Journal of General Virology (2014), 95, 434–441
DOI 10.1099/vir.0.059659-0
Whole genome analysis of epizootic hemorrhagic disease virus identified limited genome constellations and preferential reassortment Srivishnupriya Anbalagan, Elyse Cooper, Pat Klumper, Randy R. Simonson and Ben M. Hause Correspondence
Newport Laboratories, 1520 Prairie Drive, Worthington, Minnesota 56187, USA
Srivishnupriya Anbalagan [email protected] Ben M. Hause [email protected]
Received 18 September 2013 Accepted 3 November 2013
Epizootic hemorrhagic disease virus (EHDV) is a Culicoides transmitted orbivirus that causes haemorrhagic disease in wild and domestic ruminants. A collection of 44 EHDV isolated from 2008 to 2012 was fully sequenced and analysed phylogenetically. Serotype 2 viruses were the dominant serotype all years except 2012 when serotype 6 viruses represented 63 % of the isolates. High genetic similarity (.94 % identity) between serotype 1 and 2 virus VP1, VP3, VP4, VP6, NS1, NS2 and NS3 segments prevented identification of reassortment events for these segments. Additionally, there was little genetic diversity (.96 % identity) within serotypes for VP2, VP5 and VP7. Preferential reassortment within the homologous serotype was observed for VP2, VP5 and VP7 segments for type 1 and type 2 viruses. In contrast, type 6 viruses were all reassortants containing VP2 and VP5 derived from an exotic type 6 with the remaining segments most similar to type 2 viruses. These results suggest that reassortment between type 1 and type 2 viruses requires conservation of the VP2, VP5 and VP7 segment constellation while type 6 viruses only require VP2 and VP5 and are restricted to type 2-lineage VP7. As type 6 VP2 and VP5 segments were exclusively identified in viruses with type 2-derived VP7, these results suggest functional complementation between type 2 and type 6 VP7 proteins.
INTRODUCTION Epizootic haemorrhagic disease is an infectious disease caused by viruses belonging to the species Epizootic hemorrhagic disease virus (EHDV) within the genus Orbivirus and family Reoviridae (Savini et al., 2011). The virus is transmitted by insects of the genus Culicoides (Savini et al., 2011). EHDV periodically causes disease in white-tailed deer, pronghorn antelope and desert bighorn sheep (Couvillion et al., 1980; Jessup, 1985; Savini et al., 2011; Stallknecht et al., 1995). EHDV is primarily subacute in cattle with occasional outbreaks of clinical disease (Stallknecht et al., 1995). EHDV is a significant problem for the livestock industries in the Mediterranean Basin, especially Israel, Morocco, Algeria, Tunisia, Jordan and Turkey (Savini et al., 2011; Yadin et al., 2008). EHDV is both genetically and morphologically similar to Bluetongue virus (BTV), the type species of the genus that causes disease in ovine, bovine and other species (Jessup, 1985; Savini et al., 2011; Stallknecht et al., 1995; Yadin et al., 2008). Both EHDV and BTV contain a ten-segmented The GenBank accession numbers for the genomic RNA segments sequences of representative EHDV are KF570113-KF570142. Three supplementary figures and one supplementary table are available with the online version of this paper.
434
double-stranded RNA genome (Savini et al., 2011). Each segment encodes a single protein, with a total of seven structural proteins (VP1 to VP7) and four non-structural proteins (NS1, NS2, NS3 and NS3a) (Anthony et al., 2009a, b; Mecham & Dean, 1988; Savini et al., 2011). The virion outer layer is composed of VP2 and VP5 trimers that are involved in viral attachment and penetration of the host cell (Anthony et al., 2009c; Mecham & Dean, 1988; Savini et al., 2011). VP2 and VP5 are the most variable viral proteins and specificity of their interactions with neutralizing antibodies (especially VP2) determines the virus serotype (Anthony et al., 2009c; Mecham & Dean, 1988; Savini et al., 2011). VP7 forms the outer core layer and provides a surface for VP2 and VP5 attachment (Anthony et al., 2009a; Mecham & Dean, 1988; Savini et al., 2011). VP3 forms the inner subcore layer and surrounds VP1, VP4 and VP6 as well as the ten linear dsRNA segments (Anthony et al., 2009a; Mecham & Dean, 1988; Savini et al., 2011). VP3 is an RNA-binding protein which interacts with both the viral RNA genome and the minor proteins VP1, VP4 and VP6 (Anthony et al., 2009a; Mecham & Dean, 1988; Savini et al., 2011). Self-assembly of VP3 controls the size and organization of the capsid structure (Anthony et al., 2009a). VP1, a highly conserved protein, functions as a viral RNA-dependent RNA polymerase (RdRP) (Anthony et al., 2009a; Mecham & Dean, 1988; Savini et al., 2011). 059659 G 2014 SGM Printed in Great Britain
Whole genome analysis of EHDV
VP4 is the capping enzyme, and VP6 is believed to be the viral helicase (Anthony et al., 2009a; Mecham & Dean, 1988; Savini et al., 2011). Based on the antigenic and genetic analyses of the two outer capsid proteins (VP2 and VP5), eight serotypes of EHDV are proposed (Campbell & St George, 1986). EHDV serotypes 3 and 4 are reported in Africa (Allison et al., 2010; Lee et al., 1974; Moore, 1974; Moore & Kemp, 1974). EHDV serotypes 5, 6, 7 and 8 are reported in Australia (Allison et al., 2010; St George et al., 1983). EHDV-6 has been identified in Bahrain, the Sultanate of Oman, Sudan, Morocco and Algeria (Mohammed et al., 1996). Historically, in the USA only two serotypes, EHDV-1 (New Jersey strain) and EHDV-2 (Alberta strain), were identified (Allison et al., 2010). However, in 2006, a novel reassortant serotype 6 EHDV containing two variable surface antigens (VP2 and VP5) from an exotic EHDV-6 with the remaining segments derived from endemic EHDV-2 was isolated from moribund and dead white-tailed deer in Indiana and Illinois (Allison et al., 2010). Subsequently, reassortant type 6 EHDV was identified in several other states demonstrating that the virus is geographically widespread (Allison et al., 2010). This strain is currently endemic in the USA (Allison et al., 2010). Reassortment between native and exotic EHDV illustrates the plasticity of EHDV and their ability to become established in new environmental niches (Allison et al., 2010). EHDV infection can be very debilitating and is often fatal in wild ungulates, including white-tailed deer, mule deer and antelope (Savini et al., 2011). However, EHDV infection is generally not apparent or very mild in livestock (Savini et al., 2011). Despite this, numerous EHDV outbreaks and its introduction to new areas suggest EHDV is clinically and economically important (Savini et al., 2011). Consequently, surveillance is warranted in livestock and wild deer populations. Whole genome sequencing of the EHDV isolated from clinical submissions to the Newport Laboratories over a 4 year period was performed to assess EHDV genetic diversity and population dynamics.
RESULTS AND DISCUSSION Whole genome sequencing and phylogenetic analysis Tissue and blood samples were collected from white-tailed deer exhibiting clinical signs of EHDV infection and submitted to Newport Laboratories from 2008 to 2012 (Table 1). EHDV positive samples were identified using real-time quantitative reverse transcription PCR and cultured on baby hamster kidney (BHK) cells (Clavijo et al., 2010). Forty-four viruses were isolated over this time period and subjected to genome sequencing. Predicted serotypes were determined by homology of VP2 and VP5 genes to reference strains. Type 2 viruses were the most prevalent serotype over http://vir.sgmjournals.org
this time period, accounting for 57 % of the isolates (Table 2). Type 6 viruses represented 32 % of our samples; however, 63 % of EHDV were isolated in 2012. Type 1 was least prevalent, at 11 %, and was not identified in 2012. The results suggest that type 1 was not widespread in the sample set used for the studies. Although type 2 was prevalent, type 6 viruses became dominant in 2012 in the sampling area. Type 2 viruses were identified in all regions sampled and type 6 viruses were also widespread, with isolates from Texas and the Midwest. Type 1 viruses were only identified in Texas and Oklahoma. Co-circulation of type 1, 2 and 6 viruses was seen in Texas in 2010 and 2011, the state most heavily represented in our sample set. Phylogenetic analysis (VP2, VP5 and VP7) Phylogenetic comparisons were constructed in MEGA version 5.05 using the maximum-likelihood algorithm and bootstrapped using 1000 replicates. Phylogenetic trees of all ten segments of EHDV isolates are shown in Figs 1, S1, S2 and S3 (available in JGV Online). Comparison of the VP2 sequences to reference viruses identified 97.3–98.4 % nucleotide identity between type 2 viruses, 96.6298.7 % identity between type 1 viruses and 97.4–97.7 % identity between type 6 viruses (Table S1). Although we observed greater than 96 % identity between VP2 gene sequences within each serotype, there was only 47–53 % identity between type 1, 2 and 6 viruses. Similarly, comparisons of complete VP5 sequences to the published reference viruses identified 96.0–97.5 % nucleotide identity between type 2 viruses, 97.7–98.3 % identity between type 6 viruses and 96.4–96.9 % identity between type 1 viruses (Table S1). Inter-subtype sequence identity was only 64–69 %. Likewise, comparisons of complete VP7 sequences to the published reference viruses identified 97.8–98.6 % nucleotide identity between type 2 isolates, 97 % identity between type 1 isolates and only 78–80 % identity between type 1 and type 2 viruses (Table S1). Moreover, all type 1 and type 2 viruses demonstrates preferential reassortment for VP2, VP5 and VP7 within the homologous serotype (Fig. 2). In the virion, VP2 and VP5 are arranged on the outer core layer formed by VP7. Preferential reassortment observed for type 1 and type 2 viruses suggests incompatibility between the surface proteins VP2 and VP5 and the outer core protein VP7 from the heterologous serotype. Similar to previous studies, the results also demonstrate that VP2 and VP5 from type 6 are compatible with VP7 from serotype 2. In contrast, reassortment by another orbivirus, BTV, was shown to freely occur for all segments (Shaw et al., 2013). Alternate explanations for the observed preferential reassortment could be the limited size of our sample set prohibiting us from identifying all genotypes as well as lack of co-infection with multiple types preventing reassortment in vivo. Co-circulation of type 1, 2 and 6 viruses in Texas in 2010 and 2011 demonstrates that all three types were present in the same state at the same time, 435
S. Anbalagan and others
Table 1. EHDV isolates used in this study Isolate designation (case number) 8-2592-1 8-3707-1 9-2201-1 9-2317-1 9-2742-1 9-2778-1 9-2889-1 9-3132-1 9-3326-1 9-3397-1 9-3466-1 10-0235-1 10-2522-1 10-2604-1 10-2604-2 10-3180-1 10-3456-1 11-0276-2 11-1767-1 11-2140-1 11-2280-1 11-2297-1 11-2507-1 11-2778-2 11-3099-1 11-3174-1 11-3245-1 11-3366-1 12-2980-1 12-3394-1 12-3394-2 12-3422-1 12-3437-8 12-3469-1 12-3583-1 12-3587-1 12-3600-2 12-3960-1 12-3961-1 12-4087-3 12-4087-4 12-4087-5 12-4087-9 12-4350-1
State
Sample type*
Year
Oklahoma Iowa Texas Texas Indiana Missouri Texas Massachusetts Oklahoma Oklahoma Florida Texas Texas Texas Texas Kentucky Missouri Texas Texas Minnesota Louisiana Florida Texas Missouri Missouri Pennsylvania Texas Texas Texas Kansas Kansas Missouri Ohio Ohio Texas Missouri Illinois Texas Missouri Illinois Illinois Illinois Illinois Texas
Blood Tissue homogenate Spleen Spleen Spleen Spleen Spleen Spleen Spleen Spleen Spleen Spleen Blood Spleen Spleen Lung, liver Spleen Spleen Spleen Blood swabs Spleen Spleen Liver, spleen Spleen Spleen Spleen Spleen Spleen Spleen Kidney Kidney Spleen Spleen Lung, spleen, lymph node Spleen Spleen Spleen Spleen Spleen
2008 2008 2009 2009 2009 2009 2009 2009 2009 2009 2009 2010 2010 2010 2010 2010 2010 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2012 2012 2012 2012 2012 2012 2012 2012 2012 2012 2012 2012 2012 2012 2012 2012
NA NA NA NA
Spleen
*NA, Not available.
theoretically allowing for reassortment. For EHDV, reverse genetics is needed to conclusively demonstrate preferential reassortment. High genetic identity (.94 %) was observed between type 1 and type 2 reference viruses for segments encoding VP1, VP3, VP4, VP6, NS1, NS2 and NS3, preventing conclusive identification of reassortment events between these segments for type 1 and type 2 viruses (Table S1). Although 436
the phylogenetic analysis showed subclades which are supported by strong bootstrap values, the branch lengths among isolates are very short, indicating a very low level of genetic variation among the segments (Figs S1, S2 and S3). Overall the results indicate that within type 1 and type 2 viruses, unlike VP2, VP5 and VP7 there are no serotypespecific genetic traits common to segments encoding VP1, VP3, VP4, VP6, NS1, NS2 and NS3 suggesting that genes Journal of General Virology 95
Whole genome analysis of EHDV
Table 2. Number of EHDV serotypes isolated each year Serotype Year
1
2
6
2008 2009 2010 2011 2012
0 1 2 2 0
2 7 3 7 6
0 1 1 2 10
encoding internal proteins are evolving under different selective pressure. Earlier studies of EHDV from multiple hosts species (Culicoides, deer and bovine) and a wide geographical region in the USA (CA, ID, NE, LA, TX, WY) demonstrated high genetic identity (.97 %) among strains (Cheney et al., 1996; Murphy et al., 2005). Our dataset included 44 isolates of EHDV from wide geographical region in the United States (TX, IA, OK, IN, MA, FL, MO, MN, LA, PA, KS, OH, IN, KY) and similar to previous studies, we observed high genetic identity among the isolates (Murphy et al., 2005). These results were somewhat surprising as EHDV, an RNA virus, has a viral RNA polymerase that lacks a proof-reading mechanism (Holland et al., 1982). Additionally, the segmented genome facilitates reassortment within the species. However, our finding supports a growing body of evidence that EHDV, similar to many orbiviruses, is genetically stable (Murphy et al., 2005). In the mosquito-borne Yellow fever virus, a single-stranded RNA virus, there was only a 6.8 % variation among isolates collected over 45 years from a region where transmission is seasonal, multiple species are involved and regular epidemics of diseases in humans occur (Mutebi et al., 2001). Similar observations were also made for another single-stranded RNA virus, vesicular stomatitis virus, which exists in an enzootic cycle, suggesting that there is genetic stability in this virus (Stallknecht et al., 1985). Furthermore, Jenkins et al. (2002) analysed nucleotide substitution rates among 50 RNA viruses and postulated that selection pressure in multiple hosts contributes to slightly lower rates of evolution among vector-borne viruses when compared to non-vector-borne viruses. Such selection pressure might be responsible for less genetic diversity where mutations can disrupt the balance between efficient virus replication and maximal transmission. Moreover, the double-stranded RNA genome of BTV and EHDV likely plays a role in viral genetic stability. Since EHDV is a vector-borne virus, the high level of genetic identity between the isolates suggests a stable host–virus relationship. Serum neutralization assay A subset of EHDV isolates from each serotype was randomly selected and assayed by serum neutralization http://vir.sgmjournals.org
with reference antiserum (Table 3). The results indicate that the serotype determined by genetic analysis (VP2 and VP5 genes) is in agreement with results from serum neutralization assays. As predicted by the genetic analysis, there was little antigenic variation within serotype. Overall, this study found evidence for co-circulation of type 1, type 2 and type 6 in North American white-tailed deer with an increasing prevalence of type 6 in 2012. The results also indicate that VP2, VP5 and VP7 from type 1 and type 2 viruses preferentially reassort within homologous serotypes while for type 6 viruses, this restriction applies only to VP2 and VP5 from type 6 viruses reassorting with VP7 from type 2 viruses. Consistent with other studies, EHDV demonstrated little genetic and antigenic diversity, indicating a stable host–virus relationship.
METHODS Viruses. EHDVs (Table 1) used in this study were isolated from samples submitted to Newport Laboratories for isolation and genetic analysis from 2008 to 2012 as a part of the routine diagnostic testing. The samples were taken from naturally infected animals in the field, by qualified veterinarians, as a part of normal veterinary care and diagnostic testing procedures. Virus growth. Virus was isolated either from tissue or blood samples.
Filtrates from the blood samples (either whole blood or plasma) were used to inoculate BHK cells. Tissue samples were mechanically homogenized in 1 ml Dulbecco’s Modified Eagle’s Medium (DMEM) (Pellgro) containing Cipro (10 mg l21) (Invivogen). The samples were centrifuged at 5000 g for 10 min to remove debris. The supernatant containing the virus was used to inoculate 3-day-old BHK cells propagated in DMEM (Pellgro) with either plasmocin (25 mg l21) (Invivogen) or normocin (100 mg l21) (Invivogen) at 37 uC with 5 % CO2. The cultures were incubated at 37 uC in a CO2 incubator and inspected daily for cytopathic changes (CPE). If CPE was noted, mediim was collected and analysed for EHDV by real-time quantitative reverse transcription PCR using EHDV-specific primers and probe (data not shown). EHDV positive cultures were harvested for serum neutralization testing and RNA extraction. RNA isolation. BHK cells that showed 100 % CPE following EHDV
virus infection were used for RNA extraction. Cell culture supernatant (20 ml) was filtered using 0.2 mm bottle-top filters (Thermo Scientific). The filtrate was centrifuged at 50 000 g for 2 h. Supernatant was discarded and the pellet was suspended in 1000 ml water. Samples were concentrated to a final 100 ml volume using Amicon ultracentrifugal filters (0.5 ml, 50 kDa) (Millipore). Cellular DNA and RNA were removed by incubation with DNase I (25 units) [New England Biolabs (NEB)] and RNase A (25 units) (Qiagen) at 37 uC for 1 h. RNA was extracted using QiaAmp viral RNA isolation kit (Qiagen) and eluted in 35 ml of the supplied elution buffer. Genome amplification. Whole genome sequence of EHDV virus was obtained by sequence-independent cDNA synthesis methodology as described by Potgieter et al. (2009), with some modifications. In brief, 10 ml dsRNA was ligated to PC3-T7 loop (200 ng) (59-p– GGATCCCGGGAATTCGGTAATACGACTCACTATATTTTTATAGTGAGTCGTATTA–OH-39) in T4 RNA ligase buffer (NEB), 1 mM ATP (NEB), 10 % DMSO (Sigma Aldrich), 20 % polyethyleneglycol (PEG6000) (Sigma Aldrich), 40 U RNasin (Promega) and 30 U T4 RNA ligase (NEB) in a final volume of 50 ml. The reaction mix was incubated at 16 uC overnight. Ligated dsRNA was purified using RNA
437
VP5_12-3960-1
VP2 9-3466-1
99
VP7_9-2742-1 VP7_9-2778-1
0.005
VP5_11-2507-1
VP7_9-2889-1
VP2 11-2507-1
VP5_11-2280-1
VP7_9-3132-1
VP2 10-2522-1
VP5_10-2522-1
VP7_9-3326-1
VP2 10-3180-1
VP7_9-3466-1
VP2 9-3326-1
VP5_9-3326-1 VP5_9-3132-1
VP2 11-2778-2
VP5_9-2742-1
VP7_10-2604-2
VP2 11-2280-1
VP5_8-3707-1
VP7_10-3180-1
91
VP7_10-3456-1
VP5_9-3466-1
VP7_11-1767-1
VP2 Serotype 2 Alberta
VP5_10-3456-1
VP7_11-2140-1
VP2 12-4350-1
VP5_11-2297-1
VP7_11-2280-1
63 VP5_11-3174-1 100 VP5_12-3422-1
VP7_11-2778-2
VP5_12-3469-1
97 VP7_11-3099-1
VP2 9-2201-1
VP2 12-2980-1 60 VP2 9-3132-1 78
VP2 11-3245-1
VP2 12-3960-1 97 VP2 11-3366-1 86 89 VP2 9-3397-1 VP2 11-0276-2 59 VP2 10-2604-1 100 VP2 10-0235-1 VP2 Serotype1 NewJersey
61
VP7_12-3394-2
VP5_Serotype_2_Alberta
VP7_12-3422-1
VP5_8-2592-1
100
VP7_12-3600-2
VP5_12-3583-1
VP7_12-3960-1
VP5_10-2604-2
VP7_12-4087-3
VP5_9-2317-1
VP7_12-4087-4
VP5_11-1767-1
VP7_12-4087-9
VP5_12-3394-1
VP7_12-4350-1
VP5_12-3437-8
VP7_Serotype_2_Alberta
VP5_12-3587-1
81
VP7_12-3583-1
VP5_12-3600-2
VP7_12-3437-8
VP5_12-3961-1
61
VP5_12-4087-3 VP5_12-4087-4
VP5_10-2604-1 100
Journal of General Virology 95
79 58
VP5_11-3366-1 VP5_9-3397-1 VP5_10-0235-1
VP7_12-3961-1 VP7_11-3245-1
VP7_Serotype_6_CSIRO753 VP7_Serotype_1_NewJersey
VP5_Serotype_1_NewJersey VP5_11-0276-2
VP7_12-4087-5 VP7_11-2297-1
VP5_12-4087-5 VP5_12-4087-9
64 VP2 12-4087-3
VP2 9-2317-1
VP7_12-3587-1
VP5_Serotype_6_CSIRO 753
VP2 12-4087-4
62
VP7_12-3469-1
VP5_11-3099-1
VP2 12-3437-8
VP2 10-2604-2
VP7_12-3394-1
VP5_9-2889-1
77 VP2 12-3394-1 56 VP2 12-4087-9 56 VP2 12-3600-2
VP2 12-3587-1
VP7_12-2980-1
VP5_9-2201-1
VP2 11-3099-1
VP2 11-1767-1
VP7_11-3174-1
VP5_11-2140-1
VP2 Serotype 6 CSIRO 753 VP2 12-3583-1 100 93 VP2 12-4087-5 57 70 VP2 12-3961-1
VP7_11-2507-1
VP5_10-3180-1
VP2 11-2140-1
VP2 9-2889-1 93 VP2 12-3394-2
100
VP7_10-2522-1
VP5_9-2778-1
VP2 8-2592-1
95
VP7_9-2317-1
VP5_11-3245-1
VP2 9-2778-1
VP2 9-2742-1 78 VP2 8-3707-1
100
VP7_9-2201-1
VP5_12-2980-1 VP5_11-2778-2
0.05
VP7_8-2592-1
VP7
VP5_12-3394-2
VP2 10-3456-1 0.2
VP7_8-3707-1
VP5_12-4350-1
VP5
VP7_11-3366-1 VP7_11-0276-2
100 91
VP7_10-2604-1 VP7_10-0235-1 VP7_9-3397-1
Fig. 1. Maximum-likelihood phylogenetic trees of VP2, VP5 and VP7 nt sequence from EHDV isolates and the reference strains available in GenBank. The numbers indicate bootstrap confidence values after 1000 replications.
S. Anbalagan and others
438 57 VP2 12-3422-1 58 VP2 11-3174-1 VP2 12-3469-1 75 VP2 11-2297-1
VP2
Whole genome analysis of EHDV
VP2 Serotype 1 Serotype 2 Serotype 6
VP5
VP7
Fig. 2. VP2, VP5 and VP7 constellation for EHDV isolates as determined by phylogenetic analysis (see Figs 1, S1, S2 and S3). The isolate number is listed in the first column. The numbers prior to the first hyphen denote the year of isolation. The protein encoded by each gene is listed in the top row.
8-2592-1 8-3707-1 9-2201-1 9-2317-1 9-2742-1 9-2778-1 9-2889-1 9-3132-1 9-3326-1 9-3397-1 9-3466-1 10-0235-1 10-2522-1 10-2604-1 10-2604-2 10-3180-1 10-3456-1 11-0276-2 11-1767-1 11-2140-1 11-2280-1 11-2297-1 11-2507-1 11-2778-2 11-3099-1 11-3174-1 11-3245-1 11-3366-1 12-2980-1 12-3394-1 12-3394-2 12-3422-1 12-3437-8 12-3469-1 12-3583-1 12-3587-1 12-3600-2 12-3960-1 12-3961-1 12-4087-3 12-4087-4 12-4087-5 12-4087-9 12-4350-1
http://vir.sgmjournals.org
Clean & Concentrator (Zymo Research) according to the manufacturer’s instructions. To the purified ligated dsRNA, DMSO (15 % (v/ v) final concentration) was added and incubated in a thermal cycler at 95 uC for 5 min and 4 uC for 5 min. RNA was reverse transcribed to cDNA using the GoScript Reverse Transcription system (Promega). The reaction was incubated in a thermal cycler at 25 uC for 15 min, 42 uC for 60 min and 5 uC for 15 min. RNA was removed using RNaseH (2.5 units, NEB) and incubated in a thermal cycler at 37 uC for 20 min. The cDNA was annealed at 65 uC for 4 h. cDNA was amplified using primer PC2 (59-p–CCGAATTCCCGGGATCC-39) in a PCR mixture containing 16 Ex Taq buffer, 0.2 mM dNTPs (NEB), 5 ml cDNA and 2.5 U TaKaRa Ex Taq (Clontech). The samples were incubated at 72 uC for 1 min. This was followed by an initial denaturation step of 94 uC for 2 min followed by 40 cycles of 94 uC for 30 s, 67 uC for 30 s and 72 uC for 4 min. A final extension step of 72 uC for 5 min was included. Amplified products were visualized after separation on 1.5 % agarose gels (TBE) containing ethidium bromide. Sequencing and data analysis. Amplified products were sequenced using Ion Personal Genome Machine (PGM) sequencing platform (Life Technologies). In brief, the amplified products were fragmented, adaptor ligated and amplified using NEB Next Fast DNA Fragmentation and Library Prep Set 4 (NEB) according to the manufacturer’s instructions. Fragments of sizes 300–330 bp were selected using AMPure XP bead size selection (Beckman Coulter) procedure according to the manufacturer’s instructions. Quality of the amplified adaptor ligated DNA was checked using Qubit dsDNA HS Assay kit (Life Technologies). The samples were pooled in equal molar concentrations and the emulsion-based PCR was conducted using the Ion OneTouch 2 System (Life Technologies) according to the manufacturer’s instructions. Percentage of templated ion sphere particles was checked using Ion Sphere Quality Control kit (Life Technologies). The templated ion spheres were loaded on the 314 chip (Life Technologies) and sequenced on the Ion Torrent Personal Genome Machine (PGM) system using the Ion PGM Template OT2 200 kit v2 (Life Technologies) according to the manufacturer’s instructions. Sequence reads were assembled into contigs using the SeqMan NGen program (DNASTAR). Sequence data of each isolate were analysed using MEGA (version 5.05). Phylogenetic comparisons of VP1, VP2, VP3, VP4, VP5, VP6, VP7, NSP1, NSP2 and NSP3 nt sequences were conducted using maximum-likelihood analysis. Bootstrap confidence values were determined using 1000 replicates. Representative type 1, 2 and 6 viruses were submitted to GenBank under accession numbers KF570113 (type 1, VP1), KF570114 (type 1, VP2), KF570115 (type 1, VP3), KF570116 (type 1, VP4), KF570117 (type 1, VP5), KF570118 (type 1, VP6), KF570119 (type 1, VP7), KF570120 (type 1, NS1), KF570121 (type 1, NS2), KF570122 (type 1, NS3), KF570123 (type 2, VP1), KF570124 (type 2, VP2), KF570125 (type 2, VP3), KF570126 (type 2, VP4), KF570127 (type 2, VP5), KF570128 (type 2, VP6), KF570129 (type 2, VP7), KF570130 (type 2, NS1), KF570131 (type 2, NS2), KF570132 (type 2, NS3), KF570133 (type 6, VP1), KF570134 (type 6, VP2), KF570135 (type 6, VP3), KF570136 (type 6, VP4), KF570137 (type 6, VP5), KF570138 (type 6, VP6), KF570139 (type 6, VP7), KF570140 (type 6, NS1), KF570141 (type 6, NS2), KF570142 (type 6, NS3).
439
S. Anbalagan and others
Table 3. Serum neutralization titres of randomly selected EHDV isolates against serotype reference antisera Serum neutralization titre Isolate
Serotype (VP2)*
Type 1
Type 2
Type 6
2 6 2 1 1 1 2 2 2 2 6
¡4 91 ,4 181 91 512 ¡4 ¡4 ¡4 ¡4 ¡4
16 91 11 ¡4 ¡4 ¡4 23 32 23 91 ,4
¡4 512 ¡4 ¡4 ¡4 ¡4 ¡4 ¡4 ¡4 ¡4 362
8-3707-1 9-2317-1 9-2778-1 9-3397-1 10-0235 10-2604-1 11-2297-1 11-2507-1 12-3394-2 12-3960-1 12-3961-1
Campbell, C. H. & St George, T. D. (1986). A preliminary report of a
comparison of epizootic haemorrhagic disease viruses from Australia with others from North America, Japan and Nigeria. Aust Vet J 63, 233. Cheney, I. W., Yamakawa, M., Roy, P., Mecham, J. O. & Wilson, W. C. (1996). Molecular characterization of the segment 2 gene of epizootic
hemorrhagic disease virus serotype 2: gene sequence and genetic diversity. Virology 224, 555–560. Clavijo, A., Sun, F., Lester, T., Jasperson, D. C. & Wilson, W. C. (2010).
An improved real-time polymerase chain reaction for the simultaneous detection of all serotypes of epizootic hemorrhagic disease virus. J Vet Diagn Invest 22, 588–593. Couvillion, C. E., Jenney, E. W., Pearson, J. E. & Coker, M. E. (1980).
Survey for antibodies to viruses of bovine virus diarrhea, bluetongue, and epizootic hemorrhagic disease in hunter-killed mule deer in New Mexico. J Am Vet Med Assoc 177, 790–791. Holland, J., Spindler, K., Horodyski, F., Grabau, E., Nichol, S. & VandePol, S. (1982). Rapid evolution of RNA genomes. Science 215,
1577–1585. *Serotype predicted based on phylogenetic analysis of the VP2 gene.
Jenkins, G. M., Rambaut, A., Pybus, O. G. & Holmes, E. C. (2002).
Rates of molecular evolution in RNA viruses: a quantitative phylogenetic analysis. J Mol Evol 54, 156–165. Serum neutralization assay. Antisera were diluted in DMEM
serially twofold beginning at 1 : 2 and ending at 1 : 256 in a 96-well micro titre plate. Viruses were grown on BHK cells as described above. Viruses were diluted in DMEM to 200 TCID50 per 100 ml as determined by titration on a monolayer of BHK cells. Diluted virus (100 ml) was added to every well of the diluted serum. The virus and antiserum mixture was incubated at 37 uC for 1 h in a humidified incubator with ±5 % CO2. To the virus and antiserum mixture, 150 ml Vero (African green monkey) cells (105 cells ml21) in DMEM supplemented with 5 % FCS was added. Controls consisting of cells only and cell plus antiserum were also included. Plates were incubated for 4–7 days at 37 uC in a humidified incubator with ±5 % CO2. Cells were visualized using an inverted microscope. Each well was observed for typical CPE and the wells were scored either positive or negative. Wells without CPE were considered positive for the presence of antibody. End-point titres were calculated by the reciprocal of the maximum dilution showing no CPE. The assay was considered valid only when the back titration average fell within 50 and 300 TCID50 per well.
REFERENCES Allison, A. B., Goekjian, V. H., Potgieter, A. C., Wilson, W. C., Johnson, D. J., Mertens, P. P. & Stallknecht, D. E. (2010). Detection of a novel
Jessup, D. A. (1985). Epidemiology of two orbiviruses in California’s
native wild ruminants: preliminary report. Prog Clin Biol Res 178, 53–65. Lee, V. H., Causey, O. R. & Moore, D. L. (1974). Bluetongue and
related viruses in Ibadan, Nigeria: isolation and preliminary identification of viruses. Am J Vet Res 35, 1105–1108. Mecham, J. O. & Dean, V. C. (1988). Protein coding assignment for
the genome of epizootic haemorrhagic disease virus. J Gen Virol 69, 1255–1262. Mohammed, M. E., Aradaib, I. E., Mukhtar, M. M., Ghalib, H. W., Riemann, H. P., Oyejide, A. & Osburn, B. I. (1996). Application of
molecular biological techniques for detection of epizootic hemorrhagic disease virus (EHDV-318) recovered from a sentinel calf in central Sudan. Vet Microbiol 52, 201–208. Moore, D. L. (1974). Bluetongue and related viruses in Ibadan,
Nigeria: serologic comparison of bluetongue, epizootic hemorrhagic disease of deer, and Abadina (Palyam) viral isolates. Am J Vet Res 35, 1109–1113. Moore, D. L. & Kemp, G. E. (1974). Bluetongue and related viruses in Ibadan, Nigeria: serologic studies of domesticated and wild animals. Am J Vet Res 35, 1115–1120. Murphy, M. D., Howerth, E. W., MacLachlan, N. J. & Stallknecht, D. E. (2005). Genetic variation among epizootic hemorrhagic disease viruses in
reassortant epizootic hemorrhagic disease virus (EHDV) in the USA containing RNA segments derived from both exotic (EHDV-6) and endemic (EHDV-2) serotypes. J Gen Virol 91, 430–439.
the southeastern United States: 1978–2001. Infect Genet Evol 5, 157– 165.
Anthony, S. J., Maan, N., Maan, S., Sutton, G., Attoui, H. & Mertens, P. P. (2009a). Genetic and phylogenetic analysis of the core proteins
Phylogenetic and evolutionary relationships among yellow fever virus isolates in Africa. J Virol 75, 6999–7008.
VP1, VP3, VP4, VP6 and VP7 of epizootic haemorrhagic disease virus (EHDV). Virus Res 145, 187–199.
Potgieter, A. C., Page, N. A., Liebenberg, J., Wright, I. M., Landt, O. & van Dijk, A. A. (2009). Improved strategies for sequence-independent
Anthony, S. J., Maan, N., Maan, S., Sutton, G., Attoui, H. & Mertens, P. P. (2009b). Genetic and phylogenetic analysis of the non-structural
amplification and sequencing of viral double-stranded RNA genomes. J Gen Virol 90, 1423–1432.
proteins NS1, NS2 and NS3 of epizootic haemorrhagic disease virus (EHDV). Virus Res 145, 211–219.
Savini, G., Afonso, A., Mellor, P., Aradaib, I., Yadin, H., Sanaa, M., Wilson, W., Monaco, F. & Domingo, M. (2011). Epizootic haemorragic
Anthony, S. J., Maan, S., Maan, N., Kgosana, L., BachanekBankowska, K., Batten, C., Darpel, K. E., Sutton, G., Attoui, H. & Mertens, P. P. (2009c). Genetic and phylogenetic analysis of the
Shaw, A. E., Ratinier, M., Nunes, S. F., Nomikou, K., Caporale, M., Golder, M., Allan, K., Hamers, C., Hudelet, P. & other authors (2013).
outer-coat proteins VP2 and VP5 of epizootic haemorrhagic disease virus (EHDV): comparison of genetic and serological data to characterise the EHDV serogroup. Virus Res 145, 200–210.
Reassortment between two serologically unrelated bluetongue virus strains is flexible and can involve any genome segment. J Virol 87, 543–557.
440
Mutebi, J.-P., Wang, H., Li, L., Bryant, J. E. & Barrett, A. D. T. (2001).
disease. Res Vet Sci 91, 1–17.
Journal of General Virology 95
Whole genome analysis of EHDV
St George, T. D., Cybinski, D. H., Standfast, H. A., Gard, G. P. & DellaPorta, A. J. (1983). The isolation of five different viruses of the epizootic
Stallknecht, D. E., Nettles, V. F., Rollor, E. A., III & Howerth, E. W. (1995).
haemorrhagic disease of deer serogroup. Aust Vet J 60, 216–217.
Epizootic hemorrhagic disease virus and bluetongue virus serotype distribution in white-tailed deer in Georgia. J Wildl Dis 31, 331–338.
Stallknecht, D. E., Nettles, V. F., Fletcher, W. O. & Erickson, G. A. (1985). Enzootic vesicular stomatitis New Jersey type in an insular
Yadin, H., Brenner, J., Bumbrov, V., Oved, Z., Stram, Y., Klement, E., Perl, S., Anthony, S., Maan, S. & other authors (2008). Epizootic haemorrhagic
feral swine population. Am J Epidemiol 122, 876–883.
disease virus type 7 infection in cattle in Israel. Vet Rec 162, 53–56.
http://vir.sgmjournals.org
441
DOI 10.1099/vir.0.059659-0
Whole genome analysis of epizootic hemorrhagic disease virus identified limited genome constellations and preferential reassortment Srivishnupriya Anbalagan, Elyse Cooper, Pat Klumper, Randy R. Simonson and Ben M. Hause Correspondence
Newport Laboratories, 1520 Prairie Drive, Worthington, Minnesota 56187, USA
Srivishnupriya Anbalagan [email protected] Ben M. Hause [email protected]
Received 18 September 2013 Accepted 3 November 2013
Epizootic hemorrhagic disease virus (EHDV) is a Culicoides transmitted orbivirus that causes haemorrhagic disease in wild and domestic ruminants. A collection of 44 EHDV isolated from 2008 to 2012 was fully sequenced and analysed phylogenetically. Serotype 2 viruses were the dominant serotype all years except 2012 when serotype 6 viruses represented 63 % of the isolates. High genetic similarity (.94 % identity) between serotype 1 and 2 virus VP1, VP3, VP4, VP6, NS1, NS2 and NS3 segments prevented identification of reassortment events for these segments. Additionally, there was little genetic diversity (.96 % identity) within serotypes for VP2, VP5 and VP7. Preferential reassortment within the homologous serotype was observed for VP2, VP5 and VP7 segments for type 1 and type 2 viruses. In contrast, type 6 viruses were all reassortants containing VP2 and VP5 derived from an exotic type 6 with the remaining segments most similar to type 2 viruses. These results suggest that reassortment between type 1 and type 2 viruses requires conservation of the VP2, VP5 and VP7 segment constellation while type 6 viruses only require VP2 and VP5 and are restricted to type 2-lineage VP7. As type 6 VP2 and VP5 segments were exclusively identified in viruses with type 2-derived VP7, these results suggest functional complementation between type 2 and type 6 VP7 proteins.
INTRODUCTION Epizootic haemorrhagic disease is an infectious disease caused by viruses belonging to the species Epizootic hemorrhagic disease virus (EHDV) within the genus Orbivirus and family Reoviridae (Savini et al., 2011). The virus is transmitted by insects of the genus Culicoides (Savini et al., 2011). EHDV periodically causes disease in white-tailed deer, pronghorn antelope and desert bighorn sheep (Couvillion et al., 1980; Jessup, 1985; Savini et al., 2011; Stallknecht et al., 1995). EHDV is primarily subacute in cattle with occasional outbreaks of clinical disease (Stallknecht et al., 1995). EHDV is a significant problem for the livestock industries in the Mediterranean Basin, especially Israel, Morocco, Algeria, Tunisia, Jordan and Turkey (Savini et al., 2011; Yadin et al., 2008). EHDV is both genetically and morphologically similar to Bluetongue virus (BTV), the type species of the genus that causes disease in ovine, bovine and other species (Jessup, 1985; Savini et al., 2011; Stallknecht et al., 1995; Yadin et al., 2008). Both EHDV and BTV contain a ten-segmented The GenBank accession numbers for the genomic RNA segments sequences of representative EHDV are KF570113-KF570142. Three supplementary figures and one supplementary table are available with the online version of this paper.
434
double-stranded RNA genome (Savini et al., 2011). Each segment encodes a single protein, with a total of seven structural proteins (VP1 to VP7) and four non-structural proteins (NS1, NS2, NS3 and NS3a) (Anthony et al., 2009a, b; Mecham & Dean, 1988; Savini et al., 2011). The virion outer layer is composed of VP2 and VP5 trimers that are involved in viral attachment and penetration of the host cell (Anthony et al., 2009c; Mecham & Dean, 1988; Savini et al., 2011). VP2 and VP5 are the most variable viral proteins and specificity of their interactions with neutralizing antibodies (especially VP2) determines the virus serotype (Anthony et al., 2009c; Mecham & Dean, 1988; Savini et al., 2011). VP7 forms the outer core layer and provides a surface for VP2 and VP5 attachment (Anthony et al., 2009a; Mecham & Dean, 1988; Savini et al., 2011). VP3 forms the inner subcore layer and surrounds VP1, VP4 and VP6 as well as the ten linear dsRNA segments (Anthony et al., 2009a; Mecham & Dean, 1988; Savini et al., 2011). VP3 is an RNA-binding protein which interacts with both the viral RNA genome and the minor proteins VP1, VP4 and VP6 (Anthony et al., 2009a; Mecham & Dean, 1988; Savini et al., 2011). Self-assembly of VP3 controls the size and organization of the capsid structure (Anthony et al., 2009a). VP1, a highly conserved protein, functions as a viral RNA-dependent RNA polymerase (RdRP) (Anthony et al., 2009a; Mecham & Dean, 1988; Savini et al., 2011). 059659 G 2014 SGM Printed in Great Britain
Whole genome analysis of EHDV
VP4 is the capping enzyme, and VP6 is believed to be the viral helicase (Anthony et al., 2009a; Mecham & Dean, 1988; Savini et al., 2011). Based on the antigenic and genetic analyses of the two outer capsid proteins (VP2 and VP5), eight serotypes of EHDV are proposed (Campbell & St George, 1986). EHDV serotypes 3 and 4 are reported in Africa (Allison et al., 2010; Lee et al., 1974; Moore, 1974; Moore & Kemp, 1974). EHDV serotypes 5, 6, 7 and 8 are reported in Australia (Allison et al., 2010; St George et al., 1983). EHDV-6 has been identified in Bahrain, the Sultanate of Oman, Sudan, Morocco and Algeria (Mohammed et al., 1996). Historically, in the USA only two serotypes, EHDV-1 (New Jersey strain) and EHDV-2 (Alberta strain), were identified (Allison et al., 2010). However, in 2006, a novel reassortant serotype 6 EHDV containing two variable surface antigens (VP2 and VP5) from an exotic EHDV-6 with the remaining segments derived from endemic EHDV-2 was isolated from moribund and dead white-tailed deer in Indiana and Illinois (Allison et al., 2010). Subsequently, reassortant type 6 EHDV was identified in several other states demonstrating that the virus is geographically widespread (Allison et al., 2010). This strain is currently endemic in the USA (Allison et al., 2010). Reassortment between native and exotic EHDV illustrates the plasticity of EHDV and their ability to become established in new environmental niches (Allison et al., 2010). EHDV infection can be very debilitating and is often fatal in wild ungulates, including white-tailed deer, mule deer and antelope (Savini et al., 2011). However, EHDV infection is generally not apparent or very mild in livestock (Savini et al., 2011). Despite this, numerous EHDV outbreaks and its introduction to new areas suggest EHDV is clinically and economically important (Savini et al., 2011). Consequently, surveillance is warranted in livestock and wild deer populations. Whole genome sequencing of the EHDV isolated from clinical submissions to the Newport Laboratories over a 4 year period was performed to assess EHDV genetic diversity and population dynamics.
RESULTS AND DISCUSSION Whole genome sequencing and phylogenetic analysis Tissue and blood samples were collected from white-tailed deer exhibiting clinical signs of EHDV infection and submitted to Newport Laboratories from 2008 to 2012 (Table 1). EHDV positive samples were identified using real-time quantitative reverse transcription PCR and cultured on baby hamster kidney (BHK) cells (Clavijo et al., 2010). Forty-four viruses were isolated over this time period and subjected to genome sequencing. Predicted serotypes were determined by homology of VP2 and VP5 genes to reference strains. Type 2 viruses were the most prevalent serotype over http://vir.sgmjournals.org
this time period, accounting for 57 % of the isolates (Table 2). Type 6 viruses represented 32 % of our samples; however, 63 % of EHDV were isolated in 2012. Type 1 was least prevalent, at 11 %, and was not identified in 2012. The results suggest that type 1 was not widespread in the sample set used for the studies. Although type 2 was prevalent, type 6 viruses became dominant in 2012 in the sampling area. Type 2 viruses were identified in all regions sampled and type 6 viruses were also widespread, with isolates from Texas and the Midwest. Type 1 viruses were only identified in Texas and Oklahoma. Co-circulation of type 1, 2 and 6 viruses was seen in Texas in 2010 and 2011, the state most heavily represented in our sample set. Phylogenetic analysis (VP2, VP5 and VP7) Phylogenetic comparisons were constructed in MEGA version 5.05 using the maximum-likelihood algorithm and bootstrapped using 1000 replicates. Phylogenetic trees of all ten segments of EHDV isolates are shown in Figs 1, S1, S2 and S3 (available in JGV Online). Comparison of the VP2 sequences to reference viruses identified 97.3–98.4 % nucleotide identity between type 2 viruses, 96.6298.7 % identity between type 1 viruses and 97.4–97.7 % identity between type 6 viruses (Table S1). Although we observed greater than 96 % identity between VP2 gene sequences within each serotype, there was only 47–53 % identity between type 1, 2 and 6 viruses. Similarly, comparisons of complete VP5 sequences to the published reference viruses identified 96.0–97.5 % nucleotide identity between type 2 viruses, 97.7–98.3 % identity between type 6 viruses and 96.4–96.9 % identity between type 1 viruses (Table S1). Inter-subtype sequence identity was only 64–69 %. Likewise, comparisons of complete VP7 sequences to the published reference viruses identified 97.8–98.6 % nucleotide identity between type 2 isolates, 97 % identity between type 1 isolates and only 78–80 % identity between type 1 and type 2 viruses (Table S1). Moreover, all type 1 and type 2 viruses demonstrates preferential reassortment for VP2, VP5 and VP7 within the homologous serotype (Fig. 2). In the virion, VP2 and VP5 are arranged on the outer core layer formed by VP7. Preferential reassortment observed for type 1 and type 2 viruses suggests incompatibility between the surface proteins VP2 and VP5 and the outer core protein VP7 from the heterologous serotype. Similar to previous studies, the results also demonstrate that VP2 and VP5 from type 6 are compatible with VP7 from serotype 2. In contrast, reassortment by another orbivirus, BTV, was shown to freely occur for all segments (Shaw et al., 2013). Alternate explanations for the observed preferential reassortment could be the limited size of our sample set prohibiting us from identifying all genotypes as well as lack of co-infection with multiple types preventing reassortment in vivo. Co-circulation of type 1, 2 and 6 viruses in Texas in 2010 and 2011 demonstrates that all three types were present in the same state at the same time, 435
S. Anbalagan and others
Table 1. EHDV isolates used in this study Isolate designation (case number) 8-2592-1 8-3707-1 9-2201-1 9-2317-1 9-2742-1 9-2778-1 9-2889-1 9-3132-1 9-3326-1 9-3397-1 9-3466-1 10-0235-1 10-2522-1 10-2604-1 10-2604-2 10-3180-1 10-3456-1 11-0276-2 11-1767-1 11-2140-1 11-2280-1 11-2297-1 11-2507-1 11-2778-2 11-3099-1 11-3174-1 11-3245-1 11-3366-1 12-2980-1 12-3394-1 12-3394-2 12-3422-1 12-3437-8 12-3469-1 12-3583-1 12-3587-1 12-3600-2 12-3960-1 12-3961-1 12-4087-3 12-4087-4 12-4087-5 12-4087-9 12-4350-1
State
Sample type*
Year
Oklahoma Iowa Texas Texas Indiana Missouri Texas Massachusetts Oklahoma Oklahoma Florida Texas Texas Texas Texas Kentucky Missouri Texas Texas Minnesota Louisiana Florida Texas Missouri Missouri Pennsylvania Texas Texas Texas Kansas Kansas Missouri Ohio Ohio Texas Missouri Illinois Texas Missouri Illinois Illinois Illinois Illinois Texas
Blood Tissue homogenate Spleen Spleen Spleen Spleen Spleen Spleen Spleen Spleen Spleen Spleen Blood Spleen Spleen Lung, liver Spleen Spleen Spleen Blood swabs Spleen Spleen Liver, spleen Spleen Spleen Spleen Spleen Spleen Spleen Kidney Kidney Spleen Spleen Lung, spleen, lymph node Spleen Spleen Spleen Spleen Spleen
2008 2008 2009 2009 2009 2009 2009 2009 2009 2009 2009 2010 2010 2010 2010 2010 2010 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2012 2012 2012 2012 2012 2012 2012 2012 2012 2012 2012 2012 2012 2012 2012 2012
NA NA NA NA
Spleen
*NA, Not available.
theoretically allowing for reassortment. For EHDV, reverse genetics is needed to conclusively demonstrate preferential reassortment. High genetic identity (.94 %) was observed between type 1 and type 2 reference viruses for segments encoding VP1, VP3, VP4, VP6, NS1, NS2 and NS3, preventing conclusive identification of reassortment events between these segments for type 1 and type 2 viruses (Table S1). Although 436
the phylogenetic analysis showed subclades which are supported by strong bootstrap values, the branch lengths among isolates are very short, indicating a very low level of genetic variation among the segments (Figs S1, S2 and S3). Overall the results indicate that within type 1 and type 2 viruses, unlike VP2, VP5 and VP7 there are no serotypespecific genetic traits common to segments encoding VP1, VP3, VP4, VP6, NS1, NS2 and NS3 suggesting that genes Journal of General Virology 95
Whole genome analysis of EHDV
Table 2. Number of EHDV serotypes isolated each year Serotype Year
1
2
6
2008 2009 2010 2011 2012
0 1 2 2 0
2 7 3 7 6
0 1 1 2 10
encoding internal proteins are evolving under different selective pressure. Earlier studies of EHDV from multiple hosts species (Culicoides, deer and bovine) and a wide geographical region in the USA (CA, ID, NE, LA, TX, WY) demonstrated high genetic identity (.97 %) among strains (Cheney et al., 1996; Murphy et al., 2005). Our dataset included 44 isolates of EHDV from wide geographical region in the United States (TX, IA, OK, IN, MA, FL, MO, MN, LA, PA, KS, OH, IN, KY) and similar to previous studies, we observed high genetic identity among the isolates (Murphy et al., 2005). These results were somewhat surprising as EHDV, an RNA virus, has a viral RNA polymerase that lacks a proof-reading mechanism (Holland et al., 1982). Additionally, the segmented genome facilitates reassortment within the species. However, our finding supports a growing body of evidence that EHDV, similar to many orbiviruses, is genetically stable (Murphy et al., 2005). In the mosquito-borne Yellow fever virus, a single-stranded RNA virus, there was only a 6.8 % variation among isolates collected over 45 years from a region where transmission is seasonal, multiple species are involved and regular epidemics of diseases in humans occur (Mutebi et al., 2001). Similar observations were also made for another single-stranded RNA virus, vesicular stomatitis virus, which exists in an enzootic cycle, suggesting that there is genetic stability in this virus (Stallknecht et al., 1985). Furthermore, Jenkins et al. (2002) analysed nucleotide substitution rates among 50 RNA viruses and postulated that selection pressure in multiple hosts contributes to slightly lower rates of evolution among vector-borne viruses when compared to non-vector-borne viruses. Such selection pressure might be responsible for less genetic diversity where mutations can disrupt the balance between efficient virus replication and maximal transmission. Moreover, the double-stranded RNA genome of BTV and EHDV likely plays a role in viral genetic stability. Since EHDV is a vector-borne virus, the high level of genetic identity between the isolates suggests a stable host–virus relationship. Serum neutralization assay A subset of EHDV isolates from each serotype was randomly selected and assayed by serum neutralization http://vir.sgmjournals.org
with reference antiserum (Table 3). The results indicate that the serotype determined by genetic analysis (VP2 and VP5 genes) is in agreement with results from serum neutralization assays. As predicted by the genetic analysis, there was little antigenic variation within serotype. Overall, this study found evidence for co-circulation of type 1, type 2 and type 6 in North American white-tailed deer with an increasing prevalence of type 6 in 2012. The results also indicate that VP2, VP5 and VP7 from type 1 and type 2 viruses preferentially reassort within homologous serotypes while for type 6 viruses, this restriction applies only to VP2 and VP5 from type 6 viruses reassorting with VP7 from type 2 viruses. Consistent with other studies, EHDV demonstrated little genetic and antigenic diversity, indicating a stable host–virus relationship.
METHODS Viruses. EHDVs (Table 1) used in this study were isolated from samples submitted to Newport Laboratories for isolation and genetic analysis from 2008 to 2012 as a part of the routine diagnostic testing. The samples were taken from naturally infected animals in the field, by qualified veterinarians, as a part of normal veterinary care and diagnostic testing procedures. Virus growth. Virus was isolated either from tissue or blood samples.
Filtrates from the blood samples (either whole blood or plasma) were used to inoculate BHK cells. Tissue samples were mechanically homogenized in 1 ml Dulbecco’s Modified Eagle’s Medium (DMEM) (Pellgro) containing Cipro (10 mg l21) (Invivogen). The samples were centrifuged at 5000 g for 10 min to remove debris. The supernatant containing the virus was used to inoculate 3-day-old BHK cells propagated in DMEM (Pellgro) with either plasmocin (25 mg l21) (Invivogen) or normocin (100 mg l21) (Invivogen) at 37 uC with 5 % CO2. The cultures were incubated at 37 uC in a CO2 incubator and inspected daily for cytopathic changes (CPE). If CPE was noted, mediim was collected and analysed for EHDV by real-time quantitative reverse transcription PCR using EHDV-specific primers and probe (data not shown). EHDV positive cultures were harvested for serum neutralization testing and RNA extraction. RNA isolation. BHK cells that showed 100 % CPE following EHDV
virus infection were used for RNA extraction. Cell culture supernatant (20 ml) was filtered using 0.2 mm bottle-top filters (Thermo Scientific). The filtrate was centrifuged at 50 000 g for 2 h. Supernatant was discarded and the pellet was suspended in 1000 ml water. Samples were concentrated to a final 100 ml volume using Amicon ultracentrifugal filters (0.5 ml, 50 kDa) (Millipore). Cellular DNA and RNA were removed by incubation with DNase I (25 units) [New England Biolabs (NEB)] and RNase A (25 units) (Qiagen) at 37 uC for 1 h. RNA was extracted using QiaAmp viral RNA isolation kit (Qiagen) and eluted in 35 ml of the supplied elution buffer. Genome amplification. Whole genome sequence of EHDV virus was obtained by sequence-independent cDNA synthesis methodology as described by Potgieter et al. (2009), with some modifications. In brief, 10 ml dsRNA was ligated to PC3-T7 loop (200 ng) (59-p– GGATCCCGGGAATTCGGTAATACGACTCACTATATTTTTATAGTGAGTCGTATTA–OH-39) in T4 RNA ligase buffer (NEB), 1 mM ATP (NEB), 10 % DMSO (Sigma Aldrich), 20 % polyethyleneglycol (PEG6000) (Sigma Aldrich), 40 U RNasin (Promega) and 30 U T4 RNA ligase (NEB) in a final volume of 50 ml. The reaction mix was incubated at 16 uC overnight. Ligated dsRNA was purified using RNA
437
VP5_12-3960-1
VP2 9-3466-1
99
VP7_9-2742-1 VP7_9-2778-1
0.005
VP5_11-2507-1
VP7_9-2889-1
VP2 11-2507-1
VP5_11-2280-1
VP7_9-3132-1
VP2 10-2522-1
VP5_10-2522-1
VP7_9-3326-1
VP2 10-3180-1
VP7_9-3466-1
VP2 9-3326-1
VP5_9-3326-1 VP5_9-3132-1
VP2 11-2778-2
VP5_9-2742-1
VP7_10-2604-2
VP2 11-2280-1
VP5_8-3707-1
VP7_10-3180-1
91
VP7_10-3456-1
VP5_9-3466-1
VP7_11-1767-1
VP2 Serotype 2 Alberta
VP5_10-3456-1
VP7_11-2140-1
VP2 12-4350-1
VP5_11-2297-1
VP7_11-2280-1
63 VP5_11-3174-1 100 VP5_12-3422-1
VP7_11-2778-2
VP5_12-3469-1
97 VP7_11-3099-1
VP2 9-2201-1
VP2 12-2980-1 60 VP2 9-3132-1 78
VP2 11-3245-1
VP2 12-3960-1 97 VP2 11-3366-1 86 89 VP2 9-3397-1 VP2 11-0276-2 59 VP2 10-2604-1 100 VP2 10-0235-1 VP2 Serotype1 NewJersey
61
VP7_12-3394-2
VP5_Serotype_2_Alberta
VP7_12-3422-1
VP5_8-2592-1
100
VP7_12-3600-2
VP5_12-3583-1
VP7_12-3960-1
VP5_10-2604-2
VP7_12-4087-3
VP5_9-2317-1
VP7_12-4087-4
VP5_11-1767-1
VP7_12-4087-9
VP5_12-3394-1
VP7_12-4350-1
VP5_12-3437-8
VP7_Serotype_2_Alberta
VP5_12-3587-1
81
VP7_12-3583-1
VP5_12-3600-2
VP7_12-3437-8
VP5_12-3961-1
61
VP5_12-4087-3 VP5_12-4087-4
VP5_10-2604-1 100
Journal of General Virology 95
79 58
VP5_11-3366-1 VP5_9-3397-1 VP5_10-0235-1
VP7_12-3961-1 VP7_11-3245-1
VP7_Serotype_6_CSIRO753 VP7_Serotype_1_NewJersey
VP5_Serotype_1_NewJersey VP5_11-0276-2
VP7_12-4087-5 VP7_11-2297-1
VP5_12-4087-5 VP5_12-4087-9
64 VP2 12-4087-3
VP2 9-2317-1
VP7_12-3587-1
VP5_Serotype_6_CSIRO 753
VP2 12-4087-4
62
VP7_12-3469-1
VP5_11-3099-1
VP2 12-3437-8
VP2 10-2604-2
VP7_12-3394-1
VP5_9-2889-1
77 VP2 12-3394-1 56 VP2 12-4087-9 56 VP2 12-3600-2
VP2 12-3587-1
VP7_12-2980-1
VP5_9-2201-1
VP2 11-3099-1
VP2 11-1767-1
VP7_11-3174-1
VP5_11-2140-1
VP2 Serotype 6 CSIRO 753 VP2 12-3583-1 100 93 VP2 12-4087-5 57 70 VP2 12-3961-1
VP7_11-2507-1
VP5_10-3180-1
VP2 11-2140-1
VP2 9-2889-1 93 VP2 12-3394-2
100
VP7_10-2522-1
VP5_9-2778-1
VP2 8-2592-1
95
VP7_9-2317-1
VP5_11-3245-1
VP2 9-2778-1
VP2 9-2742-1 78 VP2 8-3707-1
100
VP7_9-2201-1
VP5_12-2980-1 VP5_11-2778-2
0.05
VP7_8-2592-1
VP7
VP5_12-3394-2
VP2 10-3456-1 0.2
VP7_8-3707-1
VP5_12-4350-1
VP5
VP7_11-3366-1 VP7_11-0276-2
100 91
VP7_10-2604-1 VP7_10-0235-1 VP7_9-3397-1
Fig. 1. Maximum-likelihood phylogenetic trees of VP2, VP5 and VP7 nt sequence from EHDV isolates and the reference strains available in GenBank. The numbers indicate bootstrap confidence values after 1000 replications.
S. Anbalagan and others
438 57 VP2 12-3422-1 58 VP2 11-3174-1 VP2 12-3469-1 75 VP2 11-2297-1
VP2
Whole genome analysis of EHDV
VP2 Serotype 1 Serotype 2 Serotype 6
VP5
VP7
Fig. 2. VP2, VP5 and VP7 constellation for EHDV isolates as determined by phylogenetic analysis (see Figs 1, S1, S2 and S3). The isolate number is listed in the first column. The numbers prior to the first hyphen denote the year of isolation. The protein encoded by each gene is listed in the top row.
8-2592-1 8-3707-1 9-2201-1 9-2317-1 9-2742-1 9-2778-1 9-2889-1 9-3132-1 9-3326-1 9-3397-1 9-3466-1 10-0235-1 10-2522-1 10-2604-1 10-2604-2 10-3180-1 10-3456-1 11-0276-2 11-1767-1 11-2140-1 11-2280-1 11-2297-1 11-2507-1 11-2778-2 11-3099-1 11-3174-1 11-3245-1 11-3366-1 12-2980-1 12-3394-1 12-3394-2 12-3422-1 12-3437-8 12-3469-1 12-3583-1 12-3587-1 12-3600-2 12-3960-1 12-3961-1 12-4087-3 12-4087-4 12-4087-5 12-4087-9 12-4350-1
http://vir.sgmjournals.org
Clean & Concentrator (Zymo Research) according to the manufacturer’s instructions. To the purified ligated dsRNA, DMSO (15 % (v/ v) final concentration) was added and incubated in a thermal cycler at 95 uC for 5 min and 4 uC for 5 min. RNA was reverse transcribed to cDNA using the GoScript Reverse Transcription system (Promega). The reaction was incubated in a thermal cycler at 25 uC for 15 min, 42 uC for 60 min and 5 uC for 15 min. RNA was removed using RNaseH (2.5 units, NEB) and incubated in a thermal cycler at 37 uC for 20 min. The cDNA was annealed at 65 uC for 4 h. cDNA was amplified using primer PC2 (59-p–CCGAATTCCCGGGATCC-39) in a PCR mixture containing 16 Ex Taq buffer, 0.2 mM dNTPs (NEB), 5 ml cDNA and 2.5 U TaKaRa Ex Taq (Clontech). The samples were incubated at 72 uC for 1 min. This was followed by an initial denaturation step of 94 uC for 2 min followed by 40 cycles of 94 uC for 30 s, 67 uC for 30 s and 72 uC for 4 min. A final extension step of 72 uC for 5 min was included. Amplified products were visualized after separation on 1.5 % agarose gels (TBE) containing ethidium bromide. Sequencing and data analysis. Amplified products were sequenced using Ion Personal Genome Machine (PGM) sequencing platform (Life Technologies). In brief, the amplified products were fragmented, adaptor ligated and amplified using NEB Next Fast DNA Fragmentation and Library Prep Set 4 (NEB) according to the manufacturer’s instructions. Fragments of sizes 300–330 bp were selected using AMPure XP bead size selection (Beckman Coulter) procedure according to the manufacturer’s instructions. Quality of the amplified adaptor ligated DNA was checked using Qubit dsDNA HS Assay kit (Life Technologies). The samples were pooled in equal molar concentrations and the emulsion-based PCR was conducted using the Ion OneTouch 2 System (Life Technologies) according to the manufacturer’s instructions. Percentage of templated ion sphere particles was checked using Ion Sphere Quality Control kit (Life Technologies). The templated ion spheres were loaded on the 314 chip (Life Technologies) and sequenced on the Ion Torrent Personal Genome Machine (PGM) system using the Ion PGM Template OT2 200 kit v2 (Life Technologies) according to the manufacturer’s instructions. Sequence reads were assembled into contigs using the SeqMan NGen program (DNASTAR). Sequence data of each isolate were analysed using MEGA (version 5.05). Phylogenetic comparisons of VP1, VP2, VP3, VP4, VP5, VP6, VP7, NSP1, NSP2 and NSP3 nt sequences were conducted using maximum-likelihood analysis. Bootstrap confidence values were determined using 1000 replicates. Representative type 1, 2 and 6 viruses were submitted to GenBank under accession numbers KF570113 (type 1, VP1), KF570114 (type 1, VP2), KF570115 (type 1, VP3), KF570116 (type 1, VP4), KF570117 (type 1, VP5), KF570118 (type 1, VP6), KF570119 (type 1, VP7), KF570120 (type 1, NS1), KF570121 (type 1, NS2), KF570122 (type 1, NS3), KF570123 (type 2, VP1), KF570124 (type 2, VP2), KF570125 (type 2, VP3), KF570126 (type 2, VP4), KF570127 (type 2, VP5), KF570128 (type 2, VP6), KF570129 (type 2, VP7), KF570130 (type 2, NS1), KF570131 (type 2, NS2), KF570132 (type 2, NS3), KF570133 (type 6, VP1), KF570134 (type 6, VP2), KF570135 (type 6, VP3), KF570136 (type 6, VP4), KF570137 (type 6, VP5), KF570138 (type 6, VP6), KF570139 (type 6, VP7), KF570140 (type 6, NS1), KF570141 (type 6, NS2), KF570142 (type 6, NS3).
439
S. Anbalagan and others
Table 3. Serum neutralization titres of randomly selected EHDV isolates against serotype reference antisera Serum neutralization titre Isolate
Serotype (VP2)*
Type 1
Type 2
Type 6
2 6 2 1 1 1 2 2 2 2 6
¡4 91 ,4 181 91 512 ¡4 ¡4 ¡4 ¡4 ¡4
16 91 11 ¡4 ¡4 ¡4 23 32 23 91 ,4
¡4 512 ¡4 ¡4 ¡4 ¡4 ¡4 ¡4 ¡4 ¡4 362
8-3707-1 9-2317-1 9-2778-1 9-3397-1 10-0235 10-2604-1 11-2297-1 11-2507-1 12-3394-2 12-3960-1 12-3961-1
Campbell, C. H. & St George, T. D. (1986). A preliminary report of a
comparison of epizootic haemorrhagic disease viruses from Australia with others from North America, Japan and Nigeria. Aust Vet J 63, 233. Cheney, I. W., Yamakawa, M., Roy, P., Mecham, J. O. & Wilson, W. C. (1996). Molecular characterization of the segment 2 gene of epizootic
hemorrhagic disease virus serotype 2: gene sequence and genetic diversity. Virology 224, 555–560. Clavijo, A., Sun, F., Lester, T., Jasperson, D. C. & Wilson, W. C. (2010).
An improved real-time polymerase chain reaction for the simultaneous detection of all serotypes of epizootic hemorrhagic disease virus. J Vet Diagn Invest 22, 588–593. Couvillion, C. E., Jenney, E. W., Pearson, J. E. & Coker, M. E. (1980).
Survey for antibodies to viruses of bovine virus diarrhea, bluetongue, and epizootic hemorrhagic disease in hunter-killed mule deer in New Mexico. J Am Vet Med Assoc 177, 790–791. Holland, J., Spindler, K., Horodyski, F., Grabau, E., Nichol, S. & VandePol, S. (1982). Rapid evolution of RNA genomes. Science 215,
1577–1585. *Serotype predicted based on phylogenetic analysis of the VP2 gene.
Jenkins, G. M., Rambaut, A., Pybus, O. G. & Holmes, E. C. (2002).
Rates of molecular evolution in RNA viruses: a quantitative phylogenetic analysis. J Mol Evol 54, 156–165. Serum neutralization assay. Antisera were diluted in DMEM
serially twofold beginning at 1 : 2 and ending at 1 : 256 in a 96-well micro titre plate. Viruses were grown on BHK cells as described above. Viruses were diluted in DMEM to 200 TCID50 per 100 ml as determined by titration on a monolayer of BHK cells. Diluted virus (100 ml) was added to every well of the diluted serum. The virus and antiserum mixture was incubated at 37 uC for 1 h in a humidified incubator with ±5 % CO2. To the virus and antiserum mixture, 150 ml Vero (African green monkey) cells (105 cells ml21) in DMEM supplemented with 5 % FCS was added. Controls consisting of cells only and cell plus antiserum were also included. Plates were incubated for 4–7 days at 37 uC in a humidified incubator with ±5 % CO2. Cells were visualized using an inverted microscope. Each well was observed for typical CPE and the wells were scored either positive or negative. Wells without CPE were considered positive for the presence of antibody. End-point titres were calculated by the reciprocal of the maximum dilution showing no CPE. The assay was considered valid only when the back titration average fell within 50 and 300 TCID50 per well.
REFERENCES Allison, A. B., Goekjian, V. H., Potgieter, A. C., Wilson, W. C., Johnson, D. J., Mertens, P. P. & Stallknecht, D. E. (2010). Detection of a novel
Jessup, D. A. (1985). Epidemiology of two orbiviruses in California’s
native wild ruminants: preliminary report. Prog Clin Biol Res 178, 53–65. Lee, V. H., Causey, O. R. & Moore, D. L. (1974). Bluetongue and
related viruses in Ibadan, Nigeria: isolation and preliminary identification of viruses. Am J Vet Res 35, 1105–1108. Mecham, J. O. & Dean, V. C. (1988). Protein coding assignment for
the genome of epizootic haemorrhagic disease virus. J Gen Virol 69, 1255–1262. Mohammed, M. E., Aradaib, I. E., Mukhtar, M. M., Ghalib, H. W., Riemann, H. P., Oyejide, A. & Osburn, B. I. (1996). Application of
molecular biological techniques for detection of epizootic hemorrhagic disease virus (EHDV-318) recovered from a sentinel calf in central Sudan. Vet Microbiol 52, 201–208. Moore, D. L. (1974). Bluetongue and related viruses in Ibadan,
Nigeria: serologic comparison of bluetongue, epizootic hemorrhagic disease of deer, and Abadina (Palyam) viral isolates. Am J Vet Res 35, 1109–1113. Moore, D. L. & Kemp, G. E. (1974). Bluetongue and related viruses in Ibadan, Nigeria: serologic studies of domesticated and wild animals. Am J Vet Res 35, 1115–1120. Murphy, M. D., Howerth, E. W., MacLachlan, N. J. & Stallknecht, D. E. (2005). Genetic variation among epizootic hemorrhagic disease viruses in
reassortant epizootic hemorrhagic disease virus (EHDV) in the USA containing RNA segments derived from both exotic (EHDV-6) and endemic (EHDV-2) serotypes. J Gen Virol 91, 430–439.
the southeastern United States: 1978–2001. Infect Genet Evol 5, 157– 165.
Anthony, S. J., Maan, N., Maan, S., Sutton, G., Attoui, H. & Mertens, P. P. (2009a). Genetic and phylogenetic analysis of the core proteins
Phylogenetic and evolutionary relationships among yellow fever virus isolates in Africa. J Virol 75, 6999–7008.
VP1, VP3, VP4, VP6 and VP7 of epizootic haemorrhagic disease virus (EHDV). Virus Res 145, 187–199.
Potgieter, A. C., Page, N. A., Liebenberg, J., Wright, I. M., Landt, O. & van Dijk, A. A. (2009). Improved strategies for sequence-independent
Anthony, S. J., Maan, N., Maan, S., Sutton, G., Attoui, H. & Mertens, P. P. (2009b). Genetic and phylogenetic analysis of the non-structural
amplification and sequencing of viral double-stranded RNA genomes. J Gen Virol 90, 1423–1432.
proteins NS1, NS2 and NS3 of epizootic haemorrhagic disease virus (EHDV). Virus Res 145, 211–219.
Savini, G., Afonso, A., Mellor, P., Aradaib, I., Yadin, H., Sanaa, M., Wilson, W., Monaco, F. & Domingo, M. (2011). Epizootic haemorragic
Anthony, S. J., Maan, S., Maan, N., Kgosana, L., BachanekBankowska, K., Batten, C., Darpel, K. E., Sutton, G., Attoui, H. & Mertens, P. P. (2009c). Genetic and phylogenetic analysis of the
Shaw, A. E., Ratinier, M., Nunes, S. F., Nomikou, K., Caporale, M., Golder, M., Allan, K., Hamers, C., Hudelet, P. & other authors (2013).
outer-coat proteins VP2 and VP5 of epizootic haemorrhagic disease virus (EHDV): comparison of genetic and serological data to characterise the EHDV serogroup. Virus Res 145, 200–210.
Reassortment between two serologically unrelated bluetongue virus strains is flexible and can involve any genome segment. J Virol 87, 543–557.
440
Mutebi, J.-P., Wang, H., Li, L., Bryant, J. E. & Barrett, A. D. T. (2001).
disease. Res Vet Sci 91, 1–17.
Journal of General Virology 95
Whole genome analysis of EHDV
St George, T. D., Cybinski, D. H., Standfast, H. A., Gard, G. P. & DellaPorta, A. J. (1983). The isolation of five different viruses of the epizootic
Stallknecht, D. E., Nettles, V. F., Rollor, E. A., III & Howerth, E. W. (1995).
haemorrhagic disease of deer serogroup. Aust Vet J 60, 216–217.
Epizootic hemorrhagic disease virus and bluetongue virus serotype distribution in white-tailed deer in Georgia. J Wildl Dis 31, 331–338.
Stallknecht, D. E., Nettles, V. F., Fletcher, W. O. & Erickson, G. A. (1985). Enzootic vesicular stomatitis New Jersey type in an insular
Yadin, H., Brenner, J., Bumbrov, V., Oved, Z., Stram, Y., Klement, E., Perl, S., Anthony, S., Maan, S. & other authors (2008). Epizootic haemorrhagic
feral swine population. Am J Epidemiol 122, 876–883.
disease virus type 7 infection in cattle in Israel. Vet Rec 162, 53–56.
http://vir.sgmjournals.org
441
