22. Vaccines

The generation of an immune reaction against a pathogen by vaccination follows several distinct steps that should ultimately lead to long-lasting protection against the pathogen through memory cells. These steps are uptake of the vaccine (consisting of either the entire pathogen or antigenic components thereof) by phagocytic cells, activation and migration of professional antigen-presenting cells (APCs) from infected tissue to peripheral lymphoid organs, antigen presentation to T lymphocytes, and finally activation (or inhibition) of T and B lymphocytes. The entire process is illustrated in Fig. 22.3. Below we describe the successive steps leading to an immune response to a pathogen, which are important for the design of vaccines against the pathogen.

Every immune reaction against a pathogen starts with activation of the innate immune system. This is a nonspecific fast response against antigens. Important constituents of the innate system are antigen sampling cells like macrophages and dendritic cells (DCs). The innate response does not lead to immunological memory. Phagocytic cells sense conserved microbial structures called pathogen-associated molecular patterns (PAMPs). They do this via pattern recognition receptors (PRRs) on the cell surface (for bacterial PAMPs) or in the cytoplasm (for viral PAMPs). Examples of PRRs are toll-like receptors (TLRs), scavenger receptors, and C-type lectins (Table 22.2). TLRs consist of a family of receptors, with each member recognizing different patterns of pathogens (Kawai and Akira 2010). TLRs can be found on many cells including macrophages and DCs. DCs can also engulf materials from their extracellular environment by a receptor-independent process called pinocytosis.

Besides mediating uptake of antigenic material from the surrounding tissue, PRRs also play an important role in triggering the cytokine network that will eventually influence the type of adaptive immune response that will be evoked against the pathogen. The phagocytic cells that have taken up pathogens from the infected tissue become activated and start to produce proinflammatory cytokines such as interleukin-1β, interleukin-6, and tumor necrosis factor-α as well as chemokines. The chemokines recruit more phagocytic cells such as neutrophils and monocytes to the infection site, whereas the proinflammatory cytokines induce fever and the production of acute-phase response proteins that can opsonize pathogens.

Most phagocytic cells, including DCs and macrophages, and also B cells can serve as APCs to present processed antigenic determinants to lymphocytes in the peripheral lymphoid organs. For instance, DCs that have taken up antigens from infected tissue become activated and migrate via the afferent lymphatic vessels towards nearby lymph nodes where the encounter with pathogen-specific lymphocytes can take place.

The peripheral lymphoid organs are the primary meeting place between cells of the innate immune system (APCs) and cells of the adaptive immune system (T cells and B cells). Upon interaction with APCs, pathogen-specific T cells and B cells will be activated, provided that they acquire the appropriate signals from the APCs. Besides antigen-specific binding via their antigen receptors, lymphocytes require co-stimulatory signals via interaction of accessory and co-stimulatory molecules between lymphocytes and APCs. This cell-cell interaction is essential for proper stimulation of lymphocytes, and without those accessory signals, antigen-specific T cells may become anergic. Lymphocytes receiving the appropriate signals for activation will clonally expand and generate multiple progenitors all recognizing the same antigen. Clonal expansion is a typical feature of the adaptive immune system, which will be discussed in more detail below.

The adaptive immune system is involved in elimination of pathogens in the late phase of infection and in the generation of immunological memory. It comprises B and T lymphocytes both bearing antigen-specific receptors. The adaptive immune system can be divided into humoral immunity and cell-mediated immunity (CMI) (see Fig. 22.4 and Table 22.3). The humoral response results in antibody formation (but contains cell-mediated events, see Fig. 22.4, panels a, b); CMI results in the generation of cytotoxic cells (see Fig. 22.4, panels a, c). The action of antibodies and T cells is dependent on accessory factors, some of which are mentioned in Table 22.3. In general, after infection with a pathogen or a protective vaccine, both humoral and cellular responses are generated. This indicates that both are needed for efficient protection. The balance between humoral and cellular responses, however, can differ widely between pathogens and is dependent on how the pathogen is presented to the adaptive immune system by APCs. This may have consequences for the design of a particular vaccine (see “Vaccine Design in Relation with the Immune Response”).

Antibodies are the typical representatives of humoral immunity. An antibody belongs to one of four different immunoglobulin classes (IgM, IgG, IgA, or IgE) (cf. Chap.​ 7). Upon immunization, B cells expressing specific antibodies on their cell surface (representing a fifth immunoglobulin class, IgD) bind intact antigen and are activated. The surface-bound antibodies bind specific epitopes of the pathogen, and in close cooperation with T-helper cells (T_h-cells), the B cell becomes activated eventually resulting in massive clonal proliferation. The proliferated B cells are called plasma cells and excrete large amounts of soluble antibodies (Fig. 22.4 panel b). Antibodies are able to prevent infection or disease by several mechanisms:

Antibodies are effective against certain but not all infectious microorganisms. They may have limited value when CMI is the major protective mechanism. Of the cell types that are known to exhibit cytotoxicity, two are antigen sensitized. Because of their specificity, they are of special importance with respect to vaccine design:

Other less specific cells involved in cytotoxic immune responses are natural killer cells (NK cells). They play a role in antibody-dependent cellular cytotoxicity (ADCC). NK cells recognize opsonized (antibody-coated) cells with their Fc receptors.

Besides plasma cells and cytotoxic cells, in many cases, memory B and T cells develop. Memory B cells do not produce soluble antibody, but on repeated antigen contact, their response time to develop into antibody-excreting plasma cells is shorter compared to naïve B cells.

The occurrence of different types of immune response to vaccines is the result of differences in antigen processing of the vaccine by APCs and, as a result, in the activation of T_h-cells (Figs. 22.3 and 22.4). Major histocompatibility complex (MHC) molecules play an important role in the presentation of processed antigens to T cells. Most cells expose MHC class I molecules and some also MHC class II molecules on their surface.

APCs carrying class II molecules process soluble, exogenous (extracellular) proteins or more complicated structures such as microorganisms (see Fig. 22.4, panel a). After their endocytosis, the proteins are subject to limited proteolysis before they return as peptides to the surface of the APC in combination with the class II molecules for presentation to a T-cell receptor of CD4-positive T_h-cells. The T_h-cells provide type 2 help necessary for the effector function of B cells. This type 2 help is characterized by the lymphokine pattern produced: interleukin 4 (IL-4), IL-5, IL-6, IL-10, and IL-13. These lymphokines trigger B cells, which eventually results in the production of IgM and IgG antibodies.

Cells carrying MHC class I molecules process endogenous (intracellularly produced) antigens like viral and tumor antigens and present them in combination with class I molecules on the cell surface (see Fig. 22.4, panel c). The class I-antigen combination on the APC is recognized by the T-cell receptor of CD8-positive CTLs. Th-cells provide help for the CTLs. For the induction of CMI (Fig. 22.4, panels a, c), type 1 help is needed (production of IL-2 and IL-12, interferon-γ, and tumor necrosis factor). Th-cells are CD4 positive, regardless whether they have Th1 or Th2 functions. There is increasing evidence that the Th1/Th2 balance is an important immunological parameter since some diseases coincide with Th1 (autoimmunity)- or Th2 (allergy)-type responses. Other T-helper cell phenotypes have been identified. Some play a role in autoimmune disease (e.g., T_h17-cells) or suppression of the immune response (e.g., T_reg-cells).

For the rational design of a new vaccine, understanding of the mechanisms of the protective immunity to the pathogen against which the vaccine is developed is crucial. For instance, to prevent tetanus a high blood titer of antibody against tetanus toxin is required; in mycobacterial diseases such as tuberculosis, a macrophage-activating CMI is most effective; in case of an influenza virus infection, CTLs probably play a significant role besides antibodies. Importantly, the immune effector mechanisms triggered by a vaccine and, hence, the success of immunization depend not only on the nature of the protective components but also on their presentation form, the presence of adjuvants, and the route of administration.

The presentation form of the vaccine is one of the determinants that influence the extent and type of immune response that will be evoked (Pashine et al. 2005; Pulendran and Ahmed 2006). DCs and other APCs play a pivotal role in how the antigenic determinants of a vaccine will be processed and presented to T cells in the peripheral lymphoid organs. Through various PRRs, DCs are more or less able to “sense” the type of pathogen that is encountered. This determines the set of co-stimulatory signals and proinflammatory cytokines that will be generated by APCs when presenting the antigen to Th-cells in the peripheral lymphoid organs. For instance, pathogens or vaccines containing lipoproteins or peptidoglycans will trigger DCs via TLR-2, which predominantly generates a T_h2 response, whereas stimulation of DCs through TLR-3, TLR-4, TLR-5, or TLR-8 is known to yield robust T_h1 responses. Therefore, vaccines should be formulated in such a way that the appropriate T_h response will be triggered. This can be done by presenting the antigen in its native format, as is the case for the classical vaccines, or by adding adjuvants that stimulate the desired response (see below).

The response by B cells is dependent upon the nature of the antigen and two types of antigens can be distinguished:

When the antigen is a protein, the epitopes can be continuous or discontinuous. Continuous epitopes involve linear peptide sequences (usually consisting of up to ten amino acid residues) of the protein (see Fig. 22.5, panel a). Discontinuous epitopes comprise amino acid residues sometimes far apart in the primary sequence, which are brought together through the unique folding of the protein (see Fig. 22.5, panel b). Antibody recognition of B-cell epitopes, whether continuous or discontinuous, is usually dependent on the conformation (= three-dimensional structure). T-cell epitopes, on the other hand, are continuous peptide sequences, the conformation of which does not seem to play a role in T-cell recognition.

The immunological response to a vaccine is dependent on the route of administration. Most current vaccines are administered intramuscularly or subcutaneously. Parenteral immunization usually induces systemic immunity. However, mucosal (e.g., oral, intranasal, or intravaginal) immunization may be preferred, because it may induce both mucosal and systemic immunity. Mucosal surfaces are the common entrance of many pathogens, and the induction of a mucosal secretory IgA response may prevent the attachment and entry of pathogens into the host. For example, antibodies against cholera need to be in the gut lumen to inhibit adherence to and colonization of the intestinal wall. Also, orally administered Salmonella typhi not only invades the mucosal lining of the gut but also infects cells of the phagocytic system throughout the body, thereby stimulating the production of both secretory and systemic antibodies, as well as CMI. Additional advantages of mucosal immunization are the ease of administration and the avoidance of systemic side effects (Holmgren and Czerkinsky 2005; Czerkinsky and Holmgren 2012). Up to now, however, successful mucosal immunization has only been achieved with a limited number of oral vaccines (e.g., oral polio, cholera, typhoid fever, and rotavirus vaccines) and a nasal influenza vaccine (FluMist). Most of these vaccines are based on attenuated (see later) versions of the pathogens for which the route of administration is the same as the natural route of infection.

Apart from mucosal routes, research groups are working on needle-free jet injection of powders and fluids and dermal delivery with microneedles (Kersten and Hirschberg 2004; Bal et al. 2010) (see Chap.​ 4). A prerequisite of these approaches is that they must be painless. In that case several immunizations can be given with monovalent vaccines, replacing one multivalent vaccine. Up to now, these products have not yet been registered.

Classical vaccines originate from viruses or bacteria and can be divided in live attenuated vaccines and nonliving vaccines. In addition, three vaccine generations can be distinguished for nonliving vaccines. First-generation vaccines consist of an inactivated suspension of the pathogenic microorganism. Little or no purification is applied. For second-generation vaccines, purification steps are applied, varying from the purification of a pathogenic microorganism (e.g., improved nonliving polio vaccine) to the complete purification of the protective component (e.g., polysaccharide vaccines). Third-generation vaccines are either a well-defined combination of protective components (e.g., acellular pertussis vaccine) or the protective component with the desired immunological properties (e.g., polysaccharides conjugated with a carrier protein). An overview of classical vaccines and their generations is given in Table 22.4.

Before the introduction of recombinant DNA (rDNA) technology, a first step to improved live vaccines was the attenuation of virulent microorganisms by serial passage and selection of mutant strains with reduced virulence or toxicity. Examples are vaccine strains for oral polio vaccine, measles-mumps-rubella (MMR) combination vaccine, and tuberculosis vaccine consisting of bacille Calmette-Guérin (BCG). An alternative approach is chemical mutagenesis. For instance, by treating Salmonella typhi with nitrosoguanidine, a mutant strain lacking some enzymes that are responsible for the virulence was isolated (Germanier and Fuer 1975).

Live attenuated organisms have a number of advantages as vaccines over nonliving vaccines. After administration, live vaccines may replicate in the host similar to their pathogenic counterparts. This confronts the host with a larger and more sustained dose of antigen, which means that few and low doses are required. In general, the vaccines give long-lasting humoral and cell-mediated immunity.

Live vaccines also have drawbacks. Live viral vaccines bear the risk that the nucleic acid sequence is incorporated into the host’s genome. Moreover, reversion to a virulent form may occur, although this is unlikely when the attenuated seed strain contains several mutations. Nevertheless, for diseases such as viral hepatitis, AIDS, and cancer, this drawback makes the use of classical live vaccines virtually unthinkable. Furthermore, it is important to recognize that immunization of immune-deficient children with live organisms can lead to serious complications. For instance, a child with T-cell deficiency may become overwhelmed with BCG and die.

An early approach for preparing vaccines is the inactivation of whole bacteria or viruses. A number of reagents (e.g., formaldehyde, glutaraldehyde) and heat are commonly used for inactivation. Examples of this first-generation approach are pertussis, cholera, typhoid fever, and inactivated polio vaccines. These nonliving vaccines have the disadvantage that little or no CMI is induced. Moreover, they more frequently cause adverse effects as compared to live attenuated vaccines and second- and third-generation nonliving vaccines.

Nowadays vaccines can be designed based on the information encoded by the genome of a particular pathogen (Masignani et al. 2002; Rappuoli and Covacci 2003). From many pathogens, the entire genomes have been sequenced and this number is growing (http://​cmr.​tigr.​org). The genome sequence of a pathogen provides a complete picture of all proteins that can be produced by the pathogens at any given time. Using computer algorithms, proteins that are either excreted or expressed on the surface of the pathogen, and thus most likely available for recognition by the host’s immune system, can be identified. After recombinant production and purification, these vaccine candidates can be screened for immunogenicity in mice. From these, the best candidates can be selected and used as subunit vaccines (Fig. 22.7).

A big advantage of reverse vaccinology is the ease at which novel candidate antigens can be selected without the need to cultivate the pathogen. Furthermore, by comparing genomes of different strains of a pathogen, conserved antigens can be identified that can serve as a “broad spectrum” vaccine, giving protection against all strains or serotypes of a given pathogen. One drawback of this approach is that it is limited to the identification of protein-based antigens.

Reverse vaccinology has been successfully used to identify novel antigens for a variety of pathogens, including Neisseria meningitidis, Bacillus anthracis, Streptococcus pneumoniae, Staphylococcus aureus, Chlamydia pneumoniae, and Mycobacterium tuberculosis (Sette and Rappuoli 2010).

Most classical vaccine applications are prophylactic: they prevent an infectious disease from developing. Besides prophylactic applications, vaccines may be used to treat already established diseases, such as infectious diseases, cancer, or drug addiction. Although the development of therapeutic vaccines is still in its infancy, some examples will be highlighted here.

One of the biggest problems in vaccine development is the inability to predict the efficacy and safety of new vaccines with other methods than phase 3 studies. Animal models are poorly predictive, and even immunological parameters in humans, like the induction of pathogen-neutralizing antibody responses, are not fully predictive. As a result, vaccine efficacy has to be measured in terms of reduction of disease. This is sometimes very difficult because symptoms of a disease can be caused by more than one pathogen (e.g., influenza-like illness). To measure protection, one has to detect the influenza virus in the group with influenza-like illness. In addition, to measure reduction of disease, the groups in the trial need to be very large because it is unknown who will get the disease. Sometimes tens of thousands of people are included in phase 3 trials.

Systems biology approaches that may limit these problems in the future are under development. The idea is to identify gene signatures that correlate with a protective immune response. This is done by a combination of gene expression analysis for, e.g., lymphocytes in the blood and functional assays like measurement of antibodies, cytokines, and cellular responses. Bioinformaticians try to unravel pathways and networks of genes involved in immune responses. Eventually, it may be possible to assign the activity of a limited number of genes to protective immune responses (Nakaya et al. 2011). Perhaps that such gene signatures can be used in future clinical trials (Pulendran et al. 2010). This approach would allow for reduction of the size of clinical studies, reduce the risk of late stage failure, and assess the significance of animal models in the preclinical phase of vaccine development.

Except for synthetic peptides, the antigenic components of vaccines are derived from microorganisms or animal cells. For optimal expression of the required vaccine component(s), these microorganisms or animal cells can be genetically modified. Animal cells are used for the cultivation of viruses and for the production of some subunit vaccine components and have the advantage that the vaccine components are released into the culture medium.

Three stages can be discerned in the manufacture of cell-derived vaccines: (1) cultivation or upstream processing, (2) purification or downstream processing, and (3) formulation. For the first two stages, the reader is referred to Chap.​ 3, whereas the formulation is addressed in the following section.

Second- and third-generation classical vaccines and modern vaccines are better-defined products in terms of immunogenicity, structure, and purity. This means that the products can be characterized with a combination of appropriate biochemical, physicochemical, and immunochemical techniques (see Chap.​ 2). Vaccines have to meet similar standards as other biotechnological pharmaceuticals. The use of modern analytical techniques for the design and release of new vaccines is gaining importance. Currently, animal experiments are needed for quality control of many vaccines but in vitro analytical techniques may eventually (partly) substitute preclinical tests in vivo. During the development of the production process of a vaccine component, a combination of suitable assays can be defined. These assays can subsequently be applied during its routine production.

Column chromatographic (HPLC) and electrophoretic techniques like gel electrophoresis and capillary electrophoresis provide information about the purity, molecular weight, and electric charge of the vaccine component. Physicochemical assays comprise mass spectrometry and spectroscopy, including circular dichroism and fluorescence spectroscopy. Information is obtained mainly about the molecular weight and the conformation of the vaccine component. Immunochemical assays, such as enzyme-linked immunoassays and radioimmunoassays, are powerful methods for the quantification of the vaccine component. By using monoclonal antibodies (preferably with the same specificity as those of protective human antibodies) information can be obtained about the conformation and accessibility of the epitope to which the antibodies are directed. Moreover, the use of biosensors makes it possible to measure antigen-antibody interactions momentarily, allowing accurate determination of binding kinetics and affinity constants.

Depending on their specific characteristics, vaccines are stored as solution or as a freeze-dried formulation, usually at 2–8 °C. Their shelf life depends on the composition and physicochemical characteristics of the vaccine formulation and on the storage conditions and typically is in the order of several years. The quality of the container can influence the long-term stability of vaccines, e.g., through adsorption or pH changes resulting from contact with the vial wall. The use of pH indicators or temperature- or time-sensitive labels (“vial vaccine monitors,” which change color when exposed to extreme temperatures or after the expiration date) can avoid unintentional administration of inappropriately stored or expired vaccine.