ORIGINAL ARTICLE

Immunogenicity of the capsid precursor and a nine-amino-acidsite-directed mutant of the 3C protease of foot-and-mouth diseasevirus coexpressed by a recombinant goatpox virus

Wenge Ma • Jie Wei • Yurong Wei • Huiling Guo •

Yinghong Jin • Ying Xue • Yan Wang • Zhong Yi •

Liya Liu • Jiong Huang • Lijian Wang

Received: 31 August 2013 / Accepted: 11 January 2014

� Springer-Verlag Wien 2014

Abstract The myristoylated capsid precursor mP1-2A of

foot-and-mouth disease virus (FMDV), when expressed in

mammalian cells and processed by the FMDV 3C protease,

can self-assemble into virus-like particles (VLPs). In the

present study, nine amino acids of the 3C protease were

replaced by site-directed mutagenesis to create a mutant 3C

protease, 9m3C. To coexpress mP1-2A and 9m3C and test

the resulting proteolytic processing and VLP assembly, two

recombinant goatpox viruses (rGTPVs) were constructed

by the insertion of two coding regions, one for mP1-2A and

the other for either 9m3C (rGTPV-mP1-2A-9m3C) or

Theileria protective antigen (TPA) as a control (rGTPV-

mP1-2A-TPA). The two exogenous genes were inserted

into an intergenic region between loci gp_24 and gp_24.5

of the rGTPV genome. Western blotting of cells infected

with rGTPV-mP1-2A-9m3C showed that proteins VP0,

VP1, and VP3 from the mP1-2A processed by the 9m3C

protease could be detected by polyclonal FMDV sera. As

observed by electron microscopy, the infected cells pro-

duced VLPs with a diameter of about 25 ± 2 nm. Titers of

neutralizing antibody against FMDV were significantly

higher in mice inoculated with rGTPV-mP1-2A-9m3C,

which expresses the 9m3C protease together with mP1-2A,

than mice inoculated with the control rGTPV-mP1-2A-

TPA, which does not express the protease. An ovine

immunization test determined that sheep inoculated intra-

muscularly with rGTPV-mP1-2A-9m3C produced FMDV-

specific neutralizing antibody, but its titers did not meet the

requirement of the World Organization for Animal Health.

The result indicates that further modifications of rGTPV-

mP1-2A-9m3C are necessary to produce an effective

vaccine.

Introduction

Foot-and-mouth disease virus (FMDV) belongs to the

genus Aphthovirus in the family Picornaviridae and is the

causative agent of a highly contagious disease affecting

cloven-hoofed animals. Foot-and-mouth disease (FMD)

presents a significant threat to animal health, food safety,

and international trade, and therefore prevention strategies

against the disease have been developed that include vac-

cination with inactivated whole-virus vaccines, and

slaughter policies [1]. However, concerns accompanying

the vaccination include virus escape, relatively shorter

duration of immune protection, and harmful side effects.

Alternative vaccine approaches to prevent the disease,

including peptide-fused epitopes [2] and recombinant live

vectors [3, 4], have been investigated, of which recombi-

nant poxviruses are expected to be effective alternatives.

Goatpox virus (GTPV) belongs to the genus Capripox-

virus in the family Poxviridae. GTPV replication and

assembly occur at discrete sites within the cytoplasm of the

host cell, and virions can be released by budding, exocy-

tosis, or cell lysis, but most virions are not enveloped and

are released by cell lysis. Importantly, this mode of virus

release makes it possible that exogenous antigens expressed

in the cytoplasm could be recognized by circulating B cells.

In fact, attenuated GTPV carrying exogenous antigens for

sheep, goat, and cattle immunization has been demonstrated

to be efficacious and safe [3, 5, 6]. In view of all these

W. Ma (&) � J. Wei � Y. Wei � H. Guo � Y. Jin � Y. Xue �Y. Wang � Z. Yi � L. Liu � J. Huang � L. Wang

Institute of Veterinary Medicine, Xinjiang Academy of Animal

Science, 151 Eastern Kelamayi Street, Urumqi 830000,

People’s Republic of China

e-mail: mawenge2004@aliyun.com

123

Arch Virol

DOI 10.1007/s00705-014-1984-8

advantages, recombinant GTPV (rGTPV) was employed for

the expression of the myristoylated capsid precursor (mP1-

2A) and 3C protease of FMDV in this study.

The intergenic regions of poxviruses are accepted as

suitable sites for the insertion of exogenous genes. How-

ever, few intergenic regions in GTPV have been identified

as stable, and therefore the expression of foreign genes

inserted into different intergenic regions of GTPV is often

unpredictable [6, 7]. In this study, an intergenic region

located between the homologous loci gp_24 and gp_24.5 of

GTPV strain CVCC AV41 was selected for the insertion of

FMDV mP1-2A and 3C protease genes to evaluate this

region’s ability to support the insertion and expression of

exogenous genes [8].

Seven FMDV serotypes have been identified to date,

namely A, O, C, Asia-1, and South African Territories 1, 2,

and 3. The FMDV virion is nonenveloped and has an i-

cosahedrally symmetrical capsid composed of 60 copies of

a protomer assembled from structural proteins VP1, VP2,

VP3, and VP4. Major antigenic sites exist in the VP1, VP2,

and VP3 capsid proteins [9]. Proteolytic processing of

mP1-2A by protease 3C produces the proteins VP0, VP3,

and VP1, which self-assemble to form empty capsid par-

ticles [10]. It is generally accepted that the intact virion is

most antigenic and that the empty capsid possesses similar

immunogenicity to the native virion, whereas its precursor

does not [11, 12]. The 3C protease can digest the capsid

precursor; however, apoptosis is induced in cells infected

with recombinant poxvirus expressing 3C [13]. Therefore,

in the present study, we aimed to decrease the 3C prote-

ase’s apoptotic effect on its host, so as to produce an

rGTPV that could coexpress the capsid precursor and 3C

protease of FMDV. Moreover, the rGTPV was analyzed to

determine whether the FMDV structural proteins it

expressed could be proteolytically processed to assemble

into VLPs, as well as to determine its immunogenicity.

Materials and methods

Viruses, plasmids, and sera

The cold-adapted GTPV strain CVCC AV41 as an atten-

uated live vaccine was manufactured by Xinjiang Tecon

Animal Husbandry Biotechnology Co., Ltd (XJTC,

Urumqi, China). The virus was preserved and propagated

in lamb testis (LT) cells to obtain a titer of approximately

105.5 50 % tissue culture infectious doses (TCID50)/mL.

The virus was prepared in LT cells in Dulbecco’s modified

essential medium (DMEM) supplemented with 10 % fetal

calf serum (FCS) and incubated at 37 �C and 5 % CO2.

The commercial vaccine made from inactivated FMDV

isolate Asia-1/Jiangsu/China/2005 (GenBank accession no.

EF149009) and polyclonal sera derived from convalescent

cattle infected with the FMDV were both obtained from

XJTC. The plasmid pGPT-TPA, preserved by Institute of

Veterinary Medicine, Xinjiang Academy of Animal Sci-

ence, China, includes the coding sequence of Theileria

protective antigen (TPA), the internal ribosome entry site

(IRES) from the plasmid pIRES2-EGFP (Clontech, USA),

the coding sequence of Escherichia coli guanine phos-

phoribosyltransferase (GPT) [14], and the canonical

sequences of the 11K late promoter and 7.5K early/late

promoter, both derived from vaccinia virus [3, 5, 15].

The plasmid pP1-2A includes the coding regions of P1

and 2A, which were amplified from the FMDV isolate Asia-

1/Jiangsu/China/2005. The plasmid p9m3C includes the

coding region of the 9m3C gene, which was synthesized by

Sangon Biotech (Shanghai) Co., Ltd (China). The encoded

9m3C protease differs from the native 3C protease of

FMDV isolate Asia-1/Jiangsu/China/2005 by the presence

of nine point mutations, 10K ? N, 60R ? Q, 92R ? N,

95R ? Q, 97R ? H, 101K ? Q, 126R ? N, 196R ? Q,

and 203K ? Q. These amino acid substitutions were

selected on the basis of the X-ray crystal structure of the

FMDV 3C protease [13], and the quality of the protein

structure of each mutant containing different substitutions

was estimated by a comparative modeling method [16].

Identification and amplification of homologous arms

Two 1.1-kbp fragments encompassing the intergenic region

flanked by the loci of gp_24 and gp_24.5 as homologous

recombinant arms were amplified by PCR using primer

pairs designed according to the genome of GTPV strain

Pellor (GenBank accession no. 21426072) and using the

genomic DNA of GTPV strain CVCC AV41 as a template.

The PCR products were isolated and ligated into the

plasmid pMD-18T (Takara, Japan). The primer pair tar-

geting the upstream homologous arm were 5’-GCG AGA

TCT GTC TGA CCA ATC TCT TTC AAA T-3’ (sense)

and 5’-GTC GGA TCC ATG TGG TCC TTA TTT TTT

TCA A-3’ (antisense), and those targeting the downstream

homologous arm were 5’-GCG GGA TCC CAA TGT TTG

TTT TTC TT CAA TA-3’ (sense) and 5’-CGC AGA TCT

TCA TTT ATA TAA CTT TTT AAT G-3’ (antisense),

with the restriction sites used for cloning underlined. The

two fragments were sequenced to verify identity to the

parental sequences.

Construction of transfer plasmids encoding mP1-2A

and either 9m3C or TPA

The upstream and downstream homologous PCR fragments

were ligated into the plasmid pCI (Promega, Madison, WI,

USA) through the BglII and BamHI restriction sites,

M. Wenge et al.

123

respectively. Afterwards, the recombinant plasmid was

sequenced to verify the integrity and cis-insertion of the

homologous fragments. The TPA-IRES-GPT-P11-P7.5

cassette, amplified from plasmid pGPT-TPA by PCR, was

ligated back into the plasmid through the NheI and MluI

restriction sites. Subsequently, the coding region for mP1-

2A was created by amplifying the P1-2A coding sequence

from pP1-2A by PCR and incorporating an upstream

sequence via a mutagenic oligonucleotide containing the

myristoylation consensus sequence MGXXXS [10, 17, 18].

The resulting product was inserted through the MluI and

NotI restriction sites. Afterwards, the transfer plasmid

including the TPA-IRES-GPT-P11-P7.5 cassette and the

mP1-2A coding sequence was designated as pmP1-2A-

TPA. Last, the 9m3C protease–encoding region was

obtained from p9m3C by PCR and ligated into pmP1-2A-

TPA through restriction sites BstXI and NheI. The transfer

plasmid based on pmP1-2A-TPA and including the coding

region of the 9m3C protease was designated as p9m3C-

mP1-2A. The transfer plasmids were sequenced to verify

their integrity and were introduced into E. coli strain

JM109 (Promega) by transformation for plasmid amplifi-

cation and isolation. The primer pairs for amplifying the

TPA-IRES-GPT-P11-P7.5 cassette were 5’-TC GCT AGC

CCC GTC AGA GTC ATC ATC ATG GTG T-3’ (sense)

and 5’-GA ACG CGT GTC ACT GTT CTT TAT GAT

TCT ACT T-3’ (antisense), and those used to amplify the

mP1-2A-encoding region were 5’-GCG ACG CGT GCC

ACC ATG GGA GCT GGG CAA TCC AGT CCG GCG

A-3’ (sense) and 5’-GTA GCG GCC GC ACA AAA A

TTA GAA GGG CCC AGG GTT GGA CTC CAC-3’

(antisense). The primer pairs for amplifying the 9m3C

protease–encoding region were 5’-ATA CCA CCG CGG

TGG GCC ACC ATG AGA GTG GTG CCC CAC CGA

CTG ACT TGC AAC AGA-3’ (sense) and 5’-TA GCT

AGC ACA AAA A TTA CTC GTG GTG TGG TTC GGG

ATC AAT GTG AGC CT-3’ (antisense). The initiation and

stop codons are indicated with wavy underlining, myris-

toylation consensus sequence with dashed underlining, and

restriction sites with solid underlining. A schematic

depiction of the genes and features of pmP1-2A-TPA and

pmP1-2A-9m3C is shown in Fig. 1.

Homologous recombination and screening

The procedure used for homologous recombination was

described previously [5, 8]. Briefly, confluent LT cells

were cotransfected with one of the transfer plasmids and

the parental GTPV strain CVCC AV41 using the non-

liposomal lipid formulation Effectene transfection reagent

(QIAGEN, Hilden, Germany) according to the manufac-

turer’s instructions. The rGTPV was cultured in the pre-

sence of mycophenolic acid (25 lg/mL), xanthine (250 lg/

mL), and hypoxanthine (15 lg/mL) (Sigma, St. Louis,

MO, USA). The rGTPV expressing the 9m3C protease or

TPA was screened in LT cells at 37 �C until cytopathic

effect (CPE) appeared. Individual viral plaques were iso-

lated, collected, and further amplified in vitro. The geno-

mic DNA of rGTPVs was extracted by the SDS–proteinase

K–phenol method and was amplified by duplex PCR. The

specific PCR primers used to amplify the 1.25-kbp region

including the upstream homologous sequence, part of the

pCI plasmid, and part of the 9m3C protease–encoding

region were 5’-GGA GGT TTT GAA AAA AAT AAG

GAC CAC AT-3’ (sense) and 5’-TGT AGT GTG CAT

GGA CGG AGA-3’ (antisense). Specific PCR primers for

the 670-bp region including the downstream homologous

sequence, part of the pCI plasmid, and part of the mP1-2A-

encoding region were 5’-TCA ACA ACC TCA ATG TCA

AT-3’ (sense) and 5’-GGG AAG ACA ACG TAC GGA

GA-3’ (antisense). Duplex PCR was performed by 25

cycles of 94 �C for 30 s, 50 �C for 30 s, and 72 �C for

75 s. The positive rGTPVs were propagated and screened

further in LT cells until a specific DNA band of 450 bp was

not amplified out from its template by duplex PCR. The

pure rGTPV expressing the 9m3C protease was designated

as rGTPV-mP1-2A-9m3C, and the pure rGTPV expressing

TPA, rGTPV-mP1-2A-TPA.

Fig. 1 Schematic representation of the GTPV transfer plasmids

pmP1-2A-9m3C and pmP1-2A-TPA. The intergenic region lies

between the gp_24 and gp_24.5 loci of GTPV strain CVCC AV41.

The coding regions for the myristoylated capsid precursor (mP1-2A)

and the nine-amino-acid site-directed mutant of the 3C protease

(9m3C) were derived from the plasmids pP1-2A and p9m3C by PCR.

TPA, Theileria protective antigen; P11-P7.5, canonical promoters

derived from vaccinia virus; D, myristoylation consensus sequence;

arrowheads, transcriptional directions of promoters

Recombinant goatpox virus vaccine for foot-and-mouth disease

123

SDS-PAGE and western blotting

Three days postinfection with a rGTPV, the LT cells were

frozen and thawed twice, vigorously vortexed, resolved by

12 % SDS-PAGE, and transferred onto a polyvinylidene

fluoride (PVDF) membrane (Millipore, Billerica, MA, USA)

using a semi-dry blotter (Bio-Rad, Hercules, CA, USA).

After blocking in 5 % (w/v) nonfat dry milk, the membrane

was incubated with anti-FMDV polyclonal antibody diluted

1:100 in phosphate-buffered saline (PBS), followed by

incubation with the secondary antibody horseradish peroxi-

dase (HRP)–conjugated goat anti-bovine IgG (Sigma) dilu-

ted 1:2500 in PBS. The membrane was then visualized using

hydrogen peroxide and 3,3’-diaminobenzidine tetrahydro-

chloride (DAB) (Sigma). BHK-21 cells infected with FMDV

were used as a positive control, and LT cells infected with the

parental GTPV, a negative control.

Immunization

Twenty-five BALB/c mice (female, 6-8 weeks old) were

acquired from the Experimental Animal Center of Xinjiang

Medicine University and were divided randomly into five

groups of five mice each. The mice were inoculated with

viral preparations by intramuscular injection. Each mouse

in groups 1, 2, and 3 received 0.1 mL (1 9 101.5 TCID50)

of rGTPV-mP1-2A-9m3C, rGTPV-mP1-2A-TPA, and

GTPV strain CVCC AV41, respectively. Group 4 mice

received 0.1 mL of the commercial FMD vaccine, whereas

group 5 mice received 0.1 mL of PBS to serve as mock-

immunized controls. Mice were bled from the caudal vein

for serum sample collection six times at 1-week intervals.

Sera were stored at -20 �C until analyzed.

Twelve sheep of at least 6 months of age were obtained

from areas free from FMD; the sheep had not previously

been vaccinated with FMD or GTPV vaccines and were

free from antibodies to any serotypes of FMDV or GTPV.

The sheep were divided into three groups. The five sheep in

group 1 were each inoculated intramuscularly with 0.1 mL

(104.5 TCID50) of rGTPV-mP1-2A-9m3C; the five sheep in

group 2, 0.1 mL (104.5 TCID50) of rGTPV-mP1-2A-TPA;

and the two sheep in group 3, 0.1 mL of PBS to serve as

mock-immunized controls.

Enzyme-linked immunosorbent assay (ELISA)

Indirect ELISA was performed in accordance with the

protocol described by the World Organization for Animal

Health (OIE) [19]. Briefly, 96-well flat-bottomed plates

(Corning/Costar, Tewksbury, MA, USA) were coated

overnight at 4 �C with 5 lg/mL virions of FMDV Asia-1/

Jiangsu/China/2005 inactivated by treatment with binary

ethyleneimine at 50 lL per well in 0.1 M carbonate-

bicarbonate buffer, pH 9.6. After being blocked with 5 %

bovine serum albumin in PBS, the plates were incubated

with 20-fold dilutions of test sera in duplicate for 1 h at

37 �C. HRP-conjugated anti-mouse IgG (Sigma, USA) was

diluted 1:5000, added to each well, and allowed to incubate

for 1 h at 37 �C. Hydrogen peroxide and 3,3’,5,5’-tetra-

methylbenzidine (TMB) liquid substrate (TIANGEN, Bei-

jing, China) were added to each well to develop color. The

optical density of the ELISA plates was read at a wave-

length of 450 nm.

Transmission electron microscopy

After infection with rGTPVs for 96 h, 5 9 107 LT cells

were harvested by centrifugation at 12,000 rpm, 4 �C for

10 min, washed twice with 1 mL PBS, frozen and thawed

with vigorous agitation, and vortexed three times. After

another centrifugation at 12,000 rpm, 4 �C for 10 min, the

supernatant was collected as a crude VLP preparation, to

which 10 lL of polyclonal serum was added. The mixture

was incubated at 4 �C for 1 h and centrifuged at 4 �C,

12,000 rpm for 30 min, and the resulting precipitate was

resuspended in 100 lL of PBS as a VLP-antibody com-

plex. An aliquot of the complex was layered onto a copper

grid (300–400 mesh) and allowed to bind for 10 min at

room temperature. Upon washing twice with 30 lL TBS

(150 mM NaCl, 100 mM Tris-HCl, pH 7.5), the grid was

incubated in 1 % phosphotungstic acid for 5 min at room

temperature before it was slowly dried. For the analysis and

documentation of VLP samples, a transmission electron

microscope type H-600 equipped with an AMT camera

system (Advanced Microscopy Techniques Corp., Woburn,

MA, USA) was used.

Virus neutralization test (VNT)

VNTs were performed according to the protocol described

by the OIE [19]. Sera were inactivated at 56 �C for 30 min

and were serially diluted twofold starting from a 1 to 2

dilution in a volume of 50 lL per well. All dilutions were

performed in quadruplicate. Fifty microliters of 2 9 103

TCID50/mL FMDV Asia-1/Jiangsu/China/2005 was added

to each well, and controls were set up according the stan-

dard protocol. After incubation at 37 �C for 1 h, 50 lL of

106 cells/mL BHK-21 cells in Eagle’s DMEM containing

10 % FCS was added to each well. The plates were incu-

bated at 5 % CO2 at 37 �C for 3 days and fixed and stained

with naphthalene blue black solution. Endpoint titers were

determined and expressed as the reciprocal of the final

dilution of serum present in the serum-virus mixture where

50 % of the wells were protected [20]. VNTs of the

parental GTPV were also performed according to the

protocol described by the OIE [19].

M. Wenge et al.

123

Results

Construction of transfer plasmids for homologous

recombination

All plasmid constructs were verified by sequencing. The

upstream and downstream homologous sequences of the

transfer plasmids had 99.4 % and 99.5 % identity,

respectively, to the homologous regions of GTPV strain

Pellor. The full-length pmP1-2A-9m3C construct was

approximately 10.6 kbp and was confirmed by restriction

digestion with MluI and NotI to produce fragments of 2.3

and 8.3 kbp, and with MluI and NheI to produce fragments

of 2.1 and 8.5 kbp. The full-length pmP1-2A-TPA con-

struct was approximately 10.7 kb and was confirmed by

restriction digestion with MluI and NotI to produce frag-

ments of 2.3 and 8.4 kbp, and with MluI and NheI to

produce fragments of 2.2 and 8.5 kbp (data not shown).

Screening of recombinant GTPVs

LT cells were cotransfected with a transfer plasmid and the

parental GTPV strain CVCC AV41. Twenty-four viral

plaques were selected until CPE appeared in the presence

of mycophenolic acid, xanthine, and hypoxanthine and

were cultivated in 24-well plates. According to the proce-

dure, the rGTPVs were selected until pure. The purified

viruses were designated as rGTPV-mP1-2A-TPA and

rGTPV-mP1-2A-9m3C, corresponding to their respective

plasmid origins of pmP1-2A-TPA and pmP1-2A-9m3C,

and were used as stock viruses. The viruses were confirmed

to contain the integral expression cassette by PCR and were

used to express their encoded heterologous proteins in LT

cells. The genomic DNAs of the stock viruses were

extracted for use as PCR templates to detect the correct

insertion of the heterologous genes between loci gp_24 and

gp_24.5. Analysis showed that two specific DNA bands of

670 bp were amplified from both templates and that no

such band was present in the DNA from the parental

GTPV. Further analysis revealed that a specific PCR

product of 1.25 kbp was amplified from the template of

rGTPV-mP1-2A-9m3C, whereas no such product was

obtained from the parental GTPV and rGTPV-mP1-2A-

TPA. In contrast, a specific DNA band of 450 bp was

amplified only from the template of the parental GTPV

(data not shown).

Analysis of mP1-2A expression and processing

by 9m3C protease

Western blotting showed that P1, VP0, VP3, and VP1

capsid proteins expressed in LT cells infected with rGTPV-

mP1-2A-9m3C could be detected by polyclonal FMDV

sera. This is consistent with the profile for FMDV-infected

cells. In contrast, there were no specific bands corre-

sponding to VP0, VP3, and VP1 in the rGTPV-mP1-2A-

TPA–infected sample. No specific bands were detected in

wild-type GTPV (Fig. 2). As observed by electron

microscopy, only the mP1-2A cleaved by the 9m3C pro-

tease was able to form VLPs, the diameter of which was

about 25 ± 2 nm. In the same field, a GTPV particle was

captured as a reference, its diameter being about 160 nm

(Fig. 3).

Antigenic assay of immunized mice

Sera collected from mice in each inoculation group were

pooled for ELISA and VNT analyses. The ELISA results

Fig. 2 SDS-PAGE and western blot analysis of expressed FMDV

capsid proteins from recombinant GTPVs. Lane 1, LT cells infected

with 20th-passage rGTPV-mP1-2A-9m3C; lane 2, inactivated FMDVs

propagated in BHK-21 cells; lane 3, prestained protein ladder

(Fermentas, Vilnius, Lithuania; now part of Thermo Fisher Scientific,

Pittsburgh, PA, USA); lane 4, rGTPV-mP1-2A-TPA; lane 5, LT cells

infected with 10th-passage rGTPV-mP1-2A-9m3C; lane 6, GTPV

strain CVCC AV41

Fig. 3 Transmission electron microscopy observation of VLPs

expressed in LT cells infected with rGTPV-mP1-2A-9m3C. The

diameter of these VLPs is about 25 ± 2 nm. In the same field, a

GTPV particle, with a diameter of about 160 nm, is captured as a

reference. Bar, 100 nm

Recombinant goatpox virus vaccine for foot-and-mouth disease

123

showed that there was no significant difference between the

antibody titers of the five groups in the first week po-

stimmunization (wpi), and the titers were slightly increased

by 2 wpi. The antibody titers of mice inoculated with

rGTPV-mP1-2A-9m3C or with the commercial FMDV

vaccine were significantly higher at 3 wpi than those of the

mice inoculated with rGTPV-mP1-2A-TPA or with the

parental GTPV (Fig. 4).

Neutralizing antibody titers in mice were determined by

the VNT assay, and the result showed that the group

inoculated with rGTPV-mP1-2A-TPA developed hardly

detectable neutralizing antibody titers, whereas the group

inoculated with the parental GTPV did not produce any

detectable titer. The neutralizing antibody titers for the

group inoculated with rGTPV-mP1-2A-9m3C increased

from 1:4 at 2 wpi to a maximum of 1:16 at 3 and 4 wpi

(Fig. 4). The neutralizing antibody titers for the group

inoculated with the commercial FMDV vaccine were 1:4,

1:32, and 1:64 at the corresponding time points (Fig. 5).

These results suggest that an antibody response was gen-

erated to the FMDV protein delivered by the rGTPV

vectors.

Immunogenicity of mP1-2A cleaved by 9m3C protease

in sheep

While the two non-immunized sheep did not develop

neutralizing antibodies and sheep inoculated with rGTPV-

mP1-2A-TPA produced neutralizing antibody against only

GTPV, all five sheep inoculated with rGTPV-mP1-2A-

9m3C developed either high-level neutralizing antibodies

against GTPV or specific neutralizing antibodies against

FMDV. The maximum titer produced was 1:32 against

FMDV. However, the titers of antibody against FMDV at

1, 3, 5, 7, and 9 weeks postvaccination (wpv) followed a

trend of, in succession, an initial rapid increase, a shorter

stationary phase, and a sharp decline, and the persistence

period of the neutralizing antibody was about 2 months

(Table 1). The lower titer of neutralizing antibody against

FMDV and its shorter duration did not meet the require-

ments of the World Organization for Animal Health (OIE).

Discussion

It is generally accepted that protective immunity against

FMDV is principally due to the production of neutralizing

antibodies. In this regard, recombinant viruses possess

significant advantages over conventional inactivated virus

vaccines or peptides. It has been shown that the empty

capsid particle of FMDV is a potent immunogen [4, 11].

However, previous attempts to isolate recombinant vac-

cinia viruses constitutively expressing the coding region of

mP1-2A together with the 3C protease of FMDV were

unsuccessful because of the induction of apoptosis in the

host cells by the protease [3]. These reported results could

reflect the toxicity of the 3C protease and the resulting

lower yields of the capsid precursor, and not the immu-

nogenic potential of the empty capsid [13]. We also tried to

construct a rGTPV that coexpresses mP1-2A and the native

3C protease, but the expression failed (data not shown). To

avoid the toxicity of the 3C protease, we employed the

GTPV strain CVCC AV41 to express mP1-2A together

with the 9m3C protease of FMDV. Fortunately, stable

recombinant GTPV expressing the cDNA cassettes of

mP1-2A and the 9m3C protease was isolated successfully

in vitro. Our findings suggest that the catalytic activity of

the 9m3C protease is decreased relative to that of 3C, and

consequently the cells were not induced to undergo apop-

tosis excessively and were able to support the complete life

cycle of rGTPV-mP1-2A-9m3C. However, the apoptotic

effect may occur at a later point in time.

Intergenic regions are accepted as suitable sites for the

insertion of heterologous genes in poxviruses. However,

few intergenic regions have been determined to be stable

insertion sites. In this study, an intergenic region located

Fig. 4 ELISA analysis of FMDV-specific antibody titers of mice

inoculated with rGTPV-mP1-2A-9m3C, rGTPV-mP1-2A-TPA, and

parental GTPV strain CVCC AV41. Sera from all mice were assayed

at a dilution of 1:20. Titers are represented as the average of five sera

in each group

Fig. 5 Virus neutralization titers in mice inoculated with rGTPV-

mP1-2A-9m3C, rGTPV-mP1-2A-TPA, and parental GTPV strain

CVCC AV41. Sera from all mice were serially diluted twofold. The

values represent the average of five sera in each group

M. Wenge et al.

123

between the homologous loci gp_24 and gp_24.5 of the

GTPV genome proved to be an ideal site for the insertion

of foreign genes. In the present study, we were able to

express the FMDV mP1-2A and 9m3C protease genes. In

rGTPV-mP1-2A-9m3C, the coding sequences of mP1-2A

and 9m3C are under the control of the canonical P11-P7.5

promoters. With this combination of promoters, heterolo-

gous gene expression can be initiated effectively and can

occur continually at a relatively high level in host cells.

Indeed, the immune response elicited by the recombinant

viruses demonstrated the efficacy of the combined pro-

moters, whether for expression analysis or animal trials.

In this study, the antibody titers elicited by the recom-

binant virus expressing the 9m3C protease were slightly

lower than those for the inactivated whole-virus vaccine. In

contrast, the recombinant virus expressing the uncleaved

capsid precursor did not elicit a comparable response

(Fig. 4). This suggests that the processed capsid precursor

expressed by rGTPV is a promising immunogen.

Although rGTPV-mP1-2A-9m3C was able to express

FMDV mP1-2A and 9m3C protease proteins from the

novel intergenic region, and animal trials indicated that the

cleaved mP1-2A induced significantly higher levels of

neutralizing antibody against FMDV than the uncleaved

mP1-2A, the titers of neutralizing antibody did not meet

the standard required by the OIE. It could be that rGTPV-

mP1-2A-9m3C could complete its life cycle to produce

more heterologous proteins, but perhaps because the 9m3C

protease caused subtle induction of apoptosis in the host

cells, the yields of VLPs were not high enough to stimulate

the animals to produce fully protective antibody titers.

According to OIE’s requirements for FMDV vaccines, the

titer of neutralizing antibody stimulated by a vaccine needs

to be at least 1:64 for the vaccine to be permitted for use. It

is also possible that some of the sheep had been infected by

GTPV naturally before they were inoculated with rGTPV-

mP1-2A-9m3C, but their specific antibodies were not

detectable. The preexisting cellular immunity could have

inhibited the diffusion and propagation of the recombinant

virus in the host, thus preventing the recombinant virus

from stimulating those sheep to produce high enough

protective antibody titers [21]. These results indicate that

further modifications of the recombinant virus vector used

in this study are necessary to produce a more effective

alternative to the current FMDV vaccine.

Acknowledgments This study was supported by the National Nat-

ural Science Foundation of China (ID: 30760181, 31060346) and

Special Fund for Agro-Scientific Research in the Public Interest (ID:

201103008).

References

1. Grubman MJ, Baxt B (2004) Foot-and-mouth disease. Clin

Microbiol Rev 17:465–493

2. Taboga O, Tami C, Carrillo E, Nunez JI, Rodrıguez A, Saız JC,

Blanco E et al (1997) A large-scale evaluation of peptide vac-

cines against foot-and-mouth disease: lack of solid protection in

cattle and isolation of escape mutants. J Virol 71:2606–2614

3. Abrams CC, King AM, Belsham GJ (1995) Assembly of foot-

and-mouth disease virus empty capsids synthesized by a vaccinia

virus expression system. J Gen Virol 76:3089–3098

4. Claudine P, Abhay K, Alison B, Terry J, Jingshan R, Silvia L, Ian

MJ, Elizabeth EF, David IS, Bryan C (2013) Rational engineering

Table 1 Neutralizing antibody titers against GTPV and FMDV in sheep at 0, 1, 3, 5, 7, and 9 weeks postvaccination

Group No. Sheep No. Neutralizing antibody titers at

0 wpv 1 wpv 3 wpv 5 wpv 7 wpv 9 wpv

G/F G/F G/F G/F G/F G/F

1 420717 -/- 1:22/1:22 1:25/1:23 1:25/1:22 1:25/1:22 1:25/-

1 420725 -/- 1:22/1:22 1:25/1:25 1:26/1:24 1:26/1:24 1:26/1:23

1 420714 -/- 1:23/1:22 1:25/1:24 1:25/1:23 1:25/1:22 1:25/-

1 420732 -/- 1:23/1:22 1:25/1:22 1:26/1:22 1:26/1:22 1:25/-

1 420739 -/- 1:22/1:22 1:25/1:23 1:26/1:23 1:26/1:22 1:25/1:22

2 420721 -/- 1:23/- 1:25/- 1:25/- 1:26/- 1:26/-

2 420736 -/- 1:23/- 1:26/- 1:26/- 1:25/- 1:24/-

2 420718 -/- 1:22/- 1:25/- 1:25/- 1:25/- 1:25/-

2 420738 -/- 1:22/- 1:25/- 1:26/- 1:25/- 1:25/-

2 420723 -/- 1:23/- 1:25/- 1:26/- 1:26/- 1:26/-

3 420730 -/- -/- -/- -/- -/- -/-

3 420715 -/- -/- -/- -/- -/- -/-

Antibody titers are expressed as dilution ratios

Titers below 1:4 are designated by ‘-’

Wpv week(s) postvaccination, G/F neutralizing antibody titer against GTPV / neutralizing antibody titer against FMDV

Recombinant goatpox virus vaccine for foot-and-mouth disease

123

of recombinant picornavirus capsid to produce safe, protective

antigen. PLoS Pathog 9:e1003255

5. Berhe G, Minet C, Le Goff C, Barrett T, Ngangnou A, Grillet C,

Libeau G, Fleming M, Black DN, Dialloet A (2003) Develop-

ment of a dual recombinant vaccine to protect small ruminants

against peste-des-petits-ruminants virus and capripoxvirus

infections. J Virol 77:1571–1577

6. Pastoreta PP, Vanderplasschen A (2003) Poxviruses as vaccine

vectors. Comp Immunol Microbiol Infect Dis 26:343–355

7. Tulman ER, Afonso CL, Lu Z, Zsak L, Kutish GF, Rock DL

(2001) Genome of lumpy skin disease virus. J Virol

75:7122–7130

8. Ma WG, Wang F, Yu L, Xue Y, Wang Y, Wei J, Huang J, Fu ZH

(2010) Identification of an intergenic region that is not essential

for replication of goatpox virus. Arch Virol 155:1337–1341

9. Fry EE, Stuart DI, Rowlands DJ (2005) The structure of foot-and-

mouth disease virus. Curr Top Microbiol Immunol 288:71–101

10. Stewart G, Tobias JT, Armando A, Richard AK, David JR (2009)

Foot-and-mouth disease virus assembly: processing of recombi-

nant capsid precursor by exogenous protease induces self-

assembly of pentamers in vitro in amyristoylation-dependent

manner. J Virol 83:11275–11282

11. Lewis SA, Morgan DO, Grubman MJ (1991) Expression, pro-

cessing of foot-and-mouth disease virus capsid structure in het-

erologous system: induction of a neutralizing antibody response

in guinea pig. J Virol 65:6572–6580

12. Ansardi DC, Porter DC, Morrow CD (1991) Coinfection with

recombinant vaccinia viruses expressing poliovirus P1 and P3

proteins results in polyprotein processing and formation of empty

capsid structures. J Virol 65:2088–2092

13. Birtley BJ, Knox SR, Jaulent AM, Brick P, Leatherbarrow RJ,

Curry S (2005) Crystal structure of foot-and-mouth disease virus

3C protease: new insights into catalytic mechanism and cleavage

specificity. J Biol Chem 280:11520–11527

14. Falkner FG, Moss B (1988) Escherichia coli gpt gene provides

dominant selection for vaccinia virus open reading frame

expression vectors. J Virol 62:1849–1854

15. Boyle DB (1992) Quantitative assessment of poxvirus promoters in

fowlpox and vaccinia virus recombinants. Virus Genes 6:281–290

16. Benkert P, Biasini M, Schwede T (2011) Toward the estimation

of the absolute quality of individual protein structure models.

Bioinformatics 27:343–350

17. Chow M, Newman JF, Filman D, Hogle JM, Rowlands DJ,

Brown F (1987) Myristylation of picornavirus capsid protein VP4

and its structural significance. Nature 327:482–486

18. Ansardi DC, Porter DC, Morrow CD (1992) Myristylation of

poliovirus capsid precursor P1 is required for assembly of sub-

viral particles. J Virol 66:4556–4563

19. OIE Reference Laboratories, Manual of Diagnostic Tests and

Vaccines for Terrestrial Animals (2013) (Version adopted May

2012, chapter2.1.5[FMDV], chapter2.7.14[GTPV]). http://www.

oie.int/en/international-standard-setting/terrestrial-manual/access-

online/

20. Finney DJ (1978) Statistical methods in biological assay. In:

English, 3rd edn. Macmillan Publishing Co., Inc, New York,

pp 394–401

21. Belyakov IM, Moss B, Strober W, Berzofsky JA (1991) Mucosal

vaccination overcomes the barrier to recombinant vaccinia

immunization caused by pre existing poxvirus immunity. Proc

Natl Acad Sci USA 96:4512–4517

M. Wenge et al.

123