Plants, edible vaccine producing bioreactors
Through the centuries plants have proved to be extremely flexible organisms that adjust to man’s requirements both for agricultural and ornamental purposes. About twenty years ago early studies on the recombinant DNA method also found the possibility of introducing genes resulting from other vegetable species and of bacterial or animal origin into vegetable tissue to ensure their expression in this context.
This discovery has opened new horizons concerning the extensive number of applications, which could be proposed. Early studies naturally focused on improving plants’ defence against pathogens and infesting organisms. To this end resistant genes of bacterial or fungal origin or isolated from other plants naturally endowed with the desired defence have been introduced. It takes little to move from this stage to imagining the possibility of making plants produce other protein types, which are not necessarily to its advantage. In particular, the possibility of producing molecules for health purposes (i.e. antigens for the production of vaccines, antibodies or proteins with a pharmacological function) have given rise to considerable interest

How to Produce Innovative Vaccines
The three main categories of vaccines available today are:
Attenuated vaccines, formed by pathogens with reduced virulence;
Inactive vaccines, formed by microrganisms, whose virulence has been neutralized;
Subunit vaccines, formed by purified elements of microrganisms. In this case the vaccine will only hold part (formed by one or more proteins) of the original organism, which will be adequate to trigger an immune reaction;
Recombinant vaccines. This last category is based on recognizing antigen molecules, which can induce an immune response, on isolating the corresponding gene and on producing the vaccine in a heterologous expression system, which has so far been a microganism or animal tissue.
The best method of expression will naturally be the one that ensures a safe final product, optimal biological activity and low production costs. Antigen expression in mammal cells has the advantage of giving suitable products, but the procedure is expensive and not free of danger (possible presence of pathogens derived from the animal used). The use of microrganisms as a method of expression enables a more extensive production but it also has limits when secondary changes typical of eukaryote cells (i.e. glycosylation) or a special fold in the protein produced are required.
A new method of expression based on the integration of adequate genes in superior plants has been recently proposed. This would offer considerable advantages for the product’s safety thanks to the absence of contamination with animal pathogens or toxins that could be present in vaccines expressed in animal or bacteria cells. Given the ease with which plants produce plentiful biomass on an agricultural and industrial scale, the approach would considerably cut down costs. Lastly vaccines made from plants would not depend on the “cold chain”, which is required today for conservation and distribution and which could enable oral administration as an alternative to intravenous injections (Daniell et al., 2001; Sala et al., 2003).
The possibility of administering edible vaccines expressed in plants has given rise to great interest even in the veterinary field, especially for animals bred for food purposes, considering the social and economic implications related to their health.
The possibility of directly providing these animals with substances that have pharmacological activity in the form of food would greatly simplify drug administration. When vaccines too must be precisely dosed, transgenic plants that express the gene can be easily dehydrated, the active principle can be dosed and hence administered after mixing it with the daily ration. This would considerably cut down costs and would avoid the stress induced in animals by the injection method.
Studies on vaccines produced in plants have developed in many directions in the past 10 years. Described below are some examples of the first results obtained with this technology: the expression of the hepatitis B specific antigen in tobacco and lettuce (Mason et al., 1992; Eheani et al., 1997), of the rabies specific antigen in tomatoes (McGarvey et al., 1995), of a cholera specific antigen in tobacco and potatoes (Arakawa et al., 1997) and of a cytomegalovirus specific antigen in tobacco (Tackaberry et al., 1999). Experiments conducted on laboratory animals have revealed these vaccines’ capacity to stimulate the immune system by inducing the synthesis of antibodies against hepatitis B (Thanavakam et al., 1995), the Norwalk virus (Mason et al., 1996) and bacterial enteritis (Haq et al., 1995).
In early experimentation on human volunteers, the vaccine formed by E. coli’s enterotoxin’s subunit B expressed in tobacco produced an immune response both on mucous tissue and systemically in individuals orally treated with an adequate dosage of transgenic potato (Tacket et al., 1998). The achievement of mucous tissue immunity is one of the points in favour of edible vaccines. Many pathogens penetrate the body through mucous tissue. The first defences are hence the ones present on the very mucous tissue that cover airways, the digestive system and the urogenital one. The possibility of building chimerical genes, which express detoxificated forms of the cholera toxin (CT) and of E. coli’s thermolabile enterotoxin (LT), bound as adjuvants to antigen proteins (Di Tommaso et al., 1996; Kong et al., 2001) has been designed to further improve vaccine potential expressed in plants to evoke a mucous response
Besides vegetable cells are surrounded by a cellulose wall that protects them from the action of gastric juices. In non-ruminant animals vegetable tissues are carried directly to the intestinal lumen where they undergo slow lysis and can interact with a slow release natural system.

Vegetable Transformation Methods
There are essentially two plant transformation methods: one based on the use of Agrobacterium and the biolistic particle delivery system. The former is based on vegetable pathogens A. tumefaciens and A. rhizogenes’ property to integrate their DNA (T-DNA) with infected cells’ nuclear genome (de la Riva et al., 1988). The introduction of exogenous genes into the adequately modified T-DNA of Agrobacterium cells and the following infection of a vegetable tissue lead to the study gene’s stable integration in the plant’s genome and to the production of a transgenic protein (Figure 1).

The application of Agrobacterium-mediated transformation, first limited to tobacco and to few other species, which are the infection’s natural targets, has now been extended to most vegetable species marked by agronomic interest, including Graminae and legumen (Lee et al., 2001; Chikwamba et al., 2002). This opens interesting new prospects for the development of edible vaccines for both human and veterinary use.
The second approach is based on the microprojectile bombardment method (Taylor and Fauquet, 2002). Selected DNA sequences are precipitated onto metal microparticles and bombarded against the vegetable tissue with a special tool (particle gun) at an accelerated speed. Microparticles penetrate the walls and release the exogenous DNA inside the cell where it will be integrated in the nuclear genome through mechanisms that have yet to be entirely cleared,
Vegetable cells have cytoplasmic organelles called chloroplasts, which contain chlorophyll, generally known for their photosynthetic function. These organelles, which, like mitochondria, are supposed to derive from ancient bacterial predecessors and which have penetrated a larger cell as symbionts, have an independent chromosome complement, but their characteristics are typical of prokaryote cells. The biolistic particle delivery system “shoots” adequately processed DNA particles, which, penetrate into the chloroplast and integrate with its genome. The chloroplast’s transformation is an interesting alternative to nuclear transformation (Maliga, 2002; Daniell et al., 2002). In fact, according to some published data on the transformation of tobacco for the expression of the Bacillus thuringiensis’ (BT) insecticide toxin, the introduction of exogenous genes into the chloroplast’s genome has led to a collection of active recombinant proteins amounting to 47% of total soluble proteins (Decosa et al., 2001). This type of transformation offers the advantage of producing many transgene copies per cell.

Besides, unlike what occurs in the nucleus, the chloroplast’s transformation is based on the exogenous gene’s insertion for homologous recombination. The transgene can thus be inserted in a precise point of the plastid chromosome. This avoids positions that can have negative effects on the plant’s growth, which often occur in nuclear transformations, following the random insertion of a gene in the nuclear genome.
However so far chloroplast genetic engineering has been carried out only in tobacco and partly in the potato. We can hope that in time this method will be extended to other species that are more interesting for the production of edible vaccines (i.e. corn, lettuce and clover).
Our laboratories at the University of Milan’s Department of Biology are currently studying the possibility of making plants express various proteins marked by antigenic activity both for human and veterinary use. In the framework of cooperation with the Pasteur Institute in Paris, a polyepitope (a molecule comprising a series of peptide fragments endowed with immunostimulating activity) with antigenic features against human melanoma was expressed in tobacco leaves. It is currently at an early experimentation stage to evaluate its level of immunogenicity.
An earlier stage of this process is research targeted at producing veterinary vaccines. This study, conducted in collaboration with many Institutes of the University of Milan’s Faculty of Veterinary Medicine, first focuses on checking a new vaccine producing system’s potential through a step by step comparison of “traditional” vaccines and vaccines expressed in plants.

Research Stages
This goal was achieved by marking out diseases which already had marketed protein-based vaccines (the antigen) with tested immunogenicity. Concerning proteins, it is important to know their structure, amino acid sequence and especially the nucleotide sequence of the gene they derive from. This data must be carefully evaluated because, though every living being’s DNA only comprises 4 nucleotides and though the genetic code is more or less universal, every organism has its “preferences” and some special features in its protein translation and maturing systems. For example, some amino acids are encoded by more than one DNA triplet, but every organism preferentially uses certain triplets compared to others. Hence the guest organism must present all the triplets required for the protein’s translation. Other problems could result from proteins that require glycosylation, since the chemical structure of saccharide groups added by plants during a protein’s maturing phase can slightly differ from animal structures and this could cause the formation of an antibody that does not perfectly meet requirements. One must be aware of such points ahead to adequately modify the gene, which must be introduced into the host plant (i.e. with site-specific mutation).
Current studies are developing along various lines and focus on transforming plants into defence agents against diseases caused by many organisms and with equally distant target organisms. A few examples are filariasis in animals, and also in humans, caused by a nematode worm, a special form of bovine enteropathy caused by Escherichia coli bacteria and a degenerative disease in horses’ respiratory and reproductive systems caused by a herpes virus strain. A protein, which, reproduced with traditional methods, can be used as a vaccine has already been marked out and extensively studied for all these diseases.
In the study’s early phase we chose to use tobacco as a vegetable system, the best known and most studied system, to minimize the variable factors involved. Naturally in this case experimentation on animals must be conducted with protein extracts free of alkaloids normally present in plant cell juice. The following phase envisages the transformation of an edible plant (i.e. rice).

If the genes are adequate for expression in the chosen host, they must be placed in the plant under the control of an active promoter, which can be recognized by vegetable polymerases. This promoter can be either constitutive, in other words it can be functional in all host tissue, or tissue-specific, in which case the protein will only be expressed in a certain tissue of the plant (i.e. the seeds’ reserve tissue). The currently used promoter is CaMV 35S, which is a DNA sequence derived from the cauliflower mosaic virus. It encourages the very virus’ reproduction in any part of the plant where the infection has occurred. It is a good practice to place a transcription terminator downstream of the gene to facilitate the polymerase’s detachment from the DNA filament.

The promoter-gene-terminator complex is then inserted in a T-DNA in a plasmid vector that can replicate both in Escherichia coli and in Agrobacterium tumefaciens. As all plasmids used in genetic engineering these vectors, called binary vectors, carry genes for resistance to a certain antibiotic or weedkiller to enable transformed cell selection. All the early phases of genetic manipulation are conducted on detoxificated strains of Escherichia coli, a fast multiplying bacterium that is much more flexible. Agrobacterium cells are transformed only at the end of the process.
Parts of the leaf can be infected once adequate checks have been performed on selective soil to ascertain the correct insertion of the plasmid, which carries the gene of interest into the final bacterial host.
Infected “leaf disks” are placed in strictly sterile conditions, once again in selective soil, while waiting for a callus to form, which is cell proliferation caused by the Agrobacterium (Figure 1). At this point the presence of adequate phytohormones in the soil and the typical totipotence of non differentiated vegetable cells enable the offshoot of a miniature plant to develop from cells in whose genome the T-DNA has penetrated with its resistance and study gene (Figure 2, 3 and 4). Roots will issue after a certain lapse of time (Fig. 5) and the plant can be then moved from agarized soil to normal soil (Figure 6).

 

Concerning the two trends mentioned as an example - filariasis and equine herpes virus - studies have reached this stage and analyses are currently underway to evaluate the two proteins’ level of expression in GM tobacco plants.
As whoever is experienced in laboratory research in the biological field can easily deduce, the above described study is all but free of difficulties. But we too, like other groups who are conducting such research in the world, are firmly convinced that it is a highly promising course, both as an alternative to already existing solutions and to solve some problems that have yet to find an adequate solution.

Prof.ssa Barbara Basso
CNR - Universitá degli Studi di Milano
Dipartimento di Biologia
“Luigi Gorini”- Milano