Introduction

Even as we move towards the second millennium, we continue to be confronted by both old diseases like tuberculosis, hepatitis, Lyme disease and new ones like AIDS. Other disorders including cancer and genetic blood diseases brought about by environmental pollutants and stressful lifestyles are proving to be extremely costly for public sector health management. Furthermore, viral, bacterial and protozoan pathogens continue to inflict high levels of morbidity and mortality in humans as well as in commercial animal production systems. Prevention by strategic administration of efficacious vaccines appears to be the most economical approach in alleviating, if not, eradicating many of these problems.¹

Protection of immunized recipients against disease should be the principal criterion for vaccine efficacy. Unfortunately, the cost of vaccine production and market demand can frequently influence research into vaccine development. Nowhere is this more evident than in the poultry industry where efficacious vaccines are sold for only a few cents per unit dose. Consequently, vaccines for animal diseases caused by bacteria are often packaged as multivalent bacterins designed to confer protection against a number of disease-causing agents or a range of different serotypes of the same pathogen. As the level of exposure to endemic pathogens is usually low under field conditions, such vaccines are rarely put to the test. It would be unrealistic to expect a comprehensive immune response to be generated against the plethora of potential antigenic epitopes present in multi-component vaccines. One is tempted to ask the question whether vaccines incorporating antigenic determinants of a specific pathogen may do more than just promote adaptive immunity against that pathogen. Could vaccines containing the appropriate combination of adjuvant and antigen not only 'prime' for acquired immunity but also serve to activate the innate immune system? And, would dependence of the adaptive immune response on activation of innate immunity² dictate the pattern of subsequent innate immune responses?

Adjuvants

Adjuvants are immunopotentiating substances and may include any natural or synthetic products. Immune responses are enhanced in recipient animals when vaccinated with adjuvanted T-dependent or T-independent antigens. The use of adjuvants in vaccines is particularly important when the antigen has low immunogenicity. This applies to subunit and recombinant peptide antigens whose structure and conformation are less complex than inactivated, but still intact, viruses and bacteria. Ultimately, the usefulness of an adjuvant is dependent upon its safety profile and whether it is able to stimulate long-lasting immunity. Table 1 depicts selected examples of commercially available adjuvants. Without exception, all have been reported to stimulate either humoral or cellular immunity and some appear to stimulate both. Interestingly, only a few publications have alluded to the inflammatory properties of adjuvants; for example, the granuloma-inducing properties of Freund's complete adjuvant are legendary. None, for instance, have investigated the role of adjuvants in mobilizing inflammatory cells of the myeloid-monocytic lineage, even though they are the major players in an innate immune response.

Table 1 - Examples of different arbitrary classes of adjuvants and their immunomodulatory function*.

Full table (97K)

Antigens

Bacterins and inactivated viral particles possess highly complex structures with potentially, many immunogenic epitopes. The immune response in vaccinated outbred species to these antigens is surprisingly consistent and directed against only a few immunodominant epitopes. There is, however, more variability in antibody responses against less immunogenic determinants of structurally complex vaccines. In an effort to minimize antigenic competition and to elicit only specific immune responses against protective antigens, the recent trend in vaccine development has focused on the search for purified and well-characterized subunit vaccine antigens. When presented in a soluble form, these antigens do not stimulate naive T cells to respond in vitro. ¹⁵ However, the addition of antigen-presenting cells (APC) dramatically increases T cell help, and the cellular immune response profile including antibody production is also enhanced.

Soluble antigens are invariably processed in membrane-bound acidic endosomal compartments. The processed peptides are thus always presented in association with class II MHC molecules to elicit CD4⁺ T cell help. When delivered, particulate antigens generally elicit strong humoral responses against conformational epitopes. Intracellular processing of such antigens by B cells and macrophages only rarely, if at all, route processed peptides via class I MHC. To generate cytotoxic T cell (CTL) responses, the antigenic peptides must be derived from intracellular processing of newly synthesized proteins within proteasomes in the cell cytoplasm. Peptides are transported to the endoplasmic reticulum where they form complexes with MHC class I molecules for presentation to CD8⁺ cytotoxic T cells. Therefore, infectious agents that proliferate intracellularly elicit better CTL responses because immunogenic peptides associate readily with class I MHC. Consequently, the use of live or attenuated bacterial and viral vaccine vectors that can replicate intracellularly are more effective at stimulating CTL immunity.¹⁶ Unfortunately, this partitioning of antigen processing pathways cannot adequately explain observations of CTL responses to some adjuvanted vaccine antigens. Per Peterson's group at the Scripps Institute have clearly demonstrated that MHC class I molecules can acquire antigen in the phagocytic pathway by trafficking from the trans-Golgi network to the phagosome.¹⁷ This mechanism may also operate in vivo and provide a model to explain why liposomes,¹⁸ immune stimulating complexes (ISCOM)¹⁹ and non-ionic triblock copolymer adjuvants¹¹ can stimulate humoral immunity as well as both classes I and II CMI. There are also examples where particulate antigens formed by covalent linkage to synthetic microspheres promote both humoral and cellular immunity in the absence of adjuvant.²⁰ Trans-trafficking of internalized particulate antigen-bearing microspheres between intracellular compartments may account for the cytotoxic responses elicited in these systems.

Routes for vaccine delivery

Most vaccines are administered either intramuscularly or subcutaneously. Adjuvanted antigen when delivered as a bolus by these routes is localized at the site of injection. This provides an antigen depot from which trafficking macrophages acquire, process and ultimately present antigen to T cells. Peripheral routes of immunization are not known to stimulate mucosal immunity and, in particular, mucosal IgA. The paradigm that mucosal routes of immunization best elicit mucosal immunity has been challenged in recent years, with several groups^{7, 21} demonstrating that intradermal immunization can also elicit mucosal as well as systemic immune responses. Indeed, intradermal immunization has been reported to elicit better immunity against hepatitis B vaccine in healthcare workers who failed to respond to intramuscular vaccination.²² The skin is also a more effective route for promoting antigen-specific responses to DNA vaccines than either intrasplenic or intranasal immunization.²³ One reason for the potency of the skin as an immunization route may be attributed to the presence of specialized APC known as Langerhans cells (LC).

Epidermal LC are of bone marrow (BM) origin and usually comprise about 2–5% of cells in the mammalian epidermis.²⁴ Langerhans cells constitutively express class II and their primary sentinel role in the skin is to entrap antigens permeating the epidermis and to process and present these to T cells.²⁵ To do this, LC activated by encounter with antigen, retract their dendrites and alter their surface adhesins to facilitate migration from the skin to draining lymph nodes. While there, they re-assume more macrophage-like or dendritic morphology and eventually mature to become interdigitating cells.²⁶ Contact sensitizers have been used to examine the rates of LC migration from the skin via the lymphatics to the draining lymph nodes. With the chemical carcinogen DMBA ([7,12-dimethylbenz(a)anthracene]), LC migration commenced 24 h after exposure and reached a peak after 5 days. In contrast, TNCB (2,4,6-trinitrochlorobenzene) initiated migration within minutes of topical application and reached a peak by 12 h.²⁷ Deposition of antigen on or in the skin can, therefore, provide an efficient method for stimulating immunity. It would be interesting to determine whether the addition of an adjuvant to the antigen for skin immunization would contribute further in enhancing the immune response.

Development of a murine vaccine/infection model to assess the efficacy of intradermal immunization against a porcine bacterial pathogen that causes pleuropneumonia

Actinobacillus pleuropneumoniae (Ap) is a Gram-negative coccobacillus and the aetiological agent of porcine pleuropneumonia. Pathogenic isolates of Ap excrete a number of haemolysins (Apx toxins) that are considered important virulence factors. Vaccines that contain outer membrane proteins (OMP) of Ap as well as those that also incorporate one of the Apx toxins, have been found to protect animals against lethal challenge. Mice infected intranasally with Ap develop lung pathology very similar to that seen in pigs. This includes massive infiltration of polymorphonuclear cells to the lung with extensive oedema. The murine model provides an appropriate system to test the hypothesis that systemic and mucosal immunity generated by intradermal vaccination may protect against lethal challenge and, also, to elucidate some of the mechanisms that may contribute to such protection.

An important consideration in the development of the murine Ap vaccine model is the role of lipopolysaccharide (LPS) as an immunoregulator in its own right. Lipopolysaccharide is a major structural component in the outer membranes of Gram-negative bacteria and is normally co-purified when outer membrane vesicles are extracted by buoyant density ultracentrifugation in sucrose gradients. Recognition of LPS by LPS-binding molecules such as CD14,²⁸ CD11/CD18 integrin family,²⁹ the scavenger receptor on macrophages³⁰ or the 80 kDa protein on the surface of polymorphonuclear leucocytes (PMN)³¹ represents a critical step in the mobilization of host defences. Consequently, all vaccination regimens in the Ap mouse model included an unvaccinated control group (group Con) injected intradermally with PBS and an antigen control group (group Ag) intradermally immunized with only outer membrane proteins/LPS purified from A. pleuropneumoniae (serovar 1).

Immune-stimulating complexes or ISCOM are cage-like structures composed of lipids and the saponin Quil A. Hydrophobic antigens such as OMP/LPS are readily incorporated into ISCOM while hydrophilic antigens such as ovalbumin are less easily incorporated. However, both hydrophilic and hydrophobic antigens can be first incorporated into liposomes and then re-incorporated into ISCOM with very high efficiency (J. Chin unpubl. data). These liposome/ISCOM hybrids (LIH) appear structurally identical to ISCOM when examined by electron microscopy (B. Morein pers. comm.). The immunomodulatory and protective effects of OMP/LPS vaccine was assessed by incorporation into LIH adjuvant and injected intradermally into mice (vaccine group or group Vac). Additionally, to assess the possible contribution of the LIH adjuvant to cellular mobilization, an adjuvant control group (group Adjv) was included in these studies.

All groups of mice were vaccinated on two occasions at days 0 and 7, respectively. Four mice from each group were killed at various time points postimmunization (Figure 1). Anti-OMP antibody levels were determined in respiratory tract washings and blood samples by ELISA. Leucocytes were prepared from the thymus, spleen and collagenase-digested lung tissues.³² Bone marrow cells were obtained from femurs by flushing. Cells were stained with monoclonal antibodies against cell surface markers characteristic of thymocytes (Thy-1⁺), B cells (B220⁺), granulocytes (Gr-1⁺), myelomonocytic cells (Mac-1⁺) and T cells (CD4⁺or CD8⁺). All of the aforementioned were obtained from Pharmingen, San Diego, CA, USA. Samples were analysed by flow cytometry to obtain a temporal profile of changes in various leucocyte subpopulations in different organ compartments. Pluripotent haematopoietic progenitor cells (HPC) were identified as Thy-1^low CD4^- CD8^- B220^- Mac-1^- Gr-1^- and hereafter referred to as Thy-1^low Lin^- to designate the absence of expression of lineage markers.³³ Based on cell counts, the result for each Ag, Adjv and Vac group was expressed as a percentage shift relative to control-grouped mice. Shifts to the left or right designate, respectively, a relative decrease or increase in cell numbers for each population.

Figure 1.

Timeline showing intradermal immunization (syringe/needle), and samp-ling points for flow cyto-metry (unfilled arrows) and serology (filled arrows).

Full figure and legend (15K)

Vaccine efficacy was determined by challenging mice (groups Con, Adjv, Vac) intranasally at week 5 (Figure 1, star) with A. pleuropneumoniae (10⁸ CFU in 20 L). Clinical signs based on morbidity and laboured breathing were used to compile a clinical score for each mouse. Severely affected mice with a high clinical score were killed with pentobarbital. Death was not used as an end-point in these experiments.

Inflammatory cells in different organ compartments are modulated by intradermal delivery of liposome/ISCOM-adjuvanted vaccines

Figure 2 shows the shift in total cell numbers in the (a) BM, (b) thymus, (c) spleen and (d) lung at 3, 14 and 21 days postimmunization (PI) for mice vaccinated with Adjv (hatched bars), Ag (stippled bars) and Vac (filled bars) groups, respectively. Vaccination significantly decreased cell numbers in the BM by more than 50% in all three groups of mice at day 3 (P<0.05). The biggest shift was seen in mice vaccinated with adjuvanted antigen, and the least change was observed in mice given only antigen. By day 21, cell numbers had returned to normal (0% shift relative to controls) with the exception of group Vac mice, which displayed an increase of 50% relative to controls (P<0.01). Flow cytometry was used to track the identity of the cell population in the BM associated with vaccine-induced shifts in cell numbers (Figure 3). Within days, vaccination with antigen, adjuvant or vaccine consistently caused a 40–60% decline (P<0.01) in HPC even though this population accounted for less than 0.5% of total BM cells in the femur. While HPC numbers remained depressed in both antigen- and adjuvant-treated mice for up to 21 days PI, group Vac mice showed an increase in HPC at this time (Figure 3a). At day 3, vaccination also depressed a second population of pro- and pre-B cells carrying the Thy^- B220⁺ phenotype. By day 21 PI, this population had rebounded above control levels in both adjuvant- and antigen-vaccinated mice. Mice receiving the vaccine displayed the most significant increases in pro- and pre-B cells (P<0.001) at this time (Figure 3b). The numbers of Mac-1⁺ Gr-1⁺ cells representing immature and mature granulocyte lineage cells fell after intradermal vaccination in all groups of mice with the exception of group Vac animals. These actually increased at day 3 PI, returned to baseline or control levels at day 14 and surged upwards (P<0.05) again at day 21 (Figure 3c). Vaccination reduced the numbers of BM promonocytes and monocytes (Mac-1⁺ Gr-1^-) in all groups at day 3 PI. This population then increased above the controls at day 14, with the largest increase seen in mice vaccinated with adjuvanted antigen.

Figure 2.

Percentage shifts in total cell numbers in (a) the bone marrow, (b) thymus, (c) spleen and (d) lung at 3, 14 and 21 days postimmunization for groups of mice vaccinated with () Adjv, () Ag and () Vac, respectively.

Full figure and legend (21K)

Figure 3.

Percentage shifts in (a) Thy⁺Lin^-, (b) B220⁺, (c) Gr-1⁺Mac-1⁺ and (d) Gr-1^-Mac-1⁺ cell populations in the bone marrow at 3, 14 and 21 days postimmunization for groups of mice vaccinated with () Adjv, () Ag and () Vac, respectively.

Full figure and legend (25K)

Thymocyte cell numbers were decreased in mice vaccinated with either adjuvant or vaccine (P<0.002). Once again, intradermal administration of vaccine caused the most dramatic decline in thymocyte numbers. Repopulation of the thymus by progenitor cells gradually returned cell numbers to normal levels by day 14 in adjuvant-vaccinated mice, but recovery occurred far more slowly in mice given adjuvanted antigen. Intradermal administration of antigen alone had no effect on thymus cell numbers. Figure 4 shows relative changes in different thymocyte populations carrying either the two single-positive markers for CD4⁺ CD8^- and CD4^- CD8⁺, or the double-positive CD4⁺CD8⁺ (DP) and double-negative CD4^- CD8^- phenotypes (DN). All four populations were reduced by vaccination. The DP population was most sensitive to vaccination with adjuvant and particularly so with adjuvanted antigen. Antigen alone had little or no impact on the numbers of DP thymocytes.

Figure 4.

Percentage shifts in (a) CD4^-CD8^-, (b) CD4⁺ CD8⁺, (c) CD4⁺CD8^- and (d) CD4^-CD8⁺ cell populations in the thymus at 3, 14 and 21 days postimmunization for groups of mice vaccinated with () Adjv, () Ag and () Vac, respectively.

Full figure and legend (23K)

Relative to the controls, vaccination caused slight decreases in spleen cell numbers at day 3 PI in all vaccinated groups. By day 14, cell numbers had begun to rise in both antigen- and adjuvant-vaccinated mice. Cell numbers in the spleen of group Vac mice increased significantly (P<0.02) by two- to three-fold. The increase in spleen cell numbers observed in mice given vaccine was caused primarily by an increase in granulocytes (Gr-1⁺ Mac-1⁺; Figure 5a) and monocytes (Gr-1^- Mac-1⁺; Figure 5b). The indigenous population of T (CD4⁺or CD8⁺ phenotype) and B cells (B220⁺) were not affected by vaccination (J. Chin unpubl. data).

Figure 5.

Percentage shifts in granulocyte ((a) Gr-1⁺Mac-1⁺) and monocyte ((b) Gr-1^-Mac-1⁺) cell populations in the spleen at 3, 14 and 21 days postimmunization for groups of mice vaccinated with () Adjv, () Ag and () Vac, respectively.

Full figure and legend (9K)

Only minor changes in cell numbers recruited to the lung were observed at day 3 PI in mice given either antigen or adjuvant. Intradermal injection with adjuvanted antigen caused the greatest increase in cellular recruitment to the lung. This population was confined primarily to granulocytes (Figure 6) that were recruited to the lung as early as 3 days PI, with further increases by day 14 and gradually falling to control levels by day 21.

Figure 6.

Percentage shift in granulocytes (Gr-1⁺Mac-1⁺) in the lung at 3, 14 and 21 days postimmunization for groups of mice vaccinated with () Adjv, () Ag and () Vac, respectively.

Full figure and legend (17K)

Antibody response to skin immunization

The temporal production of antibodies against OMP antigens of A. pleuropneumoniae is depicted in Figure 7. The data is plotted as a ratio of ELISA reactivity in mice vaccinated with either adjuvanted antigen or antigen alone versus unvaccinated control mice for each time point that was sampled. The serum IgG, IgA and IgM levels in group Vac mice paralleled those seen in mice vaccinated intradermally with only antigen. Serum IgG and IgM levels peaked at 3 weeks post-vaccination and were about 11 and eight times that found in control mice. A low-level rise in serum IgA was detectable throughout the 20-week period and is consistent with previous experiences with intradermal immunization using different antigens in other species.²¹

Figure 7.

Time course showing ELISA reactivity of serum (a) IgG, (b) IgA and (c) IgM against Actinobacillus pleuropneumoniae outer membrane proteins for groups of mice vaccinated with () Ag and () Vac, respectively.

Full figure and legend (22K)

Modulation of leucocyte cell populations following lethal bacterial challenge in vaccinated and unvaccinated mice

The effect of bacterial challenge on changes in the cell population of the BM, spleen and lung was examined in unvaccinated mice and mice that had been intradermally vaccinated with either adjuvanted antigen or antigen. When challenged intranasally with a lethal dose of A. pleuropneumoniae, control mice displayed a significant reduction in BM cell numbers within 12 h of challenge alone (Figure 8a). Mice vaccinated with only antigen had slightly lower levels of cellular loss while group Vac mice exhibited the least loss in BM cells. This pattern of attenuated BM cell loss was reflected in the HPC population (Thy⁺ Lin^-; Figure 8b), as well as in the myeloid (Gr-1⁺ Mac-1⁺; Figure 8c,d) and pre-B (B220⁺) precursor populations (Figure 8e).

Figure 8.

Percentage shifts in (a) total bone marrow cells, and (b) Thy⁺Lin^-, (c) Gr-1⁺Mac-1⁺ and (d) Gr-1^-Mac-1⁺ (e) B220⁺ cell populations in the bone marrow at 12 h post-challenge for groups of unvaccinated mice (() group Con), and mice previously vaccinated with () Ag and () Vac, respectively.

Full figure and legend (23K)

Figure 9 and Figure 10 show the relative changes in cell numbers in the spleens and lungs of unvaccinated control mice and group Vac and group Ag mice. Group Vac mice responded to pathogenic challenge with significantly increased rates of recruitment of granulocytes/monocytes to the spleen and lung. The level of recruitment in these cell populations was not significantly different between group Ag and unvaccinated control mice.

Figure 9.

Percentage shifts in total cell numbers in (a) the spleen, and (b) granulocytes (Gr-1⁺Mac-1⁺) and (c) monocytes (Gr-1^-Mac-1⁺) in the spleen at 12 h post-challenge for groups of unvaccinated mice (() group Con), and mice previously vaccinated with () Ag and () Vac, respectively.

Full figure and legend (14K)

Figure 10.

Percentage shifts in total cell numbers in (a) the lung, (b) granulocytes (Gr-1⁺Mac-1⁺) and (c) monocytes (Gr-1^-Mac-1⁺) in the lung at 12 h post-challenge for groups of unvaccinated mice (() group Con), and mice previously vaccinated with () Ag and () Vac, respectively.

Full figure and legend (26K)

Skin response to intradermal immunization

No distinctive polarization of Th1 and Th2 responses in intradermally vaccinated skin

In phylogenetic terms, phagocytic cells involved in non-specific inflammatory reactions probably represent the earliest form of protective immunity. Garside and Mowat³⁴ proposed that further evolution of divergent non-specific responses to handle different types of infection have led to the polarization of the T helper response. Phagocytic macrophages that have been engaged by antigen and/or pathogen, produce potent cytokines such as IL-12. This cytokine activates both T and NK cells to secrete IFN-, thereby polarizing CMI towards Th1-dependent immunity within a local micro-environment. In contrast, encounters with antigen and/or pathogen at mucosal surfaces such as the lung rapidly led, within hours, to the recruitment of mast cells. These cells potentially polarize CMI towards Th2-dependent immunity by virtue of their capacity to produce very high levels of IL-4. Activated Th2 lymphocytes further produce IL-10 and this cytokine down-regulates Th2 cells by suppressing the production of macrophage-derived IL-12.³⁵ Reverse transcription–polymerase chain reaction (RT-PCR) analyses for Th1 and Th2 cytokines in mRNA prepared from skin biopsies taken from mice vaccinated with LIH/OMP displayed no such polarization with detectable levels of both IFN- and IL-4 and low levels of transforming growth factor (TGF-). In contrast, IL-4 predominated in the skin of mice vaccinated with only antigen (J Chin, unpubl. data).

Pro-inflammatory cytokines in the skin

Wheal and flare responses that accompany intradermal injection of skin and mucosal adjuvant 4 (SAMA4)-adjuvanted antigens in the skin are associated with massive infiltration of leucocytes to the delivery site. These inflammatory responses are short-lived, lasting between 2 and 5 days, and probably reflect local production of chemotactic cytokines such as IL-1 and TNF-. In a series of cell mixing experiments,³⁶ Schreiber et al. showed that keratinocytes are a major source of IL-1, granulocyte macrophage colony stimulating factor (GM-CSF) and TNF- while LC preferentially produce IL-6. These primary cytokines can, in turn, signal the production of a wider range of secondary cytokines including the 'C-X-C' class of chemokines such as ENA78 (epithelial neutrophil-activating protein-78), IP-10 (IFN- inducible protein 10), and GCP-2 (granulocyte chemotactic protein-2) which are potent chemotactic factors for polymorphonuclear leucocytes. Also produced in the skin are 'C-C' chemokines such as MCP-1/MCAF (monocyte-chemotactic and activating factor), MIP-1/MIP-1 (macrophage inflammatory proteins) and RANTES (regulated and normal T cell expressed and secreted) which are chemotactic for monocytes, basophils, eosinophils and mast cells.³⁷ Pro-inflammatory mediators in the adjuvant and/or antigen are most likely responsible for inducing the production of dermal cytokines. However, tissue injury arising from skin immunization may also signal the production of inflammatory cytokines. It is likely that the composition of cytokines in the dermal micro-environment may determine the developmental pathway of BM-derived precursor cells into different types of class II MHC⁺ cells in the skin.

Increased inflammatory responses in the skin elicited by synergism between adjuvant and antigen is responsible for enhancing immunity

Based on the results from the Ap intradermal immunization model, neither adjuvant nor antigen administration was able to induce apoptosis in the thymus and BM compartments even though pro-inflammatory components such as Quil A and LPS were present in the adjuvant and antigen preparations, respectively. At the same time, compared to adjuvanted antigens, single delivery of either adjuvant or antigen was unable to enhance the migratory activity of innate immune cells in various organ compartments. For instance, while the liposome/ISCOM adjuvant may provide the necessary inflammatory stimuli to induce the maturation of Langerhans or dendritic cells, they would still fail to accumulate large numbers of peptide-loaded MHC class II molecules and retain their ability to stimulate T cells³⁸ in the absence of antigen. It is not clear whether activated APC continue to migrate from the skin to secondary lymphoid organs but no T cell help would be elicited even if they did. It is tempting to speculate that the antibody responses elicited by intradermal immunization with OMP antigen is the result, in part, to inflammatory properties associated with LPS in the preparation. However, similar responses elicited by the non-inflammatory antigen, ovalbumin,³² suggest that activation of Langerhans cells in the skin may also arise by the production of pro-inflammatory cytokines initiated by tissue injury from the injection process. Clearly, inflammatory components in the adjuvant and antigen can synergistically interact to increase dermal inflammatory responses in the skin. This, in turn, promotes more effective recruitment of APC, enhances antigen presentation and migration of dentritic cells (DC) or LC to the peripheral lymph nodes, resulting in the activation of T cell help.³⁹

Although phagocytosis or pinocytosis of vehicular lipid adjuvants such as liposomes and ISCOM containing entrapped antigen may facilitate entry into the endosomal pathway, APC can utilize alternative mechanisms for antigen uptake. Increased plasma extravasation at sites of tissue injury lead to an accumulation of high levels of ₂-macroglobulin (₂-M).⁴⁰ Binding of the proteinase inhibitor ₂-M to tissue proteases generates a transient proteinase-activated state that allows ₂-M to entrap and form covalent linkages with different proteins. These complexes are readily endocytosed after binding of the complex to ₂-M receptors on macrophages and other cells. Such pathways have been shown to operate in vivo with concomitant enhancement of antibody production.⁴¹ Immune complexes formed by association of antigen with C3b or C4b can also deliver covalently bound antigen to APC.⁴² C3b and C4b, like ₂-M, are evolutionarily related thioester-containing proteins that undergo conformation and functional changes upon limited proteolysis.⁴³ Thioester-mediated covalent bonds with proteins, cell-surface carbohydrates or immune complexes are formed during such conformational change and provide an antigen source for APC. These different mechanisms provide a way for the innate immune system (via LC/APC) to recognize, process and present antigen with the appropriate costimulatory signal, and without the requirement that antigens are only internalized if entrapped in adjuvanted vehicles.⁴⁴ In contrast, administration of antigen by an intravenous route (which does not solicit antigen recognition by the innate immune system), results in tolerance because of deletion of T and B cell clones.⁴⁵

Only adjuvanted antigen and adjuvant but not antigen significantly down-regulate precursor T and B cells

Our studies show conclusively that only intradermal vaccination with adjuvanted antigens can induce disruption of lymphoid DP (CD4 and CD8) thymocytes in the thymus and B220⁺ progenitor B cells in the BM. The sharp decline in the two cell populations was not observed when mice were vaccinated with antigen or adjuvant alone. Such declines cannot be attributed to increased export of cells from the two organs because the disappearance of cells was not accompanied by an increase in these populations in the circulation (J. Chin unpubl. data). Therefore, it is likely that increased antigen loading in the skin, initiated by the synergistic interaction of adjuvant and antigen with APC, may generate signals that activate apoptosis in the thymus and BM. It is not clear at this point as to why only certain cell populations undergo apoptosis but this may partly be caused by the fact that both DP thymocytes and BM B220⁺ cells possess high levels of corticosteroid receptors and are, therefore, vulnerable to steroids generated by vaccine-induced stress.

The changes in haemotopoietic activity in the BM initiated by vaccination highlights its capacity to participate in an immune response. The BM can assume some of the functions of a secondary lymphoid tissue when splenectomized mice were first treated with Mel-14 mAb⁴⁶ to disrupt lymphocyte trafficking to lymph nodes, and then challenged with influenza virus. Disruption of the precursor T and B cell pool may generate the necessary signals to restore the imbalance by increasing production of Thy-1^lowLin^- HPC.

Skin exposure to antigen results in rapid depletion of LC and skin immune responses are compromised if these cells are not replenished.⁴⁷ The up-regulation of HPC in BM following intradermal immunization probably represents an important facet of vaccination as these pluripotent cells can refurbish the peripheral requirement for APC. Both murine and human epidermal LC are derived from precursors in the BM.⁴⁸

Intradermal vaccination with adjuvanted antigen increases haematopoietic progenitor cells in the bone marrow

The HPC in the BM are located within specific stroma niches.⁴⁹ Although HPC constitute only very small numbers of the total cell population in the BM, their biological activity is dependent upon the type of activation signals produced by stromal cells in the micro-environmental niche. Induction and expression of appropriate surface adhesion molecules would thus instruct HPC to proliferate, remain in the BM or emigrate. A recently characterized chemokine, stromal cell-derived factor (SDF-1), appears to fulfil an important role in HPC mobilization.⁵⁰ The changes seen in the HPC population of intradermally vaccinated mice may well represent a physiological response similar to peripheral mobilization of HPC after chemotherapy and/or exogenous cytokine administration. These events occur if vaccination affects the production of G-CSF and IL-3, two cytokines known to alter the mobilization kinetics of BM cells.⁵¹ The initial decrease in HPC at day 3 PI may reflect increased export of this population by supplying progenitors to the thymus and spleen. Over a longer time period (up to 21 days PI), increased production of HPC in the BM results in an oversupply of this population. Coincidentally, increased numbers of HPC in the BM of vaccinated mice are temporally associated with delayed increases in pro- and pre-B cells as well as myeloid precursors. These changes are most apparent following treatment with adjuvanted antigen and less so with either adjuvant or antigen alone. Therefore, the synergistic action of adjuvanted antigens can influence the BM compartment, most likely by regulating the production of specific cytokines.

Intradermal vaccination of mice with adjuvanted antigen attenuates loss of BM precursors and enhances granulocyte recruitment to the lung following lethal pathogenic challenge

A comparative study of kinetic changes of haemopoietic stem cells in the spleen and BM of mice infected with malaria, reported a relationship between increased production of multipotent stem cells and reduced lethality of the protozoan challenge strain.⁵² Virulent Plasmodium berghei decreased CFU-GM in the BM while avirulent Plasmodium yoelii significantly increased this population. As lethality occurred only 18 days after exposure, samples of BM were not analysed at shorter time intervals. In the case of A. pleuropneumoniae, which causes death within 24 h, a similar reduction in HPC was noted in unvaccinated mice as soon as 12 h post-challenge. The level of cell loss in mice previously vaccinated with antigen was only slightly less than that seen in unvaccinated controls. This destruction of BM cells including progenitor HPC was significantly attenuated in mice previously vaccinated with adjuvanted antigen. Reduced levels of cellular destruction in the BM of group Vac mice would most likely assist in maintaining haematopoiesis and enable these animals to recover from infection by turning on production of lymphocytes and myeloid/monocytic precursors sooner. Vaccination with adjuvanted antigen also appears to enhance innate immunity to lethal bacterial challenge by facilitating the recruitment of granulocytes and monocyte/macrophages to the spleen and lung more efficiently than unvaccinated controls or mice receiving only antigen. These findings indicate that it may now be possible to assess adjuvants not only upon the basis of their ability to enhance adaptive immune responses, but also to mobilize cellular changes in lymphoid organs and to alter rates of haematopoiesis in the BM.

References

1. Ertl HCJ, Xiang Z. Novel vaccine approaches. J. Immunol. 1996; 156: 3579–82. | PubMed | ISI | ChemPort |
2. Parrish CR, O'Neill ER. Dependence of the adaptive immune response on innate immunity: Some questions answered but new paradoxes emerge. Immunol. Cell Biol.. 1997; 75: 523–7. | PubMed | ChemPort |
3. Strassmann G, Kambayashi T, Jacob CO, Sredni D. The immunomodulator AS-101 inhibits IL-10 release and augments TNF and IL-1 release by mouse and human mononuclear phagocytes. Cell. Immunol. 1997; 176: 180–5. | Article | PubMed | ChemPort |
4. Herbert WJ. Methods for the preparation of water-in-oil and multiple emulsions for use as antigen adjuvants; and notes on their use in immunization procedures. In: Weir DM, Blackwell (eds). Oxford: Blackwell Science, 1967; 1207–17.
5. Cooper PD, Steele EJ. The adjuvanticity of gamma inulin. Immunol. Cell Biol. 1988; 66: 345–52. | PubMed | ChemPort |
6. Czop JK, Austen KF. A -glucan inhibitable receptor on human monocytes: Its identity with the phagocytic receptor for particulate activators of the alternative complement pathway. J. Immunol. 1985; 134: 2588–93. | PubMed | ISI | ChemPort |
7. Daynes RA, Araneo BA. Prevention and reversal of some age-associated changes in immunologic responses by supplemental dehydroepiandrosterone sulfate therapy. Ageing Immunol. Infect. Dis. 1992; 3: 135–54.
8. McKnight AJ. Effects of IL-12 on helper T cell-dependent immune responses in vivo. J. Immunol. 1994; 153: 2172–9.
9. Tao MH, Levy R. Idiotype/granulocyte-macrophage colony stimulating factor fusion protein as a vaccine for B-cell lymphoma. Nature 1993; 362: 755–8. | Article | PubMed | ISI | ChemPort |
10. Kalish ML. Murine IgG isotype responses to the Plasmodium cynomolgi circumsporozoite protein (NAGG): I. Effects of carrier, copolymer adjuvants and nontoxic LPS on isotype distribution. Infect. Immun. 1991; 59: 2750–7. | PubMed |
11. Ke Y, McGraw CL, Hunter R, Kapp JA. Non-ionic triblock copolymers facilitate delivery of exogenous proteins into the MHC Class I and Class II processing pathways. Cell. Immunol. 1997; 176: 113–21. | Article | PubMed | ChemPort |
12. Marx PA, Compans RW, Gettie A et al. Protection against vaginal SIV transmission with microencapsulated vaccine. Science 1993; 260: 1323–6. | Article | PubMed | ISI | ChemPort |
13. Elson CO, Holland SP, Dertzbaugh MT, Cuff CF, Anderson AO. Morphologic and functional alterations of mucosal T cells by cholera toxin and its B subunit. J. Immunol. 1995; 154: 1032–40. | PubMed | ChemPort |
14. Schmidt W, Buschle M, Zauner W et al. Cell-free tumor antigen peptide-based cancer vaccines. Proc. Natl Acad. Sci. USA 1997; 94: 3262–7. | Article | PubMed | ChemPort |
15. Plebanski M, Burtles SM. In vitro primary responses of human T cells to soluble protein antigens. J. Immunol. Methods 1994; 170: 15–25. | Article | PubMed | ISI | ChemPort |
16. Schafer R, Pornoy DA, Brassell SA, Paterson Y. Induction of a cellular immune response to a foreign antigen by a recombinant Listeria monocytogenes vaccine. J. Immunol. 1992; 149: 53–9. | PubMed | ISI | ChemPort |
17. De Bruijn MLH, Jackson MR, Peterson P. Phagocyte-induced antigen-specific activation of unprimed CD8⁺ T cells in vitro. Eur. J. Immunol. 1995; 25: 1274–85. | PubMed | ChemPort |
18. Harding CV, Collins DS, Kanagawa O, Unanue ER. Liposome-encapsulated antigens engender lysosomal processing for class II MHC presentation and cytosolic processing for class I presentation. J. Immunol. 1991; 147: 2860–6. | PubMed | ChemPort |
19. Sjolander A, Van T, Land B, Bengtsson KL. ISCOM containing purified Quillaja saponins upregulate both Th1-like and Th2-like immune responses. Cell. Immunol. 1997; 177: 69–76. | Article | PubMed | ChemPort |
20. Sedlik C, Perraut R, Bonnemains B, Leclerc C. Antigens linked to synthetic microspheres induce immune responses in primates in the absence of adjuvant. Immunobiology 1996; 195: 105–18. | PubMed | ChemPort |
21. Chin JC, San Gil F, Novak M et al. Manipulating systemic and mucosal immune responses with skin-deliverable adjuvants. J. Biotechnol. 1996; 44: 13–19. | Article | PubMed | ChemPort |
22. Levitz RE, Cooper BW, Regan HC. Immunization with high-dose intradermal hepatitis B vaccine in healthcare workers who failed to respond to intramuscular vaccination. Infect. Control Hosp. Epidemiol. 1995; 16: 88–91. | PubMed | ChemPort |
23. Raz E, Carson DA, Parker SE et al. Intradermal gene immunization: The possible role of DNA uptake in the induction of cellular immunity to viruses. Proc. Natl Acad. Sci. USA 1994; 91: 9519–23. | Article | PubMed | ChemPort |
24. Stingl G, Tamaki K, Katz SI. Origin and function of epidermal Langerhans cells. Immunol. Rev. 1980; 53: 149–74. | Article | PubMed | ISI | ChemPort |
25. Cohen PJ, Katz SI. Cultured human Langerhans cells process and present intact protein antigens. J. Invest. Dermatol. 1992; 99: 331–6. | Article | PubMed | ChemPort |
26. Hoefsmit ECM, Duijvestijn AM, Kamperkijk EWA. Relation between Langerhans cells, veiled cells and interdigitating cells. Immunobiology 1982; 161: 255–65. | PubMed | ISI | ChemPort |
27. Dandie GW, Watkins FY, Ragg SJ, Holloway PE, Muller HK. The migration of Langerhans cells into and out of lymph nodes draining normal, carcinogen and antigen-treated sheep skin. Immunol. Cell Biol. 1994; 72: 79–86. | PubMed | ISI | ChemPort |
28. Wright SD, Ramos RA, Tobias PS, Ulevitch RJ, Mathison JC. CD14, a receptor for complexes of LPS and LPS binding protein. Science 1990; 249: 1431–4. | Article | PubMed | ISI | ChemPort |
29. Wright SD, Levin SM, Jong MTC, Chad Z, Kabbash G. CR3 (CD11b/CD18) expresses one binding site for Arg-Gly-Asp-containing peptides and a second site for bacterial lipopolysaccharide. J. Exp. Med. 1989; 169: 175–81. | Article | PubMed | ChemPort |
30. Hampton RY, Golenbock DT, Penman M, Krieger M, Raetz CRH. Recognition and plasma clearance of endotoxin by scavenger receptors. Nature 1991; 352: 342–4. | Article | PubMed | ISI | ChemPort |
31. Roeder DJ, Lei MG, Morrison DC. Endotoxic lipopolysaccharide-specific binding proteins on lymphoid cells of various animal species: Association with endotoxin susceptibility. Infect. Immun. 1989; 57: 1054–8. | PubMed | ChemPort |
32. San Gil F, Turner B, Mullbacher A et al. Flow cytometric analysis of cellular changes in mice after intradermal inoculation with a liposome/ISCOM adjuvanted vaccine. Scand. J. Immunol. 1998; (in press).
33. Spangrude GJ, Heimfeld S, Weissman IL. Purification and characterization of mouse hematopoietic stem cells. Science 1988; 241: 58–62. | Article | PubMed | ISI | ChemPort |
34. Garside P, Mowat A. Polarization of Th-cell responses: A phylogenetic consequence of non-specific immune defence? Immunol. Today 1995; 16: 220–3. | Article | PubMed | ChemPort |
35. D'Andrea A, Aste-Amezaga M, Valiante NM, Ma X, Kubin M, Tinchieri G. Interleukin 10 inhibits human lymphocyte interferon gamma production by suppressing natural killer cell stimulatory factor/IL-2 synthesis in accessory cells. J. Exp. Med. 1993; 178: 1041–8. | Article | PubMed | ChemPort |
36. Schreiber S, Kilgus O, Payer E et al. Cytokine pattern of Langerhans cells isolated from murine epidermal cell cultures. J. Immunol. 1992; 149: 3525–34.
37. Schroder J-M. Cytokine networks in the skin. J. Invest. Dermatol. 1995; 105: 20S–4S. | Article | PubMed | ChemPort |
38. Cella M, Engering A, Pinet V, Pieters J, Lanzavecchia A. Inflammatory stimuli induce accumulation of MHC class II complexes on dendritic cells. Nature 1997; 388: 782–7. | Article | PubMed | ISI | ChemPort |
39. Watts C. Inside the gearbox of the dendritic cell. Nature 1993; 362: 755–8. | Article |
40. Giannopoulou C, Di Felice R, Andersen E, Cimasoni G. Synthesis of 2-macroglobulin in human gingiva: A study of the concentration of macroglobulin and albumin in gingival fluid and serum. Arch. Oral Biol. 1990; 35: 13–18. | PubMed | ChemPort |
41. Chu CT, Oury TD, Enghild JJ, Pizzo SV. Adjuvant-free in vivo targeting: Antigen delivery by ₂-macroglobulin enhances antibody formation. J. Immunol. 1994; 152: 1538–45. | PubMed | ChemPort |
42. Arvieux J, Yssel H, Colomb MG. Antigen-bound C3b and C4b enhance antigen-presenting cell function in activation of human T-cell clones. Immunology 1988; 65: 229–36. | PubMed | ISI | ChemPort |
43. Sottrup-Jensen L, Petersen TE, Magnusson. Trypsin-induced activation of the thiol esters in ₂-macroglobulin generates a short-lived intermediate that can react rapidly to incorporate not only methylamine or putresine but also proteins lacking proteinase activity. FEBS Lett. 1981; 128: 123–6. | Article | PubMed | ChemPort |
44. Medzhitov R, Janeway CA. Innate immunity: Impact on the adaptive response. Curr. Opin. Immunol. 1997; 9: 4–9. | Article | PubMed | ISI | ChemPort |
45. Bretscher P, Cohn M. A theory of self–nonself discrimination: Paralysis and induction involve the recognition of one and two determinants on an antigen, respectively. Science 1970; 163: 1042–9.
46. Tripp RA, Topham DJ, Watson SR, Doherty PC. Bone marrow can function as a lymphoid organ during a primary immune response under conditions of disrupted lymphocyte trafficking. J. Immunol. 1997; 158: 3716–20. | PubMed | ISI | ChemPort |
47. Wood GM, Qu M, Ragg SJ, Muller HK. Chemical carcinogens and antigens induce immune suppression via Langerhans cell depletion. Immunology 1996; 88: 134–9. | PubMed |
48. Cooper KD, Breathnach SM, Caughman SW, Palini AG, Waxdal MJ, Katz SI. Fluorescence microscopic and flow cytometric analysis of bone marrow-derived cells in human epidermis: A search for the human analogue of the murine dendritic Thy-1⁺ epidermal cell. J. Invest. Dermatol. 1985; 85: 546–51. | Article | PubMed | ChemPort |
49. Schofield R. The relationship between the spleen colony-forming cell and the haemopoietic stem cell. Blood Cells 1978; 4: 7–25. | PubMed | ISI | ChemPort |
50. Aiuti A, Webb IJ, Bleul C, Springer T, Gutierrez-Ramoz JC. The chemokine SDF-1 is a chemoattractant for human CD34⁺ hematopoietic progenitor cells and provides a new mechanism to explain the mobilization of CD34⁺ progenitors to peripheral blood. J. Exp. Med. 1997; 185: 111–20. | Article | PubMed | ISI | ChemPort |
51. Guillaume T, D'Hondt V, Symann M. IL-3 and peripheral blood stem cell harvesting. Stem Cells 1993; 11: 173–81. | PubMed | ChemPort |
52. Asami M, Owhashi M, Abe T, Nawa Y. A comparative study of the kinetic changes of hemopoietic stem cells in mice infected with lethal and non-lethal malaria. Int. J. Parasitol. 1992; 22: 43–7. | Article | PubMed | ChemPort |
53. Vogel FR, Powell MF. A compendium of vaccine adjuvants and excipients. In: Powell MF, Newman MJ (eds). New York: Plenum Publishing, 1–92.

Acknowledgements

The authors would like to thank Drs G Eamens and S Djordjevic (NSW Agriculture) for their continued interest and participation in the use of the skin as a route for vaccine delivery in pigs. Experimental findings from many pig vaccine trials have provided the impetus to elucidate the mechanism of adjuvant action in a mouse disease model and this has been facilitated by support from the laboratories of Drs M Walker (University of Wollongong) and A Mullbacher (Australian National University). We would also like to thank Johnson & Johnson for initially funding this research.