Live bacteria - key to the vaccine pill
Live bacteria are a promising oral alternative to needle-based vaccination. Rocky Cranenburgh, head of research at Cobra, describes the development of live bacterial vaccines (LBVs) designed to combat a range of diseases
Live bacteria are a promising oral alternative to needle-based vaccination. Rocky Cranenburgh, head of research at Cobra, describes the development of live bacterial vaccines (LBVs) designed to combat a range of diseases
The bacterial cells in your body outnumber your own cells by several orders of magnitude, and have a major role in digestion and the uptake of nutrients. The concept of "friendly bacteria" as probiotics to aid digestive health is well established, and recent research at the Washington University School of Medicine has demonstrated the potential role of the gut microbes in the current obesity epidemic. Cobra is exploring the use of genetically modified bacteria to deliver vaccines orally via the gut, with the aim of developing a "vaccine pill" that will replace injection using a needle.
A vaccine pill is clearly a favourable, pain-free option to a needle, especially for people with needle phobia, but it is in the developing world that this technology would have the most significant benefits. Vaccination programmes can be difficult to coordinate in sparsely populated areas with limited infrastructure, and the frequent need for more than one dose to be administered to offer protection (prime and booster doses) makes patient compliance a real problem. There is also the requirement for refrigeration of many vaccines prior to administration, and the difficulty of safe disposal of contaminated needles. A capsule containing a live bacterial vaccine (LBV) is stable at ambient temperature and much easier to distribute and administer. Veterinary vaccination is another area where LBVs, which can be administered in drinking water, have a considerable advantage over needles in terms of reducing the cost and time for vaccinating farm animals.
In addition to these factors, the LBV approach offers cheaper vaccine production over current manufacturing methods. Many vaccines are produced by recombinant DNA technology, where a gene that encodes the protein antigen is inserted into an organism that can be cultured in large fermenters (e.g. Escherichia coli, yeast or mammalian cells). These are harvested and processed to obtain a pure form of the protein. Much of the cost of vaccine production, indeed the production of any drug, is incurred in this downstream purification. The LBV approach effectively eliminates the downstream purification, as the bacteria themselves are the vaccine. They simply need to be concentrated by centrifugation, washed, freeze-dried and encapsulated. Conventional vaccines have to be formulated with a chemical called an adjuvant to stimulate a protective immune response, but LBVs are naturally immunostimulatory, so this is not required.
There are already a few orally delivered LBVs on the market, but these are all attenuated (weakened) forms of pathogenic bacteria used to vaccinate against the diseases that they cause in their pathogenic state. The Vivotif vaccine from Berna Biotech, that protects against typhoid fever, consists of a capsule containing live attenuated Salmonella enterica Typhi strain Ty21a and has had over 20 years of safe use in humans, although it is weakly immunogenic and requires three doses within a week. An improved single-dose strain, Ty800, is currently being developed by AVANT Immunotherapeutics. Other strains of Salmonella are used to vaccinate livestock against Salmonellosis by administration to their drinking water, such as the S. enteriditis and S. enterica Typhimurium strains (SALMONELLA VAC E and T respectively) developed by poultry vaccine specialist Lohmann Animal Health. Berna also market a single-dose human cholera vaccine Orochol that consists of live attenuated Vibrio cholorae strain CVD 103-HgR.
The application of live bacteria as vaccines has laid the foundation for the next stage: the delivery of genetically inserted antigens from other pathogenic species. These genetically modified LBVs are ingested and deliver their cargo of foreign antigen across the lining of the gastrointestinal tract to elicit a protective immune response. There are several species that have been intensively researched for oral recombinant antigen delivery, principally attenuated pathogens such as Salmonella spp., Shigella flexneri, Listeria monocytogenes, Vibrio cholerae and Yersinia enterocolitica. Non-pathogenic commensal species have also been studied, including Escherichia coli, Bacillus subtilis, Lactococcus spp., Lactobacillus casei and Streptococcus gordonii.
Cobra is involved in an international consortium lead by the Royal Holloway, University of London, to investigate the ability of Bacillus subtilis spores to deliver foreign antigens. B. subtilis is a transient component of the gut and the soil, and forms spores in response to nutrient depletion that are very resistant to adverse environmental conditions. These are currently taken as a probiotic to aid digestive health, but we are constructing strains with antigens from TB and malaria fused to the spore coat protein. These spores resist the stomach acid following ingestion, germinate in the small intestine and resporulate to generate spores loaded with the recombinant antigen. We are also investigating a variety of mutations that will prevent spores from germinating in the environment, as this would represent the release of a genetically modified micro-organism.
To adapt a live attenuated bacterium for recombinant vaccine delivery, the gene encoding the antigen is inserted into the bacterium, where it is expressed and the antigenic protein produced. The antigen may be coupled to a secretion signal sequence or fused with a protein that is normally exported, enabling it to be transported from the cytoplasm onto the cell wall or into the external environment. A range of antigens (mainly surface proteins) have been evaluated using LBVs in preclinical and clinical trials against diseases such as HIV/AIDS, HPV, malaria, tuberculosis, bubonic plague, anthrax and enterotoxigenic E. coli (ETEC).
There are two approaches for maintaining the foreign antigen gene in the host LBV cell. The first is to insert the antigen gene into the chromosome; invariably as a single copy (a bacterial chromosome is usually a big loop of double-stranded DNA at one copy per cell). This has the advantage of genetic stability, as it is transmitted on the chromosome as the cell divides. However, the disadvantage is that the amount of antigen produced from a single copy gene may or may not be sufficient to achieve a protective immune response. An efficient vaccine requires the protective antigen to be stably expressed at high levels within the LBV, so the second and most commonly used approach is to maintain the gene on a plasmid. A plasmid is an autonomously replicating loop of double-stranded DNA, a bit like a mini version of the chromosome. The plasmids used for protein expression are typically present in tens of copies per cell, and so will produce significantly more protein than from a single copy chromosomal gene.
The approach of delivering a DNA vaccine in an LBV is an interesting alternative to producing a conventional antigen protein. With a DNA vaccine, the plasmid needs to be present in hundreds of copies per cell, and it is this plasmid that the LBV releases into the recipient's cells, where it is expressed to produce the antigen protein. The advantages of this approach are that it allows post-translational modifications (such as glycosylation) that bacteria cannot perform, and enables the production of large, multi-epitope proteins, which may be difficult in bacteria.
Once the LBVs are genetically fitted with their recombinant antigen, either in their chromosomes or on multi-copy plasmids, they are formulated for delivery and ingested (see figure 1). The formulation (e.g. an antacid buffer or acid-resistant capsule) prevents the bacteria from being killed by the gastric hydrochloric acid (although B. subtilis spores are naturally resistant to this). Once they have made it to the small intestine, they are able to invade the lining of the gastrointestinal tract through specialised M-cells and enter lymphatic nodules called Peyer's patches. There, they are engulfed by the immune system's antigen presenting cells (APC). Salmonella, Listeria and Shigella are all able to replicate even once they have been internalised in a phagosome. Salmonella are able to modify the phagosome and remain there for a limited time, but the other two species can escape into the host cell cytoplasm. The bacteria then secrete antigens, either within the phagosome or the cytoplasm, which are displayed by MHC class I molecules on the surface of the APC, stimulating CD8+ T lymphocytes, leading to a cellular response where cytotoxic T lymphocytes are directed to attack the pathogen. Bacteria that remain in the phagosome are eventually killed when it fuses with an enzyme-packed lysozome; their contents presented via MHC class II molecules, leading to CD4+ T-lymphocytes inducing B cells to differentiate and produce antibodies against the pathogen. Salmonella are also able to induce strong mucosal (secretory IgA) antibody responses. Vibrio cholerae can elicit the production of strong systemic (serum IgG) and mucosal antibody responses despite being non-invasive.
Whether an LBV strain is growing in a nutrient broth culture, or in the stressful environment inside their mammalian host, any cells that manage to lose their plasmids have a greatly reduced metabolic burden and are able to out-compete the plasmid-containing cells. This means there is a strong selection pressure favouring plasmid loss, making plasmid maintenance a significant problem with LBVs. Traditionally, an antibiotic resistance gene is placed on the plasmid and the strain is grown in a medium containing the corresponding antibiotic, ensuring that plasmid-free cells are killed. However, antibiotics and their resistance genes are now banned for LBV applications by regulatory authorities including the FDA, so I will not discuss their other associated disadvantages further.
This introduces a requirement for antibiotic-free plasmid maintenance systems. The most common approach is to complement a mutation in an essential gene in the host by introducing a plasmid with a functioning gene. Although a logical approach, the bacteria experience a metabolic burden from expressing a considerable excess of the essential protein than is required for survival, thus reducing their fitness. In addition, the high levels of protein expressed mean that plasmids can be lost over several cell divisions without the bacteria feeling the effect until it is too late.
Another approach uses post-segregational killing mechanisms (e.g. the hok-sok system). Here the plasmid contains genes for both a toxin and an anti-toxin. The toxin is highly stable, but the anti-toxin is less stable and therefore needs to be constantly produced to counteract the toxin's effects and keep the cell alive. If the plasmid is lost, the rapid breakdown of the anti-toxin will lead to the cell being killed. However, the system does not enrich the culture with those cells that maintain the plasmid; merely kills those that have lost it. Bacteria frequently escape the killing effect of the toxin, and these cells rapidly grow to dominate the culture, leading to plasmid loss. Another disadvantage is the lack of an associated selection mechanism, often requiring post-segregational killing mechanisms to be coupled with auxotrophic complementation.
Cobra's Operator-Repressor Titration (ORT) technology removes the requirement for antibiotic resistance markers and genes for stable plasmid maintenance, but retains the advantages of being a plasmid system, i.e. high levels of antigen expression. ORT was developed for plasmid maintenance in E. coli, but has been adapted to LBVs and termed ORT-VAC. It has been demonstrated in vivo to overcome the problem of plasmid loss, and is therefore highly promising for the development of LBVs. As the selective gene is only present as a single copy on the chromosome, with only the antigen gene present on the plasmid, the metabolic pressure on the cell is significantly reduced.
In an ORT-VAC bacterium, the essential gene dapD has been engineered to be under the control of the lac operator. In the absence of an inducer such as the lactose analogue IPTG, the LacI repressor protein binds to the operator site, blocking transcription of dapD. The DapD protein stabilises the cell wall, so its absence eventually causes cell lysis. However, if ORT-VAC cells are transformed with a high copy number plasmid containing the lac operator (and the antigen of interest), LacI binds to the lac operator sequences instead of the chromosomal lac operator. The presence of plasmid thus switches on dapD expression, so cells can survive for as long as high numbers of plasmids are maintained (See figure 2).
In collaboration with the UK Defence Science and Technology Laboratory, we tested the ability of the F1 antigen of Yersinia pestis, carried by an ORT-VAC strain of Salmonella Typhimurium, to provide protection against bubonic plague in mice. Proof-of-principle results showed that after only a single dose of the vaccine, mice achieved a high level of immunity against the disease. Plasmid maintenance was also investigated in vivo, and the ORT-VAC strain was able to stably maintain a very high copy number plasmid that was rapidly lost from a conventional strain of Salmonella Typhimurium.
Cobra is currently collaborating with the US Naval Medical Reseach Center in Washington DC to evaluate their malaria DNA vaccines, using an ORT-VAC Salmonella Typhimurium strain for vaccine delivery. Cobra is also leading a consortium to continue the development of ORT-VAC using different LBV species for delivery of vaccines against HIV and influenza. Future applications could involve ORT-VAC delivering prodrugs or therapeutic agents, and the natural tropism of Salmonella for solid tumours could be used to selectively deliver anti-cancer agents.
As a contract manufacturing organisation, Cobra seeks to use ORT-VAC as a platform technology to offer customers a faster and more effective route to vaccine delivery than is currently available, simultaneously making vaccines cheaper to produce and easier to distribute and administer.