Plant Production of Vaccines

Final Report Summary - PLAPROVA (Plant Production of Vaccines)

Executive Summary:
Introduction
The overall aim of project PLAPROVA was to exploit recent advances in the technologies for the rapid expression of proteins in plants to screen a range of vaccine candidates. The project focussed on veterinary diseases which are particularly relevant to both the EU and Russia, including Avian influenza virus (AIV), Human and Bovine papillomaviruses (HPV and BPV) Bluetongue virus (BTV), Porcine respiratory and reproductive syndrome virus (PRRSV) and Foot-and-mouth-disease virus (FMDV). It concentrated on the production of virus-like particles (VLPs) as these are known to be potent stimulators of the immune system.

Plant expression systems
To enable the rapid production of VLPs in plants, both replicating (plant virus-based) and non-replicating (Agrobacterium-based) expression systems were used. The replicating vectors were based on potato virus X (PVX) and Tobacco mosaic virus (TMV). The non-replicating systems were based mainly on the CPMVHT system which was refined to create the pEAQ-HT series of vectors during the course of the project.

Vaccine candidates
3 categories of vaccine candidates were successfully expressed in the course of PLAPROVA:
1. Simple VLPs which spontaneously form when a single polypeptide is expressed
2. Complex VLPs which only form when several polypeptides are co-expressed
3. Antigenic sequences which are fused to a carrier polypeptide which forms VLPs

1. Simple VLPs which spontaneously form when a single polypeptide is expressed.
Two simple VLPs were successfully expressed using the pEAQ-HT system. These were from the papillomaviruses HPV and BPV. The L1 proteins of HPV-8 and BPV-1 were successfully expressed in N. benthamiana and shown to form VLPs. The VLPs from BPV-1 were shown to be immunogenic in rabbits and immunogenicity testing of the HPV-8 is ongoing.

2. Complex VLPs which only form when several polypeptides are co-expressed
All four structural proteins of BTV serotype 8 (BTV-8) were successfully co-expressed in N. benthamiana using the pEAQ-HT and shown to assemble into VLPs. A protocol for the efficient purification of BTV-8 VLPs from plant tissue was devised and the purified particles shown to be stable on storage. The plant-produced VLPs were found to be able to confer protective immunity in sheep, thereby constituting an experimental vaccine.
The most successful approach to the creating of a plant-based FMDV vaccine was the expression in plants using a PVX vector of a polyepitope protein consisting of the fused antigenic sites of virus. The protein purified was shown to protect guinea pigs against FMDV challenge.
Expression of antigenic proteins from PRRSV in plants required the engineering of proteins from which the trans-membrane domains of the proteins had been removed. The assessment of the immunological properties of the modified protein is ongoing.

3. Antigenic sequences which are fused to a carrier polypeptide which forms VLPs
HBcAg and HPV-8 particles expressed in plants were identified as potentially suitable carrier molecules. It was shown that the M2e epitope AIV can be fused to HBcAg to create VLPs which can protect mice and chickens (partially) against AIV challenge. Three plant viruses were investigated as potential carriers - Alternanthera mosaic virus (AltMV; a filamentous virus), Cowpea chlorotic mottle virus (CCMV; an icosahedral virus) and TMV (a rod-shaped virus). All could be used as carriers of foreign epitopes. TMV particles expressing the M2e were shown to protect mice against AIV challenge.

Project Context and Objectives:
Summary of project context and objectives

At the start of PLAPROVA in Jan. 2009, the use of plants as bioreactors for the production of pharmaceutical proteins stood at a cross-roads. On the one hand, the previous 5 years had seen considerable advances in the technologies for expressing proteins and extracting them in an active form from plants. This culminated in 2006 in Dow Agroscience obtaining regulatory approval for a plant-expressed vaccine against Newcastle disease of poultry. On the other hand, most of the successes had concerned the production of well-characterised antigens and antibodies which had already been produced using previously established methods such as mammalian cell culture. From a scientific standpoint, this approach clearly made sense as it was important to establish the principal that plant-produced pharmaceuticals were comparable in safety and efficacy to their conventionally-produced counterpoints. Furthermore, given the time-lags associated with the production of lines of stably transformed plants, it was essential that proteins with previously characterised pharmacological properties were expressed as only a few candidates could be examined. However, the downside was that the plant-expressed proteins which were most highly developed and, in some cases, undergoing clinical testing were in direct competition with existing products. This is likely to hinder their commercial development, a point illustrated by the fact that Dow Agroscience has not marketed their Newcastle disease vaccine.

Given the above, it was clear that for plants to fulfil their potential as a means of producing pharmaceutical proteins, it was imperative that methods were developed for the rapid production and characterisation of plant-expressed proteins. This was particularly important for cases for which it was not possible and/or practical to produce the protein by other means. In these cases, there was likely to be far less pre-existing information available about the efficacy of the final product and it was essential that the properties of a number of variants were quickly assessed to determine which were the most suitable for further development.

Stable genetic transformation (either nuclear or plastid) is unsuitable for these studies in view of the time taken to obtain expression. It was one of the major aims of PLAPROVA to refine transient expression technologies so that they could be used to produce sufficient material in a sufficiently short time frame to enable pharmacological studies of a large number of vaccine candidate variants to be undertaken. This would permit optimisation of the methods of antigen presentation of a variety of different potential immunogens. At the same time, methods would be developed to allow the rapid translation of the information gained through the transient studies into larger scale production systems for the most promising candidates. A particular advantage of the transient approach is that yields can easily reach 10-30% of TSP - approximately 10-1000 times higher than usually obtained through nuclear transformation. Furthermore these high levels of expression can be achieved in the time-frame of a few days, potentially enabling a high-throughput analysis. At the start of the project, the consortium members had access to state-of-the-art transient expression systems previously developed by members of the consortium. These had already been used to express a number of proteins at levels up to and exceeding 1g/Kg leaf fresh weight. The availably of efficient transient expression system from the very start of the project ensured that it got off to a rapid start.

By using the transient expression systems available to the consortium at the start of the project, it was anticipated that it would be possible to go from cloning of constructs to the expression of milligram amounts of candidate immunogens within 2-3 weeks. This speed of analysis meant that it would be possible to screen a wide range of vaccine candidates within the lifetime of the project. Thus the consortium was able to evaluate potential vaccines against a number of diseases of great and increasing importance to both the EU and Russia. The PLAPROVA consortium concentrated on the expression of proteins which form polypeptide complexes or virus-like particles (VLPs). Due to their immunological properties these are particularly suitable as candidate vaccines. Furthermore, because of their particulate nature, they are relatively easy to purify.

In summary, PLAPROVA was a highly innovative project which aimed to exploit recent developments in plant expression systems to conduct a high-throughput analysis of the viability of producing both simple and complex virus-like particles (VLPs) in plants for vaccination purposes. It had two essential components

(1) The use and refinement of plant expression systems capable of expressing high levels of candidate immunogens within a short time-frame

(2) The identification of candidate immunogens and the immunological characterisation of the plant-expressed proteins.

These two essential components involved expertise which was not available at a single institution. To undertake the project, it was necessary to assemble a consortium with expertise in plant expression systems, animal virology and immunology. Thus an FP7 project was considered to be an ideal vehicle for undertaking the studies proposed. Given the international dimension of the diseases to be examined a SICA project, involving the EU, Russia and an International Cooperation Partner Country (ICPC) was considered to be particularly appropriate. All the partners within the assembled PLAPROVA consortium were considered world experts in their fields. The consortium thus created consisted of 11 institutions, 6 of which were based in a total of 5 EU countries, 4 within Russia and 1 in South Africa in order to obtain the relevant expertise. The outline of the project was discussed between the potential partners prior to the preparation of the proposal. As a result of this, the role of each partner was clearly defined and each member of the consortium was fully aware of their part in the project. In addition, all partners were aware of the need to closely co-ordinate their activities and agreed to adhere to the management policy for the project. In addition, it was recognized that co-ordination of a project involving Partners both within and outside the EU required special skills and therefore an appropriate management structure was devised. The importance of regular meetings of the whole consortium was recognised as was the value of exchanges of both material and scientist between partner laboratories.

The principle scientific and technological objectives of the project were:

1. To use high-level transient expression technologies recently developed by partners in the consortium to express candidate vaccines in plants
The fact that members of the PLAPROVA consortium had previously developed highly effective transient expression systems for plants meant that an immediate start could be made at the commencement of the project on the expression of candidate vaccines. The consortium anticipated that through the use of these existing systems it would be possible to produce milligram quantities of the some candidate vaccines within a few months of commencement of the project.

2. To Refine methods for the rapid expression of candidate immunogens
Members of the PLAPROVA consortium would refine methods for the rapid high-level transient expression of candidate immunogens in plants with particular emphasis on the development of high-throughput methods. These methods will be based on either Agrobacterium-mediated expression or on the use of replicating plant virus-based vectors. In particular, the consortium planned to develop methods for the simultaneous expression of multiple polypeptides within the same cell and in the appropriate subcellular compartment. Ways in which the stoichiometry of various polypeptides could be controlled would also be investigated. A vital aspect of this work was that any new expression technology developed would be made available to all members of the consortium.

3. To identify candidate immunogens for expression in plants.
It has been generally recognised that multivalent and especially particulate structures make the most efficient immunogens as they stimulate a repetoire of B- and T-cell responses. Thus PLAPROVA focussed on the production of virus-like particles (VLPs) and multi-protein complexes in plants. Furthermore such entities are generally easier to purify than individual proteins. If the antigenic sequences from a given pathogen did not themselves form particles they would be fused to an appropriate self-assembling macromolecule. The targets included both simple structures in which the VLPs consists of a single polypeptide species, such as those from Hepatitis B or Human or Bovine papilloma virus (HPV and BPV) and more complex, multi-chain VLPs, such as those from Foot-and mouth disease virus (FMDV), Avian influenza virus (AIV) Bluetongue virus (BTV) and Porcine Respiratory and Reproductive Syndrome Virus (PRRSV), viral diseases which are of emerging importance within the EU, South Africa and Russia. In addition to natural VLPs, the consortium planned to create antigenic chimaeras in which antigenic sequences are expressed on the surface of a carrier protein (such as HBcAg cores or plant virus particles).

4. Assessment of the ability of expressed proteins to self-assemble into VLPs
The candidate immunogens identified would be expressed using both the existing high-level plant expression systems and those developed during the project. The levels of expression would be quantitated and the ability of the proteins to form appropriate assemblies, such as VLPs, determined. The properties of the VLPs, including the stoichiometry of the various polypeptides, would be assessed. Clearly, the precise methods of analysis would depend on the structures being assembled.

5. Determination of antigenic and immunogenic properties of plant-expressed VLPs
Assembled complexes would be purified from plant tissue using methods appropriate to the particular VLP. For many of the VLPs expressed, methods could be adapted from previous work undertaken with purification from mammalian, avian or insect cells. As well as having advantages in terms of their immunological efficacy, VLPs tend to be more easily separable from host proteins than single subunit proteins due to their characteristically large sizes. Thus methods based on ultracentrifugation and/or size exclusion chromatography would be particularly effective methods of purification. Furthermore, for a number of the VLPs to be expressed, such as those from HBcAg, BPV and BTV, detailed protocols had already been devised for their purification from animal cell culture. Preliminary studies showed that these can be successfully adapted for their purification from plant tissue. Furthermore, the recent upsurge in interest in plant-made pharmaceuticals had led to the development of improved methods for eliminating plant-specific contaminants (such as polysaccharides) from protein extracts. The consortium would make use of these developments to produce VLPs of sufficient purity to be administered to experimental animals so that their efficacy can be assessed.

6. Development of methods for scaling up production desirable host plants
The consortium anticipated that most, if not all, of the initial transient expression studies would be carried out on a small scale in Nicotiana species. For the subsequent commercial development, it will be necessary to identify and develop methods both for the efficient introduction of genes encoding the immunogenic proteins into plants so that production can be scaled up and for their expression in most suitable plant species or tissue within a given plant species. To this end, the transient methods would be tested and modified to achieve expression in target issue and plant species. In particular, the consortium would take advantage of recent developments in the expression of proteins in root tissue.

7. Addressing intellectual property (IP) and regulatory issues
If the advances anticipated by this project were to be translated into practical and commercially viable products, it was essential that IP and regulatory issues were addressed from its commencement. The co-ordinator and several participants already had experience with obtaining patent protection for IP and in negotiating licensing agreements. Furthermore, through participation in the EU FP6 Pharma-Planta project, the coordinators team had previously consulted with the appropriate European regulatory agencies, EFSA and EMEA.

Project Results:
Description of the main S&T results/foregrounds

Overall strategy of Work plan

PLAPROVA was an innovative project which aims to exploit recent developments in plant expression systems to conduct a high-throughput analysis of the viability of producing both simple and complex virus-like particles (VLPs) in plants to develop novel vaccines. To undertake the project, it was necessary to assemble a consortium with expertise in plant expression systems, animal virology and immunology. To achieve its ends, the project was divided into 11 Workpackages.

Workpackages 1-3 were concerned with the development of efficient systems for the transient expression of candidate vaccines in plants. These would then be used for the expression of candidate vaccines in plants.

Workpackages 4-9 were concerned with the production of proteins which could serve as candidate vaccines for specific diseases and a determination of the immunological properties of the expressed proteins.

Workpackage 10 concerned the development of methods for upscaling production,

Workpackage 11 concerned effective management of PLAPROVA and the exploitation of the results of the project.

The description of the main results is broken down according to Workpackages described in the original proposal. However, it should be noted that the project was highly interactive and there was much cross-over of activity between the various Workpackages.

Workpackage 1: Use of the CPMV-HT system for high-throughput expression of candidate immunogens in plants

The aim of this workpackage was to see if a recently developed transient plant expression system, termed CPMV-HT would be a suitable method for the expression of VLPs in plants. This system is based on a deleted version of Cowpea mosaic virus (CPMV) RNA-2 (delRNA-2) in which the gene to be expressed is flanked by the 5 and 3 UTRs of RNA-2. This system had the advantage that it gives very high levels of protein expression in infiltrated leaves without the need for virus replication. Thus it does not suffer the problems associated with replicating systems such as genetic drift, limitations to the size of the genes that can be expressed and the problem of virus exclusion which severely hampers the expression of multiple proteins from replicating viral vectors. In addition, work was undertaken to refine the expression system to make it user-friendly so that it could be readily used both by other members of the PLAPROVA consortium and by researchers in other fields. This was achieved by reducing the size of the vector backbone and introducing sequences which enabled the easy and quick insertion of foreign genes. This resulted in the creation of the pEAQ vector series of vectors which are modular in design and allow the expression of multiple proteins from the same vector. Furthermore the expression levels of individual proteins can be modulated by changes to the 5UTR.

The system proved highly effective and was used to express VLPs from Hepatitis B core antigen (HBcAg), Human papilloma virus (HPV), Bovine papillomavirus (BPV) and bluetongue virus (BTV) as well as more complex proteins from Porcine respiratory and reproductive syndrome virus (PRRSV), Foot-and-mouth disease virus (FMDV) and Avian influenza virus (AIV). Progress on the expression of these proteins is described in the relevant workpackage. The pEAQ vector series also proved effective for the expression of proteins outside the initial objectives of PLAPROVA. These included the expression of RNA-free VLPs from CPMV

Because it concerned the use and development of a pre-existing expression system WP1 only ran for the first 24 months of the project.

Significant Results

Improved CPMV-HT system (the pEAQ vector system) for the easy and quick expression of target proteins in plants. pEAQ vectors made available to the members of the PLAPROVA consortium.
N. benthamiana tissue expressing simple VLPs from HBcAg and HPV in plants
Demonstration that the CPMV-HT system can be used to modulate translational efficiency thereby permitting the expression of proteins at different stoichiometries.
Production of pEAQ-based constructs designed to express complex VLPs from PRRSV, FMDV, AIV and BTV.
Demonstration that large quantities of CLPs and VLPs from BTV-8 could be purified from N. benthamiana leaves infiltrated with pEAQ vectors expressing structural proteins. .

Workpackage 2: Refining Agrobacterium-mediated delivery for high-level expression.

Agrobacterium-mediated gene transfer forms an essential component of all transient expression systems, either replicating viral vectors (such as those based on PVX or TMV) or the non-replicating CPMV-HT system. Thus to maximise expression levels it is highly desirable to use Agrobacterium strains capable of efficient delivery of expression plasmids contained within them. Workpackage 2 concerned the exploration of potential method for making Agrobacterium-mediated gene transfer more efficient. The results of these studies could then be deployed in conjunction with both replicating and non-replicating expression vectors. Unfortunately, the development of improved agrobacterium strains proved problematic and the only approach for improved expression proved to be the deployment of multiple suppressors of gene silencing. As with WP1, WP2 only ran for the first 24 months of the project.

Significant Results

Plasmid supplementation with the VirD and VirE genes does not give improvements in gene transfer efficiencies in Agrobacterium strains expressing functional versions of these proteins.
The use of plasmids expressing VirG does give improvements in gene transfer efficiencies in Agrobacterium strains used by consortium members.
The expression of a variety of silencing suppressors gives improvements in product yields and the provision of suitable constructs to other members of the consortium.

Workpackage 3: Development and exploitation of plant virus-based epitope-presentation systems

Since it is generally recognised that multivalent, and especially particulate structures, make the most efficient immunogens as they stimulate a repetoire of B- and T-cell responses, PLAPROVA concentrated on the expression of VLPs in plants. In a number of cases, expression of the native protein from the pathogen of interest will spontaneously self-assemble into a VLP. However, for a number of important pathogens of interest to PLAPROVA, this does not occur and it is necessary to fuse the antigenic sequences to an appropriate self-assembling macromolecule in order to produce VLPs. The aim of WP3 was to create self-assembling epitope-presentation systems based on plant viruses which could be used to create potential vaccines against a number of potential diseases including avian influenza virus. Three different plant viruses, Alternanthera mosaic virus (AltMV), Cowpea chlorotic mottle virus (CCMV) and Tobacco mosaic virus (TMV).

AltMV is a member of the potexvirus group of plant viruses. These are filamentous helical viruses whose coat proteins (CP) can be reassembled into VLPs free of RNA. Thus it should be possible to assemble VLPs after expression of the CP in plants. To verify that that the
AltMV CP can be used to present foreign epitopes, a histidine tag and sequences from the M2e epitope from AIV (details of this are given in the description of WP6) were fused to the C-terminus of the CP. Expression of the modified CP in plants was initially achieved using a commercially Agrobacterium binary vector which was available at the start of the project. The results demonstrated that the recombinant AltMV CP carrying M2e peptide are stable in plants, soluble and, making use of the his-tag, can be isolated by affinity chromatography. Though these results were encouraging, the level of expression was insufficient for further studies. Therefore the AltMV CP-M2e fusion proteins were inserted into a replicating potato virus X-based vector and used to infect N. benthamiana plants. The results showed that the expressed protein reacted with antibodies specific for M2e and recombinant AltMV CP-M2e accumulated to the levels of up to 1 mg per 1 g of leaf tissue. A method for obtaining VLPs from the recombinant AltMV CP derivatives expressed in plants by means of forced pH changes in leaf extracts was developed. The resulting VLPs were shown to be recognized by anti M2e antibodies, demonstrating that ALtMV CP is a suitable carrier of foreign epitopes.

As an alternative to the CP from helical virus, AltMV, the CP from the icosahedral virus CCMV was investigated. Once synthesized, the CCMV CP units assemble into VLPs consisting of 180 subunits that are permeable for substrate and products, and this permeability can be altered by changing pH. However, knowledge of CCMV VLPs as a potential nano-antigen carrier for vaccine development was still limited at the onset of this project. Initial work on CCMV CP VLP formation was performed in the baculovirus insect cell expression system and transiently in plants using the CPMV-HT (pEAQ-HT) expression vector. Although good expression levels and VLP formation were observed when the baculovirus insect cell expression system was being used, this system was rather time consuming for the rapid construction and testing of a larger range of VLP constructs. In contrast, no VLPs could be discerned when the CCMV CP gene was expressed in planta using the CPMV-HT system. The reason for this is unclear. As an alternative to plant-based expression , the CCMV CP gene was expressed in E.coli . These studies revealed high expression levels that could be folded relatively easy into large amounts of VLPs using a straightforward denaturation and renaturation procedure. Despite the easy production of large amounts of VLPs from E.coli expressed CCMV CP, analysis of a large set of (>12) His-tag and/or M2e epitope insertions/fusions in the in loops in the CP that were predicted to be exposed and anticipated to most likely maintain VLP formation, failed to give VLPs. Only fusions to the N- and C-terminus of CCMV CP, but also in a combined N-C-terminal setting, maintained VLP formation. Due to the possibility that larger epitope fusions (> ~40 aa) might still abrogate VLP formation, a second CCMV-VLP system is being developed which will allow the non-covalent attachment of epitopes via interacting coils.

In addition to the work on the RNA-free VLPs from AltMV and CCMV, the ability of RNA-containing particles of TMV, produced via infection with the virus, to present the M2e epitope was investigated. The TMV CP was modified to have insertion of the M2e sequence into the surface loop between 154 and 155 amino acid residues. Two vectors that contain complete genome of TMV with insertion of the sequence of M2e peptide in the CP gene were constructed. One of them contained the native M2e peptide, while another - its modified version. It was described earlier that cysteine residues in heterologous peptide might impede efficient assembly of chimeric particles. Therefore we introduced additional mutations in the natural sequence of Influenza epitope: cysteine residues at the positions 17 and 19 were changed to serine residues. The constructs were named TMV-M2e and TMV-M2e-serine. Virus particles were isolated from N. benthamiana plants systemically infected with TMV-M2e and TMV-M2e-serine. The results showed that cysteine-to-serine substitutions in M2e sequence considerably increased the stability of recombinant CP incorporated into virions. The genetic stability of TMV-M2e-serine was analysed by serial virus passages and subsequent analysis of the virus progeny for the presence of the M2e peptide fused to CP, as well as RT-PCR analysis of the progeny RNA genomes to verify the presence of M2e-serine-coding sequence. These experiments proved the genetic stability of TMV-M2e-serine vector.

Significant Results
Demonstration that native and His-tagged versions of AltMV CP can form VLPs in vitro
Expression of AltMV CP fused to M2e epitope in plants
Demonstration that CCMV can be used to produce VLPs in plants
Determination of optimum conditions for AltMV VLP formation in vitro
Demonstration of antigenicity of M2e epitope expressed on AltMV VLPs
Development of a method for the production of plant virus-based VLPs based on AltMV and PVX
Creation of TMV-based chimaeras presenting the M2e epitope from influenza virus
Development of a method for producing CCMV-based chimaeras for presenting epitopes

Workpackage 4: Creation and expression of modified Hepatitis B core particles in plants

The core antigen of hepatitis B virus (HBcAg) has attracted considerable attention as potential carrier for heterologous antigenic. HBcAg is the product of the HBV C-gene and is composed of between 183 and 185 amino acids (approximately 21 kDa), depending on sero-subtype. HBcAg has been shown to be capable of self-assembly into core-like particles when expressed in a number of heterologous systems. These core-like particles have antigenic structures identical to those of native HBcAg particles and it is possible to display foreign peptides at several different positions on their surface. Since core particles are potent inducers of B- and T-cell immune responses, the immunogenicity of the inserted foreign epitope is enhanced significantly in HBcAg-based chimaeras. As a consequence, HBcAg-based chimaeras have been developed as experimental vaccines against a number of diseases, including HBV itself.

To maximise the levels of HBcAg that can be produced in plants, synthetic codon-optimized genes have been created Full-length (HBcAg (aa1 to183) and fully truncated HBcAg (aa1 to 149) were expressed in N. benthamiana from pEAQ-HT In addition the synthetic HBcAg genes contained unique restriction sites within N terminus and immunodominant loop. A library of HBcAg molecules C-terminal modifications and antigenic chimaeras containing expressing the M2e epitope as created. The library of HBcAg mutants and chimaeras was expressed in plants using the pEAQ-HT vector system and the constructs assessed for their ability to form nucleic acid-free VLPs. Extracted and purified protein samples (HBc183, HBc149, HBc149M2e and HBc183M2e) were examined by electron microscopy in order to detect correctly assembled wild type and chimaeric HBcM2e VLPs. Efficient VLP formation with the non-modified full length-version of HBcAg (183aa) was found and there was an indication of VLPs formation from the chimaeras construct HBc149M2e.

The immunogenicity of HBc183 and chimaeric HBc149M2e extracted from inoculated N. benthamiana plants was assessed in BALB/c mice using subcutaneous immunization. Mice immunized subcutaneously with 50 µg HBc183 with adjuvant showed strong immune response against HBcAg antibodies after second immunization and produce high titre IgG while mice administered HBc149M2e with adjuvant showed induction of IgG specific for M2e at day 22. These results demonstrated that HBcAg expressed in plants can act as a carrier of foreign epitopes and that the epitope is presented in a form which is recognised by antibodies.

Significant Results
Expression of high levels of assembled HBcAg particles in plants using pEAQ-HT vector system
Generation of a library of HBcAg C-terminal deletion mutants
Construction of chimaeric HBcAg genes bearing the AIVM2e epitope and insertion of genes into pEAQ-HT vectors
Demonstration of plant expression of HBcAg-M2e protein
Demonstration of immunogenicity of plant-expressed HBcAg and HBcAg-M2e VLPs in mice

Detailed description of results in Workpackage

Task 4.1: Improvement of HBcAg gene sequence:

WP5: Expression of papillomavirus-based VLPs in plants

Papillomaviruses (PVs) are small non-enveloped viruses harbouring a double-stranded DNA molecule (8 Kb). Human papilloma viruses (HPVs) are a group of more than 100 different types of virus. Some types of HPV can increase the risk of developing cervical cancer, particularly numbers 16, 18, 30 and 33. In addition, a relationship has recently been discovered between non-melanoma skin cancer (NMSC) and HPV infection. NMSC, including basal cell and squamous cell carcinoma (SCC), is the most frequent cancer in the Caucasian population and its incidence dramatically increased worldwide in the last decades, exposure to sunlight being the most important environmental risk factor. SCC is often preceded by benign lesions called actinic keratosis (AK). Vaccines are now available to prevent infection with the certain types of HPV. However, they are expensive to produce and provide protection against only a subset of variants. Furthermore, for cutaneous HPVs, causing NMSC, no vaccine strategies are currently available. Thus there is a continuing need for new vaccines against HPV. In addition to affording protection against HPV itself, HPV VLPs can be used as carriers for additional antigenic sequences.

Bovine papillomavirus (BPV) is highly prevalent, with around 50% of cattle being estimated to bear lesions in the UK. A large survey of bovine papillomatosis in Scotland established that BPV causes teat and penile fibropapillomas (BPV1), common cutaneous warts and oesophageal fibropapillomas (BPV2), papillomas of upper gastrointestinal tract (BPV4), rice grain fibropapillomas of the udder (BPV5) and frond papillomas of teats (BPV6). Persistant BPV-induced lesions are at very high risk of progression to cancer. BPV-caused diseases are not only health problems but have also causes significant economic damage. Vaccination with BPV VLPs offers a potential route to protecting animals against infection.

This Workpackage involved using rapid transient approaches to produce VLPs from the structural proteins (particularly L1) of papillomaviruses HPV and BPV. The L1 protein of HPV-8, a prototype member of cutaneous HPVs, was transiently expressed in N. benthamiana using two expression systems, a deconstructed Tobacco mosaic virus (TMV)-based vector (Icon Genetics, Bayer) and pEAQ-HT. The HPV-8 L1 gene was cloned in both vectors as a full gene (L1) and as a 3terminally truncated version encoding for an L1 protein with 22 C-terminal aa deleted (L1?C22). Western blotting with an anti-HPV8-L1 polyclonal antibody confirmed that L1 and L1?C22 were expressed in N. benthamiana agroinfiltrated leaves using both expression systems. Comparing the two expression systems, higher expression levels were obtained with pEAQ-HT. On the basis of these results, pEAQ-HT was used as an expression system for further work. The approximate yield of L1 and L1?C22 protein was about 60 mg/kg fresh leaf weight for L1 and 240 mg/kg fresh leaf weight for L1?22. In both cases, VLP-like structures could be detected by electron microscopy of crude leaf extracts. For the purification of VLPs made of L1 and L1?22 from N. benthamiana leaves, a method commonly used for purifying the Begomovirus Tomato yellow leaf curl Sardinia virus (TYLCSV) was found to be most effective. It was estimated that the level of HPV-8 L1?22 expression (as a wild-type gene) was sufficient to proceed with immunization assays without need to synthesize the gene and optimize its codon usage.

As a result of the success in producing HPV-8 VLPs, engineering of L1 protein with heterologous epitopes to produce chimeric HPV L1 capsomers or VLPs as a multi-type vaccine platform was undertaken. For this, a synthetic L1 gene and its chimeric variants were designed, based on the L1 gene of HPV-16, a high-risk HPV. This HPV species was selected instead of HPV-8 because of its greater health impact and because of the wide panel of reagents suitable for protein and VLP characterisation. As a donor of heterologous epitope, Influenza A virus M2e was selected as an example of target disease relevant to PLAPROVA.
As a master gene, HPV-16 L1?C22 lacking the 3 22 amino acids was synthesized with a human codon usage. Western blotting with an anti-HPV16 Ll monoclonal antibody confirmed the expression of L1?C22 in N. benthamiana leaves using a pEAQ-HT vector containing the master gene (MG) and the four chimeric variants. The approximate yield of L1?C22 protein without foreign epitopes was 1 g/kg fresh leaf weight, while in the case of chimaeras the yields were somewhat lower. The most efficient chimeras were selected for further purification and immunization tests.

In the case of BPV, it was found that codon optimization of BPV-1 L1 substantially increased its expression levels in plants using the pEAQ vector system. The Begomovirus extraction procedure was used to isolate BPV L1 VLPs from N. benthamiana plants. Electron microscope analysis of the extracts revealed that expression of L1 led to the formation of small 30nm T=1 VLPs, Western blot analysis of the isolated VLPs revealed that a highly pure VLPs preparation at an estimated concentration of around 250mg/ kg fresh leaf weight- a level beyond that required for economic production had been obtained. The immunogenicity of the L1 VLPs was investigated by immunizing rabbits. A very strong cross reaction of the VLPs with the post-immunization antisera was obtained, which confirms that the plant produced VLPs are highly immunogenic. Thus a potential candidate vaccine against BPV has, for the first time, been produced in plants. Furthermore, FLAG and SP1 (sheep mite antigen) epitopes have been inserted into the BPV L1 protein, effectively replacing the amino acid region 128-137. This construct was transiently expressed in N. benthamiana and it was shown by electron microscopy that epitope insertion into this region does not disrupt VLP assembly.

Significant Results
Expression of assembled HPV-8 particles in plants
Expression of HPV L2 protein in plants
Expression of chimaeric versions of L1 proteins in plants
Production of BPV constructs
Expression of assembled BPV particles in plants
Expression of chimaeric versions of HPV and BPV L1 proteins in plants
Demonstration of immunogenicity of BPV particles produced in plants.

Workpackage 6: Development and expression of avian influenza immunogens in plants

Even though H5 influenza virus had never been shown to transmit between humans, concerns were raised over the possibility that a mutant or reassortant virus capable of efficient human-to-human transmission could trigger major influenza pandemic. The danger of recurring emergence of highly-virulent bird-like influenza virus in human population dictates the necessity of search for efficient protective methods. Recombinant vaccines against influenza are currently based on the highly immunogenic virion surface proteins, - haemagglutinin (HA) and neuraminidase (NA). Conventional influenza vaccines, including subunit vaccines based on HA and/or NA, must be developed every year to follow the antigenic drift and shift of the virus. Because of high variability of HA and NA, it has not been possible to develop a broad range vaccine against influenza. Unlike HA and NA, one of the minor virion proteins, M2, is highly conservative with almost no variations in all influenza strains of human origin being isolated since 1933. M2 is a membrane protein and its extracellular domain, M2e, is the promising vaccine candidate. The protein itself is, however, poorly immunogenic and needs to be attached to a carrier molecule.

In this Workpackage, the ability of several different carrier molecules to present AIV epitopes was investigated. The use of the plant viruses AltMV, CCMV and TMV to present the M2e epitope has already been described in Workpackage 3 and the work on the expression of HBcAg-M2e in plants has been described in Workpackage 4.

To determine whether expression of M2e on the surface of a carrier molecule can provide protection against AIV, BALB/c mice (48 weeks old) were immunized with the plant-produced preparation of M2eHBc particles. The particles were found to induce anti-M2e antibodies. To assess the protective effect of the vaccine, experimental and control mice were challenged with the mouse-adapted influenza strain A/Chicken/Kurgan/05/2005(H5N1). Upon infection the body weight of the immunized animals decreased significantly less than in the control mice, indicating that immunization with the candidate vaccine significantly relieved the course of the illness. The data showed a 90% protective effect of M2eHBc. In the control group of mice only half of the animals survived under the conditions of the infection. Although the main area of potential application of vaccine candidates developed within this project is the human health care, they may be also used for veterinary purposes. Therefore, in the second set of experiments immunogenicity and protective activity of candidate M2eHBc avian influenza vaccine was investigated in another model animals chickens. To assess the protective effect of the vaccine, experimental and control chickens were challenged with a highly pathogenic avian influenza strain A/Kurgan/1/05. It was observed that all chickens died although the death was delayed in immunised group (6 vs 4 days post challenge). In case of intranasal challenge, the survival of immunized chicken was 40%, while all control chickens died. Thus the M2eHBc particles only partly protected chickens against the lethal challenge with highly pathogenic avian influenza strain. However, the main purpose of influenza vaccine if it should be applied to chickens would be prevention of the spread of epidemic rather than complete protection of individual animal. Further experiements showed that though immunisation is unable to protect chickens from the lethal influenza infection but it can greatly reduce the spread of infection due to reduced excretion of virus from infected animals. To assess the immunogenicity of the M2e epitope presented on the surface of TMV, a preparation of TMV-M2e-serine isolated from plants was used for mice immunization. Immunisation of mice with TMV-M2e-serine resulted in 100% protection from the homologous influenza virus A/PR/8/34 (H1N1) upon infection Thus, TMV-M2e-serine preparation administered intraperitoneally show high immunogenicity and protection from infection with homologous influenza virus.

As an alternative to the creation of anti-AIV vaccine by genetically fusing the M2e epitope to self-assembling VLP, a synthetic polyepitope protein was created. This consisted of tandem repeats from the HA protein. The synthetic protein was expressed in E.coli and gave levels of expression of 1.5 - 2 mg per 1 liter of cultured cells. To increase the potential immunogenicity of the polyepitope protein, it was chemically linked to assembled particles of TMV. Data obtained in mice showed that the subcutaneous immunization of complex TMV-HA complex actively induced antibodies against the hemagglutinin of the virus that can interact with the whole virions of influenza virus.

Significant Results
Successful expression in plants of M2e epitope on surface of AltMV CP
Successful expression in plants of M2e epitope on surface of assembled HBcAg cores
Purification of assembled HBcAg-M2e particles from plants
Demonstration of antigenicity of assembled HBcAg-M2e particles produced in plants
Design and expression of a poly-epitope for influenza virus
Demonstration of antigenicity of polyepitope when linked to TMV particles
Plant-produced HBc particles carrying the M2e peptide of AIV are highly immunogenic.
Immunization of mice with plant-produced M2eHBc candidate vaccine protects the animals against the lethal challenge with avian influenza virus.
Partial protection of chickens against avian influenza virus by plant-produced M2eHBc candidate vaccine

Workpackage 7: Development of novel FMDV vaccines

FMDV is a highly contagious and devastating disease that affects all species of cloven-hoofed animals, including economically important species such as cattle, pigs and sheep. The disease is recognised by the World Organisation for Animal Health (OIE) as a major constraint to international trade. The current FMDV vaccine is an inactivated whole-virus preparation that is formulated with adjuvant prior to use in the field. It is estimated that about one billion doses of vaccine are used annually to reduce the number of outbreaks in enzootic areas. However, there are concerns about safety, cost-efficiency, stability and potency of existing vaccines. Thus the development of subunit anti-FMDV vaccines based on FMDV capsid proteins, peptides and epitopes have been the subject of intense research over the past years. Various subunit vaccine candidates, mainly based on immunodominant epitopes from the VP1 capsid proteins, have shown good protection in lab animals. However, none of these experimental vaccines reached the efficiency of current killed vaccines in protecting the livestock.

To develop more efficient FMDV subunit vaccines, two approaches were taken by PLAPROVA. In the first approach hybrid protein genes, combining within a single polypeptide chain both well-established B-and T-cell epitopes (VP1 21-40, 136-160, 200-213) and recently discovered T-cell epitopes (2C 68-76, 3D 1-115, 421-460) from the most widespread O serotype of FMDV isolate were constructed. The nucleotide sequence was optimised for expression in plants. The recombinant protein, HPE, was expressed in N. benthamiana plants using a replicating PVX-based vector. The protein was expressed at the level of 0.5-1% of the total plant protein, as estimated by SDS-PAGE, but appeared to be insoluble. Thus the protein (containing 6-HIS tag) was purified from plant tissue using metal-chelate affinity chromatography under denaturing conditions. This method allowed high purity HPE to be obtained suitable. The immunogenicity of the plant-produced HPE protein was evaluated in guinea pigs. A single immunization of guinea pigs by emulsion vaccine containing the plant-produced HPE protein in immunizing dose of 350 or 120 µg, respectively, induced in the animals the formation of virus-neutralizing antibodies to FMD virus type O/Taiwan/99. No evidence of infection was seen demonstrating that a single immunization of guinea pigs with a dose of 120 µg can induce immune response in the animals which confers resistance to the control infection with the homologous adapted FMD virus type O/Taiwan/99.

An alternative to the use of a polyepitope protein was the creation of artificial empty capsids of FMDV. The development of "empty capsid"-based FMDV vaccine approach requires co-expression of the polyprotein P1 (encoding VP0, VP3 and VP1 capsid proteins) from FMDV and a protease able to cut the polyprotein into individual capsid components. The assembly of the "empty capsid" occurs simultaneously with the processing. In the first approach adopted expression of the native FMDV P1 precursor was accompanied by expression of 3C protease of FMDV which is able to process the polyprotein. In the second approach we modified the cleavage sites within P1 polyprotein to be recognised by CPMV protease 24K, which is known to be less toxic than 3C. For expression of the precursor proteins and appropriate proteases four viral vectors were contructed in pEAQ-HT and PVX vectoe. The first vector, pEAQ-HT-P1, contained the native P1 polyprotein, while the second vector, pEAQ-HT-24KmP1, contained the modified P1 with CPMV 24K cleavage sites. The 3C protease is known to be very toxic for bacteria and it was not possible to clone this gene in the vector pA7248AMV. To prevent synthesis of 3C in E.coli and A. tumefaciens an intron was introduced into the 3C gene. The last of constructed viral vectors, pEAQ-HT-CPMV/24K, is based on pEAQ-HT and contained the gene for 24K protease of CPMV.

For expression of capsid proteins in plants, vectors encoding the precursors, were introduced into N. benthamiana leaves by agroinfiltration either alone or together with a second vector encoding their cognate protease (co-infiltration by two agrobacterial strains was used in that cases). The presence and processing of recombinant capsid proteins was analysed by western-blotting with antibodies against FMDV. The data showed that (i) the modified P1 polyprotein carrying the recognition sites for 24K protease is expressed in N. benthamiana plants, and (ii) this polyprotein is processed into shorter proteins if 24K protease is co-expressed. However, no VLP-like particles were observed by electron microscopy of protein samples isolated for agroinfiltrated plant leaves. Probably, processing of P1 polyprotein in plant cells and/or some steps of assembly of capsids differs from that occurring in animal cells thus precluding production of capsid VLPs. Therefore further projects aiming to develop plant-produced FMDV vaccine should focus on production of HPE polyprotein rather than on empty capsids.

Significant Results
Successful expression of FMDV multi-epitope fusion protein in E. coli and plants.
Demonstration that multi-epitope fusion protein can give an immune response in guinea pigs.
Expression of FMDV P1 precursor of the viral coat proteins in plants.
Demonstration of correct cleavage of P1 precursor by proteinase.
Demonstration that plant-produced multi-epitope fusion protein can protect immunised guinea pigs against infection by FMDV

Workpackage 8: Development of plant-based PRRSV vaccines

Porcine reproductive and respiratory syndrome (PRRS) is currently one of the major health concerns of the pig industry. It is endemic in the EU and other parts of the world and is associated with considerable economic losses. In fact, the causative agent, porcine reproductive and respiratory syndrome virus (PRRSV) is considered the animal virus which causes the highest losses in the pig industry throughout the World. Unfortunately, the problem has reached an acute level by the recent appearance in China of a highly virulent strain of PRRS ("blue ear disease") that in the past year has caused the sacrifice of 30 million pigs. Current vaccines against PRRSV have several drawbacks. Modified live vaccines protect against challenge with homologous isolates, but generally have a limited effect against challenge with heterologous viruses. Furthermore, live vaccines provide partial protection against clinical disease but did not prevent infection and, more importantly, they can revert to virulence. Non-infectious PRRSV vaccines produced in plants at low cost, on the other hand, could be effective in prevention of both infection and disease if high antigen doses can be produced at low cost by expressing a combination of two antigens inducing either neutralizing antibodies (GP5) or cellular responses (M protein).

Initial work involved attempts to express full-length GP5 and M proteins in plants using the pEAQ-HT vector system. Specific antibodies were used in western blot analyses carried out on plant extracts, but no protein was detected in any plant tissues. Furthermore infiltration with GP5 constructs led to necrosis in the leaves. In an attempt to alleviate this problem , it was decided to remove the transmembrane domains of the proteins, as these were believed to be responsible for the lack of expression in plants. GP5 contains an N-terminal putative signal peptide, followed by an ectodomain of approximately 35 aa containing the glycosylation sites and the neutralizing epitope. The rest of the protein is made by a hydrophobic region with transmembrane domains and a hydrophilic C-terminus of approximately 70 residues. Similarly, the M protein contains an ectodomain containing the cysteine involved in the dimer formation, and a transmembrane region followed by a hydrophilic inner domain. It was decided to express in plants only the ectodomains of both proteins, eliminating signal peptides and transmembrane regions. Sequences of GP5oi and Moi, obtained in the same way, were cloned in pEAQ-HT vectors and infiltrated in Nicotiana benthamiana leaves. Western blots showed the expression of GP5oi protein, whereas no Moi protein was detected, using specific antibodies.

To test whether the codon usage could have an effect on expression, the same engineered structures of GP5 and M proteins, without transmembrane domain, were codon-optimized, changing the codons according to the plant codon usage. In the new sequences, called GP5Synt and MSynt respectively, the outer and inner domains were joined without the central domain,and a Histidine tag was added at the C-terminus, in order to ease the purification process. Despite the lack of transmembrane domain, GP5oi and GP5Synt were insoluble and it was possible to detect them only by using a strong solubilization buffer. Another set of vectors carrying the new sequences GP5Synt and MSynt was built, in order to facilitate the expression of the heterodimer in plants. In this two vectors, the genes were either cloned in two separate cassettes within the same T-DNA or fused as one open reading frame in a pEAQ-HT derivative vector. In both cases, only GP5Synt was detected. Since the best results were obtained with the single protein GP5Synt, we decided to focus on that for the production of a plant-based vaccine. The plant-expressed protein was purified using an IMAC chromatography. Fractions were pooled and concentration was determined by ELISA assays. Up to 5 mg/Kg FW were purified from infiltrated plants. This material was sent for immunological testing. Due to the difficulties of producing sufficient material for testing, this work is still ongoing.

At the same time as the plant expression studies were being undertaken, other correlates of PRRSV protection were analysed by expressing different proteins in mammalian cells using recombinant Transmisible gastroenteritis virus (TGEV) vectors. TGEV vectors expressing PRRSV GP2a, GP3 and GP4, which are minor structural proteins exposed on virus surface assembled which may play a role on protection against PRRSV. A tricistronic rTGEV vector has been constructed, expressing each protein from a separate transcription regulating sequence (TRS). Unfortunately, this rTGEV vector was not stable as GP3 gene was lost in early stages. This result is in agreement with previous data obtained when expressing GP3 protein, alone or in combination with GP5 protein, using rTGEV system. To produce antibodies specifically recognizing PRRSV proteins GP2, GP3 and GP4 each gene was cloned in a baculovirus system for their expression in insect cells. Baculoviruses expressing high levels of each protein have been generated, and proteins have been purified. Polyclonal antibodies recognizing GP4 protein by immunofluorescence, and GP3 protein both by Western blot and immunofluorescence were successfully obtained.

rTGEV vectors expressing full length GP5, GP3 or GP4 proteins were only partially stable. As an alternative approach, rTGEVs expressing small GP5, GP3 and GP4 protein domains, containing the epitopes involved in protection were generated. In all cases, stability after 16 passages in tissue culture was significantly improved, i.e. up to 100% in the case of GP3 fragment. Previous evaluation of rTGEVs co-expressing PRRSV GP5 and M proteins indicated that M protein expression was stable, with at least 95% of infected cells expressing M protein for more than 10 passages in tissue culture. As M protein would be used as a platform for expression of chimeric proteins including domains involved in protection, further confirmation of the stability of rTGEV expressing M protein was required. The rTGEV expressing M protein was passaged 8 times in tissue culture, and 30 isolated viral clones were analyzed. From the analyzed clones, 100% expressed M protein mRNA, in agreement with the previous results. Virus was passaged 8 more times in tissue culture, and re-analyzed. 100% of the isolated clones expressed M protein mRNA, confirming M protein stability in rTGEV system.

In vivo immunological experiments using the TGEV vector expressing PRRSV GP5 and M proteins have been conducted. All animals presented a high antibody response against TGEV, indicating that the vector infected targeted susceptible tissues as expected. Also, vaccinated animals showed a clear antibody response against the PRRSV antigens (i.e. GP5 and M proteins). After a challenge with a virulent PRRSV isolate, a fast recall response was observed, as vaccinated animals induced higher antibody titers against PRRSV antigens and earlier than control animals. Nevertheless, the immune response did not provide full protection against PRRSV, most likely due to the low levels of neutralizing antibodies produced before challenge.

Significant Results

Production plant expression constructs for the expression of PRRSV proteins in plants
Demonstration of the difficulty of expressing full-length PRRSV proteins in plants
Production of constructs designed to counter the toxicity of PRRSV
TGEV vectors designed express minor proteins from PRRSV Demonstration of plant expression of constructs expressing deleted versions of GP5
Demonstration of the difficulty of the expression of M protein using TGEV
Partial protection by TGEV constructs expressing GP5 and M proteins
TGEV vectors expressing minor proteins from PRRSV

Workpackage 9: Creation of plant-expressed BTV vaccines

Bluetongue is a non-contagious, insect-transmitted, viral disease of domestic and wild ruminants. At present 24 serotypes of the virus are recognized. The virulence and mortality rate of the different virus strains vary considerably depending also on the infected species. Vaccination is regarded as one of the most effective ways of controlling and eventually eradicating bluetongue disease in affected areas. BTV is a complex, multi-layer virus which was a challenging target for expression using the high-level transient plant systems. The VPs 2 and 5 provide an outer shell laid on to the foundation provided by assembly of VP3 and VP7, which constitute the essentials of the inner shell: these latter two proteins expressed together can form core-like particles (CLPs). VP2 and VP5 do not form a structure without 3 and 7. Most neutralising antibody activity is against VP2, but VP5 elicits neutralising antibodies as well - and importantly, the two proteins can be used alone or in combination as a vaccine if produced via baculovirus. However, assembled particles are more immunogenic, and it would be highly desirable to express all 4 proteins simultaneously.

The gene sequences of the four structural proteins (VP2, VP3, VP5 and VP7) of BTV-8 were obtained and Nicotiana codon-optimised versions of these genes synthesized by the commercial company Geneart (Germany). These genes were cloned into 7 different expression vectors to screen for the one that yields the highest levels of intact BTV8 VLPs in N. benthamiana plant tissue. All four VP genes were successfully cloned into each of the 7 different vectors, either individually or in various combinations, and electroporated into the appropriate A. tumefaciens strains. Based on the results most of the subsequent work concentrated on using the pEAQ-HT system for the production of BTV-8 VLPs. To test for efficiency of expression of the four BTV-8 structural genes, the pEAQ-based constructs for VP2, VP3, VP5 and VP7 were initially expressed individually. Agroinfiltrated tissue was harvested 8 dpi and clarified extract separated on SDS-PAGE. All four constructs expressed well enough for bands to be visible on Coomassie-stained gels.

To test for the formation of core-like particles (CLPs), the codon-optimised constructs for VP3 and VP7 were co-infiltrated and harvested 9 dpi. The plant extracts were analysed on sucrose or iodixanol gradients. As expected, VP3 and VP7 co-sedimented in the gradient. Purified BTV-8 CLPs could be seen by electron microscopy using negative staining. The particles had the spiky appearance characteristic of BTV cores. To determine whether full VLPs can be produced using a similar manner, the four BTV-8 structural proteins were co-expressed and subjected to analysis as described above for CLPs. Coomassie staining of density gradient fractions revealed a high yield and co-sedimentation of the expressed BTV-8 structural proteins. The purified particles were observed using by electron microscopy. With a mean diameter of 87 nm, VLPs have a less distinct surface than CLPs. The appearance of VLP preparations was variable, often showing an abundance of partially (dis-)assembled VLPs and CLPs, with only few VLP structures. This was due excess of VP3 was found to be produced, leading to an excess of CLPs, rather than the desired VLPs. To modulate the relative expression of the structural genes, a version of pEAQ-HT with a modified 5 UTR was usd to drive VP3 expression. (See results from Workpackage 1). Expression of VP3 from the modified leader greatly reduced the level of VP3 produced and boosted the yield of VLPs as opposed to CLPs. This strategy was therefore adopted for the expression of BTV-8 VLPs for immunological analysis. Furthermore, methods of purification of VLPs were refined. The stability of the BTV-8 VLPs after storage at 4°C for 8.5 months was investigated. Comparison of the banding pattern showed that very little degradation of the VPs had taken place. Comparisons of the VLPs using TEM after storage at 4°C for 14 months showed less VLPs per frame than seen immediately after production, but there were still a lot of whole particles visible which were approximately 60 µm in diameter, similar to the documented size of CLPs. This indicates that the particles are very stable over time, and augers well for the production of a commercially viable vaccine

The ability of purified BTV8 VLPs produced in N. benthamiana to stimulate an immune response in sheep was investigated. Two BTV-free sheep were inoculated with N. benthamiana-produced VLPs to test whether they produced an antibody response to BTV. These results showed that the sheep mounted an immune response to the plant-produced VLPs. The ability of the plant-produced BTV 8 VLPs to protect sheep from a live BTV challenge was also carried out. One group of 5 sheep was inoculated with 50µg of plant-produced BTV 8 VLPS. A second group of 5 sheep was inoculated with the BTV 8 attenuated live monovalent vaccine manufactured by Onderstepoort Biological Products. A third group of sheep was inoculated with PBS as a negative control and a fourth group was inoculated with 200µg CLPs. The Clinical Reaction Index (CRI) of the sheep inoculated with the BTV 8 plant-produced VLPs was 0 (all the sheep survived the challenge) showing that the VLPs protected the animals to the same level as the live attenuated monovalent BTV strain (commercially available vaccine Onderstepoort Biological Products BTV8.0) which was also 0. The CRI of the sheep inoculated with the BTV 8 plant-produced CLPs showed an elevated CRI of 10.82 but this was still less than the CRI measured in the animals which received PBS as a negative control. These results are highly significant in showing that plant-produced BTV 8 VLPs can fully protect sheep (one of the major animals affected by BTV infection) against BTV and is the first time that this has been shown.

Significant Results

Successful expression of all four structural proteins of BTV-8 in N. benthamiana
Demonstration that co-expression of VP3 and VP7 leads to the formation of CLPs in plants
Demonstration that co-expression of VP3 and VP7 together with VP2 and VP5 leads to the formation of VLPs in plants
Demonstration that modulation of translational efficiency improves VLP (as opposed to CLP) formation.
Development of an efficient protocol for the production of plant-produced VLPs Demonstration that plant-produced BTV VLPs are stable on storage
Demonstration that plant-produced BTV VLPs can elicit protective immunity in sheep

Workpackage 10: Scale-up of transient expression systems.

This Workpackage was intended to develop methods for the scale-up of the production of proteins based on the transient approaches developed and refined during WPs 1, 2 and 3. Developing methods for scaling-up production was regarded as an important step for the commercial development of plant-made vaccines. The approaches included the development of a new TMV vector based on strain U1which gave very high levels of marker gene expression and the determination of the minimum Agrobacterium titre that can efficiently deliver constructs into tissue. Minimizing the amount of Agrobacterium culture necessary for construct delivery is important for cost reasons when production is scaled up.
Attempts 1 to transform N. benthamiana with pEAQ vectors designed to express GFP did not result in successful generation of line of transgenic plants. It was hypothesised that this was due to the presence of the powerful suppressor of silencing, P19, on the T-DNA of this construct. While P19 has previously been shown to be extremely effective in increasing expression levels in transient expression studies, it was also known to cause harmful developmental effects when expressed as a transgene. The address this problem, Partner 1 obtained a mutant version of P19 (mP19, containing a R43W change) Transient expression studies showed that though less potent than wt P19, P19/R43W was an effective suppressor using the CPMV-HT expression system. Stable transformation of N. benthamiana with pEAQ-GFP constructs harbouring mP19 yielded plants that expressed elevated accumulation of GFP compared with the levels found in the absence of a suppressor. Transgenic expression of mP19 caused no or minimal morphological defects and plants produced normal-looking flowers and fertile seed. Thus expression of mP19 is developmentally harmless to plants while providing a suitable platform for transient or transgenic over-expression of value-added genes in plants without hindrance by RNA silencing. This finding has been exploited to develop transgenic plants expressing the antibody 2G12 and human gastric lipase (hGL).

The use of aeroponic agroinjection (AA) technology for scaling up production of proteins was investigated. Following aeroponic agro-infiltration of Agrobacterium strain GV3101 carrying a tobacco rattle virus (TRV) carrying GFP, GFP expression was observed in the root tissue Therefore the TRV vector system has the capacity for in planta expression of foreign proteins, and may be used in conjunction with the aeroponic agro-infiltration high through-put root transformations. The efficiency of T-DNA delivery into plant roots using a variety of A. tumefaciens strains and also an armed A. rhizogenes strain. It was found that while all Agrobacterium types were able to induce GFP fluorescence in roots to a similar level, armed A. rhizogenes was the most potent.

Significant Results
Development of an efficient tobamovirus vector based on the U1 strain of TMV
Determination of minimum Agrobacterium titre for efficient delivery of TMV vector
Demonstration that pEAQ-HT expression system can be adapted for use in stable transgenic plants
Demonstration of utility of modified pEAQ-HT expression system for use in stable transgenic plants
Development of a TRV-based vector for the expression of proteins in roots

Workpackage 11: Management of consortium

Consortium management has been led by JIC (P1). CB-RAS (P8) has also contributed
significantly, particularly in liaison with the Russian partners. Other partners have contributed where appropriate, for example, in organising meetings. Consortium management activities have included the following:

1. General project management
Project management was managed and coordinated by the dedicated project manager, working closely with George Lomonossoff and the JIC Contracts and Finance departments to form the project management team. The project manager was responsible for the non-scientific management of the project. This includes the overall administrative, legal and financial management. Specific tasks in this reporting period have included the following:
Coordinating, contributing to and supporting the first and second periodic reports submitted in months 20 and 38.
Obtaining progress reports from partners at month 24 for internal progress monitoring.
Managing interim payments to partners after the approval of the first periodic report.
Contributing to the organisation of General Assembly meetings in Norwich, UK in Jan. 2009, Moscow, Russia in July 2009, Varna, Bulgaria in June 2010 and Torino, Italy in October 2011 and giving a presentation on WP11 and all project-management- and reporting- related issues.
Maintaining the list of contact details and individuals involved in the project for each beneficiary, and making this available to the consortium.
Preparing guidance and templates for the beneficiaries for the second periodic report and the final report. Detailed guidance was needed as several beneficiaries had little no prior experience of the Framework Programmes and reporting.
Preparing material for the project website.
Providing general advice, information and assistance to beneficiaries on issues relating to the management and administration of the project, for example on eligible costs, deadlines, contractual issues. Much of this fell to the project manager as the administration in several of the beneficiaries was not experienced in FP7.

This work was all directly ascribable to the project, and would not have been carried out if JIC was not coordinating PLAPROVA. It was separate from and additional to the project managers usual work of general research grant support which is included in JICs overhead calculation.

Project website
The project website is at http://www.plaprova.eu. It was established in January 2009 and has been updated regularly. Project partners were encouraged to contribute content. The website is in two parts: a public site that includes general, non-confidential information about the project and the participants; and the private, password-protected wiki which is only accessible to members of the PLAPROVA consortium. The private site includes copies of contractual documents; guidance notes on publication and material transfer procedures; project news; reports; and agenda, minutes and presentations from project meetings. JIC expects to maintain and update the project website for at least 3 years after the project end date.

Overall assessment of results of PLAPROVA

PLAPROVA has provided a wonderful demonstration of the power of transient expression methods for the production of novel vaccines in plants. Within the space of 3 years it has:

Developed highly efficient systems for the transient expression of proteins in plants. These have applications beyond the immediate objectives of the project.

Demonstrated that VLPs can be efficient produced and assembled in plants using transient expression technologies. This has implications for the production of a wide variety of VLPs in the future.

Shown that VLPs produced in plants are immunogenic and, where tested, can stimulate protective immunity in animals.

Demonstrated that a multidisciplinary project involving groups from the EU, Russia and South Africa was a highly effective way to achieve scientific progress.

As a result, PLAPROVA has made a huge contribution to advancing the concept of producing vaccines in plants.

Potential Impact:
Potential impact

The overall goal of PLAPROVA was to develop and exploit practical systems for the high-throughput expression of candidate immunogens in plants. This involved exploiting new and emerging research opportunities and technologies to contribute to a major aspect of both human and animal health the development of novel vaccines which can be produced at low cost in plants. This goal was intended a major contribution to the development of a European Knowledge-Based Bio-Economy (KBBE). In that it proposes the use of plants as bioreactors for the production of high value proteins, it is highly relevant to Activity 2.3 Area 2.3.1 Improved biomass and plant-based renewables.

In terms of delivery of its scientific objectives, PLAPROVA proved to be highly effective. This was, in large part, due to the efficient operation of the consortium and the excellent collaboration between its members. Thus the project demonstrated how effective collaboration between the EU, Russia and South Africa can be and it is hope that it will pave the way for future collaborative work.

As testament to the overall impact of PLAPROVA, the co-ordinator (Prof. G.P. Lomonossoff) has been shortlisted for the U.K. Biotechnology and Biological Sciences Research Council (BBSRC) Innovator of the Year award. This shortlisting is based on his work on developing A system for the rapid production of vaccines and pharmaceutical proteins in plants.

The overall goal of the project was addressed by taking a number of specific steps as outlined below:

1. The exploitation and further development of systems for the rapid high-level expression of immunogens in plants. A particularly notable development was the creation of a series of small and user-friendly binary vectors based on the CPMV-HT system, the pEAQ series, which allow the rapid expression of proteins in plants without virus replication. These vectors have proved to be extremely useful to the members of the PLAPROVA consortium. For example the vectors were successfully used to VLPs from HBcAg, HPV, BPV and BTV. However, their utility is not restricted proteins of interest to the PLAPROVA consortium but also to others in the field of plant-based protein expression. Thus the co-ordinator of PLAPROVA discussed the licensing and commercial exploitation of discoveries related to the CPMV-HT with Plant Biosciences Ltd. (PBL), a technology interaction and Intellectual Property Management Company based at the JIC, Norwich. As a result, the system has been licensed to a number of commercial enterprises including Medicago inc. of Quebec City Canada who are utilising it to produce plant-based influenza vaccines for human use. In addition, more than 120 Material Transfer Agreements (MTAs) have been signed with academic laboratories worldwide, enabling the pEAQ vectors to be widely distributed and applied to a wide variety of different research project. Thus the pEAQ vectors have the potential to revolutionise the use of plants as bioreactors.

2. The rapid expression and characterisation of immunogenic proteins which can act as carriers of antigenic sequences. The PLAPROVA consortium successfully used plants to produce VLPs consisting of a single protein from HBcAg, HPV and BPV as well as from a number of plant viruses. These are important disease targets and availability of a cheap source of these VLPs could have a major impact on the availability of both prophylaxis and therapy. For example, candidate prophylactic HPV vaccines based on the structural L1 protein of either the mucosal HPV16 or the cutaneous HPV8 have been obtained. Although vaccines are now available to prevent infection with the certain types of HPV, they are expensive to produce and provide protection against only a subset of variants. Thus there is a continuing need for new vaccines against HPV. In addition to affording protection against HPV itself, HPV VLPs can be used as carriers for additional antigenic sequences. Thus the plant-expressed HPV VLPs have the potential for preventing cancer in humans caused by HPV and therefore could have considerable impact in the medical field. Likewise the availability of highly immunogenic structures of BPV has the potential to have a positive impact on farmers since BPV is highly prevalent, with around 50% of cattle being estimated to bear lesions in the UK. BPV-caused diseases are not only health problems but have also causes significant economic damage, e.g cows with teat papillomas cannot be milked, young calves cannot suckle. Occasionally, entire herds have to be culled if the papillomatosis does not regress. Thus the BPV VLPs have a significant potential impact on animal health with concomitant economic benefits.

As well as expressing VLPs which have an intrinsic ability to self-assemble, the PLAPROVA consortium also investigated the use of proteins with an intrinsic ability to self-assemble as platforms for epitope display. These proteins included the coat proteins from the plant viruses AltMV, CCMV and TMV as well hepatitis B core particles (HBcAg). All these systems proved effective at presenting the M2e epitope from AIV. In the case of HBcAg- and TMV-M2e fusions, it was shown that presented epitope was able to protect mice against challenge with AIV. Furthermore, chickens could also be partly protected against AIV challenge. This is highly significant since even though AIV has never been shown to transmit between humans, concerns have been raised over the possibility that a mutant or reassortant virus capable of efficient human-to-human transmission could trigger major influenza pandemic. The danger of recurring emergence of highly-virulent bird-like influenza virus in human population dictates the necessity of search for efficient protective methods. Thus the results have huge implications for the control of influenza epidemics.

3. The identification and characterisation of target proteins for expression in plants.
A significant achievement of PLAPROVA was the creation and expression in plants of a polyepitope protein (HPE) consisting of various epitopes from the capsid proteins of FMDV. When injected into guinea pigs, this synthetic protein was shown to confer protective immunity and is thus a potential FMDV vaccine. This work could have huge impact in the field of veterinary medicine. . FMDV is a highly contagious and devastating disease that affects all species of cloven-hoofed animals, including economically important species such as cattle, pigs and sheep. The disease is recognised by the World Organisation for Animal Health (OIE) as a major constraint to international trade. In outbreak situations, effective control strategies must, thus, be put in place to stop the spread of the virus. The disease is endemic in many parts of the world and outbreaks of all seven serotypes of FMDV have recently occurred all over the world with multibillion dollar losses. The 2001 FMDV epidemic in Northern Europe led to the change in the EU FMDV control policy, permitting emergency vaccination to block FMDV spread. The current FMDV vaccine is an inactivated whole-virus preparation that is formulated with adjuvant prior to use in the field. It is estimated that about one billion doses of vaccine are used annually to reduce the number of outbreaks in enzootic areas. However, concerns about safety, cost-efficiency, stability and potency of existing vaccines has stimulated further research aimed at a deeper understanding of immune responses during FMD infection and development of novel FMDV vaccines, such as the HPE protein produced in plants by PLAPROVA.

Porcine reproductive and respiratory syndrome (PRRS) is currently one of the major health concerns of the pig industry. It is endemic in the EU and other parts of the world and is associated with considerable economic losses. In fact, the causative agent, porcine reproductive and respiratory syndrome virus (PRRSV) is considered the animal virus which causes the highest losses in the pig industry throughout the World. Unfortunately, the problem has reached an acute level by the recent appearance in China of a highly virulent strain of PRRS ("blue ear disease") that in the past year has caused the sacrifice of 30 million pigs. The new PRRSV strain spreads remarkably quickly by unknown means and is characterized by very high mortality. Currently the syndrome is having a devastating impact on the Chinese swine industry. Different types of vaccines are commercially available which are widely used. Nevertheless, these vaccines reduce the clinical symptoms, but do not prevent infection and, hence, allow maintenance and spread of the virus mainly due to its high genetic variability that allows viral escape. Furthermore, the available vaccines most likely will be partially protective for the European and North-American virus isolates but not to the new Chinese virus. PRRS is not a notifiable disease in the EU or subject to other harmonised disease control measures. However, third countries that are free of the disease impose safeguard measures for imports from EU affected countries. Thus the work of PLAPROVA in the development of a new generation of safe and efficient vaccines that induce immune responses differentiate vaccinated from naturally infected animals is desperately needed and, if successfully concluded would have an enormous impact on agriculture worldwide..

4. The exploitation of efficient methods for the expression of heteromeric VLPs. The pEAQ vector system has been shown to be a very effective method for producing heteromeric complexes in plants. Thus it was applied to the creation of novel vaccines against BTV which have been shown to be capable of conferring protective immunity in sheep. Bluetongue is a non-contagious, insect-transmitted, viral disease of domestic and wild ruminants. At present 24 serotypes of the virus are recognized. The virulence and mortality rate of the different virus strains vary considerably depending also on the infected species. Vaccination is regarded as one of the most effective ways of controlling and eventually eradicating bluetongue disease in affected areas. It reduces clinical signs in affected animals resulting in lower mortality and reduced economic losses, and it prevents the spread of the disease amongst livestock. Vaccines against bluetongue can either be inactivated vaccines or modified live virus (MLV) vaccines. Inactivated vaccines, when administered in 2 separate doses, are able to fully protect animals for a long period. Modified live vaccines generate protective immunity after a single inoculation, and they have been proven effective in preventing clinical BT in the areas where they are used.

Under EU legislation, Member States can apply vaccination as a control measure against bluetongue. Member States that wish to carry out a bluetongue vaccination campaign must inform the Commission. Bluetongue vaccination has been successfully used in a number of European countries which have been affected by the disease. Italy, Spain, France and Portugal have all used vaccination as a means of controlling and eradicating outbreaks of the bluetongue virus. For certain strains of the bluetongue virus, including BTV-8 which
caused outbreaks in Northern Europe, no vaccine has been available up to now. Vaccination does not immediately protect the animal from infection if there is a virus circulating. When the vaccines are administered to uninfected animals, the onset of a protection is observed only after a certain period, depending on the biological properties of the vaccines. For this reason, EU legislation lays down movement restrictions and/or controls for vaccinated animals to ensure that they do not contribute to the spread of the disease. However, no strategy is currently available to distinguish vaccinated from infected animals on the basis of serology. The demonstration that plant-produced BTV VLPs can confer protective immunity in sheep will thus have a major impact on vaccination strategies.

Dissemination Activities

From the commencement of PLAPROVA, the co-ordinator and all partners were acutely aware that this project has a very definite potential for producing results that will be suitable for commercial exploitation.

There are three aspects of PLAPROVA which have immediate commercial potential:

1. The development of systems with enhanced characteristics for the high-level rapid production of immunogens in plants
2. The development of novel VLPs capable of presenting foreign antigenic sequences.
3. Novel vaccines consisting of multiple polypeptides produced in plants.

In view of the possibilities for exploitation, there is a clear need for efficient handling of Intellectual Property matters. To ensure this occurs, a framework for dealing with such matters formed part of the consortium agreement which was drafted before the start of the project. Intellectual Property matters formed a significant topic of the first meeting of the consortium. Subsequently, the project participants kept in regular contact to discuss the prospects for the dissemination and/or exploitation of the results of the consortium as they arise. The decision as to what Intellectual Property might need protection made use of the high level of experience in such matters available within the consortium.

The co-ordinator (Partner 1) has discussed the licensing and commercial exploitation of discoveries related to the CPMV-HT with Plant Biosciences Ltd. (PBL), a technology interaction and Intellectual Property Management Company based at the JIC, Norwich. Indeed, in view of the clear potential shown by the preliminary data on the use of Cowpea mosaic virus (CPMV), the co-ordinator has already filed patent applications with PBL. The work demonstrating the protective immunity of plant-produced BTV particles was conducted in collaboration with Onderstepoort Biological Products (OBP) a licenced animal vaccine manufacturer in South Africa. OBP realise the implications of the results of this research on the agriculture sector of South Africa as well as on that of other neighbouring countries, and are very keen to see it go further. Partner 7 intends to apply to a government agency in South Africa (Department of Science and Technology) for further research funding to refine and further test the results of this research. There are other factors which we would like to investigate in order to make the vaccine even more cost effective. These include factors such as testing immunogenicity and efficacy in a range of animals other than sheep that are also susceptible to BTV, dose testing to minimise the dose required to retain the same efficacy of the vaccine, further development of vaccines which can protect against multiple serotypes, as well as testing less purified preparations of the vaccine which could also reduce the cost of production.

In terms of more general dissemination all the partners in PLAPROVA have communicated their results at appropriate scientific meetings (either orally or through posters), seminars and though publications in the scientific literature. A detailed list of these activities, which mainly, but not exclusively, involved dissemination to a scientific audience, is provided in Section 4.2.

In terms of more general dissemination, the following activities are of particular note:

The setting up of the website for the project website http://www.plaprova.eu/ to which all members of the consortium contributed

The production of a poster at the start of the project by Partner 1 describing the aims of PLAPROVA. This was made available to all members of the consortium for presentation at any appropriate venue.

The arranging of a workshop by Partner 6 entitled Future and perspectives in Plant Molecular Farming held in Varna, Bulgaria on 04/06/10.

A presentation by Partner 1 on PLAPROVA to an open meeting in Brussels on 25/10/11.

A publication in International innovation in Feb. 2012 describing PLAPROVA.

The involvement of several members in the EU COST action FA0804 Molecular Farming: plants as a production platform for high value proteins. This enabled dissemination of the results of PLAPROVA to others in the field of plant-made pharmaceuticals.

List of Websites:
http://www.plaprova.eu/

final1-p38-40-plaprova.pdf

Documents connexes

final1-p38-40-plaprova.pdf

Final Report Summary - PLAPROVA (Plant Production of Vaccines)

Documents connexes

Partager cette page

Télécharger