"Diagnosis of Foot-and-Mouth disease virus by automated RT-PCR"
Appendix 25 Diagnosis of Foot-and-Mouth disease virus by automated RT-PCR Scott M. Reid, Nigel P. Ferris, Geoffrey H. Hutchings and Soren Alexandersen Institute for Animal Health, Pirbright Laboratory, Ash Road, Pirbright, Woking, Surrey GU24 0NF, United Kingdom Abstract Automated 5' nuclease probe-based (TaqMan®) reverse transcription polymerase chain reaction (RT-PCR) procedures using a MagNA Pure LC were evaluated for foot-and-mouth (FMD) virus diagnosis during the United Kingdom (UK) 2001 epidemic. Epithelial suspensions (ES), serum or whole blood, milk and oesophageal-pharyngeal fluid (“probang”) samples submitted to the OIE/FAO World Reference Laboratory for Foot-and-Mouth Disease (WRL for FMD), Pirbright, were tested as well as supernatant fluids following inoculation of cell cultures with ES. New programmes have been inserted into the software of the MagNA Pure LC to improve the extraction of nucleic acids from test samples and controls and to increase the speed, flexibility, capacity and reproducibility of the RT and PCR procedures without harming the assay sensitivity. Like the initial programmes for automation, the new programmes enabled definitive diagnostic results to be achieved on first passage cell culture supernatant fluids but a 96-well PCR assay can now be performed by 2 people within an extended working day. Our results indicate that our real-time automated RT-PCR could be recommended instead of virus isolation during an outbreak to accelerate FMD diagnosis. The positive-negative acceptance criteria for the testing of probangs by automated RT-PCR is under consideration. Preliminary results from experimentally infected animals show that the virus can be detected in probangs but different assay acceptance criteria to that based on the testing of ES and cell culture supernatant fluids will have to be used. 1. Introduction If laboratory investigations are to play a part in FMD diagnosis following the control policies introduced by the UK government during the 2001 epidemic then the time to perform tests will have to be considerably accelerated. In the WRL for FMD, Pirbright, an evaluation is being made of the application of automated RT-PCR technology for the rapid and accurate diagnosis of the disease. Automated RT-PCR procedures for the diagnosis of FMD virus using tissue epithelium, serum or whole blood, milk and probang samples were initially evaluated on samples submitted to the WRL for FMD during the UK 2001 epidemic and from animals experimentally infected with the FMD virus serotype O UK 2001. A primer/probe set designed for the intended detection of all 7 serotypes of FMD virus was used (Reid et al., 2002) and the results of this evaluation were directly compared to the routine diagnostic tests of ELISA and virus isolation in cell culture (Reid et al., 2001a; 2001b). Automated programmes increased the speed and capacity of the RT- PCR assays by enabling larger panels of clinical samples (together with positive and negative 203 controls) to be tested simultaneously by RT-PCR. In an extended working day, one operator could realistically obtain results from 64 test (and control) samples by performing 2 runs of a 32- well RT-PCR assay. The Pirbright Laboratory have further developed and evaluated automated RT-PCR procedures for FMD virus diagnosis by creating new MagNA Pure LC programmes in attempts to improve the efficiency of the extraction of total nucleic acid from test and control samples and to increase the speed, flexibility and reproducibility of the automated procedures for RT and PCR without affecting the sensitivity achieved by the initial programmes. 96-well automated assays would greatly increase the diagnostic capacity and consistency of RT-PCR and could easily be performed on a routine basis. The results from assays involving new automated programmes for the nucleic acid extraction, RT and PCR steps and their comparison with the routine diagnostic procedures of ELISA and virus isolation in cell culture are presented in this study. 2. Materials and methods 2.1. Sample preparation, ELISA and virus isolation ES and other suspensions of samples submitted for FMD virus diagnosis from the UK 2001 FMD epidemic were prepared, tested by ELISA and inoculated onto primary calf thyroid cell cultures (Ferris and Dawson, 1988). IB-RS-2 cells were also used for virus isolation on some sample submissions. ES was similarly prepared and tested from overseas sample submissions. Serum or whole blood ("blood") and probang samples from the epidemic were similarly inoculated onto primary calf thyroid cell cultures. Samples showing a cytopathic effect (CPE) were harvested and the FMD virus specificity of the supernatant fluids was confirmed by ELISA. Supernatant fluids of a selection of cell cultures not showing a recognisable CPE on first and second passage following inoculation with ES or other suspensions were collected. 2.2. Total nucleic acid extraction Prior to the extraction procedure, 0.2 ml of ES was added to 1 ml TRIzol® Reagent (Life Technologies, UK). Blood and probang samples were added to an equal volume of lysis/binding buffer (Roche, UK) and each sample mixed for 10-15 sec. Cell culture supernatant fluids were either added to an equal volume of lysis/binding buffer or 0.2 ml added to 1 ml of TRIzol® Reagent. Samples were placed in batches of 32 inside a MagNA Pure LC (Roche, UK) programmed to extract total nucleic acid to a final elution volume of 0.1 ml. A new total nucleic acid extraction programme was also used to provide a final elution volume of 0.05 ml and therefore achieve a greater concentration of extracted nucleic acid. 2.3. Reverse transcription The MagNA Pure LC was programmed to mix 6 µl of each nucleic acid with 9 µl RT mix in batches of 32 samples. For the 96-well RT-PCR assay (called the ‘fast RT-PCR protocol’), this RT process was carried out on 3 consecutive sets of 32 nucleic acids in a single 96-well plate and the RT of 96 samples was completed by placing the plate in a PTC-100TM thermal cycler (MJ Research, Inc.) and incubating successively at 48oC for 45 min, 95oC for 5 min and 20oC for at least 20 min. A faster RT programme was later created to achieve the same function but which could mix 32 samples of nucleic acid with RT mix in a shorter time period of around 5 min. The 204 slower RT programme was only used for 96-well assays (‘fast RT-PCR protocol’) and the newer faster one has mainly been used for 32-well assays. 2.3. PCR amplification by 5' nuclease probe-based reaction Redundant primers and a fluorogenic 5' nuclease probe were as described previously (Reid et al., 2001a; 2001b; 2002). RT products were added to the PCR mix by an automated process in which the MagNA Pure LC was programmed to add 7 µl cDNA from batches of either 32 or 96 samples (as in the 96-well ‘fast RT-PCR protocol’) to 18 µl PCR mix containing 0.9 pmol/µl each of the forward and reverse primers and 0.3 pmol/µl of probe. This was followed by PCR amplification in a GeneAmp® 5700 Sequence Detection System (Applied Biosystems) using the amplification programme described previously (Reid et al., 2001a; 2001b; 2002). A quantitative value was assigned to each PCR reaction and positive/negative cut-off criteria was applied which included a ‘borderline’ region. Samples falling within this region require retesting by the PCR to determine their status. 2.4. Reproducibility of automated RT-PCR The intra-assay reproducibility of the 96-well ‘fast RT-PCR protocol’ was monitored by incorporating FMD-positive and FMD-negative control samples within each set of 32 samples. Intra-assay reproducibility of the same automated RT-PCR procedure was also assessed by testing a single cell culture grown virus preparation in every well of the 96-well plate. Inter-assay reproducibility of this protocol was determined by testing a batch of 32 samples in more than one assay run. 3. Results 3.1. Automated 96-well ‘fast RT-PCR protocol’ The results of the 96-well ‘fast RT-PCR protocol’, ELISA and virus isolation on ES, other suspensions, blood and probang samples are summarised in Table 1 (section A). RT-PCR detected more positive sample submissions than ELISA and was positive on more ES than virus isolation. Blood samples positive by virus isolation were generally also positive by RT-PCR but only one probang was positive by virus isolation. This probang was borderline by the RT-PCR as were 2 other probangs submitted from other premises but which were negative by virus isolation. Table 2 (section A) compares the results of the ‘fast RT-PCR protocol’ with virus isolation on cell culture supernatant fluids inoculated with ES or other suspensions. Fifteen supernatant fluids produced a CPE on first passage cell culture and were positive by RT-PCR while 17 supernatant fluids which did not produce a CPE after first (and then second) passage (and therefore not tested by ELISA) were negative by RT-PCR. 3.2. 32-well RT-PCR protocol with alternative programmes for nucleic acid extraction and 205 reverse transcription The results from a 32-well RT-PCR using the alternative programme for extraction of total nucleic acid from test and control samples to a final elution volume of 0.05 ml followed by the more rapid programme for automated RT on ES and cell culture supernatant fluids are shown in Table 1 (section B) and Table 2 (section B) respectively. The RT-PCR had an equivalent sensitivity to virus isolation and a higher sensitivity than ELISA for detection of FMD virus in ES. Four supernatant fluids produced a CPE on first passage cell culture and were RT-PCR positive. Fifty one supernatant fluids which did not produce a CPE after first and second cell culture passage (not tested by ELISA) were negative by RT-PCR. 3.3. Reproducibility of automated RT-PCR The standard deviation was low when the cell culture grown virus preparation was tested in all 96 wells by the ‘fast RT-PCR protocol’ and the inter-assay reproducibility of this procedure was also good based on the re-testing of the batch of 32 samples and the re-testing of other selected samples (data not shown). 4. Discussion Automated 5' nuclease probe-based RT-PCR can provide FMD diagnostic results more rapidly than conventional RT-PCR or non-automated 5' nuclease probe-based RT-PCR methods but further development of the methodology has increased test capacity and consistency to achieve 96 results in a single plate by one person within 2 typical working days or by 2 people in around 12 hours. The sensitivity of this 96-well assay (‘fast RT-PCR protocol’) compared well with virus isolation (the so-called gold standard) for the testing of ES and blood while the more recently-developed 32-well RT-PCR protocol had an equivalent sensitivity to virus isolation on ES. Both RT-PCR methods detected FMD viral RNA in positive cell culture supernatant fluids on first passage (showing a recognisable CPE) and were also clearly negative on first passage cell cultures not showing a CPE and which did not produce a CPE on second passage so that definitive diagnostic results could be achieved by RT-PCR on first passage cell cultures alone. The described automated RT-PCR could be recommended instead of virus isolation during an FMD outbreak to achieve diagnostic results within a much smaller time-scale. More samples will be tested to confirm that passaging of samples in cell culture may not be necessary for the issue of laboratory results although it would still be useful to isolate samples in cell culture for confirmation of a first outbreak in a previously FMD-free country. The improved nucleic acid extraction and RT programmes used in the 32-well RT-PCR can be applied for testing batches of 96 (or 64) test/control samples in a single plate to provide diagnostic results in an even shorter time-scale than that of the 96-well ‘fast RT-PCR protocol’ (total assay time cut from 12 to 10-11 hours). This new 96-well RT-PCR assay will have to be evaluated on more ES and cell culture supernatant fluids but it has already achieved promising results on probangs taken from experimentally infected animals even though the positive/negative acceptance criteria has not been firmly established (S. M. Reid et al., unpublished results). This is likely to differ from that currently used for the testing of ES and cell culture supernatant fluids. Acknowledgements 206 The authors thank Sylvia Grierson (Veterinary Laboratories Agency, Addlestone) and Antoinette Cherubini (Central Veterinary Research Laboratory, Abbotstown, Dublin) for their help. This work was supported financially by the Department for Environment, Food & Rural Affairs (DEFRA), UK. References Reid, S. M., Ferris, N. P., Hutchings, G. H., Zhang, Z., Belsham, G. J., Alexandersen, S., 2002. Detection of all seven serotypes of foot-and-mouth disease virus by real-time, fluorogenic reverse transcription polymerase chain reaction assay. J. Virol. Methods 105, 67-80. Reid, S. M., Ferris, N. P., Hutchings, G. H., Zhang, Z., Belsham, G. J., Alexandersen, S., 2001a. Diagnosis of foot-and-mouth disease by real-time fluorogenic PCR assay. Vet. Record 149, 621-623. Reid, S. M., Ferris, N. P., Hutchings, G. H., Alexandersen, S., 2001b. Evaluation of automated RT-PCR systems to accelerate FMD diagnosis. Report of the Session of the Standing Technical Committee of the European Commission for the Control of Foot-and-Mouth Disease, Island of Moen, Denmark, 12-15 September, 2001. Rome: FAO 2001 Appendix 25, pp. 118-125. Ferris, N. P., Dawson, M., 1988. Routine application of enzyme-linked immunosorbent assay in comparison with complement fixation for the diagnosis of foot-and-mouth and swine vesicular diseases. Vet. Microbiol. 16, 201-209. 207 Table 1. Comparison of the 96-well ‘fast RT-PCR protocol’ (section A) and the new automated 32-well RT-PCR protocol (section B) with ELISA and virus isolation for the testing of epithelial suspensions (ES), other suspensions, blood and probang samples Test procedure Type of material and the number of samples positive (FMD virus), negative (NVDa) or borderline ES Bloodb Probangs Otherc FMD NVD Borderline FMD NVD Borderline FMD NVD Borderline FMD NVD Borderline A ELISA 104 104 8 0 3 0 Virus isolation 124 92 0 26 226 0 1 159 0 0 3 0 RT-PCR 134 80 2 27 220 5 0 157 3 0 3 0 B ELISA 13 17 0 Virus isolation 14 16 0 RT-PCR 14 15 1 a NVD, no virus detected. b Blood and probang samples are not tested directly by ELISA but inoculated onto primary calf thyroid cell culture. ELISA then used to test supernatant fluids from cell cultures showing a CPE. c Suspensions prepared from “bone marrow” samples. 208 Table 2. The results of the 96-well ‘fast RT-PCR protocol’ (section A) and the newer automated 32-well RT-PCR protocol (section B) on cell cultures inoculated with epithelial suspensions (ES) Test procedure Ratio of number of samples positive (FMD virus) or NVDa in passages of cell culture following inoculation with ES First passage (0-48 hr) Second passage (48-72 hr) Positive NVD Positive NVD A Virus isolation 15/15 0/17 NT b 0/17 RT-PCR 15/15 0/17 NT NT B Virus isolation 4/4 0/51 NT 0/51 RT-PCR 4/4 0/51 NT NT a NVD, no virus detected. b NT, not tested. 209