at 95°C followed by 1 min at 60°C.

Second-step PCR with LightCycler. The second-step PCR was performed on the Roche LightCycler using the FastStart LC DNA hybridization probes kit (Roche Molecular Biochemicals, Indianapolis, Ind.). The master mix was prepared according to the package insert instructions, and then 17 µl was added to 3 µl of sample cDNA. The concentration of MgCl2 was 5 mM. Samples were placed in capillary tubes and then subjected to 40 cycles of amplification at 95°C for 10 s and then 60°C for 13 s.

Enterovirus standard materials. Two lots of enterovirus Armored RNA containing 5' untranslated region sequence from enterovirus were used for standard preparation, a custom preparation of 154 bp (2.0 x 1013 copies/ml) and a newer commercial preparation of 263 bp (Ambion RNA Diagnostics, Austin, Tex.). The custom preparation was purchased in 1998 and had been stored for 5 years at -70°C. The two lots of Armored RNA performed equally when they were used to create standard curves of 8.7 x 107, 8.7 x 105, 8.7 x 103, and 8.7 x 102 copies/ml for the real-time assays. For routine assays, the 8.7 x 107-copies/ml standard was prepared in batches (either RNA or cDNA) from the custom preparation, which was purchased in 1998. The 8.7 x 105-, 8.7 x 103-, and 8.7 x 102-copies/ml standards were prepared from the 8.7 x 107- copies/ml standard daily by dilution in Tris buffer (10 mM, pH 8.0). For studies to determine the assay's analytical sensitivity, a 1:1.7 dilution of the 870-copies/ml standard was made daily to produce a sample containing 510 copies/ml, and this sample was run in quadruplicate wells on seven consecutive runs. Initial evaluation of the real-time assays was done using dilutions of aliquots of 1.0 x 106 copies of enterovirus Armored RNA material/ml originally prepared by dilution of the custom lot and stored at -70°C for 2 years. All standards were assayed in duplicate for generation of the standard curve.

Enterovirus control materials. Cox B5 culture fluid derived from a patient isolate previously typed by direct fluorescent antibody was used to make high and low enterovirus controls. The high enterovirus control with a mean concentration of 3.3 x 106 copies/ml was stored at -70°C in 250-µl aliquots, whereas the low control, with a mean concentration of 2.2 x 103 copies/ml, was prepared daily by dilution of the high control. For studies to determine the assay's analytical sensitivity, a 1:4 dilution of the low control was made daily to produce a sample containing 550 copies/ml, and this sample was run in quadruplicate wells on seven consecutive runs. For routine runs, high and low controls were run in duplicate.

Computer analysis tools. Tms of primers and probes were determined using Oligo Melting Temperature version 5.0 (http://blocks.fhcrc.org/oligo_melt.html) (3, 24). Enterovirus and rhinovirus sequences were retrieved from the GenBank database of the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov/) by a Taxonomy search and used to create two data files. For both databases, sequences containing either 5' untranslated region or the entire genomic sequence were retrieved. To simplify alignment and analysis, the first 70 to 100 bp and all base pairs past bp 700 were deleted to isolate sequence in the area of interest. Alignments of sequences were done with GeneDoc software version 2.6.2, available through the Pittsburgh Supercomputing Center (http://www.psc.edu/biomed/genedoc/).


arrow
RESULTS
 
Bioinformatics analysis of candidate primers and probes. More than 60 distinct serotypes of human enteroviruses are known to exist. For this reason, only a few of the previously published primer-probe sets have actually been tested in the laboratory for reactivity against all enterovirus serotypes (9, 11, 18, 32). We used a bioinformatics approach in an effort to predict the specificity of the more recently published real-time primers and probes. First, we determined those serotypes of enterovirus most likely to be of clinical relevance, using surveillance reports for the years 1993 to 1999 published by the Centers for Disease Control and Prevention in Morbidity and Mortality Weekly Report (4, 5). By ranking the serotype frequencies from highest to lowest, the top 14 serotypes for each year from 1 to 14 were determined to be the following: echovirus (Echo) 11, Echo 30, coxsackievirus (Cox) A9, Echo 6, Echo 9, Cox B2, Echo 7, Cox B3, Cox B4, Cox B5, Echo 25, enterovirus 71, Echo 17, and Echo 22. Based on these frequencies, we extracted from GenBank all available sequences up to a total of 6 for each of these serotypes and created a database with 45 sequences. We then determined the degree of matching with previously published primers and probes.

Significant homology in the primer and probe sequence regions was observed with all serotypes in the database (Fig. 1). Most published primer and probe sequences for detection of enterovirus were found to amplify very similar sequences within the 5' untranslated region of the virus. Essentially all sequences in the database would be predicted to amplify with most published primer and probe sets. However, because of some base pair differences and some significant Tm differences, not all primer and probe sets would be predicted to amplify all sequences or work with all methodologies. Based on the alignment information and the Tm constraints of primer and probe design for real-time PCR, for further development we selected the Verstrepen 5' primer and probe (30) to be used in conjunction with the Rotbart 1990 3' primer (21).

Our analysis revealed that two of the Echo 6 sequences in our initial database had C-to-T changes at the 3' end of the 5' primer sequence (position 479). This mismatch may lead to decreased or absent amplification of Echo 6 viruses possessing this C-to-T change. To evaluate the frequency of this C-to-T change, we extracted all Echo 6 sequences available in GenBank. Of the 12 available sequences, 3 had the C-to-T change, while the remaining 9 had the common C. Thus, based on alignment data, we would predict that the selected primer and probe sequence should amplify most Echo 6 viruses and nearly all of the most clinically relevant enterovirus sequences.

As a complementary approach, we then checked our primer and probe sequences against all sequences in GenBank using Taxonomy BLAST analysis. Compared to the 5' primer used in our earlier liquid hybridization assay, the real-time 5' primer identified more identical sequences (513 versus 413). Similarly, the real-time probe identified more identical sequences than did the earlier liquid hybridization probe (363 versus 344). BLAST searches using the real-time 5' primer identified sequences for all enterovirus serotypes. For several enterovirus serotypes (serotypes 13, 14, 15, 16, 17, 20, 21, 24, 29, 31, 32, and 33), sequence information is not available for the binding region for the 3' primer and probe. However, based on the Taxonomy BLAST searches done with the remaining serotypes, we would predict that the various enterovirus serotypes should have very similar sequences in the 5' primer, 3' primer, and probe areas. Therefore, assays with the new primer-probe set would be expected to detect nearly all enterovirus serotypes.

The rhinoviruses are genetically very similar to the enteroviruses, and thus, cross-reactivity is common between these viruses. We therefore performed a similar analysis of our primers and probe using rhinovirus sequences from GenBank. Significantly fewer rhinovirus sequences were present in GenBank, and only 53 sequences included the 5' untranslated region bound by our primers and probe. Of these 53 sequences, only 21 had serotype identification, and only 16 of the possible 110 serotypes were represented. We compared all available rhinovirus sequences with those of our chosen enterovirus primers and probe. After alignment of the rhinovirus sequences, areas of sequence similarity to our enterovirus primers and probe were identified at bp 453 to 471, 536 to 563, and 578 to 597 (Fig. 2). At the site of the enterovirus 3' primer similarity, all but one rhinovirus sequence (serotype 1B) would be expected to prime. At the probe site, essentially all amplified sequences should be detected. However, at the 5' primer site, most rhinovirus sequences had CCT at the 3' end. In contrast, the primer and six sequences, two of which were serotype 87, had TCC. This mismatch between the primer and the majority of rhinovirus sequences would be expected to result in a lack of priming with most rhinoviruses. Thus, based on this rhinovirus alignment analysis, the majority of rhinovirus sequences should not be amplified by the primer-probe set because of mismatching with the 5' primer.



View larger version (44K):
[in this window]
[in a new window]
  		FIG. 2. Rhinovirus sequences with similarity to enterovirus 5' primer, 3' primer, and probe regions. All rhinovirus sequences containing the 5' untranslated region bound by our enterovirus real-time primers and probe were aligned with sequences from our primers and probe using GeneDoc version 2.6.2. Shown for each region is the consensus sequence, any sequences varying from the consensus sequence, and the alignment and sequences of our selected real-time primer-probe set. Superscript a: numbering based on rhinovirus 1B sequence, GenBank accession no. 221708.

One-step real-time RT-PCR assay: sensitivity. We then utilized the Verstrepen 5' primer and probe (30) and the Rotbart 1990 3' primer (21) to evaluate a one-step real-time RT-PCR assay using rTth DNA polymerase. The one-step real-time RT-PCR method was used to determine the endpoint sensitivity for a Cox B5 culture and an enterovirus Armored RNA stock preparation. By culture, the Cox B5 fluid was positive to a dilution of 10-7, while by our older liquid hybridization method, it was positive to a dilution of 10-8. However, when the Cox B culture fluid sample was tested by the rTth assay method, the Cox B5 fluid was positive only to a 10-5 dilution. The control Armored RNA preparation was positive only to a 10-2 dilution, implying a detection limit of approximately 1.0 x 104 copies/ml. The rTth method was then used to analyze the RNA preparations isolated from 60 patient samples. Of the 21 samples previously positive for enterovirus by liquid-phase hybridization, only 11 were positive by this one-step real-time method. These data indicate that this one-step real-time assay is clearly not as sensitive as our existing liquid-phase hybridization assay.

Two-step real-time assays: sensitivity. We next evaluated two separate two-step real-time PCR assays, both utilizing identical primers and probes to amplify cDNA that was prepared from patient RNA using M-MLV. One assay was performed using an in-house prepared master mix (12) with detection by the ABI 7700 (TaqMan), and the other was performed using a Roche DNA amplification kit with detection by the LightCycler. Both assays produced good standard curves with dilutions of enterovirus Armored RNA and had similar threshold cycles (Ct). The LightCycler and ABI 7700 assays yielded identical titers, with the Cox B5 culture positive to a dilution of 10-7 and the Armored RNA control solution positive to a dilution of 10-5. Thus, similar sensitivities were seen with the two different reagent/instrument assays, and this sensitivity was 2 to 3 logs better than that observed with the rTth one-step assay. In an effort to maximize sensitivity, we modified the TaqMan assay to use 10 µl of cDNA rather than the 3 µl used initially. This further improved the TaqMan assay sensitivity by 1 log for both the Cox B5 culture fluid and the Armored RNA control solution.

To determine the sensitivity of our assay, we compared the two-step TaqMan assay using 10 µl of sample to the liquid-phase hybridization method by evaluating serial threefold dilutions around the endpoint titers of the Cox B5 and Echo 6 culture fluids (Table 1). These initial results indicated approximately equal sensitivity between the liquid-phase hybridization and TaqMan assays.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Comparison of TaqMan Ct and quantitation with liquid-phase hybridizationa

To further define the analytical sensitivity of the two-step TaqMan assay, we evaluated standard material with a concentration of 510 copies/ml and Cox B5 control material with a concentration of 550 copies/ml. These dilutions of the standard material and control material were both positive in 28 of 28 wells tested on seven runs, with Ct coefficients of variation (CVs) of 6.1% for the former and 4.6% for the latter. Therefore, we have determined our assay's reproducible lower limit of detection to be 510 copies/ml or better, although positive results with lower concentrations of virus are frequently observed.

Patient sample comparisons. Sixty patient-derived samples were then analyzed by the liquid-phase hybridization assay and both two-step real-time methods, and an additional 14 were evaluated by the two-step real-time methods alone. The two-step real-time assays yielded very similar results for all patient samples, with an r2 value of 0.90 (Fig. 3). Dilutions of the culture fluids demonstrated linearity of both real-time assays as well as good correlation between these two methods, from 104 to 109 copies/ml. In the comparison with the liquid hybridization assay, 24 samples were positive by all three methods, and 36 samples were negative by all three methods. One sample was negative by liquid-phase hybridization but positive repeatedly with both real-time methods. This discrepant sample measured 8.0 x 103 copies/ml by TaqMan PCR; it was subsequently cultured, and Cox B3 was found. The positive result for this sample suggests that the primer-probe set utilized in the real-time assays may have a broader reactivity than our earlier primer-probe set for Cox B3.



View larger version (22K):
[in this window]
[in a new window]
  		FIG. 3. Comparison of LightCycler versus TaqMan results with patient-derived specimens. Thirty-eight enterovirus-positive patient-derived samples submitted to the clinical laboratory between June 2001 and December 2002 were evaluated using the two-step LightCycler and TaqMan methods. The correlation between the two methods is shown.

Two-step TaqMan assay: reproducibility. To evaluate the reproducibility of the two-step TaqMan assay over time in a clinical setting, we analyzed quality control data obtained over 1 month as the assay was performed 5 days per week by different medical technologists. The threshold cycles for three of the standard curve points were stable over time, with CVs of 3.5% for the high standard, 2.8% for an intermediate standard, and 2.5% for the low standard (Fig. 4A). The Ct values for the high and low Cox B5 control samples were also stable over time, with CVs of 2.4 and 7.6%, respectively. The quantitative results had CVs of 37% for the high control and 62% for the low control (Fig. 4B).



View larger version (23K):
[in this window]
[in a new window]
  		FIG. 4. Quality control data for two-step TaqMan method. Shown are quality control data obtained from 19 runs of the two-step TaqMan method over a 1-month period for high (8.7 x 107 copies/ml), intermediate (8.7 x 103 copies/ml), and low (8.7 x 102 copies/ml) standards (A) and high positive (3.3 x 106 copies/ml) and low positive (2.2 x 103 copies/ml) controls (B).

Enterovirus specificity and rhinovirus cross-reactivity. As noted previously, rhinovirus has a genomic structure very similar to that of enterovirus, and thus cross-reactions with enterovirus primer-probe sets are possible. Our previous liquid-phase hybridization assay had been shown to give false-positive results with all five of the rhinovirus serotypes (1B, 12, 15, 38, and 56) tested. To investigate cross-reactivity with the TaqMan and liquid-phase hybridization assays, comparison testing was done with American Type Culture Collection rhinoviruses of 18 different serotypes (Table 2). All 18 rhinovirus serotypes were detected by our earlier liquid hybridization enterovirus assay. In contrast, by the TaqMan method, about two-thirds of the rhinovirus serotypes were negative, and only one-third of the serotypes were positive. The positive rhinovirus serotypes appeared to be randomly scattered throughout the Rhinovirus genus and were not associated with rhinovirus receptor subgroups, serotype subgroups, or relatedness of the genetic sequence in the VP4/2 region (25). Overall, the results demonstrated that our real-time TaqMan assay for enterovirus has far less cross-reactivity with rhinoviruses than did our older liquid-phase hybridization method.


View this table:
[in this window]
[in a new window]
  		TABLE 2. Rhinovirus cross-reactivity by TaqMan versus liquid-phase hybridization


arrow
DISCUSSION
 
Rapid, accurate diagnosis of enterovirus infection is essential for the appropriate clinical management of patients presenting with meningeal symptoms. Various studies have emphasized the cost savings of up to US $2,700 per patient that can be realized with timely diagnosis of enterovirus infection (20, 28). In response to this, our laboratory has developed and extensively evaluated a rapid, reproducible two-step real-time quantitative TaqMan PCR assay for the detection of enterovirus RNA in CSF.

We carefully selected primers and a probe for enterovirus based on bioinformatics analysis of the sequences of the 14 most clinically relevant enterovirus serotypes in the United States (4, 5) as well as available rhinovirus sequences. Our analysis and experimental results confirm the sensitivity of the primers and probe for these enterovirus serotypes. Based on our bioinformatics analysis and the similarity of our primers and probe to those used in previous studies evaluating extensive enterovirus panels (18, 19, 31), the primers and probe used in our study should also be highly sensitive for the detection of other enterovirus serotypes less commonly seen in the clinical setting. It will be important for future studies to validate the selected primer-probe set against the 60-plus enterovirus serotypes, since the potential of such molecular testing to detect serotypes not easily cultured may completely revise our knowledge of the clinical epidemiology of enterovirus serotypes and infection.

Employing this primer-probe combination, we developed and evaluated two rapid two-step real-time PCR methods for the quantification of enterovirus in CSF, one for the ABI 7700 TaqMan platform and the other for the Roche LightCycler platform. The two-step real-time methods were essentially equivalent to each other and to the older liquid-phase hybridization method with respect to sensitivity. All three methods were superior to the one-step RT-PCR method using rTth. The two-step TaqMan method using 10 µl of cDNA demonstrated the highest degree of analytical sensitivity, with a lower limit of detection of 510 copies enterovirus/ml of CSF. This degree of analytical sensitivity compares favorably to those of various recently published enterovirus real-time methods, which have suggested sensitivities ranging from 1,250 copies/ml (30) to 3,500 copies/ml (17). The contrasting insensitivity of the one-step rTth real-time method may be due to disparate efficiencies between the different reverse transcriptase systems used (rTth versus M-MLV) or to inherent differential efficiencies of one-step versus two-step methods. Another study which compared a two-step with a single-step real-time enterovirus method showed that the former could be more sensitive than the later by 3.2 orders of magnitude (7). To our knowledge, other groups reporting better success with one-step RT-PCR than our group either did not use rTth DNA polymerase (17, 18, 29, 30) or used rTth but on the LightCycler platform (19, 31). Because we were unable to achieve reasonable sensitivity using rTth on the ABI 7700 platform during our initial evaluation, we chose not to further pursue development of the assay using rTth on the LightCycler platform. However, it should be noted that the sensitivity of our two-step TaqMan assay is better than any of the sensitivities reported by others using a one-step approach. Further work will be necessary to evaluate and maximize the efficiency of one-step real-time RT-PCR.

The two-step TaqMan assay demonstrated excellent reproducibility over time in a clinical setting with different medical technologists, as indicated by good CVs for Ct results of the standards (3.5% for the high standard, 2.8% for an intermediate standard, and 2.5% for the low standard) which are comparable to those reported by others (17), as well as by good CVs for quantitative control materials. This suggests good performance over the entire dynamic range of the assay with an upper limit of at least 1010 copies/ml. The two-step real-time assay's ability to reproducibly quantify enteroviral load over a wide range represents a significant improvement over our previous qualitative liquid-phase hybridization method.

Using our real-time primer-probe set for enterovirus, we observed cross-reactivity with about one-third of the tested rhinovirus serotypes. Cross-reactivity with rhinovirus has also been described for other real-time PCR enterovirus assays (18). We therefore reviewed our clinical experience regarding enterovirus-rhinovirus cross-reactivity. From 1,205 samples submitted for enterovirus testing during the years 1998 to 2001, 156 (12.9%) were positive by our enterovirus liquid-phase hybridization assay. Of the positives, 83% were CSF specimens and the rest were other sample types. Because of possible cross-reactivity, our laboratory's policy was to routinely retest all positive samples with a rhinovirus-specific probe. Upon retesting using the rhinovirus-specific probe, only 1 of the 156 positive patient-derived specimens proved to contain rhinovirus. This single specimen was a nasal swab. From a clinical standpoint, there have been no published papers reporting isolation of rhinovirus from CSF (19). Because of the lack of rhinovirus detection in clinical CSF specimens and the much lower rate of cross-reactivity with rhinoviruses using our new real-time TaqMan assay, we have discontinued the confirmatory rhinovirus testing of our clinical samples.

Recently several papers have documented significant cost savings that could potentially be realized by utilizing rapid PCR-based testing in the diagnostic work-up of aseptic meningitis due to enterovirus. One of these studies found that rapid detection of enterovirus and appropriate clinical management during an outbreak of enterovirus meningitis in adults reduced the mean length of hospital stay from 103 to 80 h and reduced the mean duration of antibacterial treatment from 115 to 69 h (28). Another study found that pediatric patients for whom positive enterovirus PCR results were available within 24 h had 20 h less of antibiotic treatment and US $2,798 less in hospital fees than patients with positive results available after 24 h (20). Our real-time PCR assay for enterovirus has an assay time of approximately 4 h and thus should allow realization of such cost savings to the health care system.

In summary, our laboratory has developed a rapid and highly sensitive real-time TaqMan assay for the quantification of enterovirus RNA in CSF. The assay is faster, safer, less expensive, and more specific than earlier assays based on liquid-phase hybridization. Rapid and sensitive diagnosis of enterovirus infections should result in timely and appropriate management of patients presenting with clinical meningitis, with significant cost savings to the patient and the health care system.


arrow
ACKNOWLEDGMENTS
 
We thank Ka Wing Ng, Ederlyn E. Atienza, Abby Hamad, Anne M. Cent, and Grace Koon for their technical assistance and Rhoda Ashley-Morrow for useful discussions. K. K.-Y. Lai also thanks James S. Fine and Sum P. Lee for their encouragement.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Fred Hutchinson Cancer Research Center, 1100 Fairview Ave. N, D3-100, Seattle, WA 98109. Phone: (206) 667-6793. Fax: (206) 667-4411. E-mail: kjerome{at}fhcrc.org. Back


arrow
REFERENCES
 
    1
  1. Abe, A., K. Inoue, T. Tanaka, J. Kato, N. Kajiyama, R. Kawaguchi, S. Tanaka, M. Yoshiba, and M. Kohara. 1999. Quantitation of hepatitis B virus genomic DNA by real-time detection PCR. J. Clin. Microbiol. 37:2899-2903.[Abstract/Free Full Text]
    2. 2
  2. Berlin, L. E., M. L. Rorabaugh, F. Heldrich, K. Roberts, T. Doran, and J. F. Modlin. 1993. Aseptic meningitis in infants 2 years of age: diagnosis and etiology. J. Infect. Dis. 168:888-892.[Medline]
    3. 3
  3. Breslauer, K. J., R. Frank, H. Blocker, and L. A. Marky. 1986. Predicting DNA duplex stability from the base sequence. Proc. Natl. Acad. Sci. USA 83:3746-3750.[Abstract/Free Full Text]
    4. 4
  4. Centers for Disease Control and Prevention. 2000. Enterovirus surveillance—United States, 1997-1999. Morb. Mortal. Wkly. Rep. 49:913-916.[Medline]
    5. 5
  5. Centers for Disease Control and Prevention. 1997. Nonpolio enterovirus surveillance—United States, 1993-1996. Morb. Mortal. Wkly. Rep. 46:748-750.
    6. 6
  6. Chapman, N. M., S. Tracy, C. J. Gauntt, and U. Fortmueller. 1990. Molecular detection and identification of enteroviruses using enzymatic amplification and nucleic acid hybridization. J. Clin. Microbiol. 28:843-850.[Abstract/Free Full Text]
    7. 7
  7. Corless, C. E., M. Guiver, R. Borrow, V. Edwards-Jones, A. J. Fox, E. B. Kaczmarski, and K. J. Mutton. 2002. Development and evaluation of a 'real-time' RT-PCR for the detection of enterovirus and parechovirus RNA in CSF and throat swab samples. J. Med. Virol. 67:555-562.[CrossRef][Medline]
    8. 8
  8. Glimaker, M., B. Johansson, P. Olcen, A. Ehrnst, and M. Forsgren. 1993. Detection of enteroviral RNA by polymerase chain reaction in cerebrospinal fluid from patients with aseptic meningitis. Scand. J. Infect. Dis. 25:547-557.[Medline]
    9. 9
  9. Halonen, P., E. Rocha, J. Hierholzer, B. Holloway, T. Hyypia, P. Hurskainen, and M. Pallansch. 1995. Detection of enteroviruses and rhinoviruses in clinical specimens by PCR and liquid-phase hybridization. J. Clin. Microbiol. 33:648-653.[Abstract]
    10. 10
  10. Heim, A., and J. Schumann. 2002. Development and evaluation of a nucleic acid sequence based amplification (NASBA) protocol for the detection of enterovirus RNA in cerebrospinal fluid samples. J. Virol. Methods 103:101-107.[CrossRef][Medline]
    11. 11
  11. Kessler, H. H., B. Santner, H. Rabenau, A. Berger, A. Vince, C. Lewinski, B. Weber, K. Pierer, D. Stuenzner, E. Marth, and H. W. Doerr. 1997. Rapid diagnosis of enterovirus infection by a new one-step reverse transcription-PCR assay. J. Clin. Microbiol. 35:976-977.[Abstract]
    12. 12
  12. Limaye, A. P., M. L. Huang, W. Leisenring, L. Stensland, L. Corey, and M. Boeckh. 2001. Cytomegalovirus (CMV) DNA load in plasma for the diagnosis of CMV disease before engraftment in hematopoietic stem-cell transplant recipients. J. Infect. Dis. 183:377-382.[CrossRef][Medline]
    13. 13
  13. Lina, B., B. Pozzetto, L. Andreoletti, E. Beguier, T. Bourlet, E. Dussaix, L. Grangeot-Keros, B. Gratacap-Cavallier, C. Henquell, M. C. Legrand-Quillien, A. Novillo, P. Palmer, J. Petitjean, K. Sandres, P. Dubreuil, H. Fleury, F. Freymuth, I. Leparc-Goffart, D. Hober, J. Izopet, H. Kopecka, Y. Lazizi, H. Lafeuille, P. Lebon, M. Aymard, et al. 1996. Multicenter evaluating of a commercially available PCR assay for diagnosing enterovirus infection in a panel of cerebrospinal fluid specimens. J. Clin. Microbiol. 34:3002-3006.[Abstract]
    14. 14
  14. Locatelli, G., F. Santoro, F. Veglia, A. Gobbi, P. Lusso, and M. S. Malnati. 2000. Real-time quantitative PCR for human herpesvirus 6 DNA. J. Clin. Microbiol. 38:4042-4048.[Abstract/Free Full Text]
    15. 15
  15. Mackay, I. M., K. E. Arden, and A. Nitsche. 2002. Real-time PCR in virology. Nucleic Acids Res. 30:1292-1305.[Abstract/Free Full Text]
    16. 16
  16. Modlin, J. F. 1996. Update on enterovirus infections in infants and children. Adv. Pediatr. Infect. Dis. 12:155-180.[Medline]
    17. 17
  17. Monpoeho, S., M. Coste-Burel, M. Costa-Mattioli, B. Besse, J. J. Chomel, S. Billaudel, and V. V. Ferre. 2002. Application of a real-time polymerase chain reaction with internal positive control for detection and quantification of enterovirus in cerebrospinal fluid. Eur. J. Clin. Microbiol. Infect. Dis. 21:532-536.[CrossRef][Medline]
    18. 18
  18. Nijhuis, M., N. Van Maarseveen, R. Schuurman, S. Verkuijlen, M. De Vos, K. Hendriksen, and A. M. Van Loon. 2002. Rapid and sensitive routine detection of all members of the genus enterovirus in different clinical specimens by real-time PCR. J. Clin. Microbiol. 40:3666-3670.[Abstract/Free Full Text]
    19. 19
  19. Rabenau, H., A. Clarici, G. Muhlbauer, A. Berger, A. Vince, S. Muller, E. Daghofer, B. Santner, E. Marth, and H. Kessler. 2002. Rapid detection of enterovirus infection by automated RNA extraction and real-time fluorescence PCR. J. Clin. Virol. 25:155-164.[CrossRef][Medline]
    20. 20
  20. Robinson, C. C., M. Willis, A. Meagher, K. E. Gieseker, H. Rotbart, and M. P. Glode. 2002. Impact of rapid polymerase chain reaction results on management of pediatric patients with enteroviral meningitis. Pediatr. Infect. Dis. J. 21:283-286.[CrossRef][Medline]
    21. 21
  21. Rotbart, H. A. 1990. Enzymatic RNA amplification of the enteroviruses. J. Clin. Microbiol. 28:438-442.[Abstract/Free Full Text]
    22. 22
  22. Rotbart, H. A. 2002. Treatment of picornavirus infections. Antivir. Res. 53:83-98.[CrossRef][Medline]
    23. 23
  23. Rotbart, H. A., M. H. Sawyer, S. Fast, C. Lewinski, N. Murphy, E. F. Keyser, J. Spadoro, S. Y. Kao, and M. Loeffelholz. 1994. Diagnosis of enteroviral meningitis by using PCR with a colorimetric microwell detection assay. J. Clin. Microbiol. 32:2590-2592.[Abstract/Free Full Text]
    24. 24
  24. Rychlik, W., W. J. Spencer, and R. E. Rhoads. 1990. Optimization of the annealing temperature for DNA amplification in vitro. Nucleic Acids Res. 18:6409-6412.[Abstract/Free Full Text]
    25. 25
  25. Savolainen, C., S. Blomqvist, M. N. Mulders, and T. Hovi. 2002. Genetic clustering of all 102 human rhinovirus prototype strains: serotype 87 is close to human enterovirus 70. J. Gen. Virol. 83:333-340.[Abstract/Free Full Text]
    26. 26
  26. Sawyer, M. H., D. Holland, N. Aintablian, J. D. Connor, E. F. Keyser, and N. J. Waecker, Jr. 1994. Diagnosis of enteroviral central nervous system infection by polymerase chain reaction during a large community outbreak. Pediatr. Infect. Dis. J. 13:177-182.[Medline]
    27. 27
  27. Schutten, M., B. van den Hoogen, M. E. van der Ende, R. A. Gruters, A. D. Osterhaus, and H. G. Niesters. 2000. Development of a real-time quantitative RT-PCR for the detection of HIV-2 RNA in plasma. J. Virol. Methods 88:81-87.[CrossRef][Medline]
    28. 28
  28. Tattevin, P., S. Minjolle, C. Arvieux, V. Clayessen, R. Colimon, J. Bouget, and C. Michelet. 2002. Benefits of management strategy adjustments during an outbreak of enterovirus meningitis in adults. Scand. J. Infect. Dis. 34:359-361.[CrossRef][Medline]
    29. 29
  29. Verstrepen, W. A., P. Bruynseels, and A. H. Mertens. 2002. Evaluation of a rapid real-time RT-PCR assay for detection of enterovirus RNA in cerebrospinal fluid specimens. J. Clin. Virol. 25 (Suppl.):39-43.[Medline]
    30. 30
  30. Verstrepen, W. A., S. Kuhn, M. M. Kockx, M. E. Van De Vyvere, and A. H. Mertens. 2001. Rapid detection of enterovirus RNA in cerebrospinal fluid specimens with a novel single-tube real-time reverse transcription-PCR assay. J. Clin. Microbiol. 39:4093-4096.[Abstract/Free Full Text]
    31. 31
  31. Watkins-Riedel, T., M. Woegerbauer, D. Hollemann, and P. Hufnagl. 2002. Rapid diagnosis of enterovirus infections by real-time PCR on the LightCycler using the TaqMan format. Diagn. Microbiol. Infect. Dis. 42:99-105.[CrossRef][Medline]
    32. 31
  32. Webster, A. D., H. A. Rotbart, T. Warner, P. Rudge, and N. Hyman. 1993. Diagnosis of enterovirus brain disease in hypogammaglobulinemic patients by polymerase chain reaction. Clin. Infect. Dis. 17:657-661.[Medline]
    33. 32
  33. Zoll, G. J., W. J. Melchers, H. Kopecka, G. Jambroes, H. J. van der Poel, and J. M. Galama. 1992. General primer-mediated polymerase chain reaction for detection of enteroviruses: application for diagnostic routine and persistent infections. J. Clin. Microbiol. 30:160-165.[Abstract/Free Full Text]

Journal of Clinical Microbiology, July 2003, p. 3133-3141, Vol. 41, No. 7
0095-1137/03/$08.00+0     DOI: 10.1128/JCM.41.7.3133-3141.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:

  • Mahony, J. B. (2008). Detection of Respiratory Viruses by Molecular Methods. Clin. Microbiol. Rev. 21: 716-747 [Abstract] [Full Text]  
  • Kost, C. B., Rogers, B., Oberste, M. S., Robinson, C., Eaves, B. L., Leos, K., Danielson, S., Satya, M., Weir, F., Nolte, F. S. (2007). Multicenter Beta Trial of the GeneXpert Enterovirus Assay. J. Clin. Microbiol. 45: 1081-1086 [Abstract] [Full Text]  
  • Tang, Y.-W., Sefers, S. E., Li, H., Kohn, D. J., Procop, G. W. (2005). Comparative Evaluation of Three Commercial Systems for Nucleic Acid Extraction from Urine Specimens. J. Clin. Microbiol. 43: 4830-4833 [Abstract] [Full Text]  
  • Landry, M. L., Garner, R., Ferguson, D. (2005). Real-Time Nucleic Acid Sequence-Based Amplification Using Molecular Beacons for Detection of Enterovirus RNA in Clinical Specimens. J. Clin. Microbiol. 43: 3136-3139 [Abstract] [Full Text]  
  • Hourfar, M. K., Roth, W. K., Seifried, E., Schmidt, M. (2004). Comparison of Two Real-Time Quantitative Assays for Detection of Severe Acute Respiratory Syndrome Coronavirus. J. Clin. Microbiol. 42: 2094-2100 [Abstract] [Full Text]  

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lai, K. K.-Y.
Right arrow Articles by Jerome, K. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lai, K. K.-Y.
Right arrow Articles by Jerome, K. R.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»