Previous Article | Next Article ![]()
Journal of Clinical Microbiology, December 2006, p. 4464-4470, Vol. 44, No. 12
0095-1137/06/$08.00+0 doi:10.1128/JCM.00276-06
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
Alerts and Emergencies Unit, National Center of Microbiology, Instituto de Salud Carlos III, Majadahonda, Madrid, Spain,1 Arbovirus and Imported Viral Diseases Laboratory, Diagnostic Microbiology Service, National Center of Microbiology, Instituto de Salud Carlos III, Majadahonda, Madrid, Spain2
Received 8 February 2006/ Returned for modification 10 April 2006/ Accepted 6 October 2006
|
|
|---|
|
|
|---|
Several molecularly based diagnostic methods to detect the DNA of VARV and differentiate it from other orthopoxviruses were published before September 2001. These methods were based on the analysis of the pattern lengths of the restriction endonuclease digestion products of previously amplified genomes (13, 22).
Soon after this date, as an alternative to these time-consuming methods, several real-time PCR protocols for the detection and identification of orthopoxvirus DNA were described (5, 8, 17, 18, 19, 24).
In this study we describe a real-time PCR based on TaqMan 3'-minor groove binder (MGB) chemistry (Applied Biosystems) to simultaneously detect VARV and differentiate it from other orthopoxviruses in a single reaction tube. Moreover the method uses an internal control to detect false negatives.
A careful review of previously published methods for the real-time amplification of variola virus is also described.
|
|
|---|
Rabbitpox virus (RPXV), Utrecht strain, ectromelia virus (ECTV), Moscow strain, and CPXV, Brighton strain, were obtained from the American Type Culture Collection.
DNAs from Yersinia enterocolitica, Yersinia pseudotuberculosis, Bacillus anthracis, Salmonella typhi, Burkholderia mallei, Francisella tularensis subsp. holarctica, Brucella melitensis, Brucella suis, and Brucella abortus and from varicella-zoster virus and herpes simplex virus 1 were obtained to ascertain the specificity of the method.
Quality control panels. A first panel, containing 15 lyophilized human samples, spiked with some orthopoxviruses, was provided by the European Network for Imported Viral Diseases (ENIVD). The samples contained camelpox virus (CMLV), CPXV (81/02 and Brighton strains), VACV (Elsree and modified virus Ankara strains), ECTV, and serial dilutions of MPXV (Lam87 strain) (15). Moreover, 10-fold dilutions of MPXV, CMLV, CPXV, and VACV were obtained from P7, P9, P12, P13, and P15 samples of this panel to determine the sensitivity limits of each orthopoxvirus.
The second orthopoxvirus PCR European quality control panel, similar to the previous one, was sent to our laboratory to form part of a number of quality assurance studies by ENIVD. The panel included serial dilutions of MPXV, CPXV, VACV, tanapox virus (genus Yatapoxvirus), and parapox virus (genus Parapoxvirus). Also, four samples included inhibitor factors to test for possible false-negative results (16).
Primers and probes. The highly conserved genomic region of the crmB gene was the target used to design a real-time PCR assay. Nucleotide sequences for VARV and other orthopoxviruses were obtained from GenBank databases and then aligned using MEGA 2 software (9) (Fig. 1). Primers (Formia 1 and Formia 3) and probes were selected using the Primer Express software (Applied Biosystems). Each probe contained a different fluorescent report dye at the 5' end: 6-carboxyfluorescein (FAM; VAR130, specific for VARV), 6-carboxyrhodamine (VIC; OPOX143; generic for orthopoxviruses), and NED (IC; specific for the internal control). MGB and nonfluorescent quencher (NFQ) were added to the 3' end of each (Table 1).
![]() View larger version (53K): [in a new window] |
FIG. 1. Alignment of orthopoxvirus and probe sequences within the crmB gene. BPXV, buffalopox virus; GBLV, taterapox virus. VIR 130 is specific probe for variola virus; ortopox 143 is generic probe. Accession numbers are shaded.
|
|
View this table: [in a new window] |
TABLE 1. Primer and probe sequences
|
Construction of an internal control plasmid. A DNA fragment was obtained by an overlap extension PCR (18) carried out with ICS and ICA primers (Table 1). The product was cloned and sequenced as previously described. As expected, the cloned DNA contained the binding sites for the Formia 1 and Formia 3 primers, flanking the heterologous sequence selected as the IC probe.
DNA extraction. DNA from VACV was isolated from the infected-cell cultures from preparations of viruses with a guanidinium thiocyanate lysis buffer by following a previously described procedure (3).
DNA from samples included in the ENIVD quality controls and from RPXV, ECTV, and CPXV was extracted using a QIAamp RNA minikit (QIAGEN, Hilden, Germany) by following the manufacturer's instructions. Each tube used in the assay contained 100 copies of internal control plasmid.
Real-time PCR assay. Genome amplifications were carried out in triplicate in an ABI PRISM 7000 sequence detection system (Applied Biosystems, Foster City, CA) in a final volume of 50 µl including 25 µl of TaqMan 2x Universal PCR Master Mix, No Amperase (Applied Biosystems), 900 nM of Formia 1 primer, 300 nM of Formia 3 primer, 200 nM of each of three fluorogenic TaqMan MGB probes, and 5 µl of extracted DNA. Thermal cycling was performed as follows: 1 cycle at 95°C for 10 min, followed by 45 cycles of 95°C for 15 s and 60°C for 1 min.
|
|
|---|
![]() View larger version (34K): [in a new window] |
FIG. 2. Detection of chimerical VARV DNA by TaqMan real-time PCR. Amplification plots (cycle number versus delta Rn) of serial 10-fold dilutions (106 to 10 copies per tube) of chimerical VARV DNA tested in triplicate in the same experiment. NC is the negative control. Rn, the normalized reporter signal, is the fluorescence signal of the reporter dye divided by the fluorescence signal of the passive, internal reference dye. Delta Rn is determined by the formula Rn+ Rn, where Rn+ is the Rn value for a reaction involving all components, including the template, and Rn is the value for an unreacted sample.
|
![]() View larger version (41K): [in a new window] |
FIG. 3. Detection of VACV (Western Reserve strain) by TaqMan real-time PCR. The amplification plot was realized on 10-fold dilutions (106 to 1 PFU per tube), assayed in triplicate, of extracted DNA from a VACV stock. See the Fig. 2 legend for Delta Rn definition.
|
![]() View larger version (23K): [in a new window] |
FIG. 4. Linearity of the TaqMan PCR assay. The standard curve is generated by plotting the observed Ct value, measured in triplicate, against (a) the log10 of the input copy number (106 to 10 copies per tube) of chimerical VARV DNA and (b) the log of serial dilutions (106 to 1 PFU per tube) of VACV.
|
Specificity of the assay. The specificity of the probes was determined by assessing virus-specific fluorescence signals with the FAM, VIC, and NED reporter dyes developed in each reaction and using as templates chimerical VARV DNA and genomic DNA from MPXV (Lam87 strain), CMLV, RPXV (Utrecht strain), CPXV (81/02 and Brighton strains), ECTV (Moscow strain), and VACV (Elstree, Western Reserve, and modified virus Ankara strains). The VAR130 probe can recognize specifically recognize VARV DNA, whereas other different orthopoxvirus DNAs were recognized only by OPOX143. Neither ECTV, an orthopoxvirus not infecting humans that lacks the selected crmB region, nor tanapox virus and parapox virus, other poxviruses not belonging to the Orthopoxvirus genus, were detected. When samples containing amplification inhibitors were assayed, the IC probe showed a negative signal, identifying potential false-negative results (Table 2).
|
View this table: [in a new window] |
TABLE 2. Second orthopoxvirus PCR European quality control panel
|
Reproducibility of real-time PCR assay. In order to evaluate the intra-assay reproducibility, a dilution end point standard curve was drawn in triplicate. Ct values were plotted against the log10 of the known input copy number dilutions (107 to 100 copies) of chimerical VARV DNA; the variation coefficient average was 0.4%.
To determine the interassay reproducibility, Ct values were measured in triplicate and in three different days; a variation coefficient average of 0.34% was obtained.
In multiple assays 100 copies of VARV DNA were regularly detected; however, 10 copies were detected in some experiments. Amplification efficiencies were similar, as indicated by the slopes of the curves, the values of which ranged between 3.32 and 3.76.
|
|
|---|
In the past, electron microscopy assays, virus cultures, and serology methods have been used in the laboratory diagnosis of smallpox. Additionally, classical molecular methods such as PCR or nested PCR, DNA endonuclease cleavage, and genome sequencing have proved highly efficient, although they are time-consuming and could generate false positives due to possible contamination events. Recently, an oligonucleotide-based microchip technology (10, 11) capable of detecting and differentiating members of the genus Orthopoxvirus has been described. However, its implementation in other laboratories could be difficult, due to the need for technical experts to prepare this kind of assay. Therefore, the applicability of these methods in antibioterrorism programs is currently being discussed.
A bioterrorism response plan depends on the immediate clinical suspicion of infectious disease and on rapid and accurate laboratory identification of the variola virus. Early diagnosis is essential to control VARV infection, mainly in the initial phases. Based on the rapidity and efficiency of real-time PCR assays, several laboratories involved in biodefense programs are turning to them for the early diagnosis of VARV infection.
Recently, several different real-time PCR assays for orthopoxvirus detection have been described. Most of them use generic detection of orthopoxviruses and posterior melting analysis for VARV identification. Others differentiate VARV from other members of the Orthopoxvirus genus by means of VARV-specific assays. However, it would appear that none of these assays may have actually combined robust and specific VARV identification in the presence of an internal control.
In this study we describe a real-time PCR assay for the rapid detection and simultaneous differentiation of VARV from other orthopoxviruses using a specific VARV probe. We take advantage of the new generation of TaqMan probes with a conjugated MGB group and an NFQ at the 3' end. Upon hybridization to a complementary target, the MGB molecule folds into a duplex and hyperstabilizes it, allowing the use of a shorter and more specific probe that still meets the high-melting-temperature (Tm) requirement for a PCR to be designed (1, 6). This is very important when the design of a specific probe in a genome as highly conserved as that of the Orthopoxvirus genus is required. After a general analysis of Orthopoxvirus genomes, we identified the crmB gene as an attractive target for our objective; with respect to other genes analyzed, it contains a homologous region for the design of generic primers and probes and a polymorph target region where a specific probe for VARV can be selected.
The primers used in this assay succeeded in amplifying not only VARV but also several orthopoxviruses, including all those capable of causing disease in humans. We detected MPXV (Lam87 strain), CMLV, CPXV (81/02 and Brighton strains), and VACV (Elstree and modified virus Ankara strains) present in the quality control panels. We did not detect ECTV, a virus that infects mice but that is not considered to be an infectious agent for humans (12), since its genome is deleted in the selected target region (21). Parapox and tanapox viruses, included in the second ENIVD panels and not belonging to the Orthopoxvirus genus, rendered negative results (Table 2).
The specificity of the method has also been proved with other DNAs from agents that should be considered in the differential diagnosis of smallpox. None of these produced fluorescent signals with the VAR130 or OPOX143 probe.
Moreover, the assay includes an internal control to detect the potential presence of inhibitors in clinical specimens that may lead to false-negative results. One hundred copies of the internal control plasmid were inserted into the buffer for DNA extraction, amplified in the same reaction tube, and distinguished by the specific probe for the plasmid. The internal control had no influence on the performance of the assay, nor did it decrease sensitivity. Additionally, the robustness of the internal control, which had been shown to be effective in detecting inhibitor effects, was analyzed by testing a panel of 10 negative human samples in triplicate (clinical samples routinely tested for herpesvirus and enterovirus). This was done by using another internal-control PCR (4). The internal control obtained in this study showed very high efficiency, failing in only 1 out of 10 samples (data not shown).
External quality assurance studies were conducted through the ENIVD in order to ascertain the diagnostic proficiency of the laboratories involved in orthopoxvirus diagnosis. We feel that the results obtained in the second ENIVD quality control panel (Table 2) and the good score achieved, based on sensitivity, specificity, rapidity, and ability to detect inhibitor effects of the samples, show that this assay is a sensitive, specific, and rapid tool for the diagnosis of VARV infection. Comparison of results with a conventional generic nested PCR for orthopoxvirus (23), previously designed in our laboratory and confirmed by two ENIVD quality control panels, showed the same sensitivity limit. Also, the possibility of obtaining synthetic DNA with the same properties as genomic DNA to be used as positive controls makes this method easily transferable to other laboratories. This is a very important factor in establishing a rapid response to a possible biological alert.
There is only one previous description of a real-time assay including an internal control (19); nevertheless, this method, based on fluorescence resonance energy transfer probes and Tm analysis, could pose problems of specificity, since other orthopoxviruses (GenBank accession numbers: CPXV, AF375084, AF375085, AF375086, AF377878, AF377879, AF377880, AF377881, AF377882, AF377883, AF377884, AF377885, and AF377886; taterapox virus, AF375093; elephantpox virus, AF375090) have the same sequence as VARV at the probe sites. In addition, it may be the case that many of the previously published methods, based on real-time PCR and the use of specific probes, require reconsideration on a theoretical basis.
Sofi Ibrahim et al. (24) proposed the use of VARV-specific TaqMan probes designed on the hemagglutinin gene, but there are 7 out 32 VARV sequences that present one mismatch with these probes (GenBank accession numbers AF375129, AF375130, AF375137, AF375138, AF375141, AF375142, and Y16780). The use of short probes with an internal mismatch could theoretically dramatically decrease the sensitivity of the method. Recently, Kulesh et al. (8) also proposed two alternative methods based on other genes. One of these involved the use of a specific probe on the B9R gene, but two CPXV strains (GenBank accession numbers AY519982 and AY519984) have the same sequence as the proposed VARV-specific probe; moreover, one VARV strain (GenBank accession number AY552594) has one mismatch with the probe. The first real-time method for identifying VARV published (5) proposes the use of probes with sequences of VARV; however, these sequences are also present in CPXV (GenBank accession numbers AF375084, AF377878, AF377879, AF377880, AF377881, AF377882, AF377883, AF377886, AF375084, AF375086, and AF375087), CMLV (GenBank accession numbers AF375079, AF375080, AF375081, AF375082, AF438165, and AY009089) and taterapox virus (GenBank accession number AF375093).
Another method based on melting analysis (18) also poses theoretical problems because it is based on the presence of one mismatch between the VARV sequences and the probe; unfortunately all the known equivalent sequences of CMLV also have one mismatch (GenBank accession numbers AF438165, AY009089, AY223496, AY299081, AY299082, AY299083, AY299084, AY299085, AY299086, AY299087, and X75156).
Finally, the three methods recently proposed by Nitsche et al. (17) are also based on the melting analysis of probes selected on the rpo18, VETF, and A13L genes; none of them have theoretical problems, but only three VARV sequences are known for these genes.
The method proposed here has been designed on the basis of 129 sequences of orthopoxviruses for the crmB gene, including 15 VARV sequences; the VAR130 probe was selected because it represents a specific sequence for VARV that is present in all the known VARV genomes in the crmB gene but not in the other orthopoxviruses, which show between 2 and more than 10 mismatches with the probe (Fig. 1). Clearly, new knowledge of sequences of VARV or other orthopoxviruses could also affect the proficiency of this design in the future. That is why, as in the case of previous studies, we emphasize the need for conventional molecular methods, such as identification by sequencing the amplified product, in order to confirm a positive results of orthopoxvirus infection.
This study was partially funded by the Instituto de Salud Carlos III. A. Negredo and F. Molero are engaged under an agreement between the Spanish Defense Ministry and the Instituto de Salud Carlos III for the development of safe and sensitive molecular tools for the detection of viruses with the potential to be used as biological weapons.
Published ahead of print on 25 October 2005. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2010 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»