Sequencing and Resolution of Amplified Herpes Simplex Virus DNA with Intermediate Melting Curves as Genotype 1 or 2 by LightCycler PCR Assay

DNA from 101 specimens containing herpes simplex virus (HSV)produced atypical intermediate melting curves compared withthose expected for HSV type 1 or HSV type 2 subsequent to real-timePCR. Nucleic acid sequence analysis of amplified target DNArevealed 1- or 3-bp polymorphisms in the probe region whichallowed designation of these viruses as HSV genotype 1 or HSVgenotype 2. These two subpopulations of HSV were also identifiedas HSV genotype 1 or HSV genotype 2 using another commerciallyavailable PCR method. Amplified HSV target DNA producing intermediatemelting curves could be designated as HSV genotype 1 or HSVgenotype 2 without performing sequencing or another PCR methodwith 96/101 (95%) specimens by adding known intermediate HSVDNA characteristic for the two subpopulations as controls.

INTRODUCTION

Top

Introduction

Herpes simplex virus (HSV) causes superficial and systemic infectionswithin every major organ system of the body in both normal andimmunocompromised patients. Over a recent 5-year period, HSVwas the most common virus recovered in the diagnostic laboratoryusing conventional and shell vial cell culture methods (5).Nevertheless, HSV is rarely recovered from cerebrospinal fluid(CSF) specimens; PCR was demonstrated to have at least equalsensitivity to cell culture recovery of the virus from brainbiopsy tissue and is now recognized as the "gold standard" forthe laboratory diagnosis of central nervous system infectionsdue to this virus (8).

We converted culture-based assays and conventional PCR (gelelectrophoresis and Southern blotting or enzyme-linked immunosorbentassay) used for CSF specimens to real-time molecular amplificationsof this virus target nucleic acid for dermal and genital specimensbeginning in May 2000 and later for CSF specimens. Implementationof LightCycler real-time PCR (Roche Molecular Biochemicals,Indianapolis, Ind.) in the routine laboratory became possible,since amplification and detection of target HSV DNA occurs ina closed system, thereby virtually eliminating the threat ofcarry over amplicon contamination events. Trend analysis inour laboratory has provided an increased sensitivity for theroutine detection of HSV infections from genital (34.3% to 38.4%;12% increase) and dermal (23.6% to 27.7%; 17% increase) specimenscompared to shell vial cell cultures, in agreement with theexperience of others (2, 4, 6, 9). The assay also provides simultaneousgenotyping of HSV using melting curve analysis of the fluorescenceresonance energy transfer (FRET) hybridization probes. However,some amplified DNA of HSV exhibits atypical melting temperatures(T_m) that occur between the predicted T_ms obtained with HSVgenotype 1 and genotype 2. We sequenced the intermediate HSVPCR products to determine the polymorphism(s) responsible forthe atypical melting curves and to designate these as HSV genotype1 or 2.

(This report was presented in part at the 42nd Annual Meetingof IDSA [poster 991; session 105], 30 September to 3 October2004, Boston, Mass.)

MATERIALS AND METHODS

Top

Materials and Methods

This study was approved by the Institutional Review Board ofthe Mayo Foundation. Nucleic acid from specimens previouslyyielding intermediate melting curves with amplified target DNAof HSV (n = 101; genital, 56; dermal, 41; cerebrospinal fluid,4) were extracted using the MagNA Pure LC (Roche Molecular Biochemicals,Indianapolis, Ind.) automated extraction system according tothe manufacturer's instructions. HSV DNA was amplified by LightCyclerPCR using herpes simplex virus 1/2 Primer/Hyb Probes ASR, whichincluded HSV type 1 and type 2 reference DNA as controls (GenBankaccession number M16321, catalog no. 3 315 177; Roche AppliedScience, Indianapolis, IN). After amplification, the PCR productwas processed for DNA sequencing with shrimp alkaline phosphataseand exonuclease I. DNA sequence was determined bidirectionallyusing dye terminator chemistry on an ABI PRISM 377 DNA Sequencer(Applied Biosystems, Foster City, CA) with HSV primers (3).A second assay was used to resolve the specimens with targetHSV DNA producing intermediate melting curves (RealArt HSV 1/2LC PCR Reagent; Artus GmbH, Hamburg, Germany) as HSV genotype1 or HSV genotype 2.

RESULTS

Top

Results

Two subpopulations within the 101 specimens containing HSV DNAwith intermediate T_ms were identified based on the DNA sequencesobtained and were designated intermediate HSV type A (55%) andintermediate HSV type B (45%). Of these specimens (genital,56; dermal, 41; CSF, 4), 5 (dermal, 3; genital, 2) containedHSV intermediate DNA that could not be resolved as genotype1 or 2. Intermediate HSV type A had one polymorphism, and typeB had three polymorphisms (Table 1) in the probe region. Thepolymorphisms resulted in an altered FRET probe melting curvewith a peak T_m of 61.8°C for type A and 62.7°C for typeB, which fall between the T_m of HSV genotype 1 (55.3°C)and the T_m of genotype 2 (69.7°C) (Fig. 1). PCR resultsusing another LightCycler assay (RealArt HSV 1/2 LC PCR Reagent)found all intermediate HSV type A strains to be HSV genotype1 and all 101 of the intermediate HSV type B strains to be HSVgenotype 2.

View this table:
[in this window]
[in a new window]
TABLE 1. Polymorphisms in the probe region observed in HSV intermediate strains compared to reference strains

View larger version (25K):
[in this window]
[in a new window]
FIG. 1. Melting curves of HSV genotype 1, HSV genotype 2, and the intermediate HSV.

DISCUSSION

Top

Discussion

An advantage of the LightCycler hybridization FRET probes isthat they allow PCR product detection even in the presence ofpolymorphisms. This is especially important for dependable detectionof viruses due to the large number of polymorphisms often presentin viral specimens. Nevertheless, some specimens containingamplified DNA of HSV had polymorphisms under the FRET probesthat produced atypical melting curves which precluded designationas HSV genotype 1 or HSV genotype 2. We emphasize that all ofthe T_m patterns were diagnostic for HSV DNA. Using primer andprobe sequences within the DNA polymerase gene, Anderson etal. found that 15 of 745 (2.0%) genital and cutaneous specimenscontained HSV DNA that produced melting curve peaks that werenot specific for either genotype of HSV (1). In their practice,these melting curves from 15 samples conformed to six differentT_m patterns. We found a 2.5-fold higher incidence of these intermediatestrains at the Mayo Clinic using the Roche HSV1/2 primer/HybProbes ASR comprising two patterns. For example, during a recent9-month period of time, 164 of 3,269 (5.0%) specimens containingHSV demonstrated intermediate melting curves.

For diagnostic service to clinicians, specific resolution ofintermediate specimens of HSV into specific genotypes of thevirus may have importance in patient management. For example,the genotype of HSV (1 or 2) may influence the frequency ofreactivation of infection and is important for the managementof recrudescent infection since it confirms that the recurrentinfection is due to the same HSV genotype (10). In addition,infection with HSV type 1 can influence the clinical manifestationsthat may develop after subsequent infection with HSV type 2.In one study, for example, four patients who developed HSV type1 genital lesions first had a more prolonged episode and morefrequent recurrences following the acquisition of HSV type 2(7).

In the absence of commercially available reference DNA of intermediateHSV-1 and HSV-2, several general approaches can be used to characterizethese specimens detected by the Roche ASR. The most accuratemethod is to sequence the PCR amplicon from intermediate specimens,but this is not always feasible. A second approach is to retestthe intermediate specimens with another reference assay. A thirdapproach would be to retest intermediate specimens in the sametest run with specific target controls representing polymorphicHSV, genotypes characteristic for HSV-1 and HSV-2. The two intermediategenotypes can be specifically identified by small (0.9°C)but characteristic and reproducible melting curve peak T_m differences.

For routine laboratory practice we have optimized the thirdoption. For example, we attempted to differentiate the two intermediateHSV strains by determining a melting temperature and were successfulin achieving 100% specificity and more than 95% sensitivityon one LightCycler instrument; however, the variation in T_mwas too great to achieve consistent differentiation when differentreagent lots and instruments were compared. We were able toresolve and designate the two subpopulations of amplified HSVDNA that produced intermediate melting curves by using intermediatestrains with known sequence and genotype as controls withinthe run and designate the unknown intermediate HSV specimensas HSV genotype 1 and HSV genotype 2 by comparing the meltingtemperatures of these unknown specimens to the melting temperaturesof known intermediate strains. Using known intermediate strainsas controls with each run enabled us to correctly type 96 specimens(95%) out of the 101 intermediate specimens.

Conclusion. HSV specimens that cannot be initially placed into genotype1 or 2 by melting curve analysis using LightCycler PCR werefound to contain polymorphisms. These intermediate specimenscan be designated as HSV genotype 1 or HSV genotype 2 by addingknown intermediate strains as controls or by testing with anotherPCR method.

ACKNOWLEDGMENTS

M. J. Espy, J. R. Uhl, and T. F. Smith have codeveloped licensedanalyte-specific reagents for HSV PCR with Roche Molecular Biochemicals,Indianapolis, Ind.

FOOTNOTES

* Corresponding author. Mailing address: Division of Clinical Microbiology, Mayo Clinic, 200 First St., SW, Rochester, MN 55905. Phone: (507) 284-8146. Fax: (507) 284-4272. E-mail: tfsmith{at}mayo.edu.

REFERENCES

Top