Journal of Virological Methods 153 (2008) 263–265
Contents lists available at ScienceDirect
Journal of Virological Methods
journal homepage: www.elsevier.com/locate/jviromet
Laboratory tests for evaluating the level of attenuation of bluetongue virus
P. Franchi, M.T. Mercante, G.F. Ronchi, G. Armillotta, S. Ulisse, U. Molini, M. Di Ventura,
R. Lelli, G. Savini, A. Pini∗
Istituto Zooprofilattico Sperimentale dell’Abruzzo e del Molise “G. Caporale”, Via Campo Boario, 64100 Teramo, Italy
Received 17 April 2008
Received in revised form 3 July 2008
Accepted 17 July 2008
Available online 9 September 2008
Bovine fetal aorta endothelial cell line
a b s t r a c t
One of the most important steps when preparing a live attenuated vaccine is the assessment of the level
of attenuation in target animals. It is costly and time consuming as it requires, on each occasion, a large
number of susceptible animals and contained accommodation. This study assessed the consistency of the
bovine foetal aorta endothelial (BFA) cell line and newborn mice for evaluating the attenuation level of
BTV4, BTV9 and BTV16 Italian field isolates. Following serial passages in BHK21c13or Vero cell cultures,
BTV attenuated clones demonstrated a reduced replication capability in the BFA cells compared to the
homologous virulent strains. Similarly, following intracerebral inoculation, the attenuated clones were
completely innocuous to newborn mice contrary to the homologous virulent strains which killed all ani-
and capable of preventing clinical symptoms and viraemia in sheep following challenge with homologous
virulent virus. The two assays may be valuable indicators of the gradual changes occurring in the BTV
population leading to virus attenuation, they can predict the safety of a BTV attenuated vaccine and, in
turn, reduce the number of sheep and cattle required to assess the level of attenuation attained.
© 2008 Elsevier B.V. All rights reserved.
ing ruminants, primarily sheep and to a lesser extent cattle and
goats. Buffaloes, camels, deer and antelope could also be infected.
The virus causes vascular endothelial damage leading to increased
capillary permeability and fragility.
Bluetongue virus (BTV) belongs to the Orbivirus genus, Reoviri-
dae family, 24 serotypes have been identified so far. The virus is
pogonidae family. It has a worldwide distribution and since 1999
has spread into Southern Europe affecting Greece, Italy, Corsica,
Spain, the Balkans, Portugal and more recently Central and North-
ern Europe. The following six virus serotypes are circulating at
present in Europe: BTV1, BTV2, BTV4, BTV8, BTV9 and BTV16.
Although measures such as the ban of livestock movements
and/or vector controls could be useful for limiting the spread of
infection, vaccination is likely to be the only effective tool to con-
in Africa over the last 60 years and is still used in some European
countries (Savini et al., 2007). The vaccine may induce temperature
reaction, clinical signs and viraemia. If some of these side effects
might have no relevance in the African context, they could repre-
∗Corresponding author. Tel.: +39 0861 332481; fax: +39 0861 332251.
E-mail address: email@example.com (A. Pini).
sent an obstacle to its use in European animal breeds that in some
instances are highly sensitive to the virus.
Another problem faced in dealing with BTV attenuated vac-
cines, is the level of attenuation of the virus serotypes used in
their formulation. In this regard, it has to be mentioned that the
use of the Ondersterpoort Biological Products live attenuated vac-
cine prepared with serotype 16 was banned in Italy because of its
pathogenicity for local sheep breeds (Italian Ministry of Health,
sequences of homologous BTV serotypes (Potgieter et al., 2005).
Having available attenuated European BTV isolates would indeed
solve some of the above-mentioned problems.
Under the circumstances, it was decided to proceed with the
attenuation of Italian BTV4, BTV9 and BTV16 field isolates. One of
the most important steps when preparing a live attenuated vaccine
is the assessment of its level of attenuation in target animals. It is
costly and time consuming as it requires, on each occasion, a large
number of susceptible animals and contained accommodation.
In vitro and/or in vivo laboratory models, to be used as comple-
mentary assays to the target animals’ inoculation, could represent
a valid tool in studying BTV attenuation after serial passages in
In this study, bovine fetal aorta endothelial cell line (BFA) (Euro-
pean Collection of Cell Cultures, Salisbury, Wiltshire, UK) and
0166-0934/$ – see front matter © 2008 Elsevier B.V. All rights reserved.
P. Franchi et al. / Journal of Virological Methods 153 (2008) 263–265
newborn mice were used for evaluating the attenuation of BTV4,
BTV9 and BTV16 Italian field isolates.
Primary equine endothelial cells have been used for character-
ising the virulence of the viral arteritis virus (Moore et al., 2002)
whereas cattle pulmonary artery endothelial cell line for replicat-
ing the epizootic hemorrhagic diseases virus, an Orbivirus closely
related to BTV (Quist et al., 1997; Gaydos et al., 2002). The choice of
BFA cell line, as possible indicator of the level of attenuation, was
supported by the fact that virulent BTV, as already stated, causes
vascular endothelial damages in susceptible species.
On the other hand, it was thought to be appropriate to conduct
a parallel study with newborn mice. It has been already shown that
virulent BTV causes encephalopathy when inoculated in newborn
mice (Carr et al., 1994) and that, following subcutaneous inocula-
tion, they may be used as a model to ascertain the virulence of field
isolates (Waldvogel et al., 1986, 1987).
Bluetongue virus serotype 4, BTV9, and BTV16 were serially
passed on cell cultures. BTV9 was passed on Vero cells whereas
BTV4 and BTV16 on BHK21c13. Both cell lines were obtained from
cells were used from passages 124 to 142, whereas the BHK21c13
were from 59 to 79. Confluent monolayers, in 25cm2Falcon plas-
tic flasks, were infected with BTV, 72h later, when the cytopathic
effect (CPE) affected 90–100% of the monolayer, the cell lysates
were harvested. The virus suspensions were then centrifuged at
1942×g for 15min and titrated; 103TCID50were used for infect-
ing confluent monolayers in each subsequent passage. It occurred
approximately a week to both, complete a passage and determine
the titre of the virus suspension harvested. Every 10 passages, the
virus suspensions were checked for purity and serotype identity.
Plaque purified cloning was performed according to Blanchard-
Channel and Stott (1989). No differences in shape and size (from
1.5 to 2.0mm) were observed among the plaques produced by the
BTV isolates included in this study. Although differing with what
observed by other authors, this finding could provide evidence
of the genetic stability of the BTV selected clones (Howell et al.,
1967). Infectivity titres were determined as described by Reed and
els were tested on BFA cells for their replication capability. In this
study, the BFA cells were used from the 14th to 19th passage. Con-
fluent cell monolayers, cultivated in 25cm2Falcon plastic flasks,
were infected with 50TCID50of either the virus culture under test-
ing or the homologous virulent field isolate at the second passage.
Cultures were incubated at 37◦C and, because in this cell line BTV
harvested and centrifuged at 1942×g for 15min. The supernatants
containing the virus were collected and titrated on Vero cells.
In parallel, litters of 10 Swiss mice, 3 days old, were inoculated
cultures used for infecting BFA cells. Animals were observed for 20
days pi. Tests in both, mice and BFA cells, were repeated on three
different occasions and results given are the averages of the three
nificance of the in vivo and in vitro models, experimental batches of
vaccine were produced according to the European Pharmacopoeia,
2005 and tested in sheep for safety and potency according to meth-
ods described (OIE, 2004).
The average titres as assessed on Vero following replication on
BFA cells were 104.2TCID50/ml for the Italian virulent BTV9 at the
second passage and 101.3TCID50/ml for clone 27 at the 35th serial
Mice inoculated with the virulent BTV9 at the second passage
died within the first 10 days pi whereas those inoculated with
either clone 27 or Minimal Essential Medium (Biowest, Nuaillè,
genicity and protection to challenge. The presence and titre of the
virus in the blood were assessed by titration on Vero cells, accord-
ing to the methods described by Savini et al. (2004) and the OIE
manual of diagnostic tests and vaccines for terrestrial animals. The
vaccine resulted innocuous, viraemia could not be demonstrated,
it was immunogenic and capable of preventing clinical symptoms
and viraemia following challenge with the BTV9 virulent strain.
Conversely, clinical symptoms and viraemia were demonstrated in
lished). For BTV4, 10 clones were selected following 62 passages
on BHK21c13, but for 5 of them the test on BFA cells did not show
any difference in infectivity when compared with the titre of the
homologous virulent virus at the second passage. Of the remain-
ing five, clone number 4 gave the most promising results achieving
a titre 102.2TCID50/ml lower than the titre obtained with the vir-
ulent homologous isolate. However, when it was inoculated into
mice, mortality was in the range of 26% within the first 10 days
pi. It was therefore decided to pass clone number 4 on BHK21c13
10 more times. At the 72nd passage and following replication on
BFA cells, clone number 5 showed an average infective titre of
102.3TCID50/ml, whereas the average titre of the virulent virus at
the second passage was 105.1TCID50/ml.
Mice inoculated with clone 5 survived over an observation
period of 20 days, whereas those inoculated with the virulent iso-
late died within the first 10 days pi. As for BTV9, similar results
were achieved when six sheep were immunised with an exper-
imental batch of vaccine produced in BHK21c13monolayer using
clone number 5 at the 72nd passage. No symptoms of BT nei-
ther viraemia could be demonstrated in the immunised animals,
antibody response to serotype 4 was detected by serum neutral-
isation test on day 14 post-vaccination (pv), with peak titres of
1/80–1/320 on day 21. Following challenge with a BTV4 virulent
field isolate on day 70 pv, viraemia was detected in the four unvac-
cinated sheep from days 7 to 11, but not in the six immunised
For BTV16, several clones were selected and tested following
42, 62 and 78 passages on BHK21c13. After the BFA test, only those
homologous virulent isolate. However, mortality in mice remained
high, in the range of 62% for clones isolated after 62 passages and of
34% for a clone number 25 isolated at the 78th passage. At present,
the serotype is undergoing further passages.
In this study, the sensitivity of BFA cells and the pathogenicity
be used as indicators of the level of attenuation attained by BTV in
the course of passages in non-conventional hosts. Waldvogel et al.
not the result of cell culture history but was a characteristic of the
that serial passages in non-conventional hosts facilitate the isola-
tion of clones with modified replicating capacity on BFA cells and
pathogenicity for mice and sheep. Detection of virulence changes
needs a variable number of serial passages in relation to serotype
and perhaps virus isolate.
The two assays may be valuable indicators of the gradual
changes occurring in BTV population leading to virus attenuation,
they can predict the safety of a BTV attenuated vaccine and, in turn,
reduce the number of sheep and cattle required to assess the level
of attenuation attained.
P. Franchi et al. / Journal of Virological Methods 153 (2008) 263–265
Both models could also be useful when testing attenuated vac-
cine strains for reversion to virulence in target animals.
However, although promising, these preliminary results need
to be confirmed through trials involving larger number of target
animals and the other serotypes circulating in Italy and Europe.
Blanchard-Channel, M., Stott, J.L., 1989. Bluetongue virus propagation and plaque
assay: variation due to medium and serum supplement. J. Virol. Methods 24,
Carr, M.A., De Mattos, C.C., Demattos, C.A., Osburn, B.J., 1994. Association of blue-
tongue virus gene segment 5 with neuroinvasiveness. J. Virol. 68, 1255–1257.
Gaydos, J.K., Davidson, W.R., Elvinger, F., Mead, D.G., Howerth, E.W., Stallknecht, D.E.,
2002. Innate resistance of epizootic hemorrhagic disease in white-tailed deer. J.
Wild. Dis. 38, 713–719.
virus. Onderstepoort J. Vet. Res. 34, 317–332.
Italian Ministry of Health, 2005. Nota del 19 gennaio 2005, no. 1720.
Bluetongue—provvedimenti ed impiego.
Moore, B.D., Balasuriya, U.B.R., Hedges, J.F., Maclachlan, N.J., 2002. Growth of a high
virulent, a moderate virulent and an avirulent strain of equine arteritis virus
in primary equine endothelial cells and predictive of their virulence to horses.
Virology 298, 39–44.
Office International des Epizooties (OIE), 2004. OIE Standards Commission (Eds).
Manual of Diagnostic Tests and Vaccines for Terrestrial Animals, fifth ed.
Potgieter, A.C., Monaco, F., Mangan, O., Nomikou, K., Yadin, H., Savini, G., 2005. VP2
segment sequence analysis of some isolates of bluetongue virus recovered in
the Mediterranean basin during the 1998–2003 outbreaks. J. Vet. Med. Ser. B 52,
Quist, C.F., Howerthe, W., Bounous, D.J., Stallknecht, D.E., 1997. Cell-mediated
immune response and IL-2 production in white-tailed deer experimen-
tally infected with epizootic hemorrhagic disease virus. Vet. Immunol.
Immunopathol. 56, 283–297.
Reed, L.J., Muench, H., 1938. A simple method of estimating fifty percent endpoints.
Am. J. Hyg. 27, 493–497.
Monovalent modified vaccine against bluetongue virus serotype 2: immunity
Third International Symposium. Veterinaria Italiana, vol. 40. Taormina, 26–29
October, pp. 664–667.
Savini, G., Maclachlan, N.J., Sanchez-Vizcaino, J.M., Zientara, S., 2008. Vaccines
against bluetongue in Europe. Comp. Immunol. Microbiol. Infect. Dis. 31,
Waldvogel, A.S., Stott, J.L., Squire, K.R.E., Osburn, B.I., 1986. Strain-dependent
virulence characteristics of bluetongue virus serotype 11. J. Gen. Virol. 67,
Waldvogel, A.S., Anderson, C.A., Higgins, R.J., Osburn, B.I., 1987. Neurovirulence of
the UC-2 and UC-8 strains of bluetongue virus serotype 11 in newborn mice.
Vet. Pathol. 24, 404–410.