Journal of General Virology (2005), 86, 907917

DOI 10.1099/vir.0.80718-0

Development of bovine herpesvirus 4 as an

expression vector using bacterial artificial
chromosome cloning
L. Gillet,13 V. Daix,13 G. Donofrio,2 M. Wagner,3 U. H. Koszinowski,4
B. China,5 M. Ackermann,6 N. Markine-Goriaynoff1
and A. Vanderplasschen1
1,5

Department of Infectious and Parasitic Diseases (B43b)1 and Food Sciences Department
(B43b)5, Faculty of Veterinary Medicine, University of Lie`ge, B-4000 Lie`ge, Belgium

Correspondence
A. Vanderplasschen
A.vdplasschen@ulg.ac.be

Dipartimento di Salute Animale, Facolta` di Medicina Veterinaria, Sezione di Malattie Infettive

degli Animali, Universita` degli Studi di Parma, I-43100 Parma, Italy

Department of Pathology, Harvard Medical School, Boston, MA 02115, USA

Department of Virology, Max von Pettenkofer-Institut, Ludwig-Maximilians-Universitat

Munchen, D-81377 Munich, Germany

Institute for Virology, University of Zurich, CH-8057 Zurich, Switzerland

Received 25 October 2004

Accepted 20 December 2004

Several features make bovine herpesvirus 4 (BoHV-4) attractive as a backbone for use as a
viral expression vector and/or as a model to study gammaherpesvirus biology. However, these
developments have been impeded by the difficulty in manipulating its large genome using classical
homologous recombination in eukaryotic cells. In the present study, the feasibility of exploiting
bacterial artificial chromosome (BAC) cloning and prokaryotic recombination technology for
production of BoHV-4 recombinants was explored. Firstly, the BoHV-4 genome was BAC
cloned using two potential insertion sites. Both sites of insertion gave rise to BoHV-4 BAC
clones stably maintained in bacteria and able to regenerate virions when transfected into permissive
cells. Reconstituted virus replicated comparably to wild-type parental virus and the loxP-flanked
BAC cassette was excised by growing them on permissive cells stably expressing Cre
recombinase. Secondly, BoHV-4 recombinants expressing Ixodes ricinus anti-complement
protein I or II (IRAC I/II) were produced using a two-step mutagenesis procedure in Escherichia
coli. Both recombinants induced expression of high levels of functional IRAC molecules in the
supernatant of infected cells. This study demonstrates that BAC cloning and prokaryotic
recombination technology are powerful tools for the development of BoHV-4 as an expression
vector and for further fundamental studies of this gammaherpesvirus.

INTRODUCTION
Bovine herpesvirus 4 (BoHV-4) is a gammaherpesvirus that
has been isolated throughout the world from healthy cattle
as well as those exhibiting a variety of diseases (Thiry et al.,
1992). Its genome has a B-type structure consisting of a long
unique region (L-DNA) flanked by a total of 15 (on average,
as determined for the 66-p-347 strain) polyrepetitive DNA
(prDNA) elements distributed randomly at both ends of
the genome (Fig. 1a). The recent sequencing of BoHV-4
confirmed that it is a member of the genus Rhadinovirus
encoding a relatively reduced set of open reading frames
(ORFs) homologous to cellular genes (Zimmermann et al.,
2001).
3These authors contributed equally to this work.

0008-0718 G 2005 SGM

Printed in Great Britain

Several features make BoHV-4 attractive as a backbone for a

viral vector and/or as a model to study gammaherpesvirus
biology: (i) its genome is less complex than those of several
other herpesviruses (Zimmermann et al., 2001); (ii) it allows
the stable insertion of additional genetic material up to at
least 10?5 kb (see below); (iii) in contrast to several other
rhadinoviruses, it is easy to propagate in cell culture (Thiry
et al., 1992); (iv) wild-type virus exhibits limited or no
pathogenicity in natural and experimental hosts (Thiry
et al., 1992); (v) in contrast to most gammaherpesviruses,
BoHV-4 is able to replicate in a broad range of host species
both in vitro and in vivo (Thiry et al., 1992); (vi) in some
cells, BoHV-4 has been shown to establish non-replicative
persistent infection, allowing expression of transgenes for
weeks without affecting the viability of the expressing cells
907

L. Gillet and others

Fig. 1. Schematic representation of the strategies followed to produce two infectious BoHV-4 BAC plasmids. (a, b) The
EcoRI restriction map of the entire BoHV-4 V. test strain is shown at the top. A loxP-flanked BAC cassette was inserted into
the XhoI (a) or BstEII (b) sites located, respectively, at the left and the right ends of BoHV-4 V. test strain L-DNA. These
regions do not contain any ORFs and are located, respectively, in the EcoRI G and EcoRI I restriction fragments of the
BoHV-4 V. test strain genome. (c) Flowchart of stages performed to produce BoHV-4 BAC plasmids, to control their
infectivity and to demonstrate the possibility of removing the loxP-flanked BAC cassette from the genome of reconstituted
virus.
908

Journal of General Virology 86

BAC cloning of BoHV-4

(Gillet et al., 2004); and (vii) a small animal model (rabbit) is

available (Thiry et al., 1992).
Applied and fundamental research on herpesviruses requires
genomic modifications such as insertion of transgenes and/
or deletion of viral ORFs. Both types of modification have
been applied to BoHV-4 by homologous recombination in
eukaryotic cells (Gillet et al., 2004; Markine-Goriaynoff
et al., 2004). However, this approach is extremely slow
and laborious. Further studies on BoHV-4 require a faster
approach to modify its genome. Manipulation of large
herpesvirus genomes has recently been facilitated by using
bacterial artificial chromosome (BAC) vectors (Messerle
et al., 1997; Wagner et al., 2002). These vectors allow the
maintenance and mutagenesis of the viral genome in
Escherichia coli followed by the reconstitution of progeny
virions by transfection of the BAC plasmid into permissive
eukaryotic cells.
In the present study, we addressed the feasibility of exploiting BAC cloning and prokaryotic recombination technology
for further applied and fundamental research on BoHV-4.
Firstly, in order to BAC clone the BoHV-4 genome, two
non-coding regions located at both extremities of the LDNA were selected for insertion of a loxP-flanked BAC
cassette. Both sites of insertion led to the production of
BoHV-4 BAC clones stably maintained in bacteria and able
to regenerate virions when transfected into permissive
cells. Reconstituted viruses replicated comparably to wildtype parental virus and the loxP-flanked BAC cassette was
excised by growing them on permissive cells stably expressing Cre recombinase. Secondly, prokaryotic recombination
technology and one of the BoHV-4 BAC clones produced
were used to generate BoHV-4 recombinants expressing
functional Ixodes ricinus anti-complement protein I or II
(IRAC I/II) (Daix et al., 2001). IRACs are inhibitors of the
alternative pathway of the complement system and are
secreted in the saliva of I. ricinus ticks.
Taken together, the data of the present study demonstrated
that BAC cloning and prokaryotic recombination technology are powerful tools for the development of BoHV-4 as
an expression vector and for further fundamental studies of
this gammaherpesvirus.

METHODS
Cell lines and virus strains. MadinDarby bovine kidney

(MDBK; ATCC CCL-22), embryonic bovine lung (EBL; DSMZ ACC

192) and 293T cells (ATCC CRL-11268) were cultured in minimum
essential medium (MEM; Invitrogen) containing 10 % fetal calf
serum (FCS; BioWhittaker). The BoHV-4 V. test (Thiry et al., 1981)
strain was used throughout this study.
Production of a stable cell line expressing Cre recombinase.

EBL cells stably expressing Cre recombinase fused to a nuclear localization signal (EBL NLSCre) were produced as follows. Briefly,
NLSCre was excised by NheI and SpeI digestion from the pGBFNLS-Cre vector (provided by Dr D. Pirottin, University of Lie`ge,
Belgium) and cloned into the XbaI site of the pEFIN3 bicistronic
expression vector (provided by Dr M. Parmentier, Free University of
http://vir.sgmjournals.org

Brussels, Belgium) resulting in pEFIN3-NLS-Cre. After linearization

by ScaI digestion, the plasmid was transfected into EBL cells. After
2 weeks of selection under G418 (final concentration 200 mg ml21;
Invitrogen), all resistant cells expressed NLSCre (data not shown)
and were used without further cell cloning.
BAC cloning of BoHV-4. Two strategies of insertion were used to
BAC clone BoHV-4 genome. A BAC cassette was inserted either into
the far left XhoI site or into the far right BstEII site of BoHV-4
L-DNA. These regions of insertion do not contain any ORFs
(Zimmermann et al., 2001). The BAC cassette was obtained from a
modified pBeloBAC vector in which additional restriction sites were
inserted upstream and downstream of the loxP sequences (GenBank
accession no. AY665170) (Fig. 1a and b). Insertion of the BAC
cassette into the BoHV-4 genome was performed by homologous
recombination as described previously (Markine-Goriaynoff et al.,
2004). The recombination cassettes were produced as follows. For
the left insertion (Fig. 1a), the EcoRI restriction fragment G of the
BoHV-4 V. test strain genome was first cloned into the pGEM-T
Easy vector (Promega) resulting in pGEM-T-EcoRI G. The BAC
cassette was then released by PmeI digestion from the modified
pBeloBAC vector and introduced into the blunted XhoI site of
pGEM-T-EcoRI G resulting in pGEM-T-EcoRI G BAC. Similarly, for
the right insertion (Fig. 1b), the EcoRI restriction fragment I of the
BoHV-4 V. test strain genome was first cloned into the pGEM-T
Easy vector resulting in pGEM-T-EcoRI I. The BAC cassette released
from the modified pBeloBAC vector by PmeI digestion was then
ligated into the blunted BstEII site of pGEM-T-EcoRI I resulting in
pGEM-T-EcoRI I BAC. The vectors pGEM-T-EcoRI G BAC and
pGEM-T-EcoRI I BAC were used to produce BoHV-4 V. test BAC G
and V. test BAC I recombinant strains, respectively (Fig. 1c). EGFPexpressing recombinant viruses were enriched by six rounds of
plaque purification. The viral genomes were then transferred into
bacteria as described elsewhere (Smith & Enquist, 2000).
Southern blotting. Southern blot analysis was performed as
described previously (Markine-Goriaynoff et al., 2003).
Antibodies. The mouse monoclonal antibody (mAb) 35 raised
against BoHV-4 early-late glycoprotein complex gp6/gp10/gp17 was
used in this study (Dubuisson et al., 1991). Mouse sera raised
against IRAC I and II (anti-IRAC I and anti-IRAC II) were also used
in this study (Daix et al., 2001).
Indirect immunofluorescent staining. Cells grown on glass

coverslips were fixed in PBS containing 4 % (w/v) paraformaldehyde

(Merck) for 10 min on ice and then for 20 min at 20 uC. After washing with PBS, samples were permeabilized in PBS containing 0?1 %
(w/v) NP-40 (Fluka) at 37 uC for 10 min. Immunofluorescent staining (incubation and washes) was performed in PBS containing 10 %
FCS. Samples were incubated at 37 uC for 45 min with one of the
following primary antibodies: mAb 35 (diluted 1 : 1000), anti-IRAC I
(diluted 1 : 100) or anti-IRAC II (diluted 1 : 100). After three washes,
samples were incubated at 37 uC for 30 min with R-phycoerythrin
(PE)-conjugated F(ab9)2 goat anti-mouse Ig (5 mg ml21; Dako) as
the secondary conjugate. Samples were mounted as described elsewhere (Vanderplasschen et al., 2000).
Multi-step growth curves. Triplicate cultures of MDBK cells were

infected at an m.o.i. of 0?5 p.f.u. per cell. After an incubation period

of 1 h, cells were washed and then overlaid with MEM containing
5 % FCS. Supernatant of infected cultures was harvested at successive intervals after infection and the amount of infectious virus was
determined by plaque assay on MDBK cells as described previously
(Vanderplasschen et al., 1993).
Production of BoHV-4 recombinants expressing IRAC I or
IRAC II by mutagenesis in bacteria. Mutagenesis of wild-type V.

test BAC G plasmid was performed by homologous recombination

909

L. Gillet and others

in bacteria using a modified version of the shuttle plasmid pST76KSR expressing recA recombinase (Hobom et al., 2000). Briefly,
pBluescript SK(+) vector (Stratagene) was digested with NspI and
BsrFI to release the lacZ gene containing the multiple cloning site.
The latter fragment was then inserted into pST76K-SR digested with
AvaI and SphI, resulting in pST76KSR-lacZ. Plasmids to induce
homologous recombination were constructed as follows. Firstly,
IRAC I and II ORFs with their stop codon deleted were inserted
into the pcDNA4-TO His/Myc vector (Invitrogen) digested with
BamHI and XhoI, resulting in pcDNA4-TO IRAC I and II. Secondly,
DNA fragments containing human cytomegalovirus (HCMV)

910

immediate-early (IE) promoter fused to IRAC (I or II) ORF and

bovine growth hormone poly(A) [HCMV IEIRAC I/IIBGHpoly(A)]
were released from the latter plasmids by PvuII and BspHI digestion.
After filling in both ends with Klenow, the fragments were inserted
into the blunted XhoI site of pGEM-T-EcoRI I resulting in pGEM-TEcoRI I IRAC I or II. Finally, pGEM-T-EcoRI I IRAC I or II plasmids
were digested with NotI and EcoNI. After blunting, the appropriate
fragments were inserted into the SmaI site of the pST76KSR-lacZ
shuttle vector resulting in pST76KSR-lacZ IRAC I or II. Recombinant
BoHV-4 BAC plasmids were produced by a two-step mutagenesis procedure as described elsewhere (Hobom et al., 2000) (see also Fig. 3).

Journal of General Virology 86

BAC cloning of BoHV-4

Concentration of cell supernatant. Cell supernatants (15 ml per

75 cm2 flask) were collected 2 days post-infection or post-transfection,

centrifuged at 100 000 g for 2 h and concentrated to a final volume
of 250 ml using Amicon Centricon Plus-20 columns (10 000 nominal
molecular weight limit; Millipore).
Immunoblotting. Concentrated cell supernatants were diluted four
times in NuPAGE LDS sample buffer (Invitrogen) containing a
reducing agent. After electrophoresis on NuPAGE Novex 412 %
Pre-cast Bis-Tris gels (Invitrogen), proteins were transferred to
nitrocellulose membranes and identified with specific antibodies
and a chemiluminescence peroxidase substrate kit (Sigma-Aldrich).
Mouse anti-IRAC I and anti-IRAC II sera were used as primary antibodies at a dilution of 1 : 3000.
Anti-complement alternative pathway assay. The ability of

IRAC molecules to inhibit the alternative pathway of the complement system was tested using the AH50 assay (Coligan et al., 1992).
Briefly, 50 ml unsensitized rabbit erythrocytes (Erab) [26108 cells
ml21 in ice cold gelatin/veronal-buffered saline with MgCl2 and
EGTA (GVB/Mg EGTA)] were added on ice to 20 ml human serum
and increasing amounts of the concentrated cell supernatant to be
tested. The final volume of each sample was made up to 150 ml with
GVB/Mg EGTA buffer. For background and total lysis samples,
100 ml GVB/Mg EGTA buffer and 100 ml water were added to the
50 ml Erab, respectively. After an incubation period of 60 min at
37 uC, 1?2 ml ice-cold 0?15 M NaCl was added to each sample to
stop haemolysis. After centrifugation at 1250 g for 10 min at 4 uC,
the supernatant was collected and its optical density measured at
412 nm (OD412). Relative haemolysis of each sample was calculated
as: (test sample OD4122background sample OD412)/(total lysis
sample OD4122background sample OD412).
Confocal microscopy analysis. Confocal microscopy analyses

were performed with a TCS SP confocal microscope (Leica) as

described previously (Vanderplasschen & Smith, 1997).
Statistical analysis. Statistical comparisons were assessed by

analysis of variance. When F ratios were significant (P<0?05), the

method of Scheffes post hoc tests was used.

RESULTS
As mentioned above, BoHV-4 has several intrinsic features
that make it attractive as a backbone for a viral vector and/or
as a model to study gammaherpesvirus biology. However,
these developments have been impeded by the difficulty of

manipulating its large genome. Manipulation of large

herpesvirus genomes has recently been facilitated by using
BAC vectors and prokaryotic recombination technology. In
the present study, we addressed the feasibility of applying
these techniques to BoHV-4.
Cloning of BoHV-4 genome in E. coli
The approach depicted in Fig. 1(c) was used to BAC
clone BoHV-4. Two sites located in non-coding regions of
BoHV-4 L-DNA were selected for insertion of the BAC
cassette (Fig. 1a and b). Simultaneous transfection of
BoHV-4 V. test strain genome and pGEM-T-EcoRI G
BAC or pGEM-T-EcoRI I BAC plasmids into MDBK cells
generated the BoHV-4 V. test BAC G and V. test BAC I
recombinant strains, respectively. The molecular structures
of these recombinant strains were confirmed by a combined
HindIII restriction endonuclease/Southern blot approach
(Fig. 2a) and by sequencing of the regions used to target
recombination (data not shown).
Circular intermediates of BoHV-4 V. test BAC G and V.
test BAC I genomes were isolated from infected cells and
electroporated into E. coli DH10B to generate BoHV-4 V.
test BAC G and I plasmids, respectively. These plasmids
were characterized as described above by a combined
HindIII restriction endonuclease/Southern blot approach
(Fig. 2a). This method confirmed that BoHV-4 V. test
BAC G and I plasmids were two BAC clones of BoHV-4
genome.
Stability of the BoHV-4 genome in E. coli
To assess the stability of BoHV-4 genome as a BAC
plasmid, E. coli DH10B containing BoHV-4 V. test BAC G
or I plasmids was serially cultured for 20 consecutive days,
each day representing approximately 36 generations. After
various periods of culture, the BAC plasmids were isolated
and characterized by HindIII endonuclease digestion. No
difference was observed among plasmids grown for various
periods of time, demonstrating the stability of V. test BAC
G and I plasmids in E. coli (data not shown).

Fig. 2. Characterization of BoHV-4 BAC plasmids and derived strains. (a) Characterization of BoHV-4 BAC plasmids and
derived BoHV-4 strains by a combined restriction endonuclease/Southern blot approach. V. test BAC G and V. test BAC I
plasmids and the genome of BoHV-4 V. test BAC G, V. test BAC I, V. test BAC G excised and V. test BAC I excised strains
were analysed by HindIII restriction (left panels) and further tested by Southern blotting using a probe corresponding to nt
14309940 of the modified pBeloBAC plasmid (right panels). Arrows indicate restriction fragments containing the BAC
cassette. Marker sizes (MS) in kb are indicated on the left. (b) Characterization of BoHV-4 strains derived from BoHV-4 BAC
plasmids by confocal analysis of viral plaques. MDBK cells grown on glass coverslips were infected with BoHV-4 wild-type V.
test (iiii), V. test BAC G (ivvi) and V. test BAC G excised (viiix) strains and then overlaid with MEM containing 5 % FCS
and 0?6 % (w/v) carboxymethylcellulose (Sigma-Aldrich) to obtain isolated plaques. Three days after infection, plaques were
revealed by indirect immunofluorescent staining using mAb 35 and PE-conjugated goat anti-mouse Ig as the primary and
secondary antibodies, respectively. Each set of three horizontal panels (iiii, ivvi and viiix) represents analysis of the same
plaques. EGFP fluorescence is shown in (i), (iv) and (vii), and PE fluorescence in (ii), (v) and (viii). The merged EGFP and PE
signals are shown in (iii), (vi) and (ix). The side of each panel corresponds to 75 mm. Similar results were obtained with the
strains derived from V. test BAC I plasmid (data not shown). (c) Replication kinetics of BoHV-4 strains derived from BoHV-4
BAC plasmids compared with the parental BoHV-4 V. test strain, determined as described in Methods. The data presented
are the means of triplicate measurements.
http://vir.sgmjournals.org

911

L. Gillet and others

Fig. 3. Production of BoHV-4 recombinant strains expressing IRAC I or II by mutagenesis of V. test BAC G plasmid in
bacteria. (a) The IRAC I or II expression cassette was inserted in the BstEII site of the V. test BAC G plasmid corresponding
to the far left BstEII site of BoHV-4 L-DNA. The insertion was performed by homologous recombination between V. test BAC
G and pST76KSR-lacZ IRAC I or II plasmids. (b) Flowchart of stages performed to produce BoHV-4 strains expressing IRAC
I or II.

Reconstitution of infectious virus from BoHV-4

BAC plasmids and excision of the BAC cassette
from reconstituted virus
The usefulness of BAC cloning technology for manipulation
of large DNA viruses requires the ability to reconstitute
infectious virus from the BAC plasmid. Consequently, we
tested whether infectious particles could be produced by
electroporation of BoHV-4 V. test BAC plasmids (Fig. 2b).
HindIII restriction analysis of the DNA of reconstituted
912

viruses revealed restriction profiles identical to the patterns

observed for BoHV-4 V. test BAC plasmids (Fig. 2a).
To excise the BAC cassette, BoHV-4 V. test BAC G and I
reconstituted viruses were propagated in EBL NLSCre
cells to generate BoHV-4 V. test BAC G and I excised
strains, respectively. Deletion of the BAC cassette was confirmed by a combined restriction endonuclease/Southern
blot approach (Fig. 2a) and by monitoring the expression
of EGFP [Fig. 2b, compare (iv) and (vii)].
Journal of General Virology 86

BAC cloning of BoHV-4

Finally, in order to investigate the putative effect of the

recombination processes described above (insertion/
excision of the BAC cassette) on BoHV-4 growth in vitro,
BoHV-4 wild-type V. test, V. test BAC G or I and V. test
BAC G or I excised strains were compared using the growth
assay described in Methods (Fig. 2c). All viruses tested
exhibited similar growth curves (P0?05).
Production of BoHV-4 recombinant strains
expressing IRAC molecules by mutagenesis in
bacteria
In the second part of this study, we tested whether the V.
test BAC G plasmid described above could be modified
by mutagenesis in bacteria in order to produce BoHV-4
recombinants to be used as in vitro expression vectors. To
challenge this approach, we decided to produce BoHV-4
recombinants expressing IRAC I or II proteins under the
control of the HCMV IE promoter. To this end, IRAC
expression cassettes were inserted into the far right BstEII
site of BoHV-4 L-DNA using the V. test BAC G plasmid and
a two-step replacement procedure (Fig. 3a). The resulting
modified plasmids, called V. test BAC G IRAC I or II, were
analysed by a combined HindIII restriction endonuclease/
Southern blot approach (Fig. 4a). IRAC I ORF probe, which
only cross-reacts very weakly with the IRAC II sequence,
hybridized to the expected 4?25 kb band in the HindIII
profile of the V. test BAC G IRAC I plasmid. The IRAC II
ORF probe, which cross-reacts with the IRAC I sequence,
hybridized to the expected 3?95 kb and 232 bp bands in the
HindIII profile of the V. test BAC G IRAC II plasmid. V. test
BAC G IRAC I and II plasmid structures were further
confirmed by sequencing of the regions used to target
recombination (data not shown). Next, V. test BAC G IRAC
I and II plasmids were transfected into permissive EBL
cells to reconstitute BoHV-4 V. test BAC G IRAC I and
II recombinant viruses, respectively. These viruses were
grown on EBL NLSCre to generate BoHV-4 V. test BAC
G IRAC I and II excised strains in which the BAC cassette
had been deleted. The molecular structure of the four
recombinant strains produced was analysed using the
combined HindIII restriction endonuclease/Southern blot
approach described above (Fig. 4a). Once it had been determined that the recombinants had the correct molecular
structure, IRAC expression was tested (Fig. 4b). Immunostaining of BoHV-4 V. test BAC G IRAC I or II and V. test
BAC G IRAC I or II excised plaques revealed that both
transgenes were expressed during the virus replication cycle
[Fig. 4b, (ii), (v), (viii) and (xi)].
The results presented above suggested that infection of
cells by BoHV-4 V. test BAC G IRAC I or II excised strains
could represent an alternative to cell transfection for the
expression of secreted functional IRAC molecules. To test
this hypothesis, MDBK and 293T cells were either transfected with pcDNA4-TO-IRAC I or II, or infected with
BoHV-4 V. test BAC G IRAC I or II excised strains (Fig. 5).
The pEGFP-N1 vector and the BoHV-4 V. test strain were
used as negative controls, respectively. Western blot analysis
http://vir.sgmjournals.org

of concentrated cell supernatants revealed that infection

and transfection of 293T cells led to comparable high level
of IRAC I and II protein expression. It is important to note
that transfection of 293T cells by IRAC-encoding vectors
represented the most efficient expression system among
many tested (data not shown). In MDBK cells, expression
of IRACs induced by transfection was barely detectable,
while expression induced by infection was comparable to
the level observed in 293T cells. Finally, anti-complement
assays performed with concentrated supernatants suggested
that IRAC molecules resulting from transfection or infection are comparably active.
Taken together, the results presented above demonstrate
that the BoHV-4 BAC clones produced in the present study
can be manipulated in bacteria for production of efficient
BoHV-4 expression vectors.

DISCUSSION
BAC cloning has speeded up fundamental and applied
research on several herpesviruses. In the present study, we
BAC cloned the BoHV-4 genome and demonstrated the
quality of the plasmids obtained for production of BoHV-4
recombinants using prokaryotic recombination technology.
The recombinants produced were used to illustrate the
potential of BoHV-4 as an in vitro expression vector.
The BoHV-4 BAC plasmids produced in this study exhibited several interesting features for the production of
BoHV-4 recombinant vectors. Firstly, it was possible to
propagate them stably in bacteria for at least 720 generations. Despite this relative high number, recombination
events were not detected, as described, for example, for
murid herpesvirus 4 BAC (Adler et al., 2000). Secondly,
infectious BoHV-4 recombinant viruses could be efficiently
reconstituted by transfection of the BoHV-4 BAC plasmid
into permissive cells. This feature contrasts with HCMV
(Borst et al., 1999) BAC plasmids, which required the
expression of transactivating factor for efficient production
of virions, and with BoHV-1 (Mahony et al., 2002) and
herpes simplex virus 1 (Stavropoulos & Strathdee, 1998)
BAC plasmids requiring drug treatment to promote IE
gene expression. Thirdly, a major advantage of herpesvirus
vectors is their very large cloning capacity when compared
with other viral vectors (Kootstra & Verma, 2003; Pfeifer &
Verma, 2001; Verma & Somia, 1997). The data presented
in this study demonstrated that at least 10?5 kb of foreign
DNA could be inserted into the BoHV-4 wild-type genome
without affecting its replication properties (Fig. 4b).
The present study illustrates the potential of BoHV-4 as
an in vitro expression vector. However, several properties
of BoHV-4 make it a good candidate for the development
of recombinant vaccines for cattle. Firstly, in contrast to
BoHV-1, there is no eradication scheme against BoHV-4.
Secondly, bovine monocytes and macrophages support
BoHV-4 replication (M. Lambot, unpublished data) suggesting that expression of the transgene in those cells should
913

L. Gillet and others

Fig. 4. Characterization of BoHV-4 recombinant strains expressing IRAC I or II. (a) Characterization of BoHV-4 recombinant
strains expressing IRACs using a combined restriction endonuclease/Southern blot approach. The DNA of the various
intermediates described in Fig. 3(b) was analysed by HindIII restriction (left panel) and further tested by Southern blotting
using IRAC I or II ORF probes (middle and right panels). Arrows indicate the restriction fragments containing IRAC ORFs.
Marker sizes (MS) in kb are indicated on the left. (b) Immunodetection of IRACs expressed by BoHV-4 recombinant strains.
MDBK cells were infected at an m.o.i. of 0?01 with BoHV-4 V. test BAC G IRAC I (iiii), V. test BAC G IRAC I excised (ivvi),
V. test BAC G IRAC II (viiix) and V. test BAC G IRAC II excised (xxii) strain. After an incubation period of 48 h at 37 6C,
cells were analysed by indirect immunofluorescent staining as described in Methods. Anti-IRAC I (ivi) and anti-IRAC II (viixii)
mouse sera were used as primary antibodies and revealed using PE-conjugated goat anti-mouse secondary antibodies. Cells
were then examined by confocal microscopy for EGFP (i, iv, vii and x) and R-PE (ii, v, viii and xi) signals. The merged EGFP
and R-PE signals are shown in (iii), (vi), (ix) and (xii). The side of each panel corresponds to 150 mm.
914

Journal of General Virology 86

BAC cloning of BoHV-4

Fig. 5. Ability of IRAC molecules expressed by BoHV-4 recombinants to inhibit the alternative pathway of the complement
system. MDBK and 293T cells grown in 75 cm2 flask were transfected with pEGFP-N1 or pcDNA4-TO IRAC I or II plasmids,
or infected at an m.o.i. of 1 p.f.u. per cell with BoHV-4 V. test, V. test BAC G IRAC I excised or V. test BAC G IRAC II excised
strain. Forty-eight hours after transfection or infection, cell-culture supernatants were collected and concentrated. Proteins
contained in 5 ml of concentrated supernatants were resolved by SDS-PAGE, blotted and detected with anti-IRAC I or
anti-IRAC II mouse sera as described in Methods. The ability of concentrated cell supernatants to inhibit the alternative
complement pathway was determined by the addition of various amounts of concentrated cell supernatant to the test. Data
are expressed as the percentage of lysis observed compared with the addition of mock-transfected or mock-infected
concentrated cell supernatant and represent the means of triplicate measures. SDs were lower than 0?2.

induce a strong and complete immune response. Thirdly,

the immunity developed against the vector should confer
cross-protection against alcelaphine herpesvirus 1, the
causative agent of malignant catarrhal fever (Rossiter et al.,
1989), which has an important economic impact in Africa.
Lastly, development of BoHV-4 as a recombinant vaccine
will benefit from the rabbit model (Naeem et al., 1990).
The technology applied in the present study for BoHV-4
mutagenesis could contribute to its development as a
recombinant vaccine in the following ways: (i) a major
disadvantage of herpesviruses in the context of vaccine
development is their ability to establish latent infection and
to reactivate. Several strategies could be adopted to prevent BoHV-4 latent infection such as overexpression of
ORF50 (encoding IE2) or deletion of ORF73 (encoding
LANA) (Fowler & Efstathiou, 2004). Strategies based on
ORF73 deletion could be compatible with the use of the
ORF73 regulatory region to provide long-term transgene
expression, as described recently for herpesvirus saimiri
(Giles et al., 2003). (ii) BAC technology could allow the
production of non-replicative vector with no dissemination
http://vir.sgmjournals.org

potential. (iii) The safety of BoHV-4 could be improved

by deletion of non-essential genes responsible for its deleterious effects. For example, our recent study showed that
non-replicative infection by BoHV-4 protects infected cells
against TNF-a-induced apoptosis (Gillet et al., 2004), an
effect mainly due to ORF71 encoding BoHV-4 v-FLIP
(F. Minner, unpublished data). Deletion of the latter gene
among others could improve further the safety and the
cloning capacity of the vector. (iv) In the context of viral
vaccine development, BACs represent a new form of
vaccine that combines the advantages of DNA vaccines
and modified live viruses, since virus can be reconstituted
in vivo after administration of infectious DNA (Meseda
et al., 2004; Petherbridge et al., 2003; Suter et al., 1999;
Tischer et al., 2002). Moreover, recent work (Cicin-Sain
et al., 2003) showed that infectious viruses could be reconstituted in vivo after direct transfer of the viral genome from
bacteria to the vaccinated animal. These authors postulated
that a bacterial vaccination vector delivering an attenuated,
yet infectious virus might present the basis for efficient
vaccines that are easy to store, distribute and administer.
915

L. Gillet and others

(v) Finally, the maximum cloning capacity of BoHV-4 could

be reached by deletion of all non-essential ORFs. BAC-based
technology could be used both to identify non-essential
genes and to delete them. The former goal could be reached
by random transposon mutagenesis of the BoHV-4 genome
in E. coli (Brune et al., 1999).
BAC cloning has speeded up fundamental and applied
research on several herpesviruses. In the present study,
we have BAC cloned the BoHV-4 genome and demonstrated the potential of prokaryotic recombination technology for the production of BoHV-4 recombinants. This
step forward will greatly facilitate further research on this
gammaherpesvirus.

Giles, M. S., Smith, P. G., Coletta, P. L., Hall, K. T. & Whitehouse, A.

(2003). The herpesvirus saimiri ORF 73 regulatory region provides

long-term transgene expression in human carcinoma cell lines.

Cancer Gene Ther 10, 4956.
Gillet, L., Minner, F., Detry, B. & 7 other authors (2004).

Investigation of the susceptibility of human cell lines to bovine

herpesvirus 4 infection: demonstration that human cells can
support a nonpermissive persistent infection which protects them
against tumor necrosis factor alpha-induced apoptosis. J Virol 78,
23362347.
Hobom, U., Brune, W., Messerle, M., Hahn, G. & Koszinowski, U. H.
(2000). Fast screening procedures for random transposon libraries of

cloned herpesvirus genomes: mutational analysis of human cytomegalovirus envelope glycoprotein genes. J Virol 74, 77207729.
Kootstra, N. A. & Verma, I. M. (2003). Gene therapy with viral

vectors. Annu Rev Pharmacol Toxicol 43, 413439.

ACKNOWLEDGEMENTS
A. V., N. M.-G. and L. G. are a Senior Research Associate, Postdoctoral
Researcher and Research Fellow of the Fonds National Belge de la
Recherche Scientifique (FNRS), respectively. Thanks are due to Dr
D. Pirottin (University of Lie`ge, Belgium) and Dr M. Parmentier (Free
University of Brussels, Belgium) for providing the pGBF-NLS-Cre
and pEFIN3 plasmids, respectively. This work was supported by
the following grants: Recherche dinitiative du Ministe`re de la
Region Wallonne programme no. 14628 convention 315543 Region
Wallonne and Service public et Federal sante publique, securite
de la chane alimentaire et environnement (Belgium) programme
no. S-6146. M. W. and U. H. K. were supported by the Deutsche
Forschungsgemeinschaft through SBF 455. M. W. is supported by a
Human Frontiers Science Program long-term fellowship.

Mahony, T. J., McCarthy, F. M., Gravel, J. L., West, L. & Young, P. L.

(2002). Construction and manipulation of an infectious clone of

the bovine herpesvirus 1 genome maintained as a bacterial artificial

chromosome. J Virol 76, 66606668.
Markine-Goriaynoff, N., Georgin, J. P., Goltz, M., Zimmermann, W.,
Broll, H., Wamwayi, H. M., Pastoret, P. P., Sharp, P. M. &
Vanderplasschen, A. (2003). The core 2 b-1,6-N-acetylglucosaminyl-

transferase-mucin encoded by bovine herpesvirus 4 was acquired

from an ancestor of the African buffalo. J Virol 77, 17841792.
Markine-Goriaynoff, N., Gillet, L., Karlsen, O. A., Haarr, L.,
Minner, F., Pastoret, P. P., Fukuda, M. & Vanderplasschen, A.
(2004). The core 2 b-1,6-N-acetylglucosaminyltransferase-M

encoded by bovine herpesvirus 4 is not essential for virus replication

despite contributing to post-translational modifications of structural
proteins. J Gen Virol 85, 355367.
Meseda, C. A., Schmeisser, F., Pedersen, R., Woerner, A. & Weir,
J. P. (2004). DNA immunization with a herpes simplex virus 2

REFERENCES

bacterial artificial chromosome. Virology 318, 420428.

Adler, H., Messerle, M., Wagner, M. & Koszinowski, U. H. (2000).

Messerle, M., Crnkovic, I., Hammerschmidt, W., Ziegler, H. &

Koszinowski, U. H. (1997). Cloning and mutagenesis of a herpes-

Cloning and mutagenesis of the murine gammaherpesvirus 68

genome as an infectious bacterial artificial chromosome. J Virol 74,
69646974.

Naeem, K., Caywood, D. D., Werdin, R. E. & Goyal, S. M. (1990).

Borst, E. M., Hahn, G., Koszinowski, U. H. & Messerle, M. (1999).

Cloning of the human cytomegalovirus (HCMV) genome as an

infectious bacterial artificial chromosome in Escherichia coli: a new
approach for construction of HCMV mutants. J Virol 73, 83208329.
Brune, W., Menard, C., Hobom, U., Odenbreit, S., Messerle, M. &
Koszinowski, U. H. (1999). Rapid identification of essential and

nonessential herpesvirus genes by direct transposon mutagenesis. Nat

Biotechnol 17, 360364.
Cicin-Sain, L., Brune, W., Bubic, I., Jonjic, S. & Koszinowski, U. H.
(2003). Vaccination of mice with bacteria carrying a cloned

virus genome as an infectious bacterial artificial chromosome. Proc

Natl Acad Sci U S A 94, 1475914763.
Evaluation of pregnant rabbits as a laboratory model for bovid
herpesvirus-4 infection. Am J Vet Res 51, 640644.
Petherbridge, L., Howes, K., Baigent, S. J., Sacco, M. A., Evans, S.,
Osterrieder, N. & Nair, V. (2003). Replication-competent bacterial

artificial chromosomes of Mareks disease virus: novel tools for

generation of molecularly defined herpesvirus vaccines. J Virol 77,
87128718.
Pfeifer, A. & Verma, I. M. (2001). Gene therapy: promises and

problems. Annu Rev Genomics Hum Genet 2, 177211.

herpesvirus genome reconstituted in vivo. J Virol 77, 82498255.

Rossiter, P. B., Gumm, I. D., Stagg, D. A., Conrad, P. A.,

Mukolwe, S., Davies, F. G. & White, H. (1989). Isolation of bovine

Coligan, J. E., Kruisbeek, A. M., Margulies, D. H., Shevach, E. M. &

Strober, W. (editors) (1992). Current Protocols in Immunology. New

herpesvirus-3 from African buffaloes (Syncerus caffer). Res Vet Sci 46,
337343.

York: Greene Publishing Associates and Wiley Interscience.

Smith, G. A. & Enquist, L. W. (2000). A self-recombining bacterial

artificial chromosome and its application for analysis of herpesvirus
pathogenesis. Proc Natl Acad Sci U S A 97, 48734878.

Daix, V., Godfroid, E., Pastoret, P. P., Bollen, A. &

Vanderplasschen, A. (2001). New anti-complement proteins.

Patent no. PCT 01870143.3.

Dubuisson, J., Pastoret, P. P. & Thiry, E. (1991). Temporal control of

bovine herpesvirus type 4 glycoprotein synthesis. J Gen Virol 72,

14291434.

Stavropoulos, T. A. & Strathdee, C. A. (1998). An enhanced

packaging system for helper-dependent herpes simplex virus vectors.
J Virol 72, 71377143.

Fowler, P. & Efstathiou, S. (2004). Vaccine potential of a murine

Suter, M., Lew, A. M., Grob, P., Adema, G. J., Ackermann, M.,
Shortman, K. & Fraefel, C. (1999). BAC-VAC, a novel generation of

gammaherpesvirus-68 mutant deficient for ORF73. J Gen Virol 85,

609613.

(DNA) vaccines: a bacterial artificial chromosome (BAC) containing

a replication-competent, packaging-defective virus genome induces

916

Journal of General Virology 86

BAC cloning of BoHV-4

protective immunity against herpes simplex virus 1. Proc Natl Acad
Sci U S A 96, 1269712702.

Vanderplasschen, A., Bublot, M., Dubuisson, J., Pastoret, P. P. &

Thiry, E. (1993). Attachment of the gammaherpesvirus bovine

Thiry, E., Pastoret, P. P. & Calberg-Bacq, C. M. (1981). Isolement

dun herpesvirus chez un taureau infertile. Ann Med Vet 125, 143.

herpesvirus 4 is mediated by the interaction of gp8 glycoprotein with

heparinlike moieties on the cell surface. Virology 196, 232240.

Thiry, E., Bublot, M., Dubuisson, J., Van Bressem, M. F., Lequarre,
A. S., Lomonte, P., Vanderplasschen, A. & Pastoret, P. P. (1992).

Vanderplasschen, A., Markine-Goriaynoff, N., Lomonte, P. & 8

other authors (2000). A multipotential b-1,6-N-acetylglucosaminyl-

Molecular biology of bovine herpesvirus type 4. Vet Microbiol 33,

7992.

transferase is encoded by bovine herpesvirus type 4. Proc Natl Acad

Sci U S A 97, 57565761.

Tischer, B. K., Schumacher, D., Beer, M. & 7 other authors (2002).

Verma, I. M. & Somia, N. (1997). Gene therapy promises, problems

and prospects. Nature 389, 239242.

A DNA vaccine containing an infectious Mareks disease virus

genome can confer protection against tumorigenic Mareks disease in
chickens. J Gen Virol 83, 23672376.
Vanderplasschen, A. & Smith, G. L. (1997). A novel virus binding

assay using confocal microscopy: demonstration that the intracellular

and extracellular vaccinia virions bind to different cellular receptors.
J Virol 71, 40324041.

http://vir.sgmjournals.org

Wagner, M., Ruzsics, Z. & Koszinowski, U. H. (2002). Herpesvirus

genetics has come of age. Trends Microbiol 10, 318324.

Zimmermann, W., Broll, H., Ehlers, B., Buhk, H. J., Rosenthal, A.
& Goltz, M. (2001). Genome sequence of bovine herpesvirus 4,

a bovine rhadinovirus, and identification of an origin of DNA

replication. J Virol 75, 11861194.

917