- Email: [email protected]
Rapid screening of embryonated chicken eggs for bluetongue virus infection with an antigen capture enzyme linked immunosorbent assay
Journal of Virological Methods 75 (1998) 39 – 46
Rapid screening of embryonated chicken eggs for bluetongue virus infection with an antigen capture enzyme linked immunosorbent assay M. Hosseini a, R.A. Hawkes b, P.D. Kirkland b,*, R.J. Dixon a a
Department of Veterinary Clinical Sciences, Uni6ersity of Sydney, Camden, NSW 2570, Australia b Elizabeth Macarthur Agricultural Institute, NSW Agriculture, Camden, NSW 2570, Australia Received 29 January 1998; received in revised form 16 June 1998; accepted 16 June 1998
Abstract The sensitivity and specificity of an antigen capture ELISA have been compared with virus isolation in cell culture. Bluetongue virus (BLU) (serotype 23) from the blood of a sheep was titrated by inoculating embryonated chicken eggs (ECEs) and detecting viral antigen in chicken embryo livers using an antigen capture enzyme linked immunosorbent assay (ELISA) (Stanislawek et al., 1996. Detection by ELISA of bluetongue antigen directly in the blood of experimentally infected sheep. Vet. Microbiol. 52, 1–12). Five days after inoculation of ECEs with lysed red blood cells from the infected sheep the embryo livers were harvested and homogenised. The supernatant from the homogenate was used in the antigen capture ELISA to determine which livers were infected and the virus titre calculated as CEID50/ml packed red blood cells. These results were compared with a standard cell culture isolation protocol which passaged the liver homogenate supernatant through Aedes albopictus cells and up to three passages in BHK21 cells. The antigen capture ELISA showed 100% sensitivity and specificity with no false negatives or false positives when compared to cell culture isolation of the virus. The major advantage of the combination of ECE inoculation and antigen capture ELISA is the reduction in the time to less than 7 days from a maximum of 35 days for the ECE/cell culture system. The procedure is easy to undertake, cost effective and does not require expensive specialist cell culture facilities. © 1998 Elsevier Science B.V. All rights reserved. Keywords: Bluetongue virus; Antigen capture ELISA, Embryonated chicken eggs
1. Introduction
* Corresponding author. Tel.: +61-2-46406331; Fax: +612-46406429; E-mail: [email protected]
A number of procedures are used currently to detect bluetongue virus (BLU) in blood or tissues from infected animals. These include inoculation of embryonated chicken eggs (ECEs) followed by
0166-0934/98/$ - see front matter © 1998 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 6 - 0 9 3 4 ( 9 8 ) 0 0 0 9 6 - 2
M. Hosseini et al. / Journal of Virological Methods 75 (1998) 39–46
40
passaging through insect and mammalian cell culture; direct inoculation of insect and/or mammalian cells; sheep inoculation; nucleic acid probes or polymerase chain reaction (PCR) to detect BLU virus nucleic acid; and, antigen capture enzyme linked immunosorbent assay (ELISA) to detect BLU virus antigen either directly or after amplification in cell culture (Mecham et al., 1990; McColl and Gould, 1991; Gard and Kirkland, 1993; Mecham, 1993; Afshar, 1994). Many of these are both time consuming and expensive, or require specialist facilities. The method accepted generally for testing of animals for export and other regulatory purposes involves the inoculation of ECEs and passaging through cell culture. This method is believed to be as sensitive as inoculating sheep for the detection of live virus (Foster et al., 1972; Goldsmit et al., 1975). ELISA technology to detect virus antigen has a number of advantages including being economical to use and the short time to determine a result. However, an antigen capture ELISA to detect BLU virus directly from sheep blood has been found to have low sensitivity. This sensitivity was increased to a certain degree by the amplification of BLU virus through cell culture (Mecham, 1993). More recently, Stanislawek et al. (1996) describe an improved antigen capture ELISA using a monoclonal antibody (Mab) that was able to detect BLU virus in the blood of animals that were sampled during the period of highest antigenaemia. This report applies the antigen capture ELISA of Stanislawek et al. (1996) to detect virus amplified in chick embryo livers after inoculation of ECEs with infected sheep blood. This approach maintains the high sensitivity achieved by inoculation of ECEs, but shortens the antigen/virus detection process and avoids the need for cell culture after amplification in ECE.
2. Materials and methods
2.1. Source of BLU 6irus An Australian isolate of BLU-23 from a sen-
tinel calf was directly passaged once in a Merino sheep. One ml of heparinised blood from this animal was collected 8 days after inoculation and injected subcutaneously into a second Merino sheep. Subsequently, heparinised blood samples were taken at 5, 7, 10 and 12 days post-inoculation (DPI) and used in this study.
2.2. Amplification of BLU 6irus in embryonated chicken eggs BLU virus amplification in ECE followed the protocol of Gard and Kirkland (1993). Briefly, packed red cells from the BLU infected sheep were washed in 0.2 M phosphate-buffered saline (PBS) pH 7.3 and centrifuged at 1000×g for 10 min. Seventy five ml of the packed red cells were then lysed by diluting 1/10 in sterile distilled water at 4°C. A series of log10 dilutions were prepared in sterile normal saline (0.85% NaCl in double distilled water) and 100 ml of each dilution was inoculated intravenously into five 10–11 day old ECEs. The inoculated ECEs were then kept at 33.5°C for 5 days. Embryos dying within 24 h of inoculation were discarded. Embryos dying after 24 h and those surviving for 5 days after inoculation were collected. Embryo livers were harvested into sterile 1.5 ml microcentrifuge tubes containing 1.0 ml of PBST (0.5% Tween20 in PBS) and homogenised by aspirating three to four times with a 5 ml syringe and 18 G needle. The liver homogenates were divided and frozen at − 80°C or refrigerated at 4°C. Before further processing, these homogenates were thawed if frozen, and all were centrifuged at 10 000× g for 1 min. The supernatants were then tested in the antigen capture ELISA as the sample or subjected to virus isolation in cell culture as described below.
2.3. Antigen capture enzyme linked immunosorbent assay The antigen capture ELISA method was essentially that developed by Stanislawek et al. (1996) with minor modifications. Half the wells of a microtitre plate were coated with 50 ml/well of diluted (1/1000) BLU group reactive Mab
M. Hosseini et al. / Journal of Virological Methods 75 (1998) 39–46
41
(CSIRO/BTV/20E9B7G2, Lunt et al., 1988) in 0.1M carbonate buffer (pH 9.6). Unlike the protocol of Stanislawek et al. (1996), the other wells were coated with an irrelevant Mab (Murray Valley Encephalitis: MVE 4B6A2, Hawkes et al., 1989) also diluted 1/1000. After being left at 4°C overnight and washed with PBST, 50 ml of the supernatant of each chick embryo liver-homogenate were added to two wells on the BLUMab side and two wells on the irrelevant Mab side. Supernatant from a liver homogenate of a noninfected 15 day chicken embryo was used as the negative control and undiluted cell culture media from BLU 23 and BLU1-infected BHK21 cells were the positive controls. The plates were incubated for 1 hr at room temperature and then shaken for 1 h at 37°C. Wells were washed in PBST and then a rabbit polyclonal anti BLU-20 core antibody (a kind gift of R. Lunt, Australian Animal Health Laboratory, Geelong) diluted 1/ 2000 in skim milk buffer (1% skim milk in PBST) was added (50 ml/well). The plates were shaken for 30 min at 37°C, washed with PBST and horseradish peroxidase conjugated anti-rabbit IgG (1/3000, Silenus, Cat. No. RAH) added (50 ml/well). After shaking at 37°C for 30 min, the plates were again washed and 50 ml of TMB substrate (3% 3%-3%,5%-5%-tetraethylbenzidine in 1.0 M citric acetate buffer, pH 6.0 and 1/2000 dilution 30% H2O2) was added to each well and left 15 – 20 min. The reaction was stopped by the addition of 50 ml 1M H2SO4 to each well and the optical density (OD) read at 450 nm. Whether a sample was positive or negative was determined by the following formula:
culture (Gard and Kirkland, 1993). Heparinised blood was inoculated into ECE intravenously and embryo livers harvested as described above. Supernatant from the chick embryo liver homogenate was diluted from 10 − 1 to 10 − 6 and 100 ml was inoculated into each of 2 tube cultures of Aedes albopictus (AA) mosquito cells (C6/36) when the cells were 80–90% confluent. Cells were maintained in a modified MM (Mitsuhashi and Maramorosch) medium (Williams, 1994) with 2% foetal calf serum (FCS) and incubated at 25°C. After 1 week the AA cells were scraped from the duplicate tubes and the cells and medium pooled for each specimen. The pooled suspension was passaged to two BHK-21 cell cultures at 90% confluency. BHK-21 cells were maintained in Medium 199 with 2% FCS at 37°C. All tubes were examined daily for 7 days for cytopathic effects (CPE). Those tubes which did not show CPE after 7 days were passaged two more times in BHK-21 cells. Cells not showing CPE after three passages, were considered negative for BLU virus. Virus titre in the original blood sample was calculated from the number of positive/negative replicates at each dilution using the Poisson distribution (Kleczkowski, 1968) as described by Roberts et al. (1988). The titre was expressed as 50% chicken embryo infective doses (CEID50)/ml of packed red blood cells.
Signal to noise (S/N) ratio
Several experiments were done to determine the specificity of screening chick embryo liver homogenates by antigen capture ELISA in comparison with cell culture and to assess the effects of different storage temperatures for liver homogenates.
=
mean O.D. of sample with BLU ×Mab mean O.D. of sample with irrelevant ×Mab
A signal to noise ratio of 2:1 or greater was defined as positive and a ratio of less than 2:1 was considered negative.
2.4. Bluetongue 6irus isolation using ECE and cell culture BLU virus was isolated using ECE and cell
2.5. Comparing the detection of BLU in chicken embryo li6ers by either the antigen capture ELISA or 6irus isolation in cell culture
2.5.1. Sensiti6ity and specificity of the antigen capture ELISA The specificity of the antigen capture ELISA was compared with virus isolation in cell culture using samples screened in the antigen capture
42
M. Hosseini et al. / Journal of Virological Methods 75 (1998) 39–46
ELISA. Thirty two negative and 4 positive undiluted supernatants of chick embryo liver homogenates were selected at random. These homogenates were held at 4°C overnight and were passaged once in AA cells and three times in BHK-21 cells. Another four negative and three positive samples that had been frozen at − 80°C were thawed and passaged through cell culture as described.
antigen capture ELISA undiluted, or at dilutions of 10 − 1 and 10 − 2.
2.6. Statistical analysis An ANOVA test was used for comparison between groups where appropriate.
3. Results
2.5.2. Effect of freezing and thawing of li6er homogenates The effect of freezing and thawing liver homogenates on the antigen capture ELISA and virus isolation results was tested. Three positive and four negative samples as determined by the antigen capture ELISA and frozen at −80°C were thawed. These were used in the antigen capture ELISA and subjected to virus isolation in cell culture. 2.5.3. The effect of holding temperature on ELISA and 6irus isolation results The effect of holding liver homogenates at 4 or −80°C was determined on four fresh homogenates that were positive for BLU virus in the antigen capture ELISA. These homogenates were divided into two equal volumes and kept at either 4 or −80°C overnight. The next day supernatants were collected after centrifugation, diluted (10 − 1 to 10 − 6) in sterile normal saline and passaged through AA and BHK-21 cells (four tubes/dilution) to determine the virus titre (TCID50/0.1ml) in each homogenate. This series of diluted supernatants was also tested in the antigen capture ELISA. 2.5.4. The effect of different holding temperatures on antigen capture ELISA results The effect of holding temperatures on S/N ratios was further investigated. Fifteen fresh liver homogenates that were positive for BLU virus were divided into three similar samples. One sample was centrifuged immediately and the supernatant tested in the antigen capture ELISA. The other two were left overnight either at 4 or − 80°C. Both of these samples were centrifuged the next day and the supernatants tested in the
3.1. The sensiti6ity and specificity of the antigen capture ELISA compared with 6irus isolation by cell culture Virus was not isolated from any fresh or frozen sample that was negative in the antigen capture ELISA (Table 1). Similarly, virus was isolated in cell culture from all the fresh and frozen samples that were positive in the antigen capture ELISA.
3.2. The effect of freezing and thawing on the antigen capture ELISA results compared with 6irus isolation The holding temperature (− 80 or 4°C) of samples overnight did not change the final result (Table 1). However, the infectivity of BLU was reduced by freezing. All positive samples held at 4°C and which were positive in the antigen capture ELISA showed CPE on the first passage in BHK-21 cells, whereas positive samples that had been frozen showed CPE on the second or third passage in cells. Table 1 Comparison of the sensitivity and specificity of the antigen capture ELISA with the isolation of bluetongue virus by cell culture Antigen capture ELISA
Negative (refrigerated) Positive (refrigerated) Negative (frozen) Positive (frozen)
Cell culture Positive
Negative
0 4 0 3
32 0 4 0
M. Hosseini et al. / Journal of Virological Methods 75 (1998) 39–46
43
Table 2 The effect of holding temperature of the liver homogenate on the S/N ratio of the antigen capture ELISA compared with the effect of holding temperature on virus titre determined by cell culture Sample no.a
Dilution
Titreb
Signal/noise ratio 4°C
−80°C
4°C
−80°C
316
100 10−1 10−2
9.1 1.4 0.8
7.9 2.3 1.12
103.5
101.5
318
100 10−1 10−2
4.0 1.6 0.8
7.6 2.3 1.19
102.1
100.4
326
100 10−1 10−2
4.0 1.26 0.74
8.1 2.1 1.0
102.28
100.89
334
100 10−1 10−2
4.9 1.48 0.88
7.4 2.2 1.0
102.13
100.89
a
These samples were obtained from liver-homogenates of ECEs inoculated with 10−4 and 10−6 dilution of packed red blood cells from the infected sheep. b Log10TCID50/0.1 ml of supernatant of liver homogenate.
3.3. The effect of holding temperature of the li6er homogenate on S/N ratios in the antigen capture ELISA and on 6irus titre determined by cell culture While the infectivity determined in cell culture of all 4 refrigerated samples was higher than the frozen samples (PB 0.008), the S/N ratios in the antigen capture ELISA of frozen samples appeared higher than the refrigerated samples (Table 2). However, this difference in S/N did not reach statistical significance, due, in part, to the limited number of replicates. The higher S/N may indicate that freezing/thawing releases antigen detectable in the antigen capture ELISA, but inactivates infectious virus. The only exception was sample No. 316 where the S/N ratio of the undiluted frozen sample was slightly lower. Virus titres were reduced in frozen samples by up to 102.0 TCID50/ml.
3.4. The effect of different holding temperatures of li6er homogenates on S/N 6alues in the antigen capture ELISA The effect of holding temperature on S/N val-
ues was investigated further using a larger number of samples. The results are summarised in Table 3. S/N ratios were markedly higher after overnight storage. The mean S/N ratio for fresh was less than for refrigerated which in turn was much less than for frozen samples. This difference was significant (PB 0.004) between fresh and refrigerated samples. The difference was highly significant (PB 0.001) between fresh and frozen or refrigerated and frozen samples. Therefore, S/N ratios can be influenced by how liver homogenates are stored after collection but this did not change whether the samples were classified as positive or negative. 4. Discussion The antigen capture ELISA can replace the cell culture step for determining the presence or absence of BLU virus in sheep blood after amplification through ECE. The screening of embryo livers in the ELISA was shown to have identical sensitivity and specificity to virus isolation in cell culture. The procedure is easy to carry out, cost-effective and rapid compared with cell culture. Up to 33 days (BLU-amplification in ECE 5
0.1939 0.014 NDe ND
2.80 90.23 ND ND
0.069 90.002 ND ND
0.463 9 0.02 0.124 9 0.009 0.173 9 0.016 0.065 90.001 No. samples with S/N 2.0
b
3.73 9 0.21 2.66 9 0.20f
Mean OD irrel Mean S/N
Mean OD rel
Mean S/Nd
Mean OD relb Mean OD irrelc
Refrigerated (4°C) overnight
Fresh
Sample: supernatant from liver-homogenate of ECE. Mean optical density 9 standard error of 15 samples on the relevant Mab side of the plate. c Mean optical density 9standard error of 15 samples on the irrelevant Mab side of the plate. d S/N ratio. e Not done. f Twelve of the 15 samples had S/N ratio 2. g Thirteen of the 15 samples had S/N ratio 2.
a
100 10−1 10−2
Samplea dilution
Table 3 The effect of holding temperatures of liver homogenates on S/N ratios determined in the antigen capture ELISA
Mean OD irrel
0.4299 0.022 0.070 90.001 0.1609 0.012 0.060 90.001 No. samples with S/N 2.0
Mean OD rel
Frozen (−80°C) overnight
6.13 9 0.303 2.67 9 0.196g
Mean S/N
44 M. Hosseini et al. / Journal of Virological Methods 75 (1998) 39–46
M. Hosseini et al. / Journal of Virological Methods 75 (1998) 39–46
days, mosquito and mammalian cell culture 28 days) may be required for the detection of BLU in cell culture (i.e. before a sample can be declared confidently as negative). If the antigen capture ELISA is used the time required is only 6 days (BLU-amplification in ECE 5 days, antigen capture ELISA 1 day). In addition, cell culture is expensive, labour intensive and requires specialist facilities. Therefore, the antigen capture ELISA is useful for the screening of chicken embryos for evidence of the infectivity of samples. If a sample is positive, virus isolation could then be done selectively to serotype the strain of BLU. This approach is suitable where bluetongue virus isolation is required for regulatory purposes, for epizootiological surveys, diagnosis of clinical cases or the titration of wild strains of BLU that have not been adapted to cell culture, for example challenge virus in vaccine trials. The antigen capture ELISA demonstrated 100% sensitivity and specificity. There were no false positives or false negatives of refrigerated or frozen liver samples when they were passaged through cell culture. Because the capture mAb used in the assay (anti VP7) is restricted in reactivity to members of the BLU group, the antigen capture ELISA should not detect BLU-virus-related orbiviruses such as epizootic haemorrhagic disease of deer. The S/N ratio was influenced by how the chicken embryo liver was stored. For example, fresh livers had lower S/N ratios than those refrigerated overnight, which in turn, were lower than those frozen. Nevertheless, although there were higher S/N values from frozen samples, frozen livers negative for virus were also negative in the antigen capture ELISA. These altered S/N ratios did not alter the classification of a sample. Interestingly, prolonged storage at 4°C did increase OD values for the irrelevant mAb but did not influence the final outcome. This difference needs further investigation but probably involves the liberation of non-specific antigens from livers stored overnight at 4°C. The antigen capture ELISA could be designed without using an irrelevant antibody by utilising a strictly analogous negative control antigen with a relevant capture antibody.
45
It may be necessary to serotype an isolate after determining a positive result in the antigen capture ELISA. This requires isolation in cell culture. Although virus titre was highest in liver samples that had not been frozen, practical considerations may determine the use of frozen specimens instead. These should still yield virus in culture. Before implementation as a routine screening method, there is a need to examine a wide range of BLU serotypes utilising a large number of infected and noninfected samples tested in parallel by virus isolation. In this study, the degree of amplification in ECE of BLU-23 from sheep blood was sufficient to be detected by the antigen capture ELISA. However, viral antigen may not be expressed in livers of chicken embryos to the same level by other serotypes or isolates. Non cell culture adapted (wild) BLU usually needs to be passaged through ECEs and the use of direct cell culture inoculation for the isolation of these viruses is generally not considered an option. The antigen capture ELISA in combination with ECE inoculation could prove to be a cost and time effective protocol for the detection of BLU and for determination of BLU virus titres in blood samples. While other antigen or nucleic acid detection procedures such as immunoblotting, PCR or nucleic acid probes could be used to screen embryo tissues, they generally take as long or longer than the antigen capture ELISA, are not suitable for the screening of large numbers of samples and are usually more expensive.
Acknowledgements M. Hosseini was supported by the Iranian Ministry of Culture and Higher Education. The Australian Centre for International Agricultural Research (ACIAR) also contributed funds towards this project.
References Afshar, A., 1994. Bluetongue: laboratory diagnosis. Comp. Immunol. Microbiol. Infect. Dis. 17, 221 – 242. Foster, N.M., Luedke, A.J., Metcalf, H.E., 1972. Bluetongue
46
M. Hosseini et al. / Journal of Virological Methods 75 (1998) 39–46
in sheep and cattle: efficacy of embryonating chicken eggs in viral isolations. Am. J. Vet. Res. 33, 77–81. Gard, G.P., Kirkland, P.D., 1993. Bluetongue virology and serology. In: Corner L.A., Bagust T.J. (Eds.), Australian Standard Diagnostic Techniques for Animal Diseases. CSIRO for the Standing Committee on Agriculture and Resource Management, East Melbourne, Australia. Goldsmit, L., Barzilai, E., Tadmor, A., 1975. The comparative sensitivity of sheep and chicken embryos to bluetongue virus and observations on viraemia in experimentally infected sheep. Aust. vet. J. 51, 190–196. Hawkes, R.A., Roehrig, J.T., Hunt, A.R., Moore, G.A., 1989. Antigenic structure of the Murray Valley encephalitis virus E glycoprotein. J. Gen. Virol. 69, 1105–1109. Kleczkowski, A., 1968. Experimental design and statistical methods of assay. In: Maramorosch, K., Koprowski, H. (Eds.), Methods in Virology, vol. 4. Academic Press, New York, pp. 705 – 713. Lunt, R.A., White, J.R., Blacksell, S.D., 1988. Evaluation of a monoclonal antibody blocking ELISA for the detection of group-specific antibodies to bluetongue virus in experimental and field sera. J. Gen. Virol. 69, 2729–2740.
McColl, K.A., Gould, A.R., 1991. Detection and characterisation of bluetongue virus using the polymerase chain reaction. Virus Res. 21, 19 – 34. Mecham, J.O., 1993. Detection of bluetongue virus from blood of infected sheep by use of an antigen-capture enzyme-linked immunosorbent assay after amplification of the virus in cell culture. Am. J. Vet. Res. 54, 370 – 372. Mecham, J.O., Dean, V.C., Wigington, J.G., Nunamaker, R.A., 1990. Detection of bluetongue virus in Culicoides 6ariipennis by antigen capture ELISA assay. J. Med. Entomol. 27, 602 – 606. Roberts, P.C., Etchison, J.R., Bond, C.W., 1988. A rapid, quantitative assay for titration of bovine virus diarrhoeamucosal disease virus. Vet. Microbiol. 18, 209 – 217. Stanislawek, W.L., Lunt, R.A., Blacksell, S.D., Newberry, K., Hooper, P.T., White, J.R., 1996. Detection by ELISA of bluetongue antigen directly in the blood of experimentally infected sheep. Vet. Microbiol. 52, 1 – 12. Williams, C.F., 1994. Studies on the Australian bluetongue viruses. Master of Science in Veterinary Science Thesis, University of Sydney, Australia.
. .
Rapid screening of embryonated chicken eggs for bluetongue virus infection with an antigen capture enzyme linked immunosorbent assay M. Hosseini a, R.A. Hawkes b, P.D. Kirkland b,*, R.J. Dixon a a
Department of Veterinary Clinical Sciences, Uni6ersity of Sydney, Camden, NSW 2570, Australia b Elizabeth Macarthur Agricultural Institute, NSW Agriculture, Camden, NSW 2570, Australia Received 29 January 1998; received in revised form 16 June 1998; accepted 16 June 1998
Abstract The sensitivity and specificity of an antigen capture ELISA have been compared with virus isolation in cell culture. Bluetongue virus (BLU) (serotype 23) from the blood of a sheep was titrated by inoculating embryonated chicken eggs (ECEs) and detecting viral antigen in chicken embryo livers using an antigen capture enzyme linked immunosorbent assay (ELISA) (Stanislawek et al., 1996. Detection by ELISA of bluetongue antigen directly in the blood of experimentally infected sheep. Vet. Microbiol. 52, 1–12). Five days after inoculation of ECEs with lysed red blood cells from the infected sheep the embryo livers were harvested and homogenised. The supernatant from the homogenate was used in the antigen capture ELISA to determine which livers were infected and the virus titre calculated as CEID50/ml packed red blood cells. These results were compared with a standard cell culture isolation protocol which passaged the liver homogenate supernatant through Aedes albopictus cells and up to three passages in BHK21 cells. The antigen capture ELISA showed 100% sensitivity and specificity with no false negatives or false positives when compared to cell culture isolation of the virus. The major advantage of the combination of ECE inoculation and antigen capture ELISA is the reduction in the time to less than 7 days from a maximum of 35 days for the ECE/cell culture system. The procedure is easy to undertake, cost effective and does not require expensive specialist cell culture facilities. © 1998 Elsevier Science B.V. All rights reserved. Keywords: Bluetongue virus; Antigen capture ELISA, Embryonated chicken eggs
1. Introduction
* Corresponding author. Tel.: +61-2-46406331; Fax: +612-46406429; E-mail: [email protected]
A number of procedures are used currently to detect bluetongue virus (BLU) in blood or tissues from infected animals. These include inoculation of embryonated chicken eggs (ECEs) followed by
0166-0934/98/$ - see front matter © 1998 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 6 - 0 9 3 4 ( 9 8 ) 0 0 0 9 6 - 2
M. Hosseini et al. / Journal of Virological Methods 75 (1998) 39–46
40
passaging through insect and mammalian cell culture; direct inoculation of insect and/or mammalian cells; sheep inoculation; nucleic acid probes or polymerase chain reaction (PCR) to detect BLU virus nucleic acid; and, antigen capture enzyme linked immunosorbent assay (ELISA) to detect BLU virus antigen either directly or after amplification in cell culture (Mecham et al., 1990; McColl and Gould, 1991; Gard and Kirkland, 1993; Mecham, 1993; Afshar, 1994). Many of these are both time consuming and expensive, or require specialist facilities. The method accepted generally for testing of animals for export and other regulatory purposes involves the inoculation of ECEs and passaging through cell culture. This method is believed to be as sensitive as inoculating sheep for the detection of live virus (Foster et al., 1972; Goldsmit et al., 1975). ELISA technology to detect virus antigen has a number of advantages including being economical to use and the short time to determine a result. However, an antigen capture ELISA to detect BLU virus directly from sheep blood has been found to have low sensitivity. This sensitivity was increased to a certain degree by the amplification of BLU virus through cell culture (Mecham, 1993). More recently, Stanislawek et al. (1996) describe an improved antigen capture ELISA using a monoclonal antibody (Mab) that was able to detect BLU virus in the blood of animals that were sampled during the period of highest antigenaemia. This report applies the antigen capture ELISA of Stanislawek et al. (1996) to detect virus amplified in chick embryo livers after inoculation of ECEs with infected sheep blood. This approach maintains the high sensitivity achieved by inoculation of ECEs, but shortens the antigen/virus detection process and avoids the need for cell culture after amplification in ECE.
2. Materials and methods
2.1. Source of BLU 6irus An Australian isolate of BLU-23 from a sen-
tinel calf was directly passaged once in a Merino sheep. One ml of heparinised blood from this animal was collected 8 days after inoculation and injected subcutaneously into a second Merino sheep. Subsequently, heparinised blood samples were taken at 5, 7, 10 and 12 days post-inoculation (DPI) and used in this study.
2.2. Amplification of BLU 6irus in embryonated chicken eggs BLU virus amplification in ECE followed the protocol of Gard and Kirkland (1993). Briefly, packed red cells from the BLU infected sheep were washed in 0.2 M phosphate-buffered saline (PBS) pH 7.3 and centrifuged at 1000×g for 10 min. Seventy five ml of the packed red cells were then lysed by diluting 1/10 in sterile distilled water at 4°C. A series of log10 dilutions were prepared in sterile normal saline (0.85% NaCl in double distilled water) and 100 ml of each dilution was inoculated intravenously into five 10–11 day old ECEs. The inoculated ECEs were then kept at 33.5°C for 5 days. Embryos dying within 24 h of inoculation were discarded. Embryos dying after 24 h and those surviving for 5 days after inoculation were collected. Embryo livers were harvested into sterile 1.5 ml microcentrifuge tubes containing 1.0 ml of PBST (0.5% Tween20 in PBS) and homogenised by aspirating three to four times with a 5 ml syringe and 18 G needle. The liver homogenates were divided and frozen at − 80°C or refrigerated at 4°C. Before further processing, these homogenates were thawed if frozen, and all were centrifuged at 10 000× g for 1 min. The supernatants were then tested in the antigen capture ELISA as the sample or subjected to virus isolation in cell culture as described below.
2.3. Antigen capture enzyme linked immunosorbent assay The antigen capture ELISA method was essentially that developed by Stanislawek et al. (1996) with minor modifications. Half the wells of a microtitre plate were coated with 50 ml/well of diluted (1/1000) BLU group reactive Mab
M. Hosseini et al. / Journal of Virological Methods 75 (1998) 39–46
41
(CSIRO/BTV/20E9B7G2, Lunt et al., 1988) in 0.1M carbonate buffer (pH 9.6). Unlike the protocol of Stanislawek et al. (1996), the other wells were coated with an irrelevant Mab (Murray Valley Encephalitis: MVE 4B6A2, Hawkes et al., 1989) also diluted 1/1000. After being left at 4°C overnight and washed with PBST, 50 ml of the supernatant of each chick embryo liver-homogenate were added to two wells on the BLUMab side and two wells on the irrelevant Mab side. Supernatant from a liver homogenate of a noninfected 15 day chicken embryo was used as the negative control and undiluted cell culture media from BLU 23 and BLU1-infected BHK21 cells were the positive controls. The plates were incubated for 1 hr at room temperature and then shaken for 1 h at 37°C. Wells were washed in PBST and then a rabbit polyclonal anti BLU-20 core antibody (a kind gift of R. Lunt, Australian Animal Health Laboratory, Geelong) diluted 1/ 2000 in skim milk buffer (1% skim milk in PBST) was added (50 ml/well). The plates were shaken for 30 min at 37°C, washed with PBST and horseradish peroxidase conjugated anti-rabbit IgG (1/3000, Silenus, Cat. No. RAH) added (50 ml/well). After shaking at 37°C for 30 min, the plates were again washed and 50 ml of TMB substrate (3% 3%-3%,5%-5%-tetraethylbenzidine in 1.0 M citric acetate buffer, pH 6.0 and 1/2000 dilution 30% H2O2) was added to each well and left 15 – 20 min. The reaction was stopped by the addition of 50 ml 1M H2SO4 to each well and the optical density (OD) read at 450 nm. Whether a sample was positive or negative was determined by the following formula:
culture (Gard and Kirkland, 1993). Heparinised blood was inoculated into ECE intravenously and embryo livers harvested as described above. Supernatant from the chick embryo liver homogenate was diluted from 10 − 1 to 10 − 6 and 100 ml was inoculated into each of 2 tube cultures of Aedes albopictus (AA) mosquito cells (C6/36) when the cells were 80–90% confluent. Cells were maintained in a modified MM (Mitsuhashi and Maramorosch) medium (Williams, 1994) with 2% foetal calf serum (FCS) and incubated at 25°C. After 1 week the AA cells were scraped from the duplicate tubes and the cells and medium pooled for each specimen. The pooled suspension was passaged to two BHK-21 cell cultures at 90% confluency. BHK-21 cells were maintained in Medium 199 with 2% FCS at 37°C. All tubes were examined daily for 7 days for cytopathic effects (CPE). Those tubes which did not show CPE after 7 days were passaged two more times in BHK-21 cells. Cells not showing CPE after three passages, were considered negative for BLU virus. Virus titre in the original blood sample was calculated from the number of positive/negative replicates at each dilution using the Poisson distribution (Kleczkowski, 1968) as described by Roberts et al. (1988). The titre was expressed as 50% chicken embryo infective doses (CEID50)/ml of packed red blood cells.
Signal to noise (S/N) ratio
Several experiments were done to determine the specificity of screening chick embryo liver homogenates by antigen capture ELISA in comparison with cell culture and to assess the effects of different storage temperatures for liver homogenates.
=
mean O.D. of sample with BLU ×Mab mean O.D. of sample with irrelevant ×Mab
A signal to noise ratio of 2:1 or greater was defined as positive and a ratio of less than 2:1 was considered negative.
2.4. Bluetongue 6irus isolation using ECE and cell culture BLU virus was isolated using ECE and cell
2.5. Comparing the detection of BLU in chicken embryo li6ers by either the antigen capture ELISA or 6irus isolation in cell culture
2.5.1. Sensiti6ity and specificity of the antigen capture ELISA The specificity of the antigen capture ELISA was compared with virus isolation in cell culture using samples screened in the antigen capture
42
M. Hosseini et al. / Journal of Virological Methods 75 (1998) 39–46
ELISA. Thirty two negative and 4 positive undiluted supernatants of chick embryo liver homogenates were selected at random. These homogenates were held at 4°C overnight and were passaged once in AA cells and three times in BHK-21 cells. Another four negative and three positive samples that had been frozen at − 80°C were thawed and passaged through cell culture as described.
antigen capture ELISA undiluted, or at dilutions of 10 − 1 and 10 − 2.
2.6. Statistical analysis An ANOVA test was used for comparison between groups where appropriate.
3. Results
2.5.2. Effect of freezing and thawing of li6er homogenates The effect of freezing and thawing liver homogenates on the antigen capture ELISA and virus isolation results was tested. Three positive and four negative samples as determined by the antigen capture ELISA and frozen at −80°C were thawed. These were used in the antigen capture ELISA and subjected to virus isolation in cell culture. 2.5.3. The effect of holding temperature on ELISA and 6irus isolation results The effect of holding liver homogenates at 4 or −80°C was determined on four fresh homogenates that were positive for BLU virus in the antigen capture ELISA. These homogenates were divided into two equal volumes and kept at either 4 or −80°C overnight. The next day supernatants were collected after centrifugation, diluted (10 − 1 to 10 − 6) in sterile normal saline and passaged through AA and BHK-21 cells (four tubes/dilution) to determine the virus titre (TCID50/0.1ml) in each homogenate. This series of diluted supernatants was also tested in the antigen capture ELISA. 2.5.4. The effect of different holding temperatures on antigen capture ELISA results The effect of holding temperatures on S/N ratios was further investigated. Fifteen fresh liver homogenates that were positive for BLU virus were divided into three similar samples. One sample was centrifuged immediately and the supernatant tested in the antigen capture ELISA. The other two were left overnight either at 4 or − 80°C. Both of these samples were centrifuged the next day and the supernatants tested in the
3.1. The sensiti6ity and specificity of the antigen capture ELISA compared with 6irus isolation by cell culture Virus was not isolated from any fresh or frozen sample that was negative in the antigen capture ELISA (Table 1). Similarly, virus was isolated in cell culture from all the fresh and frozen samples that were positive in the antigen capture ELISA.
3.2. The effect of freezing and thawing on the antigen capture ELISA results compared with 6irus isolation The holding temperature (− 80 or 4°C) of samples overnight did not change the final result (Table 1). However, the infectivity of BLU was reduced by freezing. All positive samples held at 4°C and which were positive in the antigen capture ELISA showed CPE on the first passage in BHK-21 cells, whereas positive samples that had been frozen showed CPE on the second or third passage in cells. Table 1 Comparison of the sensitivity and specificity of the antigen capture ELISA with the isolation of bluetongue virus by cell culture Antigen capture ELISA
Negative (refrigerated) Positive (refrigerated) Negative (frozen) Positive (frozen)
Cell culture Positive
Negative
0 4 0 3
32 0 4 0
M. Hosseini et al. / Journal of Virological Methods 75 (1998) 39–46
43
Table 2 The effect of holding temperature of the liver homogenate on the S/N ratio of the antigen capture ELISA compared with the effect of holding temperature on virus titre determined by cell culture Sample no.a
Dilution
Titreb
Signal/noise ratio 4°C
−80°C
4°C
−80°C
316
100 10−1 10−2
9.1 1.4 0.8
7.9 2.3 1.12
103.5
101.5
318
100 10−1 10−2
4.0 1.6 0.8
7.6 2.3 1.19
102.1
100.4
326
100 10−1 10−2
4.0 1.26 0.74
8.1 2.1 1.0
102.28
100.89
334
100 10−1 10−2
4.9 1.48 0.88
7.4 2.2 1.0
102.13
100.89
a
These samples were obtained from liver-homogenates of ECEs inoculated with 10−4 and 10−6 dilution of packed red blood cells from the infected sheep. b Log10TCID50/0.1 ml of supernatant of liver homogenate.
3.3. The effect of holding temperature of the li6er homogenate on S/N ratios in the antigen capture ELISA and on 6irus titre determined by cell culture While the infectivity determined in cell culture of all 4 refrigerated samples was higher than the frozen samples (PB 0.008), the S/N ratios in the antigen capture ELISA of frozen samples appeared higher than the refrigerated samples (Table 2). However, this difference in S/N did not reach statistical significance, due, in part, to the limited number of replicates. The higher S/N may indicate that freezing/thawing releases antigen detectable in the antigen capture ELISA, but inactivates infectious virus. The only exception was sample No. 316 where the S/N ratio of the undiluted frozen sample was slightly lower. Virus titres were reduced in frozen samples by up to 102.0 TCID50/ml.
3.4. The effect of different holding temperatures of li6er homogenates on S/N 6alues in the antigen capture ELISA The effect of holding temperature on S/N val-
ues was investigated further using a larger number of samples. The results are summarised in Table 3. S/N ratios were markedly higher after overnight storage. The mean S/N ratio for fresh was less than for refrigerated which in turn was much less than for frozen samples. This difference was significant (PB 0.004) between fresh and refrigerated samples. The difference was highly significant (PB 0.001) between fresh and frozen or refrigerated and frozen samples. Therefore, S/N ratios can be influenced by how liver homogenates are stored after collection but this did not change whether the samples were classified as positive or negative. 4. Discussion The antigen capture ELISA can replace the cell culture step for determining the presence or absence of BLU virus in sheep blood after amplification through ECE. The screening of embryo livers in the ELISA was shown to have identical sensitivity and specificity to virus isolation in cell culture. The procedure is easy to carry out, cost-effective and rapid compared with cell culture. Up to 33 days (BLU-amplification in ECE 5
0.1939 0.014 NDe ND
2.80 90.23 ND ND
0.069 90.002 ND ND
0.463 9 0.02 0.124 9 0.009 0.173 9 0.016 0.065 90.001 No. samples with S/N 2.0
b
3.73 9 0.21 2.66 9 0.20f
Mean OD irrel Mean S/N
Mean OD rel
Mean S/Nd
Mean OD relb Mean OD irrelc
Refrigerated (4°C) overnight
Fresh
Sample: supernatant from liver-homogenate of ECE. Mean optical density 9 standard error of 15 samples on the relevant Mab side of the plate. c Mean optical density 9standard error of 15 samples on the irrelevant Mab side of the plate. d S/N ratio. e Not done. f Twelve of the 15 samples had S/N ratio 2. g Thirteen of the 15 samples had S/N ratio 2.
a
100 10−1 10−2
Samplea dilution
Table 3 The effect of holding temperatures of liver homogenates on S/N ratios determined in the antigen capture ELISA
Mean OD irrel
0.4299 0.022 0.070 90.001 0.1609 0.012 0.060 90.001 No. samples with S/N 2.0
Mean OD rel
Frozen (−80°C) overnight
6.13 9 0.303 2.67 9 0.196g
Mean S/N
44 M. Hosseini et al. / Journal of Virological Methods 75 (1998) 39–46
M. Hosseini et al. / Journal of Virological Methods 75 (1998) 39–46
days, mosquito and mammalian cell culture 28 days) may be required for the detection of BLU in cell culture (i.e. before a sample can be declared confidently as negative). If the antigen capture ELISA is used the time required is only 6 days (BLU-amplification in ECE 5 days, antigen capture ELISA 1 day). In addition, cell culture is expensive, labour intensive and requires specialist facilities. Therefore, the antigen capture ELISA is useful for the screening of chicken embryos for evidence of the infectivity of samples. If a sample is positive, virus isolation could then be done selectively to serotype the strain of BLU. This approach is suitable where bluetongue virus isolation is required for regulatory purposes, for epizootiological surveys, diagnosis of clinical cases or the titration of wild strains of BLU that have not been adapted to cell culture, for example challenge virus in vaccine trials. The antigen capture ELISA demonstrated 100% sensitivity and specificity. There were no false positives or false negatives of refrigerated or frozen liver samples when they were passaged through cell culture. Because the capture mAb used in the assay (anti VP7) is restricted in reactivity to members of the BLU group, the antigen capture ELISA should not detect BLU-virus-related orbiviruses such as epizootic haemorrhagic disease of deer. The S/N ratio was influenced by how the chicken embryo liver was stored. For example, fresh livers had lower S/N ratios than those refrigerated overnight, which in turn, were lower than those frozen. Nevertheless, although there were higher S/N values from frozen samples, frozen livers negative for virus were also negative in the antigen capture ELISA. These altered S/N ratios did not alter the classification of a sample. Interestingly, prolonged storage at 4°C did increase OD values for the irrelevant mAb but did not influence the final outcome. This difference needs further investigation but probably involves the liberation of non-specific antigens from livers stored overnight at 4°C. The antigen capture ELISA could be designed without using an irrelevant antibody by utilising a strictly analogous negative control antigen with a relevant capture antibody.
45
It may be necessary to serotype an isolate after determining a positive result in the antigen capture ELISA. This requires isolation in cell culture. Although virus titre was highest in liver samples that had not been frozen, practical considerations may determine the use of frozen specimens instead. These should still yield virus in culture. Before implementation as a routine screening method, there is a need to examine a wide range of BLU serotypes utilising a large number of infected and noninfected samples tested in parallel by virus isolation. In this study, the degree of amplification in ECE of BLU-23 from sheep blood was sufficient to be detected by the antigen capture ELISA. However, viral antigen may not be expressed in livers of chicken embryos to the same level by other serotypes or isolates. Non cell culture adapted (wild) BLU usually needs to be passaged through ECEs and the use of direct cell culture inoculation for the isolation of these viruses is generally not considered an option. The antigen capture ELISA in combination with ECE inoculation could prove to be a cost and time effective protocol for the detection of BLU and for determination of BLU virus titres in blood samples. While other antigen or nucleic acid detection procedures such as immunoblotting, PCR or nucleic acid probes could be used to screen embryo tissues, they generally take as long or longer than the antigen capture ELISA, are not suitable for the screening of large numbers of samples and are usually more expensive.
Acknowledgements M. Hosseini was supported by the Iranian Ministry of Culture and Higher Education. The Australian Centre for International Agricultural Research (ACIAR) also contributed funds towards this project.
References Afshar, A., 1994. Bluetongue: laboratory diagnosis. Comp. Immunol. Microbiol. Infect. Dis. 17, 221 – 242. Foster, N.M., Luedke, A.J., Metcalf, H.E., 1972. Bluetongue
46
M. Hosseini et al. / Journal of Virological Methods 75 (1998) 39–46
in sheep and cattle: efficacy of embryonating chicken eggs in viral isolations. Am. J. Vet. Res. 33, 77–81. Gard, G.P., Kirkland, P.D., 1993. Bluetongue virology and serology. In: Corner L.A., Bagust T.J. (Eds.), Australian Standard Diagnostic Techniques for Animal Diseases. CSIRO for the Standing Committee on Agriculture and Resource Management, East Melbourne, Australia. Goldsmit, L., Barzilai, E., Tadmor, A., 1975. The comparative sensitivity of sheep and chicken embryos to bluetongue virus and observations on viraemia in experimentally infected sheep. Aust. vet. J. 51, 190–196. Hawkes, R.A., Roehrig, J.T., Hunt, A.R., Moore, G.A., 1989. Antigenic structure of the Murray Valley encephalitis virus E glycoprotein. J. Gen. Virol. 69, 1105–1109. Kleczkowski, A., 1968. Experimental design and statistical methods of assay. In: Maramorosch, K., Koprowski, H. (Eds.), Methods in Virology, vol. 4. Academic Press, New York, pp. 705 – 713. Lunt, R.A., White, J.R., Blacksell, S.D., 1988. Evaluation of a monoclonal antibody blocking ELISA for the detection of group-specific antibodies to bluetongue virus in experimental and field sera. J. Gen. Virol. 69, 2729–2740.
McColl, K.A., Gould, A.R., 1991. Detection and characterisation of bluetongue virus using the polymerase chain reaction. Virus Res. 21, 19 – 34. Mecham, J.O., 1993. Detection of bluetongue virus from blood of infected sheep by use of an antigen-capture enzyme-linked immunosorbent assay after amplification of the virus in cell culture. Am. J. Vet. Res. 54, 370 – 372. Mecham, J.O., Dean, V.C., Wigington, J.G., Nunamaker, R.A., 1990. Detection of bluetongue virus in Culicoides 6ariipennis by antigen capture ELISA assay. J. Med. Entomol. 27, 602 – 606. Roberts, P.C., Etchison, J.R., Bond, C.W., 1988. A rapid, quantitative assay for titration of bovine virus diarrhoeamucosal disease virus. Vet. Microbiol. 18, 209 – 217. Stanislawek, W.L., Lunt, R.A., Blacksell, S.D., Newberry, K., Hooper, P.T., White, J.R., 1996. Detection by ELISA of bluetongue antigen directly in the blood of experimentally infected sheep. Vet. Microbiol. 52, 1 – 12. Williams, C.F., 1994. Studies on the Australian bluetongue viruses. Master of Science in Veterinary Science Thesis, University of Sydney, Australia.
. .