New Approaches for Needed Vaccines: Bacteria

1

Background

Bacteria are still one of the most common causes of infection, associated with a range of different diseases in humans and important veterinary animals. The emergence of antibiotic resistant microbes (ARM) means that the threat bacteria pose to human health is unlikely to diminish in the near future. Thus, the design of new or improved bacterial vaccines remains high on the global health agenda, particularly as we still lack effective vaccines for many bacterial disease classes. Bacteria are relatively complex life forms in comparison to viruses as they harbor genes for thousands of proteins that provide structure, support life, and drive the synthesis of complex life-associated molecules [eg, lipopolysaccharide (LPS), capsules, etc.]. Nevertheless, individual disease-associated bacteria have been successfully targeted by vaccination. Indeed, Pasteur targeted bacterial diseases such as anthrax in his very early work on vaccine development.

Since those early pioneering days, a variety of bacterial vaccines have been developed based around different formulations. Many early vaccines, for example, typhoid and cholera, were composed of “killed” whole bacterial cells that were delivered parenterally by injection. Such vaccines tended to be of moderate efficacy and were quite reactogenic, in part because of the many innate toxins and immune stimulators that whole bacterial cells harbor. Perhaps the only vaccine of this type still in common use is the whole-cell pertussis vaccine based on inactivated Bordetella pertussis bacteria. Killed whole bacterial cell vaccines have also been developed for oral use against enteric infections such as cholera and enterotoxigenic Escherichia coli (ETEC). Again these vaccines tend to have moderate efficacy, although reactogenicity is less of an issue and they have found some utility both within disease endemic regions and for travelers to such regions. Additionally, live whole cell vaccines have been developed that are based upon attenuated vaccine strains, such as Bacille de Calmette et Guérin (BCG), that can be delivered orally or by injection. The only live attenuated bacterial vaccine still finding broader utility is BCG, although more niche typhoid (Ty21a) and cholera vaccines are available in some regions.

The recognition that some bacterial diseases are toxin-driven facilitated the development of early toxoid-based vaccines for diphtheria and tetanus. These vaccines are now cornerstones of the global children’s vaccine program, even though the technology used to produce them is relatively crude by modern standards. The success of the toxoid approach stimulated a drive to replace some whole cell vaccines (eg, cholera and pertussis) with toxoid formulations. A good modern example was the development of acellular pertussis vaccines built around toxoided pertussis toxin and several other defined B. pertussis surface proteins (eg, pertactin, filamentous hemagglutinin). The development of these vaccines proved the principle of moving from whole cell to acellular bacterial vaccines based on defined antigen mixes. Other examples have followed including the recently licensed meningitis vaccine Bexsero (4CMenB), based on a combination of Neisseria proteins. Hence, in line with other vaccines, there is a trend toward bacterial vaccines of defined antigenic composition or known mechanism of attenuation, facilitating a drive toward better quality and a stronger safety profile.

Over the past decades investigations into how bacteria cause disease have intensified and many key “pathogenic mechanisms” have been defined and the proteins and other molecules that contribute to infection identified. Such studies on the molecular basis of infection/pathogenesis can contribute to vaccine design and the selection of antigens for vaccine development. Thus, the molecular toolbox for vaccine development has expanded dramatically giving rise to opportunities for new vaccination approaches based on functional genomics and structural biology. These approaches are also incentivizing interest in bacterial targets that were previously regarded as intractable or challenging.

Surface polysaccharides play a key role in the ability of pathogenic bacteria to defend themselves against attack from the host immune system. Consequently, there was an early recognition that such polysaccharides could make attractive vaccine candidates. Many polysaccharides are relatively poor immunogens and most are so-called T-cell independent antigens that are poor stimulators of T-cell immunity. Consequently such antigens are poorly immunogenic in infants, do not effectively induce antibody affinity-maturation, efficient antibody class switching or effective immune memory. Fortunately such T-cell-independent antigens can be converted to T-cell-dependency by conjugating them to carrier molecules, normally globular proteins. Such conjugates have found broad utility in the vaccine industry and now many classes of bacterial vaccines are being built using the conjugation approach.

The first conjugate vaccine of this type to be licensed was against Haemophilus influenzae type B and now conjugate vaccines against Streptococcus pneumoniae and Neisseria meningitidis have been successfully launched (see other chapters in this book). A complication of the conjugate approach is that a variety of different antigenically distinct polysaccharides can be found on different clades of the same bacteria causing the same disease (eg, S. pneumoniae ) and consequently multiple conjugates have to be formulated into the same (now multivalent) vaccine. Also, only a limited number of proteins have been used as carriers in licensed vaccines. These include tetanus toxoid and a mutant diphtheria toxin known as CRM ₁₉₇ (cross reactive material 197). These carriers are regarded as “heterologous” in the sense that they are derived from different bacteria than the target polysaccharide antigen. It may be advantageous to use “homologous” antigens from the same bacteria in certain circumstances so searching for new carrier proteins is an ongoing endeavor. Such homologous antigens have the potential to induce pathogen-specific T cells.

2

Gaps and targets

Despite the success of these approaches there are still significant challenges remaining in the field of bacterial vaccines. We do not have any licensed vaccines against a range of important pathogens including Treponema pallidum (syphilis), Chlamydia , Shigella , Klebsiella and these are urgently needed as some are developing antibiotic resistance. Also, there is evidence that some pathogens, including B. pertussis and S. pneumoniae may be escaping vaccine-induced immunity. Further, the evolution of bacterial disease is dynamic and rapid, so we can anticipate that the epidemiology of such infections will change over time. This is in part driven by the use of antibiotics and the aging human population, where increasing numbers of immunosenescent individuals present fresh challenges to vaccine developers.

A major gap exists in the area of sexually transmitted infections, where we currently have no vaccines against the more common bacterial diseases. Antigenic variation has presented a challenge to vaccine development against Neisseria gonorrhoea with early attempts to develop vaccines based on fimbriae/pili having little success. Vaccines against T. pallidum and Chlamydia should be feasible but to date progress has been hampered by a variety of challenges, including the fastidious nature of these pathogens and the lack of good animal models. However, the dramatic breakthrough made in the area of papilloma virus vaccines indicates that the problem is tractable. Another area of challenge is the healthcare associated pathogens. Here a variety of infections are emerging ranging from multiply antibiotic resistant Staphylococci through to Clostridium difficile . Many of these infections are associated with immunocompromised or aging individuals and any vaccine development against such infections must take this factor into account. We also lack effective vaccines against many of the bacteria that cause enteric infections and here a number of approaches, including the development of live oral vaccines, have met with mixed success. New vaccines would also be desirable against Mycobacterium tuberculosis and other Mycobacteria associated with infections. Here, intensive investigations are underway.

3

Classical approaches for making bacterial vaccines

One of the challenges associated with generating vaccines against bacterial pathogens is to identify potentially protective antigens among the thousands of proteins and other antigenic molecules the bacteria can produce. In the premolecular era vaccine developers either exploited whole bacterial cells or focused their attention on a limited number of tractable antigens that could be identified by simple serological or biochemical assays. The main targets that emerged from such studies were either immunogenic surface oligosaccharides or polysaccharides that could be readily purified or toxins identified as significant drivers of disease pathology.

Despite the limitations of these approaches, considerable progress was made in terms of developing vaccines against many bacteria. We have mentioned the early successes with diphtheria and tetanus toxins that were inactivated by chemical treatment to produce highly successful toxoid-based antigens. Toxoid extracts have remained in general use since the early development for these two diseases. Although attempts have been made to make improved vaccines based on either purified toxins or genetically engineered variants (eg, CRM ₁₉₇ for diphtheria toxin or tetanus toxin fragment C), these have not been adopted for the two diseases. However, as we entered the molecular genetic era in the 1970s efforts began to emerge to make improved bacterial vaccines based on more defined antigens.

The first real success in this area was the development of acellular pertussis vaccines, driven by claims that the whole-cell pertussis vaccine caused serious side effects in some individuals. Searches were undertaken to identify potential antigens for inclusion in acellular vaccines. The primary candidate was pertussis toxin that had been shown to be one of the main drivers of disease and was a highly immunogenic antigen. Pertussis toxin has a classical AB bacterial toxin structure and toxoided versions of the toxin were created that were both immunogenic and induced the production of toxin-neutralizing antibodies. In addition, genetically modified versions of the toxin were engineered using site-directed mutagenesis and these were also considered as vaccine candidates. Several other protein antigens, including the adhesins pertactin and filamentous hemagglutinin, were also developed as vaccines leading to the generation of a series of acellular vaccines that were evaluated in clinical trials in different countries. Eventually blends of these antigens were found to have reasonable efficacy in children and acellular vaccines were gradually licensed across the world over the following decade. Thus, acellular pertussis vaccines became the first of a new type of bacterial vaccine based on combinations of defined bacterial proteins.

Again, as biochemistry improved, vaccine developers began to search for other antigens that might be exploitable as acellular vaccines and attention turned to the polysaccharides that were expressed at the bacterial surface as capsular materials. A series of polysaccharide based vaccines were generated against diseases including H. influenzae (type B), N. meningitidis (A, C) and S. pneumoniae (multiple capsular types). Although these vaccines had efficacy against disease they generally induced relatively short-lived protection due to the T-cell independent nature of the antigens and they were gradually replaced with conjugate versions, many of which are now licensed vaccines.

Thus, a combination of acellular and conjugate based approaches began to fill some of the gaps in the bacterial vaccine repertoire. However, many pathogens did not produce readily tractable polysaccharide antigens or obvious toxins that could induce broad immunity. Thus, new approaches were required to start to target these remaining classes of pathogens.

4