Journal of Virological Methods 183 (2012) 86–89

Contents lists available at SciVerse ScienceDirect

Journal of Virological Methods

j ournal homepage: www.elsevier.com/locate/jviromet

Short communication

Novel approach for the generation of recombinant African swine fever virus from

a field isolate using GFP expression and 5-bromo-2!-deoxyuridine selection

a b a,

Raquel Portugal , Carlos Martins , Günther M. Keil ∗

a

Friedrich-Loeffler-Institut, Südufer 10, 17493 Greifswald-Insel Riems, Germany

b

Laboratório de Doenc¸ as Infecciosas, CIISA, Faculdade de Medicina Veterinária, Technical University of Lisbon, Lisbon, Portugal

a b s t r a c t

Article history: Generation of African swine fever virus (ASFV) recombinants has so far relied mainly on the manipulation

Received 4 August 2011

of virus strains which had been adapted to growth in cell culture, since field isolates do not usually

Received in revised form 14 March 2012

replicate efficiently in established cell lines. Using wild boar lung cells (WSL) which allow for propagation

Accepted 21 March 2012

of ASFV field isolates, a novel approach for the generation of recombinant ASFV directly from field isolates

Available online 4 April 2012

was developed which includes the integration into the viral thymidine kinase (TK) locus of an ASFV p72-

promoter driven expression cassette for enhanced green fluorescent protein (EGFP) embedded in a 16 kbp

Keywords:

mini F-plasmid into the genome of the ASFV field strain NHV. This procedure enabled the monitoring of

African swine fever virus recombinants

recombinant virus replication by EGFP autofluorescence. Selection for the TK-negative (TK−) phenotype

Field isolate

of the recombinants on TK− Vero (VeroTK−) cells in the presence of 5-bromo-2!-deoxyuridine (BrdU)

Green fluorescent protein

+

Thymidine kinase led to efficient isolation of recombinant virus due to the elimination of TK wild type virus by BrdU-

BrdU selection phosporylation in infected VeroTK− cells. The recombinant NHV-dTK-GFP produced titres of both cell-

associated and secreted viral progeny in WSL cells similar to parental NHV indicating that insertion of

large heterologous sequences into the viral TK locus and EGFP expression do not impair viral replication

in these cells. In summary, a novel method has been developed for generation of ASFV recombinants

directly from field isolates, providing an efficacious method for further manipulations of wild-type virus genomes.

© 2012 Elsevier B.V. All rights reserved.

African swine fever virus (ASFV), or as proposed recently African remain largely unknown which underlines the need to develop

swine fever asfivirus (Van Regenmortel et al., 2010) is classified as strategies to facilitate the study and manipulation of this complex

the sole member of the family Asfarviridae, genus Asfivirus (Dixon virus.

et al., 2005). The size of the double stranded DNA genome varies Generation of ASFV recombinants has relied mainly on the

between 170 and 190 kbp, depending on the virus isolate. The mutagenesis of cell-culture adapted viruses since field isolates –

African swine fever virus (ASFV) infects all members of the Suidae with the exception of COS-1 cells (Hurtado et al., 2010) – do not

family. In domestic pigs and wild boars it causes African swine fever grow well in cultured cells. On the other hand, working with pri-

(ASF), a highly contagious hemorrhagic disease with high mortality mary swine macrophages, the natural host cells of ASFV, has proven

rates for which no efficacious vaccine is available (for review see to be difficult. In this report the use of a new cell line derived from

Tulman et al., 2009). Therefore it constitutes a major threat for pig wild boar lung cells (WSL, provided by the Collection of Cell Lines

husbandry worldwide, highlighted particularly by the recent intro- in Veterinary Medicine, FLI Insel Riems, Germany) is described,

duction of ASFV into Caucasian countries (Rowlands et al., 2008; which is suitable for efficient propagation of several ASFV field

Costard et al., 2009; Rahimi et al., 2010) and its ongoing spread in isolates (unpublished results), in a novel approach for generation

the affected area. of recombinant ASFV directly from the field isolate NHV, a non-

The ASFV genome contains approximately 150 open read- fatal, non-haemadsorbing ASFV strain, isolated from a pig infected

ing frames (ORFs) coding for proteins with functions at both chronically (Vigário et al., 1974) which provided the basis for a use-

the cellular and the viral replication and morphogenesis levels ful and reliable infection model for studies on the mechanisms of

(Yánez˜ et al., 1995) which account for the high complexity of the protective immunity (Leitao et al., 2001). To this end, heterologous

virus-host interactions. Pathogenesis and virulence determinants sequences encompassing the gene for enhanced green fluorescent

protein (EGFP) were integrated into the thymidine kinase (TK) locus

of NHV Cells infected with the TK-negative, EGFP-positive recom-

binant virus could be detected easily by fluorescence microscopy.

∗ Corresponding author. Tel.: +49 38351 71272; fax: +49 38351 71151.

E-mail address: Guenther.Keil@fli.bund.de (G.M. Keil). Subsequently, viral mutants were selected positively on a

0166-0934/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jviromet.2012.03.030

R. Portugal et al. / Journal of Virological Methods 183 (2012) 86–89 87

TK-negative Vero cell line using BrdU to eliminate TK-positive wild

type virus.

For generation of recombinant ASFV, a transfer plasmid con-

taining the EGFP ORF under transcriptional control of the promoter

from the gene encoding vp72 (López-Otín et al., 1990), the major

viral structural protein, and flanked by segments of the viral TK

gene, was constructed (Fig. 1). The resulting plasmid pASFV-dTK-

EGFP-BAC-Lox had been designed initially for cloning of the ASFV

genome as a bacterial artificial chromosome. It contains the viral

TK-spanning locus from nt 43,995 to nt 50,739 with the same 316 bp

deletion within the TK ORF as described by Moore et al. (1998). The

TK ORF flanking sequences are both about 3 kbp in size to provide

longer sequence segments for homologous recombination as used

in previous constructs to target the same genomic region (Moore

et al., 1998).

To generate recombinants, 4 ␮g of plasmid pASFV-dTK-EGFP-

BAC-Lox were transfected into semi-confluent WSL cells in 6-well

6

plates (approximately 10 cells per well) using the FuGene HD

transfection reagent as recommended by the supplier (Roche,

Mannheim, Germany). The medium was removed 5 h after trans-

fection and the cells were infected with NHV at an MOI of 2. The

inoculum was removed 1 h after adsorption. Cells were washed

with culture medium and incubated further in fresh medium for

3 days, when autofluorescing foci of rounded and granulated cells

indicated productive replication of recombinant virus. Infected

cells from these foci were collected by aspiration and re-inoculated

onto WSL cells after one 70 C freeze/thaw cycle. Cells from aut-

− ◦

ofluorescing foci were harvested as mentioned above and used for

infection of bromodeoxyuridine (BrdU)-resistant Vero (VeroTK−)

cells in presence of 50 ␮g/ml BrdU for positive selection of recom-

binants. VeroTK− cells were selected using a strategy employed by

Bello et al. (1987) for MDBK cells and Kit et al. (1966) for HeLa cells

and kindly provided by Roland Riebe, FLI, Insel Riems, Germany.

Appearance of autofluorescent cells was monitored daily. At 7 dpi

cultures were harvested and after 2 freeze/thaw cycles aliquots of

Fig. 1. Construction of recombination plasmid pASFV-dTK-EGFP-BAC Lox. (A)

the virus/cell suspension were again added to Vero TK− cells and Schematic representation of the ASFV genome region containing the TK ORF.

incubated in the presence of 50 ␮g/ml BrdU. Nucleotide numbers are given in kilobases (kb), names and direction of transcrip-

tion of contained ORFs are indicated. The location of the TK-ORF (K196R) is shown in

Fluorescent foci which consisted of only a few cells and thus

bold. (B) Plasmid constructions. The left (TK-L) and right (TK-R) segments of the ASFV

were considerably smaller than foci in WSL cell cultures, were col-

TK gene and respective flanking sequences were amplified by PCR from infected

lected and passaged again on Vero TK cells in medium containing

− macrophages DNA. Primers used were TK-L+ (GTG GGC GTA TAG ATA AGG ATA TC)

50 ␮g/ml BrdU. After two further rounds of positive selection, GFP- and TK-L (TAA GGT ACC GTG TTT TAA TAG TTT TGT CTC GGG TG) amplifying a

−

3207 bp fragment from nt 43,995 to nt 47,201 (TK-L), and primers TK-R+ (TGA CCC

expressing infected cells were freeze/thawed twice and used for

GGG CGT AAG AAC GCA GAC AAG ACG C) and TK-R (CCT GCT CGT GTT ACT TAT

the infection of WSL cells to obtain high titre stocks of recombi- −

GAA AC) amplifying a 3236 bp fragment from nt 47,504 to nt 50,739 (TK-R). All

nant virus. Large fluorescent foci readily developed and finally led

nucleotide numbers are from GenBank accession # U18466.1. Both amplicons were

to the isolation of the ASFV recombinant NHV-dTK-GFP. To test sequentially cloned into plasmid vector pSP73 (Promega) using established standard

for homogeneity of the recombinant virus preparation, WSL cul- procedures. TK-L and TK-R were blunt ended with Klenow polymerase. TK-L was

then cleaved with Acc65I and inserted into pSP73 cleaved with Acc65I and EcoRV.

tures on coverslips were inoculated with approximately 100 PFU.

TK-R was cleaved with SmaI, and cloned into the TK-L containing plasmid after cleav-

At 4 days p.i., cells were fixed and stained for detection of ASFV

age with SmaI to yield pspASFV-dTK, containing the viral TK-spanning locus from nt

infected cells by indirect immunofluorescence using mouse mon-

43,995 to nt 50,739 with a 304 bp deletion from 47,202 to 47,503. Plasmid pspASFV-

oclonal antibody C18 directed against the early viral protein vp30 dTK was then used for the integration of the synthetic sequence GGT ACC GTA TAC

GCG GCC GC CCC GGG

(kindly provided by Linda Dixon, Pirbright, UK). Fig. 2A shows that A TAA CTT CGT ATA ATG TAT GCT ATA CGA AGT TAT , which

contains a loxP site (shown in bold) flanked by recognition sequences for Acc65I,

all foci with green autofluorescence (left picture) were recognized

BstZ17Iand NotI, and for SmaI (printed in italics). A p72-EGFP expression cassette

also by the anti-vp30 antibodies (red fluorescence, right picture).

was afterwards also integrated, containing the ASFV vp72 promoter region (López-

No non-autofluorescing foci were observed among more than 100 Otín et al., 1990) from nt 218 to nt 430 (GenBank accession # M34142.1) fused to

infected cell foci, indicating that the recombinant virus stock was the EGFP ORF from plasmid pEGFP-N1 (Clontech), resulting in pASFV-dTK-EGFP-

Lox. The 6385 bp plasmid pMBO131 (kindly provided by W. Fuchs, FLI) was cloned

essentially free from contaminating parental NHV virus. This con-

into the blunt-ended NotI site of pASFV-dTK-EGFP-Lox, resulting in the final trans-

clusion was supported by the results of PCR assays (Fig. 2B) which

fer plasmid construct pASFV-dTK-EGFP-BAC-Lox. Only relevant restriction enzyme

revealed that no wild type virus genomes were detectable in NHV-

cleavage sites are indicated. Segments and plasmids are not drawn to scale. All PCR

dTK-GFP infected WSL cells. reaction conditions are available upon request.

To test whether TK-deletion and EGFP-expression affect in vitro

replication of the recombinant, WSL cells were infected with NHV-

dTK-EGFP or parental NHV. At the time points indicated in Fig. 3, NHV, indicating that insertion of the foreign sequences into the

titres of cell-associated virus and infectivity released into the cul- viral TK locus and expression of EGFP do not impair viral repli-

ture medium were determined by titration on WSL cells. As shown, cation and release of infectious virions in WSL cells. It should be

NHV-dTK-GFP produced similar or even slightly higher titres of noted that the increased virus yield, although statistically not sig-

both cell associated and secreted viral progeny in comparison to nificant, might reflect an adaptation process for virus replication in

88 R. Portugal et al. / Journal of Virological Methods 183 (2012) 86–89

Fig. 2. Homogeneity of ASFV recombinant NHV-dTK-GFP. (A) WSL cells, grown on coverslips, were infected with an appropriate dilution of NHV-dTK-GFP stock virus and

fixed 4 days p.i. with 3% paraformaldehyde in PBS for 20 min, permeabilized with 0.2% Triton X-100 in 3% paraformaldehyde/PBS for 10 min, washed 4 times for 5 min with

PBS and incubated with anti-vp30 monoclonal antibody C18 for 1 h at room temperature. After 4 washes with PBS for 5 min each, bound antibodies were visualized by

incubation with Alexa Fluor 555-conjugated goat anti-mouse serum (Invitrogen, Karlsruhe, Germany) for 1 h. The coverslips were mounted on microscope slides with 1,4-

diazabicyclo[2.2.2]octane (DABCO) in PBS/glycerol and fluorescing cells were photographed and pictured using a Zeiss Axioskop fluorescence microscope with CCD camera

and AxioVision software, respectively. GFP autofluorescence is shown in the left panel, bound Alexa Fluor 555 is shown in the right panel. (B) PCR amplification with primers

CTT ATT CAT TGC ATT TAC ATG CTC G and ACA ACA TGT TAC GTA CAG TTC AC which target TK ORF sequences flanking the p72EGFP-BAC-Lox insert (see Fig. 1) on whole-cell

DNA extracted from WSL cells infected with wild type NHV (lane 1), non-infected WSL cells (lane 2), whole-cell DNA extracted from WSL cells infected with NHV-dTK-GFP

(lane 3), no-template control (lane 4). M: 1 kbp ladder (Invitrogen). PCR reaction conditions are available upon request.

Fig. 3. Infectious replication of NHV-dTK-GFP and wild type NHV in WSL cells is comparable. WSL cultures were infected with wild type NHV (closed circles) or NHV-dTK-GFP

(closed triangles) at an MOI of 0.5. Cultures were washed with medium after 1 h adsorption. At the times indicated, culture supernatants were collected and adherent cells

were washed with medium which was added to the respective supernatants. Fresh medium corresponding to the total supernatant volume was added to the adherent cells.

Cell cultures and supernatants were stored at 70 C until titration. Virus titres were determined after 2 freeze/thaw cycles (cell associated virus) by inoculation of serial

− ◦

dilutions on WSL cell monolayers in 96-well tissue culture plates. Values are means of three independent determinations. Standard deviations are indicated.

R. Portugal et al. / Journal of Virological Methods 183 (2012) 86–89 89

cell culture. Since NHV replicates only inefficiently in Vero TK− cells be prevented? Philosophical Transactions of the Royal Society of London B:

Biological Sciences 364, 2683–2696.

which might also contribute to cell culture adaptation, attempts are

Dixon, L.K., Escribano, J.M., Martins, C., Rock, D.L., Salas, M.L., Wilkinson, P.J., 2005. In:

currently being made to select a TK-deficient variant of the WSL

Fauquet, C.M., Mayo, M.A., Maniloff, J, Desselberger, U., Ball, L.A. (Eds), Vírus Tax-

cell line which, however is highly sensitive to BrdU already at a onomy, VIIIth Report of the ICTV, Elsevier/Academic Press, London, pp. 135–143.

Garcia-Escudero, R., Andrés, G., Almazán, F., Vinuela,˜ E., 1998. Inducible gene expres-

concentration of 0.5 ␮g/ml.

sion from African swine fever virus recombinants: analysis of the major capsid

The ASFV TK locus has been used previously for generation of

protein p72. Journal of Virology 72, 3185–3195.

recombinant viruses from cell culture adapted isolates since TK García, R., Almazán, F., Rodríguez, J.M., Alonso, M., Vinuela,˜ E., Rodríguez, J.F., 1995.

is non-essential for viral growth in cultured cells (García et al., Vectors for the geneic manipulation of African swine fever virus. Journal of

Biotechnology 40, 121–131.

1995; Brun et al., 1999; Garcia-Escudero et al., 1998). Disrup-

Hurtado, C., Bustos, M.J., Carrascosa, A.L., 2010. The use of COS-1 cells for studies

tion of the viral TK locus from pathogenic ASFV isolates partially of field and laboratory African swine fever virus samples. Journal of Virological

adapted to Vero cells has been reported to impair viral growth Methods 164, 131–134.

Kit, S., Dubbs, D.R., Frearson, P.M., 1966. HeLa cells resistant to bromodeoxyuridine

in macrophages after a low MOI inoculation (0.01) but not after

and deficient in thymidine kinase activity. International Journal of Cancer 1,

infection at a MOI of 10–20, and to decrease pathogenicity for 19–30.

swine (Moore et al., 1998). Whether NHV-dTK-EGFP exhibits Leitao, A., Cartaxeiro, C., Coelho, R., Cruz, B., Parkhouse, R.M.E., Portugal, F.C., Vigario,

J.D., Martins, C.L.V., 2001. The non-haemadsorbing African swine fever virus

a similar phenotype in macrophages and in vivo needs to be

isolate ASFV/NH/P68 provides a model for defining the protective anti-virus

analyzed.

immune response. Journal of General Virology 82, 513–523.

In summary, a novel method was developed for the genera- López-Otín, C., Freije, J.M., Parra, F., Méndez, E., Vinuela,˜ E., 1990. Mapping and

sequence of the gene coding for protein p72, the major capsid protein of African

tion of ASFV recombinants directly from field isolates, providing a

swine fever virus. Virology 175, 477–484.

method for further manipulations of such wild type virus genomes.

Moore, D.M., Zsak, L., Neilan, J.G., Lu, Z., Rock, D.L., 1998. The African swine fever virus

thymidine kinase gene is required for efficient replication in swine macrophages

Acknowledgments and for virulence in swine. Journal of Virology 72, 10310–10315.

Rahimi, P., Sohrabi, A., Ashrafihelan, J., Edalat, R., Alamdari, M., Masoudi, M., Mostofi,

S., Azadmanesh, K., 2010. Emergence of African swine fever virus northwestern

We thank Thomas C. Mettenleiter for valuable comments and Iran. Emerging Infectious Diseases 16, 1946–1948.

Rowlands, R.J., Michaud, V., Heath, L., Hutchings, G., Oura, C., Vosloo, W., Dwarka,

suggestions on the manuscript, Roland Riebe for supplying WSL

R., Onashvili, T., Albina, E., Dixon, L.K., 2008. African swine fever virus isolate,

and Vero TK-cells, and Linda Dixon and Walter Fuchs for providing

Georgia, 2007. Emerging Infectious Diseases 14, 1870–1874.

materials and Anette and Craig Beidler for proofreading. This work Tulman, E.R., Delhon, G.A., Ku, B.K., Rock, D.L., 2009. African swine fever virus. Current

was funded through the EU-Project # 211691, FP7-KBEE-2007-1- Topics in Microbiology and Immunology 328, 43–87.

3-05. Van Regenmortel, M.H., Burke, D.S., Calisher, C.H., Dietzgen, R.G., Fouquet, C.M.,

Ghabrial, S.A., Jahrling, P.B., Johnson, K.M., Holbrook, M.R., Horzinek, M.C., Keil,

G.M., Kuhn, J.H., Mahy, B.W., Martelli, G.P., Pringle, C., Rybicki, E.P., Skern, T., Tesh,

References R.B., Wahl-Jensen, V., Walker, P.J., Weaver, S.C., 2010. A proposal to change exist-

ing virus species names to non-Latinized binomials. Archives of Virology 155,

Bello, L.J., Whitbeck, J.C., Lawrence, W.C., 1987. Map location of the thymidine kinase 1909–1919.

gene of bovine herpesvirus 1. Journal of Virology 61, 4023–4025. Vigário, J.D., Terrinha, A.M., Moura Nunes, J.F., 1974. Antigenic relationships among

Brun, A., Rodríguez, F., Parra, F., Sobrino, F., Escribano, J.M., 1999. Design and strains of African swine fever virus. Archiv fur Die Gesamte Virusforschung 45,

construction of African swine fever virus chimeras incorporating foreign viral 272–277.

epitopes. Archives of Virology 144, 1287–1298. Yánez,˜ R.J., Rodriguez, J.M., Nogal, M.L., Yuste, L., Enriquez, C., Rodriguez, J.F., Vinuela,˜

Costard, S., Wieland, B., de Glanville, W., Jori, F., Rowlands, R., Vosloo, W., Roger, E., 1995. Analysis of the complete nucleotide sequence of African swine fever

F., Pfeiffer, D.U., Dixon, L.K., 2009. African swine fever: how can global spread virus. Virology 208, 249–278.