Iniso

R.

a Anb An

1. I

pesfoudise(Thdominfedisefev(Bainfesev

Veterinary Microbiology 162 (2013) 315–320

A R

Artic

Rece

Rece

Acce

Keyw

Bov

50 u

Npro

Phy

*

1

037

http

creased phylogenetic diversity of bovine viral diarrhoea virus type 1lates in England and Wales since 2001

Strong a,*, J. Errington b,1, R. Cook a,1, N. Ross-Smith a, P. Wakeley a, F. Steinbach a

imal Health and Veterinary Laboratories Agency – Weybridge, Addlestone, Surrey KT15 3NB, United Kingdom

imal Health and Veterinary Laboratories Agency – Penrith, Cumbria CA11 9RR, United Kingdom

ntroduction

Bovine viral diarrhoea virus (BVDV) is a member of thetivirus genus, family Flaviviridae. The genus comprisesr recognised species: BVDV type 1 and type 2, Borderase virus (BDV) and Classical swine fever virus (CSFV)

iel et al., 2005). BVDV causes major economic losses inestic ruminants worldwide (Houe, 1999). BVDV

ction in cattle usually causes no symptoms or a mildase depicted by mild diarrhoea, inappetance, transient

er, reduction in milk yield and reproductive problemsker, 1995). Immunosuppression associated with BVDVction gives rise to a high incidence of morbidity and in

ere cases mortality. Vertical transmission of the virus

may result in either the abortion of the foetus or the birthof persistently infected (PI) calves immunotolerant toBVDV, which are a constant source of infection to othersusceptible animals. Such animals may also succumb tothe fatal mucosal disease, if infected with a cytopathicstrain of BVDV (Baker, 1995).

Pestiviruses have a positive sense RNA genome ofapproximately 12.3 kilobases with a single open readingframe (ORF) flanked by 50 and 30 untranslated regions(UTR). Classification of novel pestivirus isolates is usuallybased on the phylogenetic analysis of partial sequencesfrom the 50UTR, N-terminal autoprotease (Npro) orenvelope glycoprotein (E2) region of the genome (Thielet al., 2005). There are two distinct species of BVDV,namely BVDV-1 and BVDV-2. BVDV-1 is further dividedinto potentially 16 genetic subtypes, 1a through to 1p(Vilcek et al., 2001; Jackova et al., 2008; Nagai et al.,2008; Xue et al., 2010) and BVDV-2 into 3 subtypes, a–c(Tajima et al., 2001).

T I C L E I N F O

le history:

ived 1 May 2012

ived in revised form 30 August 2012

pted 5 September 2012

ords:

ine viral diarrhoea virus

ntranslated region

region

logenetic analysis

A B S T R A C T

Currently, there are two recognised genotypes of Bovine viral diarrhoea virus (BVDV), type

1 and type 2. These genotypes are divided into subtypes based on phylogenetic analysis,

namely a–p for BVDV-1 and a–c for BVDV-2. Within this study, the genetic heterogeneity

of BVDV-1 in England and Wales was investigated and compared to the situation in 1996/

1997. Viral RNA was extracted from 316 blood samples collected between 2004 and 2009

that were previously identified as BVDV-1 positive. A region of the 50 untranslated region

(UTR) was amplified by RT-PCR and the PCR products were sequenced. Phylogenetic

analysis of the 50UTR demonstrated the existence of five subtypes of BVDV-1 circulating in

England and Wales, namely BVDV-1a (244 samples), BVDV-1b (50), BVDV-1e (3), BVDV-1f

(1) and BVDV-1i (18). Phylogenetic analysis of the nucleotide sequence for the Npro region

of the viral genome supported the classification obtained with the 50UTR. Given the fact

that only three subtypes were detected in 1999 this report supports the notion that the

restocking of cattle from continental Europe, after the mass culling during the Foot-and-

Mouth outbreak in 2001 and slaughter of cattle due to bovine tuberculosis infection, has

increased the genetic diversity of BVDV-1 subtypes in England and Wales in the past

10 years.

� 2012 Elsevier B.V. All rights reserved.

Corresponding author. Tel.: +44 1932341111; fax: +44 1932347046.

E-mail address: rebecca.strong@ahvla.gsi.gov.uk (R. Strong).

These authors made equal contribution to the work.

Contents lists available at SciVerse ScienceDirect

Veterinary Microbiology

jo u rn al ho m epag e: ww w.els evier .c o m/lo cat e/vetmic

8-1135/$ – see front matter � 2012 Elsevier B.V. All rights reserved.

://dx.doi.org/10.1016/j.vetmic.2012.09.006

R. Strong et al. / Veterinary Microbiology 162 (2013) 315–320316

A previous study of the genetic diversity of BVDV-1 inEngland and Wales showed that there were threecirculating subtypes, with 1a being the predominantsubtype (Vilcek et al., 1999). Following the mass cullingof animals during the Foot-and-Mouth Disease outbreak of2001, there has been significant movement within the UKand from continental Europe. The impact of these eventson the BVDV status of the UK herd was never investigated.It is important though to ascertain the genetic diversity ofBVDV in a given geographical area to ensure that effectivecontrol strategies can be employed and optimal protectionwill be afforded by vaccination. In this study, we presentdata on the circulating BVDV-1 subtypes in England andWales, demonstrating a significant increase in geneticdiversity since 2001.

2. Materials and methods

2.1. Field samples

Blood samples from cattle exhibiting BVDV-like symp-toms such as diarrhoea, wasting or abortions weresubmitted to the Veterinary Laboratories Agency (VLA)for BVDV testing by antigen ELISA (SERELISA1 BVD/MD Ag,Synbiotics, France). Viral RNA from BVDV-1 positivesamples received by VLA from 2004 to 2006 and from2008 to 2009 was extracted using the MagNA Pure LCextraction robot (Roche-Diagnostics, UK) using the RocheTotal Nucleic Acid isolation kit, according to the manu-facturer’s protocol. The presence of viral RNA was

confirmed using a real time RT-PCR assay based on anassay by McGoldrick et al. (1999) using probes that candiscriminate between BVDV-1, BVDV-2 and BDV.

2.2. RT-PCR and DNA sequencing

A region of the 50UTR was amplified using the Qiagen onestep RT-PCR kit with primers V324 and V326 (Vilcek et al.,1994) using the cycling conditions 50 8C for 20 min (RT step),95 8C for 15 min, followed by 35 cycles at 96 8C for 10 s, 50 8Cfor 5 s, and 60 8C for 4 m. For selected isolates, a region of theNpro gene was amplified using the Qiagen one step RT-PCRkit as described previously with either the BD1 and BD2primer set or the BD1 and BD4 primer set (Vilcek et al., 1997,2001). For some isolates, it was necessary to perform anested two-step RT-PCR to amplify the partial Npro gene. Inbrief, reverse transcription was carried out using theSuperscript1 III reverse transcription kit (Invitrogen, UK)as per the manufacturer’s protocol, followed by PCR withprimers 320F and 1040R (Strong et al., 2010) using the KODHot start DNA polymerase kit (Novagen) and the recom-mended cycling conditions. Subsequently, 1 ml of the firstround PCR was added to a second round nested PCR mixcontaining the primers, BD1 and BD3 (Vilcek et al., 2001).The purified amplicons were sequenced on both strandsusing the V324/V326 primer pair for the 50UTR and eitherthe BD1/BD3 primer pair or the BD1/BD4 primer pair for theNpro region. Sequencing was performed using the Big Dyecycle sequencing technology and the automated ABI3730DNA sequencer.

Fig. 1. Geographic distribution of BVDV-1 samples in England and Wales. (a) The geographic locations of samples analysed based on the farm’s County

Parish Holding number from which the sample was taken. The cattle holding density/km2 for England and Wales is depicted as detailed in the figure. (b)

Each BVDV-1 subtype is indicated by a different symbol as depicted in the legend.

2.3.

thewaBVDana(TaKimrepvalet aparNAmeBVDacc

3. R

3.1.

to t200somTheparholgooand

Tab

Sam

Npro

Ye

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

R. Strong et al. / Veterinary Microbiology 162 (2013) 315–320 317

Genetic typing

The 50UTR sequence corresponding to bases 131–371 of NADL reference strain (accession number: M31182)s aligned with BVDV reference strains representative of

V-1 subtypes, 1a to 1p using ClustalW. Phylogeneticlysis was carried out using the MEGA5 programmura et al., 2011) applying the Neighbour-joiningura two-parameter method (Kimura, 1980) using 1000

licates for calculation of bootstrap values. Bootstrapues of over 70% are shown (Hillis and Bull, 1993; Olveral., 2010). Analysis of the nucleotide sequence for the

tial Npro region corresponding to bases 386–770 of theDL reference strain was analysed using the samethod and parameters. Nucleotide sequences of the

V strains have been submitted to GenBank withession numbers JQ920013–JQ920347.

esults

Field samples

The 316 samples included in this study were submittedhe VLA between the period of 2004–2006 and 2008–9. The samples were from 298 different farms, withe farms submitting multiple samples over this time.

geographic location of the farms were based on countyish holding (CPH) numbers and plotted on a cattledings density map for England and Wales. There wasd correlation between the samples tested by the VLA

thus ensuring that the selection was representative for theUK situation.

3.2. Phylogenetic analysis of the 50UTR region of BVDV-1

isolates

The nucleotide sequence of a region of the 50UTRranging from 239 to 243 bases in length was obtained forall 316 BVDV-1 samples. The nucleotide similarityamongst these isolates was in the range 86.1–100%. Thesequences were aligned with BVDV-1 reference strains andphylogenetic analysis performed. The BVDV-1 subtype foreach sample submitted is summarised in Table 1. Aphylogenetic tree of a representative group of isolates isshown in Fig. 2. Phylogenetic analysis of the 316 sequencesindicated that there are currently five circulating subtypesof BVDV-1 in England and Wales, namely 1a, 1b, 1e, 1i andone isolate (sample 2561) that clusters closely toreferences strains in subtypes -1f and -1g. The nucleotidehomologies between these isolates and the referencestrains BVDV-1a (NADL, Oregon), -1b (Osloss, 24-15), 1e(3186V6, 10-84-Fr), 1i (23-15) and BVDV-2 (NY-93) were86.1–98.3%, 87.4–100%, 86.4–99.2%, 86.2–99.2% and 73.2–78.5%, respectively. The sequence identity within the samesubtypes was 90–100%. The 50UTR sequence for sample2561 was submitted for a BLAST search using the blastnalgorithm of NCBI. It shared a 98% identity to the ‘‘1f-like’’strain isolated from a mousedeer (Tragulus javanicus) inCopenhagen Zoo (Grondahl et al., 2003) and 97% identity tostrain 22146/81 isolated in Lower Saxony, Germany which

le 1

ple identifier and proposed genotype based on the partial 50 UTR sequence for the 316 BVDV-1 samples utilised in this study. The isolates selected for

sequencing are indicated in bold italics.

ar Sample identifier Genotype

04 286, 287, 289, 292 1a

04 284 1b

04 282 1e

04 283 1i

05 357, 360, 361, 364, 366, 367, 372, 373, 378, 380, 381, 384, 385, 392, 393, 395, 396, 397, 398, 400, 405, 406, 415,

416, 420, 422, 423, 425, 426, 428, 429, 431, 432, 433, 435, 441, 451, 453, 456, 459, 460, 461, 463, 466, 467, 468,

469, 472, 473, 475, 478, 479, 481, 560, 645, 686, 774, 828

1a

05 358, 362, 365, 368, 370, 402, 407, 427, 430, 462, 482, 565, 816 1b

05 596 1e

05 356, 363 1i

06 853, 857 1a

08 2175, 2176, 2178, 2180, 2183, 2184, 2185, 2190, 2191, 2192, 2193, 2194, 2195, 2209, 2210, 2211, 2212, 2213,

2215, 2216

1a

08 2181, 2186 1i

09 2218, 2221, 2222, 2223, 2224, 2225, 2227, 2228, 2230, 2233, 2234, 2235, 2237, 2239, 2240, 2245, 2251, 2252,

2254, 2255, 2256, 2258, 2259, 2262, 2263, 2266, 2267, 2270, 2271, 2272, 2276, 2277, 2278, 2280, 2283, 2306,

2312, 2313, 2314, 2316, 2318, 2321, 2322, 2323, 2325, 2341, 2342, 2343, 2344, 2345, 2347, 2348, 2365, 2366,

2368, 2370, 2371, 2375, 2376, 2380, 2381, 2387, 2395, 2396, 2397, 2400, 2401, 2404, 2405, 2406, 2408, 2410,

2412, 2413, 2414, 2418, 2420, 2422, 2425, 2427, 2434, 2438, 2439, 2443, 2445, 2449, 2456, 2460, 2461, 2462,

2465, 2466, 2467, 2468, 2470, 2471, 2473, 2475, 2476, 2477, 2479, 2483, 2484, 2502, 2505, 2507, 2508, 2509,

2510, 2512, 2513, 2516, 2517, 2520, 2521, 2522, 2523, 2524, 2525, 2527, 2542, 2547, 2550, 2551, 2552, 2553,

2563, 2564, 2566, 2568, 2569, 2571, 2573, 2574, 2583, 2587, 2589, 2590, 2591, 2592, 2603, 2618, 2623, 2625,

2626, 2634, 2635, 2639, 2649, 2661, 2663, 2674, 2678, 2690, 2691, 2692, 2696, 2715, 2720, 2722

1a

09 2229, 2231, 2236, 2253, 2303, 2315, 2319, 2358, 2362, 2372, 2374, 2379, 2386, 2391, 2403, 2423, 2424, 2428,

2429, 2430, 2448, 2454, 2455, 2458, 2472, 2474, 2480, 2503, 2518, 2560, 2588, 2605, 2606, 2636, 2641, 2727

1b

09 2687 1e

09 2561 1f

09 2219, 2232, 2346, 2359, 2383, 2399, 2402, 2415, 2426, 2478, 2485, 2541, 2544 1i

s considered a BVDV-1f strain in a previous analysis

the density of cattle in England and Wales (Fig. 1a) wa

R. Strong et al. / Veterinary Microbiology 162 (2013) 315–320318

(Tajima et al., 2001). The predominant subtype was BVDV-1a (244/316 samples). Sequence data from some of thesamples shared 100% identity across the 50UTR regionanalysed. Interestingly, samples 435 (BVDV-1a) and 2518(BVDV-1b) showed 100% identity to UK strains 28-1 and24-15, respectively, which were isolated during the laststudy in England and Wales (Vilcek et al., 1999).

3.3. Nucleotide sequencing and phylogenetic analysis of the

Npro region of selected BVDV-1 isolates

The Npro sequence for 19 samples representing thevarious BVDV-1 subtypes was obtained to verify the BVDV-1 classification determined from the 50UTR sequences. Theamino acid similarity amongst these isolates was in therange 78.2–100%. Phylogenetic analysis was performed bycomparing the 385 nucleotide region of Npro correspondingto bases 386–770 of the 19 samples with data available forBVDV-1 reference strains representing subtypes 1a–1p(Fig. 3). The classification of BVDV-1 strains based on the50UTR sequences was supported by the phylogeneticanalysis of the Npro sequences. The comparison of theseamino acid sequence of the BVDV-1 strains to the referencestrains BVDV-1a (NADL, Oregon), -1b (Osloss, 24-15), 1e(3186V6, 26-V639), 1i (23-15) and BVDV-2 (NY-93)revealed 87.5–97.7%, 85.2–100%, 86.7–92.2%, 86.7–100%and 71.9–75% identity, respectively. The sequence identitywithin the same subtypes was 90.6–100%. Sample 2561again clustered close to but more distantly related to thesubtype 1f reference strains. For subtypes 1a, 1b and 1i, thesamples sequenced were those that had shown 100%identity over the 50UTR region sequenced. For subtype 1aisolates 2566, 2587 and 2690, there was 97–98% identitywithin the Npro sequences and for samples 2387, 2414 and2420, a 93–95% identity. These sequence differencesresulted in amino acid changes. For subtype 1b isolates,97% identities was observed amongst the samples,although two of the samples were 100% identical andwere received from the same farm within a 1 monthperiod. For samples 2186, 2383 and 2486 (1i isolates) 100%identity was observed for the Npro sequences whilst 98–99% identity was shown for samples 2402, 2415 and 2541,resulting in amino acid differences. This data confirms thatthe Npro region is more divergent than the 50UTR sequenceanalysed.

The geographic distribution of the BVDV-1 subtypeswas plotted (Fig. 1b). Four subtypes were present in Wales

Fig. 2. Phylogenetic typing of selected BVDV-1 isolates from England and

Wales in the 50UTR. Fifty-five BVDV-1 samples representing the five

different subtypes circulating in England and Wales are shown. Sample

2561 is italicised with underlining and an asterisk. Reference strains are

shown in bold and the accession numbers for these strains on NCBI

(AF091605), 28-1 (AF298061); BVDV-1b – Osloss (M96687), 24-15

(AF298060); BVDV-1c – Bega (AF049221), Manasi (EU159702); BVDV-1d

– F-Au (AF298065), 16-111-Fr (AF298056); BVDV-1e – 3186V6

(AF298062), 10-84-Fr (AF298054); BVDV-1f – J-Au (AF298067), W-Au

(AF298073), 22146/81 (AJ304386); BVDV-1g – A-Au (AF298064), L-Au

(AF298069); BVDV-1h – G-Au (AF298066), KM-SK (AF298068); BVDV-1i

– 23-15 (AF298059); BVDV-1j – KS86-1ncp (AB078950), Deer

(AB040132); BVDV-1k – Suwa (AF117699), Rebe (AF299317); BVDV-

1m – ZM-95 (AF526381), TY05 (GU120242); BVDV-1n – Shitara-02-06

(AB359930), So CP-75 (AB359929); BVDV-1o – IS25CP-01 (AB359931),

AQGN96B15 (AB300691); BVDV-1p – BJ0702 (GU120248), BJ0703

(GU120249); BVDV-2 – NY-93 (AF502399). The tree was computed by

the neighbour-joining kimura two-parameter method using the MEGA5

program. Bootstrap values of over 70% are shown for 1000 replicate data

Genbank are as follows: BVDV-1a – NADL (M31182), Oregon C24V sets.

Fig.

Nine

circu

in b

as fo

1b –

519

(AF2

(AF2

G-A

(AB0

BVD

(AF5

CP-7

BJ07

The

met

show

R. Strong et al. / Veterinary Microbiology 162 (2013) 315–320 319

compared to five in England. Subtypes 1a and 1b werespread throughout England and Wales. The 1i isolates wereabsent from the northern most part of England and Walesand subtype 1e was only present in three distinct counties(Lancashire, Devon and Monmouthshire). The isolate thatclustered closely with the BVDV-1f reference strain wasisolated from a farm in North Yorkshire. From the samplesanalysed Devon and Monmouthshire have the greatestdiversity of BVDV-1 subtypes.

4. Discussion

In this study, the genetic diversity of BVDV-1 isolatescollected from England and Wales over a 5 year period wasanalysed. The previous study carried out on 62 samplesfrom cattle herds in 1996–1997 indicated the presence ofonly three different subtypes of BVDV-1 including 59isolates of subtype 1a, 1 subtype 1b isolate and 2 isolateslater classified as subtype 1i (Vilcek et al., 1999, 2001). In2001, Graham et al. hypothesised that ‘‘lifting of the cattleimportation restrictions following the introduction of theSingle European Market in 1993 would increase theincidence of BVDV-1b in England’’. This is indeed reflectedin the current study where 51 samples (16%) of the 316samples were classified as BVDV-1b compared to 1 out of62 samples in the 1999 study. The incidence of BVDV-1iisolates has also increased in the past 10 years from 3% ofsamples in the 1999 study to 6% of samples analysed in thisstudy. Two additional subtypes, namely 1e and 1f, are nowcirculating in England and Wales. The increase in geneticdiversity between the two studies may lie with the smallstudy size, only 62 samples in the 1999 study, compared to316 samples analysed in this study. However, it is likelythat this increase is due to increased cattle movement andimportation over the last decade. During the 2001 Foot-and-Mouth outbreak, 581,802 cattle were culled (FMDData Archive, DEFRA, 2004). Restocking of the farmsinvolved cattle bred in the UK as well as importation ofcattle from across Europe (RADAR CTS dataset, DEFRA,2012a). In the last 10 years, there has been significantimportation of animals from Ireland, Denmark andGermany in which BVDV-1b is the predominant subtype(Tajima et al., 2001; Graham et al., 2001; Uttenthal et al.,2005). BVDV-1i seems to be restricted to the UK where ithas spread from one county (Dorset) in 1999 to a further 10counties in the past 10 years. Importation of cattle fromcountries such as Italy and Switzerland, where BVDV-1e isthe predominant subtype, has been recorded (Giammarioliet al., 2008; Stalder et al., 2005) and also from France andDenmark (RADAR CTS dataset, DEFRA, 2012a), whereBVDV-1e is reported to be circulating at a low level(Jackova et al., 2008; Uttenthal et al., 2005). There was noevidence for any direct importation of cattle from Europeinto these farms where subtype 1e and 1f were identified.The high diversity of BVDV-1 subtypes in Devon supportsthe hypothesis that restocking, whether from within theUK or from Europe, has contributed to the greater diversityof subtypes. In this county alone, over 67,000 cattle wereculled during the FMD outbreak in 2001 and further cattleslaughtered as a result of bovine TB, which started in the

1m

1p

1o

1g

1f

1h

1i

1j

1a

1k

1e

1n

1c

1l

1b

1d

BVDV-2

*

ZM-95

TY05

BJ070 2

BJ070 3

IS25 CP-01

A-Au

L-Au

256 1 J-Au

W-Au

G-Au

23 -15

240 2

241 5

254 1

218 6

234 6

238 3

Dee r-GB1

KS86 -1ncp

256 6

258 7

269 0

Oregon C24 V

NAD L

238 7

241 4

242 0

CH -Suwa

26 -V63 9

3186 V6

268 7

SoCP-75

Shitara-02 -06

Bega

51 9

TR-27

TR-29

Osloss

24 -15

251 8

239 1

230 3

242 3

247 2

F-Au

NY-93

100

100

100

100

99

100

100

100

98

82

98

100

100

100

100

87

99

98

0.05

3. Genetic typing of selected BVDV-1 samples in the Npro region.

teen BVDV-1 samples representing the five different subtypes

lating in England and Wales are shown. Reference strains are shown

old and the accession numbers for these strains on NCBI Genbank are

llows: BVDV-1a – NADL (M31182), Oregon C24V (AF091605); BVDV-

Osloss (M96687), 24-15 (AF287280); BVDV-1c – Bega (AF049221),

(AF144464); BVDV-1d – F-Au (AF287284); BVDV-1e – 3186V6

87282), 26-V639 (AF287282); BVDV-1f – J-Au (AF287286), W-Au

87290); BVDV-1g – A-Au (AF287283), L-Au (AF287287); BVDV-1h –

u (AF287285); BVDV-1i – 23-15 (AF287279); BVDV-1j – KS86-1ncp

78950), Deer-GB1 (U80902); BVDV-1k – CH-Suwa (AY894998);

V-1l – TR-27 (EU163975), TR-29 (EU163977); BVDV-1m – ZM-95

26381), TY05 (GU120258); BVDV-1n – Shitara-02-06 (AB359930), So

5 (AB359929); BVDV-1o – IS25CP-01 (AB359931); BVDV-1p –

02 (GU120260), BJ0703 (GU120261); BVDV-2 – NY-93 (AF502399).

tree was computed by the neighbour-joining kimura two-parameter

hod using the MEGA5 program. Bootstrap values of over 70% are

n for 1000 replicate data sets.

R. Strong et al. / Veterinary Microbiology 162 (2013) 315–320320

South-West of England. In 2010 alone, 34,300 cattle wereslaughtered across England and Wales (DEFRA, 2012b).

The 50UTR region is highly conserved amongst pesti-viruses and most extensively studied. Thus there are alarge number of available sequences for this region. Thehighly conserved nature of the 50UTR was reflected withinthis study as several samples exhibited 100% sequenceidentity over the region amplified. Samples submittedfrom the same farm showed 99–100% sequence identitythus demonstrating good genetic conservation at the herdlevel. The majority of identical sequences resulted fromfarms in different counties.

To obtain a better resolution of phylogeny, the Npro

sequences for 19 samples which included those that hadshown 100% identity within the 50UTR, one subtype 1eisolate and sample 2561 that by 50UTR sequence clusteredbetween BVDV-1f and -1g subtypes were analysed.Generally, farms from which samples were identified ashaving 100% sequence identity over the 50UTR region,showed nucleotide differences within the Npro region thatinferred amino acid changes. Two samples (435 and 2518)showed 100% sequence identity to two strains, namely 28-1 and 24-15 isolated in the previous study by Vilcek et al.(1999). There was no correlation between the farms fromwhich the samples or the reference strains were taken.Unexpectedly, Npro sequencing of sample 2518 revealedthat this isolate was 100% identical to the reference strain24-15 over both the 50UTR and the Npro region amplified.Sample 2561 clustered next to BVDV-1f but more distantthan the current reference strains are to each other.

In this study, the genetic diversity of BVDV-1 withinEngland and Wales has been ascertained. The restocking offarms after the FMD outbreaks of 2001 and 2007 andfollowing the slaughter of animals potentially infected bybovine TB over the last 10 years appears to have had animpact on the genetic diversity of BVDV-1 subtypes.

Acknowledgements

This work was supported by DEFRA under researchcontract OD0345 and surveillance contract ED1000. Theauthors thank Adnan Younas for producing the geogra-phical maps and Rose Nicholson for providing data fromthe RADAR cattle tracing system.

References

Baker, J.C., 1995. The clinical manifestations of bovine viral diarrheainfection. Vet. Clin. North Am. Food Anim. Pract. 11, 425–445.

DEFRA, 2004. Animal Health and Welfare: FMD Data Archive: Statistics –Cattle Slaughtered per Month, Listed by County. DEFRA.

DEFRA, 2012a. RADAR Cattle Tracing System (CTS). DEFRA.DEFRA, 2012b. Bovine TB in Great Britain: Regional Statistics in 2010.

DEFRA.Giammarioli, M., Pellegrini, C., Casciari, C., Rossi, E., De Mia, G.M., 2008.

Genetic diversity of bovine viral diarrhea virus 1: Italian isolates

clustered in at least seven subgenotypes. J. Vet. Diagn. Invest. 20,783–788.

Graham, D.A., McLaren, I.E., Brittain, D., O’Reilly, P.J., 2001. Genetic typingof ruminant pestivirus strains from Northern Ireland and the Republicof Ireland. Res. Vet. Sci. 71, 127–134.

Grondahl, C., Uttenthal, A., Houe, H., Rasmussen, T.B., Hoyer, M.J., Larsen,L.E., 2003. Characterisation of a pestivirus isolated from persistentlyinfected mousedeer (Tragulus javanicus). Arch. Virol. 148, 1455–1463.

Hillis, D.M., Bull, J.J., 1993. An empirical test of bootstrapping as a methodfor assessing confidence in phylogenetic analysis. Syst. Biol. 42,182–192.

Houe, H., 1999. Epidemiological features and economical importance ofbovine virus diarrhoea virus (BVDV) infections. Vet. Microbiol. 64,89–107.

Jackova, A., Novackova, M., Pelletier, C., Audeval, C., Gueneau, E., Haffar, A.,Petit, E., Rehby, L., Vilcek, S., 2008. The extended genetic diversity ofBVDV-1: typing of BVDV isolates from France. Vet. Res. Commun. 32,7–11.

Kimura, M., 1980. A simple method for estimating evolutionary rates ofbase substitutions through comparative studies of nucleotidesequences. J. Mol. Evol. 16, 111–120.

McGoldrick, A., Bensaude, E., Ibata, G., Sharp, G., Paton, D.J., 1999. Closedone-tube reverse transcription nested polymerase chain reaction forthe detection of pestiviral RNA with fluorescent probes. J. Virol.Methods 79, 85–95.

Nagai, M., Hayashi, M., Itou, M., Fukutomi, T., Akashi, H., Kida, H., Sakoda,Y., 2008. Identification of new genetic subtypes of bovine viraldiarrhea virus genotype 1 isolated in Japan. Virus Genes 36, 135–139.

Olvera, A., Busquets, N., Cortey, M., de Deus, N., Ganges, L., Nunez, J.I.,Peralta, B., Toskano, J., Dolz, R., 2010. Applying phylogenetic analysisto viral livestock diseases: moving beyond molecular typing. Vet. J.184, 130–137.

Stalder, H.P., Meier, P., Pfaffen, G., Wageck-Canal, C., Rufenacht, J., Schaller,P., Bachofen, C., Marti, S., Vogt, H.R., Peterhans, E., 2005. Geneticheterogeneity of pestiviruses of ruminants in Switzerland. Prev.Vet. Med. 72, 37–41 discussion 215–219.

Strong, R., La Rocca, S.A., Ibata, G., Sandvik, T., 2010. Antigenic and geneticcharacterisation of border disease viruses isolated from UK cattle. Vet.Microbiol. 141, 208–215.

Tajima, M., Frey, H.R., Yamato, O., Maede, Y., Moennig, V., Scholz, H.,Greiser-Wilke, I., 2001. Prevalence of genotypes 1 and 2 of bovineviral diarrhea virus in Lower Saxony, Germany. Virus Res. 76, 31–42.

Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., Kumar, S., 2011.MEGA5: molecular evolutionary genetics analysis using maximumlikelihood, evolutionary distance, and maximum parsimony methods.Mol. Biol. Evol. 28, 2731–2739.

Thiel, H.-J., Collett, M.S., Gould, E.A., Heinz, F.X., Meyers, G., Purcell, R.H.,Rice, C.M., Houghton, M., 2005. Family Flaviviridae. In: Fauquet,C.M., Mayo, M.A., Maniloff, J., Desselberger, U., Ball, L.A. (Eds.),Virus Taxonomy – Eighth Report of the International Committeeon Taxonomy of Viruses. Academic Press, San Diego, pp. 981–998.

Uttenthal, A., Stadejek, T., Nylin, B., 2005. Genetic diversity of bovine viraldiarrhoea viruses (BVDV) in Denmark during a 10-year eradicationperiod. APMIS 113, 536–541.

Vilcek, S., Drew, T.W., McGoldrick, A., Paton, D.J., 1999. Genetic typing ofbovine pestiviruses from England and Wales. Vet. Microbiol. 69,227–237.

Vilcek, S., Herring, A.J., Herring, J.A., Nettleton, P.F., Lowings, J.P., Paton,D.J., 1994. Pestiviruses isolated from pigs, cattle and sheep can beallocated into at least three genogroups using polymerase chainreaction and restriction endonuclease analysis. Arch. Virol. 136,309–323.

Vilcek, S., Nettleton, P.F., Paton, D.J., Belak, S., 1997. Molecular character-ization of ovine pestiviruses. J. Gen. Virol. 78 (Pt 4), 725–735.

Vilcek, S., Paton, D.J., Durkovic, B., Strojny, L., Ibata, G., Moussa, A., Loitsch,A., Rossmanith, W., Vega, S., Scicluna, M.T., Paifi, V., 2001. Bovine viraldiarrhoea virus genotype 1 can be separated into at least elevengenetic groups. Arch. Virol. 146, 99–115.

Xue, F., Zhu, Y.M., Li, J., Zhu, L.C., Ren, X.G., Feng, J.K., Shi, H.F., Gao, Y.R.,2010. Genotyping of bovine viral diarrhea viruses from cattle in Chinabetween 2005 and 2008. Vet. Microbiol. 143, 379–383.