- Email: [email protected]
Biological safety concepts of genetically modified live bacterial vaccines
Vaccine 25 (2007) 5598–5605
Biological safety concepts of genetically modified live bacterial vaccines Joachim Frey ∗ Institute of Veterinary Bacteriology, Laenggassstrasse 122, CH-3001 Bern, Switzerland Received 21 June 2006; received in revised form 23 November 2006; accepted 27 November 2006 Available online 5 December 2006
Abstract Live vaccines possess the advantage of having access to induce cell-mediated and antibody-mediated immunity; thus in certain cases they are able to prevent infection, and not only disease. Furthermore, live vaccines, particularly bacterial live vaccines, are relatively cheap to produce and easy to apply. Hence they are suitable to immunize large communities or herds. The induction of both cell-mediated immunity as well as antibody-mediated immunity, which is particularly beneficial in inducing mucosal immune responses, is obtained by the vaccine-strain’s ability to colonize and multiply in the host without causing disease. For this reason, live vaccines require attenuation of virulence of the bacterium to which immunity must be induced. Traditionally attenuation was achieved simply by multiple passages of the microorganism on growth medium, in animals, eggs or cell cultures or by chemical or physical mutagenesis, which resulted in random mutations that lead to attenuation. In contrast, novel molecular methods enable the development of genetically modified organisms (GMOs) targeted to specific genes that are particularly suited to induce attenuation or to reduce undesirable effects in the tissue in which the vaccine strains can multiply and survive. Since live vaccine strains (attenuated by natural selection or genetic engineering) are potentially released into the environment by the vaccinees, safety issues concerning the medical as well as environmental aspects must be considered. These involve (i) changes in cell, tissue and host tropism, (ii) virulence of the carrier through the incorporation of foreign genes, (iii) reversion to virulence by acquisition of complementation genes, (iv) exchange of genetic information with other vaccine or wild-type strains of the carrier organism and (v) spread of undesired genes such as antibiotic resistance genes. Before live vaccines are applied, the safety issues must be thoroughly evaluated case-by-case. Safety assessment includes knowledge of the precise function and genetic location of the genes to be mutated, their genetic stability, potential reversion mechanisms, possible recombination events with dormant genes, gene transfer to other organisms as well as gene acquisition from other organisms by phage transduction, transposition or plasmid transfer and cis- or trans-complementation. For this, GMOs that are constructed with modern techniques of genetic engineering display a significant advantage over random mutagenesis derived live organisms. The selection of suitable GMO candidate strains can be made under in vitro conditions using basic knowledge on molecular mechanisms of pathogenicity of the corresponding bacterial species rather than by in vivo testing of large numbers of random mutants. This leads to a more targeted safety testing on volunteers and to a reduction in the use of animal experimentation. © 2006 Elsevier Ltd. All rights reserved. Keywords: Live vaccines; Bacteria; Genetically modified organisms; Release; Environment; Survival; Stability; Mutations; Recombination; Plasmids; Complementation; Safety
1. Introduction Vaccines have had a major impact on the improvement of animal health, animal productivity and safety in public health by reducing zoonotic diseases. Vaccines have played ∗
Tel.: +41 31 631 2414; fax: +41 31 631 2634. E-mail address: [email protected].
0264-410X/$ – see front matter © 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.vaccine.2006.11.058
an important role in the eradication of certain diseases in countries or on entire continents that have implemented vaccination programs combined with accompanying control measures such as import restrictions of animals and animal derivatives, control of animal movements as well as diagnostic and surveillance facilities [1]. Thereby efficacy and safety of the vaccines used are the two essential features. An ideal vaccine does not cause disease or negative side
J. Frey / Vaccine 25 (2007) 5598–5605
effects in the host while it is inducing an immune response that is capable to protect against the pathogen. In veterinary medicine, the vaccines should allow serological differentiating of vaccinated animals from infected animals, a feature that is most important in eradication campaigns of epizootics [2–4]. Live vaccines induce cell-mediated immunity, as well as systemic antibody-mediated immunity and hence are considered highly efficient. They are particularly efficacious to prevent from infectious respiratory or enteric pathogens that cause the most frequent severe infectious diseases and epidemics of herd animals. Furthermore, live vaccines are easy to produce and to apply to entire herds as they may be administrated orally by feed or dispensed in water. However, several attenuated microorganisms were shown to be unstable and occasionally to revert to virulence [3,5,6]. Hence, live vaccines require particular attention regarding their safety as they are generally derived from pathogens by attenuation and as they multiply and may propagate, to some extent, in the vaccinated host and in the environment. Before the introduction of recombinant gene technology methods, currently named genetic engineering, that allows the generations of genetically modified organisms (GMOs), conventional live vaccines were obtained by random mutagenesis that do not incorporate deliberate genetic modifications, followed by experimental selection from their virulent ancestors. For these processes multiple passages of the microorganism under laboratory conditions on growth medium, cell cultures or eggs, chemical or physical random mutagenesis that resulted in attenuating mutations of unknown target genes were used. Safety assessment was mainly restricted to observations of adverse reactions and stability of the vaccine in experimental animals or volunteers, demonstration of the extent of vaccine shedding, genetic stability and lot-to-lot consistency. Basic requirements for live vaccines for animals are: • contain the necessary antigenic determinants to induce protective immunity; • possess a high immunogenic potential; • be safe for administration without risk of clinical infection in the vaccinee; • be safe for humans; • induce a serological response that allows differentiation of vaccinated from infected animals; • be stable for long-term storage; • not to be contraindicated for release in the environment; • to be traceable in the animal population and in the environment. Conventional bacterial live vaccines for animals were used widely in the past or still are used extensively [7] (Table 1). Many conventional bacterial live vaccines do not meet all of the above stated requirements although they have been used worldwide since many years. Some of them have shown to be able to revert to virulence or to cause non-target effects and were withdrawn from the market. Environmental safety aspects of conventional live vaccines were mostly not considered. Prospective safety considerations of randomly
5599
Table 1 Conventional live bacterial vaccines used for animals Disease
Vaccine strain
Brucellosis of ruminants
Brucella abortus, strain 19, strain RB51 Brucella suis, strain 2 Bacillus anthracis, strain Sterne Mycobacterium paratuberculosis strain 316F Mycoplasma mycoides subsp. mycoides SC, strain T1/44 Salmonella enterica serov. Gallinarum, strain R9 Pasteurella multocida (various strains) Mannheimia (Pasteurella) haemolytica (various strains) Bordetella bronchiseptica (various strains) Clostridium perfringens
Brucellosis of swine Anthrax (bovine, ovine, equine) Johne’s disease Contagious bovine pleuropneumonia (CBPP) Avian salmonellosis Poultry cholera Cattle pasteurellosis Swine atropic rhinitis Bovine clostridiosis
attenuated live vaccines are hardly feasible, due to the largely unknown genetic background. Furthermore, environmental monitoring was generally not done as the conventional vaccines were not designed to be traced in the environment and virtually no data exist on their fate, once they leave the vaccinated individual. The current knowledge of molecular mechanisms of pathogenicity and novel techniques of genetic engineering, however, has been exploited for the construction of targeted and stable mutations. This in turn enables the development of efficacious and safe vaccines that can clearly be distinguished from the virulent pathogens [8]. Using genetic engineering, biological safety has become a conceptional issue that is considered during the genetic construction of live vaccine strains, rather than leaving the matter to chance and selecting safe mutants by testing them on experimental animals or volunteers. The most obvious strategy to obtain stable and safe attenuations in live vaccine strains that permitted to differentiate vaccinated from infected animals is the creation of defined attenuating deletion mutations by recombinant gene technology. However, new candidate live vaccine strains prepared by recombinant DNA technology have been contested by several radical opponents of recombinant DNA technology, which resulted in a broad suspicion of the public towards any products of recombinant DNA technology and in particular toward release of GMOs into the environment. Currently the national and international regulatory agencies are commissioned to control and survey proposed clinical trials and licensing of any new live vaccine taking into account medical safety and environmental safety aspects. For this purpose, vaccine strains constructed by recombinant DNA technology have the advantage over conventional vaccines that they allow initial risk assessment to be made based on the precisely introduced attenuating mutations and other known genetic modifications already under in vitro conditions prior to clinical trials. Nevertheless, live GMO vaccines should be scrutinized by the same way as live vaccines that were prepared by other means.
5600
J. Frey / Vaccine 25 (2007) 5598–5605
2. Regulatory aspects Since the introduction of medical products based on or containing genetically modified organisms (GMO) to the market, national and international regulatory agencies survey the biological safety of GMO products, to humans, animals and the environment. In contrast to inactivated or subunit vaccines, live vaccines require special attention concerning their fate in the environment, as they are disseminated as live organisms by the vaccinees. Live GMO vaccines represent a deliberate release of GMO into the environment and currently their regulatory restrictions in many countries are significantly stricter than those for the release of conventional live vaccines where environmental aspects were not specifically considered. This is in strong contrast to biological safety considerations where defined genetic modifications allow for much better control and safety assessment than random, mostly unknown mutations, in a conventionally attenuated vaccine. Many regulatory authorities request an environmental safety assessment for live GMO vaccines. This includes an assessment of potential risks of the GMO to cause harm to target or non-target organisms directly or indirectly by transferring genetic traits to other organisms or acquiring genetic traits that would make the GMO harmful. In Europe, the European Medicines Agency (EMEA) grants, in close collaboration with national authorities, the licences for GMO-containing or GMO-derived medical products for human and animal use. The EMEA bases its authorisation procedures for live GMO vaccines on current directives of the European Union which include: (a) Directives “on the contained use of genetically modified micro-organisms” No. 90/219/EEC (28 October 1991) and No. 98/81/EC (5 December 1998), (b) Directive “on the deliberate release into the environment of genetically modified organisms” No. 2001/18/EC (12 March 2001) as well as on (c) the regulation EC No. 726/2004 of the European Parliament and of the Council from 31 March 2004 laying down Community procedures for the authorisation and supervision of medicinal products for human and veterinary use and establishing a European Medicines Agency. According to EU Directive 2001/18/EC “on the deliberate release into the environment of genetically modified organisms”, recital 31, “genetically modified medicinal products that are to be placed on the market are not covered by the Directive and hence have not to be labelled as GMO”. However, an environmental risk assessment should be carried out according to the council regulation EC No. 726/2004. This regulation implemented by the EMEA also refers to medicinal products containing or consisting of GMOs while the “note for Guidance” No. 3BR1a (12 January 1995) provides the requirements for an assessment of medicinal products that contain or are derived from GMOs. The Center for Biologics Evaluation and Research (CBER) of the US Food and Drug Administration (FDA) is the regulatory agency charged with insuring the safety, purity, efficacy and potency of vaccine products for human and veterinary use in the US, providing the requirements for
assessment of GMO (and also non-GMO) derived medicinal products [9]. This includes in particular (a) the identification of potential hazards, (b) the assessment of consequences of each individual hazard, (c) the evaluation of the consequences of each individual hazard, (d) the estimation of the likelihood and consequences and (e) the selection and assignment of risk management by the introduction of appropriate control measures. In principle, such environmental risk assessments are appropriate to all medical products that contain or are based on live organisms and not exclusively to GMOs.
3. Attenuating mutations The attenuating mutations constitute the central part of the biosafety concept in constructing live vaccines. Attenuation should be sufficient to decrease or possibly eliminate undesirable vaccination reactions and signs of disease. Hence, an inherent property of the vaccine strain must be that the attenuating mutations are not reversible. Furthermore, the attenuation should be made in such a way that it permits the vaccine strain to be sufficiently invasive and to induce both a strong primary and long lasting memory immune response, but not a persistent carrier state of the vaccine [8]. Attenuations are generally made in two groups of genes, (a) genes involved in central metabolic pathway functions (housekeeping genes) such as aroA (5-enolpyruvylshikimate 3-phosphate synthase), which affects the aromatic biosynthesis pathway, cAMP receptor protein (crp) and adenylate cyclase (cya), affecting the cyclic AMP pathway [10–13] and (b) genes involved directly in virulence including toxin genes such as ctxAB (cholera toxin Vibrio cholerae) [14], as well as genes involved in the regulation of virulence genes such as phosphorylated transcriptional activator for regulation of virulence genes (phoP). Both types of attenuations or even combinations of them including attenuations in a central metabolic pathway gene plus a virulence gene [15] have been used to construct vaccine strains to protect either against homologous pathogens [14] or against heterologous pathogens by expressing particular recombinant genes in heterologous vaccination strains [8,12,16–18].
4. Attenuations by deletion of central metabolic pathway genes Deletions of genes in the central metabolic pathway such as aroA, which blocks biosynthesis of aromatic amino acids, render the bacterium unable to replicate unless cultivated on special growth medium under in vitro conditions. Hence, such organisms are unable to propagate in the environment and persist in the vaccinee for a limited time, which is sufficient to induce immune responses. Genes involved in central metabolic pathways are stably located on the chromosome and are generally not carried on or in the vicinity of mobilizable elements. Therefore, deletion mutations created in
J. Frey / Vaccine 25 (2007) 5598–5605
genes such as aroA, aroB and crp that include the deletion of complete genes or at least large segments of the corresponding genes are very stable. Reversions, even when the vaccine strain is growing in presence of large amounts of closely related wild-type bacterial species that carry their own corresponding functional genes, are most unlikely to occur. Due to their long evolution in the specific species, housekeeping genes show species specificity and relatively little horizontal gene transfer capacity. Therefore, complementation of deletion mutants by uptake of analogous genes from surrounding bacteria that are in close contact with them in the vaccinated individual or in the environment, are unlikely to occur. Furthermore, many studies have shown that deletion mutants in central metabolic pathway genes have a very short survival rate under natural conditions and disappear rapidly compared to the homologous wild-type strains. Deletion mutants of genes in the basic branch of the aromatic biosynthesis pathway have been found to be safe and suitable for vaccinations in various bacterial species such as Aeromonas hydrophila [19], Listeria monocytogenes [20], Salmonella typhimurium [10], Shigella flexneri [15], Pseudomonas aeruginosa [21], Shigella dysenteriae [22], Bordetella pertussis [23], Neisseria gonorrhoeae [24], Bacillus anthracis [25], avian pathogenic Escherichia coli [13], Salmonella abortusovis [26] and many more. A recombinant ΔaroA strain of Streptococcus equi has been developed as a safe live vaccine against strangles in horses [27]. This mutant, when applied at ≥108 colonyforming units per dose in the inner side of the upper lip, protects horses against strangles. The vaccine strain S. equi ΔaroA, TW928, was derived from virulent wild-type S. equi that was isolated from a lymph node abscess of a foal with strangles, by recombinant deletion of an essential 1 kb segment of the aroA gene. The vaccine strain was shown to be free of vector derived DNA sequences, in particular antibiotic resistance determinants [27]. This latter is a basic requirement for the release of GMOs outside of restricted confinements in many countries. The mutant was shown to require special growth medium supplemented with aromatic pathway precursors phenylalanine, tryptophan, tryosine, p-aminobenzoic acid, or 2,3-dihydroxybenzoic acid and it could not be isolated from vaccinated horses at post-mortem examinations 4 weeks post-vaccination with two submucosal applications of 109 colony forming units in the inner side of the upper lip [27]. Other characteristics of S. equi such as hemolysis, capsule biosynthesis and sugar fermentation were shown to be unchanged in the vaccine strain TW928 compared to wildtype S. equi. The vaccine strain induced high antibody titers in horses and provided nearly in all test animals full protection against clinical signs of strangles after challenge with a heterologous virulent strain of S. equi [27]. Furthermore, no S. equi could be isolated from vaccinated horses after challenge, while in non-vaccinated horses challenged with wild-type S. equi, the challenge strain could be isolated from retropharyngeal lymph nodes [27]. This revealed that the ΔaroA S. equi not only protected the animals against clinical signs of stran-
5601
gles but also against infection with wild-type S. equi. Since the vaccine strain did not propagate in vaccinated animals, it is not shed and presents no potential risk of spreading to non-target animals. S. equi ΔaroA is the first and currently the only licensed GMO bacterial live vaccine for veterinary medical applications. It has recently been introduced successfully into the market under the brand name Equilis StrepE® , Intervet, and also represents the first vaccine against S. equi and the first commercial bacterial GMO vaccine for animals to receive EU authorisation.
5. Attenuations by targeted mutagenesis of virulence genes Deletion of virulence genes in pathogens results in loss of pathogenicity. In many cases, the deletion or inactivation of a single gene encoding the primary virulence principle, such as the main toxin, leads to entirely non-pathogenic organisms. Hence, virulence genes can be used as targets for deletions in constructing safe recombinant live vaccines. However, since antibodies against the entire or against part of the main virulence attributes, e.g. the major toxin, often are essential to confer protective immunity, strategies to develop vaccines are generally aimed at the inactivation or partial deletion of toxin genes thereby leaving the antigens that are essential to induce protective antibodies within the vaccine strain. This can be either a truncated peptide of the toxin lacking the enzymatically active domain, a toxin protein with a point mutation rendering the enzymatic domain inactive or deletions of nearly the entire toxin genes leaving in the vaccine strain only the genes encoding accessory toxin subunits such as adhesions or translocation domains. This approach is valid for subunit vaccines such as tetanus and botulinum vaccines [28–31] as well as for live vaccines such as cholera vaccine [14]. However, the inactivating mutations or partial deletions in toxin genes must be carefully designed in order to avoid reversion or suppression of the attenuation by secondary mutations. A well-known example of this is the site directed deletion mutation of Glu-148 of the active site of diphtheria toxin that results in inactivation of the toxin and attenuation of the corresponding strain of Corynebacterium diphteriae. However, this attenuation can be suppressed either by substituting glutamic acid for Val-147 or by a five codon deletion, which places Glu-142 in the corresponding position to the active-site and resituates the toxic activity [32]. The modification of coding sequences to construct genetic toxoids for live vaccines must therefore be carefully designed in the light of the possibility of various reversion elements. By far the best way to design safe live vaccines that avoid reversions is by introducing of larger deletion mutations at the relevant attenuating loci such as by deleting entire genes that encode enzymatically toxic subunits. For example, deletions have been made of nearly the entire ctxA genes that encode the active part of the cholera toxin thus leaving in the vaccine strain only the genes encoding the protective
5602
J. Frey / Vaccine 25 (2007) 5598–5605
antigenic CtxB subunits of the cholera toxin, which is responsible for the docking of the receptor to the host cells [14,33]. In such mutants, reacquisition of functional virulence genes is impossible by reversion but still could occur, theoretically, by acquisition of complementing genes from bacteria in their microenvironment. The V. cholerae vaccine strain CVD 103-HgR is the first and currently only GMO registered live vaccine for human use introduced under the name Orochol® by Berna Biotech. Biosafety aspects of this orally administered live vaccine have been exhaustively studied in a risk assessment that included environmental safety, since the vaccine is known to be released into the environment [34]. This thorough risk assessment was claimed for the registration of the vaccine by the European authorities in spite the fact that the vaccine was already on the Swiss market and used before hand. The environmental risk assessment included a global and genetic characterization of the strain, the study of its natural cryptic prophages and plasmids, analysis for the absence of residual sequences from construction, environmental release, capacity to survive and establish in the environment, the assessment of potential reversion of the vaccine strain to a toxigenic state via phage promoted transfection, plasmid acquisition or activity of transposable elements and finally, study on the impact of foreign introduced genes to the vaccinee and the environment [34]. Considering the fact that no environmental risk assessment was required for licensing conventional (non-GMO) vaccines, it is questionable whether this extensive risk assessment was justified from the scientific point of view for the introduction of a GMO vaccine to the market or if it rather reflected the general political and social mistrust towards a new technology, respectively, its products the GMOs. Nevertheless it provided an excellent opportunity to demonstrate that live GMO vaccines can be designed to provide very high safety standards, which most probably cannot be reached by random mutagenesis.
6. Genetic stability An important trait of the GMO is its genetic stability over repeated generations during the production process and during the presence of the live vaccine in the vaccinee, as well as during a potential persistence in the environment. From this point of view, it must be recognised that the mutation leading to the non-pathogenic phenotype must not revert via recombination events with possible dormant genes or by reactivation of the latter. Attenuating deletion mutations should therefore be constructed, if possible, on genomic loci that are known to be stable and not to be flanked by known active mobile elements. Furthermore, the strain to be attenuated must be genetically well characterized and gene duplications must be known. Hence, for the construction of the live cholera vaccine strain V. cholerae CVD 103-HgR, 95% of both chromosomal copies of the ctxA gene, which encodes the toxic A subunit (CtxA), had to be deleted [35].
However, not only duplications of active genes but also dormant copies of genes must be considered. Studies to improve the biological safety of the currently used live vaccine strain T1/44 of Mycoplasma mycoides subsp. mycoides SC, the etiological agent of contagious bovine pleuropneumonia (CBPP), were initiated in our laboratory since the currently used non-GMO vaccine was shown to revert to pathogenicity [36]. The studies revealed that selected target genes for deletion mutations are located on large segments of the genome as tandem repeats [37]. Furthermore, several genes that could be considered as targets for attenuations are present in the form of an active copy on one of the repeats and dormant copy on the other [37,38]. This shows that profound genetic knowledge of the strain to be selected for introductions of attenuating deletions for novel live vaccines is necessary. In certain cases, therefore, double deletions will be necessary, even if only one of the duplicated genes is active, in order to avoid recombination events that could lead to reversions by complementation. Furthermore, carefully conducted studies are necessary to show that the genome of the constructed vaccine does not show genome rearrangements of genetic modifications, other than the ones purposely performed for its construction.
7. Survival and dissemination Many live vaccine strains transiently colonize the vaccinated individual, a property that favours the induction of protective immune response. Vaccine strains must be designed in a way that avoids the establishment of a carrier state. Transient excretion of the live vaccine strain from vaccinees, however, must be accepted as a normal status and should be included in environmental safety considerations. For the risk assessment, spreading of live vaccines including GMOs from for example the stool of vaccinees must be compared with the wild-type pathogens from diseased individuals and the risk of creating epidemics. Since vaccine strains are designed to multiply only to a very limited extent in the vaccinee, in contrast to pathogens, the numbers of secreted live vaccine micro-organisms is generally negligible. In an extensive environmental safety assessment for the V. cholerae live vaccine strain CVD 103-HgR, Viret et al. [34] noted infection by toxigenic V. cholerae O1 culminates with a daily release of 1012 fully pathogenic vibrios into the environment over several days. In contrast, the GMO CVD 103-HgR is only excreted by about 20–30% of the vaccinees, during a maximum of 7 days and at maximum of 2 × 104 organisms on the peak day. This is approximately 106 to 108 times less than a person affected by cholera. Considering the risk/benefit ratio derived from the use of such a vaccine, one has to consider the fact that shedding of infectious pathogens such as V. cholerae O1 in the volunteer challenge model is reduced significantly by more than 104 as compared to the non-vaccinated placebo group. Hence, the use of vaccines in endemic regions can have an indirect positive effect on
J. Frey / Vaccine 25 (2007) 5598–5605
the environmental pool of pathogenic organisms. In contrast, in a study where chicken were vaccinated orally with a live aroA− mutant of S. typhimurium and subsequently inoculated with wild-type S. typhimurium or Salmonella enteritidis, the vaccination failed from protecting the chicken to be infected with and subsequently to shed wild-type S. typhimurium or S. enteritidis, while the vaccine strain could be detected in faeces up to 26 days post-vaccination [39]. This shows that infection and subsequent shedding of the pathogen cannot always be reduced or even avoided by vaccination with a live vaccine. In this particular case, however, it must be noted that S. typhimurium and S. enteritidis are not pathogens of chicken. Hence, the chicken health is not affected by these Salmonella enterica serotpyes, but their absence in poultry would be favoured from point of view of food hygiene. In spite of generally very low excretion rates of correctly attenuated vaccine organisms, their capacity to survive, establish and disseminate in the environment, and a potential for gene transfer that could take place in the vaccinee or in the environment, are relevant biological safety issues. Therefore, live vaccine strains should be traceable in the vaccinee and in the environment, a requirement that should be addressed during the strain construction.
8. Gene transfer to other organisms Gene transfer from live vaccine strains to other organisms could occur via phages or via mobile genetic elements such as plasmids and transposons. Such elements are integral genetic parts of most parental microbial strains that are used for the design of live vaccines. In this respect, one must assess (a) whether these genetic mobile elements and prophages are actively transmitting genes to other organisms and (b) if so, what genes are most likely to be transmitted. A basic knowledge of the genes that are located on prophage, plasmids and transposons is necessary for an extensive safety assessment. Genes that should be absent or, at least not be located on mobile elements in live vaccines, are antibiotic resistance genes and major toxin genes. Antibiotic resistance can be determined either phenotypically or by genetic means as for example by using gene chips [40], which are particularly useful in detecting dormant resistance genes such as the inactive beta-lactamase genes in B. anthracis wild-type and vaccine strains [40]. The presence of toxin and virulence genes on mobile elements must be assessed for each situation separately. In the case of cholera, it must be noted that the cholera toxin genes in V. cholerae are located on prophages and are actively transmitted via phages [41] a principle that is also found in other pathogenic bacteria. In the vaccine strain CVD 103-HgR the gene of the active subunit ctxA of the cholera toxin is deleted and the resident prophage contains in addition a deletion mutation Δcep in a structural phage gene, which results in the sole production of defective, non-infectious vibrio phages from the vaccine strain that are unable to
5603
form plaques on a sensitive indicator strain [34]. Although phages and prophages are little known for most animal pathogens, genes flanking virulence attributes can be analyzed for similarity with prophages, transposons, insertion sequences or plasmid replication genes. Furthermore, plasmid analysis and plaque formation tests can be considered prior to setting up strategies to design modified live bacterial vaccines.
9. Gene transfer to the vaccine strain The risk of reversion of a deletion mutation in a vaccine strain by phage-mediated conversion or by plasmid or transposons mediated cis- or trans-complementation must be considered in spite of their low probability due to the low number and short duration of persistence of vaccine organisms in the vaccinee or in the environment. This needs a careful case-by-case analysis taking into account the presence of bacteriophages or plasmids carrying genes that can potentially complement the vaccine strain to a wild-type pathogen or even to a new pathogenic strain. For this analysis, phage exclusion, transfer mechanisms, conjugative repression as well as plasmid incompatibility must be considered. In the exemplary biosafety study of V. cholerae CVD 103-HgR live vaccine [34,42], the risk of reversion of the vaccine strain to a toxigenic strain via CTX phage-mediated conversion or via conjugation with plasmids that carry E. coli heat-labile toxin genes eltAB was studied in detail. The experimental study showed that a risk associated with a possible acquisition of a CTX phage by CVD 103-HgR is very low, probably due to exclusion from the resident prophage, and that the potentially resulting toxigenic state is only transient as the strain very rapidly returns to a non-toxigenic state [34,43]. Hence, the potential of a CTX phage-mediated conversion event to virulence of CVD 103-HgR was estimated to represent a very low risk for the use of CVD 103-HgR as a live vaccine [34]. Furthermore, CTX phage-mediated conversion may occur only in areas where V. cholerae is endemic so that the outcome of such a conversion would be negligible in terms of biosafety. Plasmid borne trans-complementation of deletion mutants is an important issue in biosafety assessment. The heat-labile toxin (LT) of enterotoxigenic E. coli (ETEC) has a mechanism of action that is essentially identical to that of cholera toxin (CT). The corresponding LT encoding gene eltAB that is present in ETEC on large conjugative plasmids, so-called Ent-plasmids, shares about 80% sequence identity to ctxAB. This implies a possible conversion of the vaccine strain to a pathogenic phenotype via conjugative acquisition of the eltAB genes. A plasmid stability study using the conjugative plasmid pCG86 revealed that the plasmid is rapidly lost in CVD 103-HgR in the absence of artificial selective pressure and could no longer be detected after 55 generations. Furthermore, the authors could demonstrate that even under artificial selective pressure, eltAB is only weakly expressed
5604
J. Frey / Vaccine 25 (2007) 5598–5605
in V. cholerae and the toxin titers of the vaccine strain carrying the E. coli plasmid were 200 times lower than that of wild-type V. cholerae [34]. The authors concluded from these experiments that Ent-plasmids could trans-complement the ΔctxA deletion in the vaccine strain but would solely result in a transient very low pathogenic state that is lost rapidly [34]. Generally such assessments considerably enhance the knowledge of biological safety of live vaccines. However, they require a broad basic knowledge of the molecular mechanisms of pathogenicity and presuppose that the attenuating mutations are precisely introduced by means of genetic engineering.
Acknowledgments I am grateful to Jean-Franc¸ois Viret, Crucell-Berna Biotech, Bern, Switzerland, and to Ruud Segers and Ton Jacobs, Intervet International Boxmeer, The Netherlands, Nikolaus Kriz, EMEA, London, UK, Karoline Dorsch, EFBS, BAFU, Bern, Switzerland, Hans-Peter Ottiger and Raphael Fricker, IVI, Mittelh¨ausern, Switzerland, for invaluable scientific advice and to Andrea Hitz and Sarah Burr, Institute of Veterinary Bacteriology, Bern, for valuable editorial help.
References 10. Impact of exogenous genes in live vaccines The introduction of foreign genes provides a broad potential to improve vaccine strains and to expand protective activity against more diseases. They also provide the vaccine strain diagnostic advantages. Foreign genes that are recombined into vaccine strains may theoretically influence the balance of the ecosystem by, for instance, providing the recombinant host new properties to persist or replicate in microenvironments where the ancestor bacterium could not establish. Safety assessments must therefore consider the security measures that were taken in the course of introducing the foreign genes, the influence such genes could play in the final a-virulence of the vaccine strain and the potential survival advantage in particular situations. A precondition to guarantee the highest possible biological safety for vaccines carrying foreign genes is a well-defined carrier strain that contains stable and efficacious attenuations and that itself acts, to a large extent, as biological safety containment.
11. Concluding remarks Modern techniques of genetic analysis and genetic engineering allow the construction of precise, well-characterized deletions and genetic modifications for the construction of GMO bacteria of known virulence and attenuation. This enables the production of safe attenuated live bacterial vaccines whose characteristics are well understood. The latter is a prerequisite for any biological safety assessment. Hence, the important feature in evaluating the biological safety of a live vaccine for the vaccinated individual, entire communities and herds and for the environment does not depend on whether the organism is a GMO or not. Rather it depends solely on the sum of scientific knowledge of the agent and the disease with which it is associated and on the precise genetic features of the vaccine strain. While GMO bacterial live vaccines are still very rare in veterinary medicine, research for new GMO bacterial live vaccines and their introduction into the marked should be favoured, considering their advantage in biological risk assessment and disease control.
[1] Desmettre P. Live bacterial vaccines for animals. Dev Biol Stand 1995;84:221–5. [2] Bowersock TL, Martin S. Vaccine delivery to animals. Adv Drug Deliv Rev 1999;38(2):167–94. [3] Zeman D, Neiger R, Nietfield J, Miskimins D, Libal M, Johnson D, et al. Systemic Pasteurella haemolytica infection as a rare sequel to avirulent live Pasteurella haemolytica vaccination in cattle. J Vet Diagn Invest 1993;5(4):555–9. [4] Henderson LM. Overview of marker vaccine and differential diagnostic test technology. Biologicals 2005;33(4):203–9. [5] Hopkins BA, Olson LD. Comparison of live avirulent PM-1 and CU fowl cholera vaccines in turkeys. Avian Dis 1997;41(2):317–25. [6] Alexanderson S. Advantages and disadvantages of using live vaccines. Risks and control measures. Acta Vet Scand Suppl 2006;90:89–100. [7] Entriken TL. Veterinary pharmaceuticals and biologicals. 12th edition Montvale, NJ, USA: Veterinary Health Care Communications— Thomson Health Care; 2002. [8] Curtiss III R. Bacterial infectious disease control by vaccine development. J Clin Invest 2002;110(8):1061–6. [9] Kenimer JG. Vaccines: current regulatory issues. Curr Opin Biotechnol 1997;8(3):357–60. [10] Hoiseth SK, Stocker BA. Aromatic-dependent Salmonella typhimurium are non-virulent and effective as live vaccines. Nature 1981;291:238–9. [11] Valentine PJ, Devore BP, Heffron F. Identification of three highly attenuated Salmonella typhimurium mutants that are more immunogenic and protective in mice than a prototypical aroA mutant. Infect Immun 1998;66(7):3378–83. [12] Dougan G, Hormaeche CE, Maskell DJ. Live oral Salmonella vaccines: potential use of attenuated strains as carriers of heterologous antigens to the immune system. Parasite Immunol 1987;9(2):151–60. [13] Kariyawasam S, Wilkie BN, Gyles CL. Construction, characterization, and evaluation of the vaccine potential of three genetically defined mutants of avian pathogenic Escherichia coli. Avian Dis 2004;48(2):287–99. [14] Levine MM, Kaper JB. Live oral vaccines against cholera: an update. Vaccine 1993;11(2):207–12. [15] Kotloff KL, Noriega F, Losonsky GA, Sztein MB, Wasserman SS, Nataro JP, et al. Safety, immunogenicity, and transmissibility in humans of CVD 1203, a live oral Shigella flexneri 2a vaccine candidate attenuated by deletions in aroA and virG. Infect Immun 1996;64(11): 4542–8. [16] Favre D, L¨udi S, Stoffel M, Frey J, Horn MP, Dietrich G, et al. Expression of enterotoxigenic Escherichia coli colonization factors in Vibrio cholerae. Vaccine 2006;24(20):4354–68. [17] Curtiss III R, Woodrow GC, Levine MM, editors. New generation vaccines. NY, USA: Marcel Dekker, Inc.; 1990. p. 161–88. Attenuated Salmonella strains as live vectors for the expression of foreign antigens. [18] Ward SJ, Douce G, Figueiredo D, Dougan G, Wren BW. Immunogenicity of a Salmonella typhimurium aroA aroD vaccine expressing
J. Frey / Vaccine 25 (2007) 5598–5605
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29] [30]
a nontoxic domain of Clostridium difficile toxin A. Infect Immun 1999;67(5):2145–52. Vivas J, Carracedo B, Ria˜no J, Razquin BE, L´opez-Fierro P, Acosta F, et al. Behavior of an Aeromonas hydrophila aroA live vaccine in water microcosms. Appl Environ Microbiol 2004;70(5):2702–8. Stritzker J, Janda J, Schoen C, Taupp M, Pilgrim S, Gentschev I, et al. Growth, virulence, and immunogenicity of Listeria monocytogenes aro mutants. Infect Immun 2004;72(10):5622–9. Priebe GP, Meluleni GJ, Coleman FT, Goldberg JB, Pier GB. Protection against fatal Pseudomonas aeruginosa pneumonia in mice after nasal immunization with a live, attenuated aroA deletion mutant. Infect Immun 2003;71(3):1453–61. Walker JC, Verma NK. Cloning and characterisation of the aroA and aroD genes of Shigella dysenteriae type 1. Microbiol Immunol 1997;41(10):809–13. Roberts M, Maskell D, Novotny P, Dougan G. Construction and characterization in vivo of Bordetella pertussis aroA mutants. Infect Immun 1990;58(3):732–9. Chamberlain LM, Strugnell R, Dougan G, Hormaeche CE, Demarco de Hormaeche R. Neisseria gonorrhoeae strain MS11 harbouring a mutation in gene aroA is attenuated and immunogenic. Microb Pathog 1993;15(1):51–63. Ivins BE, Welkos SL, Knudson GB, Little SF. Immunization against anthrax with aromatic compound-dependent (Aro-) mutants of Bacillus anthracis and with recombinant strains of Bacillus subtilis that produce anthrax protective antigen. Infect Immun 1990;58(2):303–8. Uzzau S, Marogna G, Leori GS, Curtiss III R, Schianchi G, Stocker BA, et al. Virulence attenuation and live vaccine potential of aroA, crp cdt cya, and plasmid-cured mutants of Salmonella enterica serovar Abortusovis in mice and sheep. Infect Immun 2005;73(7): 4302–8. Jacobs AA, Goovaerts D, Nuijten PJ, Theelen RP, Hartford OM, Foster TJ. Investigations towards an efficacious and safe strangles vaccine: submucosal vaccination with a live attenuated Streptococcus equi. Vet Rec 2000;147(20):563–7. Clayton MA, Clayton JM, Brown DR, Middlebrook JL. Protective vaccination with a recombinant fragment of Clostridium botulinum neurotoxin serotype A expressed from a synthetic gene in Escherichia coli. Infect Immun 1995;63(7):2738–42. Byrne MP, Smith LA. Development of vaccines for prevention of botulism. Biochimie 2000;82(9–10):955–66. Tavallaie M, Chenal A, Gillet D, Pereira Y, Manich M, Gibert M, et al. Interaction between the two subdomains of the C-terminal part of
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41] [42]
[43]
5605
the botulinum neurotoxin A is essential for the generation of protective antibodies. FEBS Lett 2004;572(1–3):299–306. Fairweather NF, Lyness VA, Maskell DJ. Immunization of mice against tetanus with fragments of tetanus toxin synthesized in Escherichia coli. Infect Immun 1987;55(11):2541–5. Killeen KP, Escuyer V, Mekalanos JJ, Collier RJ. Reversion of recombinant toxoids: mutations in diphtheria toxin that partially compensate for active-site deletions. Proc Natl Acad Sci USA 1992;89(13):6207–9. Mekalanos JJ, Swartz DJ, Pearson GD, Harford N, Groyne F, de Wilde M. Cholera toxin genes: nucleotide sequence, deletion analysis and vaccine development. Nature 1983;306(5943):551–7. Viret JF, Dietrich G, Favre D. Biosafety aspects of the recombinant live oral Vibrio cholerae vaccine strain CVD 103-HgR. Vaccine 2004;22(19):2457–69. Kaper JB, Lockman H, Baldini MM, Levine MM. Recombinant nontoxinogenic Vibrio cholerae strains as attenuated cholera vaccine candidates. Nature 1984;308(5960):655–8. Mbulu RS, Tjipura-Zaire G, Lelli R, Frey J, Pilo P, Vilei EM, et al. Contagious bovine pleuropneumonia (CBPP) caused by vaccine strain T1/44 of Mycoplasma mycoides subsp. mycoides SC. Vet Microbiol 2004;98(3–4):229–34. Westberg J, Persson A, Holmberg A, Goesmann A, Lundeberg J, Johansson KE, et al. The genome sequence of Mycoplasma mycoides subsp. mycoides SC type strain PG1T , the causative agent of Contagious Bovine Pleuropneumonia (CBPP). Genome Res 2004;14(2):221–7. Bischof DF, Vilei EM, Frey J. Genomic differences between type strain PG1 and field strains of Mycoplasma mycoides subsp. mycoides smallcolony type. Genomics 2006;88(5):633–41. Tan S, Gyles CL, Wilkie BN. Evaluation of an aroA mutant Salmonella typhimurium vaccine in chickens using modified semisolid Rappaport Vassiliadis medium to monitor faecal shedding. Vet Microbiol 1997;54(3–4):247–54. Perreten V, Vorlet-Fawer L, Slickers P, Ehricht R, Kuhnert P, Frey J. Microarray-based detection of 90 antibiotic resistance genes of Grampositive bacteria. J Clin Microbiol 2005;43(5):2291–302. Waldor MK, Mekalanos JJ. Lysogenic conversion by a filamentous phage encoding cholera toxin. Science 1996;272(5270):1910–4. Favre D, Viret JF. Biosafety evaluation of recombinant live oral bacterial vaccines in the context of European regulation. Vaccine 2006;24(18):3856–64. Favre D, Struck M-M, Stanley J, Cryz Jr SJ, Viret JF. Further molecular characterization and stability of the live oral attenuated cholera vaccine strain CVD 103-HgR. Vaccine 1996;14(6):526–31.
Biological safety concepts of genetically modified live bacterial vaccines Joachim Frey ∗ Institute of Veterinary Bacteriology, Laenggassstrasse 122, CH-3001 Bern, Switzerland Received 21 June 2006; received in revised form 23 November 2006; accepted 27 November 2006 Available online 5 December 2006
Abstract Live vaccines possess the advantage of having access to induce cell-mediated and antibody-mediated immunity; thus in certain cases they are able to prevent infection, and not only disease. Furthermore, live vaccines, particularly bacterial live vaccines, are relatively cheap to produce and easy to apply. Hence they are suitable to immunize large communities or herds. The induction of both cell-mediated immunity as well as antibody-mediated immunity, which is particularly beneficial in inducing mucosal immune responses, is obtained by the vaccine-strain’s ability to colonize and multiply in the host without causing disease. For this reason, live vaccines require attenuation of virulence of the bacterium to which immunity must be induced. Traditionally attenuation was achieved simply by multiple passages of the microorganism on growth medium, in animals, eggs or cell cultures or by chemical or physical mutagenesis, which resulted in random mutations that lead to attenuation. In contrast, novel molecular methods enable the development of genetically modified organisms (GMOs) targeted to specific genes that are particularly suited to induce attenuation or to reduce undesirable effects in the tissue in which the vaccine strains can multiply and survive. Since live vaccine strains (attenuated by natural selection or genetic engineering) are potentially released into the environment by the vaccinees, safety issues concerning the medical as well as environmental aspects must be considered. These involve (i) changes in cell, tissue and host tropism, (ii) virulence of the carrier through the incorporation of foreign genes, (iii) reversion to virulence by acquisition of complementation genes, (iv) exchange of genetic information with other vaccine or wild-type strains of the carrier organism and (v) spread of undesired genes such as antibiotic resistance genes. Before live vaccines are applied, the safety issues must be thoroughly evaluated case-by-case. Safety assessment includes knowledge of the precise function and genetic location of the genes to be mutated, their genetic stability, potential reversion mechanisms, possible recombination events with dormant genes, gene transfer to other organisms as well as gene acquisition from other organisms by phage transduction, transposition or plasmid transfer and cis- or trans-complementation. For this, GMOs that are constructed with modern techniques of genetic engineering display a significant advantage over random mutagenesis derived live organisms. The selection of suitable GMO candidate strains can be made under in vitro conditions using basic knowledge on molecular mechanisms of pathogenicity of the corresponding bacterial species rather than by in vivo testing of large numbers of random mutants. This leads to a more targeted safety testing on volunteers and to a reduction in the use of animal experimentation. © 2006 Elsevier Ltd. All rights reserved. Keywords: Live vaccines; Bacteria; Genetically modified organisms; Release; Environment; Survival; Stability; Mutations; Recombination; Plasmids; Complementation; Safety
1. Introduction Vaccines have had a major impact on the improvement of animal health, animal productivity and safety in public health by reducing zoonotic diseases. Vaccines have played ∗
Tel.: +41 31 631 2414; fax: +41 31 631 2634. E-mail address: [email protected].
0264-410X/$ – see front matter © 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.vaccine.2006.11.058
an important role in the eradication of certain diseases in countries or on entire continents that have implemented vaccination programs combined with accompanying control measures such as import restrictions of animals and animal derivatives, control of animal movements as well as diagnostic and surveillance facilities [1]. Thereby efficacy and safety of the vaccines used are the two essential features. An ideal vaccine does not cause disease or negative side
J. Frey / Vaccine 25 (2007) 5598–5605
effects in the host while it is inducing an immune response that is capable to protect against the pathogen. In veterinary medicine, the vaccines should allow serological differentiating of vaccinated animals from infected animals, a feature that is most important in eradication campaigns of epizootics [2–4]. Live vaccines induce cell-mediated immunity, as well as systemic antibody-mediated immunity and hence are considered highly efficient. They are particularly efficacious to prevent from infectious respiratory or enteric pathogens that cause the most frequent severe infectious diseases and epidemics of herd animals. Furthermore, live vaccines are easy to produce and to apply to entire herds as they may be administrated orally by feed or dispensed in water. However, several attenuated microorganisms were shown to be unstable and occasionally to revert to virulence [3,5,6]. Hence, live vaccines require particular attention regarding their safety as they are generally derived from pathogens by attenuation and as they multiply and may propagate, to some extent, in the vaccinated host and in the environment. Before the introduction of recombinant gene technology methods, currently named genetic engineering, that allows the generations of genetically modified organisms (GMOs), conventional live vaccines were obtained by random mutagenesis that do not incorporate deliberate genetic modifications, followed by experimental selection from their virulent ancestors. For these processes multiple passages of the microorganism under laboratory conditions on growth medium, cell cultures or eggs, chemical or physical random mutagenesis that resulted in attenuating mutations of unknown target genes were used. Safety assessment was mainly restricted to observations of adverse reactions and stability of the vaccine in experimental animals or volunteers, demonstration of the extent of vaccine shedding, genetic stability and lot-to-lot consistency. Basic requirements for live vaccines for animals are: • contain the necessary antigenic determinants to induce protective immunity; • possess a high immunogenic potential; • be safe for administration without risk of clinical infection in the vaccinee; • be safe for humans; • induce a serological response that allows differentiation of vaccinated from infected animals; • be stable for long-term storage; • not to be contraindicated for release in the environment; • to be traceable in the animal population and in the environment. Conventional bacterial live vaccines for animals were used widely in the past or still are used extensively [7] (Table 1). Many conventional bacterial live vaccines do not meet all of the above stated requirements although they have been used worldwide since many years. Some of them have shown to be able to revert to virulence or to cause non-target effects and were withdrawn from the market. Environmental safety aspects of conventional live vaccines were mostly not considered. Prospective safety considerations of randomly
5599
Table 1 Conventional live bacterial vaccines used for animals Disease
Vaccine strain
Brucellosis of ruminants
Brucella abortus, strain 19, strain RB51 Brucella suis, strain 2 Bacillus anthracis, strain Sterne Mycobacterium paratuberculosis strain 316F Mycoplasma mycoides subsp. mycoides SC, strain T1/44 Salmonella enterica serov. Gallinarum, strain R9 Pasteurella multocida (various strains) Mannheimia (Pasteurella) haemolytica (various strains) Bordetella bronchiseptica (various strains) Clostridium perfringens
Brucellosis of swine Anthrax (bovine, ovine, equine) Johne’s disease Contagious bovine pleuropneumonia (CBPP) Avian salmonellosis Poultry cholera Cattle pasteurellosis Swine atropic rhinitis Bovine clostridiosis
attenuated live vaccines are hardly feasible, due to the largely unknown genetic background. Furthermore, environmental monitoring was generally not done as the conventional vaccines were not designed to be traced in the environment and virtually no data exist on their fate, once they leave the vaccinated individual. The current knowledge of molecular mechanisms of pathogenicity and novel techniques of genetic engineering, however, has been exploited for the construction of targeted and stable mutations. This in turn enables the development of efficacious and safe vaccines that can clearly be distinguished from the virulent pathogens [8]. Using genetic engineering, biological safety has become a conceptional issue that is considered during the genetic construction of live vaccine strains, rather than leaving the matter to chance and selecting safe mutants by testing them on experimental animals or volunteers. The most obvious strategy to obtain stable and safe attenuations in live vaccine strains that permitted to differentiate vaccinated from infected animals is the creation of defined attenuating deletion mutations by recombinant gene technology. However, new candidate live vaccine strains prepared by recombinant DNA technology have been contested by several radical opponents of recombinant DNA technology, which resulted in a broad suspicion of the public towards any products of recombinant DNA technology and in particular toward release of GMOs into the environment. Currently the national and international regulatory agencies are commissioned to control and survey proposed clinical trials and licensing of any new live vaccine taking into account medical safety and environmental safety aspects. For this purpose, vaccine strains constructed by recombinant DNA technology have the advantage over conventional vaccines that they allow initial risk assessment to be made based on the precisely introduced attenuating mutations and other known genetic modifications already under in vitro conditions prior to clinical trials. Nevertheless, live GMO vaccines should be scrutinized by the same way as live vaccines that were prepared by other means.
5600
J. Frey / Vaccine 25 (2007) 5598–5605
2. Regulatory aspects Since the introduction of medical products based on or containing genetically modified organisms (GMO) to the market, national and international regulatory agencies survey the biological safety of GMO products, to humans, animals and the environment. In contrast to inactivated or subunit vaccines, live vaccines require special attention concerning their fate in the environment, as they are disseminated as live organisms by the vaccinees. Live GMO vaccines represent a deliberate release of GMO into the environment and currently their regulatory restrictions in many countries are significantly stricter than those for the release of conventional live vaccines where environmental aspects were not specifically considered. This is in strong contrast to biological safety considerations where defined genetic modifications allow for much better control and safety assessment than random, mostly unknown mutations, in a conventionally attenuated vaccine. Many regulatory authorities request an environmental safety assessment for live GMO vaccines. This includes an assessment of potential risks of the GMO to cause harm to target or non-target organisms directly or indirectly by transferring genetic traits to other organisms or acquiring genetic traits that would make the GMO harmful. In Europe, the European Medicines Agency (EMEA) grants, in close collaboration with national authorities, the licences for GMO-containing or GMO-derived medical products for human and animal use. The EMEA bases its authorisation procedures for live GMO vaccines on current directives of the European Union which include: (a) Directives “on the contained use of genetically modified micro-organisms” No. 90/219/EEC (28 October 1991) and No. 98/81/EC (5 December 1998), (b) Directive “on the deliberate release into the environment of genetically modified organisms” No. 2001/18/EC (12 March 2001) as well as on (c) the regulation EC No. 726/2004 of the European Parliament and of the Council from 31 March 2004 laying down Community procedures for the authorisation and supervision of medicinal products for human and veterinary use and establishing a European Medicines Agency. According to EU Directive 2001/18/EC “on the deliberate release into the environment of genetically modified organisms”, recital 31, “genetically modified medicinal products that are to be placed on the market are not covered by the Directive and hence have not to be labelled as GMO”. However, an environmental risk assessment should be carried out according to the council regulation EC No. 726/2004. This regulation implemented by the EMEA also refers to medicinal products containing or consisting of GMOs while the “note for Guidance” No. 3BR1a (12 January 1995) provides the requirements for an assessment of medicinal products that contain or are derived from GMOs. The Center for Biologics Evaluation and Research (CBER) of the US Food and Drug Administration (FDA) is the regulatory agency charged with insuring the safety, purity, efficacy and potency of vaccine products for human and veterinary use in the US, providing the requirements for
assessment of GMO (and also non-GMO) derived medicinal products [9]. This includes in particular (a) the identification of potential hazards, (b) the assessment of consequences of each individual hazard, (c) the evaluation of the consequences of each individual hazard, (d) the estimation of the likelihood and consequences and (e) the selection and assignment of risk management by the introduction of appropriate control measures. In principle, such environmental risk assessments are appropriate to all medical products that contain or are based on live organisms and not exclusively to GMOs.
3. Attenuating mutations The attenuating mutations constitute the central part of the biosafety concept in constructing live vaccines. Attenuation should be sufficient to decrease or possibly eliminate undesirable vaccination reactions and signs of disease. Hence, an inherent property of the vaccine strain must be that the attenuating mutations are not reversible. Furthermore, the attenuation should be made in such a way that it permits the vaccine strain to be sufficiently invasive and to induce both a strong primary and long lasting memory immune response, but not a persistent carrier state of the vaccine [8]. Attenuations are generally made in two groups of genes, (a) genes involved in central metabolic pathway functions (housekeeping genes) such as aroA (5-enolpyruvylshikimate 3-phosphate synthase), which affects the aromatic biosynthesis pathway, cAMP receptor protein (crp) and adenylate cyclase (cya), affecting the cyclic AMP pathway [10–13] and (b) genes involved directly in virulence including toxin genes such as ctxAB (cholera toxin Vibrio cholerae) [14], as well as genes involved in the regulation of virulence genes such as phosphorylated transcriptional activator for regulation of virulence genes (phoP). Both types of attenuations or even combinations of them including attenuations in a central metabolic pathway gene plus a virulence gene [15] have been used to construct vaccine strains to protect either against homologous pathogens [14] or against heterologous pathogens by expressing particular recombinant genes in heterologous vaccination strains [8,12,16–18].
4. Attenuations by deletion of central metabolic pathway genes Deletions of genes in the central metabolic pathway such as aroA, which blocks biosynthesis of aromatic amino acids, render the bacterium unable to replicate unless cultivated on special growth medium under in vitro conditions. Hence, such organisms are unable to propagate in the environment and persist in the vaccinee for a limited time, which is sufficient to induce immune responses. Genes involved in central metabolic pathways are stably located on the chromosome and are generally not carried on or in the vicinity of mobilizable elements. Therefore, deletion mutations created in
J. Frey / Vaccine 25 (2007) 5598–5605
genes such as aroA, aroB and crp that include the deletion of complete genes or at least large segments of the corresponding genes are very stable. Reversions, even when the vaccine strain is growing in presence of large amounts of closely related wild-type bacterial species that carry their own corresponding functional genes, are most unlikely to occur. Due to their long evolution in the specific species, housekeeping genes show species specificity and relatively little horizontal gene transfer capacity. Therefore, complementation of deletion mutants by uptake of analogous genes from surrounding bacteria that are in close contact with them in the vaccinated individual or in the environment, are unlikely to occur. Furthermore, many studies have shown that deletion mutants in central metabolic pathway genes have a very short survival rate under natural conditions and disappear rapidly compared to the homologous wild-type strains. Deletion mutants of genes in the basic branch of the aromatic biosynthesis pathway have been found to be safe and suitable for vaccinations in various bacterial species such as Aeromonas hydrophila [19], Listeria monocytogenes [20], Salmonella typhimurium [10], Shigella flexneri [15], Pseudomonas aeruginosa [21], Shigella dysenteriae [22], Bordetella pertussis [23], Neisseria gonorrhoeae [24], Bacillus anthracis [25], avian pathogenic Escherichia coli [13], Salmonella abortusovis [26] and many more. A recombinant ΔaroA strain of Streptococcus equi has been developed as a safe live vaccine against strangles in horses [27]. This mutant, when applied at ≥108 colonyforming units per dose in the inner side of the upper lip, protects horses against strangles. The vaccine strain S. equi ΔaroA, TW928, was derived from virulent wild-type S. equi that was isolated from a lymph node abscess of a foal with strangles, by recombinant deletion of an essential 1 kb segment of the aroA gene. The vaccine strain was shown to be free of vector derived DNA sequences, in particular antibiotic resistance determinants [27]. This latter is a basic requirement for the release of GMOs outside of restricted confinements in many countries. The mutant was shown to require special growth medium supplemented with aromatic pathway precursors phenylalanine, tryptophan, tryosine, p-aminobenzoic acid, or 2,3-dihydroxybenzoic acid and it could not be isolated from vaccinated horses at post-mortem examinations 4 weeks post-vaccination with two submucosal applications of 109 colony forming units in the inner side of the upper lip [27]. Other characteristics of S. equi such as hemolysis, capsule biosynthesis and sugar fermentation were shown to be unchanged in the vaccine strain TW928 compared to wildtype S. equi. The vaccine strain induced high antibody titers in horses and provided nearly in all test animals full protection against clinical signs of strangles after challenge with a heterologous virulent strain of S. equi [27]. Furthermore, no S. equi could be isolated from vaccinated horses after challenge, while in non-vaccinated horses challenged with wild-type S. equi, the challenge strain could be isolated from retropharyngeal lymph nodes [27]. This revealed that the ΔaroA S. equi not only protected the animals against clinical signs of stran-
5601
gles but also against infection with wild-type S. equi. Since the vaccine strain did not propagate in vaccinated animals, it is not shed and presents no potential risk of spreading to non-target animals. S. equi ΔaroA is the first and currently the only licensed GMO bacterial live vaccine for veterinary medical applications. It has recently been introduced successfully into the market under the brand name Equilis StrepE® , Intervet, and also represents the first vaccine against S. equi and the first commercial bacterial GMO vaccine for animals to receive EU authorisation.
5. Attenuations by targeted mutagenesis of virulence genes Deletion of virulence genes in pathogens results in loss of pathogenicity. In many cases, the deletion or inactivation of a single gene encoding the primary virulence principle, such as the main toxin, leads to entirely non-pathogenic organisms. Hence, virulence genes can be used as targets for deletions in constructing safe recombinant live vaccines. However, since antibodies against the entire or against part of the main virulence attributes, e.g. the major toxin, often are essential to confer protective immunity, strategies to develop vaccines are generally aimed at the inactivation or partial deletion of toxin genes thereby leaving the antigens that are essential to induce protective antibodies within the vaccine strain. This can be either a truncated peptide of the toxin lacking the enzymatically active domain, a toxin protein with a point mutation rendering the enzymatic domain inactive or deletions of nearly the entire toxin genes leaving in the vaccine strain only the genes encoding accessory toxin subunits such as adhesions or translocation domains. This approach is valid for subunit vaccines such as tetanus and botulinum vaccines [28–31] as well as for live vaccines such as cholera vaccine [14]. However, the inactivating mutations or partial deletions in toxin genes must be carefully designed in order to avoid reversion or suppression of the attenuation by secondary mutations. A well-known example of this is the site directed deletion mutation of Glu-148 of the active site of diphtheria toxin that results in inactivation of the toxin and attenuation of the corresponding strain of Corynebacterium diphteriae. However, this attenuation can be suppressed either by substituting glutamic acid for Val-147 or by a five codon deletion, which places Glu-142 in the corresponding position to the active-site and resituates the toxic activity [32]. The modification of coding sequences to construct genetic toxoids for live vaccines must therefore be carefully designed in the light of the possibility of various reversion elements. By far the best way to design safe live vaccines that avoid reversions is by introducing of larger deletion mutations at the relevant attenuating loci such as by deleting entire genes that encode enzymatically toxic subunits. For example, deletions have been made of nearly the entire ctxA genes that encode the active part of the cholera toxin thus leaving in the vaccine strain only the genes encoding the protective
5602
J. Frey / Vaccine 25 (2007) 5598–5605
antigenic CtxB subunits of the cholera toxin, which is responsible for the docking of the receptor to the host cells [14,33]. In such mutants, reacquisition of functional virulence genes is impossible by reversion but still could occur, theoretically, by acquisition of complementing genes from bacteria in their microenvironment. The V. cholerae vaccine strain CVD 103-HgR is the first and currently only GMO registered live vaccine for human use introduced under the name Orochol® by Berna Biotech. Biosafety aspects of this orally administered live vaccine have been exhaustively studied in a risk assessment that included environmental safety, since the vaccine is known to be released into the environment [34]. This thorough risk assessment was claimed for the registration of the vaccine by the European authorities in spite the fact that the vaccine was already on the Swiss market and used before hand. The environmental risk assessment included a global and genetic characterization of the strain, the study of its natural cryptic prophages and plasmids, analysis for the absence of residual sequences from construction, environmental release, capacity to survive and establish in the environment, the assessment of potential reversion of the vaccine strain to a toxigenic state via phage promoted transfection, plasmid acquisition or activity of transposable elements and finally, study on the impact of foreign introduced genes to the vaccinee and the environment [34]. Considering the fact that no environmental risk assessment was required for licensing conventional (non-GMO) vaccines, it is questionable whether this extensive risk assessment was justified from the scientific point of view for the introduction of a GMO vaccine to the market or if it rather reflected the general political and social mistrust towards a new technology, respectively, its products the GMOs. Nevertheless it provided an excellent opportunity to demonstrate that live GMO vaccines can be designed to provide very high safety standards, which most probably cannot be reached by random mutagenesis.
6. Genetic stability An important trait of the GMO is its genetic stability over repeated generations during the production process and during the presence of the live vaccine in the vaccinee, as well as during a potential persistence in the environment. From this point of view, it must be recognised that the mutation leading to the non-pathogenic phenotype must not revert via recombination events with possible dormant genes or by reactivation of the latter. Attenuating deletion mutations should therefore be constructed, if possible, on genomic loci that are known to be stable and not to be flanked by known active mobile elements. Furthermore, the strain to be attenuated must be genetically well characterized and gene duplications must be known. Hence, for the construction of the live cholera vaccine strain V. cholerae CVD 103-HgR, 95% of both chromosomal copies of the ctxA gene, which encodes the toxic A subunit (CtxA), had to be deleted [35].
However, not only duplications of active genes but also dormant copies of genes must be considered. Studies to improve the biological safety of the currently used live vaccine strain T1/44 of Mycoplasma mycoides subsp. mycoides SC, the etiological agent of contagious bovine pleuropneumonia (CBPP), were initiated in our laboratory since the currently used non-GMO vaccine was shown to revert to pathogenicity [36]. The studies revealed that selected target genes for deletion mutations are located on large segments of the genome as tandem repeats [37]. Furthermore, several genes that could be considered as targets for attenuations are present in the form of an active copy on one of the repeats and dormant copy on the other [37,38]. This shows that profound genetic knowledge of the strain to be selected for introductions of attenuating deletions for novel live vaccines is necessary. In certain cases, therefore, double deletions will be necessary, even if only one of the duplicated genes is active, in order to avoid recombination events that could lead to reversions by complementation. Furthermore, carefully conducted studies are necessary to show that the genome of the constructed vaccine does not show genome rearrangements of genetic modifications, other than the ones purposely performed for its construction.
7. Survival and dissemination Many live vaccine strains transiently colonize the vaccinated individual, a property that favours the induction of protective immune response. Vaccine strains must be designed in a way that avoids the establishment of a carrier state. Transient excretion of the live vaccine strain from vaccinees, however, must be accepted as a normal status and should be included in environmental safety considerations. For the risk assessment, spreading of live vaccines including GMOs from for example the stool of vaccinees must be compared with the wild-type pathogens from diseased individuals and the risk of creating epidemics. Since vaccine strains are designed to multiply only to a very limited extent in the vaccinee, in contrast to pathogens, the numbers of secreted live vaccine micro-organisms is generally negligible. In an extensive environmental safety assessment for the V. cholerae live vaccine strain CVD 103-HgR, Viret et al. [34] noted infection by toxigenic V. cholerae O1 culminates with a daily release of 1012 fully pathogenic vibrios into the environment over several days. In contrast, the GMO CVD 103-HgR is only excreted by about 20–30% of the vaccinees, during a maximum of 7 days and at maximum of 2 × 104 organisms on the peak day. This is approximately 106 to 108 times less than a person affected by cholera. Considering the risk/benefit ratio derived from the use of such a vaccine, one has to consider the fact that shedding of infectious pathogens such as V. cholerae O1 in the volunteer challenge model is reduced significantly by more than 104 as compared to the non-vaccinated placebo group. Hence, the use of vaccines in endemic regions can have an indirect positive effect on
J. Frey / Vaccine 25 (2007) 5598–5605
the environmental pool of pathogenic organisms. In contrast, in a study where chicken were vaccinated orally with a live aroA− mutant of S. typhimurium and subsequently inoculated with wild-type S. typhimurium or Salmonella enteritidis, the vaccination failed from protecting the chicken to be infected with and subsequently to shed wild-type S. typhimurium or S. enteritidis, while the vaccine strain could be detected in faeces up to 26 days post-vaccination [39]. This shows that infection and subsequent shedding of the pathogen cannot always be reduced or even avoided by vaccination with a live vaccine. In this particular case, however, it must be noted that S. typhimurium and S. enteritidis are not pathogens of chicken. Hence, the chicken health is not affected by these Salmonella enterica serotpyes, but their absence in poultry would be favoured from point of view of food hygiene. In spite of generally very low excretion rates of correctly attenuated vaccine organisms, their capacity to survive, establish and disseminate in the environment, and a potential for gene transfer that could take place in the vaccinee or in the environment, are relevant biological safety issues. Therefore, live vaccine strains should be traceable in the vaccinee and in the environment, a requirement that should be addressed during the strain construction.
8. Gene transfer to other organisms Gene transfer from live vaccine strains to other organisms could occur via phages or via mobile genetic elements such as plasmids and transposons. Such elements are integral genetic parts of most parental microbial strains that are used for the design of live vaccines. In this respect, one must assess (a) whether these genetic mobile elements and prophages are actively transmitting genes to other organisms and (b) if so, what genes are most likely to be transmitted. A basic knowledge of the genes that are located on prophage, plasmids and transposons is necessary for an extensive safety assessment. Genes that should be absent or, at least not be located on mobile elements in live vaccines, are antibiotic resistance genes and major toxin genes. Antibiotic resistance can be determined either phenotypically or by genetic means as for example by using gene chips [40], which are particularly useful in detecting dormant resistance genes such as the inactive beta-lactamase genes in B. anthracis wild-type and vaccine strains [40]. The presence of toxin and virulence genes on mobile elements must be assessed for each situation separately. In the case of cholera, it must be noted that the cholera toxin genes in V. cholerae are located on prophages and are actively transmitted via phages [41] a principle that is also found in other pathogenic bacteria. In the vaccine strain CVD 103-HgR the gene of the active subunit ctxA of the cholera toxin is deleted and the resident prophage contains in addition a deletion mutation Δcep in a structural phage gene, which results in the sole production of defective, non-infectious vibrio phages from the vaccine strain that are unable to
5603
form plaques on a sensitive indicator strain [34]. Although phages and prophages are little known for most animal pathogens, genes flanking virulence attributes can be analyzed for similarity with prophages, transposons, insertion sequences or plasmid replication genes. Furthermore, plasmid analysis and plaque formation tests can be considered prior to setting up strategies to design modified live bacterial vaccines.
9. Gene transfer to the vaccine strain The risk of reversion of a deletion mutation in a vaccine strain by phage-mediated conversion or by plasmid or transposons mediated cis- or trans-complementation must be considered in spite of their low probability due to the low number and short duration of persistence of vaccine organisms in the vaccinee or in the environment. This needs a careful case-by-case analysis taking into account the presence of bacteriophages or plasmids carrying genes that can potentially complement the vaccine strain to a wild-type pathogen or even to a new pathogenic strain. For this analysis, phage exclusion, transfer mechanisms, conjugative repression as well as plasmid incompatibility must be considered. In the exemplary biosafety study of V. cholerae CVD 103-HgR live vaccine [34,42], the risk of reversion of the vaccine strain to a toxigenic strain via CTX phage-mediated conversion or via conjugation with plasmids that carry E. coli heat-labile toxin genes eltAB was studied in detail. The experimental study showed that a risk associated with a possible acquisition of a CTX phage by CVD 103-HgR is very low, probably due to exclusion from the resident prophage, and that the potentially resulting toxigenic state is only transient as the strain very rapidly returns to a non-toxigenic state [34,43]. Hence, the potential of a CTX phage-mediated conversion event to virulence of CVD 103-HgR was estimated to represent a very low risk for the use of CVD 103-HgR as a live vaccine [34]. Furthermore, CTX phage-mediated conversion may occur only in areas where V. cholerae is endemic so that the outcome of such a conversion would be negligible in terms of biosafety. Plasmid borne trans-complementation of deletion mutants is an important issue in biosafety assessment. The heat-labile toxin (LT) of enterotoxigenic E. coli (ETEC) has a mechanism of action that is essentially identical to that of cholera toxin (CT). The corresponding LT encoding gene eltAB that is present in ETEC on large conjugative plasmids, so-called Ent-plasmids, shares about 80% sequence identity to ctxAB. This implies a possible conversion of the vaccine strain to a pathogenic phenotype via conjugative acquisition of the eltAB genes. A plasmid stability study using the conjugative plasmid pCG86 revealed that the plasmid is rapidly lost in CVD 103-HgR in the absence of artificial selective pressure and could no longer be detected after 55 generations. Furthermore, the authors could demonstrate that even under artificial selective pressure, eltAB is only weakly expressed
5604
J. Frey / Vaccine 25 (2007) 5598–5605
in V. cholerae and the toxin titers of the vaccine strain carrying the E. coli plasmid were 200 times lower than that of wild-type V. cholerae [34]. The authors concluded from these experiments that Ent-plasmids could trans-complement the ΔctxA deletion in the vaccine strain but would solely result in a transient very low pathogenic state that is lost rapidly [34]. Generally such assessments considerably enhance the knowledge of biological safety of live vaccines. However, they require a broad basic knowledge of the molecular mechanisms of pathogenicity and presuppose that the attenuating mutations are precisely introduced by means of genetic engineering.
Acknowledgments I am grateful to Jean-Franc¸ois Viret, Crucell-Berna Biotech, Bern, Switzerland, and to Ruud Segers and Ton Jacobs, Intervet International Boxmeer, The Netherlands, Nikolaus Kriz, EMEA, London, UK, Karoline Dorsch, EFBS, BAFU, Bern, Switzerland, Hans-Peter Ottiger and Raphael Fricker, IVI, Mittelh¨ausern, Switzerland, for invaluable scientific advice and to Andrea Hitz and Sarah Burr, Institute of Veterinary Bacteriology, Bern, for valuable editorial help.
References 10. Impact of exogenous genes in live vaccines The introduction of foreign genes provides a broad potential to improve vaccine strains and to expand protective activity against more diseases. They also provide the vaccine strain diagnostic advantages. Foreign genes that are recombined into vaccine strains may theoretically influence the balance of the ecosystem by, for instance, providing the recombinant host new properties to persist or replicate in microenvironments where the ancestor bacterium could not establish. Safety assessments must therefore consider the security measures that were taken in the course of introducing the foreign genes, the influence such genes could play in the final a-virulence of the vaccine strain and the potential survival advantage in particular situations. A precondition to guarantee the highest possible biological safety for vaccines carrying foreign genes is a well-defined carrier strain that contains stable and efficacious attenuations and that itself acts, to a large extent, as biological safety containment.
11. Concluding remarks Modern techniques of genetic analysis and genetic engineering allow the construction of precise, well-characterized deletions and genetic modifications for the construction of GMO bacteria of known virulence and attenuation. This enables the production of safe attenuated live bacterial vaccines whose characteristics are well understood. The latter is a prerequisite for any biological safety assessment. Hence, the important feature in evaluating the biological safety of a live vaccine for the vaccinated individual, entire communities and herds and for the environment does not depend on whether the organism is a GMO or not. Rather it depends solely on the sum of scientific knowledge of the agent and the disease with which it is associated and on the precise genetic features of the vaccine strain. While GMO bacterial live vaccines are still very rare in veterinary medicine, research for new GMO bacterial live vaccines and their introduction into the marked should be favoured, considering their advantage in biological risk assessment and disease control.
[1] Desmettre P. Live bacterial vaccines for animals. Dev Biol Stand 1995;84:221–5. [2] Bowersock TL, Martin S. Vaccine delivery to animals. Adv Drug Deliv Rev 1999;38(2):167–94. [3] Zeman D, Neiger R, Nietfield J, Miskimins D, Libal M, Johnson D, et al. Systemic Pasteurella haemolytica infection as a rare sequel to avirulent live Pasteurella haemolytica vaccination in cattle. J Vet Diagn Invest 1993;5(4):555–9. [4] Henderson LM. Overview of marker vaccine and differential diagnostic test technology. Biologicals 2005;33(4):203–9. [5] Hopkins BA, Olson LD. Comparison of live avirulent PM-1 and CU fowl cholera vaccines in turkeys. Avian Dis 1997;41(2):317–25. [6] Alexanderson S. Advantages and disadvantages of using live vaccines. Risks and control measures. Acta Vet Scand Suppl 2006;90:89–100. [7] Entriken TL. Veterinary pharmaceuticals and biologicals. 12th edition Montvale, NJ, USA: Veterinary Health Care Communications— Thomson Health Care; 2002. [8] Curtiss III R. Bacterial infectious disease control by vaccine development. J Clin Invest 2002;110(8):1061–6. [9] Kenimer JG. Vaccines: current regulatory issues. Curr Opin Biotechnol 1997;8(3):357–60. [10] Hoiseth SK, Stocker BA. Aromatic-dependent Salmonella typhimurium are non-virulent and effective as live vaccines. Nature 1981;291:238–9. [11] Valentine PJ, Devore BP, Heffron F. Identification of three highly attenuated Salmonella typhimurium mutants that are more immunogenic and protective in mice than a prototypical aroA mutant. Infect Immun 1998;66(7):3378–83. [12] Dougan G, Hormaeche CE, Maskell DJ. Live oral Salmonella vaccines: potential use of attenuated strains as carriers of heterologous antigens to the immune system. Parasite Immunol 1987;9(2):151–60. [13] Kariyawasam S, Wilkie BN, Gyles CL. Construction, characterization, and evaluation of the vaccine potential of three genetically defined mutants of avian pathogenic Escherichia coli. Avian Dis 2004;48(2):287–99. [14] Levine MM, Kaper JB. Live oral vaccines against cholera: an update. Vaccine 1993;11(2):207–12. [15] Kotloff KL, Noriega F, Losonsky GA, Sztein MB, Wasserman SS, Nataro JP, et al. Safety, immunogenicity, and transmissibility in humans of CVD 1203, a live oral Shigella flexneri 2a vaccine candidate attenuated by deletions in aroA and virG. Infect Immun 1996;64(11): 4542–8. [16] Favre D, L¨udi S, Stoffel M, Frey J, Horn MP, Dietrich G, et al. Expression of enterotoxigenic Escherichia coli colonization factors in Vibrio cholerae. Vaccine 2006;24(20):4354–68. [17] Curtiss III R, Woodrow GC, Levine MM, editors. New generation vaccines. NY, USA: Marcel Dekker, Inc.; 1990. p. 161–88. Attenuated Salmonella strains as live vectors for the expression of foreign antigens. [18] Ward SJ, Douce G, Figueiredo D, Dougan G, Wren BW. Immunogenicity of a Salmonella typhimurium aroA aroD vaccine expressing
J. Frey / Vaccine 25 (2007) 5598–5605
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29] [30]
a nontoxic domain of Clostridium difficile toxin A. Infect Immun 1999;67(5):2145–52. Vivas J, Carracedo B, Ria˜no J, Razquin BE, L´opez-Fierro P, Acosta F, et al. Behavior of an Aeromonas hydrophila aroA live vaccine in water microcosms. Appl Environ Microbiol 2004;70(5):2702–8. Stritzker J, Janda J, Schoen C, Taupp M, Pilgrim S, Gentschev I, et al. Growth, virulence, and immunogenicity of Listeria monocytogenes aro mutants. Infect Immun 2004;72(10):5622–9. Priebe GP, Meluleni GJ, Coleman FT, Goldberg JB, Pier GB. Protection against fatal Pseudomonas aeruginosa pneumonia in mice after nasal immunization with a live, attenuated aroA deletion mutant. Infect Immun 2003;71(3):1453–61. Walker JC, Verma NK. Cloning and characterisation of the aroA and aroD genes of Shigella dysenteriae type 1. Microbiol Immunol 1997;41(10):809–13. Roberts M, Maskell D, Novotny P, Dougan G. Construction and characterization in vivo of Bordetella pertussis aroA mutants. Infect Immun 1990;58(3):732–9. Chamberlain LM, Strugnell R, Dougan G, Hormaeche CE, Demarco de Hormaeche R. Neisseria gonorrhoeae strain MS11 harbouring a mutation in gene aroA is attenuated and immunogenic. Microb Pathog 1993;15(1):51–63. Ivins BE, Welkos SL, Knudson GB, Little SF. Immunization against anthrax with aromatic compound-dependent (Aro-) mutants of Bacillus anthracis and with recombinant strains of Bacillus subtilis that produce anthrax protective antigen. Infect Immun 1990;58(2):303–8. Uzzau S, Marogna G, Leori GS, Curtiss III R, Schianchi G, Stocker BA, et al. Virulence attenuation and live vaccine potential of aroA, crp cdt cya, and plasmid-cured mutants of Salmonella enterica serovar Abortusovis in mice and sheep. Infect Immun 2005;73(7): 4302–8. Jacobs AA, Goovaerts D, Nuijten PJ, Theelen RP, Hartford OM, Foster TJ. Investigations towards an efficacious and safe strangles vaccine: submucosal vaccination with a live attenuated Streptococcus equi. Vet Rec 2000;147(20):563–7. Clayton MA, Clayton JM, Brown DR, Middlebrook JL. Protective vaccination with a recombinant fragment of Clostridium botulinum neurotoxin serotype A expressed from a synthetic gene in Escherichia coli. Infect Immun 1995;63(7):2738–42. Byrne MP, Smith LA. Development of vaccines for prevention of botulism. Biochimie 2000;82(9–10):955–66. Tavallaie M, Chenal A, Gillet D, Pereira Y, Manich M, Gibert M, et al. Interaction between the two subdomains of the C-terminal part of
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41] [42]
[43]
5605
the botulinum neurotoxin A is essential for the generation of protective antibodies. FEBS Lett 2004;572(1–3):299–306. Fairweather NF, Lyness VA, Maskell DJ. Immunization of mice against tetanus with fragments of tetanus toxin synthesized in Escherichia coli. Infect Immun 1987;55(11):2541–5. Killeen KP, Escuyer V, Mekalanos JJ, Collier RJ. Reversion of recombinant toxoids: mutations in diphtheria toxin that partially compensate for active-site deletions. Proc Natl Acad Sci USA 1992;89(13):6207–9. Mekalanos JJ, Swartz DJ, Pearson GD, Harford N, Groyne F, de Wilde M. Cholera toxin genes: nucleotide sequence, deletion analysis and vaccine development. Nature 1983;306(5943):551–7. Viret JF, Dietrich G, Favre D. Biosafety aspects of the recombinant live oral Vibrio cholerae vaccine strain CVD 103-HgR. Vaccine 2004;22(19):2457–69. Kaper JB, Lockman H, Baldini MM, Levine MM. Recombinant nontoxinogenic Vibrio cholerae strains as attenuated cholera vaccine candidates. Nature 1984;308(5960):655–8. Mbulu RS, Tjipura-Zaire G, Lelli R, Frey J, Pilo P, Vilei EM, et al. Contagious bovine pleuropneumonia (CBPP) caused by vaccine strain T1/44 of Mycoplasma mycoides subsp. mycoides SC. Vet Microbiol 2004;98(3–4):229–34. Westberg J, Persson A, Holmberg A, Goesmann A, Lundeberg J, Johansson KE, et al. The genome sequence of Mycoplasma mycoides subsp. mycoides SC type strain PG1T , the causative agent of Contagious Bovine Pleuropneumonia (CBPP). Genome Res 2004;14(2):221–7. Bischof DF, Vilei EM, Frey J. Genomic differences between type strain PG1 and field strains of Mycoplasma mycoides subsp. mycoides smallcolony type. Genomics 2006;88(5):633–41. Tan S, Gyles CL, Wilkie BN. Evaluation of an aroA mutant Salmonella typhimurium vaccine in chickens using modified semisolid Rappaport Vassiliadis medium to monitor faecal shedding. Vet Microbiol 1997;54(3–4):247–54. Perreten V, Vorlet-Fawer L, Slickers P, Ehricht R, Kuhnert P, Frey J. Microarray-based detection of 90 antibiotic resistance genes of Grampositive bacteria. J Clin Microbiol 2005;43(5):2291–302. Waldor MK, Mekalanos JJ. Lysogenic conversion by a filamentous phage encoding cholera toxin. Science 1996;272(5270):1910–4. Favre D, Viret JF. Biosafety evaluation of recombinant live oral bacterial vaccines in the context of European regulation. Vaccine 2006;24(18):3856–64. Favre D, Struck M-M, Stanley J, Cryz Jr SJ, Viret JF. Further molecular characterization and stability of the live oral attenuated cholera vaccine strain CVD 103-HgR. Vaccine 1996;14(6):526–31.