Journal of General Virology

Induction of a protective response in swine vaccinated with DNA encoding foot-and-mouth disease virus empty capsid proteins and the 3D RNA polymerase

Leticia Cedillo-Barrón,¹† Mildred Foster-Cuevas,¹‡ Graham J. Belsham,¹ François Lèfevre² and R. Michael E. Parkhouse¹§

¹ Institute for Animal Health, Pirbright Laboratory, Pirbright, Surrey GU24 0NF, UK
² INRA Virology et Immunologie moléculaires, INRA, 78350 Jouy-en-josas, France.

This work focuses on the development of a potential recombinant DNA vaccine against foot-and-mouth disease virus (FMDV). Such a vaccine would have significant advantages over the conventional inactivated virus vaccine, in particular having none of the risks associated with the high security requirements for working with live virus. The principal aim of this strategy was to stimulate an antibody response to native, neutralizing epitopes of empty FMDV capsids generated in vivo. Thus, a plasmid (pcDNA3.1/P1–2A3C3D) was constructed containing FMDV cDNA sequences encoding the viral structural protein precursor P1–2A and the non-structural proteins 3C and 3D. The 3C protein was included to ensure cleavage of the P1–2A precursor to VP0, VP1 and VP3, the components of self-assembling empty capsids. The non-structural protein 3D was also included in the construct in order to provide additional stimulation of CD4⁺ T cells. When swine were immunized with this plasmid, antibodies to FMDV and the 3D polymerase were synthesized. Furthermore, neutralizing antibodies were detected and, after three sequential vaccinations with DNA, some of the animals were protected against challenge with live virus. Additional experiments suggested that the antibody response to FMDV proteins was improved by the co-administration of a plasmid encoding porcine granulocyte–macrophage colony-stimulating factor. Although still not as effective as the conventional virus vaccine, the results encourage further work towards the development of a DNA vaccine against FMDV.

Foot-and-mouth disease virus (FMDV) causes a highly contagious vesicular disease in cloven-hoofed animals, as illustrated dramatically by the recent outbreak in the UK. The virus is a member of the genus Aphthovirus of the family Picornaviridae, and includes seven serotypes. The genome is a positive-sense, single-stranded RNA of 8.5 kb with one large open reading frame. Its translation yields a polyprotein that is subsequently processed by virus-encoded proteases to produce the structural and non-structural proteins necessary for virus assembly and replication. One of the initial polyprotein cleavages, mediated by the 2A protein, is a co-translational cleavage at the N terminus of the 2B protein. The P1–2A precursor is processed by the viral protease 3C^pro to produce the structural proteins VP0, VP3 and VP1. These proteins can then self-assemble to form icosahedral empty capsid particles (Ryan et al., 1989; Belsham et al., 1990; Abrams et al., 1995), which consist of 60 copies of each protein. Encapsidation of viral RNA to produce mature virions is accompanied by the cleavage of VP0 to VP2 and VP4.

Currently, control of the disease includes slaughter of infected and exposed animals and vaccination with a chemically inactivated virus vaccine. Such conventional vaccines, although generally effective, do have some disadvantages. These include thermal instability, the short-term nature of protection and extra cost due to the high-security containment necessary for their preparation. A particularly serious problem is that several outbreaks in Europe have been attributed to incomplete inactivation of the virus or to the escape of live virus from vaccine production laboratories. Thus, the development of recombinant FMDV vaccines, including peptide (DiMarchi et al., 1986; Geysen et al., 1984; Francis et al., 1987) and DNA vaccines (Ward et al., 1997; Chinsangaram et al., 1998), as potentially safe alternatives has been explored. Vaccination with DNA encoding protective antigens offers the potential to manipulate the induced immune response through the simultaneous delivery of adjuvants, cytokines and other proteins, enhancing the vaccination efficacy (Kim et al., 1997; Somasundaram et al., 1999; Tuteja et al., 2000). In this study, we have taken advantage of the fact that natural empty particles, lacking nucleic acid and containing VP1, VP3 and VP0 (the uncleaved precursor form), are able to induce an antibody response similar to that induced by infectious FMDV particles in guinea pigs (Rowlands et al., 1975). Significantly, these empty capsids induce neutralizing antibody of the same specificity as the whole virus. This is presumably because these particles resemble natural virus particles antigenically including, importantly, the major target of neutralizing antibodies, the G–H loop of VP1 (Rweyemamu et al., 1979; Grubman et al., 1985; Geysen et al., 1984; Francis et al., 1985). In principle, therefore, empty capsids appear to be a good candidate for the development of a recombinant vaccine.

Although protection against FMDV generally correlates with neutralizing antibodies (van Bekkum, 1969), a T cell response is quite clearly necessary for effective immunity. While it is true that FMDV structural proteins do stimulate T cell responses (Garcia-Valcarcel et al., 1996), we have shown recently that the non-structural proteins 3D and 2B are potent stimulators of cellular and humoral specific immune responses in the natural host (Foster-Cuevas, 1996; Foster et al., 1998; Collen et al., 1998). Furthermore, these non-structural proteins also induce high proliferative T cell and delayed-type hypersensitivity (DTH) responses when administered as DNA vaccines (L. Cedillo-Barrón and M. Foster-Cuevas, unpublished data).

Taking this into account, we have constructed a plasmid (pcDNA3.1/P1–2A3C3D) that contains all of the structural and non-structural proteins necessary for the formation of FMDV empty capsids, i.e. the P1 sequence plus the 2A^pro and 3C^pro proteases necessary for processing of the P1 polyprotein into VP1, VP3 and VP0. In addition, the 3D RNA polymerase was included as a potent source of T cell epitopes. This construct, when tested in pigs, induced neutralizing antibody and protection. Finally, we have shown that co-administration of a plasmid encoding porcine granulocyte–macrophage colony-stimulating factor (GM-CSF) can improve the antibody responses to the virus significantly.

Cells and virus. COS-7 cells (ATCC-CRL 1651) were grown in DMEM supplemented with 10 % heat-inactivated foetal calf serum, L-glutamine (0.29 mg/ml), penicillin (200 U/ml) and streptomycin (0.2 mg/ml).

FMDV O₁Kaufbeuren (O₁K) was prepared by infecting confluent monolayers of baby hamster kidney (BHK) cells. The infected cultures were incubated on a rocking platform until cell lysis was observed (18–24 h) and the lysate was then clarified by centrifugation (3000 r.p.m., 20 min).

Construction of recombinant plasmid. The P1–2A3C3D cassette, which contains FMDV serotype O₁K sequences, includes the complete P1–2A, 3C and 3D coding sequences with the addition of an ATG initiation codon. It was subcloned into pcDNA3.1 downstream of cytomegalovirus (CMV) and T7 promoters. Plasmid pCAK (Belsham et al., 1990) carries the P1–2A region from FMDV O₁K. The P1–2A sequence was excised by using EcoRI and XhoI endonucleases. At the same time, the 3C and 3D coding sequences were excised from pT7S3 (which contains the full-length infectious cDNA of FMDV O₁K; Ellard et al. 1999) by digestion with HpaI, blunt-ended and then digested with XhoI. Both fragments were then ligated into the pcDNA3.1 vector that had been digested with ApaI, blunt-ended and then cut with EcoRI. The resultant plasmid was called pcDNA3.1/P1–2A3C3D.

Expression of transfected recombinant plasmid. Expression of the FMDV cassette in pcDNA3/P1–2A3C3D driven by the CMV promoter was verified by transfection of COS-7 cells (in 35 mm wells) using Fu-GENE 6 reagent (Boehringer Mannheim). Two days after transfection, cells were analysed for expression of FMDV proteins by labelling with [³⁵S]methionine/[³⁵S]cysteine and immunoprecipitation and the result was confirmed by indirect immunofluorescence. For immunofluorescence, monolayers of cells were grown on cover slips and, after transfection with pcDNA3.1/P1–2A3C3D, were fixed in 4 % paraformaldehyde in PBS (1 h at room temperature). Samples were then blocked with 10 % normal goat serum (NGS) (1 h, room temperature) and then stained with MAb 2B against VP1 (provided by E. Brocchi, Brescia, Italy) followed by fluorescein-conjugated goat anti-mouse serum. For [³⁵S]methionine cell labelling, the cells were washed 48 h after transfection and labelled with 0.1 mCi [³⁵S]Promix (Amersham) (methionine and cysteine) in met/cys-free supplemented medium and then lysed with RIPA buffer (100 mM Tris–HCl, pH 8.3–8.6, 2 % Triton X-100, 150 mM NaCl, 0.6 M KCl, 5 mM EDTA, 1 % aprotinin, 3 mM PMSF, 1 µg/ml leupeptin and 5 µg/ml soybean trypsin inhibitor). For immunoprecipitation reactions, the lysate was pre-cleared with fixed cells of Staphylococcus aureus Cowan strain (1 h at 4 °C). After centrifugation, the labelled FMDV proteins were immunoprecipitated from the lysate with a hyperimmune guinea pig anti-FMDV antibody immobilized on Protein A–Sepharose beads (Pharmacia). The bound radioactive proteins were then detected by SDS–PAGE and autoradiography. Expression driven by the T7 promoter was analysed in COS-7 cells infected with the recombinant vaccinia virus vTF7-3 (Fuerst et al., 1986) and then transfected with pcDNA3.1/P1–2A3C3D. After 24 h, the cells were labelled with [³⁵S]met/cys and cell extracts were prepared and immunoprecipitated as described above.

Animals. Thirty-two Large White cross-bred Landrace pigs weighing 20–25 kg were used for the study. All animal infectivity experiments were performed within a high-containment isolation compound.

Vaccines and vaccination. Inactivated FMDV vaccine was prepared from type O₁ Lausanne FMDV antigen concentrate kept in the International Vaccine Bank at Pirbright. Vaccine was prepared as a water-in-oil-in-water emulsion using Montanide ISA 206 (Seppic) and contained 2.4 µg 146S antigen per 2 ml dose. Pigs were given a single dose by the intramuscular route, immediately caudal to the ear (Salt et al., 1998).

For DNA vaccination, recombinant plasmids were purified from Escherichia coli JM109 using the Qiagen plasmid purification Giga kit. DNA was diluted to the required concentration in endotoxin-free Dulbecco's PBS (DPBS, Sigma). Pigs were injected three times at 3–4 week intervals with 300 µg recombinant plasmid pcDNA3.1/P1–2A3C3D and, when specified, 200 µg porcine pcDNA3/GM-CSF diluted in 2.5 ml DPBS. Aliquots (1 ml) were injected, by using a 27 gauge needle, into the tibialis cranialis muscle of both legs and 0.25 ml aliquots were administered intradermally into the dorsal surface of the ears. Each injection was preceded by intramuscular administration of 10 mM carditoxin (Latoxan) in 0.5 ml endotoxin-free DPBS.

The experiments consisted of four groups, each containing four pigs: group 1, conventional FMDV vaccine; group 2, plasmid pcDNA3/GM-CSF; group 3, plasmid pcDNA3.1/P1–2A3C3D; group 4, plasmids pcDNA3.1/P1–2A3C3D and pcDNA3/GM-CSF. Serum samples were taken at regular intervals and assayed for antibodies against intact FMDV and FMDV 3D polymerase by ELISA and for neutralizing antibody as described previously (Golding et al., 1976).

Challenge of animals. Two experiments were performed to determine the ability of the plasmid pcDNA3.1/P1–2A3C3D to induce an immune response against FMDV proteins. Vaccination and experimental infection of pigs were performed in an identical manner in each experiment, as described above. Between 15 and 18 days after the third immunization with DNA, all of the pigs were challenged with live virus. Challenge virus was prepared by serial passage of FMDV O₁ Lausanne in pigs. Virulent FMDV was recovered from lesions and titrated on bovine thyroid (BTY) cells. Test animals were injected in the heel bulb with 10^5.5 TCID₅₀ in 0.1 ml M25 saline. The FMDV O₁ Lausanne and O₁K isolates are quite closely related in terms of antigenic properties and P1 nucleotide sequence, and therefore efficient protection against the virus challenge occurs. After the virus challenge in experiment A, the animals from different groups were isolated in different rooms, whereas, in the second experiment (B), all groups were housed in the same room in order to expose the vaccinated animals to high levels of virus-containing aerosol from the infected, unvaccinated animals. Animals were examined daily and protection was scored by time of appearance and severity of lesions. Total protection was defined as complete absence of lesions. Animals were classified as partially protected against FMDV if the development of FMDV lesions was delayed and lesions were restricted in distribution. As animals were inoculated with FMDV in one rear foot, the development of a single foot lesion was considered to be a significant level of protection. The four groups included control animal groups vaccinated with conventional inactivated FMDV vaccine and unprotected animals similarly immunized with an unrelated plasmid (pcDNA3/GM-CSF).

Detection of anti-3D antibodies. Detection of serum antibodies to the FMDV 3D protein was performed by ELISA using polyvinyl plates (Nunc) coated with purified recombinant GST–3D (2.5 µg/ml) in 0.1 M carbonate/bicarbonate buffer, pH 9.6, overnight at 4 °C. After blocking with 0.05 % Tween 20/3 % skimmed milk/1 % NGS in PBS, plates were incubated with duplicate double serial dilutions of test sera for 1 h at 37 °C. Goat anti-porcine Ig peroxidase conjugate (Sigma) at 1:5000 dilution was then added for 1 h, followed by the substrate (0.01 % hydrogen peroxide in phosphate/citrate buffer). Absorbances were determined at 492 nm. Recombinant GST–3D fusion protein was purified from bacterial inclusion bodies by preparative SDS–PAGE (10 %) as described previously by Katrak et al. (1992).

Detection of antibodies to FMDV. Ninety-six well flat-bottomed plates (Nunc Immunoplate II) were coated with rabbit anti-O₁ Lausanne FMDV antiserum in 0.1 M carbonate/bicarbonate buffer, pH 9.6, overnight at 4 °C. Plates were washed three times and 50 µl clarified tissue culture fluid containing FMDV was added to each well. The plates were incubated for 60 min at 37 °C and then serial dilutions of test sera followed by peroxidase-labelled goat-anti-porcine IgG (Sigma) and substrate were applied sequentially as described above.

Detection of neutralizing antibodies. Sera taken from all vaccinated pigs on day 0, before immunization, and on day 63, after the last DNA injection but before virus challenge, were analysed for neutralizing antibody titres by using a micro-neutralization assay with monolayers of primary BTY cells. Double dilutions of antibody were reacted with 0.5 log₁₀ dilution of virus for 1 h. Cells were then added as indicators of residual infectivity and microplates were incubated at 37 °C for 3 days prior to fixing and staining (Golding et al., 1976). The end-point titres were calculated as the reciprocal of the last serum dilution to neutralize 100 TCID₅₀ homologous FMDV in 50 % of the wells.

Delayed-type hypersensitivity (DTH) reaction. T cell stimulation in vivo was analysed by the DTH test (Foster et al., 1998; Bautista & Molitor, 1997). Pigs were injected intradermally in the ventral abdomen with 7.5 µg purified inactivated O₁ FMDV in 0.1 ml PBS or 20 µg purified GST–3D RNA polymerase in 0.1 ml PBS. Purified recombinant GST and PBS were injected as negative controls. Skin induration was measured 48 h after antigen injection.

Expression of pcDNA3.1/P1–2A3C3D plasmid

In order to demonstrate appropriate expression of the FMDV capsid precursor and processing to the three capsid proteins 1AB (VP0), 1C (VP1) and 1D (VP3), transfected COS-7 cells were analysed by biosynthetic labelling, co-precipitation and SDS–PAGE (Fig. 1 a) and by immunofluorescence using monoclonal anti-VP1 antibody (Fig. 1 b). As seen in Fig. 1(a) (lanes 2 and 3), the predicted components of FMDV empty capsids were synthesized in transfected cells. Furthermore, immunofluorescence demonstrated expression of VP1 protein (Fig. 1 b). While no formal proof of empty capsid formation was obtained, previous work (Abrams et al., 1995) suggests that this may indeed have occurred. As a negative control, transfection of the same cells with the parental plasmid vector yielded negative results (Fig. 1 a; lane 4). Similar negative results were obtained with cells infected with vTF7-3 alone (Fig. 1 a; lane 1).

Fig. 1. Expression of pcDNA3.1/P1–2A3C3D. (a) Lanes: 1, COS-7 cells infected with vaccinia virus vTF7-3 (negative control); 2, cells transfected with pcDNA3.1/P1–2A3C3D and co-infected with vTF7-3; 3, uninfected cells transfected with pcDNA3.1/P1–2A3C3D; 4, cells transfected with pcDNA3.1 (negative control). The cells were labelled for 24 h (as described in Methods) with [³⁵S]Promix and cell extracts were immunoprecipitated with guinea pig anti-FMDV antiserum. (b) Uninfected COS-7 cells were transfected with pcDNA3.1/P1–2A3C3D and stained with anti-VP1 MAb B2, as described in Methods.

Antibody response to FMDV and the RNA polymerase 3D after DNA vaccination

In the first experiment, with pig groups housed in separate rooms (experiment A), the group inoculated with pcDNA3.1/P1–2A3C3D alone developed low levels of antibodies to FMDV (Fig. 2). However, two of four animals co-immunized with porcine pcDNA3/GM-CSF showed a response to FMDV after the second injection that was comparable to that observed in some of the animals in the group vaccinated with conventional FMDV vaccine. No antibodies to FMDV were detected in the negative-control group (Fig. 2). In the same experiment, antibodies to 3D were detected in only one of four animals immunized with pcDNA3.1/P1–2A3C3D alone (Fig. 3). This response was enhanced by co-injection of porcine pcDNA/GM-CSF, with all four of the animals having detectable antibody, albeit with wide variations. Neither the vaccinated nor the negative-control group showed a significant response to 3D (Fig. 3).

In the second experiment, with all of the animal groups being housed together (experiment B), three of four animals inoculated with pcDNA3.1/P1–2A3C3D developed a good antibody response to FMDV (Fig. 4). This response was not increased by co-injection of porcine pcDNA3/GM-CSF. Similarly, in pigs immunized with pcDNA3.1/P1–2A3C3D, three of the four animals synthesized antibodies to 3D and, once again, the response was increased only slightly in the group co-injected with porcine pcDNA3/GM-CSF (Fig. 5). In the negative control group injected with pcDNA3/GM-CSF alone, one animal inexplicably had antibodies to the whole virus, but the rest of the group were negative, as expected.

Fig. 2. Specific antibody response to FMDV after DNA vaccination. Results are presented from experiment A, in which animal groups were housed separately. This experiment employed a course of three injections of plasmid at intervals of 3–4 weeks (indicated by arrows). The challenge virus was given 2 weeks after the last injection and the different animal groups were subsequently housed in separate boxes. Sera from immunized animals were sampled regularly and tested for antibodies at 1:80 dilution. Specific antibody response to whole FMDV was measured by capture ELISA (see Methods). The results are presented for individual animals as A₄₉₂. A group of pigs vaccinated with conventional vaccine was used as positive controls and animals vaccinated with pcDNA3/GM-CSF alone were used as negative controls.

Induction of neutralizing antibodies before and after DNA vaccination

Neutralizing antibody determinations were performed with serum from day 0 (prior to vaccination) and 2–3 weeks after the last immunization (day 64) (Table 1). In experiment A, with the animal groups housed separately, two of four pigs immunized with pcDNA3.1/P1–2A3C3D alone developed significant titres of neutralizing antibody. One of these reached titres that were higher than any animal immunized with the conventional vaccine. The four animals co-injected with the plasmid encoding porcine GM-CSF showed high titres of neutralizing antibodies, similar to the animals vaccinated with plasmid alone and the conventional inactivated FMDV vaccine. The negative-control animals immunized with porcine pcDNA3/GM-CSF did not develop neutralizing antibodies to FMDV and were susceptible to subsequent virus challenge (Table 1). In experiment B, with all of the animals housed together, those immunized with pcDNA3.1/P1–2A3C3D alone developed titres of neutralizing antibodies similar to those observed in the group of animals vaccinated with inactivated FMDV vaccine. Co-injection with pcDNA3/GM-CSF did not change the level of neutralizing antibodies significantly. In this experiment, serum from one pig in the group immunized with pcDNA/GM-CSF contained antibodies to whole virus, as detected by capture ELISA, and was also positive for virus-neutralizing antibodies. However, this animal was not protected against subsequent challenge with live virus (Table 1). There was no obvious correlation between levels of neutralizing antibody and anti-virus antibodies detected by the capture ELISA.

Table 1. Titres of neutralizing antibodies and protection results

Neutralizing antibody was measured in the two groups (A and B) of animals, as described in Methods and the legends to Figs 2 and 4. Serum samples were taken on day 0 and day 64 before the challenge with live FMDV. Titres are expressed as log₁₀ TCID₅₀/ml, measured on BTY cells and calculated by the Spearman–Karber method. Severity of symptoms was based on daily monitoring, completed on day 10 post-challenge, and is scored as: none, no lesions and no temperature or 1 day with temperature above 40 °C and no lesions; mild, 1 day with temperature above 40 °C, lesions on one foot or small lesions in the snout; severe, lesions on all four feet, gums and snout.

* This animal presented a small lesion (2 mm) at the site of virus injection.
† This animal presented a small lesion (2 mm) in the snout by day 8 post-challenge.

T cell responses in vivo measured by DTH skin test

This test was performed after DNA vaccination and before the live virus challenge. Similar results were obtained in both experiments A and B (animals housed separately or together). In each experiment, one pig in the group vaccinated with pcDNA3.1/P1–2A3C3D alone was reactive to whole FMDV and another to the RNA polymerase (Table 2). Similarly, in the group co-vaccinated with the porcine pcDNA3/GM-CSF plasmid, only one pig was reactive. In experiment B, with all the animals housed together, two animals responded to the 3D polymerase and none to the whole virus. None of the animals in the negative control group of experiment B showed a DTH response to any of the antigens tested. All of the animals in the FMDV-vaccinated group of experiment B showed reactivity against the virus but not against the 3D polymerase (Table 2).

Fig. 3. Specific antibody response to 3D RNA polymerase after DNA vaccination. Results are presented from experiment A, in which animal groups were housed separately. This experiment was performed as described in Fig. 2 and with the same positive and negative controls. The plasmid injections are indicated by arrows. Specific antibody response to GST–3D was measured by indirect ELISA and the results are presented for individual animals as A₄₉₂.

Challenge results after DNA vaccination

In the two experimental designs (A and B), the positive (FMDV-vaccinated) and negative (pcDNA3/GM-CSF) controls followed expectations, yielding protection or disease, respectively (Table 1). The negative-control animals that were not protected developed fever, lameness and vesicles in all four feet and in the snout by day 4. In experiment A, in which the different groups were housed separately, two of four animals vaccinated with plasmid pcDNA3.1/P1–2A3C3D alone were totally protected, without any signs of FMD. Upon co-immunization with the plasmid pcDNA3/GM-CSF, three of four animals were totally protected, and even the remaining animal had only one punctate lesion, at the site of inoculation. Although these numbers are small, the result does encourage further exploration of the potential of GM-CSF as an adjuvant in DNA vaccination against FMDV, particularly in view of the significant effect on anti-FMDV antibody production (see below). In experiment B, with the animals housed together and thus exposed to conditions of high virus aerosol after virus challenge, two of the four pigs injected with pcDNA3.1/P1–2A3C3D alone were partially protected, with greatly reduced clinical symptoms that appeared 2 days later than in the negative-control group. Specifically, there was only one small lesion in one animal and two lesions, one in the snout and one on the foot, in the other animal. When plasmid pcDNA3/GM-CSF was co-administrated with pcDNA3.1/P1–2A3C3D, three of four animals were partially protected, once again with just a few small lesions appearing late, 6–7 days after challenge. It is important to mention that, by day 8 post-challenge, one of the protected control animals vaccinated with inactivated FMDV vaccine showed one small lesion on the snout, presumably because of the high prevailing levels of virus resulting from proximity to infected animals.

Fig. 4. Specific antibody response to FMDV after DNA vaccination. Results are presented from experiment B, in which animals groups were housed together. The plasmid was injected three times at 3–4 week intervals (indicated by arrows). Challenge virus was given 2 weeks after the last plasmid injection. Animals were subsequently exposed to the high levels of virus aerosol produced by non-protected animals. Sera from immunized animals were sampled regularly and tested for antibodies at 1:80 dilution. Specific antibody response to whole FMDV was measured by capture ELISA (see Methods). The results are presented for individual animals as A₄₉₂. A group of pigs vaccinated with conventional vaccine was used as positive controls and animals vaccinated with pcDNA3/GM-CSF alone were used as negative controls.

Effect of co-administration of a plasmid encoding porcine GM-CSF on anti-virus antibody production

Results obtained when the plasmid encoding porcine GM-CSF was co-administered with pcDNA3.1/P1–2A3C3D were generally inconclusive (see above). There was, however, a significant positive effect on the stimulation of antibodies against virus proteins. Statistical comparisons of antibody titres against whole virus for two groups of eight pigs, one group vaccinated with pcDNA3.1/P1–2A3C3D and the other group vaccinated with pcDNA3.1/P1–2A3C3D plus pcDNA3/GM-CSF, before immunization (day 0) and after the first (day 17), second (day 28) and third (day 56) DNA immunizations showed that the group of animals that received pcDNA3.1/P1–2A3C3D plus pcDNA3/GM-CSF had statistically significant higher antibody levels against FMDV after the second (P=0.008) and third (P=0.04) DNA immunizations (Fig. 6).

FMDV is still an important pathogen of worldwide significance. It frequently escapes from endemic to non-endemic areas, as occurred in Japan in March 2000 and in the UK in 2001. The development of new recombinant vaccines is still an objective of immense practical importance. The use of DNA vaccines is a relatively recent approach in the vaccinology field that overcomes many of the problems associated with conventional immunization; for example, the potential danger associated with the production and distribution of attenuated live vaccines. In addition, the proteins expressed by recombinant plasmids are synthesized, processed and presented intracellularly, which is more similar to natural infection than is administration of the conventional FMDV vaccine.

Fig. 5. Specific antibody response to 3D RNA polymerase after DNA vaccination. Results are presented from experiment B, in which animal groups were housed together. This experiment was performed as described in Fig. 2 and with the same positive and negative controls. The plasmid injections are indicated by arrows. Specific antibody response to GST–3D was measured by indirect ELISA and the results are presented for individual animals as A₄₉₂.

Recently, the use of a DNA vaccine for FMDV serotype A has been reported. The plasmid encoded the capsid precursor together with protein 3C, but yielded variable protection and antibodies (Ward et al., 1997; Chinsangaram et al., 1998). In this study, we report experiments that extend this approach and suggest that DNA vaccination based on expression and processing of FMDV capsid precursors may be an alternative to the conventional inactivated whole-virus vaccine (Belsham, 1993; Roosien et al., 1990; Lewis et al., 1991; Saiz et al., 1994; Sanz-Parra et al., 1999; Mayr et al., 1999). The plasmid constructed (pcDNA3.1/P1–2A3C3D) contained all of the virus sequences required for correct production and processing of the FMDV serotype O₁ capsid proteins. Protein 3C was included in the construct to achieve cleavage of protein precursor P1–2A and 3D. In addition to the T cell and B cell epitopes present within the P1–2A precursor and 3C protease (Collen, 1994; Rodriguez et al., 1994), the 3D polymerase was included in the construct as an additional potent source of T cell epitopes (Foster-Cuevas, 1996; Collen et al., 1998). Although we have only demonstrated appropriate synthesis and processing of the capsid (and 3D) polypeptide chains from the transfected constructs in this particular study, and have not formally demonstrated assembly of empty capsids in vivo, previous electron microscopy observations with a similar construct (Abrams et al., 1995) did demonstrate formation of empty capsids. Thus, it seems probable that these were also produced in vivo in these studies as a result of the DNA vaccination. Consistent with this, animals vaccinated with pcDNA3.1/P1–2A3C3D alone induced antibody responses against the whole virus and against the 3D polymerase, and promising results were obtained in the protection experiments.

Table 2. DTH after DNA vaccination

For intradermal challenge, the antigen (recombinant GST–3D or FMDV vaccine) was injected intradermally in 100 µl PBS. Skin induration was measured 48 h after the injection. A positive result was recorded when the increase in skin thickness was 2 mm compared with normal skin. The data represent the numbers of animals that tested positive in each group of four.

Although three doses of DNA were administered, vaccination of pigs with plasmid pcDNA3.1/P1–2A3C3D alone gave promising results, particularly considering that the animals were challenged with a high dose (10^5.5 TCID₅₀) of FMDV O₁ Lausanne. This virus isolate, despite being related to O₁K in terms of antigenic properties and P1 nucleotide sequence, gives a more acute infection with clearer clinical symptoms than O₁K (J. Salt, personal communication). Thus, in experiment A, with the different groups housed separately, 2/4 animals were totally protected against high doses of virus challenge and in experiment B, where all animals were housed together, 2/4 were partially protected. In spite of the small number of pigs, the result of the second experiment (experiment B) becomes even more significant as one of the control pigs vaccinated with conventional vaccine developed a small lesion in the snout, presumably due to the high prevailing level of virus challenge encountered by all the animals.

Despite some exceptions (Sanz-Parra et al., 1999), there is generally a correlation between T cell-dependent virus-neutralizing antibodies and protection (McCullough et al., 1992; Francis et al., 1987; Collen et al., 1989) and, indeed, there was a correlation between protection and levels of neutralizing antibodies in our experiments. In contrast, however, levels of anti-virus antibodies detected by ELISA did not correlate consistently with neutralizing antibody levels and, consequently, with protection. Additionally, some animals with efficient anti-3D responses by ELISA were not protected.

Fig. 6. Effect of GM-CSF on antibody responses against FMDV after DNA vaccination. Pigs were injected with pcDNA3.1/P1–2A3C3D () or pcDNA3.1/P1–2A3C3D plus pcDNA3/GM-CSF (). Antibody responses to FMDV were compared on day 0 and on days 17, 28 and 56 after DNA vaccination (first, second and third injections, respectively). Results are expressed as mean absorbance values of eight animals in each group. Statistical analysis of the results was done with Student's t-test. Probabilities (p) are indicated for each time-point.

Despite the fact that this limited number of experiments has not established a clear correlation between DTH and protective immunity, it is clear from the observed synthesis of antibodies and the necessary concomitant generation of memory B cells that the DNA vaccination has stimulated T cell help. All of the conventionally vaccinated animals gave positive DTH responses to the virus, which indicates a Th1-type response, but this does not necessarily preclude a Th2 response. On the other hand, only 1/4 of the animals vaccinated with the DNA construct tested positive for DTH to the virus and there was no difference between the two experimental designs (experiments A and B).

A significant result in this study was the stimulatory effect of co-administration of a plasmid encoding porcine GM-CSF. Statistical analysis of the results showed that there was a significant increase (P=0.008) in anti-virus antibody production in those animals that received P1–2A3C3D plus GM-CSF. The increase in anti-FMDV antibody was statistically significant after the second and third DNA inoculations. In summary, this study suggests that DNA vaccination is a feasible strategy for stimulating porcine T cell and B cell responses against FMDV; its efficiency may be improved by co-administration of GM-CSF, perhaps through enhancing antigen presentation. Further work may improve the efficiency of this approach, with a reduction in the number of vaccinations required being particularly desirable.

This work was supported by EECFAIR grant 1317. We are grateful to Dr Paul Barnett from the International Vaccine Bank, IAH, Pirbright, for useful discussions and for providing FMDV antigen and to Bob Statham for preparation of FMDV vaccine. We would also like to thank Sarah J. Cox for performing the neutralizing antibody test. We are also grateful to the IAH animal caretakers for their help with animal handling.

† Present address: Centro de Investigacion y de Estudios Avanzados del IPN, Departamento de Biomedicina Molecular, Av. IPN #2508 Col. Zacatenco, 07360 México DF.

‡ Present address: Sir William Dunn School of Pathology, University of Oxford, South Parks Road, Oxford OX1 3RE, UK.

§ Present address: Gulbenkian Institute for Science, Rua da Quinta Grande 6, Apartado 14, P-2781 Oeiras Codex, Portugal.

---

References

Abrams, C. C., King, A. M. Q. & Belsham, G. J. (1995). Assembly of foot-and-mouth disease virus empty capsids synthesized by a vaccinia virus expression system. Journal of General Virology 76, 3089–3098.

Bautista, E. M. & Molitor, T. W. (1997). Cell mediated immunity to porcine reproductive and respiratory syndrome virus in swine. Viral Immunology 10, 83–94.

Belsham, G. J. (1993). Distinctive features of foot-and-mouth disease virus, a member of the picornavirus family; aspects of virus protein synthesis, protein processing and structure. Progress in Biophysics and Molecular Biology 60, 241–260.

Belsham, G. J., Brangwyn, J. K., Ryan, M. D., Abrams, C. C. & King, A. M. Q. (1990). Intracellular expression and processing of foot-and-mouth disease virus capsid precursors using vaccinia virus vectors: influence of the L protease. Virology 176, 524–530.

Chinsangaram, J., Beard, C., Mason, P. W., Zellner, M. K., Ward, G. & Grubman, M. J. (1998). Antibody response in mice inoculated with DNA expressing foot-and-mouth disease virus capsid proteins Journal of Virology 72, 4454–4457.

Collen, T. (1994). Foot and mouth disease (aphthovirus): viral T cell epitopes. In Cell-Mediated Immunity in Ruminants, pp. 173–197. Edited by B. M. L. Goddeeris & W. I. Morrison. Boca Raton, FL: CRC Press.

Collen, T., Pullen, L. & Doel, T. R. (1989). T cell-dependent induction of antibody against foot-and-mouth disease virus in a mouse model. Journal of General Virology 70, 395–403.

Collen, T., Baron, J., Childerstone, A., Corteyn, A., Doel, T. R., Flint, M., Garcia-Valcarcel, M., Parkhouse, R. M. E. & Ryan, M. D. (1998). Heterotypic recognition of recombinant FMDV proteins by bovine T-cells: the polymerase (P3Dpol) as an immunodominant T-cell immunogen. Virus Research 56, 125–133.

DiMarchi, R., Brooke, G., Gale, C., Cracknell, V., Doel, T. R. & Mowat, N. (1986). Protection of cattle against foot-and-mouth disease by a synthetic peptide. Science 232, 639–641.

Ellard, F. M., Drew, J., Blakemore, W. E., Stuart, D. I. & King, A. M. Q. (1999). Evidence for the role of His-142 of protein 1C in the acid-induced disassembly of foot-and-mouth disease virus capsids. Journal of General Virology 80, 1911–1918.

Foster, M., Cook, A., Cedillo, L. & Parkhouse, R. M. E. (1998). Serological and cellular immune responses to non-structural proteins in animals infected with FMDV. Veterinary Quarterly 20 (Suppl. 2), S28–S30.

Foster-Cuevas, M. (1996). Immunodeterminants of foot-and-mouth disease virus. PhD thesis, University of Hertfordshire, UK.

Francis, M. J., Fry, C. M., Rowlands, D. J., Brown, F., Bittle, J. L., Houghten, R. A. & Lerner, R. A. (1985). Immunological priming with synthetic peptides of foot-and-mouth disease virus. Journal of General Virology 66, 2347–2354.

Francis, M. J., Hastings, G. Z., Syred, A. D., McGinn, B., Brown, F. & Rowlands, D. J. (1987). Non-responsiveness to a foot-and-mouth disease virus peptide overcome by addition of foreign helper T-cell determinants. Nature 330, 168–170.

Fuerst, T. R., Niles, E. G., Studier, F. W. & Moss, B. (1986). Eukaryotic transient-expression system based on recombinant vaccinia virus that synthesizes bacteriophage T7 RNA polymerase. Proceedings of the National Academy of Sciences, USA 83, 8122–8126.

Garcia-Valcarcel, M., Doel, T., Collen, T., Ryan, M. & Parkhouse, R. M. E. (1996). Recognition of foot-and-mouth disease virus and its capsid protein VP1 by bovine peripheral T lymphocytes. Journal of General Virology 77, 727–735.

Geysen, H. M., Meloen, R. H. & Barteling, S. J. (1984). Use of peptide synthesis to probe viral antigens for epitopes to a resolution of a single amino acid. Proceedings of the National Academy of Sciences, USA 81, 3998–4002.

Golding, S. M., Hedger, R. S. & Talbot, P. (1976). Radial immuno-diffusion and serum-neutralisation techniques for the assay of antibodies to swine vesicular disease. Research in Veterinary Science 20, 142–147.

Grubman, M. J., Morgan, D. O., Kendall, J. & Baxt, B. (1985). Capsid intermediates assembled in a foot-and-mouth disease virus genome RNA-programmed cell-free translation system and in infected cells. Journal of Virology 56, 120–126.

Katrak, K., Mahon, B. P., Jones, W. C., Brautigam, S. & Mills, K. H. (1992). Preparative separation of foreign antigens for highly efficient presentation of T cells in vitro. Journal of Immunological Methods 156, 247–254.

Kim, J. J., Bagarazzi, M. L., Trivedi, N., Hu, Y., Kazahaya, K., Wilson, D. M., Ciccarelli, R., Chattergoon, M. A., Dang, K., Mahalingam, S., Chalian, A. A., Agadjanyan, M. G., Boyer, J. D., Wang, B. & Weiner, D. B. (1997). Engineering of in vivo immune responses to DNA immunization via codelivery of costimulatory molecule genes. Nature Biotechnology 15, 641–646.

Lewis, S. A., Morgan, D. O. & Grubman, M. J. (1991). Expression, processing, and assembly of foot-and-mouth disease virus capsid structures in heterologous systems: induction of a neutralizing antibody response in guinea pigs. Journal of Virology 65, 6572–6580.

McCullough, K. C., De Simone, F., Brocchi, E., Capucci, L., Crowther, R. J. & Kihm, U. (1992). Protective immune response against foot-and-mouth disease. Journal of Virology 66, 1835–1840.

Mayr, G. A., Chinsangaram, J. & Grubman, M. J. (1999). Development of replication-defective adenovirus serotype 5 containing the capsid and 3C protease coding regions of foot-and-mouth disease virus as a vaccine candidate. Virology 263, 496–506.

Rodriguez, A., Saiz, J. C., Novella, I. S., Andreu, D. & Sobrino, F. (1994). Antigenic specificity of porcine T cell response against foot-and-mouth disease virus structural proteins: identification of T helper epitopes in VP1. Virology 205, 24–33.

Roosien, J., Belsham, G. J., Ryan, M. D., King, A. M. Q. & Vlak, J. M. (1990). Synthesis of foot-and-mouth disease virus capsid proteins in insect cells using baculovirus expression vectors Journal of General Virology 71, 1703–1711.

Rowlands, D. J., Sangar, D. V. & Brown, F. (1975). A comparative chemical and serological study of the full and empty particles of foot-and-mouth disease virus. Journal of General Virology 26, 227–238.

Rweyemamu, M. M., Terry, G. & Pay, T. W. F. (1979). Stability and immunogenicity of empty particles of foot-and-mouth disease virus. Archives of Virology 59, 69–79.

Ryan, M. D., Belsham, G. J. & King, A. M. Q. (1989). Specificity of enzyme–substrate interactions in foot-and-mouth disease virus polyprotein processing. Virology 173, 35–45.

Saiz, J. C., Cairo, J., Medina, M., Zuidema, D., Abrams, C., Belsham, G. J., Domingo, E. & Vlak, J. M. (1994). Unprocessed foot-and-mouth disease virus capsid precursor displays discontinuous epitopes involved in viral neutralization. Journal of Virology 68, 4557–4564.

Salt, J. S., Barnett, P. V., Dani, P. & Williams, L. (1998). Emergency vaccination of pigs against foot-and-mouth disease: protection against disease and reduction in contact transmission. Vaccine 16, 746–754.

Sanz-Parra, A., Vázquez, B., Sobrino, F., Cox, S. J., Ley, V. & Salt, J. S. (1999). Evidence of partial protection against foot-and-mouth disease in cattle immunized with a recombinant adenovirus vector expressing the precursor polypeptide (P1) of foot-and-mouth disease virus capsid proteins. Journal of General Virology 80, 671–679.

Somasundaram, C., Takamatsu, H., Andreoni, C., Audonnet, J. C., Fischer, L., Lèfevre, F. & Charley, B. (1999). Enhanced protective response and immuno-adjuvant effects of porcine GM-CSF on DNA vaccination of pigs against Aujeszky's disease virus. Veterinary Immunology and Immunopathology 70, 277–287.

Tuteja, R., Li, T. C., Takeda, N. & Jameel, S. (2000). Augmentation of immune responses to hepatitis E virus ORF2 DNA vaccination by codelivery of cytokine genes. Viral Immunology 13, 169–178.

van Bekkum, J. G. (1969). Correlation between serum antibody level and protection against challenge with FMD virus. In Session of the Research Group of the Standing Technical Committee of the European Commission for the Control of Foot-and-Mouth Disease, Brescia, Italy, pp. 38–40. Rome: FAO.

Ward, G., Rieder, E. & Mason, P. W. (1997). Plasmid DNA encoding replicating foot-and-mouth disease virus genomes induces antiviral immune responses in swine. Journal of Virology 71, 7442–7447.

---

JGV Direct table of contents

© 2001 SGM

This article is now available in the July 2001 print issue of JGV (vol. 82, 1713–1724). The complete issue of the journal may be seen in electronic form on JGV Online.