In this study, we demonstrated that concurrent exposure to a dectin-1 agonist could suppress the predominant Th2 response induced by epicutaneous sensitization with protein antigen. We also showed that co-administering a dectin-1 agonist influenced the cytokine expression profiles in draining LNs and decreased the migration of epidermal LCs. To our knowledge, the effects of fungal components on epicutaneous sensitization with protein antigen have not been previously reported.
The effect of dectin-1 signaling on the activation of DCs has been investigated using in vitro culture systems. Carmona et al. and LeibundGut-Landmaun et al. both reported that adding a dectin-1 agonist to DC culture activated the DCs, as reflected by increased surface expressions of CD80, CD86, and CD40 [17, 18]. At first glance, our results showing a comparable expression of co-stimulators on LCs in draining LNs would seem to be in conflict with previous reports. However, dectin-1 was shown to be expressed on langerin- dermal DCs, but not langerin+ epidermal LCs . In addition, although epidermal keratinocytes do not constitutively express dectin-1, it can be induced by β-glucan . Therefore, in our experimental system, epidermal LCs could not directly respond to co-administered curdlan. Instead, they were indirectly influenced by cytokines that were released by surrounding keratinocytes that responded to curdlan via their induced dectin-1. Moreover, IL-18 and IL-1β have been shown to govern epidermal LC migration from the epidermis into draining LNs [22, 23]. Our results showing decreased expressions of IL-18 and IL-1β in the patched skin of mice that were co-administered curdlan with OVA are in agreement with the decreased migration of epidermal LCs into draining LNs.
Among all of the cytokines analyzed, IL-18 is always the most abundantly expressed in skin. IL-18, a member of the IL-1 family, is produced as a biologically inactive precursor and becomes active after its cleavage by caspase-1 . Keratinocytes constitutively express IL-18 and the IL-18 receptor . The expression levels of IL-18 by keratinocytes are up-regulated when exposed to double-stranded RNA, β-defensin, and cathelicidin, whereas its expression is suppressed by vitamin D3 [28–30]. Functionally, IL-18 can induce CXCL10 production by keratinocytes, IL-4 and IL-13 secretion by mast cells and basophils, and the migration of epidermal LCs [24, 31, 32]. Moreover, AD-like inflammatory skin lesions develop spontaneously in transgenic mice that overexpress mature IL-18 in their skin . Single nucleotide polymorphisms of the IL-18 gene have been shown to be associated with AD . Our results that showed decreased IL-18 expressions in patched skin and decreased migration of epidermal LCs into draining LNs when curdlan was co-administered with OVA are in agreement with a lower level of induced Th2 response. Moreover, our results that showed no significant differences of IL-18 expressions in the draining LNs between OVA and OVA plus curdlan groups support the notion that IL-18 does not play a critical role at the priming site during epicutaneous sensitization with protein antigen.
The expression level of IL-1β by epidermal keratinocytes is one of the decisive factors for generating protective Th1 immunity during experimental Leishmaniasis, along with IL-12, IL-4, IL-6, and osteopontin . After OVA patch application, the mRNA expression levels of IL-1α, IL-1β, IL-6, IL-18, and TGF-β of patched skin increased rapidly when compared with naïve BALB/c mice, whereas the expressions of TNF-α and IL-23 were at levels similar to naïve mice (data not shown). We could not detect any significant expressions of IFN-γ, IL-4, IL-10, IL-12 (p35), or IL-12 (p40) in patched skin. The lack of IL-4 and IL-12 expressions in patched skin might contribute to the predominant Th2 responses. The significance of the induced expressions of IL-1α, IL-1β, and IL-6 in OVA-patched skin is obscure at present. However, we suggest that they function as potentiators and enhancers of cutaneous inflammatory responses, similar to IL-18.
The effects of fungal components on adaptive immunity has been investigated using in vivo systems. Dillon et al. reported that zymosan induced regulatory APCs and immunological tolerance in an experimental mouse system using immunization by intravenous injections of zymosan plus OVA . Karumuthil-Melethil et al. reported that treating pre- and early hyperglycemic NOD mice with zymosan resulted in suppressing insulitis, leading to a significant delay in hyperglycemia . Both reports showed that zymosan treatment-induced suppression was associated with enhanced suppressor functions of CD4+CD25+ T regulatory cells, which produced large amounts of IL-10 upon activation. Our results showing suppression of a predominant Th2 response are in agreement with these observations, although we did not detect large amounts of IL-10 in LN cell or splenocyte reactivation cultures (data not shown). In contrast, Yoshitomi et al. reported that a single intraperitoneal injection of zymosan could trigger severe chronic arthritis in SKG mice, which failed to develop the disease in microbially clean conditions . Moreover, Kobayashi et al. reported that an asthma-related environmental fungus, Alternaria, produced potent Th2 adjuvant effects in the airways . Possible reasons for the remarkable discrepancies between this report and our results are the following. First, Th2 skewing is stronger and absolute with epicutaneous sensitization, while it is only preferential with inhalation priming, as evidenced by detectable antigen-specific IgE in only a minority of inhalation-sensitized mice and that some mice also exhibited antigen-specific IgG2a responses . Second, marked differences in the IL-4 and IL-13 dependencies of Th2 responses generated on the skin or in the airways have been emphasized . Third, the study by Kobayashi et al. used fungal extracts that may have induced signaling other than dectin-1. Fourth, during inhalation sensitization, fungal elements can be approached by dectin-1 expressing macrophages and DCs in the alveoli, which is quite a different situation from non-dectin-1-expressing epidermal LCs in the skin.