Design of primer pairs with group- and serotype-specificities.
One of the advantages of SYBR Green-based real-time RT-PCR is that it is relatively easy to design and test primer pairs suitable for RT-PCR analysis. Primers were selected based on highly conserved regions of dengue virus genome. After nucleotide sequence alignment using DNA Star software (Perkin-Elmer), potential target regions were identified in the regions of core, NS3, NS5, and 3′ noncoding genes. More than 30 primer pairs were synthesized and screened for initial evaluation. Among these primers tested, a set of group- and serotype-specific primer pairs in the core gene region was found to be most sensitive. The primers were designed to take into account of mismatches among strains and to avoid primer-dimer formations. The final sequences and the genomic locations of these primers are shown in Table . The condition of thermal profile for one-step SYBR Green-based RT-PCRs and the optimal concentrations of primer pairs were then optimized to increase the sensitivity and specificity. The optimal concentrations for each of the primer pairs were titrated and found to be 50 and 50 nM for the DN-F and DN-R pair (group specific), 50 and 50 nM for the DN-F and D1-R pair (DEN-1 specific), 50 and 50 nM for the DN-F and D2-R pair (DEN-2 specific), 50 and 25 nM for the DN-F and D3-R pair (DEN-3 specific), and 50 and 50 nM for the DN-F and D4-R pair (DEN-4 specific), respectively. The Tm values of primer-dimers were found to be below 75°C, whereas the Tm values of each of these amplicons from the group- and serotype-specific primer pairs were in the range of 79 to 83°C depending on the various strains tested. No primer-dimers were detected as demonstrated by melting curve analysis if optimal primer concentrations were followed.
Sensitivity of SYBR Green I-based one-step RT-PCR.
To ascertain the detection limits of the one-step SYBR Green-based RT-PCR method using group- and serotype-specific primer pairs in dengue diagnosis, we tested 10-fold serial dilutions of seed viruses that had been quantitated by plaque forming assay. Figure and Fig. showed the standard curves of four dengue prototype viruses using group- and serotype-specific primer pairs, respectively. The detection limit of each group-specific primer pair was between 4.1 and 43.5 PFU/ml for various dengue serotypes. The detection limits of the serotype-specific primer pairs were calculated to be 10 PFU/ml for DEN-1, 4.6 PFU/ml for DEN-2, 4.1 PFU/ml for DEN-3, and 5 PFU/ml for DEN-4.
FIG. 1. Standard curves of four dengue virus serotypes tested by one-step SYBR Green-based quantitative RT-PCR using group-specific primer pair. Standard curves were generated from the amplification plots of each of the four dengue strains representing DEN-1 (more ...)
FIG. 2. Standard curves of four dengue virus serotypes tested by one-step SYBR Green-based quantitative RT-PCR using serotype-specific primer pair. Standard curves were generated from the amplification plots of each of the four dengue strains representing DEN-1 (more ...) Specificity of SYBR Green I-based one-step RT-PCR.
To verify that the dengue group- and serotype-specific primer pairs can be used to detect and differentiate the majority strains of four dengue virus serotypes, we had selected and tested 16 dengue virus strains from our virus bank. These included 4 prototype strains collected from the American Type Culture Collection and 12 local isolates from patients with imported or indigenous dengue during 1987 to 2000. Most of the strains had been sequenced at the core and envelope gene regions and could be assigned to distinct genotypes. Among these, D4 8700544 was an unique isolate which could neither be grouped into the DEN-4 genotype I nor II based on the sequences in the core and envelope regions (data not shown). Table showed the results of SYBR Green I-based one-step RT-PCR analysis. It was demonstrated that these group- and serotype-specific primer pairs were highly specific in the detection and differentiation of various dengue virus strains. None of these primer sets, however, amplified West Nile virus (Eg101 strain), yellow fever virus (17D vaccine strain), and Japanese encephalitis virus (JaGAr strain). This is in contrast to a panel of primer pairs with broad specificity to flavivirus or restricted specificity to West Nile virus, yellow fever virus or Japanese encephalitis virus, which had been shown to successfully detect and differentiate each of these corresponding viruses (data not shown). Figure shows the results of agarose gels demonstrating that this group- and serotype-specific primer pairs are highly specific in the detection and differentiation of various strains since the amplicons generated have the correct size with expected sequences (data not shown). The 171 bp that appear in Fig. were the expected products for all four dengue serotypes using group-specific primer pair (DN-F/DN-R), whereas the 193-, 204-, 204-, and 132-bp amplicons were the expected products for each of the DEN-1, DEN-2, DEN-3, and DEN-4 serotype, respectively, using type-specific primer pairs. The 375-bp amplicon appeared in Fig. was found to be a longer product sharing the same sequences with 171-bp amplicon at its 5′ end with additional DEN-1 sequences following it.
FIG. 3. Analysis of amplification products of SYBR Green-based real-time RT-PCR using group- and serotype-specific primer pairs. The RT-PCR products were observed in ethidium bromide-stained agarose gel. Input viral RNAs were extracted from each of the 16 representative (more ...) Evaluation of SYBR Green I-based one-step RT-PCR for the detection of dengue virus RNA in the acute-phase serum samples.
To evaluate the SYBR Green I-based one-step RT-PCR in the clinical diagnosis of dengue virus RNA, we analyzed 193 acute-phase serum samples collected from confirmed dengue patients during the period between 1998 and 2002. These acute-phase serum samples selected for analysis were collected from confirmed dengue patients with either positive dengue virus isolation or positive serological test without taking into account of RT-PCR result. The results are shown in Table and revealed that 83 versus 67% out of 193 acute-phase serum samples tested were positive for one-step SYBR Green-based RT-PCR method and cell culture method, respectively. The positive rates for each of four dengue serotypes were found to be 63% for DEN-1, 67% for DEN-2, 77% for DEN-3, and 60% for DEN-4 using the virus isolation method, versus 75% for DEN-1, 82% for DEN-2, 92% for DEN-3, and 100% for DEN-4 using the one-step SYBR Green-based RT-PCR. Further analysis showed that the sera that were dengue virus negative by virus isolation and dengue virus positive by the SYBR Green-based RT-PCR assay had relatively fewer amplicons than those samples that were positive by both virus isolation positive and the SYBR Green-based RT-PCR assay (Fig. ).
Comparison of results obtained by one-step SYBR Green-based RT-PCR assay and cell culture method for detection of dengue virus in acute-phase seraa
FIG. 4. Comparisons of dengue viral titers between the serum samples negative for dengue virus by virus isolation and positive for dengue virus by the SYBR Green-based RT-PCR method (VI-PCR+) and those positive by both virus isolation and the SYBR Green-based (more ...) Comparison of sensitivities of the one-step SYBR Green-based RT-PCR method with cell culture assays in the detection of dengue virus in the acute-phase sera with dengue-specific IgM and/or IgG antibodies.
To further address the detection limit of the one-step SYBR Green-based RT-PCR, we had reclassified the above 193 serum samples with the criterion of presence or absence of dengue-specific IgM and/or IgG antibodies. The results are listed in Table . Compared with the cell culture method, the one-step SYBR Green-based RT-PCR method could detect twice as many acute-phase serum samples with positive dengue-specific IgM and/or IgG antibodies as the cell culture method. Two factors may account for the lack of sensitivity of the cell culture method: (i) a greater volume of serum samples is available (4 μl for cell culture versus 20 μl for real-time RT-PCR was added for each reaction, respectively) (Fig. ) and (ii) real-time RT-PCR is less influenced by neutralizing IgM and/or IgG antibodies than cell culture method (Table ). This result is very encouraging and suggests that one-step SYBR Green-based RT-PCR could replace the conventional cell culture method as the new gold standard in future clinical diagnosis of dengue virus infection.
Comparison of results obtained by one-step SYBR Green-based RT-PCR assay and cell culture method for antibody detection in acute-phase seraa
The application of PCR in molecular diagnosis has gradually replaced traditional cell culture method as the gold standard for virus detection. The real-time PCR assays have many advantages comparing to conventional PCR including simplicity, rapidity, sensitivity, and low contamination. There are currently five main chemistries used for the detection of PCR product during real-time PCR (16
). Among these, the most widely used format is the 5′→ 3′ nuclease oligoprobe (TaqMan assay) although this is most likely due to its commercial maturity. The TaqMan real-time PCR is highly specific due to the sequence-specific hybridization of the probe. Theoretically, it has the potential to develop multiplex up to four fluorophores in a single tube. This has led to a great expectation to analyze multiple pathogens and serotypes in clinical diagnosis. Ideally, a four-color multiplex protocol could be developed to detect and differentiate the four dengue serotypes using TaqMan one-step real-time RT-PCR. This goal, however, had not been realized at the present time.
Although the SYBR Green-based real-time PCR assays is less specific than the TaqMan real-time RT-PCR, it has the advantages of simplicity in primer design and universal RT-PCR protocols suitable for multiple target sequences. We have found that the optimization of primer concentration is critical in preventing primer-dimers and nonspecific amplification of other unrelated gene products. Indeed, nonspecific products such as primer-dimers and nonspecific amplification of other unrelated gene products could be occasionally detected at higher primer concentrations. However, these nonspecific products could be prevented if optimal concentrations were followed using the real-time RT-PCR protocol reported in this study. To assure the specificity of amplicons produced from SYBR Green I-based real-time RT-PCR in daily routine screening, both flavivirus-specific (FL-F, GCC ATA TGG TAC ATG TGG; FL-R, TGT CCC ATC CTG CGG TAT CAT [200 to 100 nM]) and dengue-specific primer pairs were used for each of the serum samples tested. Those serum samples positive for initial screening will then be tested for serotype by each of the four serotype-specific primer pairs. Excellent correlation was found among these six primer pairs over a thousand of serum samples tested so far. To further confirm the positive results of the SYBR Green-based RT-PCR method in clinical diagnosis, various confirmatory tests could be used. These include (i) virus isolation, (ii) TaqMan RT-PCR assay, and (iii) capture IgM and IgG ELISA to detect increased IgM and/or IgG antibodies using paired serum samples. We considered capture IgM and IgG ELISA the most suitable method for a confirmatory testing for the use of the SYBR Green-based RT-PCR method in clinical diagnosis due to its simplicity and reliability. In addition, most dengue virus laboratories routinely perform this test for their suspected dengue patients.
In an attempt to develop a universal diagnostic real-time RT-PCT protocol for arbovirus, we have taken an alternative approach using a SYBR Green I-based detection system. Kuno (13
) was the first to propose a two-stage procedure for the systemic diagnosis of arbovirus using old conventional RT-PCR. In the first stage, broadly group-reactive primers are used to narrow the range of possible etiologic agents; in the second stage, virus-specific primers are employed for identification. It was emphasized that multiple group-reactive primers is essential to ensure the successful screening of a clinical sample against a multiple number of potential etiologic agents. Through careful design and testing, we have successfully developed a one-step RT-PCR system that can be used to detect and differentiate several flaviviruses, including dengue virus, Japanese encephalitis virus, Yellow fever virus, and West Nile virus (data not shown).
In the present study, we reported the development of group- and serotype-specific one-step SYBR Green I-based real-time RT-PCR assay for the screening and typing of dengue virus RNA. Analysis of clinical acute-phase serum samples demonstrated that the one-step SYBR Green-based RT-PCR method could detect twice as many acute-phase serum samples with positive dengue-specific IgM and/or IgG antibodies than cell culture method. The greatest advantage of this system is that one can actually detect and differentiate multiple viruses in a single run using the universal RT-PCR protocol with a panel of group- and type-specific primer sets. Our results suggest that one-step SYBR Green I-based RT-PCR system using the universal RT-PCR protocol should have great potential in the clinical diagnosis and epidemiological surveillance of viral diseases.