T. Kakinuma, Department of Dermatology, Faculty of Medicine, University of Tokyo, 7?3-1 Hongo, Bunkyo-ku, Tokyo, 113?8655, Japan. E-mail: KAKINUMAT-DER@h.u-tokyo.ac.jp
Atopic dermatitis (AD) is a chronic and relapsing inflammatory skin disease characterized by the predominant infiltration of T cells, eosinophils and macrophages in lesional skin. Recently, macrophage-derived chemokine (MDC)/CCL22, a CC chemokine, was identified as a selective chemoattractant for CC chemokine receptor 4 (CCR4)-expressing cells, in addition to thymus and activation-regulated chemokine (TARC). We have previously reported that serum TARC levels correlate with the severity of AD. In this report, we investigated the participation of MDC in AD. First, we measured serum MDC levels in 45 patients with AD, 25 patients with psoriasis vulgaris and 25 healthy controls. Serum MDC levels in AD patients were significantly higher than those in healthy controls and psoriasis patients. Furthermore, the increases in serum MDC levels in AD patients were greater in the severely affected group than in the moderate or mild groups. We compared serum MDC levels in 11 AD patients, before and after treatment, and observed a significant decrease after treatment. Moreover, the serum MDC levels significantly correlated with the Scoring AD (SCORAD) index, serum soluble (s) E-selectin levels, serum soluble interleukin-2 receptor (sIL-2R) levels, serum TARC levels and eosinophil numbers in peripheral blood. Our study strongly suggests that serum MDC levels have a notable correlation with disease activity and that MDC, as well as the CC chemokine TARC, may be involved in the pathogenesis of AD.
Atopic dermatitis (AD) is a chronic or chronically-relapsing skin disorder characterized by the infiltration of T cells, eosinophils, mast cells and macrophages in lesional skin [1,2]. Enhanced serum IgE levels, specific IgE environmental allergens such as house dust mites, and blood eosinophilia are also present in the majority of AD patients. It has been proposed that Th2-type cells play a key role in the pathogenesis of AD because of the increased expression of Th2-related cytokines, such as IL-4 and IL-5, in lesional skin  and the high responsiveness of peripheral blood mononuclear cells to IL-4, but not IL-2 . Previously, it was shown that serum soluble (s) E-selectin and serum sIL-2 receptor (R) significantly correlate with the disease activity of AD [5–7].
Macrophage-derived chemokine (MDC), newly termed CCL22 , is a CC chemokine that potently serves as a chemoattractant for monocytes, monocyte-derived dendritic cells (DCs) and natural killer (NK) cells . MDC is a ligand for CC chemokine receptor 4 (CCR4) , and is chemotactic for a fraction of CD4 + CD45RO + T cells polarized to produce Th2-type cytokines . We have previously shown that in NC/Nga mice, a mouse model for human AD, dermal DCs are immunoreactive for MDC, and that the immunoreactivity of dermal DCs for MDC was abolished by topical corticosteroid treatment . This indicates that the DC is a main source of MDC in lesional skin of AD. Very recently, we reported that levels of thymus and activation-regulated chemokine (TARC), another ligand for CCR4, in AD sera significantly correlate with disease activity . Moreover, it is reported that serum MDC levels in AD patients are higher than those in healthy controls , although the precise involvement of the high levels of MDC in AD has not yet been fully identified.
We measured serum MDC levels in a large number of patients with AD, and compared them with levels in psoriasis vulgaris patients and healthy controls. We also examined the correlation between serum MDC levels, disease severity and the change in serum MDC levels in AD patients, before and after treatment. In addition, we compared serum MDC levels with laboratory data for AD disease markers such as serum soluble (s) E-selectin, serum soluble interleukin-2 receptor (sIL-2R), serum TARC levels and eosinophil numbers in peripheral blood.
MATERIALS AND METHODS
Samples and reagents
Sera of 45 AD patients clinically diagnosed according to the criteria by Hanifin and Rajka , 25 psoriasis vulgaris patients and 25 healthy controls were examined. The 45 AD patients were treated with topical corticosteroid therapy in combination with oral antihistamine drugs. Twenty of the 45 patients had a personal history of respiratory allergy, whereas the patients with psoriasis vulgaris and the healthy controls had no allergies and their total serum IgE levels were within the normal range. After venous blood was drawn, serum was separated and stored at – 20°C until use.
The disease severity of AD was determined by the modified SCORAD (SCORing Atopic Dermatitis) index system [15,16]. In this study, we also divided the AD patients into three groups, mild, moderate and severe, according to the proposal for severity grading of AD using only objective criteria .
The serum levels of MDC, sIL-2R, sE-selectin and TARC were measured using an ELISA system. A 96-well microplate coated with either murine MoAb against human MDC (R & D Systems Inc., Mineapolis, MN, USA), sIL-2R (R & D Systems Inc.), sE-selectin (MedSystems, Vienna, Austria) or TARC (Techne Corp., Mineapolis, MN, USA) was used. Each serum was added to the microplate and incubated. After the removal of unbound material by aspiration and washing, a conjugate of MDC, sIL-2R, sE-selectin or TARC was detected by its reaction with a substrate solution, which yielded a coloured product proportional to the amount of conjugate. The optical density of each well was measured at 450 nm.
The data were analysed using the Mann–Whitney U-test. A P value < 0·05 was considered to be statistically significant.
Correlation coefficients were determined using the Spearman rank correlation test. All comparisons were made after logarithmic transformation. Statistical analysis was also calculated and a P value < 0·05 was considered to be statistically significant.
Serum MDC levels in AD patients were significantly higher than those in psoriasis vulgaris patients or healthy controls
The average level of serum MDC in the AD patients was 3037·9 ± 300·4 pg/ml. This elevation was statistically significant compared with that in psoriasis vulgaris patients (888·0 ± 63·3 pg/ml, P < 0·001) or healthy controls (686·7 ± 42·8 pg/ml, P < 0·001) (Fig. 1a). Thus, the value of serum MDC in AD patients was remarkably higher than that in psoriasis vulgaris patients and healthy controls.
Next, we divided the 45 AD patients into three groups, mild (n = 10), moderate (n = 22) and severe (n = 13), and compared the serum MDC levels among these different groups. The serum MDC levels of the severe AD group (5375·8 ± 548·8 pg/ml) were significantly higher than those of the mild AD group (1181·3 ± 82·0 pg/ml, P < 0·001) or the moderate AD group (2500·3 ± 203·3 pg/ml, P < 0·001) (Fig. 1b).
Serum MDC levels decreased after treatment with topical corticosteroid treatment and antihistamine drugs
We measured serum MDC levels in 11 patients with AD before and after treatment. The serum MDC levels of the AD patients were high before treatment (4149·1 ± 716·0 pg/ml, SCORAD 60·2 ± 15·3) and significantly decreased after treatment in accordance with the improvement of the eruption (1397·5 ± 196·8 pg/ml, SCORAD 29·6 ± 10·0, P < 0·001) (Fig. 1c). Serum sE-selectin, sIL-2R levels and eosinophil numbers in peripheral blood significantly decreased in parallel with the changes in serum MDC levels during treatment (data not shown).
Serum MDC levels correlated with disease activity in AD patients
The serum MDC levels of AD patients were compared with serum sE-selectin, sIL-2R, IgE levels, eosinophil numbers in peripheral blood and SCORAD. The serum sE-selectin and sIL-2R levels of AD patients were significantly higher than those in the healthy controls (data not shown), which is consistent with previous reports [5–7]. The serum MDC levels significantly correlated with SCORAD (r = 0·68) (Fig. 2a), serum sE-selectin levels (r = 0·66), serum sIL-2R levels (r = 0·65), serum TARC levels (r = 0·72) (Fig. 2b) and eosinophil numbers in peripheral blood (r = 0·74).
In this study, we examined the relevance of MDC in AD and obtained results as follows: (i) serum MDC levels in AD patients were significantly higher than those in psoriasis vulgaris patients or healthy controls; (ii) serum MDC levels in the severe group of AD patients were significantly higher than those in the mild or moderate groups; (iii) serum MDC levels in AD patients decreased after treatment in accordance with the improvement of the eruption; and (iv) serum MDC levels in AD patients significantly correlated with SCORAD, serum sE-selectin levels, serum sIL-2R levels, serum TARC levels and eosinophil numbers in peripheral blood.
MDC (CCL22) is a member of the CC chemokine family and one of the ligands for CCR4 . It is constitutively expressed in DCs , macrophages  and thymic epithelial cells , and acts as a chemoattractant for CCR4-expressing cells such as memory T cells. It is also a chemoattractant for NK cells and eosinophils, even though they have little or no CCR4 expression [9,18].
Previous in vitro studies revealed that MDC production from monocytes is stimulated by IL-4 and IL-13, whereas it is inhibited by IFN-γ[19,20]. Moreover, MDC production from activated T cells is found to be preferentially associated with a Th2-type cytokine profile and inversely related to IFN-γ production in vitro[14,21]. From these data, it is suggested that MDC expression is preferentially involved in Th2-type reactions.
Previously, it was reported that MDC production is up- regulated at the mRNA and protein level during an allergic reaction in mouse airway hyperreactivity and lung inflammation . In addition, we have clarified that in NC/Nga mice, regarded as a mouse model for human AD, MDC is observed in lesional and, to a lesser extent, non-lesional skin . Thus, it is suggested that MDC may be involved in allergic diseases. Indeed, a recent study revealed increased serum MDC levels in patients with AD and immunoreactive MDC in the lesional skin . These results strongly suggest the participation of MDC in diseases involving infiltration of predominantly Th2-type cells, although little has been elucidated about the precise function of MDC in AD. In this report, we confirmed that the serum MDC levels of AD patients were significantly higher than those of psoriasis vulgaris patients or healthy controls; we demonstrated that the serum MDC levels in AD were higher in the severe group than in the moderate or mild groups, and that these high levels decreased after treatment in accordance with the improvement of the eruption in the AD patients. In addition, we demonstrated the correlation of serum MDC levels with other laboratory data such as serum sIL-2R, sE-selectin and TARC levels, and eosinophil numbers in peripheral blood, which are reported to be disease markers for AD [5–7,13]. The serum MDC levels significantly correlated with SCORAD, a measure of clinical severity of AD. Thus, these data clearly show that serum MDC levels may be an important marker for the disease activity of AD.
It was recently reported that immunoreactive MDC is observed in CD3+ T lymphocytes and CD1a+ DCs in the lesional skin of AD patients , whereas we have reported that TARC, another ligand for CCR4, is highly expressed in the lesional keratinocytes of AD patients . Thus, TARC and MDC are differentially expressed in the lesional skin of AD, although both chemokines attract CCR4-expressing cells.
In conclusion, we have confirmed that serum MDC levels clearly reflect the disease activity of AD. From these data, we strongly suggest that MDC, as well as TARC, may be involved in the pathogenesis of AD.
This work was supported in part by Health Science Research Grants from the Ministry of Health and Welfare and by grants from the Ministry of Education, Science and Culture, Japan.