Development and Evaluation of Real Time RT-PCR Assays for Detection and Typing of Bluetongue Virus

Bluetongue virus is the type species of the genus Orbivirus, family Reoviridae. Bluetongue viruses (BTV) are transmitted between their vertebrate hosts primarily by biting midges (Culicoides spp.) in which they also replicate. Consequently BTV distribution is dependent on the activity, geographic distribution, and seasonal abundance of Culicoides spp. The virus can also be transmitted vertically in vertebrate hosts, and some strains/serotypes can be transmitted horizontally in the absence of insect vectors. The BTV genome is composed of ten linear segments of double-stranded (ds) RNA, numbered in order of decreasing size (Seg-1 to Seg-10). Genome segment 2 (Seg-2) encodes outer-capsid protein VP2, the most variable BTV protein and the primary target for neutralising antibodies. Consequently VP2 (and Seg-2) determine the identity of the twenty seven serotypes and two additional putative BTV serotypes that have been recognised so far. Current BTV vaccines are serotype specific and typing of outbreak strains is required in order to deploy appropriate vaccines. We report development and evaluation of multiple ‘TaqMan’ fluorescence-probe based quantitative real-time type-specific RT-PCR assays targeting Seg-2 of the 27+1 BTV types. The assays were evaluated using orbivirus isolates from the ‘Orbivirus Reference Collection’ (ORC) held at The Pirbright Institute. The assays are BTV-type specific and can be used for rapid, sensitive and reliable detection / identification (typing) of BTV RNA from samples of infected blood, tissues, homogenised Culicoides, or tissue culture supernatants. None of the assays amplified cDNAs from closely related but heterologous orbiviruses, or from uninfected host animals or cell cultures.


Introduction
analysis, decision to publish, or preparation of the manuscript. Deltamune Pty Ltd provided support in the form of salaries for authors ACP and IMW, but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the 'author contributions' section.
The bluetongue viruses were initially identified and distinguished from the other orbiviruses by serological methods, including Agar gel immuno-diffusion (AGID) tests and enzyme-linked immunosorbent assay (ELISA) [33]. However, these methods are slow, labour intensive and require virus isolation and/or access to standard reagents (antibodies and antigens) that may themselves represent a potential biosecurity risk.
Virus neutralisation tests (VNT) and serum neutralisation tests (SNT) are the "gold standard" method for BTV serotyping. However, serological procedures are expensive and timeconsuming and are often associated with poor sensitivity. Unlike conventional PCR-based approaches, quantitative real-time PCR chemistries use fully automated detection systems that eliminate the need for post-amplification sample handing, hence displaying very high sensitivities and broad dynamic ranges after optimization particularly for mixed infections. Highly sensitive and specific molecular 'typing' assays, targeting Seg-2 (including conventional, RT-PCR, sequencing and probe-hybridisation methods) have previously been reported for identification of 26 BTV and 9 AHSV serotypes [21,42,43]. However, the detection of novel serotypes and further introductions of exotic serotypes and topotypes into new geographic areas, requires a complete set of Seg-2 specific TaqMan assays for accurate, sensitive and specific serotype identification of all of the known BTV types, to help in surveillance as well as the rapid design and implementation of effective vaccination and control programmes. Full-length Seg-2 sequence data from multiple isolates of individual types, including strains belonging to both eastern and western topotypes, were used to design a most complete set of 'serotype-specific' primers and probes to detect and identify Seg-2 from all known serotypes of BTV with the exception of putative BTV-28. The specificity of each assay was evaluated using a wide range of BTV isolates from the Orbivirus Reference Collection (ORC) [44] at The Pirbright Institute (TPI).
The real-time RT-PCR assays that are described in this manuscript provide rapid and reliable BTV serotype identification and are suitable for high through put diagnostic systems. They are not invalidated by mixed infections and have been used to identify incursions of multiple BTV types into Europe, India, Australia, Middle East and the USA [10,11,[45][46][47].

Primers and probe design
Seg-2 nucleotide sequences of BTV reference strains that were generated during this study using the previously described method [25] and nucleotide sequences that are already available in the public domain (GenBank) were used to design primers and probes for the detection and typing of BTV RNA (S1 Table). For this the Seg-2 sequences were aligned and analysed collectively and separately for each of the 27+1 BTV serotypes using MEGA v. 6 [48]. Unique regions in Seg-2 were identified for each serotype, as targets for the primers and probes ( Table 2). The primers and probes for each serotype specific assay, were also checked using the available sequence data to ensure no cross-reactions with the genome segments of related BTV serotypes or any of the 29 heterologous Orbivirus species [25,44,[49][50][51][52]. Probes were labelled at their 5' and 3' ends with 6-carboxyfluorescein (6-FAM) and Black Hole Quencher-1 (BHQ-1) respectively. All oligonucleotides were synthesised by Eurogentec, UK. Primers were PAGE purified and probes were HPLC purified.

Virus isolates
A total of 1063 BTV isolates (from ORC) representing all 27+1 of the known BTV serotypes, and strains from different areas of the world, were used for evaluation of the real-time RT-PCR assays (S2 Table) [44]. Other Orbivirus species isolates were also used to assess the specificity of the BTV-serotype-specific assays. All of the isolates were grown in BHK-21 clone 13 cells (European Collection of Animal cell Cultures [ECACC-84100501]), Vero cell monolayers (ECACC-84113001), or in KC cells derived from Culicoides sonorensis [53]. Infected KC cells were harvested 7 days post infection, whereas the infected mammalian cells were harvested when~80% cytopathic effect (CPE) was observed. In these studies the virus isolates were generated from diagnostic samples (blood and/or spleen or lung), taken from naturally infected animals as part of normal veterinary surveillance in the respective countries. Established reference strains of BTV were also included in these studies [54]. Neither animals were infected nor Table 2. List of primers and probes for BTV type-specific assays.

RNA isolation
dsRNA was extracted, either from tissue culture supernatant, blood or other clinical samples, using a MagVet TM universal purification kit (Kingfisher TM , Life Technologies), or from BTV infected cells (using Trizol Reagent 1 , Invitrogen, UK) [55]. In the first method, 100 μl of infected-tissue-culture supernatant/blood/clinical sample was added to 250 μl of the lysis buffer (NM1). Total nucleic acid (80 μl) was extracted from this solution using the Kingfisher TM platform, with magnetic beads and Kingfisher TM , automates. Each of the isolates was handled with care to avoid any cross-contamination. Total nucleic acid from uninfected tissue culture supernatants, sheep blood or homogenised Culicoides was also extracted using Kingfisher TM (Life Technologies) system.

RT-PCR reactions
RT-PCR reactions were performed as described previously by [56]. Briefly, a fragment of the targeted genome segment was amplified with different primers-probe sets using SuperScript III/ Platinum Taq One-Step qRT-PCR Kit (Invitrogen, UK). RNA samples were always heat denaturated (95°C for 3 min) prior to addition to the reaction mix [37]. Amplification was carried out in MX3005p (Stratagene, UK) using the conditions: 55°C for 30 min, 95°C for 10 min followed by 45 cycles of 95°C for 15 s and 60°C for 1 min. Fluorescence was measured during a 60°C annealing/extension step. Cycle threshold (Ct) values were measured as the point at which the sample fluorescence signal crossed a threshold value (the background level). Negative results (for assays that did not exceed this level of signal) are reported as 'No Ct' . All precautions were taken to avoid accidental contamination. The composition of the individual optimised assays using either one or two sets of primers and probes is presented in Table 3.

Evaluation of RT-PCR assays for diagnostic specificity
A panel of (1063) isolates representing BTV serotypes 1 to 24, -26, -27 and -29, from different geographical locations, were obtained from the ORC (S2 Table; [44] and tested in each case in triplicate. The Seg-2 based type-specific assays were tested in vitro with both homologous and heterologous serotypes to ensure diagnostic specificity and sensitivity. Isolates for BTV-25, and -28 were not yet available from the ORC for testing in vitro, so the primers and probes for BTV-25 were tested against them by sequence comparisons only, while for BTV-28 Seg-2 sequence data was not available for design of primers. Representative isolates of other Orbivirus species, including African horse sickness (AHSV), Epizootic haemorrhagic disease virus (EHDV), Equine encephalosis virus (EEV) and Peruvian horse sickness virus (PHSV) (as listed in S2 Table), were used to evaluate and confirm the diagnostic specificity the BTV-specific assays. The typing assays are also used extensively in the European and OIE reference laboratory at Pirbright and by colleagues in laboratories in the  Table). To ensure that no false positives are caused by cross-reactions with the host-species, total nucleic acid was extracted and tested from uninfected BHK, Vero and KC cell-culture supernatants, sheep and cattle blood and homogenised C. sonorensis, with uniformly negative results (S2 Table).

Evaluation of sensitivity and efficiency of the assays
The analytical sensitivity of the Seg-2 type-specific assays for the European serotypes (BTV-1, -2, -4, -6, -8, -9, -11, and -16) were assessed using a dilution series of the quantified dsRNA genome of reference strains of each BTV serotype (as listed in S1 Table). Additionally, this selection of the assays for the European serotypes were also assessed for their sensitivity and efficiency, using a recombinant plasmid DNA having insert sizes ranging from 78-127 bps (Seg-2 specific amplicons for each of the eight European serotypes).
The dsRNA standards were prepared as follows: viral dsRNA (of RSArrrr/01 or other reference strains) was extracted as previously described [55], assessed for any ssRNA remains in 1% agarose gel electrophoresis and the concentration of dsRNA was determined with NanoDrop (Thermo Fisher Scientific, USA). To test analytical sensitivity, a 10-fold dilution series of dsRNA (10 10 to 10°copies per μl) was made in a sample of RNA extracted from uninfected Table 3

dsRNA (μl) 3
Total volume (μl) 25 1 SuperScript III/ Platinum Taq One-Step qRT-PCR Kit (Invitrogen, UK). 2 The duplex format of the assay is for use with sets of primers and probes for the same serotype (eastern and western strains). BHK cells supernatant, then tested in triplicate. The number of dsRNA copies was calculated with the formula: Y = (X/(a x 680)) x 6.022 x 10 23 , where: Y = molecules/μl; X = g/μl dsRNA; a = viral genome length in nucleotides; 680 is the average molecular weight per nucleotide of dsRNA.
To test analytical sensitivity of the Seg-2 type-specific assay using recombinant plasmid DNA, a 10-fold dilution series of plasmid DNA (10 10 to 10°copies per μl) was made in TE dilution buffer and then tested in triplicate. The number of plasmid DNA copies were calculated with the formula: Y = (X/(a x 660)) x 6.022 x 10 23 , where: Y = molecules/μl; X = g/μl dsDNA; a = plasmid plus insert length in nucleotides; 660 is the average molecular weight per nucleotide of dsDNA.
Viral dsRNA dilution series were used to generate standard curves by linear regression methods, setting Ct values as dependent, and the dsRNA concentrations as independent variables. The slope of the standard curves for the optimised serotype-specific assays, were then used to estimate the efficiency of the individual assays in detection of each of the reference strain. The efficiency was calculated by the formula E% = (10 −1/slope −1)×100. Efficiencies of Seg-2 assays were estimated on the basis of standard curves plotting Ct values against corresponding log plasmid copy number per reaction.

Design of serotype specific primers-probes for different BTV 'types'
Sequence data generated for Seg-2 of multiple field, reference and vaccine strains as well as those available publically in GenBank (S1 Table), were compared to select regions unique to each of the virus types for the design of primers and probes that could be used to distinguish and detect each of the 27 + 1 BTV serotypes ( Table 2).
The sequences and positions for the serotype-specific primers for BTV serotypes 1-27 and putative BTV-29 are given in Table 2. Seg-2 sequence data for assay design for putative BTV28 was unavailable. In some cases (e.g. BTV-7, -18, -21 and -24) the number of available virus strains, topotypes and / or sequences for each serotype is limited. This imposes limits on the designs of the current assays, as well as on their evaluation and testing, suggesting that in order to maintain their efficiency the re-design of some assays may be necessary, as additional isolates and new topotypes of each serotype become available.
In some cases, for example during testing of the BTV-1 specific assays, some of the eastern topotype strains from Australia (AUS2003/01, AUS2005/02, AUS2009/02) were not detected by the original primers and probes. The type 1 assay was therefore redesigned with a new probe and three out of four primers (Table 2). Similarly the BTV-4, BTV-12 and BTV-21 assays were redesigned to maintain their efficiency for the detection of Indian, Zimbabwean and Brazilian strains ( Table 2).

Evaluation of RT-PCR assays targeting Seg-2, for type-specificity
The RNAs of the 27 monotypic BTV reference strains (BTV-1 to 24, -26, -27 and -29), were tested in triplicate, with 'typing-assays' targeting Seg-2 of each BTV serotypes (S2 Table). Due to non-availability of RNA for BTV-25 only sequence based comparison/evaluation was done for this serotype. As the sequence data for BTV-28 was not available, the assay for this serotype could not be designed. In every case amplification was only observed with the homologous assay confirming the specificity of the assays for each serotype (S2 Table). The specificity of the assays was further evaluated using field and/or vaccine strain isolates of each serotype, where possible representing different topotypes collected from diverse geographic locations (from the ORC). Positive signals were only obtained with the homologous serotype (S2 Table).
RT-PCR 'typing' results for majority of the isolates tested were subsequently verified by sequencing and phylogenetic analyses of Seg-2 (S2 Table). With some strains the original primers and probes (as mentioned in table 2) showed relatively poor reactivity. These included some of the isolates of BTV-1 from Australia (AUS1996/03; AUS2003/01; AUS2005/02; AUS2009/02) and BTV-4 strains isolated in 2014 from Southern India (IND2014/24), BTV-12 from India (IND2012/01), Brazil (BRA2002/01) and Zimbabwe (ZIM2003/04) and BTV-21 from India (IND2007/09) (S1 Table). The Seg-2 sequence data from these strains showed variations in the primer and/or probe 'foot-prints' . Additional sets of primers and probes were therefore designed and tested for RNA isolated from blood and cell culture grown samples of these serotypes (Table 2).
Nucleic acid preparations derived from uninfected hosts species (sheep and cattle blood and C. sonorensis) or uninfected cell culture supernatants (KC cells, BHK cells and Vero cells), or dsRNA from other closely related Orbivirus species (EEV, PHSV, EHDV and AHSV) (S2 Table) gave 'no Ct' values in any of the type-specific assays. Sequence comparison of Seg-2 sequence data for 29 Orbivirus species [44], indicated that neither primers nor probes of the BTV type-specific assays, would bind to, or amplify their RNA.
Some of the BTV strains originating from India, Israel, Turkey, Europe and, Africa, were tested using serotype-specific TaqMan probe based qRT-PCR assays giving positive results for more than one serotype, reflecting the co-circulation of multiple serotypes [44]. For example, The Indian isolate IND2005/06 contains BTV-2 and BTV-9; Indian isolate IND2005/05 contains BTV-1, BTV-10 and BTV-23; Israeli isolate (ISR2010/22) contains BTV-15 and BTV-8; isolate ISR2010/16 contains BTV-4 and BTV-8. Isolates TUR2000/06 and TUR2000/07 from Turkey contain BTV-9 and BTV-16 [44]. These real-time RT-PCR data were confirmed by conventional Seg-2 specific qRT-PCR assay [21] and Seg-2 sequencing of the mixed BTV strains, showing more than one distinctive consensus sequences in each case. Comparison of the sequence data to the Seg-2 dataset of the 29 BTV reference strains, confirmed the real-time RT-PCR typing results.

Use of RT-PCR assays for typing of BTV isolates from disease outbreaks
The type-specific qRT-PCR assays described here, were used for the primary identification, or confirmation of BTV serotype for multiple (>1000) diagnostic samples, and field isolates of BTV. These type specific qRT-PCR assays were used during August-September 2006 for the identification of BTV-8 in clinical samples from animals in the Maastricht region of northern Europe, and again in 2007 from the UK. Late in 2008/early 2009, BTV-6 was identified in the Netherlands [11], BTV-11 in Belgium [57], BTV-25 in Switzerland [17] and BTV-26 in Kuwait [14], BTV-27 in Corsica [13], and BTV-29 from South Africa [12]. The assays and primers described here are now in routine use in the OIE reference laboratory at The Pirbright Institute, to detect and identify any of the BTV serotypes as listed on the ORC web pages [44] in diagnostic samples. The results of these assays can be (are) followed by full genome sequencing and phylogenetic analysis of representative samples from each outbreak, to confirm serotype, and identify topotype, and genotype (revealing the lineage / origins of each genome segment and identifying reassortant strains [10,28]. The BTV serotypes involved in outbreaks that occurred in southern India during 2014 were identified using these qRT-PCR methods, including BTV-2, BTV-4 and BTV-12 (Maan et alunpublished data). BTV-1 (SPA2014/08), and BTV-4 (SPA2014/06, SPA2014/07) were identified in Spain and the BTV-4 outbreak that occurred in South-eastern Europe during 2014 and 2015, were all identified in this way (BUL2014/01 to BUL2014/14; KOS2014/04 to KOS2014/ 09) (S2 Table). Viruses were also isolated from bovine semen samples collected in 2011 from Brazil. One of the isolates (BRA2011/02), contained BTV-4, BTV-8, BTV-10 and BTV-16, while another isolate (BRA2011/01) contained BTV-4 and BTV-10 (Gasparini et al-in preparation) (S2 Table). This provides the first report of BTV-8, BTV-10 and BTV-16 in Brazil.

Analytical sensitivity and efficiency
RNA of the homologous BTV serotype was detected by all of the 'typing' assays. A selection of the assays for the European serotypes (BTV-1, -2, -4, -6, -8, -9, -11, and -16) were tested for their sensitivity and efficiency. Plasmid DNA corresponding to homologous European BTV serotype Seg-2 RNA targets were detected by respective 'typing' assays, at all nine dilutions down to 2-11 copies (S3a-S3h and S4 Tables).
Efficiency rates were calculated for the different European serotype typing assays, on the bases of dilutions series of dsRNA, giving values between 95-102% (S4 Table) reflected by a range of slope values (between -3.2 and -3.4). All eight of the Seg-2 assays for European serotypes that were tested against their homologous serotypes showed linearity (R 2 > 0.99) (S4 Table).

Discussion
Bluetongue virus causes severe and economically important diseases in domesticated and wild animals. Introduction of these viruses into areas which are usually free from the disease having immunologically naïve populations of susceptible host can cause high morbidity and mortality [62]. Even in endemic areas, BTV can cause massive reduction in overall productivity of animals [63]. Apart from emergence of new reassortant strains with novel characteristics, massive genetic and antigenic diversity among different serotypes/types make the prevention and control of BTV very difficult.
To enable the fast and accurate identification and typing of BTV we report the development of a complete panel of 'TaqMan' fluorescence-probe based quantitative real-time type-specific RT-PCR assays. These real-time RT-PCR assays are considerably more sensitive than the conventional RT-PCR assays previously published for typing BTV RNA [21,42,64]. There was also no evidence for cross-amplification of RNA from heterologous serotypes by any of the type-specific assays (Seg-2). However, the very high specificity of real-time RT-PCR primers and probes can lead to false negative results, particularly with strains containing sequence variations in the target gene [65]. Although very recently a set of real-time assays for typing 22 different BTV types have been reported by [66], the combination of type-specific diagnostic assays described here provides a more complete set of tools for rapid and accurate detection and identification of BTV to support surveillance programmes globally. As previously demonstrated for BTV, real-time RT-PCR assays are the most sensitive and reliable methods (that are currently available) for orbivirus detection and typing [67][68][69]. All of the type-specific assays targeting Seg-2 reported here are highly efficient, doubling amplicon quantity during each round of amplification in the geometric phase of the reaction and can detect similar levels of RNA to other published real-time PCR assays [40,41]. The large number of BTV isolates that were available for this study from the ORC, have made the wider validation of these assays possible. However, owing to high genetic diversity in these viruses it is very likely that refinement of these assays will be required in future. Eastern and western strains of the same serotype for BTV-1, 2, 9 and 16, were too divergent for a single set of 'common' type-specific primers and probe in each case (Table 3). Hence the assays for these serotypes use a duplex format to allow amplification of the more diverse strains (e and w) within these types (S1 Table). Some of the BTV-1 strains from Australia and BTV-4, BTV-12 and BTV-21 isolated from recent outbreaks in India showed poor reactivity with the original pairs of primers and probes (Table 2) hence the primers for these serotypes were upgraded as listed in Table 2. In certain serotypes (BTV-7, -18, -20, -21, -25, -26, -27 and -29) the specificity of amplification was not widely validated as only limited number of isolates, were available for testing, leaving some uncertainty concerning their specificity with more diverse strains of the same types. The diagnostic specificity of the typing assays described here, relates primarily to their inability to detect Seg-2 of non-homologous types, while still detecting all available isolates from the homologous BTV type. If a novel isolate of the virus is identified (e.g. by the virus-species-specificassays) that fails to amplify using the typing assays, it should be sequenced (Seg-2) to provide a basis for further development/refinement of the relevant primers and probes. Similar problems have been addressed with conventional and real-time RT-PCR assays detecting Seg-2 of BTV, in order to maintain their specificity and sensitivity [21].
Previous serological methods have detected multiple serotypes of BTV co-circulating in the endemic regions. The qRT-PCR assays described here have been used to identify several isolates containing mixed BTV types, which could not have been unambiguously typed by VNT.
Recently several novel BTV types (BTV-25-26-27) have been identified primarily by RNA sequence based methods, including RT-PCR assays, and subsequently confirmed as novel serotypes by serological typing methods [13,14,17]. Additional virus strains / isolates have been identified that represent two additional putative BTV serotypes. These include BTV-28 from the Middle East (Peter Mertens-Personal communication); and a putative BTV-29 isolated from an alpaca in South Africa [12]. It has been suggested that it would be useful to set absolute values for nucleotide and amino acid variations in Seg-2 and VP2 that can be used to define BTV serotype (Peter Mertens-Personal communication). Current data indicate that viruses within the same serotype can show up to 31.6% nucleotide variation, and 27.4% amino acid variation in Seg-2 and VP2 respectively [11,18]. Viruses belonging to different serotypes can show up to 71.5% nucleotide identity and 77.8% aa identity, with a minimum of 28.5% nucleotide variation and 22.2% aa variation between serotypes [11,18]. There is clearly an overlap that makes the identification of clear limits to serotype variation difficult, but it is important to note that this overlap is caused primarily by the existence of different topotypes of Seg-2 and VP2 within individual serotypes. However, if the major different topotypes (for example the eastern and western groups) are considered separately, then the level of variation for isolates within each serotype (also belonging to the same topotype), drops to a maximum of 21.8% nucleotide variation and 13.9% aa variation. This then gives a clear difference of 7.7% nt and 8.3% aa difference between the levels of variation for distinct serotypes and those within the same serotype and topotype. This potentially gives us useful and clear guidance for the identification of the existing serotypes, where both eastern and western strains have already been identified and sequenced. However, if a novel isolate is discovered that fall into this gap, it could represent either a novel serotype that is closely related to, but distinct from, other established serotypes, or it could represent an isolate of an existing serotype, but belonging to a different and distinct major topotype. The novel putative serotypes BTV-28 and -29 show significant sequence variations in Seg-2/VP2 when compared to isolates of previously recognised BTV serotypes, with a level of identity (up to 69.5%), confirming that they represent novel types. Even though the serological techniques are relatively insensitive and prone to cross-reactions, they remain as a gold-standard for serotype identification, particularly for confirmation of novel serotypes. Once identified in this way these new viruses would become a 'reference strains' , either for a novel serotype, or of an existing serotype but within a new major topotype. This also confirms that we need to identify reference strains for each serotype, within each of the major BTV topotypes.
To summarize, the 'TaqMan' probe based real-time RT-PCR based methods described here, represent fast, robust and reliable tools for the detection and typing of BTV. Together with sequencing studies it will help us to better understand the molecular epidemiology (distribution and spread) of these viruses. This system will support investigations of BTV outbreaks and can help to devise strategies for timely implementation of control measures for BTV, particularly vaccination programmes.
Supporting Information S1 Table. Sequence data used to design real time RT-PCR assays.