Administration of exogenous interleukin-18 and interleukin-12 prevents the induction of oral tolerance

Abstract

Interleukin-18 (IL-18), a pro-inflammatory member of the IL-1 family, has been associated with autoimmunity and allergic disease. This type of pathology is thought to be the result of a defect in immunological tolerance and is often observed in inflammatory disorders of the gut such as coeliac disease, Crohn's disease and ulcerative colitis. IL-18 has been implicated in a number of mucosal immune disorders, where it synergizes with IL-12 to induce the production of interferon-γ (IFN-γ). We have therefore investigated the effects of IL-18 and IL-12 administration on the induction of oral tolerance to ovalbumin. The suppression of specific Ig G2a production, delayed-type hypersensitivity responses and IFN-γ production by antigen-specific T cells were all abrogated by the presence of exogenous IL-12 and IL-18, suggesting that oral tolerance was broken. The expression of the co-stimulatory molecule CD80 on dendritic cells was also shown to be increased by this combination of cytokines. As dendritic cells are thought to be of major importance in the induction of tolerance, this suggests a mechanism by which tolerance to mucosal antigens may be broken in vivo.

Introduction

Interleukin-18 (IL-18) is a pro-inflammatory cytokine that appears to play a role in immunological priming intermediate between the effects of the IL-1 and IL-12 families.^1,2 Although sharing structural and downstream signalling components with IL-1, this cytokine can synergize with IL-12 (amongst others) to enhance the production of interferon-γ (IFN-γ) by the many cells of the immune system which express the IL-18 receptor (IL-18R). These include macrophages, natural killer (NK) cells, dendritic cells and most T cells, with the notable exception of T helper type 2 (Th2) cells.³ IL-18 can potentiate IFN-γ production via the up-regulation of IL-12R expression,⁴ and also by induction of IFN-γ transcription.⁵ Synergy of IL-18 with numerous cytokines can lead to production of either Th1 cytokines from NK cells and antigen-presenting cells (APCs) or Th2 cytokines from Th0 cells, as well as directly or indirectly inducing migration of epidermal Langerhan's cells⁶ and enhancing antigen presentation.

A number of studies have noted increased expression of IL-18 in intestinal epithelial cells of patients suffering from Crohn's disease,^7,8 and neutralization of IL-18 has been shown to reduce the severity of murine colitis.⁹ High levels of IL-18 have also been implicated in the pathogenesis of coeliac disease, a strongly Th1-related inflammatory disorder in which IL-12 remains undetectable in mucosal tissue.¹⁰ It is therefore plausible that the presence of IL-18, either alone or with IL-12, could contribute to the induction of such diseases via prevention of the oral tolerance normally associated with mucosally administered antigens.

Oral tolerance, an antigen-specific suppression of immune responsiveness, can be induced by administration of antigen via the oral route. The induction of immunological tolerance in this and other systems is believed to result from the presentation of antigen to T cells (signal 1) in the absence of co-stimulation (signal 2). For example, the absence of co-stimulatory signalling from an APC (for example a dendritic cell) to a naive antigen-specific T cell is thought to be the main factor in preventing the secretion of IL-2 by the T cell, leading to the development of an anergic, non-proliferative state.¹¹ In contrast, membrane-bound co-stimulatory molecules can be up-regulated on the surface of APCs by inflammatory cytokines, the presence of which is often associated with immunological priming, or diversion away from tolerance. For example, in mucosal systems the balance between IL-10/transforming growth factor-β and IFN-γ/tumour necrosis factor-α (TNF-α) can switch between tolerant and primed responses to a newly encountered antigen.¹²

Inflammatory cytokines of the IL-1 and TNF families, when co-administered with antigen, can cause an increased and prolonged primary immune response, and are also able to enhance a secondary, memory response.^13,14 This raises the possibility that these cytokines may be intercellular mediators of the inflammatory effect of adjuvants such as lipopolysaccharide or bacterial endotoxins, increasing the activity of transcription factors involved in promoting a Th1 or Th2 phenotype. Concurring with this, administration of either IL-1 or IL-12 at the same time as feeding antigen has been shown to prevent the induction of oral tolerance to that antigen,¹⁵ and in some cases this has been proposed to occur via activation of dendritic cells.^14–16 These cytokines have also been shown to be up-regulated during inflammatory responses to mucosal antigens.

IL-12, while not a pro-inflammatory cytokine per se, is necessary in most circumstances for the development of a type 1 response, characterized by IFN-γ production.¹⁷ While the role of IL-12 seems to be focused on inducing IFN-γ expression by responding T and NK cells, the role of IL-18 appears to be broadly pro-inflammatory, as it is present in epithelial wounds,¹⁸ sites of chronic and acute inflammation and during antigen presentation.¹⁹

As IL-18 has been implicated in the loss of oral tolerance to mucosal antigens in many clinical and model inflammatory disorders, we hypothesized that IL-18, working in combination with IL-12, may be able to abrogate the induction of oral tolerance to ovalbumin (OVA) in BALB/c mice. The possible mechanisms involved could include effects on antigen-presenting dendritic cells, antigen-responding lymphocytes and/or the surrounding tissues.

We show here that the concurrent presence of elevated levels of IL-12 and IL-18 has a marked suppressive effect on the induction of antigen-specific tolerance, and is able to promote priming of the immune system, provoking a more vigorous response on challenge. We also suggest that this may occur partly by augmenting IFN-γ production, as well as increasing activation of antigen-presenting dendritic cells.

Materials and methods

Animals

Six to twelve-week-old BALB/c mice were purchased from Harlan-Olac (Bicester, UK) and homozygous DO11.10 BALB/c mice (expressing the DO11.10 T-cell receptor, specific for chicken OVA peptide 323–339 in I-Ad) were maintained under specific pathogen-free standard animal house conditions at the University of Glasgow Central Research Facility in accordance with Home Office regulations.

Materials

Grade V chicken egg OVA, Complete Freund's Adjuvant (CFA), collagenase and sodium azide were purchased from Sigma-Aldrich (Dorset, UK). Cell culture reagents were purchased from Gibco BRL/Invitrogen (Paisley, UK). Immulon-4, Immulon-2 plates, TMB Microwell Substrate, MRX plate reader and the Revelation software were purchased from Dynex/Thermo Labsystems (Basingstoke, UK). Rat anti-mouse CD11c-phycoerythrin, rat anti-mouse IFN-γ, biotinylated rat anti-mouse IFN-γ and biotinylated anti-mouse immunoglobulin G2a (IgG2a) were purchased from BD Pharmingen (Oxford, UK). Biotinylated anti-mouse IgG1 heavy chain was purchased from Serotec (Oxford, UK). Mouse serum and streptavidin–horseradish peroxidase (streptavidin–HRP) were provided by Diagnostics Scotland (Edinburgh, UK). fluorescence-activated cell sorting (FACS) equipment, buffers and Cellquest software were purchased from BD Immunocytometry Systems (Oxford, UK).

Antigen and cytokine administration in vivo

For oral tolerization/priming, BALB/c mice were fed once with 200 μl phosphate-buffered saline (PBS) ± 100 mg of chicken OVA, using a stainless steel gavage needle. To administer cytokines, mice were injected intraperitoneally with 100 μl physiological saline ± IL-12 (R & D Systems, Abingdon, UK) (50 ng per dose per animal) ± IL-18 (PeproTech, London, UK) (200 ng per dose per animal), at the time-points described. For antigenic challenge, mice were injected subcutaneously in the scruff of the neck with 100 μg OVA in 100 μl 50% CFA. For delayed-type hypersensitivity (DTH) testing 14 days after challenge with OVA in CFA, all animals were injected subcutaneously in one rear footpad with 100 μl physiological saline containing 50 ng heat-aggregated OVA.²⁰ After 24 and 48 hr, the increase in footpad thickness was measured using standard skinfold callipers as previously described, and calibrated against the response to heat-aggregated OVA in a previously untreated (naïve) control group.

Sample collection

For serum collection, blood was collected from the tail vein by capillary tube, and centrifuged for 10 min at 450 g to separate the serum, which was removed and stored at − 20° until analysis.

For cell culture and analysis, peripheral lymph nodes (axillary, inguinal and cervical), and/or spleens were removed, passed through Nitex mesh (Cadisch Precision Meshes, London, UK) to create a single-cell suspension, then washed and resuspended in complete medium (RPMI-1640, 2 mm l-glutamine, 100 U/ml penicillin, 100 mg/ml streptomycin, 50 μm 2-mercaptoethanol, 10% fetal calf serum (FCS), 1·25 mg/ml fungizone). Complete medium ± 1 mg/ml OVA ± 100 ng/ml IL-18 was added to the appropriate wells for restimulation. Cells were cultured in 1 ml/well of 24-well plates, at 8 × 10⁶ cells/ml in a 37° incubator under 5% CO₂.

Dendritic cells

In vitro BALB/c mice were used as donors of bone marrow for dendritic cell culture, which was removed as previously described. Cells were cultured in 24-well plates at 4 × 10⁵ cells/ml in 1 ml of complete medium (without fungizone) plus 5% granulocyte–macrophage colony-stimulating factor (GM-CSF) (X63 hybridoma culture supernatant) in a 37° incubator under 5% CO₂. An equal volume of GM-CSF-containing medium was added after 2 days. Three days later, IL-12 (25 ng/ml) and/or IL-18 (100 ng/ml) were added to the cell culture. On the appropriate days following cytokine treatment, cells were removed from the plates by gentle pipetting and were stained for FACS analysis as described below.

Ex vivo. BALB/c mice were injected intraperitoneally with 100 μl physiological saline ± IL-12 (50 ng per animal) ± IL-18 (200 ng per animal). After 19 hr the animals were killed. Spleens were removed, chopped into small pieces, incubated in collagenase-containing RPMI-1640 for 30 min, then passed through Nitex to create a single-cell suspension. These cells were washed in RPMI-1640 then stained for FACS analysis as described below.

Enzyme-linked immunosorbent assay (ELISA)

ELISAs were performed in accordance with the instructions provided by the antibody manufacturer (Pharmingen). Plates were washed with wash buffer (PBS, 0·05% Tween-20) between each step. Samples, standards and antibodies were diluted in dilution buffer (PBS, 0·05% Tween-20, 0·2% FCS).

For cytokines, Immulon-4 plates were coated with primary capture antibodies in sodium carbonate buffer (PBS, 0·1 m sodium bicarbonate, pH 9·5) for 2 hr, then blocked with 10% FCS in PBS for 1 hr. Samples and standards were added in triplicate for 2 hr, followed by secondary detection antibodies for 1 hr, then streptavidin–HRP for 30 min. TMB Microwell Substrate was added, left to develop, and read at 605 nm using a Dynex MRX plate reader and Revelation software. All incubations were performed at room temperature.

For antibodies, Immulon-2 plates were coated with 100 μg/ml OVA in PBS overnight at 4°, blocked with 10% FCS in PBS for 1 hr at room temperature, then incubated with samples overnight at 4°. Detection antibodies were added for 2 hr at room temperature, streptavidin–HRP was added for 30 min at room temperature, then TMB Microwell Substrate was added and analysed as above.

Flow cytometry

Cells were washed and resuspended in Fc block (anti-CD16/32 hybridoma supernatant, 10% mouse serum, 0·1% sodium azide) for 10 min. Primary antibodies were added in FACS buffer (PBS, 2% FCS, 0·05% sodium azide) for 15 min. If necessary, cells were washed and resuspended with secondary antibodies in FACS buffer for a further 15 min. Cells were then washed and resuspended in FACS Flow for analysis using a FACSCalibur flow cytometer, CellQuest acquisition software and FlowJo analysis software (Treestar Inc, San Carlos, CA).

Results

Since it has been implicated in the pathogenesis of autoimmune and other inflammatory disorders, we hypothesized that IL-18, working in combination with IL-12, may be able to abrogate the induction of oral tolerance. This hypothesis was tested using oral tolerance to OVA in BALB/c mice as a model in vivo system, then dissected further using in vitro cell culture of dendritic cells and lymphocytes, and flow cytometric analysis of antigen-presenting cell surface molecule expression.

Firstly, we examined the effects of intraperitoneal IL-12 and IL-18 administration concurrent with OVA feeding on the development of subsequent immune responses, using OVA-specific DTH and antibody production as a readout of responsiveness to challenge.

Effects of IL-12 and IL-18 on oral tolerance induction in vivo

To induce oral tolerance, groups of five mice were fed 100 mg OVA in PBS on day 0, while control animals were fed PBS alone. On days − 1, 0 and + 1, the appropriate groups of OVA-fed mice were injected intraperitoneally with 50 ng IL-12, 200 ng IL-18, or a combination of both cytokines, while control OVA-fed mice were injected with saline alone. Fourteen days after feeding, all groups were injected subcutaneously in the scruff of the neck with 100 μg OVA in a 50% CFA solution. 15 and 22 days after challenge, blood samples were taken and serum was analysed for the presence of OVA-specific IgG1 and IgG2a by ELISA.

By day 15 after challenge, the OVA-specific antibody titres produced by the saline-fed animals and OVA-fed animals remained low, as did the levels of antibody produced by animals that were fed OVA and injected with IL-12 or IL-18 alone (which were not significantly lower than the OVA-fed, saline-treated group) (Fig. 1a,b). In contrast, the OVA-fed group that received both IL-12 and IL-18 (Th1-primed) had produced levels of OVA-specific IgG2a that were significantly higher than controls, thus demonstrating OVA-specific priming in this group. By day 22 (Fig. 1c,d), the control group (producing a primary response) had reached a level of antibody production equivalent to the IL-12 + IL-18-treated group, while the remaining OVA-fed groups remained lower in both OVA-specific IgG1 and IgG2a than controls, indicating that they were tolerized.

Figure 1.

The combined presence of IL-12 and IL-18 prevents the tolerogenic effect of feeding OVA on antigen-specific antibody production. BALB/c mice were fed either saline alone (control) or 100 mg OVA on day 0. On days − 1, 0 and + 1, the appropriate groups of OVA-fed mice were injected intraperitoneally with 50 ng IL-12, 200 ng IL-18, a combination of both cytokines, or saline alone. Fourteen days after feeding, all groups were injected subcutaneously in the scruff of the neck with 100 μg OVA in 50% CFA solution. 15 (a,b) and 22 (c,d) days after challenge, tail vein blood samples were taken and serum was analysed for the presence of OVA-specific IgG1 (a,c) and IgG2a (b,d) antibodies by ELISA. Results show the group means, and asterisks indicate significant differences (P < 0·05) from the control group (day 15) or the OVA-fed, saline-treated group (day 22), using Student's unpaired t-test (a,c,d: n = 5; b: n = 10).

As an alternative in vivo readout of antigen-specific tolerance and priming, groups of five animals were treated as described above, then 16 days after challenge with OVA/CFA they received 50 ng of heat-aggregated OVA in one rear footpad. The subsequent increase in footpad thickness was measured with standard skinfold callipers 24 hr and 48 hr after injection, compared to the thickness immediately before injection and calibrated relative to a group of footpad-immunized, but previously untreated controls. The footpad swelling seen in the primed control (PBS-fed) group in response to heat-aggregated OVA injection was matched at 24 hr by all the groups apart from the OVA-fed, tolerized group, which had significantly decreased levels of footpad swelling (Fig. 2a). At 48 hr, the tolerized group remained lower than controls, while the OVA-fed, IL-12 + IL-18-treated group remained significantly higher than the tolerized group (Fig. 2b), suggesting that the tolerance induced by feeding OVA was broken by the presence of these cytokines.

Figure 2.

The combined presence of IL-12 and IL-18 prevents the tolerogenic effect of feeding OVA on antigen-specific DTH responses and antigen-specific IFN-γ production. BALB/c mice were fed either saline alone (control) or 100 mg OVA on day 0. On days − 1, 0 and + 1, the appropriate groups of OVA-fed mice were injected intraperitoneally with 50 ng IL-12, 200 ng IL-18, a combination of both cytokines, or saline alone. Fourteen days after feeding, all groups were injected subcutaneously in the scruff of the neck with 100 μg OVA in 50% CFA solution. Sixteen days after challenge, all groups received 50 ng of heat-aggregated OVA in one rear footpad. The increase in footpad thickness was measured with standard skinfold callipers 24 hr (a) and 48 hr (b) after injection, the thickness immediately before injection was subtracted, and the resulting measurement was calibrated relative to a group of previously untreated controls (by subtracting the footpad swelling in the naïve group from the swelling in the other groups). Results show mean ± SEM (n = 5), and asterisks indicate significant differences (P < 0·05) from the OVA-fed, saline-treated group using Student's unpaired t-test. (c) Fifteen days after oral treatment and challenge with OVA/CFA as above, peripheral lymph nodes were removed from groups of three mice, and single cell suspensions were restimulated in vitro with 1 mg/ml OVA. Supernatants were removed at time-points over the next 96 hr, and analysed by ELISA for the presence of IFN-γ. Results show the group medians of ELISAs run in triplicate. Asterisks indicate significant differences (P < 0·05) from the OVA-fed, saline-treated group using the Mann–Whitney U-test.

In order to test in vitro for the in vivo priming or tolerization of OVA-specific T cells, groups of three animals were treated as above, then killed 15 days after challenge with OVA/CFA for removal of peripheral lymph nodes. Single cell suspensions were restimulated with 1 mg/ml OVA for up to 96 hr and supernatants were analysed by ELISA. The combined presence of IL-12 and IL-18 during OVA feeding was shown to promote a significantly higher level of IFN-γ production on antigen restimulation than in the OVA-fed control group, which themselves produced a significantly lower (tolerized) level of IFN-γ than the PBS-fed control animals (Fig. 2c).

Effects of IL-18 on co-stimulatory molecule expression by dendritic cells

There are a number of possible mechanisms by which the combination of increased levels of IL-12 and IL-18 could act to affect the induction of oral tolerance as seen above. These include activation of the innate immune system, particularly NK cells;²¹ effects on non-immune tissues,²² particularly disruption of the gastrointestinal epithelial barrier; effects directly on the differentiating T cells^23,24 or activation of the APCs.^6,15 It has been hypothesized that tolerance may be due to the absence of co-stimulatory molecules during antigen presentation, expression of which is normally up-regulated by adjuvants to induce priming.¹⁴ As APCs, particularly dendritic cells in a primary response, are thought to be the key co-ordinators of an ensuing response, we investigated the role of these cytokines in activating dendritic cells for efficient antigen presentation.

In vitro

The co-stimulatory molecules CD80 and CD86, ligands of T-cell surface molecules CD28 and CD152, are recognized as playing a major role in the induction of a primary immune response by dendritic cells. Using bone marrow-derived murine dendritic cells grown in GM-CSF-containing medium, we examined the effects of IL-18 and IL-12 on the expression of CD80 and CD86.

Bone marrow cells from BALB/c mice were cultured for 5 days in medium containing GM-CSF. 25 ng/ml IL-12 ± 100 ng/ml IL-18 was added and the cell-surface expression of CD80 and CD86 was analysed by flow cytometry. It was found that the joint presence of IL-18 and IL-12 was able to increase the level of CD80, without a large increase in the surface expression of CD86 (Fig. 3a,b).

Figure 3.

The combined presence of IL-12 and IL-18 enhances the surface expression of CD80 on bone marrow-derived dendritic cells. (a,b) Bone marrow cells from BALB/c mice were cultured in medium containing 5% GM-CSF. After 5 days, 25 ng/ml IL-12 ± 100 ng/ml IL-18 were added, and the cell-surface expression of CD80 (a) and CD86 (b) was analysed by flow cytometry over the following 4 days. (c,d) BALB/c mice were injected intraperitoneally with 50 ng IL-12 and/or 200 ng IL-18, followed by extraction of dendritic cells from the spleen 19 hr later, which were then analysed by flow cytometry for expression of co-stimulatory molecules CD80 (c) and CD86 (d). Asterisks indicate significant differences (P < 0·05) from control animals, analysed individually using the χ² sample analysis function of FlowJo. In vitro results are representative of repeated experiments, while in vivo results show the mean ± range (n = 2).

In vivo

We also tested the same combination of cytokines in order to confirm these in vitro findings directly in vivo. Intraperitoneal injection of IL-12 (50 ng) and/or IL-18 (200 ng) into BALB/c mice was followed by extraction of splenocytes 19 hr later, which were then analysed by flow cytometry for expression of co-stimulatory molecules on CD11c⁺ dendritic cells. As seen in vitro, CD80 was the most up-regulated, with a lesser increase in CD86 (Fig. 3c,d). The small but significant and reproducible increase in CD80 was dependent on administration of both IL-12 and IL-18, reinforcing the conclusion that the combined presence of both of these cytokines is necessary to be able to prevent the induction of oral tolerance.

Discussion

The induction of oral tolerance by mucosal delivery of soluble antigen has been an observed phenomenon for centuries, and has undergone many clinical trials in an attempt to prove its efficacy for therapeutic treatment of antigen- or organ-specific disorders of the immune system.²⁵ However, oral delivery of antigen can also lead to immunological priming in the presence of an adjuvant.²⁶ While this is desirable in the context of oral vaccination, this effect can cause immunopathology. For example, the presence of gastrointestinal inflammation, stress, helminth infection,²⁷ or bacterial toxins during the first encounter with an ingested antigen can provoke an immune response. Unfortunately, this can often cause problems with reactions to harmless food antigens or commensal bacteria, as has been implicated in inflammatory disorders of the intestine, where a breakdown of oral tolerance to self, food, or bacterial antigens is commonly found.^28,29

The evidence implicating IL-18 in various disease states is compelling, particularly as production of this cytokine can lead to secretion of TNF-α³⁰ and IFN-γ³¹ and release of nitric oxide.³² The association between the proximity of IL-18 production and pancreatic β cells³³ in diabetes suggests that IL-18 may prevent closure or regulation of immune- or pathogen-mediated damage and allow inflammatory reactions to continue to a state of chronic disease. Similarly, in Crohn's disease, the high levels of IL-18 found in patients are indicative of chronic inflammatory processes. In the intestinal mucosa, IL-18 and/or downstream mediators may also cause defects in the structure of the epithelial barrier, leading to the influx of large antigenic molecules and bacterial products, which could aggravate pre-existing pathology. The localization of IL-18 production to the lamina propria epithelium⁸ is circumstantial evidence of a primary role for this cytokine in mucosal inflammatory pathogenesis and, as shown in colitis studies, concurrent administration of both IL-12 and IL-18 is sufficient (independent of TNF-α) for the induction of intestinal inflammation.³¹ Coeliac disease, a strongly Th1-based deviation of immune responses to dietary proteins in the small intestine, is also associated with high levels of IL-18.¹⁰ Furthermore, blocking the presence or effects of IL-18 has been shown to be beneficial in many models of autoimmunity.^34,35

In our current studies, we have been able to show that the presence of IL-12 and IL-18, when administered intraperitoneally at time-points surrounding OVA feeding, is able to prevent the normal induction of oral tolerance. This influence is manifest as increased levels of IgG2a after challenge with OVA in CFA, and as abrogation of the reduced DTH response caused by oral tolerance. These effects suggest that a vaccination regime that includes both IL-12- and IL-18-enhancing factors could hope to prime successfully the mucosal immune system. At the same time, this also suggests that any therapy aimed at reducing mucosal inflammation such as that seen in Crohn's disease or colitis should aim to reduce the high levels of these cytokines seen both locally^7,36 and systemically in the disease state. This may involve suppressing the production of IL-18 by gastrointestinal epithelial or APCs, or promoting the secretion of IL-18 binding proteins³⁷ and/or receptor antagonists.

We have also aimed to show a mechanism by which the presence of increased levels of IL-12 and IL-18 may be involved in the induction phase of immunological priming. In order to prevent the induction of tolerance/anergy in T cells, a combination of inflammatory cytokines may be expected to up-regulate the expression of co-stimulatory molecules on APCs (particularly dendritic cells), as has been shown for IL-1α.¹⁵ Indeed, we show here that the level of CD80 is significantly increased on bone marrow-derived murine dendritic cells over the days following exposure to IL-12 and IL-18, though it should be noted that this increase is much smaller than that seen following lipopolysaccharide exposure. Similar investigations have previously shown an effect of IL-12 alone on CD80 expression by dendritic cells in vitro,³⁸ although in that case the cell culture medium was supplemented with TNF-α.

We also show that this co-stimulation-promoting effect of IL-12 and IL-18 can be seen on dendritic cells extracted from the spleen after cytokine injection into the peritoneum. The spleen is an important site of antigen presentation for both orally and intraperitoneally administered molecules, as it filters the bloodstream draining from both of these areas. Also, proliferation of antigen-specific T cells has been detected in both mucosal and peripheral lymphoid organs following oral administration of OVA.^39,40 This effect on antigen presentation in the spleen is therefore relevant to both oral (mucosal) and systemic tolerance – an effect on co-stimulatory molecules in the spleen could be important in the transfer of the effects of mucosal antigen encounter to systemic responses.

Variations in the relative levels of CD80 and CD86 have yet to be satisfactorily linked to particular effects on the immune system, possibly because they share many redundant features.⁴¹ Effects of signalling via these co-stimulatory molecules on the deviation of responding T cells towards Th1, Th2, or anergic phenotypes has been noted in some circumstances,⁴² as has a slight preference of each molecule for either CD28 (activating) or CD152 (suppressive) on the surface of APCs. A combination of these effects, along with members of the other co-stimulatory molecule families, may be responsible for the switch between tolerance and priming.

In summary, as we have shown that priming effects can be seen following intraperitoneal administration of these cytokines in vivo, above the level induced by each cytokine alone (even with the presumed presence of endogenous IL-12 or IL-18), it is proposed that the combined up-regulation of these cytokines is able to divert significantly the immune response to a mucosally administered antigen away from antigen-specific tolerance. Further studies of the molecular mechanism of this combined effect could be expected to expose further the similarities between exposure to mucosal adjuvants and the specific effects of IL-12 and IL-18.

Abbreviations

APC

antigen-presenting cell

CFA

complete Freund's adjuvant

DTH

delayed-type hypersensitivity

OVA

ovalbumin