It’s in the genes! Human genetic diversity and the response to anthrax vaccines

Keywords::

• anthrax
• genetic variation
• host response
• vaccine

Anthrax is a disease caused by a Gram-positive, aerobic, spore-forming bacillus called Bacillus anthracis. This bacterium primarily infects animals and also humans, sometimes with fatal consequences. Due to concerns over the illicit use of this organism, considerable energy has been expended in recent years in an effort to develop medical countermeasures capable of protecting all members of society [1]. Although licensed human vaccines are available, their protective efficacy has yet to be comprehensively demonstrated against the exposure route likely to be encountered during a bioterrorist attack. Currently, there are two human vaccines used in the USA and Europe – anthrax vaccine absorbed (AVA) and anthrax vaccine precipitated (AVP) – both of which are nonliving subunit vaccines and comprise the culture supernatant of an attenuated strain of B. anthracis combined with an aluminum adjuvant [1].

Concerns over the immunogenicity, safety and reactogenicity of these vaccines, coupled with the potential use of the agent by terrorist groups against civilian populations, have driven efforts to develop side effect-free, fully defined, second-generation vaccines capable of stimulating protection following minimal dosing. Research to date has focused on developing vaccines based on protective antigen (PA), the nontoxic cell binding and translocation component of the anthrax tripartite toxin, which owes its name to its ability to protect a range of animal species including primates from a lethal anthrax spore challenge. It is the major protective immunogen of AVA and AVP and is thought to stimulate protection by inducing neutralizing antibodies that prevent PA from transporting the toxic subunits of the tripartite toxin, lethal factor (LF) and edema factor, into the host cell. Indeed, on the basis of animal protection studies, the magnitude of the PA-specific toxin neutralization antibody (TNA) response has been proposed as a surrogate marker of protection.

Although AVA and AVP induce TNAs in humans, the significance of this response is presently unclear due to the lack of human protection data. It should also be borne in mind that the human race, in contrast to the majority of laboratory animals, is genetically diverse; thus, its susceptibility to disease and response to immunization with antigens, such as PA, is influenced by a multitude of host-defined factors [2–4]. Two large AVA studies, each comprising ≥1000 volunteers, have recently reported ethnic differences in both the quantity and quality of PA-specific antibody responses [5,6]. The significant correlation between ethnicity and the magnitude of the TNA response suggests a genetic role in the generation of neutralizing antibodies [6].

These differences may reflect variations in the presentation of PA epitopes by human leukocyte antigens (HLA) to the immune system. This supposition is not unreasonable given that several HLA alleles have been associated with a failure to respond effectively to vaccination [7,8]. Thus, to ensure that all at-risk individuals are fully protected, a vaccine should contain protective epitopes recognized by the broadest possible range of HLA molecules and, where, possible, replicate natural resistance [9]. In this regard, we recently compared the antigen-specific CD4 T-cell responses of natural infected and AVP-immunized individuals and observed significant differences in their responses [10]. Although the antibody response to AVP was directed primarily against PA [11], the major target of T-cell immunity was LF [10], a protein found in trace quantities in AVP and believed to be absent from AVA [1].

To further understand the manner in which the generation of epitope-specific T-cell responses differed between immunized and infected individuals, we defined the T-cell epitopes within domain IV of LF, which contains the catalytic center of a zinc metalloprotease, and identified epitopes preferentially seen in the context of bacterial infection [10]. These epitopes have been termed ‘infection-specific cryptic epitopes’, the suggestion being that infection skews the process by which specific epitopes are recognized and processed. To fully understand the protective basis of the current anthrax vaccines and underpin the development of the next-generation replacements, we need to characterize all the epitopes induced by natural infection. We have made a start and to date have identified epitopes within PA and LF, that are presented by a broad range of HLA class II alleles and thus could represent excellent candidates for inclusion in a future epitope-targeted vaccine [10]. Although vaccines of this kind may be some way off, advances such as epitope strings [12] and personalized peptide vaccination [13] indicate that this is a rapidly evolving field. It is certainly clear that the genetics of host responses to infection is an area of study that will hugely impact on rational vaccine design in the future.

Other non-HLA genes have also been shown to exert an effect on vaccine outcome, for example those encoding cytokine and cytokine receptor genes and the TLRs, a key element of the innate immune system, which recognize pathogen-specific molecules, such as bacterial cell wall components and DNA [8,14,15].

One of these receptors, TLR-9, is expressed by numerous cells of the immune system and responds specifically to short nucleotide sequences called CpG motifs commonly found in the DNA of prokaryotic micro-organisms but absent from the DNA of eukaryotic organisms [16]. Numerous studies have demonstrated the ability of synthetic oligodeoxynucleotides containing immunostimulatory CpG motifs to enhance the immune response to coadministered antigens, including AVA [17]. Indeed, the combination of CpG oligodeoxynucleotides with AVA increased the speed, magnitude and avidity of the resultant PA-specific antibody response in Rhesus macaques and stimulated a 17-fold increase in toxin-neutralizing titer compared with those animals who received AVA alone [17]. It is tempting to speculate that the activation of a pathogen-specific trigger induced the host to mount an immune response to AVA that mirrored the type of protection seen following natural infection.

Genetic variation and its effect on resistance to infection is not restricted to genes encoding immune factors but can also occur in host genes encoding proteins, which are exploited by the pathogen for their survival [2,3]. For example, a 30,000-fold variation in the sensitivity of immortalized human lymphoblastoid cells to anthrax lethal toxin was recently shown to correlate with the relative abundance of CMG2, which serves as the principal receptor for the toxin [18]. The experimental evidence suggests that these differences are inherited, with the lowest sensitivity to toxin observed in cells from the European Caucasian population.

Taken together, all of these recent observations highlight the need for the immunogenicity of the current and future PA-based anthrax vaccines to be determined in ethnically diverse human populations, particularly as a new vaccine is likely to be licensed by the US FDA for human use solely based on animal protection data, due to the sporadic nature of the disease in humans. The fact that animals can never fully represent the genetic diversity of humans highlights the requirement for further understanding of the contribution of host factors to protection and immunity.

One way to address the issue of over-dependence on a single antigen, such as PA, would be to broaden the composition of next-generation vaccines to include additional protective immunogens, such as the nontoxic regions of LF and the capsule, and thus increase the likelihood that an individual’s immune response will efficiently present at least one protective epitope [11]. In addition, the inclusion of multiple protective targets would offer the added advantages of conferring protection against strains in which PA had been genetically altered [19]. In the context of engineered strains, the ideal vaccine would be one that directly targets the spore and thus prevents the expression of foreign virulence genes introduced by genetic engineering [20].

Concerns over the illicit use of B. anthracis as a terror weapon have stimulated recent attempts to develop second-generation vaccines. Although almost two decades and millions of dollars have yielded vaccines capable of protecting experimental animals, there are concerns over the ability of these vaccines to protect all members of society and to confer a spectrum of protection as broad as the vaccines they seek to replace. To meet the challenge posed by biological threats, there should be an increase in understanding of protective immunity and the role of human genetic diversity in the spectra of responses elicited.

Financial & competing interests disclosure

L Baillie is a coinventor of a second-generation anthrax vaccine based on protective antigen, which was developed while working for the UK government. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.