Utility of real-time Taqman PCR for antemortem and postmortem diagnosis of human rabies
Abstract
Rabies, a fatal zoonotic viral encephalitis remains a neglected disease in India despite a high disease burden. Laboratory confirmation is essential, especially in patients with paralytic rabies who pose a diagnostic dilemma. However, conventional tests for diagnosis of rabies have several limitations. In the present study the utility of a real-time TaqMan PCR assay was evaluated for antemortem/postmortem diagnosis of rabies. Human clinical samples received for antemortem rabies diagnosis (CSF, saliva, nuchal skin biopsy, serum), and samples obtained postmortem from laboratory confirmed rabies in humans (brain tissue, CSF, serum) and animals (brain tissue) were included in the study. All CSF and sera were tested for rabies viral neutralizing antibodies (RVNA) by rapid fluorescent focus inhibition test (RFFIT) and all samples (except sera) were processed for detection of rabies viral RNA by real-time TaqMan PCR. All the 29 (100%) brain tissues from confirmed cases of human and animal rabies, and 11/14 (78.5%) CSF samples obtained postmortem from confirmed human rabies cases were positive by real-time TaqMan PCR. Rabies viral RNA was detected in 5/11 (45.4%) CSF samples, 6/10 (60%) nuchal skin biopsies, and 6/7 (85.7%) saliva samples received for antemortem diagnosis. Real-time TaqMan PCR alone could achieve antemortem rabies diagnosis in 11/13 (84.6%) cases; combined with RVNA detection in CSF antemortem rabies diagnosis could be achieved in all 13 (100%) cases. Real-time TaqMan PCR should be made available widely as an adjunctive test for diagnosis of human rabies in high disease burden countries like India. J. Med. Virol. 86: 1804–1812, 2014. © 2013 Wiley Periodicals, Inc.
INTRODUCTION
Despite a high disease burden, rabies, a fatal zoonotic viral encephalitis still remains one of the neglected diseases in India. With an estimated 20,000 deaths annually, India accounts for almost 36% of human deaths due to rabies world-wide [Sudarshan et al., 2007]. All of these deaths are due to infection with the Rabies virus (RABV), the causative agent of classical rabies (Order Mononegavirales, Family Rhabdoviridae, genus Lyssavirus, Species Rabies virus).
Two distinct forms of rabies–furious and paralytic- are recognized in humans. Diagnosis of the furious (encephalitic) form rarely poses diagnostic difficulties except in some instances wherein characteristic aerophobia or hydrophobia are lacking. In clinical practice, the paralytic or atypical forms constitute about 20% of human rabies cases and pose a diagnostic dilemma as they are clinically indistinguishable from Guillain–Barre syndrome (GBS) and neuroparalytic complications due to Semple-type antirabies vaccine which is still being used in few countries like Pakistan and Myanmar [Udawat et al., 2001; Gadre et al., 2010]. Early diagnosis can obviate the need for unnecessary treatment and medical tests [Coertse et al., 2010] and help in prognostication, institution of appropriate infection control measures and timely administration of pre or post-exposure prophylactic vaccination to family members of the patient and the treating medical and nursing staff [Gadre et al., 2010]. The reported survival of a young girl who developed rabies following a bat-bite using the “Milwaukee Protocol” [Willoughby et al., 2005] revived interest in the medical community to attempt therapeutic approaches for this otherwise fatal disease. Hence antemortem laboratory confirmation of the diagnosis has assumed greater significance.
Currently the WHO recommended “Gold Standard” for detection of rabies antigen is the fluorescent antibody test (FAT) on impression smears of brain tissue obtained postmortem [Dean et al., 1996]. FAT, though almost 100% sensitive for postmortem diagnosis on brain tissue does not produce reliable results on decomposed/autolysed tissues. Further, brain tissues are not available readily in all cases for testing and FAT cannot be employed in biological fluids like CSF, saliva, urine etc., precluding its use in antemortem diagnosis. Alternate standard conventional methods like rabies tissue culture infection test (RTCIT) and mouse inoculation test (MIT) require a day to several weeks for diagnosis. FAT on skin biopsy and corneal smears has been reported to have a low sensitivity [Madhusudana, 2011].
Currently, several molecular assays have been evaluated as an adjunct to conventional tests for rabies diagnosis and include conventional gel-based Reverse Transcriptase PCR (RT-PCR) assays with nested/hemi-nested protocols for detection of rabies viral RNA on clinical samples [Crepin et al., 1998; Nadin-Davis, 1998; Macedo et al., 2006; Dacheux et al., 2008; Coertse et al., 2010; Biswal et al., 2012]. A major drawback of these assays, however, is the risk of cross-contamination, which precludes their routine use for diagnosis of human and animal rabies. Real-time PCR methods which do not require post-amplification processing significantly reduce frequency of cross-contamination due the “closed-tube” nature of the assay and also allow quantification of the viral genome copies. Real-time PCR assay employing SYBR Green dye for detection of rabies viral RNA appears promising [Nagaraj et al., 2006; Hayman et al., 2011] but requires extreme care to ensure specificity of results [Nadin-Davis et al., 2009]. Real-time PCR assays using the TaqMan fluorogenic probes, however, ensure a high specificity because of the intrinsic hybridization reaction [Hughes et al., 2004; Wakeley et al., 2005; Wacharapluesadee et al., 2008; Nadin-Davis et al., 2009], have a wide range of detection and are 10–1,000 times more sensitive than traditional nested RT-PCR [Wakeley et al., 2005; Nadin-Davis et al., 2009]. However, very few studies assessing utility of these molecular techniques in antemortem diagnosis of rabies have been reported from countries with a high burden of human and animal rabies, such as India.
The National Institute of Mental Health and Neurosciences (NIMHANS) is a tertiary hospital catering exclusively to patients with neurological, neurosurgical and psychiatric disorders. Several clinical samples (antemortem and postmortem) from suspected cases of rabies are received at the Neurovirology laboratory both from our hospital and from several centers across the country for confirmation of diagnosis. As none of the presently available conventional tests for antemortem diagnosis of rabies is 100% sensitive or specific there is a need to explore utility of these newer molecular techniques, as an adjunctive test for diagnosis. A real-time PCR employing the TaqMan probe chemistry using a previously published set of primers and probe was evaluated in our laboratory to determine diagnostic efficacy and to improve turn-around time for antemortem/postmortem diagnosis of rabies.
MATERIALS AND METHODS
Ethics Statement
The samples from the Human Brain Bank used in this study were collected postmortem following written, informed consent from next of kin of the deceased, and their usage for research was approved by the NIMHANS Institutional Ethics Committee (Institutional Review Board). All other human samples (antemortem and postmortem) and animal samples (postmortem) included in this study were received at the Neurovirology laboratory for diagnostic confirmation of rabies in clinically suspected cases. None of the samples were obtained from infected patients or animals specifically for this study. All human samples used in the study were anonymized.
Human Clinical Samples
Postmortem (archival) clinical samples
Brain tissue (fresh, frozen tissues preserved at −86°C), CSF and serum samples from 18 autopsy confirmed cases of rabies viral encephalitis were sourced from the Human Brain Tissue Repository (Human Brain Bank) at NIMHANS, Bangalore, India (Tables I–III).
| Patient code/year of collection | Age in years/sex | Type of rabies | Incubation period | Site of animal bite | Vaccination | FAT brain tissue (n = 10) | Real-time TaqMan PCR | RVNA titres by RFFIT (IU/ml) | ||
|---|---|---|---|---|---|---|---|---|---|---|
| Brain (n = 12) | CSF (n = 14) | Serum (n = 12) | CSF (n = 13) | |||||||
| P1/1997 | 35/F | Paralytic | 3 months | Right leg | Nil | ND | ND | Pos | 0.5 | 0.1 |
| P2/2000 | 60/M | Paralytic | NA | NA | Nil | ND | ND | Pos | 0.5 | <0.1 |
| P3/2000 | 7/M | Paralytic | 20 days | Leg | 14 doses Semple | ND | ND | Pos | 3.5 | 15 |
| P4/2002 | 11/M | Paralytic | NA | NA | Nil | Pos | Pos | Neg | 3.5 | 0.1 |
| P5/2003 | 35/M | Paralytic | 2 years | Left leg | Nil | Pos | Pos | Neg | 3.5 | 15 |
| P6/2003 | 30/M | Paralytic | 2 months | Right leg | 1 dose Semple | Pos | Pos | Pos | 0.2 | <0.1 |
| P7/2004 | 40/M | Paralytic | 1 year | Right knee | 7 doses Semple | Pos | Pos | Pos | 3.5 | 0.2 |
| P8/2004 | 70/M | Paralytic | 2 ½ months | Right leg | Nil | Pos | Pos | Pos | ND | ND |
| P9/2004 | 7/F | Paralytic | NA | NA | 7 doses Semple | ND | ND | Pos | 2 | 0.5 |
| P10/2006 | 48/M | Paralytic | NA | NA | NA | ND | ND | Pos | 0.2 | <0.1 |
| P11/2008 | 28/M | Encephalitic | 20 days | Right thumb | 7 doses ? Cell culture vaccine | ND | Pos | Neg | 7.5 | 61 |
| P12/2011 | 35/F | Paralytic | NA | NA | Nil | Pos | Pos | Pos | 0.2 | <0.1 |
| P13/2011 | 20/M | Paralytic | NA | NA | Nil | ND | Pos | ND | ND | ND |
| P14/2008 | 46/F | Paralytic | NA | NA | Nil | Pos | Pos | Pos | 0.5 | <0.1 |
| P15/1999 | 33/M | Paralytic | 2 years | NA | NA | ND | ND | Pos | ND | <0.1 |
| P16/2008 | 25/M | Encephalitic | 1 ½ months | NA | NA | Pos | Pos | ND | ND | ND |
| P17/2008 | 30/M | Paralytic | NA | NA | NA | Pos | Pos | ND | ND | ND |
| P18/2011 | 25/M | Encephalitic | 13 days | Face | 3 doses Cell culture vaccine | Pos | Pos | ND | ND | ND |
| Total Samples Positive (%) | 10 (100%) | 12 (100%) | 11 (78.5%) | 12 (100%) | 7 (53.8%) | |||||
- RVNA, Rabies viral neutralizing antibodies; RFFIT, Rapid fluorescent focus inhibition test; FAT, Fluorescent antibody technique; NA, Data not available; ND, Test not done; Pos, Positive; Neg, Negative; Nil, No vaccine received; M, Male; F, Female.
- * Brain tissues of all 18 patients were positive for Negri bodies and rabies viral antigen by immunohistochemistry.
| Sample | Number tested | Positive by real-time TaqMan PCR |
|---|---|---|
| Rabies confirmed (by FAT and/or Histopathology)a | ||
| Human | 12 | 12 (100%) |
| Canine | 14b | 14 (100%) |
| Bovine | 3 | 3 (100%) |
| TOTAL | 29 | 29 (100%) |
| Clinically suspected rabies (GBS)c | ||
| Human | 6 | 0 |
| Controls (Head injury cases)c | ||
| Human | 14 | 0 |
- GBS, Guillain–Barre Syndrome.
- a Laboratory confirmation of Rabies was done by FAT and/or Negri bodies/immunohistochemistry for all human brain tissues and only by FAT for all animal brain tissues.
- b However, 3/14 canine brains were decomposed and FAT could not be performed on them.
- c Negative for Rabies by FAT and/or Negri bodies/immunohistochemistry on Brain tissues.
| Sample | RFFIT no. positive/no tested | Real-time TaqMan PCR no. positive/no tested |
|---|---|---|
| Laboratory confirmed Rabiesa | ||
| CSF | 7/13 (53.8%) | 11/14 (78.5%) |
| SERUM | 12/12 (100%) | NA |
| Clinically suspected Rabies (GBS)b | ||
| CSF | 0/5 | 0/5 |
| SERUM | 0/5 | 0/5 |
| Controls (Head injury cases)b | ||
| CSF | 0/5 | 0/5 |
| SERUM | 0/5 | 0/5 |
- GBS, Guillain–Barre Syndrome; RFFIT, Rapid Fluorescent Focus Inhibition Test; NA, not applicable.
- a Positive for Rabies.
- b Negative for Rabies (by FAT and/or Negri bodies/immunohistochemistry on Brain tissues).
Brain tissue obtained at autopsy (n = 12), with serum samples (n = 12) and CSF obtained by postmortem lumbar puncture (n = 14), from patients who succumbed to rabies (confirmed by FAT and/or demonstration of Negri bodies/immunohistochemical confirmation of rabies viral antigen in tissues) between January 1997 and December 2011 were included in this study.
Clinical samples from patients who had died of GBS, wherein rabies encephalitis was clinically suspected but confirmed to be negative for rabies [frontal brain tissue (n = 6), CSF and serum samples (n = 5, each)] were included as controls. Patients who succumbed to head-injury following road traffic accidents [frontal brain tissue (n = 14), CSF and serum samples (n = 5, each)] and confirmed to be negative for rabies were included as controls.
Antemortem clinical samples
Clinical samples received at the Neurovirology laboratory from October 2011 to November 2012 from 13 clinically suspected human rabies cases for antemortem laboratory confirmation [CSF (n = 11), serum (n = 8), nuchal skin biopsies (n = 10), saliva samples (n = 7)] were also included in the study (Table IV).
| Patient code | RVNA titre by RFFIT(IU/ml) | Real-time TaqMan PCR | ||||
|---|---|---|---|---|---|---|
| Serum (n = 8) | CSF (n = 11) | CSF (n = 11) | Nuchal Skin (n = 10) | Saliva (n = 7) | Brain tissue obtained postmortem (n = 2) | |
| A1 | NR | <0.1 | Positive | NR | NR | Positive |
| A2 | <0.1 | NR | NR | Positive | NR | Positive |
| A3 | <0.1 | <0.1 | Positive | NR | NR | NR |
| A4 | <0.1 | <0.1 | Positive | Positive | Positive | NR |
| A5 | NR | <0.1 | Positive | Negative | Positive | NR |
| A6 | 0.5 | 15 | Negative | Negative | NR | NR |
| A7 | 0.5 | <0.1 | Negative | Positive | Positive | NR |
| A8 | <0.1 | <0.1 | Negative | Negative | Positive | NR |
| A9 | 15 | 7.5 | Negative | Positive | NR | NR |
| A10 | NR | <0.1 | Negative | Positive | NR | NR |
| A11 | NR | 30 | Negative | NR | Negative | NR |
| A12 | 30 | 0.1 | Positive | Positive | Positive | NR |
| A13 | NR | NR | NR | Negative | Positive | NR |
| Total samples positive | 4 (50%) | 4 (36.3%) | 5 (45.4%) | 6 (60%) | 6 (85.7%) | 2 (100%) |
- NR, sample not received; RVNA, Rabies viral neutralizing antibodies; RFFIT, rapid fluorescent focus inhibition test.
Animal samples (Postmortem)
Brain tissues obtained at necropsy from 17 cases of suspected rabies in animals (3 bovine and 14 canine) received at our laboratory from January 2009 to May 2012 (confirmed by FAT, except in 3 canine brain tissues which were decomposed on arrival at the laboratory) were also included in this study (Table II).
Histopathology and immunohistochemistry
Brain tissues obtained postmortem were sampled after 4 weeks of formalin fixation. Representative sections from various neuroanatomical areas were sampled from hippocampus, frontal, temporal and occipital cortices, cerebellum, and brain stem and subjected to routine processing for paraffin embedding. Serial sections were stained with hematoxylin-eosin and immunohistochemistry by indirect immunoperoxidase method using polyclonal antibodies to rabies viral nucleocapsid was used to localize the rabies viral antigen in brain tissues as described earlier [Suja et al., 2009].
Fluorescent antibody technique (FAT)
All human and animal brain tissues included in this study were subjected to standard FAT as advocated by WHO [Dean et al., 1996]. Briefly, smears were made from cut surfaces of fresh brain tissues, fixed in cold acetone for 2 hr, air dried and treated with anti-rabies polyclonal antibodies conjugated with FITC (EMD Millipore Corporation, Temecula, CA) for 30 min at 37°C in an incubator in a humid chamber. Known rabies positive and negative brain smears were included as controls. The slides were examined under UV light using Nikon Eclipse fluorescence microscope.
Rapid fluorescent focus inhibition test (RFFIT)
All CSF and serum samples included in this study were tested for rabies virus neutralizing antibody (RVNA) titers in accordance with WHO recommended protocol with minor modifications [Smith et al., 1996]. BHK 21 cell line (ATCC CCL 10) and BHK 21 adapted CVS 11 strain of virus were used and the tests were performed in 96 well tissue culture plates. The antibody titers were expressed in International units (IU/ml) in comparison to an in house reference serum calibrated against 2nd International reference serum obtained from National Institute of Biological Standards, UK. The lower limit of detection of the assay was 0.1 IU/ml.
Real-time TaqMan PCR
Real-time TaqMan PCR using a set of primers and probe (RABVD1) targeting the nucleoprotein (N) gene described earlier by Nadin-Davis et al. [2009] (Forward Primer: ATGTAACACCYCTACAATG; Reverse Primer: GCMGGRTAYTTRTAYTCATA; Probe: FAM-CCGAYAAGATTGTATTYAARGTCAAKAATCAGGT-BHQ-1) [Nadin-Davis et al., 2009] was carried out on all the clinical samples (CSF, brain tissue, saliva, nuchal skin biopsy). Briefly, RNA extraction was performed according to manufacturer's instructions using commercial kits-QIAamp Viral RNA minikit (Qiagen, Hilden, Germany) for CSF and saliva samples, and RNeasy Minikit (Qiagen) for brain and skin biopsy samples. The one-step real-time TaqMan PCR assay was performed in a 25 µl reaction volume using the Superscript III Platinum one step qRT-PCR kit (Invitrogen, Carlsbad, CA), comprising of 12.5 µl of the 2X master mix, 1 µl enzyme mix, 5 µl of nuclease free water, and 0.5 µl each of the forward and reverse primers (20 pmol) and probe (10 pmol), and 5 µl of the RNA extracted from each sample. Amplification was carried out at 50°C for 15 min and 95°C for 2 min, followed by 40 cycles each of 95°C for 15 sec and 50°C for 1 min. Each PCR run included positive control (RNA extracted from known positive clinical samples), non-template control (nuclease free water added instead of template RNA) and extraction control (extraction performed on nuclease free water along with clinical samples). Amplification, data acquisition, and analysis were carried out using an ABI 7500 real-time PCR Instrument and software. All samples were tested in duplicate and samples with a mean threshold cycle (Ct) value of ≤38 were considered positive. To confirm sample RNA integrity, all the clinical samples were also subjected to a one step real-time TaqMan PCR assay in a separate reaction, using primers and probe for detection of 18S ribosomal RNA (rRNA) described previously [Nakahata et al., 2006].
RESULTS
Postmortem Diagnosis of Rabies
Clinical and laboratory details of the 18 patients whose postmortem samples were used in this study are shown in Table I.
A total of 29 brain tissues from 12 confirmed cases of human rabies and from 17 cases of animal rabies were tested by real-time TaqMan PCR. All the 29 (100%) brain tissues were positive for rabies viral RNA (Table II).
Real-time TaqMan PCR assay could detect rabies viral RNA in 11/14 (78.5%) CSF samples from confirmed human rabies cases. RVNA could be detected in 7/13 (53.8%) CSF samples and all 12 (100%) serum samples by RFFIT (Table III).
None of the human clinical samples from the control groups were positive for rabies viral RNA by real-time PCR (Tables II and III).
Antemortem Diagnosis of Rabies
Human clinical samples from 13 clinically suspected cases of rabies received for antemortem diagnosis were tested by real-time TaqMan PCR for detection of viral RNA. Rabies viral RNA was detected in 5/11 (45.4%) CSF samples, 6/10 (60%) nuchal skin biopsies, and 6/7 (85.7%) saliva samples. RVNA could be detected in 4/11 (36.3%) CSF, and 4/8 (50%) serum samples by RFFIT. Atleast one clinical sample (CSF/skin/saliva) was positive by real-time TaqMan PCR in 11/13 (84.6%) patients; when combined with RVNA detection in CSF, antemortem rabies diagnosis was achieved in all 13 (100%) cases (Table IV).
The real-time TaqMan PCR for 18S rRNA gene (internal control) was positive in all clinical samples.
DISCUSSION
Postmortem Diagnosis of Rabies
Archival (postmortem) human clinical samples from 1997 to 2011 (15 years) obtained from the Human Brain Bank of our Institute and animal brain tissues were included in the study for two major reasons; (i) Low submission rates of human samples for antemortem diagnosis (ii) Comparison with “Gold Standard” available, since all these cases were confirmed by FAT and/or Negri bodies/immunohistochemistry for rabies viral antigen.
A total of 29 brain tissues from 12 cases of human rabies (confirmed by histopathology and/or FAT) and from 17 cases of animal rabies (confirmed by FAT, except in 3 cases since the brains were decomposed on arrival at the laboratory) were tested by real-time TaqMan PCR. All the 29 (100%) brain tissues were positive for rabies viral RNA by real-time TaqMan PCR (Table II). Rabies viral RNA was also detected in 11/14 (78.5%) CSF samples obtained postmortem from confirmed human rabies cases. None of the human clinical samples (brain tissue, CSF) from the control groups (Tables II and III) were positive for rabies viral RNA by real-time TaqMan PCR confirming the 100% specificity of the assay. Rabies viral neutralizing antibodies (RVNA) could be detected in 7/13 (53.8%) CSF samples and all 12 (100%) serum samples by RFFIT (Table III).
Antemortem Diagnosis of Rabies
Real-time TaqMan PCR for viral RNA was positive in 5/11 (45.4%) CSF samples, 6/10 (60%) nuchal skin biopsies, and 6/7 (85.7%) saliva samples. RVNA could be detected in 4/11 (36.3%) CSF and 4/8 (50%) serum samples by RFFIT (Table IV). All the 13 patients had symptoms of furious (encephalitic) or paralytic rabies and succumbed to the illness. However, postmortem confirmation by FAT on brain tissue could be done only on two patients. Thus real-time TaqMan PCR alone (atleast one clinical sample positive in each patient) could achieve antemortem rabies diagnosis in 11/13 (84.6%) cases; combined with RVNA detection in CSF antemortem rabies diagnosis could be achieved in all 13 (100%) cases.
An interesting observation in the present study was that rabies viral RNA was detected in 45.4% of the CSF obtained antemortem, whereas the yield increased to 78.5% in CSF samples obtained post-mortem following lumbar puncture. The technique of CSF collection postmortem by lumbar puncture is identical to the procedure followed antemortem, hence possibility of potential contamination by neural tissue appears remote. A plausible scientific explanation is that post-mortem CSF represents sample at the terminal stage of illness in rabies, when the viral load in the central nervous system is known to be relatively high, as compared to the initial stages of infection wherein the viral load is lower. MR imaging studies in human rabies cases have demonstrated MR signal intensity changes in the early stages of illness, but enhancement is seen only in terminal stages once patient becomes comatose suggesting breach in blood brain barrier terminally [Laothamatas et al., 2003]. This could be responsible for the higher frequency of presence of viral RNA in CSF and hence high rates of positivity of PCR in postmortem samples.
Presently, confirmation of rabies with an almost 100% sensitivity can be achieved only by testing postmortem brain tissue by FAT. The results of the present study indicate that real-time TaqMan PCR on brain tissue is also 100% sensitive for detection of rabies viral RNA and is especially valuable in cases where brain tissue obtained postmortem by transnasal biopsy is scanty or when the tissue is decomposed or autolysed and hence false positive/negative FAT results may be obtained. However, since rabies is not a notifiable disease in many developing countries including India, postmortem brain biopsy or autopsy is rarely proposed by physicians or permitted by family members for religious and other reasons. Family members of several patients who receive a clinical diagnosis of rabies opt to take the patient home for terminal care, to avoid the high costs of hospital care, further hampering efforts at confirmation [Dacheux et al., 2008]. Antemortem confirmation of rabies in samples other than brain tissue can alleviate the requirement for brain biopsies/autopsy and the accompanying logistics and safety procedures [Coertse et al., 2010]. In cases where a diagnosis of rabies is not confirmed antemortem, CSF obtained by postmortem lumbar puncture can be tested by real-time TaqMan PCR, since it approached a sensitivity of almost 80% in our study. A postmortem nuchal skin biospsy can also be tested since 60% of the skin samples received antemortem were positive for rabies viral RNA in the present study.
As indicated in this study as well other studies [Wacharapluesadee and Hemachudha, 2002; Nadin-Davis et al., 2009; Coertse et al., 2010; Hunter et al., 2010] real-time TaqMan PCR can be performed on a range of biological samples like CSF, saliva, tears, urine, and skin biopsy for antemortem diagnosis of human rabies. The assay can be completed in less than 2 hr providing rapid results.
Molecular assays have been found beneficial for diagnosis of rabies in decomposed and archival samples [David et al., 2002; Johnson and Fooks, 2005; Biswal et al., 2007; Araújo et al., 2008] and have an important role in retrospective diagnosis and epidemiological studies. Amplicons generated using conventional PCR techniques can also be sequenced for phylogenetic analysis. In the present study, real-time TaqMan PCR could detect viral RNA in archival samples ranging from 1 to 15 years and in 3 canine brains that were decomposed on arrival in the laboratory and hence could not be tested by FAT. Since rabies infection can be acquired through organ transplants [Hellenbrand et al., 2005; Srinivasan et al., 2005] molecular assays can also be of use to test donors who are at risk of rabies. Real-time TaqMan PCR can also be used for quantification of viral RNA to assess the viral load, disease progression, and efficacy of experimental therapeutic approaches [Nadin-Davis et al., 2009; Maier et al., 2010], and may be valuable as more aggressive treatment options for rabies are explored in future.
The RABVD1 set of primer pair reported by Nadin-Davis et al. [2009] was selected because it targets the highly conserved RABV sequence around the start of the N gene ORF, a superior target for real-time PCR based diagnostic applications. This set of primers/probe earlier reported by Wakeley et al. [2005], have been suitably modified by Nadin-Davis et al. [2009] by incorporating additional redundancy at three positions in the reverse primer and, by extending the length and adding degeneracy to the sequence at four positions in the probe to take into account the most common variations observed at these target sites. The amplicon generated is small (∼110 bp) allowing for increased sensitivity. Moreover, this real-time TaqMan PCR assay using RABVD1 set was evaluated for detection of rabies viral RNA using a collection of 203 isolates representative of the world-wide diversity of RABV and was found to be the most sensitive, robust, and broadly reactive when compared with two other sets of primer-probes evaluated in the study [Nadin-Davis et al., 2009].
Since the 18S rRNA assay has been reported to be more sensitive, especially for samples like CSF, which may not contain enough cellular material, it was used to test the sample integrity and for the verification of the extraction procedure for all the clinical samples [Nadin-Davis et al., 2009]. All the human and animal samples used in this study, were positive by the 18S rRNA assay confirming the sample integrity, especially of the archival samples and ruling out any false negative results for rabies. An earlier study using the same protocol and primers/probe for rRNA detection reported a positive result in tissue samples from several lyssavirus hosts (Human, Dog, Cat, Fox, Jackal, Bat, Rodent, etc.) reiterating its suitability as an internal control target for rabies diagnosis in humans and several other animal hosts [Coertse et al., 2010].
Antibody testing in serum and CSF is of limited value since serocoversion occurs late in the course of the disease [Schuller et al., 1979]. However, demonstration of neutralizing antibodies to rabies in the serum and CSF is significant, especially in non-vaccinated individuals and is a good diagnostic marker in patients with paralytic rabies, who generally have a longer survival. The presence of neutralizing antibodies in the serum/CSF of vaccinated individuals has to be interpreted with caution and a confirmatory diagnosis can be made only by demonstrating a >fourfold rise in titres between two samples drawn 7–10 days apart [Madhusudana and Sukumaran, 2008]. In the present study, RVNA were detected in 4/11 (36.3%) CSF samples received for antemortem diagnosis and 7/13 (53.8%) CSF samples obtained postmortem, which represents the terminal stage of illness. An inverse correlation was found between detection of RVNA and presence of viral RNA in CSF samples. Three (100%) CSF samples received for antemortem diagnosis and 2/3 (66.6%) CSF samples obtained postmortem which had high titres of RVNA were negative for rabies viral RNA. Similarly, all 5 (100%) CSF samples received for antemortem diagnosis and 7/10 (70%) CSF samples obtained postmortem which were positive for viral RNA by real-time PCR had very low or undetectable titres of RVNA (Tables I and IV). Hence whenever feasible, detection of RVNA in CSF should be done along with real-time TaqMan PCR on multiple samples for rabies diagnosis, since the immune status of the patient determines the ability to detect virus in antemortem samples. Antibody testing is also a useful tool to monitor the immune response within the central nervous system and thus a possible clearance of the RABV in patients who are treated with the “Milwaukee Protocol” or other intensive interventions in future [Hunter et al., 2010].
Despite several advantages like high specificity and a reduced risk of cross-contamination, it was reported that viral genetic heterogeneity may prove to be an impediment to the development of TaqMan probe based PCR since mismatches between the target and the probe can lead to false-negative results or decreased sensitivity [Hughes et al., 2004; Wakeley et al., 2005]. However, mismatches on primer and/or probe binding sites did not affect the amplification or detection in several other studies. [Wacharapluesadee et al., 2008; Coertse et al., 2010]. In the present study also real-time TaqMan PCR was 100% sensitive in detecting rabies viral RNA from brain tissues representative of human and animal rabies infection from several states in the country and spanning about 15 years. Besides, no other lyssaviruses other than RABV has been reported from India [Nagarajan et al., 2006]. Nevertheless it is prudent to maintain continual evaluation and vigilance for genetically divergent strains in future. In fact, this assay can serve as a surveillance tool to detect the emergence of any new RABV variants if discordant results are obtained between FAT and real-time TaqMan PCR [Wacharapluesadee et al., 2008].
Despite being a promising technology, it has to be emphasized that a negative result by any molecular detection method may not rule out a diagnosis of rabies. In the absence of brain tissue, multiple samples (CSF, skin, saliva, etc.) may have to be tested with real-time TaqMan PCR in combination with other conventional techniques to approach 100% sensitivity for rabies diagnosis. A limitation of the present study is lack of serial sampling and correlation with stage of illness, since most of the samples received for antemortem diagnosis were from various hospitals across the country. Serial sampling of saliva and skin by real-time PCR and CSF/serum for RVNA detection can increase the sensitivity of antemortem rabies diagnosis.
Another limitation of molecular diagnostic tests cited frequently is that they are not a practical option in resource restricted countries. However, the fact remains that even today only a handful of specialized laboratories with required infrastructure and trained manpower carry out diagnostic testing using conventional tests for human rabies in India. As Fooks et al. [2009] have aptly pointed out in their comprehensive review on emerging technologies for detection of rabies, that though financial and logistical barriers may prevent the routine use of molecular diagnostic assays, the cost/benefit ratio should still be measured.
In conclusion, despite a high disease burden, rabies still remains one of the most neglected diseases in India. The low level of commitment to rabies control is partly attributable to lack of accurate and extensive surveillance data to indicate the disease burden [Weyer and Blumberg, 2007]. Molecular assays like real-time TaqMan PCR are a promising technology for rabies diagnosis as evidenced in the present study and should be made available widely as an adjunctive test for antemortem and postmortem diagnosis of human rabies in high disease burden countries like India.
