Objective— Atherosclerosis is an inflammatory disease. Autoimmune responses to low-density lipoproteins (LDL) contribute to its progression, whereas immunization with LDL may induce atheroprotective or proatherogenic responses. The objective of this study was to develop an atheroprotective vaccine by targeting a peptide of the LDL protein constituent apolipoprotein B-100 (apoB-100) to the nasal mucosa to induce a protective mucosal immune response. Methods and Results— A peptide comprising amino acids 3136 to 3155 of apoB-100 (p210) was fused to the B subunit of cholera toxin (CTB), which binds to a ganglioside on mucosal epithelia. The effect of nasal administration of the p210-CTB fusion protein on atherogenesis was compared with that of an ovalbumin peptide fused to CTB and with untreated controls. Immunization with p210-CTB for 12 weeks caused a 35% reduction in aortic lesion size in Apoe ^−/− mice. This effect was accompanied by induction of regulatory T cells that markedly suppressed effector T cells rechallenged with apoB-100 and increased numbers of interleukin (IL)-10 ⁺ CD4 ⁺ T cells. Furthermore, a peptide-specific antibody response was observed. Atheroprotection was also documented in apoe ^−/− mice lacking functional transforming growth factor-β receptors on T cells. Conclusion— Nasal administration of an apoB-100 peptide fused to CTB attenuates atherosclerosis and induces regulatory Tr1 cells that inhibit T effector responses to apoB-100.

Cholera toxin B subunit (CTB) is an efficient mucosal carrier molecule for the generation of mucosal antibody responses and/or induction of systemic T-cell tolerance to linked antigens. CTB binds with high affinity to GM1 ganglioside cell surface receptors. In this study, we evaluated how conjugation of a peptide or protein antigen to CTB by chemical coupling or genetic fusion influences the T-cell-activating capacity of different antigen-presenting cell (APC) subsets. Using an in vitro system in which antigen-pulsed APCs were incubated with antigen-specific, T-cell receptor-transgenic T cells, we found that the dose of antigen required for T-cell activation could be decreased >10,000-fold using CTB-conjugated compared to free antigen. In contrast, no beneficial effects were observed when CTB was simply admixed with antigen. CTB conjugation enhanced the antigen-presenting capacity not only of dendritic cells and B cells but also of macrophages, which expressed low levels of cell surface major histocompatibility complex (MHC) class II and were normally poor activators of naive T cells. Enhanced antigen-presenting activity by CTB-linked antigen resulted in both increased T-cell proliferation and increased interleukin-12 and gamma interferon secretion and was associated with up-regulation of CD40 and CD86 on the APC surface. These results imply that conjugation to CTB dramatically lowers the threshold concentration of antigen required for immune cell activation and also permits low-MHC II-expressing APCs to prime for a specific immune response.

Although dendritic cells (DCs) regulate immune responses, they exhibit functional heterogeneity depending on their anatomical location. We examined the functional properties of intestinal DCs after oral administration of cholera toxin (CT), the most potent mucosal adjuvant. Two CD11c⁺ DC subsets were identified both in Peyer’s patches and mesenteric lymph nodes (MLN) based on the expression of CD8α (CD8⁺ and CD8⁻ DCs, respectively). A third subset of CD11c⁺CD8^int was found exclusively in MLN. Feeding mice with CT induced a rapid and transient mobilization of a new CD11c⁺CD8⁻ DC subset near the intestinal epithelium. This recruitment was associated with an increased production of the chemokine CCL20 in the small intestine and was followed by a massive accumulation of CD8^int DCs in MLN. MLN DCs from CT-treated mice were more potent activators of naive T cells than DCs from control mice and induced a Th2 response. This increase in immunostimulating properties was accounted for by CD8^int and CD8⁻ DCs, whereas CD8⁺ DCs remained insensitive to CT treatment. Consistently, the CD8^int and CD8⁻ subsets expressed higher levels of costimulatory molecules than CD8⁺ and corresponding control DCs. Adoptive transfer experiments showed that these two DC subsets, unlike CD8⁺ DCs, were able to present Ags orally coadministered with CT in an immunostimulating manner. The ability of CT to mobilize immature DCs in the intestinal epithelium and to promote their emigration and differentiation in draining lymph nodes may explain the exceptional adjuvant properties of this toxin on mucosal immune responses.

We have examined the molecular pathways involved in the adjuvant action of cholera toxin (CT) and two novel nontoxic molecules, multiple-mutated CT (mmCT) and double-mutant heat-labile toxin (dmLT) on human T cell responses. Human PBMCs or isolated monocytes were stimulated in vitro with CT, mmCT, or dmLT plus a polyclonal stimulus (staphylococcal enterotoxin B) or specific bacterial Ags, and effects on expression of cytokines and signaling molecules were determined. CT, mmCT, and dmLT strongly enhanced IL-17A and to a lesser extent IL-13 responses, but had little effect on IFN-γ production or cell proliferation. Intracellular cytokine staining revealed that the enhanced IL-17A production was largely confined to CD4+ T cells and coculture experiments showed that the IL-17A promotion was effectively induced by adjuvant-treated monocytes. Relative to CT, mmCT and dmLT induced at least 100-fold lower levels of cAMP, yet this cAMP was enough and essential for the promotion of Th17 responses. Thus, inhibition of cAMP-dependent protein kinase A was abolished, and stimulation with a cAMP analog mimicked the adjuvant effect. Furthermore, CT, mmCT, and dmLT induced IL-1β production and caspase-1 activation in monocytes, which was associated with increased expression of key proinflammatory and inflammasome-related genes, including NLRP1, NLRP3, and NLRC4. Inflammasome inhibition with a specific caspase-1 inhibitor, or blocking of IL-1 signaling by IL-1 receptor antagonist, abrogated the Th17-promoting effect. We conclude that CT, mmCT, and dmLT promote human Th17 responses via cAMP-dependent protein kinase A and caspase-1/inflammasome–dependent IL-1 signaling.