measuring infectious bursal disease virus rna in blood by multiplex real-time quantitative rt-pcr

Journal of Virological Methods 85 (2000) 55–64

Measuring infectious bursal disease virus RNA in blood bymultiplex real-time quantitative RT-PCR

Adrian Moody *, Scott Sellers, Nat BumsteadInstitute for Animal Health, Compton, Newbury, Berkshire, RG20 7NN, UK

Received 5 July 1999; received in revised form 4 October 1999; accepted 4 October 1999

Abstract

A quantitative reverse transcription polymerase chain reaction (RT-PCR) protocol for assessing infectious bursaldisease virus (IBDV) RNA levels in blood was developed using the ABI PRISM™ 7700 Sequence Detection Systemcoupled with TaqMan® chemistry. To control for variations in sampling and processing between samples 28S rRNAwas co-amplified in a multiplex reaction and used to quantify total RNA. Relative quantification and standardisationwas achieved using a log10 dilution series of RNA extracted from IBDV stock. A linear relationship was observedbetween input RNA and cycle threshold values (CT) over 5 log10 dilutions for the IBDV-specific product and 6 log10

dilutions for the 28S rRNA-specific product. As a test of the assay it was used to determine whether differences insusceptibility to IBDV observed between inbred lines of chickens could be detected at the level of viral load in theblood. Viral RNA levels peaked 2 days post-infection when there was significantly less viral RNA in the blood ofresistant line 61 chickens compared with the more susceptible Brown Leghorns (P=0.01). These results demonstratethat the course of IBDV infection can be monitored by quantifying IBDV RNA extracted from blood of infectedchickens using TaqMan® technology. © 2000 Elsevier Science B.V. All rights reserved.

Keywords: TaqMan; Multiplex; Reverse transcription; PCR; IBDV; Blood

www.elsevier.com/locate/jviromet

1. Introduction

Infectious bursal disease virus (IBDV) is a smalldouble stranded RNA virus belonging to the fam-ily Birna6iridae (Kibenge et al., 1988). IBDV is animportant pathogen of commercial chickenswhere it can cause acute infections of B-lymphocytes leading to immune suppression andmortality. Generally the severity of IBDV infec-

tions has been assessed in terms of mortality orthe degree of bursal damage, and it has beendifficult to assess viral load because virulentstrains of IBDV do not replicate in tissue culture.RT-PCR has been widely used to detect IBDV(Wu et al., 1992; Lee et al., 1994; Qian andKibenge, 1994; Stram et al., 1994; Tham et al.,1995; Jackwood and Nielsen, 1997) and offers apossible route to quantifying IBDV levels(Clementi et al., 1995). A quantitative competitivePCR (QC-PCR) assay has been developed tomonitor IBDV RNA extracted from infected bur-

* Corresponding author. Tel.: +44-1635-577-266; fax: +44-1635-577-237.

0166-0934/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved.

PII: S 0166 -0934 (99 )00156 -1

A. Moody et al. / Journal of Virological Methods 85 (2000) 55–6456

sae (Wu et al., 1997), however this protocol islabour intensive and the technique has limitations(Souaze et al., 1996).

Difficulties in using the RT-PCR technique forquantification include variability in the collectionand processing of samples and variability in re-verse transcription. Variations introduced at thesestages are greatly amplified by the PCR and de-crease reproducibility (Freeman et al., 1999). Re-cently the ABI PRISM™ 7700 SequenceDetection System coupled with TaqMan® chem-istry has been shown to provide a rapid, sensitivemethod for quantifying nucleic acids (Gibson etal., 1996; Heid et al., 1996; Desjardin et al., 1998).The system detects increases in fluorescence dur-ing each cycle of the PCR generated by the hy-drolysis of a dual fluorescently-labelled sequencespecific probe, eliminating post-PCR handling ofsamples for quantification. As the system canmeasure more than one labelled probe per well itallows simultaneous PCR of unrelated target se-quences, making it possible to include amplifica-tion of a control RNA species in the samereaction.

An objective of this work was to assist in thegenomic mapping of genes associated with resis-tance to IBDV by providing a definable level ofthe severity of infection. Specific-pathogen-freeinbred lines of chickens vary in their ability torespond to challenge with IBDV as monitored bymortality and histological destruction of the bursaof Fabricius (Bumstead et al., 1993). The presentstudy was devised to see if it was possible todetect and quantify IBDV RNA levels in bloodsamples, to use this to plot a time course of viralinfection, and to see if differences between inbredlines of chickens infected with IBDV could bedetected in viral load in blood using TaqMan®

technology in a multiplex RT-PCR.

2. Materials and methods

2.1. Chickens

Two of the inbred lines of chickens maintainedat the Institute for Animal Health (IAH) wereused: the IAH Brown Leghorn line, which is

highly susceptible to IBDV, and line 61, which isresistant to IBDV (Bumstead et al., 1993).

2.2. Virus stock

Viral stocks were prepared from twenty chick-ens inoculated intra-nasally with a 1:10 dilution ofpurified IBDV strain F52/70 (Bygrave andFaragher, 1970) which were killed three dayspost-infection (DPI). Bursae were removed asepti-cally and homogenised in ice-cold phosphatebuffered saline (PBS) (1 ml/bursa) and the super-natant centrifuged at 1000×g for 10 min at 4°Cto remove remaining cellular material. The result-ing supernatant was stored at −70°C with anembryo infectious dose50 of 3×106/ml.

2.3. Animal experiments

Chickens were infected intra-nasally at 21 dayswith 100 ml (50 ml in each nostril) of virus stockdiluted 1:10 in PBS, each bird receiving a total of3×104 EID50 of virus. For RNA preparationblood samples were taken from a wing vein imme-diately into sodium citrate (3%). Samples weretaken prior to infection and daily thereafter. Twoanimal experiments were carried out using fivebirds from each pure line in the first (experiment1) and six birds from each pure line in the second(experiment 2). Both lines of chicken were in-fected and reared together in disease-secure ac-commodation at IAH and examined frequentlythroughout the infection.

2.4. Isolation of RNA

Total RNA was prepared from whole blood (50ml blood in 50 ml 3% sodium citrate) or virus stock(50 ml) using the S.N.A.P.™ Total RNA Isolationkit (Invitrogen) following the manufacturers in-structions. Briefly, sterile RNAase-free TE pH 7.5(10 mM Tris, 0.1 mM EDTA) was added to givean initial volume of 150 ml to which 450 ml ofbinding buffer and 125 ml proteinase K (20 mg/ml) was added and incubated at 37°C for 90 min.To each sample 300 ml of isopropanol was added,the solution mixed and applied to a S.N.A.P™Total RNA column and spun on a bench cen-

A. Moody et al. / Journal of Virological Methods 85 (2000) 55–64 57

trifuge (MicroSpin 12, Du Pont) at maximumspeed for 1 min. The columns were washed twicewith 600 ml NaCl (100 mM) and centrifuged for 2min until dry. The bound nucleic acid was elutedin 135 ml RNAase-free water and incubated at37°C for 10 min with 15 ml 10× DNAase bufferand 1 ml (2 units) RNAase-free DNAase: 450 ml ofbinding buffer and 300 ml of isopropanol wereadded to the samples, mixed and applied to asecond column. The column was centrifuged asbefore for 1 min, washed twice with 600 ml NaCl(100 mM) and spun for 2 min until dry. RNAbound to the resin was eluted in 125 ml ofRNAase-free water and stored at -70°C.

2.5. RT-PCR conditions

For both IBDV and 28S rRNA-specific amplifi-cation primers and probes were designed using thePrimer Express software programme (PE AppliedBiosystems). IBDV-specific oligonucleotides wereidentified from genomic segment A (Bayliss et al.,1990). The fluorescently labelled probe runs fromnucleotide 1433 to 1458 (5%–3%) and is labelledwith the reporter dye 5-carboxyfluoroscein (FAM)at the 5% end and the quencher N,N,N’,N’–te-tramethyl-6-carboxyrhodamine (TAMRA) at the3% end. Specific primers were designed to closelyflank the probe and to amplify a 69 bp product.Primer and probe sequences were:IBDV probe: 5%-(FAM)-TCCCCTGAAGATT-GCAGGAGCATTTG-(TAMRA)-3%IBDV for-ward primer: 5%-GAGGTGGCCGACCTCAACT-3%IBDV reverse primer: 5%-AGCCCGGATTAT-GTCTTTGAAG-3%

Three 28S rRNA-specific oligonucleotides weresimilarly derived from a chicken 28S rRNA se-quence (Genbank accession number X59733). Theprobe was labelled with the fluorescent reporterdye VIC (PE Applied Biosystems) on the 5% endand the quencher N,N,N’,N’-tetramethyl-6-car-boxyrhodamine (TAMRA) on the 3% end. Specificprimers were designed to flank closely the probeand amplified a 61 bp product. Primer and probesequences were:28S rRNA probe: 5%-(VIC)-AGGACCGCT-ACGGACCTCCACCA-(TAMRA)-3%28S rRNAforward primer: 5%-GGCGAAGCCAGAG-

GAAACT-3%28S rRNA reverse primer: 5%-GACGACCGATTTGCACGTC-3%

Multiplex RT-PCR was carried out usingreagents from the TaqMan® EZ RT-PCR kit (PEApplied Biosystems). Briefly, the RT-PCR mix-ture (25 ml) consisted of the following: 1× EzRT-PCR buffer [including 60 mm ROX (6-car-boxy-x-rhodamine; a fluorescent reference dye)], 3mM manganese acetate, 300 mM dATP, dCTP,dGTP, 600 mM dUTP, 200 hM of each IBDV-specific primer, 600 hM of each 28S rRNA-primer, 100 hM of each probe, 0.1 units rTthpolymerase, 0.01 units AmpErase™ UNG (uracil-N-glycosylase), 2.75 ml water and 5 ml of totalRNA. Amplification and detection of specificproducts were undertaken using the ABIPRISM™ 7700 Sequence Detection System withthe following cycle profile: 1 cycle of 50°C for 2min, 1 cycle of 96°C for 5 min, 1 cycle of 60°C for30 min, 1 cycle of 95°C for 5 min and 40 cycles of94°C for 20 sec and 59°C for 1 min. Singlecomponent RT-PCR was carried out in the sameway except the volumes of 28S rRNA primers andprobes were replaced with water for the IBDV-specific reaction and visa versa for the 28S rRNAreaction.

2.6. Detection and quantification of PCR products

Quantification was based on the increasedfluorescence detected by the ABI PRISM™ 7700Sequence Detection System (PE Applied Biosys-tems) due to hydrolysis of the target-specificprobes by the 5% nuclease activity of the rTthDNA polymerase during PCR amplification. Apassive reference dye ROX (present in the EZreaction buffer), which is not involved in amplifi-cation, was used to correct for fluorescent fluctua-tions resulting from changes in the reactionconditions for normalisation of the reporter sig-nal. Results are expressed in terms of thethreshold cycle value (CT), the cycle at which thechange in the reporter dye (DRn) passes a signifi-cance threshold. In this work the threshold valuesof DRn were taken as 0.05 for IBDV-specificproducts and 0.075 for 28S rRNA-specific prod-ucts for all reactions described.


To generate standard curves for the IBDV and28S rRNA-specific reactions total RNA extractedfrom virus stock was serially diluted in sterileRNAase-free water and dilutions made from10−1 to 10−6. Each RT-PCR experiment con-tained three no template controls, test samplesand a log10 dilution series. Each experiment wasrepeated in triplicate, with each replicate per-formed on a different day.

2.7. Statistical analysis

Regression analysis of the mean values of eightreplicate multiplex RT-PCRs for the log10 dilutedRNA was used to generate standard curves. Thecoefficient of variance for these eight replicateswas calculated for both the IBDV and 28S rRNA-specific reactions. The reproducibility of the assaywas tested by analysis of variance (ANOVA)comparing repeat runs of samples, and mean val-ues generated at individual time points were com-pared by Student t-test.

3. Results

3.1. Standard cur6es

Replicate measurements on different days werehighly repeatable with a coefficient of variationfor eight replicate multiplex RT-PCRs of log10

serially diluted RNA of less than 2.2% for theIBDV-specific reaction (Table 1).

A linear relationship was observed between theamount of input RNA and the CT values forIBDV-specific product over five log10 dilutions,from a dilution of 10−1 (CT value 21) to 10−5 (CT

value 40) (Fig. 1A). The 28S rRNA-specific reac-tion showed a linear relationship over six log10

dilutions from a CT value of 11 (10−1 dilution) toa CT value of 30 (10−6 dilution) (Fig. 1B). Re-gression analysis of the CT values generated bythe log10 dilution series produced R2 values forboth reactions in excess of 0.97. The slope of theregression lines indicates an increase of 4.6 cyclesper log10 decrease in input viral RNA and 3.9cycles per log10 decrease in input 28S rRNA (Fig.1). Reaction efficiencies calculated from the re-gression analyses were 0.78 for the 28S rRNA-specific reaction and 0.56 for the viral-specificreaction.

Analysis of variance showed no significant dif-ference in the IBDV-specific CT values betweensingle and multiplex reactions on experimentalsamples (P=0.67) and when regression analysisof the log10 dilution series was performed singleand multiplex IBDV-specific reaction generatedsimilar gradients, 4.3 and 4.6, respectively (Fig. 2).This demonstrates that the multiplex reaction didnot significantly affect the quantification of IBDVRNA using this method.

3.2. Standardisation

To control for variation in sampling and RNApreparation the CT values for IBDV-specific

Table 1Values for eight replicate assays of a standard series of log10 dilutions of total RNA extracted from virus stocka

Dilution of RNA CT for eight replicates of IBDV standard curve

1 2 43 5 % cvbSDMean876

10−1 1.720.3721.2221.0820.9220.9921.0221.1221.0421.7621.8524.25 24.79 24.19 24.15 24.4110−2 24.12 24.24 24.7 24.39 0.26 1.06

28.9428.5 28.60 0.36 1.2728.73 28.76 28.05 24.11 28.63 29.0610−3

32.78 0.71 2.1510−4 32.65 31.69 32.54 32.68 32.13 33.67 33.19 33.69404040 0.190.0739.97404039.79 4010−5 40

a Cycle threshold values shown are for IBDV-specific primer products from a multiplex reaction that also included 28SrRNA-specific primers.

b cv, Coefficient of variation.


Fig. 1. Regression of CT values on total RNA extracted from IBDV stock log10 diluted in RNAase-free water. (A) Regression ofIBDV specific product, CT=4.589(log10 dilution)+15.625: R2 0.98 (B) Regression of 28S rRNA specific product CT=3.88(log10

dilution)+5.912: R2 0.99.

product for each sample were standardised usingthe CT value of 28S rRNA product for the samesample from the multiplexed reaction. The CT

values for 28S rRNA did not alter significantlyover the course of the disease; for example, theaverage 28S rRNA CT values for six BrownLeghorn chickens in experiment 2 were 15.46,15.57 and 15.31 on day 0, day 1 and day 2post-infection, respectively. Over the correspond-ing period the mean 1BDV RNA specific CT valuedecreased from 40 to 24.5. The CT values for 28SrRNA thus appeared to be independent of viral

infection and were taken to be representative ofthe level of RNA extracted from the 50 ml bloodsample. To normalise RNA levels between sam-ples within an experiment the mean CT value for28S rRNA-specific product for each line ofchicken in each experiment was calculated bypooling values from all samples for that line(Table 2). Tube to tube variations in 28S rRNACT values about these experiment/line specificmeans were calculated and the slope of the 28SrRNA log10 dilution series regression line wasused to calculate differences in input RNA. Using


the slope of the IBDV log10 dilution series regres-sion the difference in input total RNA, as repre-sented by 28S rRNA, was then used to adjustIBDV-specific CT values.

3.3. Comparison of 6iral RNA le6els in line 61

and Brown Leghorn chickens

In experiments to compare viral levels duringinfection there were no significant differences be-tween replicate RT-PCRs on the same set ofsamples with correlations between replicatesgreater than 98%. The 28S rRNA mean CT valueswere closely similar at different time points forline 61 and Brown Leghorn chickens throughouteach experiment (Table 2).

The effect of standardising IBDV-specific CT

values to correct for tube to tube variation inRNA levels is shown (Fig. 3). The largest effectwas at 2 DPI in the Brown Leghorn chickens inexperiment 1 (Fig. 3A). The CT value, correctedfor RNA input, is four cycles higher (representingnearly tenfold more viral RNA) than the uncor-rected value. At this time point two of the five 28SrRNA-specific CT values differed from the meanby 4.5 and 7 cycles presumably due to errors inthe collection of blood or RNA preparation.However, standardisation did not dramatically al-ter the distribution of the results as a whole (Fig.3) with IBDV levels peaking 2 DPI in all chickens(Fig. 4). There was significantly less IBDV RNAin the blood of resistant line 61 birds compared to

Fig. 2. Regression lines for single (+ ---+ ) and multiplex (x—x) RT-PCR amplification using total RNA extracted from IBDVstock log10 diluted in RNAase free water. The regression equation for the single reaction was CT=4.27(log10 dilution)+17.162:R2=0.99. The regression equation for the multiplex reactions was CT=4.589(log10 dilution)+15.625: R2=0.98.

Table 2Mean cycle threshold (CT) values for 28S rRNA-specific product generated by pooling values from all birds of each pure line withinan experiment, for all sample days, to achieve an overall value for the average 28S rRNA level in 50 ml of blooda

Mean Standard error 99% Confidence limit95% Confidence limit

Experiment 1Line 61 15.51–17.3516.43* 15.18–17.680.44

15.55–17.41Brown leghorn 15.22–17.7416.48* 0.45

Experiment 215.42** 0.26 14.89–15.95 14.71–16.13Line 61

0.22Brown leghorn 15.04–16.2815.21–16.1115.66**

a The standard error of the mean and confidence limits for the appropriate degrees of freedom are shown. A Student t-test wascarried out to determine significance *P=0.97, **P=0.57.


Fig. 3. Quantification of viral RNA in 50 ml of whole blood extracted from chickens over the first 4 days following infection withIBDV F52/70. Cycle threshold values are expressed subtracted from 40 (the negative end-point), higher values represent higher levelsof viral RNA. Solid blocks represent viral RNA specific CT values before standardisation for input RNA. Lines representstandardised values for viral RNA corrected for variation in input RNA measured by 28S rRNA levels. (A) Brown Leghorn birdsin experiment 1 (B) Line 61 birds in experiment 1 (C) Brown Leghorn birds in experiment 2 (D) Line 61 birds in experiment 2.

Brown Leghorn chickens at 2 days after infection(P=0.009 and P=0.01 in experiment 1 and ex-periment 2, respectively, Fig. 4).

3.4. Relationship between 6iral le6els andmortality

No mortality was observed in experiment 1, but5/6 Brown Leghorn chickens died in experiment 2before sampling at 3 DPI. This difference in mor-tality does not appear to be due to a difference inthe amount of IBDV RNA in the two experimentsas these were closely similar at 2 DPI (P=0.3).The one surviving Brown Leghorn chicken inexperiment 2 had less IBDV-specific RNA 2 DPIthan any chicken that died, but this difference wasnot statistically significant (P=0.09). These re-sults indicate that assessing viral load alone is notsufficient to determine susceptibility within theBrown Leghorn population. However, the peakIBDV load does correspond with the onset ofmortality and the most extreme clinical signs.

4. Discussion

Quantitation of IBDV infections has proveddifficult because virulent IBDV’s do not grow intissue culture. Consequently, assessment of theseverity of infection has relied on clinical signs,mortality or detectable changes recorded at theend of the experiment, such as histological exami-nation of tissue samples (Henry et al., 1980). Suchsystems do not provide a continuous, quantitativescale of measurements and are unlikely to accu-rately reflect the quantity of IBDV present.

The use of classical molecular methods for de-tecting IBDV RNA in clinical samples, such asNorthern blotting, will suffer from a lack of sensi-tivity; while PCR based methodologies are recog-nised as being the most sensitive moleculartechniques (Clementi et al., 1995). In a previousstudy IBDV RNA isolated from infected bursaewas determined in 100-fold range using QC-PCR(Wu et al., 1997). However, this approach stillrelied on samples taken at the end of the experi-


ment and did not allow continuous monitoring ofIBDV load.

Extracting RNA from small volumes of bloodallowed repeated sampling of the same bird en-abling us to determine a time course for IBDVinfection. In addition the RT-PCR assay de-scribed here eliminates the processing steps associ-ated with end-point PCR measurements, makingthe assay faster and more robust with fewer possi-ble sources of tube to tube variation. Relativequantification was achieved by simultaneous am-

plification of a log10 dilution series of standardRNA in parallel to the unknown samples forevery RT-PCR. Cycle threshold values generatedby the log10 dilution series were highly repro-ducible in each of eight different experimentalRT-PCRs demonstrating the accuracy of the as-say over a wide range of dilutions (Table 1).These results, coupled with the high reproducibil-ity of replicate runs of test samples, suggest theassay reflects accurately differences in the amountof input RNA.

Fig. 4. Quantification of viral RNA in 50 ml of whole blood extracted from chickens over the first 4 days following infection withIBDV F52/70. Cycle threshold values are expressed as subtractions from 40 (the negative end-point), and higher values representhigher levels of viral RNA. Brown Leghorn birds. Line 61b. There is a significant difference in the level of RNA detected 2 daysafter infection. (A) Experiment 1 *P=0.009. (B) Experiment 2 **P=0.01.


The present technique, in common with previ-ous techniques, will measure both viral mRNAand genomic RNA. It seems likely that levels ofgenomic and mRNA will increase broadly in par-allel, but the much greater abundance of mRNAwill dominate the reaction contributing to thesensitivity of the assay. It may be possible toadapt the assay to detect both species indepen-dently by prior separation of viral capsids or bycarrying out a two-stage reaction including a sep-arate RT step using only the negative strand astemplate.

Possible variations in sampling and pre-RT-PCR fluctuations in RNA levels were addressedby co-amplifying 28S rRNA from the RNA sam-ples as an internal standard. The inclusion of this28S rRNA amplification did not alter the IBDV-specific reaction values (Fig. 2). Ribosomal RNAwas chosen as an internal standard because of itsabundance and likely stability during the courseof an IBDV infection and the high levels of 28SrRNA and the lack of change over infection seenin these experiments confirm that this is the case(Table 2). This standardisation process is applica-ble to other systems in which total RNA is ex-tracted as it rapidly provides information on thequantity and quality of the RNA sample.

To minimise the variation generated by multi-plex PCR the two reporter dyes with the largestdifference in emission maxima were used (PEApplied Biosystems User Bulletin c5). The twodyes are not perfectly spectrally resolved however,and mis-assignment of fluorescenes caused a lowlevel of IBDV-specifc signal to be detected at 0DPI (Fig. 3A and B).

The assay is rapid; it was possible to go from ablood sample to a CT value in less than one day,it requires relatively little manual input and allowslarge numbers of samples to be analysed in paral-lel. The oligonucleotides described have also beenused successfully to detect non-virulent IBDVPBG98 (data not shown), despite some sequencedisparity between strains in the primer and probebinding region (Bayliss et al., 1990). Such cross-reactivity could limit costs by negating the need tohave unique oligonucleotides for each viral strainwhen assessing viral challenge in isolation. But, aswith any PCR, careful design of specific oligonu-

cleotides will be required for studies in morecomplex environments.

The assay has a broad range of possible appli-cations both in vitro and in vivo: these include itsuse in studies on therapeutic agents, such as re-combinant viral proteins or vaccines, allowingeffectiveness to be analysed in terms of controllingIBDV at the molecular level. In addition it wouldbe possible to adapt the assay to simultaneouslydetect vaccine and pathogenic strains of IBDV,where strain-specific sequences are available, al-lowing comparison between replication of vaccinevirus and pathogenic virus over the course ofsuccessive infections.

The RT-PCR assay should also provide a morehumane alternative for assessing susceptibility toIBDV infection, since IBDV load in the bloodcould replace clinical signs and death as a mea-sure of susceptibility or protection.

Acknowledgements

We would like to thank Dr Adriaan van Loon(Intervet) for calculating the EID50 of the virusstock and Dr Jose F. Rodriguez for helpful sug-gestions. We would also like to thank Dr PeteKaiser and Dr Shane Burgess for their criticalinput in the preparation of this manuscript, andwe are especially indebted to Don Hooper and hisstaff, Jillian Ewers and Dr Lonneke Vervelde fortheir contribution to the animal experiments. Thisresearch was carried out with the financial sup-port of the BBSRC and EU programme FAIR3PL1502 CT96.

References

Bayliss, C.D., Spies, U., Shaw, K., Peters, R.W., Papageor-giou, A., Muller, H., Boursnell, M.E.G., 1990. A compari-son of the sequence of segment A of four infectious bursaldisease virus strains and identification of a variable regionin the VP2. J. Gen. Vir. 71, 1303–1312.

Bumstead, N., Reece, R.L., Cook, J.K.A., 1993. Genetic dif-ferences in susceptibility of chicken lines to infection withInfectious Bursal Disease Virus. Poult. Sci. 72, 403–410.

Bygrave, A.C., Faragher, J.T., 1970. Mortality associated withGumbro disease. Vet. Rec. 86, 758–759.


Clementi, M., Menzo, S., Manzin, A., Bagnarelli, P., 1995.Quantitative molecular methods in virology. Arch. Vir.140, 1523–1539.

Desjardin, L.E., Chen, Y., Perkins, M.D., Teixeira, L., Cave,M.D., Eisenach, K.D., 1998. Comparison of the ABI 7700System (TaqMan) and competitive PCR for quantificationof IS6110 DNA in sputum during treatment of Tuberculo-sis. J. Clin. Microbiol. 36 (7), 1964–1968.

Freeman, W.M., Walker, S.J., Vrana, K.E., 1999. QuantitativeRT-PCR: Pitfalls and potential. BioTechniques 26, 112–125.

Gibson, U.E.M., Heid, C.A., Williams, P.M., 1996. A novelmethod for real time quantitative RT-PCR. Genome Res.6, 995–1001.

Heid, C.A., Stevens, J., Livak, K.J., Williams, P.M., 1996.Real time quantitative PCR. Genome Res. 6, 986–994.

Henry, C.W., Brewer, R.N., Edgar, S.A., 1980. Studies onInfectious Bursal Disease in Chickens. 2. Scoring micro-scopic lesions in the bursa of Fabricius, thymus, spleen,and kidney in gnotobiotic and battery reared whiteleghorns experimentally infected with infectious bursal dis-ease virus. Poult. Sci. 59, 1006–1017.

Jackwood, D.J., Nielsen, C.K., 1997. Detection of infectiousbursal disease viruses in commercially reared chickens us-ing the reverse transcriptase/polymerase chain reaction-re-striction endonuclease assay. Avian Dis. 41, 137–143.

Kibenge, F.S.B., Dhillon, A.S., Russell, R.G., 1988. Biochem-

istry and immunology of infectious bursal disease virus. J.Gen. Virol. 69, 1757–1775.

Lee, L.H., Ting, L.J., Shien, J.H., Shien, H.K., 1994. Singletube, noninterrupted reverse transcription-PCR for detec-tion of infectious bursal disease virus. J. Clin. Microbiol.32 (5), 1268–1272.

PE Applied Biosystems User Bulletin c5: http://www2.perkin-elmer.com/ab.

Qian, B., Kibenge, F.S.B., 1994. Observations on polymerasechain reaction amplification of infectious bursal diseasevirus dsRNA. J. Virol. Methods 47, 237–242.

Souaze, F., Ntodou-Thome, A., Tran, C.Y., Rostene, W.,Forgez, P., 1996. Quantitative RT-PCR: Limits and Accu-racy. BioTechniques 21, 280–285.

Stram, Y., Meir, R., Molad, T., Blumenkranz, R., Malkinson,M., Weisman, Y., 1994. Applications of the polymerasechain reaction to detect infectious bursal disease virus innaturally infected chickens. Avian Dis. 38, 879–884.

Tham, K.M., Young, L.W., Moon, C.D., 1995. Detection ofinfectious bursal disease virus by reverse transcription-polymerase chain reaction amplification of the virus seg-ment A gene. J. Virol. Methods 53, 201–212.

Wu, C.C., Lin, T.L., Zhang, H.G., Davis, V.S., Boyle, J.A.,1992. Molecular detection of infectious bursal disease virusby polymerase chain reaction. Avian Dis. 36, 221–226.

Wu, C.C., Lin, T.L., Akin, A., 1997. Quantitative competitivepolymerase chain reaction for detection and quantificationof infectious bursal disease virus cDNA and RNA. J.Virol. Methods 66, 29–38.

.

measuring infectious bursal disease virus rna in blood by multiplex real-time quantitative rt-pcr

Documents