**6. Prostanoids as regulatory factors of Th differentiation**

In older *in vitro* studies, PGE2 has been regarded as a suppressant of Th1 cells, because of its suppressive effect on cell proliferation, differentiation and cytokine production from Th1 cells (Betz and Fox, 1991; Goodwin and Ceuppens, 1983). However, recent reports indicate that several PGE2 receptors are involved in the regulation of Th differentiation in skin immunity via multiple pathways and different directions (Figure 2).

Keratinocytes m m, h m m, h m, h m, h m, h m,h

Dendritic cells h m, h m m, h m, h m, h m

B cells m, h m, h m, h m, h m Macrophages h m, h m m, h m m m m m, h

Blood vessels h h h m, h m, h m, h m, h m,

**5. Involvement of prostanoids in the pathogenesis of AD: Analysis based on** 

As mentioned above, there are multiple pathogenetic factors of AD, but the skin barrier dysfunction and Th2 mediated immune response are its general characteristic features. Mouse ovalbumin (OVA)-induced dermatitis is one of the most frequently used AD models (He et al.; Spergel et al., 1998). In a typical mouse OVA-induced AD model, mice are first sensitized with an OVA patch using a transparent bio-occlusive dressing on a shaved and tape-stripped area of skin for one week. This sensitization to OVA is repeated in two-week intervals. After three to four sensitization cycles, the mice show elevated serum IgE and significant eosinophil and Th2-deviated lymphocyte infiltration in the skin, which is similar to the pathology of AD. In this model, COX-2 deficient mice and the administration of COX-2 inhibitor both showed enhanced eosinophil infiltration and elevated IL-4 expression in the skin lesion with elevated serum IgE and IgG1 (Laouini et al., 2005). These results suggest that COX-2-derived prostanoids play regulatory roles in the development of AD, such as Th differentiation and inflammatory cell infiltration in the skin. In the following sections, we will discuss how prostanoids are involved in Th differentiation, DC function, and

In older *in vitro* studies, PGE2 has been regarded as a suppressant of Th1 cells, because of its suppressive effect on cell proliferation, differentiation and cytokine production from Th1 cells (Betz and Fox, 1991; Goodwin and Ceuppens, 1983). However, recent reports indicate that several PGE2 receptors are involved in the regulation of Th differentiation in skin

Eosinophils h h h h m m,h m, h h h

cells h m, h m m, h m m m m

basophils m, h h m, h m, h m m m m Neutrophils h h h m m, h m

Table 1. Expression of prostanoid synthases and prostanoid receptors in the skin

T cells m,

PG; prostaglandin, S; synthase, m; mouse, h; human

inflammatory cell infiltration in skin immunity.

**6. Prostanoids as regulatory factors of Th differentiation** 

immunity via multiple pathways and different directions (Figure 2).

Modified from the reference by Tilley et al.

**animal models of AD** 

hLangerhans

Mast cells,

PGDS PGES PGFS PGIS TXS DP CRTH2 EP1 EP2 EP3 EP4 FP IP TP

h(Th2) m m m m, h m m

<sup>h</sup>m, h

For example, PGE2-EP1 signaling has been reported to facilitate Th1 differentiation in the sensitization process through the skin (Nagamachi et al., 2007). PGE2 produced by DCs in draining lymph nodes (dLNs) stimulates EP1 receptors on naïve CD4+ and CD8+ T cells and promotes Th1 and Tc1 differentiation (Nagamachi et al., 2007). PGI2-IP signaling promotes Th1 and Tc1 differentiation through a cAMP dependent mechanism (Nakajima et al., 2010). Intriguingly, IP deficient mice showed enhanced Th2 response such as elevated IgE concentration in serum in the mouse OVA-induced asthma model (Nagao et al., 2003), suggesting that lack of PGI2-IP signaling might result in a Th2 biased immune response through the inhibition of Th1 differentiation.

The regulatory mechanism of prostanoid signaling on Th differentiation is complex, because it depends on the context of immune system. For example, PGE2-EP2/EP4 signaling regulates Th1 and Th17 differentiation (Yao et al., 2009). In a weaker co-stimulation signaling through CD28, PGE2 suppresses the Th1 differentiation via EP2 and EP4 receptors. In the case of strong co-stimulation signaling, however, stimulation of EP2 and EP4 signaling conversely facilitates the Th1 differentiation through a PI3-kinase-dependent mechanism (Yao et al., 2009). These results suggest that the action of prostanoid receptor signaling can be changed in a context-dependent manner. EP2 and EP4 signaling also regulates the Th17 differentiation. Th17 is a recently identified Th subset, and can be detected in a number of diseases, including AD (Guttman-Yassky et al., 2011; Koga et al., 2008). *In vitro*, Th17 differentiation is induced from naïve T cells in the presence of IL-6 and TGF-b. In this condition, PGE2 acts on naïve T cells through EP2/EP4 signaling and suppress the Th17 differentiation in a cAMP-dependent manner. However, PGE2-EP2/EP4 signaling also acts on DCs and increases the IL-23 production from the DCs. Thus, PGE2 facilitates the expansion of Th17 (Yao et al., 2009). The blockade of EP4 signaling consistently ameliorated the disease progression in a CHS model and an EAE model, which are mediated by Th1 and Th17 cells, respectively (Yao et al., 2009). These results clearly indicate the importance of prostanoid signaling in Th differentiation *in vivo.* The facilitation effect of PGE2 on Th17 is also reported in human T cells (Boniface et al., 2009).

### **7. Prostanoids as regulatory factors of DC function**

In the initial step of sensitization in AD, allergens which enter the skin are captured by skin DCs and presented to the naïve T cells in the dLNs. Prostanoids can regulate this step by affecting the migration ability or antigen presentation ability of the skin DCs (Figure 2). PGE2, which is produced by KCs, acts on EP4 on LCs, and stimulates the migration of LCs (Kabashima et al., 2003b). Conversely, stimulation of DP on DCs inhibits the migration of skin DCs. Topical administration of DP inhibits the migration of DCs to dLNs and significantly suppresses the development of the mouse AD model (Angeli et al., 2001; Angeli et al., 2004). Prostanoids also regulate DC-T cell interaction in the priming of naïve T cells (Kabashima et al., 2003a). Cutaneous DCs produce abundant TXA2, which acts on naïve T cells and increases the motility of T cells, which impairs the stable DC-T cell interaction (Kabashima et al., 2003a). TP-deficient mice or wild-type mice treated with a TP antagonist during the sensitization period show enhanced CHS responses, indicating that TP signaling negatively regulates the priming of T cells *in vivo*.

Although the role of IgE in AD is still controversial (Guttman-Yassky et al., 2011), high serum IgE is one of the hallmarks of AD. Compared to the analysis of T cells and DCs, the reports about the role of prostanoids on B cells are relatively scarce. From the *in vivo* data using COX-2 deficient mice or IP deficient mice, which show increased IgE production in OVA sensitization

The Role of Prostanoids in Atopic Dermatitis 71

by activated mast cells. PGD2 has two types of receptors, DP and CRTH2. CRTH2 induces chemotaxis in Th2 cells, eosinophils and basophils with enhanced degranulation *in vitro* (Hirai et al., 2001; Yoshimura-Uchiyama et al., 2004). CRTH2 amplifies Th2 responses by preventing apoptosis of Th2 cells and enhancing their capacity to secrete cytokines (Nomiya et al., 2008; Xue et al., 2009). CRTH2 also amplifies eosinophil functions by mobilizing them from the bone marrow, preventing their apoptosis, and promoting their chemokinesis and degranulation (Gervais et al., 2001). CRTH2 mRNA expression is high in peripheral blood mononuclear cells of patients with AD (Hijnen et al., 2005), and circulating eosinophils and T cells in patients with AD have an increased surface expression of CRTH2 (Iwasaki et al.,

He et al. have recently reported that lack of CRTH2 signaling ameliorates the inflammation only in newly-challenged skin, while loss of this signaling in chronic challenged areas did not affect the inflammation (He et al., 2011). They used CRTH2 knockout mice in two types of AD models: one that was repeatedly sensitized with OVA for a total of seven weeks, which mimicked the chronic lesions of AD; and one that was challenged with OVA after the repeated sensitization of other skin areas for a total of seven weeks, which was supposed to mimic the acute lesions of AD. In the chronic lesions, the inflammatory cell infiltration and cytokine concentration was similar between wild-type and CRTH2 knockout mice, while in the acute lesions, such factors were significantly decreased in CRTH2 knockout mice compared with wild-type mice (He et al.). Consistently, the concentration of PGD2 increased significantly in the acute lesions, while the concentration of PGD2 in the chronic lesions was similar compared with that of non-affected skin. CRTH2 knockout mice also showed comparable levels of IgE production, indicating that CRTH2 signaling had little effect much on the antibody production process. Boehme et al. have previously reported that administration of a CRTH2 antagonist inhibited the development of chronic AD lesions in the same model (Boehme et al., 2009), but the CRTH2 antagonist-treated group also showed reduced IgE production, suggesting the possibility that administration of the CRTH2 antagonist affected the extent of sensitization non-specifically and thus lead to the reduced inflammatory cell infiltration in the skin. Collectively, blockage of PGD2-CRTH2 signaling might inhibit allergic skin inflammation elicited in patients with AD by re-exposure to

In contrast, stimulation of EP3-signaling in KCs is reported to play an anti-inflammatory role in skin inflammation by inhibiting chemokine production from KCs (Honda et al., 2009). Administration of an EP3 specific agonist suppressed a CHS response, and EP3 knockout mice showed an enhanced CHS response, suggesting that PGE2-EP3 signaling works as a negative regulator of allergic cutaneous inflammation. An anti-inflammatory role of EP3 signaling is also reported in other allergic diseases such as mouse asthma and the

It is reported that some prostanoids can modulate pruritus, a significant hallmark of AD. In human studies, PGE2 is a weak pruritogen and prolongs experimentally-induced itch (Hagermark and Strandberg, 1977; Neisius et al., 2002), although injection of PGE2 alone does not elicit itch-associated response in animal experiments (Andoh and Kuraishi, 1998). TXA2 is also reported as a mediator of itch (Andoh et al., 2007). Injection of a TP agonist alone elicited itch-associated responses. TP was expressed in both KCs and nerve fibers in

2002), suggesting the role of CRTH2 in AD.

antigens to which they have been sensitized (Figure 3).

**10. Prostanoids as itch mediators** 

allergic conjunctivitis model (Kunikata et al., 2005; Ueta et al., 2009).

(Laouini et al., 2005), it might be possible that some prostanoid signaling regulates the antibody production. *In vitro*, PGE2 drives Ig class switching to IgE by acting at EP2 and EP4 on B cells under LPS and IL-4 stimulation *in vitro* (Fedyk and Phipps, 1996). Whether such actions occur *in vivo* remains unknown, and this should be clarified in future studies.
