**A**

**Figure 4.** Effects of peridinin or fucoxanthin on the number of eosinophils in peripheral blood. Mice were sensitized with PCl, as described in the Experimental Section. Two weeks after the sensitization, 10 ΐg of peridinin, fucoxanthin or hydrocortisone was painted onto the ears and 50 ΐg of peridinin or fucoxanthin was administered i.p. to these mice 3 h before the antigen challenge. At 48 h after the antigen challenge, mice were bled from the retro-orbital plexus. The percentage of eosinophils among white blood cells was determined by making a smear on a slide glass for each mouse. Slide glasses were stained with Giemsa solution, and the percentage of eosinophils among white blood cells was estimated by counting at least 200 white blood cells from each sample. There was no significant difference in the total numbers of leukocytes among groups. Each value is expressed as the mean ± SD ( *<sup>n</sup>* = 10). Pos cont, positive control; Hydroc, hydrocortisone; Peri, peridinin; Fuco, fucoxanthin; Peri i.p., peridinin intraperitoneally; Fuco i.p., fucoxanthin intraperitoneally; Neg cont, negative control. \* An asterisk indicates a significant difference between the treatment group and the positive control ( *p* < 0.01, Tukey– Kramer's *post hoc* test). 

*2.4. Effects of Peridinin on Eotaxin Production* 

Figure 5 shows that eotaxin production was stimulated by the challenge with PCl, and this was inhibited by peridinin following painting onto the ears (30.6% inhibition), but not by i.p administration. This inhibitory effect was comparable to that of hydrocortisone (42.1% inhibition). On the other hand, fucoxanthin weakly inhibited the production of eotaxin in mice administered i.p (19.4% inhibition), whereas it had no effect in the case of painting. These results suggest that metabolites of peridinin and fucoxanthin administered i.p. may have different effects on eotaxinproducing cells, such as endothelial cells, epithelial cells and macrophages. It is also possible that there is a difference between these two compounds in terms of penetration through the skin. These differences between peridinin and fucoxanthin remain to be clarified. 

**Figure 5.** Peridinin inhibits eotaxin production in challenged skin. Mice were sensitized as described in the Experimental Section. At 48 h after the antigen challenge, ear lobes were removed, and the concentration of eotaxin in each ear lobe was measured as described in the Experimental Section. Each value is expressed as the mean ± SD ( *<sup>n</sup>* = 3). Pos cont, positive control; Hydroc, hydrocortisone; Peri, peridinin; Fuco, fucoxanthin; Peri i.p., peridinin intraperitoneally; Fuco i.p., fucoxanthin intraperitoneally; Neg cont, negative control. \* An asterisk indicates a significant difference between the treatment group and the positive control ( *p* < 0.01, Tukey–Kramer's *post hoc* test). \*\* There is a significant difference between Fuco i.p. and the positive control ( *p* < 0.05, Tukey– Kramer's *post hoc* test). Asterisks are shown only for treatment groups showing suppression. 

*2.5. Inhibitory Effects of Peridinin and Fucoxanthin on the Chemotaxis of Eosinophils toward Eotaxin* 

Next, we asked whether peridinin inhibits the migration of eosinophils using murine eosinophils isolated from IL-5 transgenic mice. Peridinin and fucoxanthin were tested in an eosinophil chemotaxis assay *in vitro*. Chemotaxis of eosinophils toward eotaxin (20 ng/mL) was suppressed by one and 

3 ΐg/mL of peridinin (57.4% and 72.8% suppression, respectively; Figure 6) and by one and 3 ΐg/mL of fucoxanthin (24.2% and 61.7% suppression, respectively; Figure 6). Peridinin showed higher suppression of the migration of eosinophils toward eotaxin than did fucoxanthin. More than 95% of eosinophils were viable after one hour of incubation for the chemotaxis toward eotaxin. The viability of eosinophils was checked by trypan blue exclusion. 

Although the painting of peridinin on earlobes decreased the level of eotaxin in the ears, the i.p. administration of peridinin did not have any effect on the production of eotaxin in the ears. In the peripheral blood, however, the number of eosinophils was suppressed significantly in either route of administration, suggesting that peridinin suppressed the number of eosinophils in the ears by reducing the number of eosinophils in the peripheral blood. Therefore, the level of eotaxin in the ears does not have a primary role in case of the i.p. administration of peridinin. Since peridinin did not change the serum level of IL-5 in either route of administration as described above, the decreased number of eosinophils in peripheral blood may be regulated by other factors. 

In contrast, the number of eosinophils in peripheral blood was not influenced by fucoxanthin in either route of administration. Although fucoxanthin suppressed the migration of eosinophils *in vitro*, this effect seems to be marginal *in vivo*, because there is an increased level of eosinophils in the peripheral blood and a higher level of eotaxin is produced in the ears. 

**Figure 6.** The effect of peridinin or fucoxanthin on the migration of eosinophils toward eotaxin in chemotaxis assays. Eosinophils prepared as described in the Experimental Section were suspended in Roswell Park Memorial Institute (RPMI) 1640 medium containing 0.1% bovine serum albumin (BSA) and placed in the Transwells of a 24-well chemotaxis chamber. The lower chamber contained RPMI 1640 medium containing 0.1% BSA and mouse eotaxin (final concentration: 20 ng/mL). Peridinin or fucoxanthin at a final concentration of 1 or 3 ΐg was added to each Transwell. Assay plates were incubated for 1 h at 37 °C in 5% CO2. After 1 h incubation, the cells that migrated across the filter to the lower chamber were counted. Each value is expressed as the mean ± SD ( *n* = 18). Pos cont, positive control; Peri 1, 1 ΐg of peridinin; Peri 3, 3 ΐg of peridinin; Fuco 1, 1 ΐg of fucoxanthin; Fuco 3, 3 ΐg of fucoxanthin; Neg cont, negative control. \* An asterisk indicates a significant difference between the treatment group and the positive control ( *p* < 0.01, Tukey– Kramer's *post hoc* test). 

## **3. Experimental Section**
