Development of Vaccines Toward the Global Control and Eradication of Foot-and-mouth Disease

Luis L Rodriguez; Cyril G Gay


Expert Rev Vaccines. 2011;10(3):377-387. 

In This Article

New Experimental Vaccine Platforms

Since structural proteins are the main antigens responsible for inducing protective responses,[36,37] several attempts have been made to improve current inactivated FMD vaccines by utilizing cloned capsid proteins expressed by rDNA technology. However, the subunit vaccines produced in Escherichia coli and peptide vaccines induce narrow immune responses that the virus easily gets around through antigenic drift mechanisms.[38] Recently, significant improvements in rDNA-based vaccines have been made offering improvements in efficacy, safety and use in disease control and eradication.[39] Most of these improvements consist of introducing mechanisms for proteinase processing of the viral capsid proteins that result in structurally more complex antigens. Some of these new approaches are described in the following sections.[40]

DNA Vaccines

Vaccination using plasmid DNA containing FMDV sequences has been reported as an efficient way to induce protective immunity in the mouse model.[41,42] However, protection by DNA vaccination in farm animals such as cattle, sheep and pigs has proven more challenging and requires multiple doses and addition of adjuvants and cytokines (e.g., GM-CSF, IL-2) to induce only partial or in some cases full protection.[42,43] In most DNA vaccines the antibody response, which is critical in protection against FMDV, is both limited and short lived.[44] Despite these shortcomings, DNA vaccines are appealing because plasmid DNA does not require high containment facilities for manufacture, is relatively stable for storage, allows for the rapid incorporation of emerging field strain sequences and allows discrimination between infected and vaccinated animals.[41,43] Another interesting feature of DNA vaccines is that they have been reported to induce protection not only against clinical disease but also prevention of the carrier state in sheep.[43] Other applications of DNA vaccines include a prime–boost approach using inactivated vaccines to improve their efficacy.[45] Using priming with a plasmid DNA-containing capsid and some of the nonstructural proteins of FMDV followed by boost with inactivated vaccine and recombinant 3D protein resulted in high antibody titers and protection of swine not only against homologous but also against heterologous challenge. Although not a practical approach at this time, this promising research should continue.

FMD Peptides

In laboratory animal models (e.g., mice, guinea pig), several FMD capsid-based peptide vaccine candidates have been shown to induce peptide-specific anti-FMDV serum-neutralizing (SN) antibody titers, and in some instances have been shown to confer protection against FMDV challenge.[46,47] Unfortunately, these positive results in laboratory animal models have not been consistently reproduced in cattle and pigs.[25] Although early studies in cattle showed promise,[48] in a large-scale synthetic-peptide vaccination study in 138 cattle using four different FMDV serotype C VP1 G-H loop-based peptides, none of the peptides, tested at several doses and vaccination schedules, conferred protection in above 40% of the vaccinated animals.[38] Notably, several mutant FMDV strains were isolated from vaccinated cattle, suggesting that peptide vaccination induced the rapid generation and selection of FMDV antigenic variants in vivo.

Efforts to improve and broaden VP1 G-H loop peptide immunogenicity through the incorporation of T helper (Th) sites and consensus residues into the hypervariable positions (UBI peptide) resulted in a high level of protection in swine following FMDV 01 Taiwan challenge.[49] A subsequent pilot study in cattle showed that the UBI peptide induced peptide-specific antibodies but relatively low SN titers, and failed to protect cattle following FMDV type O challenge at 3 weeks post-vaccination.[25] However, a peptide vaccine is now commercially available.[202] FMDV peptide vaccine adjuvanted with cholera toxin and administered transcutaneously elicited antipeptide antibodies with enhanced virus neutralizing activity in mice.[50] However, further experiments demonstrating efficacy in target species are still required. Recent studies in swine utilizing nontoxic Pseudomonas aeruginosa exotoxin A expressing the FMDV VP1 G-H loop failed to induce protective immune responses.[51]

The recent development of dendrimeric peptides containing one copy of an FMDV T-cell epitope branching out into four copies of a B-cell epitope provides potential improvements over the conventional linear peptide.[52] Pigs vaccinated with a dendrimeric peptide and subsequently challenged with FMDV did not develop significant clinical signs, appear to have abrogated systemic and mucosal FMDV replication, and did not transmit the virus to contact controls. The dendrimeric peptide used in this experiment elicited an immune response comparable to that found for control of FMDV-infected pigs. Dendrimeric designs for other FMDV serotypes and subtypes need to be developed and tested, but this new technology provides substantial promise for peptide-subunit vaccine development.

Virus-like Particles

Foot-and-mouth disease virus-like particles (VLPs) are nonreplicating, nonpathogenic particles that have structural characteristics and antigenicity similar to the parental virus. They are similar in conformation to intact virions and are formed by the self-assembly of processed capsid proteins. A critical component of VLP experimental vaccines is the ability to process the viral capsid polyprotein (P1) into cleaved products that can then assemble into VLPs. There are different protein-processing paths that have been pursued by the inclusion of nonstructural viral proteins 2A, 2B and 3C proteinase.[37,53,54] There are several expression systems for the production of VLPs, including:

  • Various mammalian cell lines, either transiently or stably transfected or transduced with viral expression vectors [55,56]

  • The baculovirus/insect cell and larvae systems [57–59]

  • Various species of yeast including Saccharomyces cerevisiae and Pichia pastori[60]

  • E. coli and other bacteria [61–63]

A yeast-derived VLP experimental FMD vaccine was initially described in 2003.[60] The capsid from a serotype O strain induced SN and ELISA titers in guinea pigs and these animals were protected against homologous challenge. More recently, co-expression of either recombinant bovine IFN-γ,[64] IL-18[65] or HSP-70[66] and VP1[64,65,67] constructs has been shown to enhance SN and cell-mediated immune responses in mice; however, no livestock vaccine efficacy studies have been reported.

Baculovirus-derived VLP experimental FMD vaccines have been shown to provide some protection against clinical disease in swine, but fail to protect against viral replication. Similar results using an E. coli-derived VLP experimental vaccine were also reported.[37] Recent reports have shown improvements by using baculovirus and silkworm larvae to express FMDV P1 and including protein 2A at strategic sites to facilitate processing of VP1–VP2 and VP0.[59] The authors provide electron microscopy evidence of VLP assembled in the larvae lysates. Furthermore, vaccines prepared in this fashion for serotypes Asia and O conferred protection when used to immunize cattle.[57] Further testing is necessary to determine the feasibility of this approach for large-scale production of FMD vaccines for livestock.

Hepatitis B virus core (HBc) particles self-assemble into capsid particles and are extremely immunogenic. However, formation of VLPs can be restricted by size and structure of heterologous antigens. The first report of the use of the HBc system for expression of amino acids 141–160 of the VP1 protein of FMDV was made over 20 years ago, and the immunogenicity of the VLP structures was reportedly similar to that of intact FMD particles.[68] Very recently, the formation of VLP in mammalian cells by modified HBc fused with specified FMDV multi-epitopes was studied. Complete VLP structures with one construct was confirmed by electron microscopy and induced both humoral (peptide- and FMDV-specific antibody) and cell-mediated immunity responses in mice.[69]

The generation of VLP experimental FMD vaccines using transgenic plants has also shown some laboratory success. Arabidopsis thaliana-transformed plant extracts expressing the FMDV VP1 gene were shown to provide protection against FMDV challenge in mice.[70] Similar studies have also been reported using transgenic potato plants[71] or alfalfa plants[72] as immunogens in serology and challenge studies in mice. Related studies using HBc to express a VP1 capsid epitope in transgenic tobacco has also been reported.[73] To date, none of these transgenic plant-derived VLP experimental vaccine candidates have been tested for efficacy and safety in cattle or swine, and the regulatory and manufacturing path for transgenic plant-derived vaccines is not well defined.

Viral-vectored FMD Vaccine Platforms

Viral vectors have been successfully used to deliver sequences coding for FMDV capsid proteins into animals. There are several examples including herpesvirus (pseudorabies), poxviruses and α-virus vectors. However, the best-documented and most effective platform uses human defective adenovirus 5 (hAd5). Unlike experimental DNA vaccines that mainly target the expression of the specific antigens to be presented to the immune system on the surface of transfected cells, the purpose of this platform is to provide the genetic information to express and process all the FMDV structural proteins, presumably resulting in the formation of virus-like particles in the hAd5-infected cells; although direct evidence of this has not been clearly shown.[74] The hAd5–FMD vaccines are the most extensively tested and were shown to be as effective as inactivated vaccines against FMD. Complete protection has been shown both in swine and cattle receiving one vaccine dose and challenged as early as 7 dpv.[75–77] The hAd5–FMD vaccine platform contains all FMDV capsid proteins and nonstructural proteins 2A and 3C, but lacks other nonstructural proteins such as 3B and 3D. These two proteins are the targets of serological tests aimed at differentiating infected from vaccinated animals, making the hAd5–FMD platform fully compatible with existing DIVA serological tests. Importantly, the hAd5–FMD platform allows for safe vaccine production without the need for high biological containment facilities, as the process does not involve the use of live FMDV. In addition, this vaccine platform does not require adaptation of field strains to vaccine production cells and precludes potential antigenic changes resulting from cell adaptation and virus replication during the vaccine manufacturing process. A hAd5–FMD serotype A vaccine is the most advanced product candidate in development. To date, this vaccine has been tested in over 150 cattle and shown to provide protection against generalized FMD disease and viremia following a single immunization and intra-dermal lingual (tongue) or contact challenge at 7 or 21 dpv.[77] Similarly to the commercially inactivated vaccines, protection as early as 4 dpv has been demonstrated in some, but not all animals.[76] These new molecular FMD vaccine candidates are currently being manufactured in experimental batches and tested in cattle in the US mainland (for the first time in US history) as part of the veterinary licensing process.[77]

cDNA-derived Inactivated FMDV Vaccine Platform

The utilization of cDNA-derived FMDV as a safe candidate platform for production of inactivated vaccines was first reported utilizing leader-deleted FMDV.[78–80] Attenuation of FMDV was achieved by manipulating the genome to eliminate the Lpro coding sequence, which is known to be involved in FMDV pathogenesis in vivo. In a pilot study, this virus platform was shown to be completely attenuated in cattle based on absence of tongue lesions following direct inoculation of susceptible cattle.[79] Once inactivated, the antigenic properties of the cDNA-derived FMDV vaccine are expected to be those obtained from inactivated wild-type virus. More recently, this platform has been improved by adding unique restriction enzyme sites for rapid swapping of capsid sequences and also by deleting specific epitopes in nonstructural proteins (NSPs) 3B and 3D allowing for serological DIVA testing [Rieder E. Pers. Comm.]. Serum from animals inoculated with this leaderless marker virus can be readily distinguished from parental FMDV-infected animals utilizing DIVA serological tests such as competitive ELISA for NSP. Accordingly, unlike conventional inactivated FMD vaccines, this platform eliminates the need to remove NSP during the manufacturing process, and since the vaccine virus is attenuated, eliminates the concern associated with the manufacture of live FMDV if the vaccine virus escapes from the manufacturing facility.

Modified Live Vaccines

In the past few years, development of live-attenuated FMD vaccines has shown limited success owing to unstable phenotype, differences in attenuation among various livestock species (e.g., attenuated in cattle but not in swine) or vaccines that were incapable of consistently inducing protective immune responses.[81–85] Some of these vaccines relied on the use of viruses selected in cell culture or in laboratory animals showing attenuated phenotypes, but the mechanisms of attenuation were largely unknown and attenuation was often incomplete and reversible.[81] Other vaccines engineered to contain attenuating mutations or gene deletions were better characterized but were too attenuated and did not consistently induce protection.[85] Owing to these problems and concerns over reversion to virulence through mutation or recombination with field viruses, live FMD vaccines have not been developed. However, new technologies such as the development of reverse-genetic cDNA systems for FMDV provides new opportunities for identifying virulence determinants in the FMDV genome. As new virulence determinants are identified, new possibilities for attenuated vaccines will arise. Translating this knowledge into vaccine candidates will require detailed understanding of the virus–host interaction and mechanisms of pathogenesis in order to address concerns about complete attenuation in all susceptible species, and even eliminating the possibility of reversion to virulence. All this is now possible given the current availability of well-established infectious cDNA FMDV systems and the ability to engineer multiple specific mutations in critical elements that regulate the virulence of the virus.[86–90] Importantly, this methodology also allows the engineering of negative antigenic markers to include DIVA capability.

Adjuvants & Biotherapeutics

The FMDV incubation period can be as short as 2 days and animals can shed virus prior to signs of generalized disease.[3,4] Since FMD vaccines generally require at least 4–7 days for protective adaptive immunity to develop, it is critical that FMD control programs address the gap in the onset of immunity provided by current vaccines to limit and control disease spread. With recent breakthroughs in our understanding of antiviral innate defense mechanisms, biotherapeutics or immunomodulators offer the potential for their use as an emergency tool to stop viral shed and spread within 24 h after administration.[91–93] When used in combination with rapid-acting vaccines. it may now be possible to elicit early short-term anti-FMDV effects until the onset of vaccine-induced protective immunity.

It has been previously shown that FMDV replication is inhibited by type I interferons (IFN-α/β).[94,95] At least two IFN-α/β-stimulated genes, double-stranded-RNA-dependent protein kinase and 2',5'oligoadenylate synthetase/RNase L, are involved in this process.[95] It has been demonstrated that an Ad5 vector containing the porcine IFN (pIFN)-α gene (Ad5–pIFN-α) induces high levels of biologically active interferon when injected in swine. Furthermore, swine inoculated with a single dose of Ad5-pIFN-α were completely protected when challenged with FMDV 1 day later.[74] The level of protection correlated with the Ad5-pIFN-α dose and the level of plasma IFN-α. Additional studies demonstrated that Ad5–pIFN-α treatment alone can protect swine from challenge for 3–5 days and can reduce viremia, virus shedding and disease severity when administered 1 day post challenge. Importantly, a combination of Ad5–pIFN-α and Ad5–FMD vaccination provided both immediate and long-term protection in swine.[96,97] Type II pIFN (pIFN-γ) also has antiviral activity against FMDV and, in combination with pIFN-α, has a synergistic antiviral effect.[92] The results indicate that the combination of type I and II interferons act synergistically to inhibit FMDV replication in vivo and confer protection against challenge. Similar studies in cattle have only shown partial but not complete protection against FMDV challenge.[91]