Indian Journal of Medical Microbiology Medknow Publications on behalf of Indian Association of Medical Microbiology ISSN: 0255-0857 EISSN: 1998-3646 Vol. 29, Num. 2, 2011, pp. 102-109

Indian Journal of Medical Microbiology, Vol. 29, No. 2, April-June, 2011, pp. 102-109

Original Article

Comparison of a conventional polymerase chain reaction with real-time polymerase chain reaction for the detection of neurotropic viruses in cerebrospinal fluid samples

M Ramamurthy¹, M Alexander², S Aaron², R Kannangai¹, V Ravi³, G Sridharan¹, AM Abraham¹

¹ Department of Clinical Virology, Christian Medical College, Vellore - 632 004, India
² Department of Neurology, Christian Medical College, Vellore - 632 004, India
³ Department of Neurovirology, NIMHANS, Bangalore - 560 029, India

Correspondence Address: A M Abraham, Department of Clinical Virology, Christian Medical College, Vellore - 632 004, India, asha_ma@cmcvellore.ac.in

Date of Submission: 12-Nov-2010
Date of Acceptance: 23-Mar-2011

Code Number: mb11027

PMID: 21654102

DOI: 10.4103/0255-0857.81777

Abstract

Keywords: CNS infections, neurotropic viruses, real-time PCR

Introduction

The neurological illnesses produced by individual herpes viruses have both common as well as distinctive features. The term neurotropic viruses used here is in the context of the ability of certain DNA and RNA viruses to produce pathology in the central nervous system (CNS). Almost all DNA viruses with the ability to reach and cause disease in the CNS are spread among humans by the respiratory route or contact with infected saliva or genital secretions. ^[1] In immunocompetent people, the viruses multiply in the upper respiratory tract, producing a brief viraemia with or without primary illness. Almost all the viruses have the ability to become latent and undergo reactivation at different times in the life of the host. Reactivation may be accompanied by overt clinical disease. The viruses reach the CNS either during primary infection or reactivation. When the viruses reach the CNS, they exhibit the potential for causing neuropathology and, hence, disease. The clinical disease may be acute, chronic or subacute. ^[2],[3] DNA viruses primarily cause infections in the CNS during virus reactivation from a latent state. The infection is predominantly endogenous in origin and sporadic cases occur throughout the year. In contrast, RNA viruses that cause neurological disease tend to cause epidemic CNS disease, as in the case of arboviruses like Japanese encephalitis and West Nile viruses, which are vector borne. ^[4] The neurological manifestations associated with DNA viruses include meningitis, encephalitis, meingoencephalitis, myelitis and radiculopathies, and these occur as a result of direct virus invasion and/or as an inflammatory response. Demyelination is seen post-viral infections as a parainfectious syndrome. ^[2]

The published literature clearly shows the importance of DNA viruses in neurological disease, with almost all the publications from the West. ^[5],[6] To the best of our knowledge, there are no reports from tropical developing countries like India exploring this problem. The conventional methods of virus isolation and serology have a limited role in investigating neurological disease. ^[7],[8] The availability of molecular testing methods for viral infections in different centres in India will provide opportunities for these much-needed studies. Our study is the first from India reporting the association of several DNA viruses with different acute/subacute or chronic neurological disease.

Materials and Methods

Inclusion criteria

Patients diagnosed to have aseptic meningitis, meningoencephalitis, acute disseminated encephalitis, transverse myelitis, Guillain-Barre΄ syndrome (GBS) and motor neuron disease were recruited for the study. CSF samples were sent by the treating physician for clinicopathological/microbiological/virological investigations as part of routine patient management. This study was carried out on samples received at the Department of Clinical Virology of a tertiary care hospital for testing and those received at the Clinical Pathology Department for CSF analysis.

Exclusion criteria

Where there was definite evidence of bacterial infection, the samples from these individuals were not tested for viral etiology. A few were however tested and included in the control group (not for the test group). The samples used for molecular testing were categorized as patient group (test) and control samples.

The study was performed on adequate numbers of CSF samples obtained from as many patients. These patients fulfilled the inclusion criteria. No repeat samples were included. Multiplex non-nested conventional PCR and real-time PCR were compared on 158 samples. Of the 158 samples, the housekeeping alpha-tubulin gene failed to amplify despite testing twice for 11 samples. These 11 samples were excluded from the analysis.

Patient group (n = 126)

This included patients whose CSF showed lymphocytic predominance with more than 5 cells/μl and whose clinical diagnoses were consistent with the categories of acute and chronic neurological illnesses of suspected viral aetiology. This group is also referred to as the "test" group, and consisted of patients with suspected viral aetiology in different acute and chronic neurological conditions. The clinical diagnoses on these individuals were meningoencephalitis, meningitis, aseptic meningitis, neonatal meningitis, transverse myelitis, acute demyelinating encephalomyelitis, encephalitis, motor neuron disease, GBS or miscellaneous conditions designated here as "others". The "others" category included diagnoses of altered sensorium (early stage of meningitis/meningoencephalits/encephalitis), myeloradiculopathy, acute severe encephalopathy and post-encephalitic sequalae.

Control group (n = 21)

CSF specimens collected from patients with neurological illnesses like benign intrathecal hypertension, tuberculous meningitis, diabetic neuropathy, bacterial meningitis, transverse myelitis of tumour aetiology and subacute sclerosing panencephalitis (SSPE) served as controls, along with CSF samples from patients with AV shunts among miscellaneous conditions, with a total WBC count of less than 5 cells/μl. SSPE was included here to serve as a negative control for PCR for DNA viruses as these samples are known to contain measles virus RNA. This category included CSF samples collected from patients with neurological illness not suspected to be of viral origin, except SSPE.

Collection and processing of samples

CSF samples were collected from patients with neurological disorders who visited this tertiary care hospital. These CSF samples were aliquoted and stored at -70 ^o C until they were tested.

DNA extraction

Viral DNA was obtained from the CSF samples using the QIAamp DNA blood mini kit (Qiagen, Hilden, Germany). Two hundred microlitres of sample was used for the extraction, which was carried out as per the manufacturer′s instructions. Nucleic acid was eluted in 60 μl of the elution buffer (AE) provided by the manufacturer.

Conventional PCR

The herpes simplex viruses 1 and 2 (HSV-1 and HSV-2), VZV, EBV, CMV and JC viruses were amplified in two multiplex PCR mixes and one uniplex mix. The first PCR mix contained primers for HSV-1, HSV-2 and CMV. The second PCR mix contained primers for EBV and VZV and the third PCR mix contained primers for JCV. The source of positive controls for viruses like HSV-1, HSV-2, EBV and CMV viruses was cell culture supernatant. VZV DNA was extracted from the live attenuated vaccine (Oka strain) with known titre (1000 PFU/1 ml) while the control used for JCV was a plasmid (obtained from Johns Hopkins University, Baltimore, USA). DNA was extracted from these positive controls using the QIAamp DNA blood mini kit (Qiagen). Sterile Milli Q water was used in PCR testing after every third sample as negative control to check for cross-contamination and integrity of PCR reagents. DNA was extracted from the characterised virus strains (HSV-1, HSV-2, CMV and VZV). Cloned plasmids were used for EBV and JCV. The virus control mixes were pooled to create templates for the different primer cocktails or used individually in the PCR reaction. In all PCR runs, known positive controls and negative extraction controls (no-template controls) were used. The PCR runs were validated only if the controls were satisfactory. All negative samples were checked for PCR inhibition by amplification of the alpha-tubulin gene using a non-nested format.

Amplification

A non-nested multiplex PCR was performed to amplify HSV-1, HSV-2, EBV, CMV and VZV ^[9] . JCV ^[10] was amplified by a hemi-nested PCR with two rounds of amplification. The primers used for the amplification were custom synthesized (Hysel India Pvt. Ltd., New Delhi, India). The nucleotide sequences of primers used for HSV- 1, HSV-2, VZV, EBV, CMV and JC viruses are provided in [Table - 1].

The amplification was performed using 50 μl reactants with 5 μl of the extracted DNA. The thermal cycling conditions for the initial Taq polymerase activation for all the three PCR reactions were at 95°C for 15 min. The cycling conditions for cocktail-1 were 94°C for 30 s, 52°C for 40 s, 72°C for 50 s for 40 cycles and final extension at 78°C for 15 min. The conditions for cocktail-2 were 94°C for 45 s, 57.5°C for 50 s, 72°C for 70 s for 40 cycles and final extension at 72°C for 7 min. Mix 3 was used in a hemi-nested PCR with two rounds of amplification. In the first round, the cycling conditions were 94°C for 3 min, 94°C for 20 s, 53°C for 20 s and 72°C for 5 min for 30 cycles. The second round PCR had similar cycling conditions, except that the annealing temperature was 58°C for 20 s. All amplifications were carried out using thermal cyclers GeneAmp PCR system 2400 (Perkin Elmer, Norwalk, CT, USA), PTC-100 (MJ Research, Waltham, MA, USA) or Mycycler (BioRad , Hercules, CA, USA).

Detection of the polymerase chain reaction product

The amplified product was detected by agarose gel electrophoresis using 2% agarose gel in 1X TBE buffer containing 0.5 μg/mL ethidium bromide (Sigma, St. Louis, MO, USA) at 120V for 50 min. Phi X174 DNA-Hae III digest (Bangalore Genei, Bangalore, India) was used as a molecular weight marker. The gels were viewed in a UV transilluminator (Mighty Bright; Hoefer Scientific Instruments, San Francisco, CA, USA). The size of the amplified product was determined using the gel documentation system (Gel Doc; BioRad, Hercules, CA, USA) using the Quantity One software version 4.1.1 (BioRad, Hercules, CA, USA). The validation of the multiplex conventional PCR was carried out by establishing the lower limit of detection using either cloned plasmids or specific viruses. The specificity of the primers was established by testing with heterologous plasmids or viruses for HSV-1, HSV-2, VZV, CMV, EBV and JCV. The JC PCR was a hemi-nested PCR and the lower limit of detection was estimated using a cloned plasmid preparation that was derived from the PCR product. The PCR product sizes were cocktail-1 (HSV-1: 147 bp, HSV-2: 227 bp, CMV: 256 bp), cocktail-2 (VZV: 275 bp and EBV: 182 bp) and mix 3 (JCV: 199 bp), respectively. The alpha-tubulin amplicon was 527 bp.

Cloning

PCR products were produced with cycling conditions specific to the EBV and JCV primers, with a final extension of 10 min at 72°C. The PCR products were verified by agarose gel electrophoresis to ensure that single, discrete bands were obtained. The TOPO TA cloning kit (Invitrogen, Carlsbad, CA, USA) was used to clone the PCR product as per the manufacturer′s instructions. The copy number of the cloned plasmids was calculated using the formula: weight of PCR fragment (in g/ml)/(660 g/mol X the number of base pairs of the PCR fragment) X (6.023 x10 ²³ ) = the number of genome copies/μl. The DNA concentration of the plasmid was determined by measuring the optical density at 260 nm with a spectrophotometer (μQuant; Biotek Instruments Inc., Winooski, VT, USA).

The cloned plasmids were serially diluted 10-fold in TE buffer (pH 8.0) within the concentration range of 10 ⁹ to 10 plasmid copies/μl. The dilutions were stored at -20°C until use. The copy numbers were estimated using PCR. Appropriate negative controls were used. Amplification shown in the highest dilution (least concentration) in at least two replicates of the triplicates tested at each dilution was taken as the lower limit of detection.

Real-time multiplex/uniplex polymerase chain reaction

The real-time multiplex and uniplex PCRs were standardised using the QuantiTect multiplex NoROX kit. The QuantiTect multiplex NoROX kit (Qiagen) eliminates the need for optimization of the concentrations of primers, Mg2 ⁺ and Taq DNA polymerase. This kit is specially pre-optimized for quantitative, real-time multiplex PCR. The pre-optimized master mix ensures that the PCR products in a multiplex reaction are amplified with the same efficiency and sensitivity as the PCR products in the corresponding single-amplification reactions.

Real-time multiplex/uniplex PCRs were standardised for HSV-1, HSV-2, VZV, EBV, CMV and JCV. The first real-time multiplex PCR mix contained primers for HSV-1, CMV and EBV. The second real-time PCR mix contained primers for VZV. The third real-time PCR mix contained JCV primers while the fourth uniplex real-time PCR contained HSV-2 primers. The primer sequences for the PCRs for HSV-1, HSV-2, VZV, EBV, CMV and JCV viruses are shown below [Table - 2]. The primers used for the amplification were custom synthesized (Hysel India Pvt. Ltd.).

Amplification by real-time multiplex/uniplex polymerase chain reaction

The real-time multiplex/uniplex PCR format was used to amplify CMV, HSV-1, EBV, VZV, JCV and HSV-2. ^[11]

The amplification by real-time multiplex/uniplex PCR was performed using 10 μl of the extract in a 25 μl volume containing 12.5 μl of the QuantiTect multiplex PCR NoROX master mix (Qiagen), with each forward and reverse primer at a concentration of 300 nM and each probe (labelled with a fluorochrome dye of any one of the seven different dyes appropriate for three channels FAM, ROX, Cy5) at a concentration of 200 nM. The uniplex real-time PCR was also performed in the same format as the multiplex real-time PCR. The reaction mixtures were amplified using the following thermal cycling conditions: an initial denaturation and polymerase activation step for 15 min at 95°C followed by 50 cycles of 94°C for 45 s and 60°C for 75 s. Amplification and detection were performed on a real-time PCR system (Rotor-Gene 3000/6000; Corbett Research, Perth, Australia). The sample identification number (ID) was recorded for every run in the Edit sample window of the onboard software of Rotor Gene 3000/6000. A known positive control was included in each run and sterile Milli Q water was included after every third sample in the sample DNA extraction step and used in the PCR testing (extraction of water as negative control was to check for cross-contamination and integrity of the PCR reagents).

DNA was extracted from the characterised virus strains for specific or heterologous viruses (HSV-1, HSV-2, CMV and VZV). Cloned plasmids were used for EBV and JCV. The virus extracts were pooled to create templates for the different primer cocktails or used individually in the PCR reaction. The PCR runs were validated only if the controls were satisfactory. Based on the published literature for the cut-off for real-time PCR endpoints, amplification within the 40 ^th cycle with a cut-off fluorescence intensity of 0.05 or more with a typical sigmoid amplification curve was considered positive. ^[12],[13]

The validation of the real-time multiplex/uniplex PCR was carried out by establishing the lower limit of detection using either cloned plasmids or specific viruses for HSV-1, HSV-2, VZV, EBV, CMV and JCV. The specificity of the primers was established by testing with heterologous plasmids or viruses. A second target region LMP2 for EBV real-time PCR was chosen based on the published literature. ^[14]

Sequencing of EBV

Sequencing primers were designed to amplify the region spanning the putative real-time PCR target region. The sequencing PCR products for EBV (188 bp) were larger than the respective real-time products. Sequencing PCR for the BNTp143 was performed as a single-round PCR. The amplification was performed using 10 μl of DNA in a 50 μl reaction mix containing 25 μl of the HotStarTaq Mastermix and 50 picomoles of each primer (Hysel India Pvt. Ltd.). Amplification was performed using the following protocol: 95°C for 15 min, 94°C for 30 s, 54°C for 30 s, 72°C for 1 min for 40 cycles followed by 72°C for 7 min. Amplification of these respective regions of the viral target genes produced a 188 bp product for EBV, which was detected using agarose gel electrophoresis. Fifteen microlitres of the product was analysed on a 2% agarose gel containing 0.5 μg/ml of ethidium bromide at 120 V for 50 min. Phi X 174 DNA-Hae III digest (Bangalore Genei) was used as a molecular weight marker. The188 bp product for EBV was visualized using a UV transilluminator (Mighty Bright; Hoeffer Scientific Instruments). The size of the amplified product was determined by a gel documentation system (Gel Doc; BioRad) using the Quantity One software version 4.1.1 (BioRad). Pre-cycling sequencing clean-up and post-cycle sequencing clean-up were carried out.

Statistical analysis

All the data points were entered in an Excel spread sheet. Statistical analysis was performed using the Epi Info (version 6.04b) software. The Chi square test for comparison of proportions was performed.

Results

The lower limit of detection for conventional PCR is shown in [Table - 3] and the lower limit of detection for real-time PCR is shown in [Table - 4].

The comparison of real-time multiplex PCR data with conventional multiplex PCR data on 158 samples tested for CMV, HSV-1, HSV-2, EBV, VZV and JCV is provided in [Table - 5].

The analysis has been carried out for the six viruses (CMV, HSV-1, HSV-2, VZV, EBV and JCV). PCRs for HSV-2 and JCV were in the uniplex format. Of these 158 samples tested, 11 were negative for alpha-tubulin. These 11 samples were removed from the database as they probably had PCR inhibitors in the CSF. The clinical characteristics of these 11 samples included seven test and four control samples. Hence, PCR results were analysed for 147 samples. Of the 147 samples, 126 were test CSF samples from acute and chronic CNS disease of suspected viral aetiology and 21 were from the control group. The overall detection rate was significantly higher for real-time PCR compared with the conventional multiplex PCR (P < 0.0001). In the entire study population, the major viruses detected using real-time PCR were EBV [50/147 (34%)], HSV-2 [16/147 (10.9%)] and VZV [10/147, (6.8%)]. No co-infections were observed in the conventional PCR in the two study groups. Real-time PCR showed 15 co-infections, both in the study and in the control groups.

Additionally, 151 samples were tested by real-time PCR alone of which 115 were test CSF samples from acute and chronic CNS disease of suspected viral aetiology and 36 were control samples. Overall, of the 298 samples tested, the EBV real-time PCR was positive in 124 (41.6%) samples. Because this was a very high positivity rate for EBV, we checked the reliability of this finding using another set of primers and probe targeting the latent membrane protein 2 (LMP2) region of EBV (i.e., a second target region) in a real-time PCR format. All the samples were negative for the second target region by real-time PCR.

Sequencing was carried out to verify the supernumerary positives for EBV. Among the 298 samples tested, 124 CSF samples were positive for EBV in the first target region (EBV-BNT p143) by real-time PCR. These samples were then tested using real-time PCR with another set of primers and probe targeting the LMP2 region of the EBV genome. All samples that were positive for the first target region turned out to be negative for the second target region. Of the samples that showed contradictory results between the first and the second target regions used for real-time PCR, 30 samples that had relatively high concentrations of virus (detected in earlier cycles) were chosen for sequencing, targeting the BNT p143 region for EBV. EBV was successfully amplified and sequenced in five (16.7%) samples of the 30 samples screened. The EBV amplification rate was low for the sequencing experiment, which is probably a reflection of the higher sensitivity of detection by real-time PCR. The copy numbers of EBV could have been lower than the detection threshold of a single-step conventional PCR. Phylogenetic analysis of the five EBV PCR positives was carried out using the maximum composite likelihood algorithm in MEGA4.

Discussion

We compared the non-nested multiplex conventional PCR with the real-time PCR for the detection of neurotropic DNA viruses in the CSF of patients with neurological disease. Several other studies from the West have compared nested and real-time PCR in a uniplex format. These studies indicate that the sensitivity of nested PCR and real-time PCR are in the same range, but the latter had a significant advantage in time saving and avoidance of the risk of cross-contamination, unlike with the nested step in conventional PCR. ^[15],[16]

Our study compared real-time multiplex/uniplex PCR data with non-nested multiplex conventional PCR performed on 147 samples (patient group = 126, control group = 21). Overall, in the entire set of samples, the real-time PCR yielded 88 (59.9%) positives and conventional PCR had six (4.1%) positives. The detection rate was significantly higher for real-time PCR compared with multiplex conventional PCR (P < 0.0001).

The non-nested multiplex conventional PCR may not be sensitive enough for clinical use as indicated by the current literature and the findings of this study. Improving the multiplex conventional PCR was difficult in our hands using the chosen target-specific primers. In view of the complex nature of optimising the multiplex nested format to achieve optimal amplification conditions, we used a non-nested multiplex format. Hence, in order to improve the rate of detection of viruses, we standardised a real-time PCR for use on CSF samples from patients with neurological disorders.

Reproducibility of the assays used in our study, especially where supernumerary positives were detected, was established by retesting samples from among those positive for EBV. A selected set of 30 samples was retested by the respective real-time PCR. All results were reproducible on repeat testing, validating the real-time PCR findings for EBV. Amplification curves, fluorescence intensity and the threshold cycles (Ct) were within the acceptable range on retesting for all the 30 samples. In addition to the 147 CSF samples tested by both assays, 151 samples were tested by real-time PCR (data not shown). Using real-time PCR, EBV supernumerary positives were tested with another set of primers and probes targeting another region (LMP2) of the EBV genome. All EBV positives for the first target region turned out to be negative by real-time PCR in the second target region-specific assays. Cumulatively, in this group of 298 samples, 30 samples were subjected to sequencing PCR for EBV of the EBV positives which gave contradictory results in real-time PCR. EBV was successfully amplified in five (16.7%) samples of the 30 screened.

Four of the five patient strains and six GenBank sequences showed a high degree of sequence homology. Previous studies have used Dot Blot hybridization for confirmation of conventional PCR findings. We chose to use a sequencing approach to cross-check on real-time PCR findings as a cost-effective alternate approach. Moreover, it gave us the advantage of direct comparison with the GenBank sequences of EBV. The EBV amplification rate was low for the sequencing experiments, which is probably a reflection of the higher sensitivity of detection by real-time PCR. Also, the number of copies of the EBV genome could have been lower than the detection threshold of a single-step conventional PCR used for sequencing. Hence, the EBV first-target primers could be used for real-time multiplex/uniplex PCR to avoid the problems with assays that we encountered.

Our data on virus detection is qualitative in nature and has given valuable information on the significant presence of DNA neurotropic viruses in various CNS infections where there was a clear clinical indication of viral aetiology. The issue of this supernumerary detection of EBV has to be carefully reviewed. None of the herpesvirus DNA-positive CSF samples we had tested showed contamination with RBC (as indicated previously); thus, we believe that the viruses were indeed in the CNS compartment. It is difficult to ascribe an aetiopathological significance to the presence of more than one virus. It is possible that the reactivation of these viruses would be the reason for the DNA demonstration in the CSF. The reactivation may be an outcome of the other underlying clinical conditions in this group other than the viral aetiologies of interest to our study.

It is our contention that the role of EBV in CNS disease among Indians has to be carefully evaluated. In the entire study population that was examined using real-time PCR, the major viruses detected were EBV (34%), HSV-2 (10.8%) and VZV (6.8%). Currently, the literature from the West has many reports of the development of quantitative assays for DNA neurotropic viruses, especially for members of the Herpesviridae.^[17],[18] Although not explicitly suggested in the recent European Federation of Neurological Society (EFNS) guidelines, ^[19] the present trend is to carry out quantitative viral load assays for Herpesviridae to examine the neurological disease association (for more definitive evidence) and to monitor therapy.

In our experience, the real-time multiplex/uniplex PCR detects a wide array of viruses with broad dynamic range, reproducibility, good lower limit of detection (sensitivity) and specificity when tested with heterologous viruses. We recommend that real-time multiplex/uniplex PCR assay be used for the detection of neurotropic viruses in CSF. However, quantitation experiments (viral load) and long-term follow-up of patients need to be established to understand CNS disease association with PCR findings in future prospective studies. Such studies will enable a clearer understanding of the role of DNA viruses in acute and chronic neurological illnesses seen in India.

Acknowledgement

research was funded by the Department of Biotechnology entitled "Molecular detection of neurotropic DNA viruses in important neurological disease seen in a tertiary hospital in India (South)". Ref no: BT/PR7530/MED/14/1021/2006.

References

Copyright 2011 - Indian Journal of Medical Microbiology

The following images related to this document are available:

Photo images