WO2010126351A1 - Molecular differentiation of infectious bursal disease virus (ibdv) strains - Google Patents

Abstract

This invention relates to the detection and distinguishing Infectious Bursal Disease Virus (IBDV) strains by a fluorescent probe based real-time polymerase chain reaction in chicken or other birds. More particularly, this invention relates to distinguishing different Infectious Bursal Disease Virus (IBDV) strains in chicken and other bird sample by novel subtype specific primers and fluorescent probe based on one-tube duplex Real-time Polymerase Chain Reaction (PCR) method.The PCR conditions are optimized in order to obtain optimum PCR parameters on the ingredients and profiles using samples containing IBDV RNA in a Taqman probe based duplex real-time PCR. Hence, for differentiation of very virulent from vaccine strains of IBDV by using Primer FWDC (nucleotide position: 2084 to 2102) and RVSC (nucleotide position: 2178 to 2197) were designed from the conserved region of VP4 of both very virulent and classical strains, respectively, to generate a 1 14 bp amplicon. A dual-labeled fluorescent probe FAM 5'- TAMRA 3', (nucleotide position: 2112 to 2133), was designed with the sequence specific to aligned very virulent IBDV strains (Probe 1), and a second dual-labeled fluorescent probe HEX 5'-TAMRA 3', (nucleotide position: 21 12 to 2133), was designed with the sequence specific to aligned classical IBDV strains (Probe 2).

Description

MOLECULAR DIFFERENTIATION OF INFECTIOUS BURSAL DISEASE VIRUS

(IBDV) STRAINS

FIELD OF INVENTION

The present invention relates to the detection and distinguishing Infectious Bursal Disease Virus (IBDV) strains by a fluorescent probe based real-time polymerase chain reaction in chicken or other birds. More particularly, this invention relates to distinguishing different Infectious Bursal Disease Virus (IBDV) strains in chicken and other bird sample by novel subtype specific primers and fluorescent probe based on one-tube duplex Real-time Polymerase Chain Reaction (PCR) method.

BACKGROUND OF THE INVENTION

Infectious bursal disease (IBD) is an acute contagious viral disease of young chickens often known as Gumboro disease. The etiological agent, IBD virus (IBDV), has a predilection for the cells of the bursa of Fabricius where the virus infects actively dividing and differentiating lymphocytes of the B-cell lineage. Thus, IBD is a fatal immunosuppressive disease causing heavy losses to the poultry industry.

The first outbreak of IBDV was reported in commercial chicken flocks in Delaware, USA. The IBDV strains, which were isolated during the outbreak, now referred to as classical serotype I isolates. The disease was also first report in Europe in 1962. And from 1966 to 1974, IBD was reported in the Middle East, Southern and Western Africa, India, the Far East and Australia. In most cases, the IBDV strains that associated with the outbreaks were of low virulence and caused only 1 to 2 % of specific mortality. However, a new IBDV strain (antigenic variant) emerged and able to cause up to 5% specific mortality in USA. The antigenic variant was recovered from flocks with selection pressure of field vaccination against classical IBDV serotype I. Although being antigenic variant these isolates have only minor amino acid changes and do not form a separate serotype.

Nevertheless, these changes occur at the VP2 conformation-dependent antigenic epitopes that are responsible for stimulating virus neutralizing antibodies. Currently, variant form of IBD has been reported outside Central America particularly in countries such as China, South America and Australia.

Since variant IBDV causes only changes at the bursa and depending on the immune status of the chickens, the disease is often manifested with subclinical signs, it is difficult to detect variant IBDV in commercial flocks. Hence, variant IBDV may be common in many countries in the world but remains undiagnosed. A second serotype - serotype Il of IBDV was identified in 1987. Serotype Il IBDV isolates are apathogenic and are recovered mainly from turkeys.

In the 1990s, IBDV isolates, which were able to break through levels of maternal antibodies that normally were protective, were reported in Europe. These isolates, the so called very virulent IBDV are causing more severe clinical signs during an outbreak which mortality approaching 100% in susceptible flocks, and are now found almost world-wide. The emergence of very virulent strains of IBDV has complicated the immunization programs against the disease. Early vaccination may result in failure due to interference with the maternal antibody, whilst its delay may cause field virus infections. Currently, outbreaks of wlBDV have been reported throughout various countries in the world.

In designing an effective disease control program one should consider the diagnostic methods use to diagnose disease caused by infectious agent. Currently, IBD can be diagnosed based on virus isolation, electron microscopy, immunofluorescence, virus neutralization, monoclonal antibody assays, and/or enzyme-linked immunosorbent assay. However, these methods have one or more disadvantages such as time consuming, labour intensive, expensive and of low sensitivity.

Recently, the reverse transcriptase polymerase chain (RT-PCR) has been used to detect IBDV based on the amplification of the central hypervariable region of the VP2 region. Subsequently, RT-PCR assay followed by restriction fragment length polymorphism (RFLP) also has been used to detect and differentiate IBDV strains.

Although, this method able to differentiate different IBDV strains, it is not automated and time consuming. Both radioactive and non-radioactive based nucleic acid probes that can differentiate IBDV strains have been used in the detection of IBDV. However, apart for academic interest, their use in diagnosing IBD is uncommon.

Fluorescence-based real-time PCR assays have been developed to provide a rapid and sensitive method for quantifying nucleic acids. In this assay, reactions are monitored by the point in time during cycling when amplification of a PCR product is first detected rather than the amount of PCR product accumulated after a fixed number of cycles. There are currently 2 general approaches in real-time PCR depending on the types of fluorescence dyes. The simplest method uses fluorescent dye, SYBR Green I that bind specifically to double stranded DNA. The major problem with SYBR Green l-based detection is that non-specific amplifications cannot be distinguished from specific amplifications. However, specific amplification can be verified by melting curve analysis.

The other dyes (Taqman, Molecular Beacons, Scorpion) rely on the hybridization of fluorescence labeled probes to the correct amplicon. Accumulation of PCR products is detected by monitoring the increase in fluorescence of the reporter dye. The threshold cycle (Ct) is defined as the fractional cycle number at which the reporter fluorescence generated by the accumulating amplicons passes a fixed threshold above baseline.

Hence, a plot of the log of initial target copy number for a set of standards versus Ct is a straight line whereby the higher the starting copy number of the nucleic acid target, the sooner a significant increase in fluorescence is observed. It has also been established that primers combination playing an important role for the Ct value prediction, where the approach is similar to the analysis of single-nucleotide polymorphism. Thus, several recent studies have used SYBR Green I based real-time PCR to differentiate different serotypes or strains of organisms based on Ct and/or Tm values. Recently, it has been established that SYBR Green I based real-time PCR for differentiation of IBDV subtypes in particularly very virulent from classical vaccine strains. A quantitative real-time PCR assay based on TaqMan has been developed to detect IBDV. It has been reported that TaqMan based real-time PCR was shown to be able to detect vaccine and wild type IBDV strains in infected chickens. In all the published work on Taqman based real-time PCR for differentiation of IBDV subtypes the primers and probes were designed based on VP2 gene sequences. Only recently, there is a study on the used of VP4 based primers and probe in Taqman PCR for differentiation of IBDV subtypes. SUMMARY OF THE INVENTION

The biological material, which is to be investigated, was obtained in some suitable matter and isolated. By means of standardized methods, RNA was isolated from the material. A defined amount of RNA was transcribed into DNA by means of conserved primers in RT-PCR assay.

One of the most common fluorescent dyes that are used for probe labeling is Taqman probe. In real-time PCR, measurements are detected during the exponential phase of the reaction typically by obtaining the threshold cycle (Ct) value. Ct value can be defined as the fractional cycle number at which there is a significant increase in fluorescence above a specified threshold.

The Ct value is also proportional to the numbers of target copies present in the samples. Different IBDV strains can be differentiated based on the Ct values by using conserved primer combinations and a mixture of subtype specific probes that able to detect different IBDV subtypes. Hence, with the incorporation of internal gene (a house keeping gene) the relative amount of IBDV can be measured.

The PCR conditions are optimized in order to obtain optimum PCR parameters on the ingredients and profiles using samples containing IBDV RNA in a Taqman probe based duplex real-time PCR. Hence, for differentiation of very virulent from vaccine strains of IBDV by using Primer FWDC (nucleotide position: 2084 to 2102) and RVSC (nucleotide position: 2178 to 2197) were designed from the conserved region of VP4 of both very virulent and classical strains, respectively, to generate a 1 14 bp amplicon. A dual- labeled fluorescent probe FAM 5'-TAMRA 3', (nucleotide position: 2112 to 2133), was designed with the sequence specific to aligned very virulent IBDV strains (Probe 1 ), and a second dual-labeled fluorescent probe HEX 5'-TAMRA 3', (nucleotide position: 21 12 to 2133), was designed with the sequence specific to aligned classical IBDV strains (Probe 2).

For quantitative application of IBDV from dual infected sample, a housekeeping gene primers-probe was designed from Gallus gallus (chicken) beta-actin sequence, β-actin primers, BAFWD (nucleotide position: 806 to 825) and BARVS (nucleotide position: 865 to 886) and dual-labeled, fluorescent probe FAM 5'-TAMRA 3', (nucleotide position: 833- 853) (Probe 3), generated an 81 bp amplicon.

For detection and differentiation purpose of sample that is very virulent strain, primer FWDC & RVSC and probe 1 showed amplification. And no amplification is detected with classical vaccine strain.

For detection and differentiation purpose of sample that is classical strain, primer FWDC & RVSC and probe 2 showed amplification. And no amplification is detected with very virulent strain.

For detection and differentiation purpose of sample that is very virulent and classical strains (dual infected), primer FWDC & RVSC and probe 1 and 2 showed amplification.

For quantitative purpose of very virulent IBDV from sample that contained both very virulent and classical strains (dual infected), primer FWDC & RVSC and probe 1 and 3 showed amplification. Using the 2^"ΔΔCT method the relative amount of the very virulent IBDV load in the dual infected samples is established. For quantitative purpose of classical IBDV from sample that contained both very virulent and classical strains (dual infected), primer FWDC & RVSC and probe 2 and 3 showed amplification. Using the 2^"MCT method the relative amount of the classical IBDV load in the dual infected samples is established.

It is an objective of the invention to provide a new method for differentiating IBDV strains and for relative quantitation of two IBDV strains based on fold increase in virus load.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 shows the primer and probes designed based on sequence alignment of very virulent and classical IBDV VP4 gene published sequences from nucleotide position 2084 to 2197.

Figure 2a shows the gradient real-time RT-PCR tested on FAM-TAMRA dual-labeled probe for wlBDV detection with different annealing temperatures.

Figure 2b shows the gradient real-time RT-PCR tested on HEX-TAMRA dual-labeled probe for classical IBDV detection with different annealing temperatures.

Figure 2c shows the gradient real-time RT-PCR tested on FAM-TAMRA dual-labeled probe for β-actin, house-keeping gene detection with different annealing temperatures.

Figure 2d shows the gradient duplex real-time RT-PCR tested on both FAM and HEX dual-labeled probes with different annealing temperatures. Results showed specific amplification for wlBDV using FAM labeled-probe whilst no amplification with HEX labeled-probe.

Figure 3 shows the detection limit of Taqman real-time RT-PCR amplification determined with 10-fold serial dilution of UPM94/273 total RNA from 1000 to 0.01 ng/reaction.

Figure 4 shows a linear relationship between threshold cycle and serially diluted of UPM94/273 total RNA in log dilution (from 1000 to 0.1 ng/reaction). Figure 5 shows the detection limit of Taqman real-time RT-PCR amplification determined with 10-fold serial dilution of D78 total RNA from 1000 to 0.01 ng/reaction.

Figure 6 shows a linear relationship between threshold cycle and serially diluted of D78 total RNA in log dilution (from 1000 to 10 ng/reaction).

Figure 7 shows the detection limit of Taqman real-time RT-PCR amplification determined with 10-fold serial dilution of total RNA from 1000 to 0.01 ng/reaction.

Figure 8 shows a linear relationship between threshold cycle and serially diluted total RNA in log dilution (from 1000 to 10 ng/reaction). The standard curve was generated from amplification of β-actin housekeeping gene with each point represented the mean of the results.

Figure 9 shows the amplification plot of Taqman based duplex real-time RT-PCR for experimental trial dual-infection IBDV samples on day 3 post infection.

Figure 10 shows the amplification plot of Taqman based duplex real-time RT-PCR for experimental trial dual-infection IBDV samples on day 5 post infection.

Figure 11 shows the amplification plot of Taqman based duplex real-time RT-PCR for bursal samples tested positive for very virulent IBDV detection.

Figure 12 shows the amplification plot of Taqman based duplex real-time RT-PCR for bursal samples tested positive for classical IBDV strains detection. Figure 13 shows the amplification plot of Taqman based duplex real-time RT-PCR for dual positive IBDV strains detection.

Figure 14 shows a real-time RT-PCR products of the positive IBDV strains detected by agarose gel electrophoressis.

Figure 15a shows the amplification plot of Taqman based duplex real-time RT-PCR for bursal samples tested negative for IBDV strain detection.

Figure 15b shows the real-time RT-PCR products of the negative IBDV samples.

Figure 16a shows the predicted amino acid sequence alignment of IBDV isolates.

Figure 16b shows the predicted amino acid sequence alignment of IBDV isolates (continue).

Figure 17 shows the relative amount of both D78 and UPM94/273 in bursa of Fabricius.

Figures 18a and 18b show the viral load fold change for very virulent and classical strain in dual positive IBDV bursal samples.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

The field of the invention is detection of IBDV in various matrix (tissue samples from bursa of Fabricus, tissue samples from embryonated chicken eggs, reconstituted vaccine stock) using primer and subtype-specific probe mixtures in a single-tube reaction. Total RNA from test sample is reverse transcribed using a pair of conserved primer. The RT product is amplified using conserved primers with subtype-specific probe where by different IBDV strains can be differentiated based the detection of amplification. For quantitative purpose, the RT product is amplified using conserved primers with a subtype-specific probe and internal gene probe where by IBDV can be detected and quantified.

Since the concentration of RNA, PCR mixtures and condition influence the specificity of the assay, the optimum condition of the PCR is established. In addition, other parameters including concentration of total RNA, primers, MgCI2 and PCR profiles were also optimized. The designed primers and probes able to show amplification with wide range of annealing temperature where differentiation of very virulent and classical strains of IBDV can be performed in single tube format duplex reaction. Furthermore, the incorporation of an internal gene (house-keeping gene) such as β-actin, quantitation of virus load fold changes in dual infected samples based on relative quantitation method can be established.

Thus, this invention also includes novel real-time PCR-based assays which do not require size determination of the PCR amplification product to confirm the specific amplification of the IBDV target nucleic acid sequence. Therefore, the invention includes simple one-tube duplex format assays which simultaneously detect and differentiate 2 targets in a single reaction. Hence, it obviate the need for complex molecular biology techniques such as restriction enzyme digestion and sequencing to confirm that the amplification product is, indeed, of IBDV strains of very virulent or vaccine strains. The methods of the invention are, therefore, less prone to operator error, faster, and may be fully automated. In describing and claiming the present invention, the following terminology will be used in accordance with the definitions set out below.

As used herein, "nucleic acid," RNA" and similar terms also include nucleic acid analogs, i.e. analogs having other than a phosphodiester backbone. For example, the so-called "peptide nucleic acids," which are known in the art and have peptide bonds instead of phosphodiester bonds in the backbone, are considered within the scope of the present invention.

The term "very virulent" strain of IBDV means samples containing IBDV that associated with high mortality in chickens and with the following characteristic amino acid residues at the positions 222 (alanine), 242 (isoleucine), 256 (isoleucine) and 294 (isoleucine) of the VP2 region, at the positions 680 (tyrosine), 685 (asparagine), 715 (serine) and 751

(aspartate) of the VP4 region and at the positions 990 (valine) and 1005 (alanine) of the

VP3 region. Typically, a very virulent IBDV usually associated with mortality up to 100% in SPF chickens, 25% mortality in broilers and 60% in layers.

The term "vaccine" strain of IBDV means samples containing IBDV that do not cause mortality in non-vaccinated commercial chickens and with the following characteristic amino acid residues at VP2 region ; 253 (histidine), 279 (asparagine), and 284 (threonine) and associated with one or more changes at the serine residues of the heptapeptide region SWSASGS at the position 326 to 332. A vaccine strain also has characteristic amino acid residues at the positions 680 (cysteine), 685 (lysine), 715 (proline) and 751 (histidine) of the VP4 region and at the positions 990 (alanine) and 1005 (threonine) of the VP3 region. Typically, a vaccine strain is an attenuated classical IBDV derived from repeated passages in embryonated eggs and/or cell cultures.

As used herein, "effective/optimum amount" means an amount sufficient to produce a selected effect. For example, an effective amount of RNA is an amount sufficient to amplify a segment of nucleic acid by PCR provided that a DNA polymerase, buffer, template, and other conditions, including temperature conditions, known in the art to be necessary for practicing PCR are also provided.

The term "test sample" as used herein, means anything suspected of containing a target sequence. The test sample can be derived from any biological source and can be used (i) directly as obtained from the source or (ii) following a pre-treatment to modify the character of the sample. The pre-treatment that can be applied for example, disrupting cells, preparing liquids from solid materials, diluting viscous fluids, filtering liquids, concentrating liquids, inactivating interfering components, adding reagents, purifying nucleic acids, and the like. Typically, the test sample will be derived from bursal tissue samples.

A "target sequence" as used herein means a nucleic acid sequence that is detected, amplified, or otherwise is complementary to one of the primers herein provided.

The term "bp" means base pair. EXAMPLES

IBDV references isolates, experimentally infected IBDV samples and IBDV suspected samples.

The assay is tested using three type of samples, a reference very virulent IBDV strain (UPM94/273), a reference classical IBDV strain (D78), experimentally infected IBDV samples from SPF chicks and 37 bursal samples from suspected cases collected from year 2003 to 2005 (Table 1 ). The UPM94/273 has been characterized as vvlBDV based on sequence analysis of segment A and B and pathogenicity studies. The virus is propagated in nine to eleven day-old specific-pathogen-free (SPF) embryonated chicken eggs. The inoculated SPF eggs were placed horizontally into the incubator at 37⁰C and monitored daily for viability by candling. Embryos that had died within the first 24 hours were discarded and considered to be non-significant. Eggs with embryonic death after 24 hours were kept in the refrigerator for at least 2 hours and then examined for pathological changes under a biohazard cabinet. The CAM from dead embryo was then collected and processed using the procedure of processing of bursa tissue. The samples were then stored at -80⁹C until used.

A classical vaccine strain, D78 (Intervet, Holland) is used as the reference classical IBDV strain. The vaccine strain D78 was reconstituted with deionised water before proceeded for RNA extraction. The stocks were used immediately or stored at -80⁰C until used.

Bursal samples from experimentally infected SPF chickens with very virulent and/or classical vaccine strains (single and dual infection) were also used in this study. Briefly, one-day old single-comb White Leghorn SPF chickens were obtained from Veterinary Research Institute, Ipoh. The chicks were reared in an experimental isolation unit. Feed and water were provided ad libitum. A total of 187 three-week-old SPF chickens were allotted to four groups: Group A and B (40 chickens per group), Group C (52 chickens) and Group D (55 chickens). Each chicken from Group B and Group C was inoculated intraorally with 0.1 ml of inoculum containing 10⁴⁵ tissue culture infective dose 50 (TCID₅₀) of D78, 10⁴⁸ EID₅₀ of UPM94/273, respectively, and reared separately. In order to mimic dual infection with different IBDV strains, chickens from Group B were vaccinated intraorally with D78 followed by infected with very virulent IBDV strain UPM94/273 6 h later. Group A was left as control uninfected group. At 1 , 3 and 5 days post inoculation (p.i.), five chickens from each group were selected at random and scarified for bursal and blood samples collection. Bursal and blood samples were also collected from chickens that survived on day 10, 17, 24 and 30 after inoculation. For each group sampling, bursal sample was processed individually. The serum samples obtained from each group were stored at -20^QC and subsequently tested with ELISA.

In case of suspected IBDV cases, a total of 37 bursal samples from suspected cases collected from commercial chicken farms through Malaysia in year 2003 to 2005 were used. The samples are listed in Table 1 .

Nd: Not determine; Y: Yes RNA Extraction and Determination of the RNA and cDNA Concentration and Purity

The samples are processed and total RNA are extracted using commercial method (Trizol) as recommended by the manufacturer. The concentration and purity of the extracted total RNA and cDNA were measured at the wavelength of 260 nm and 280 nm using a spectrophotometer.

Primers and Probes Design

A set of primers that conserved to both vaccine and very virulent strains of IBDV and two different probes each specific for vaccine and very virulent strains were designed in this study (Table 2). A set of primers-probe also been designed for the house-keeping gene (Table 3). All the primers and probe were designed with the aid of Primer Premier 5.0 software and via the interface, Primer 3 (http://www.qenome.wi.mit.edu/cqi- bin/primer/primer3 www.cqi) .

The primers were designed based on the following criteria for real-time RT-PCR; primers should be designed to amplify short amplicon as possible. The length of the amplicon should not exceed 400 bp. Ideally the amplicon should be between 100 to 150 bp. Whenever possible, primers should be selected in a region with a G/C content of 20- 80%. The five nucleotides at the 3' end should have no more than two G and/or C bases. The probe is a dual-labeled with different reporter dyes at the 5' end (FAM for very virulent IBDV and HEX for vaccine IBDV) and quencher at the 3' end (TAMRA). The length of the probe should be 20 to 30 bases. The probe-template should have a melting temperature (Tm) that was 8^QC to 10^sC higher than the Tm for primer-template. For best results, a probe with a Tm of 68⁸C to 70^sC should be used. Beside that, a G at the 5' end next to the reporter and run of four or more of the same nucleotide, especially of G was avoided. The binding strand that the probe has more C than G bases was used.

The primers FWDC (nucleotide position: 2084 to 2102) and RVSC (nucleotide position: 2178 to 2197) were designed from the conserved region of VP4 of both very virulent and classical strains, respectively, to generate a 114 bp amplicon. A dual-labeled fluorescent probe FAM 5'-TAMRA 3', ProVV (nucleotide position: 21 12 to 2133), was designed with the sequence specific to aligned very virulent IBDV strains (Probe 1 ), and a second dual- labeled fluorescent probe HEX 5'-TAMRA 3', ProCL (nucleotide position: 2112 to 2133), was designed with the sequence specific to aligned classical IBDV strains (Probe 2). The housekeeping gene primers-probe was designed from Gallus gallus (chicken) beta- actin sequence. Beta-actin primers, BAFWD (nucleotide position: 806 to 825) and BARVS (nucleotide position: 865 to 886) and dual-labeled, fluorescent probe FAM 5'- TAMRA 3', BAPro (nucleotide position: 833-853), generated an 81 bp amplicon, respectively.

Table 2: Primers- robes 114 b for IBDV subt es detection

a The sequences of the primers were conserved when compared to both very virulent and classical strains of IBDV as stated below. ^b The sequences of the probe was conserved when compared to very virulent strains, UPM94/273 (AF527039), UPM97/61 (AF247006), OKYM (D49706), UK661 (X92760), IBDKS (L42284), D6948 (AF240686), BD3/99 (AF362776), Tasik94 (AF322444), Chinju (AF508176), HK46 (AF092943), SH95 (AY134874), and TO9 (AY099456). c The sequences of the probe was conserved when compared to classical strains, D78 (AF499929), Cu-1 (X16107), Cu-I M (AF362771), Cu-IWT (AF362747), P2 (X84034), CT (AJ310185), CEF94 (AF194428), HZ-2 (AF321054), STC (D00499) and IM (AY029166).

The nucleotide variations between the probes are indicated in bold.

Table 3: Primers- robe 81 b for -actin house-kee in ene

a The sequences of the primers and dual-labeled probe are based on reference strain Gallus gallus beta-actin mRNA, complete cds with Genbank accession number L08165.

As shown in Figure 1 showed the nucleotide sequence alignment of VP4 gene from 12 very virulent IBDV strains; UPM94/273 (AF527039), UPM97/61 (AF247006), OKYM (D49706), UK661 (X92760), IBDKS (L42284), D6948 (AF240686), BD3/99 (AF362776), Tasik94 (AF322444), Chinju (AF508176), HK46 (AF092943), SH95 (AY134874), TO9 (AY099456); and 10 classical IBDV strains; D78 (AF499929), Cu-1 (X16107), Cu-I M (AF362771 ), Cu-IWT (AF362747), P2 (X84034), CT (AJ310185), CEF94 (AF194428), HZ-2 (AF321054), STC (D00499) and IM (AY029166). Figure 1 also shows the primer and probes designed based on sequence alignment of very virulent and classical IBDV VP4 gene published sequences from nucleotide position 2084 to 2197. The forward and reverse primers were conserved for different IBDV subtypes and probe sequences for very virulent (highlighted in yellow colour) and classical (highlighted in green colour) was conserved with 12 very virulent and 10 classical IBDV isolates, respectively. Within the region targeted by the probe consensus residues are shown as asterisks.

Reverse Transcriptase

The extracted RNA was transcribed and amplified using primers FWDC RVSC and subtype-specific probes (Probe l and Probe 2). The optimum condition is established using reference IBDV strains, UPM94/273 and D78. One-step duplex fluorescent probe (Taqman) real-time RT-PCR assay was optimized using conserved primers and two specific fluorescent dual-labeled probes. Probe 1 labeled with FAM'5-TAMRA'3 was specific for very virulent IBDV strains detection; while probe 2 labeled with HEX'5- TAMRA'3 was specific for classical vaccine IBDV strains detection. Probe 3 labeled FAM'5-TAMRA-'3 was specific for β-actin detection was included and use as a housekeeping gene for normalization. Primers and probes were utilized in a 50 μl reaction containing 25 μl of 2X Quantitect Probe RT-PCR Mix [HotStarTaq® DNA Polymerase, QuantiTect Probe RT-PCR buffer, dNTP mix, 8 mM MgCI₂] (Qiagen, Valencia, CA), 0.5 μl of Quantitect RT Mix [Omniscript™ Reverse Transcriptase, Sensiscript® Reverse Transcriptase] (Qiagen, Valencia, CA), primers to a final concentration of 0.4 μM, each probe to a final concentration of 0.2 μM, 19.5 μl of RNase-free water, and 2 μl (1 to 2 μg) of RNA template. The reaction was conducted in a DNA Engine Opticon™ (BIO-RAD, USA). Target genes were amplified in low-profile 0.2 ml tube stripes (MJ Research, USA). The cycling programme consists of reverse transcription at 50^gC for 30 min, initial activation step at 95^QC for 15 min followed by 35 cycles of denaturation at 94^SC for 15 s, annealing temperature at 60^sC for 30 s, extension at 76^δC for 30 s, and plate read. The fluorescence threshold limit of the DNA Engine Opticon™ System was set at 0.02.

The condition of the real-time PCR was optimized using an annealing temperature from 60^sC to 65^SC was tested separately for each probe (Figure 2a, b, c, d). It was found that annealing temperatures within this range showed positive amplification but with different amplification efficiencies. However, the annealing temperature at 60^sC has the best amplification efficiencies and was found to fulfill the aim of the optimization assay which has the highest possible increase in fluorescence and has the lowest threshold cycle (CT) value (Tables 4 and 5). The optimal annealing temperature could amplify three probes simultaneously under the same real-time RT-PCR conditions.

Gradient real-time RT-PCR was performed in order to obtain an optimal annealing temperature for very virulent, FAM-TAMRA dual-labeled probe (Figure 2a), classical, HEX-TAMRA dual-labeled probe (Figure 2b), beta-actin, FAM-TAMRA dual-labeled probe (Figure 2c) and duplex (very virulent-classical) (FAM-TAMRA and HEX-TAMRA) (Figure 2d) probes. The annealing temperatures for each probe ranging from 60^sC to 65^BC were tested. It was found that annealing temperatures within this range showed positive amplification but with different amplification efficiencies. However, the annealing temperature at 60^sC has the best amplification efficiencies and was found to fulfill the aim of the optimization assay which has the highest possible increase in fluorescence and has the lowest threshold cycle (CT) value (Tables 4 and 5). The optimal annealing temperature could amplify three probes simultaneously under the same real-time RT- PCR conditions. Table 4. Threshold cycle values for gradient real-time RT-PCR tested on dual- labeled probes with different annealing temperatures.

NA: No Amplification

Table 5. Threshold cycle values for gradient duplex real-time RT-PCR tested on dual-labeled probes for very virulent IBDV with different annealing temperatures.

NA: No Amplification

Development of the Taqman based real-time PCR to detect very virulent and vaccine strains of IBDV

After establishing the optimum condition of the real-time PCR, the detection limit of the duplex assay were tested using 10-fold serial dilution of total RNA of the IBDV reference strains; very virulent (UPM94/273), classical (D78) and housekeeping gene (beta-actin) ranging from 1000 to 0.01 ng per reaction. The developed duplex real-time PCR assay has a linear relationship over five log₁₀ concentration range from a dilution of 10^'1 (1000 ng/reaction) to 10^~5 (0.1 ng/reaction) for very virulent strain amplification (Figures 3 and 4); whilst classical strain amplification (Figures 5 and 6) and β-actin house-keeping gene (Figures 7 and 8) showed a linear relationship over three log₁₀ concentration range from a dilution of 10^"1 (1000 ng/reaction) to 10^"3 (10 ng/reaction). Hence, the detection limits using total RNA standard for very virulent strain was 0.1 ng/reaction and classical strain and housekeeping gene were 10 ng/reaction for each, respectively.

Figure 3 shows the amplification plot for UPM94/273, very virulent IBDV strain using 5'FAM-TAMRA3' dual-labeled probe; duplicate reactions for each dilution. Amplification curve 1 : log 10^"1 with 1000 ng/reaction, CT= 16.79+0.12; amplification curve 2: log 10^"2 with 100 ng/reaction, CT= 20.55+0.07; amplification curve 3: log 10³ with 10 ng/reaction, CT= 24.69+0.29; amplification curve 4: log 10⁴ with 1 ng/reaction, CT= 28.30+0.23; amplification curve 5: log 10^"5with 0.1 ng/reaction, CT= 31.47+0.33.

Figure 4 shows the standard curve was generated from amplification of very virulent strain with each point represented the mean of the results.

Figure 5 shows the amplification plot for D78, classical IBDV strain using 5'HEX- TAMRA3' dual-labeled probe; duplicate reactions for each dilution. Amplification curve 1 : log 10^'1 with 1000 ng/reaction, CT= 24.17+0.20; amplification curve 2: log 10² with 100 ng/reaction, CT= 27.22+0.03; amplification curve 3: log 10³ with 10 ng/reaction, CT= 31.14+0.14.

Figure 6 shows the standard curve was generated from amplification of classical strain with each point represented the mean of the results. Figure 7 shows the amplification plot for D78, β-actin house-keeping gene using 5'FAM- TAMRA3' dual-labeled probe; duplicate reactions for each dilution. Amplification curve 1 : log 10^"1 with 1000 ng/reaction, CT= 25.37+0.45; amplification curve 2: log 10² with 100 ng/reaction, CT= 28.36+0.06; amplification curve 3: log 10³ with 10 ng/reaction, CT= 32.10+0.22.

In the Figures, the developed duplex real-time PCR assay has a linear relationship over five logio concentration range from a dilution of 10^'1 (1000 ng/reaction) to 10^"5 (0.1 ng/reaction) for very virulent strain amplification (Figures 3 and 4); whilst classical strain amplification (Figures 5 and 6) and β-actin house-keeping gene (Figures 7 and 8) showed a linear relationship over three log₁₀ concentration range from a dilution of 10^"1 (1000 ng/reaction) to 10^"3 (10 ng/reaction). Regression analysis of the CT values generated by the log₁₀ dilution series of RNA extracted from very virulent IBDV UPM94/273, vaccine IBDV D78 and β-actin house-keeping gene produced coefficient of correlation [R²) values of 0.9979, 0.9948 and 0.9958, respectively (Figure 4, 6, 8).

Regression analysis of the CT values generated by the log₁₀ dilution series of RNA extracted from very virulent IBDV UPM94/273, vaccine IBDV D78 and β-actin housekeeping gene produced coefficient of correlation (R²) values of 0.9979, 0.9948 and 0.9958, respectively. The slope of the regression lines indicated an increase of 3.7 and 3.5 cycles per log₁₀ decrease in input viral RNA for very virulent and classical strains and 3.4 cycles per log₁₀ decrease in input β-actin house-keeping gene, respectively. As expected, the CT increased in direct proportion to the dilution of the RNA standard. Based on the calculated formula, [1 o^H/slope) - 1] x 100%. Amplification efficiencies (E) determined from the regression analyses were 86% for the very virulent specific reaction, 93% for the classical specific reaction and 97% for β-actin house-keeping gene specific reaction (Figures 4, 6 and 8 and Table 9).

The assay reproducibility was tested using two separate 10-fold dilution series (1000 to 0.01 ng/reaction) of RNA were assayed within a single run. The reproducibility of the dilution series of the RNA standard were determined by calculation the mean, standard deviation (S. D.) and coefficient of variation (CV.) separately for each RNA dilution. The reproducibility of the developed assay for replicate measurement for IBDV-specific very virulent and classical reactions and β-actin housekeeping gene were high with a coefficient of variation (CV) for duplicate real-time RT-PCR of log serially diluted RNA of less than 1 .20%. The mean + SD of very virulent reaction variability was 0.82+0.32%

(range, 0.3-1 .2%), classical variability was 0.46+0.36% (range, 0.1 -0.8%), whilst β-actin housekeeping gene variability was 0.42+0.50% (range, 0.06-1.0%), respectively. This suggested that results of the developed duplex real-time RT-PCR assay were very reproducible. The intra assay variations of the duplex real-tiem PCR assay were low and considered acceptable (Tables 6, 7 and 8).

Table 6. Intra assay variation for amplification of serially diluted RNA from very virulent IBDV strain

CT: Threshold cycle; SD: Standard deviation; CV: Coefficient of variation Table 7. Intra assay variation for amplification of serially diluted RNA from classical IBDV strain.

CT: Threshold cycle; SD: Standard deviation; CV: Coefficient of variation

Table 8. Intra assay variation for amplification of serially diluted RNA from β-actin, house-keeping gene.

CT: Threshold cycle; SD: Standard deviation; CV: Coefficient of variation

Table 9. The parameters for equation of standard curves, correlation of coefficient (Ff), and difference in slope (Δs) for β-actin and IBDVs D78 and UPM94/273.

a: Difference in slope (Δs)= slope (house-keeping gene, β-actin - IBDVs)

Evaluation of the real-time PCR using experimentally infected samples

After establishing the real-time PCR condition, the detection limits and reproducibility for the detection of very virulent and vaccine strains, UPM94/273 and D78, respectively, the assay was also performed on bursal samples collected from experimentally infected SPF chickens. Bursal samples from uninfected control chickens and chicken vaccinated with D78 followed by challenged with very virulent UPM94/273 IBDV 6 hours later were used. For each group sampling on 1 , 3, 5, 10, 17, 24 and 30 day of post infection (d.p.i) were carried out, where bursal samples of five chickens were collected and pooled. No amplification was observed for the uninfected control group on 1 , 3, 5, 10, 17, 24 and 30 d.p.i. For dual-infection group, viral RNA was detected only on day 3 and 5 p.i. whilst no amplification was observed on 1 , 10, 17, 24 and 30 d.p.i. Bursal samples on day 3 p.i. showed positive amplification when tested using both FAM and HEX subtypes specific probes, with CT values ranging 23.70+0.03 for very virulent and 26.88+0.09 for classical detection (Figure 9). The detection of dual-infection on day 3 p.i was highly repeatable with a CV less than 0.15% for very virulent and 0.35% for classical strains, respectively (Table 10). Bursal samples on day 5 p.i. also showed positive amplification when tested using both FAM and HEX subtypes specific probes, with CT values ranging 22.34+0.17 for very virulent and 26.35+0.10 for classical detection (Figure 10). Positive amplification with high repeatable was also observed for samples collected at day 5 p.i with a CV less than 0.80% for very virulent and 0.40% for classical strains, respectively (Table 10). For samples collected at day 3 and 5 p.i, the assay variation has a mean +_SD of 0.45+0.45% (range, 0.1 -0.8%) for very virulent subtype and 0.36+0.04% (range, 0.3- 0.4%) for classical subtype, respectively (Table 10).

In Figure 9, the triplicate dual-infection strain were tested using subtype specific probes 5'FAM-TAMRA3' for very virulent and 5ΗEX-TAMRA3' for classical detection. Dual- infection IBDV strains showed positive CT value for both subtype specific probes. Amplification curve 1 : Triplicate dual-infection strain tested on FAM-labeled probe with mean CT= 23.70. Amplification curve 2: Triplicate dual-infection strain tested on HEX- labeled probe with mean CT= 26.88. Triplicate dual-infection strain were tested using subtype specific probes 5'FAM- TAMRA3' for very virulent and 5ΗEX-TAMRA3' for classical detection. Dual-infection IBDV strains showed positive CT value for both subtype specific probes. Amplification curve 1 : Triplicate dual-infection strain tested on FAM-labeled probe with mean CT= 22.34. Amplification curve 2: Triplicate dual-infection strain tested on HEX-labeled probe with mean CT= 26.35.

No amplification was observed for the uninfected control group on 1 , 3, 5, 10, 17, 24 and 30 d.p.i. For dual-infection group, viral RNA was detected only on day 3 and 5 p.i. whilst no amplification was observed on 1 , 10, 17, 24 and 30 d.p.i. Bursal samples on day 3 p.i. showed positive amplification when tested using both FAM and HEX subtypes specific probes, with CT values ranging 23.70+0.03 for very virulent and 26.88+0.09 for classical detection (Figure 9). The detection of dual-infection on day 3 p.i was highly repeatable with a CV less than 0.15% for very virulent and 0.35% for classical strains, respectively (Table 10). Bursal samples on day 5 p.i. also showed positive amplification when tested using both FAM and HEX subtypes specific probes, with CT values ranging 22.34+0.17 for very virulent and 26.35+0.10 for classical detection (Figure 10). Positive amplification with high repeatable was also observed for samples collected at day 5 p.i with a CV less than 0.80% for very virulent and 0.40% for classical strains, respectively. For samples collected at day 3 and 5 p.i, the assay variation has a mean +_SD of 0.45+0.45% (range, 0.1-0.8%) for very virulent subtype and 0.36+0.04% (range, 0.3- 0.4%) for classical subtype, respectively (Table 10).

Table 10. Detection of IBDV strains by Taqman based duplex real-time RT-PCR assay for experimental trial dual-infected samples.

CT: Threshold cycle; CV: Coefficient of variation; SD: Standard deviation

Evaluation of the real-time PCR using suspected samples from outbreak cases

The performance of the developed duplex Taqman based real-time RT-PCR assay was evaluated using 37 bursal samples collected from commercial chickens suspected with IBDV. The assay uses subtypes specific probes; probe labeled with FAM was specific for very virulent IBDV detection, while probe labeled with HEX was specific for classical IBDV detection, β-actin probe was included and use as a house-keeping gene for normalization.

Nine out of the 37 bursal samples were tested positive for very virulent IBDV. These nine very virulent strains: MB017/03 (GenBank accession no. EF070157), MB079/04 (GenBank accession no. EF070160), MB123/04 (GenBank accession no. EF070163), MB127/04 (GenBank accession no. EF070164), MB022/05 (GenBank accession no. EF070168), MB027/05 (GenBank accession no. EF070170), MB031/05 (GenBank accession no. EF070172), MB032/05 (GenBank accession no. EF070175) and MB074/05 (GenBank accession no. EF070181 ) were successful detected as very virulent using the developed assay (Figure 1 1 ). All the samples that were detected positive for very virulent showed CT values ranging from 13.31 +0.15 to 27.08+0.07 when tested with 5'FAM-TAMRA3' dual-labeled probe and no amplification or CT values was detected with 5ΗEX-TAMRA3' dual-labeled probe. The developed Taqman assay detects the very virulent IBDV strains with high repeatable of a CV less than 1 .2%. The assay variation has a mean +_SD of 0.50+0.31 % (range, 0.2-1.2%) (Table 11 ). The sequences of the hypervariable region of VP2 from 9 out of 13 samples were determined to confirm the positive detected samples are very virulent IBDV strains. Four samples (MB018/05, MB019/05, MB046/05, MB051/05) that were not sequenced were confirmed positive for very virulent IBDV based on detected of expected PCR product based on agarose gel electrophoresis (Figure 14).

Figure 1 1 shows a total of nine very virulent strains were tested using subtype specific probes 5'FAM-TAMRA3' for very virulent and 5ΗEX-TAMRA3' for classical detection, duplicate reactions for each strain. Very virulent IBDV strains showed positive CT value when tested with FAM-labeled probe. No amplification or CT value was detected with HEX-labeled probe. Amplification curve 1 : Very virulent strain, MB031/05, mean CT= 13.31 ; Amplification curve 2: Very virulent strain, MB022/05, mean CT= 17.51 ; Amplification curve 3: Very virulent strain, MB032/05, mean CT= 17.55; Amplification curve 4: Very virulent strain, MB127/04, mean CT= 18.01 ; Amplification curve 5: Very virulent strain, MB079/04, mean CT= 18.82; Amplification curve 6: Very virulent strain, MB027/05, mean CT= 19.29; Amplification curve 7: Very virulent strain, MB123/04, mean CT= 19.31 ; Amplification curve 8: Very virulent strain, MB074/05, mean CT= 19.52; Amplification curve 9: Very virulent strain, MB017/03, mean CT= 22.08; NTC: no template control.

Figure 12 shows a total of four classical strains were tested using subtype specific probes 5' FAM-TAM RA3' for very virulent and 5ΗEX-TAMRA3' for classical detection, duplicate reactions for each strain. Classical IBDV strains showed positive CT value when tested with HEX-labeled probe. No amplification or CT value was detected with FAM-labeled probe. Amplification curve 1 : Classical strain, MB030/05, mean CT= 25.53; Amplification curve 2: Classical strain, MB069/05, mean CT= 27.06; Amplification curve 3: Classical strain, MB007/05, mean CT= 28.27; Amplification curve 4: Classical strain, MB020/04, mean CT= 29.22; NTC: No template control

Figure 13 shows a total of 12 dual positive strains were tested using subtype specific probes 5'FAM-TAMRA3' for very virulent and 5ΗEX-TAMRA3' for classical detection, duplicate reactions for each strain. Dual positive IBDV strains showed positive CT value for both subtype specific probes.

Amplification curve 1 : MB023/05, mean CT VV=15.25; CL= 24.49 Amplification curve 2: MB057/05, mean CT VV=16.17; CL= 23.71 Amplification curve 3: MB040/05, mean CT VV=17.91 ; CL= 22.97 Amplification curve 4: MB067/05, mean CT VV=17.94; CL= 23.67 Amplification curve 5: MB120/04, mean CT VV=18.39; CL= 22.21 Amplification curve 6: MB058/04, mean CT VV=19.59; CL= 24.09 Amplification curve 7: MB041/05, mean CT VV=20.03; CL= 22.97 Amplification curve 8: MB061/05, mean CT VV=20.27; CL= 23.73 Amplification curve 9: MB082/04, mean CT VV=20.37; CL= 24.59 Amplification curve 10: MB001/05, mean CT VV=22.41 ; CL= 21.77 Amplification curve 1 1 : MB078/04, mean CT VV=24.82; CL= 23.62 Amplification curve 12: MB033/05, mean CT VV=25.00; CL= 24.38 NTC: no template control

Table 11. Detection of IBDV strains by Taqman based duplex real-time RT-PCR assay for bursal samples tested positive for very virulent IBDV detection.

CT: Threshold cycle; CV: Coefficient of variation; NA: No amplification; SD: Standard deviation

Four bursal samples from the IBD suspected outbreak cases were successful detected as classical strains using the developed assay. These four classical strains: MB007/05 (GenBank accession no. EF070166), MB020/05 (GenBank accession no. EF070167), MB030/05 (GenBank accession no. EF070171 ) and MB069/05 (GenBank accession no. 070180) showed CT values ranging from 25.53+0.05 to 29.22+0.28 when tested with 5ΗEX-TAMRA3' dual-labeled probe and no amplification or CT values was detected with 5'FAM-TAMRA3' dual-labeled probe (Figure 12). The developed Taqman assay detects the classical IBDV strains with high repeatable of a CV less than 2.0%. The assay variation has a mean +_SD of 0.84+0.80% (range, 0.2-2.0%) (Table 12)

Twelve bursal samples from the IBD suspected outbreak cases were detected positive for both very virulent and vaccine strains of IBDV. These 12 dual positive strains: MB058/04 (GenBank accession no. EF070158), MB078/04 (GenBank accession no. EF070159), MB082/04 (GenBank accession no. EF070161 ), MB120/04 (GenBank accession no. EF070162), MB001/05 (GenBank accession no. EF070165), MB023/05 (GenBank accession no. EF070169), MB033/05 (GenBank accession no. EF070173), MB040/05 (GenBank accession no. EF070174), MB041/05 (GenBank accession no. EF070176), MB057/05 (GenBank accession no. EF070177), MB061/05 (GenBank accession no. EF070178), and MB067/05 (GenBank accession no. EF070179) were positive for very virulent and classical strains with CT value ranging from 15.25+0.30 to 24.82+0.24 and 21.77+0.18 to 24.49+0.45, respectively (Figure 13). The developed Taqman assay detects the dual infection bursal samples with high repeatable of a CV less than 2.0% for very virulent and 1.9% for classical strains. The assay variation has a mean +_SD of 0.97+0.57% (range, 0.1 -2.0%) and 0.80+0.45% (range, 0.2-1.8%) for each subtypes, respectively (Table 13).

Table 12. Detection of IBDV strains by Taqman based duplex real-time RT-PCR assay for bursal samples tested positive for classical IBDV strains detection.

CT: Threshold cycle; CV: Coefficient of variation; NA: No amplification; SD: Standard deviation

Figure 14 shows Lane 1 : 100 bp DNA Ladder (Promega), Lane 2: IBDV positive control, Lane 3: MB018/05, Lane 4: MB019/05, Lane 5: MB046/05, Lane 6: MB051/05 and Lane 7: no template control (NTC). The bands pointed by the arrow correspond to the 1 14 bp PCR product.

Table 13. Detection of IBDV strains by Taqman based duplex real-time RT-PCR assay for dual infected samples from IBDV suspected cases

CT: Threshold cycle; CV: Coefficient of variation; NA: No amplification; SD: Standard deviation A total of eight bursal samples: MB002/05, MB025/05, MB026/05, MB028/05, MB039/05, MB043/05, MB055/05, and MB058/05 were detected negative for both very virulent and classical IBDV (Figure 15a, 15b, 15c).

Figure 15a shows a total of 12 negative IBDV bursal samples were tested using subtype specific probes 5'FAM-TAMRA3' for very virulent and 5ΗEX-TAMRA3' for classical detection. Four bursal samples were detected as very virulent IBDV strains and showed positive CT value for FAM subtype specific probe. Eleven negative strains showed negative amplification or no CT for both subtype specific probes. Amplification curve 1 : MB051/05, CT VV= 16.82; CL= NA Amplification curve 2: MB019/05, CT VV= 21.73; CL= NA Amplification curve 3: MB046/05, CT VV= 22.23; CL= NA Amplification curve 4: MB018/05, CT VV= 27.1 1 ; CL= NA Amplification curve 5: Very virulent positive control, CT VV= 17.35; CL= NA NTC: no template control

Figure 15b shows the real-time RT-PCR products of the negative IBDV samples. Lane 1 : 100 bp DNA Ladder (Promega), Lane 2: IBDV positive control, Lane 3: MB002/05, Lane 4: MB025/05, Lane 5: MB026/05, Lane 6: MB028/05 and Lane 7: MB039/05 and Lane 8: MB043/05. The bands pointed by the arrow correspond to the 114 bp PCR product.

Comparisons on the performance of the developed real-time PCR with other PCR detection methods

The developed Taqman based duplex real-time RT-PCR assay was compared to the standard IBDV diagnostic method, conventional RT-PCR and previously developed SYBR Green I based one-step real-time RT-PCR for the IBDV detection. The three assays were performed in one-step or single tube format and the estimated duration for the assay to complete the RT-PCR amplification was calculated. The duplex Taqman assay has the most rapid turn out time of only 2 h and 10 min whilst the SYBR Green I based assay took 2 h and 45 min to complete the steps. The conventional RT-PCR took the longest time, around 3 and half hours for amplification and 1 h for gel electrophoresis detection. For the total of 37 IBDV samples detection, duplex Taqman assay also was the most sensitive detection assay meanwhile the SYBR Green I and classical RT-PCR assays have similar sensitivity levels (Table 14).

Table 14. Assay comparison between the developed duplex Taqman RRT-PCR with SYBR Green I RRT-PCR and conventional RT-PCR

+: Positive amplification; -: Negative amplification;

The developed Taqman based assay able to 13 samples positive for very virulent IBDV. The conventional and SYBR Green I based real-time PCR detect only 8 samples positive for very virulent IBDV. Four samples (MB018/05, MB019/05, MB046/05 and MB051/05) highlighted in yellow were detected negative by the conventional and SYBR Green I based real-time PCR.

Bursal samples that were found positive for IBDV amplification was amplified using a pair of primer that amplified the hypervariable region of VP2. The RT-PCR amplification was performed in a final volume of 20 μl using the Reverse Transcriptase System

(Promega, USA) following the method recommended by the manufacturer. The RT-PCR amplification was carried out as. Briefly, a total volume of 9 μl of premix containing 1.0 μg/μl of total RNA, 25 pmole of each G3 and G4 primers (Table 15) and 1.0 μl of 90% dimethyl sulfoxide (DMSO) was incubated at 99O for 5 min to denature the RNA. The mixture was quickly chilled on ice for 5 mins. The premix was then mixed with 1 x reaction mixture (1 1 μl) containing 2.0 μl of 10 mM of dNTP mixture, 5.0 U of AMV reverse transcriptase, 20 U of recombinant RNasin ribonuclease inhibitor, 5.0 mM of MgCI₂ and 1 x of reaction buffer [250 mM Tris-HCL pH 8.3, 250 mM KCL, 50 mM MgCI₂, 50 mM DTT, 2.5 mM spermidine]. The final reaction mixture was incubated at 42°C for 1 hour and then denatured at 99⁰C for 1 min to inactivate the reverse transcriptase. The cDNA was then chilled on ice for 2 to 3 min and then used immediately or stored at - 8O⁰C.

Table 15. VP2 primers used in RT-PCR amplification

aThe numbering of the nucleotide position were based on Bayliss et al. (1990). b Primers have been previously described by Lin et al (1993).

The cDNA amplification was undertaken in a 50 μl reaction volume containing 5.0 μl of cDNA, 2.0 mM MgCI₂, 1.0 μl of 10 mM dNTP mixture, 25 pmole of each G3 and G4 primers, 2.5 U of Taq DNA polymerase and 1 x reaction buffer [10 mM Tris-HCI, 50 mM KCI, 0.1 % Triton ® X-100, p.H 8.8] and additional dH₂O. The amplification was performed in PTC-200 DNA Peltier Thermal Cycler (MJ Research, USA). The protocol was developed as follows: pre-denaturation at 95³C for 3 min followed by 35 cycles of denaturation at 94²C for 1 min, annealing temperature at 48^SC for 1 min and extension at 72^eC for 2 min. The reaction was terminated with a final extension at 72²C for 5 min.

Agarose Gel Electrophoresis

The amplified PCR products were analyzed on agarose gel 1.0% (w/v) electrophoresis in 1 X TAE buffer at 70 V for 30 min.

Purification of PCR Products

The PCR products for positive amplification of IBDV isolates were purified by using GENE^V ALL™ GEL SV Kit (General Biosystem, Korea) following the manufacturer's instructions.

DNA Sequencing

The purified PCR products obtained from the PCR amplification were sequenced using primers G3 and G4, respectively. Each purified PCR product was sequenced twice from both directions. The sequencing was carried out using ABI PRISM^® BigDye Terminator Cycle Sequencing Ready Reaction Kit v2.0 (Perkin Elmer, USA) in an automated DNA sequencer (ABI PRISM^® 377 DNA Sequencer) following the instructions supplied by the manufacturer. The cycle sequencing was conducted with the following thermal cycle profiles; 30 cycles, each with 92^QC for 30 s, 52³C for 30 s, 70^sC for 1 min and 4°C hold. Sequence Assembly and Analysis

The sequencing data were initially aligned to the known DNA sequences using the basic

BLAST (Basic Local Alignment Search Tool) search programme of National Centre for Biotechnology Information (NCBI) (http://www.ncbi.nlm.nih.gov/BLAST). The database searches were performed using the FASTA programme and the sequences data were assembled and analysed using the Bio-Edit package (Version 3.75c) of the Cluster W

Multiple alignment (Thompson et al., 1994). The sequences were then compared with 78 selected serotype 1 IBDV from GenBank database which including 52 of very virulent and 26 classical strains of IBDV (Table 16)

All the bursal samples that were detected positive for IBDV were subjected for sequence analysis of the HVR VP2 gene. A total of 474 nucleotides of VP2 gene were determined using primer G3 and G4. The nucleotide sequences of the 25 IBDV isolates were successfully submitted to National Center Biotechnology Information (NCBI) GenBank and with accession number as listed in Table 17. The deduced amino acid sequences of the VP2 hypervariable region for all the 25 IBDV isolates were aligned and compared with the very virulent IBDV reference strain UPM94/273 and classical IBDV reference strain D78 (Figure 16). There were no deletion and insertion in the deduced amino acid sequence of any of the isolate but amino acid substitutions occurred as compared with very virulent and classical reference strains (Table 16).

Figure 16 shows the predicted amino acid sequence alignment of IBDV isolates.

Deduced amino acid sequences of the VP2 hypervariable region residues 208 to 350 of 25 IBDV isolates isolated from bursal samples from IBD suspected cases compared with the very virulent reference strain UPM94/273 and classical strain D78. A dot indicated position where the sequence is identical to that of the reference strain, UPM94/273. Hydrophilic regions which represent areas critical for antigenicity and the serine rich heptapeptide region (SWSASGS) are boxed.

As shown in Figure 16 and Table 16, similar to the very virulent reference strain UPM94/273, all the 9 very virulent isolates (MB017/03, MB079/04, MB123/04, MB127/04, MB022/05, MB027/05, MB031/05, MB032/05 and MB074/05) that were detected positive for wlBDV using the SYBR Green I based real-time PCR have the characteristic amino acids as found in other very virulent IBDV strains at position 222 (Ala), 242 (lie), 256 (lie), 294 (lie) and 299 (Ser), respectively. The sequence of VP2 region of all the wlBDV isolates have a conserved heptapeptide motif SWSASGS at the hypervariable region at position 326 to 332 (Figure 16). The positive detection from additional four samples that were negative for SYBR Green I based real-time PCR nut positive for taqman real-time PCR were confirmed by the detection of expected band of 1 14 bp on agarose gel (Figure 14). The remaining 8 samples that were detected negative with Tagman real-time PCR were free of IBDV. No band was observed on agarose gel for these 8 samples (Figures 15a, 15b, 15c).

able 16. Comparison of amino substitution at VP2 hypervariable region between classical and very virulent strains and the

^*Highlighted is the unique amino acid for the IBDV isolate at different position

I 9 Very virulent (VV); ■ 4 Classical (CL); □ 12 Dual Positive (VV & CL)

^* The deduced amino acid sequence of the VP2 hypervariable region of published serotype I strains were aligned. The alignment consisted of 78 serotype I sequences; 52 very virulent and 26 classical, attenuated and vaccine strains.

Strain names are listed followed by the database accession number in brackets:

Very virulent: UPM97/61 (AF247006), UPM94/273 (AF527039), UPM04178 (AY970665), UPM04190 (AY791998), UPM04238 (DQ000436), 89224 (AJ001942), MYGA-97 (AJ238647), TN1 -93 (AJ404327), DV86 (Z25482), 89163 (Y14956), 91168 (Y14957), K280-89 (AF159217), K406-89 (AF159218), XJ-9 (AF155123), Ehime91 (AB024076), JK1 -97 (AJ249520), C4-2 (AF076223), HD96 (AF076226), JS-18 (AF076227), K357-88 (AF159216), B2-28 (AF076225), UP1 -99 (AJ277801 ), KT1 -98 (AJ249521 ), AP1 -93 (AJ245884), RJ1 -94 (Y18682), WB1 -93 (Y18650), TP1 -96 (AJ249522), CH2-97 (AJ245885), UP1 -97 (Y18612), UP2-97 (AJ249524), CH1 -97 (AJ245886), CS89 (Z25481 ), 74-89A (Z25481 ), Tula94 (X89570), N4 (AF159207), N13 (AF159214), N14 (AF159215), N7 (AF159209), N8 (AF159210), N6 (AF159208), N9 (AF15921 1 ), HR1 -96 (AJ249518), 95072-2 (AJ001946), 91247 (AJ001944), JY86 (Z25480), 94432 (Y14955), 95072-8 (AJ001947), 96108 (AJ001948), 91 184 (AJ001943), 92309 (AJ001945), AH-2 (AF076224), D1 1 -2 (AF076228).

Classical: UPM03292 (DQ074690), 88180 (AJ001941 ), HKL6 (AF051839), BV3 (AF076235), Univax (AF076236), CEF94 (AF194428), HZ96 (AF121256), Tri-bio (AJ249523), Soroa (AF140705), J1 (D16677), D78 (Y14962), Ts (AF076230), EM3 (Y14963), HN3 (AF076229), BJ-1 (AF076231 ), Miss (AF076234), GBF-1 (D16828), K (D16678), 3212 (AF091097), Variant A (M64285), Ark (AF076232), Variant E (D10065), E/DEL (X54858), Ga (AF076233), U28 (AF091099), P3009 (AF109154). Table 17. GenBank accession number for the 25 IBDV isolates used in this study

From the previous SYBR Green I detection results, we found that MB007/05, MB020/05, MB030/05 and MB069/05 were the only samples positive for the specific classical primer detection. In Table 16, compared to classical reference strain D78, IBDV isolates MB007/05, MB020/05 and MB030/05 have the characteristic amino acids as found in other classical IBDv at position 242 (VaI), 270 (Thr), 299 (Asn). The sequence of VP2 region of all the classical isolates have a conserved heptapeptide motif SWSASGS at the hypervariable region at position 326 to 332 (Figure 16).

For samples that were detected dual positive for both very virulent and classical IBDV, there were 1 to 4 amino acid residues differences at all the 12 dual positive samples (MB058/04, MB078/04, MB082/04, MB120/04, MB001/05, MB023/05, MB033/05, MB040/05, MB041/05, MB057/05, MB061/05 and MB067/05), which the IBDV samples positive for both very virulent and classical detection when compared to reference strains (Table 16). MB078/04, MB001/05 and MB033/05 have the characteristic amino acids found in both reference strains. MB078/04 and MB033/05 have the same amino acid residues of very virulent UPM94/273 at position 279 (Asp), 294 (He) and 351 (Thr), whereas at position 222 (Pro), 242 (VaI), 256 (VaI), 270 (Thr) and 299 (Asn) similar with classical strain D78. Both isolates have a unique amino acid substitution at position 217 from serine to leucine and this amino acid was observed in major hydrophilic region (residues 212-224). MB001/05 isolate has same characteristic amino acid at position 256 (lie) with very virulent strain and at position 242 (VaI), 270 (Thr), 279 (Asn), 294 (Leu), 299 (Asn) and 351 (Ala) with classical strain D78. MB001/05 isolate has 2 unique amino acid residues differences at position 222 (A222S for very virulent and P222S for classical) and valine (V) at position 329, whereas both reference strains have alanine (A) in common, respectively. MB058/04 and MB120/04 have 2 unique amino acid residues differences (G254D and S317R). The amino acid at position 317 (Arg) was observed in second major hydrophilic region (residues 314-324). While, other 7 isolates (MB082/04, MB023/05, MB040/05, MB041/05, MB057/05, MB061/05 and MB067/05) have 4 unique (D212N, Q249E, I264M and R349C) amino acid residues changes, respectively, with references strains UPM94/273 and D78. The amino acids at positions 212 (Asn) was observed in major hydrophilic region (residues 212-224) and 249 (GIu) was observed in minor hydrophilic region (residues 248-252). In addition, the heptapeptide region SWSASGS (residues 326-332) was conserved for all the isolates excepted MB001/05 (S330V) and MB030/05 (S330V), respectively (Figure 16).

Bursal samples from two groups of experimental trial were tested using the developed assay. First group was uninfected control chickens and second group was dual-infection group (to mimic dual-infection in natural infection), where the chickens were vaccinated with D78 followed by challenged with very virulent UPM94/273 IBDV 6 hours later. For each group sampling on 1 , 3, 5, 10, 17, 24 and 30 day of post infection (d.p.i) were carried out, where bursal samples of five chickens were collected and pooled.

No amplification was observed for the uninfected control group on 1 , 3, 5, 10, 17, 24 and 30 d.p.i. For dual-infection group, viral RNA was detected only on day 3 and 5 p.i. whilst no amplification was observed on 1 , 10, 17, 24 and 30 d.p.i. Bursal samples on day 3 p.i. showed positive amplification when tested using both FAM and HEX subtypes specific probes, with CT values ranging 23.70+0.03 for very virulent and 26.88+0.09 for classical detection (Figure 9). The detection of dual-infection on day 3 p.i was highly repeatable with a CV less than 0.15% for very virulent and 0.35% for classical strains, respectively (Table 10). Bursal samples on day 5 p.i. also showed positive amplification when tested using both FAM and HEX subtypes specific probes, with CT values ranging 22.34+0.17 for very virulent and 26.35+0.10 for classical detection (Figure 10). Positive amplification with high repeatable was also observed for samples collected at day 5 p.i with a CV less than 0.80% for very virulent and 0.40% for classical strains, respectively. For samples collected at day 3 and 5 p.i, the assay variation has a mean +_SD of 0.45+0.45% (range, 0.1-0.8%) for very virulent subtype and 0.36+0.04% (range, 0.3- 0.4%) for classical subtype, respectively (Table 10).

Virus quantitaion based on viral load fold change

To compare the amplification efficiency of both very virulent and vaccine IBDV strains with the house-keeping gene, β-actin, a serial 10 fold dilution of RNA derived from D78 and UPM94/273 infected samples from experimentally infected chickens (Group D) was used. The standard curve for β-actin, D78 and UPM94/273 were calculated (Figures 4, 6, 8). The equations and the parameters of the standard curves were shown in Table 9. If the amplification efficiency of D78 and UPM94/273 is very close and similar to that of the house-keeping gene, β-actin, the difference in slope (Δs) of standard curves, subtracting the slope of D78 and UPM94/273 from the slope of corresponding β-actin, will approach 0 and the ΔΔCT calculation for the relative quantitation of the target may be used. As shown in Table 9, Δs value of β-actin to D78 and UPM94/273 were -0.123 and -0.348, respectively. Both values were closed to 0.

Beta-actin was used to normalize the amount of RNA used in the reverse transcription reactions. Furthermore, the β-actin gene was used to standardize the PCR parameters to ensure the gene expression or viral load was unaffected by the experimental treatment. Experimental variability data obtained from the β-actin gene for dual-infection IBDV samples and experimental trial samples on day 3 and 5 p.i were summarized in Table 18. The detection of β-actin from the dual infected samples obtained from outbreak cases showed CT values ranging from 20.45 ± 0.23 to 23.21 +0.22 with CV less than 3.15%. The assay variation has a mean ± SD of 1.15 ± 0.77% (range, 0.05- 3.15%). Meanwhile, the detection of β-actin from the dual infected experimental trial samples on day 3 p.i. showed CT values 26.21+0.07 with CV 0.27%; on day 5 p.i. showed CT values 22.09 ± 0.09 with CV 0.39%. The assay variation has a mean ± SD of 0.33+0.08%, respectively.

Table 18. Detection of β-actin house-keeping gene from dual positive IBDV bursal samples obtained from suspected IBD cases and experimental trial infections by Taqman based duplex real-time RT-PCR assay

CT: Threshold cycle; CV: Coefficient of variation; P.I: Post infection; SD: Standard deviation

Viral Load Fold Change of Dual Positive IBDV Bursal Samples

The relative quantitation of very virulent and classical IBDV strains from dually infected bursal samples from experimental infected samples and suspected IBDV outbreak cases were determined using a standard curve and then expressed relative to a single calibrator sample. Very virulent strain in the dual infection samples was used as a calibrator. Duplex real-time RT-PCR was performed on the corresponding RNA synthesized from each sample.

The data were analyzed using 2^"MCT method, where ΔΔCT= (CT,_Targ_erCT _βa)c_L-(CT,_Targ_et- CT,_Pa)vvi. The data were presented as the fold change in gene and normalized to β-actin gene and relative to calibrator. The mean CT values for both target and β-actin gene were determined (Tables 19 and 20). For the very virulent calibrator, ΔΔCT equals zero and 2° equals one, so that the fold change in the dual positive bursal sample relative to the very virulent equals one. For classical strain in the dual positive bursal samples, evaluation of 2^"^⁰¹ indicated the fold change in gene relative to the very virulent strain.

For the experimental trial samples, the mean SD and CV were determined in triplicates of bursal samples collected at day 3 and 5 p.i (Table 20). The very virulent calibrator was set with the mean fold change equals to one. The viral load fold change showed that the very virulent strain was higher than the classical strain in the samples for day 3 p.i. (1.00 fold viral load of very virulent strain versus 0.11 fold viral load of classical strain) and day 5 p.i. (1.00 fold viral load of very virulent strain versus 0.08 fold viral load of classical strain), respectively (Figure 17). Figure 17 shows the viral load fold change of both vaccine and very virulent strains at day 3 and 5 p.i. was determined using the 2^~^^cτ equation. The viral load fold change showed that the very virulent strain was higher than the classical strain in the samples for day 3 p.i. (1.00 fold viral load of very virulent strain vs 0.1 1 fold viral load of classical strain) and day 5 p.i. (1.00 fold viral load of very virulent strain vs 0.08 fold viral load of classical strain), respectively.

Figure 18 on the other hand, shows the relative amount of both vaccine and very virulent strain for total 12 samples were determined using the 2^"MCT equation. Among the 12 samples, only three samples: MB078/04, MB001/05 and MB033/05 showed lower mean CT values in classical strain than the very virulent strain. Lower CT values represented for higher viral load in the sample. Therefore, the viral load fold change in classical strain was 2.297, 1.558 and 1.534 higher compared to very virulent viral load, respectively. The other nine samples (MB058/04, MB082/04, MB120/04, MB023/05, MB057/05, MB067/05, MB040/05, MB041/05, and MB061/05) were with higher viral load in very virulent compared to the classical fold change.

The viral load was also measured in 12 bursal samples that were obtained from suspected IBD cases that were detected positive for both very virulent and vaccine strains of IBDV. The changes in the viral load in these 12 dual positive IBDV bursal samples correlated to the CT values (Table 19). Among the 12 samples, only three samples: MB078/04, MB001/05 and MB033/05 showed lower mean CT values in classical strain than the very virulent strain. Lower CT values represented for higher viral load in the sample. Therefore, the viral load fold change in classical strain was 2.297, 1 .558 and 1.534 higher compared to very virulent viral load, respectively. The other nine samples were with higher viral load in very virulent compared to the classical fold change. Figures 18a and 18b showed the viral load fold change for very virulent and classical strain in dual positive IBDV bursal samples.

Table 19. Viral load fold change of dual positive IBDV bursal samples using the 2^"ΔΔCT mthod.

Ul

βa: Beta-actin gene; CL: Classical; VV: Very virulent. a. VV ΔΔCT= (CT.TargerCT,_βa)vv-(CT,τargerCT,_βa)vv

= VV ΔCT-VV ΔCT b. CL ΔΔCT= (CT,τarget-CT_iβa)cL-(CT,τargerCT,_βa)w

= CL ΔCT - VV ΔCT c. Viral load fold change = 2^"MCT

For the very virulent calibrator, ΔΔCT equals zero and 2° equals one, so that the fold change in the dual positive bursal sample relative to the very virulent equals one. For classical strain in the dual positive bursal samples, evaluation of 2^"^^CT indicated the fold change in gene relative to the very virulent strain.

Among the 12 samples, only three samples: MB078/04, MB001/05 and MB033/05 showed lower mean CT values in classical strain than the very virulent strain. Lower CT values represented for higher viral load in the sample. Therefore, the viral load fold change in classical strain was 2.297, 1.558 and 1.534 higher compared to very virulent viral load, respectively. The other nine samples (MB058/04, MB082/04, MB120/04, MB023/05, MB057/05, MB067/05, MB040/05, MB041/05, and MB061/05) were with higher viral load in very virulent compared to the classical fold change.

Table 20. Viral load fold change of experimental trial dual-infection IBDV samples on day 3 and 5 p.i using the 2^'ΔΔCT method.

βa: Beta-actin gene; CL: Classical; VV: Very virulent.

Ul a. VV ΔΔCT= (CT,TargerCT,_βa)_Vv-(CT,TargerCT,_Pa)vv Ul

= VV ΔCT- VV ΔCT b. CL ΔΔCT= (CT,τarget-CTβ_a)cL-(CT,τargerCT,βa)vV

= CL ΔCT-VV ΔCT c. Viral load fold change = 2^'^^C1

The viral load fold change showed that the very virulent strain was higher than the classical strain in the samples for day 3 p.i. (1.00 fold viral load of very virulent strain vs 0.1 1 fold viral load of classical strain) and day 5 p.i. (1.00 fold viral load of very virulent strain versus 0.08 fold viral load of classical strain), respectively.

Claims

1. A method of detecting Infectious Bursal Disease Virus (IBDV) strains in a bird sample, wherein said method includes the steps of: a) providing a sample from the bird that is infected with IBDV strains, b) amplifying said IBDV strains by means of a polymerase chain reaction (PCR) using at least one oligonucleotide primer pair, wherein said oligonucleotide primer pair and probe are selected to give specific single amplification (very virulent or classical strain) or dual infection with relative quantitation (very virulent and classical strain); and c) detecting said IBDV strains.

2. The method according to claim 1 , wherein said primers are selected from the group of SEQ ID NO;1 to SEQ ID NO;2.

3. The method according to claim 1 or 2, wherein said PCR is selected from real-time (RT)-PCR, in particular subtype-specific Taqman probe from the groups of SEQ IB NO; 3 (for very virulent strain detection) to SEQ ID NO. 4 (for classical strain detection).

4. The method according to any of claims 1 to 3, wherein said IBDV strains in sample are further isolated.

5. The method according to any of claims 1 to 4, wherein said IBDV strain is selected from the very virulent strains UPM 94/273.

6. The method according to any of claims 1 to 4, wherein said IBDV strain is selected from the vaccine strains D78.

7. The method according to any of claims 1 to 6, wherein detecting said IBDV strains comprises detecting an amplified product in the size of 114 base pair (bp).

8. The method according to any of claims 1 to 7, wherein detecting said IBDV strains comprises mixing with dual labeled Taqman probe.

9. The method according to claim 8, wherein said step of amplifying the very virulent strains and the vaccine strains by real time reverse transcriptase polymerase chain reaction (PCR) includes: a) reverse transcription at least 50⁰C for 30 min b) initial activation step at least 95 °C for 15 min c) amplification step of at least 35 cycles with the following steps, d) at least 94 °C for at least 15 seconds, e) at least 60 °C for at least 30 seconds, f) at least 76 °C for at least 30 seconds,

10. The method according to any one of the preceding claims, wherein the bird is a chicken.

1 1. The use of oligonucleotide primer pairs consisting of [5'(SEQ ID NO;1 )3'];[5'(SEQ ID NO;2)3']; and probe for very virulent strain [5'(SEQ ID NO;3)3']; and probe for classical strain [5'(SEQ ID NO;4)3']for detecting and/or differentiating IBDV in a bird sample.

12. A diagnostic assay kit adapted or assembled to perform the method of any of the above.