Cytokine secretion profile of human keratinocytes exposed to Malassezia yeasts

Authors


  • Editor: Artur Ulmer

Correspondence: Yoshio Ishibashi, Department of Immunobiology, Meiji Pharmaceutical University, Noshio, Kiyose, Tokyo 204-8588, Japan. Tel.: +81 424 95 8741; fax: +81 424 95 8612; e-mail: yishibas@my-pharm.ac.jp

Abstract

The lipophilic yeast Malassezia is an exacerbating factor in atopic dermatitis (AD). Among organisms of the Malassezia species, Malassezia globosa and Malassezia restricta are particularly dominant on the skin of AD patients. However, the precise role of Malassezia yeasts in the pathophysiology of AD remains uncertain. Keratinocytes play a critical role in cutaneous inflammatory and immune responses by secreting cytokines. In this study, we attempted to determine the cytokine secretion profiles of human keratinocytes that were exposed to Malassezia yeasts. The human keratinocyte cell line PHK16-0b was cocultivated with M. globosa or M. restricta for 24 h, and the resulting cytokine secretion profile was analysed using a cytokine antibody array. The keratinocytes responded to the two Malassezia species with different Th2-type cytokine profiles, i.e. M. globosa induced IL-5, IL-10 and IL-13 secretion from the keratinocytes, whereas M. restricta induced IL-4 secretion. Similar results were obtained with primary normal human epidermal keratinocytes. cDNA microarray analysis confirmed that IL-5, IL-10, and IL-13 mRNAs were induced only by M. globosa, while IL-4 mRNA expression was induced only by M. restricta. These findings suggest that M. globosa and M. restricta play a synergistic role in triggering or exacerbating AD by stimulating the Th2 immune response.

Introduction

The genus Malassezia constitutes a family of lipophilic yeasts comprising 11 different species including Malassezia globosa and Malassezia restricta (Sugita et al., 2005). These organisms are members of the normal human cutaneous microbial communities, but they are also associated with several skin diseases such as pityriasis versicolor, seborrhoeic dermatitis, Malassezia folliculitis and atopic dermatitis (AD) (Ashbee & Evans, 2002; Faergemann, 2002). Recently, using a nonculture-based method (nested PCR), we demonstrated that M. globosa and M. restricta were detected in c. 90% AD patients, while other Malassezia species were detected in <40% AD patients, suggesting that M. globosa and M. restricta may play a particularly important role in the pathogenesis of AD (Sugita et al., 2001). However, the precise mechanism by which these two species trigger or exacerbate AD remains unclear.

AD is a pruritic inflammatory skin disease that is characterized by elevated plasma levels of IgE against many types of allergens as well as the infiltration of mast cells and eosinophils (Wuthrich, 1978; Leung et al., 2004). The development and pathogenesis of AD are mainly associated with the skewing of the immune responses towards the Th2-type cytokine response, although chronic AD occasionally appears to be associated with the Th1-type cytokine response (Hamid et al., 1994; Leung, 2000). Th2-type cytokines, including interleukin (IL)-4, IL-5, IL-6, IL-10 and IL-13, promote eosinophil growth, migration and activation, mast cell differentiation, and IgE production. On the other hand, Th1-type cytokines, including interferon (IFN)-γ, IL-2, IL-12 and tumour necrosis factor (TNF)-α, promote cell-mediated immunity (Chen et al., 2004).

Keratinocytes play a critical role in cutaneous inflammatory and immune responses by secreting a variety of cytokines (Esche et al., 2004; Albanesi et al., 2005). These cells secrete several interleukins (IL-1, IL-3, IL-6, IL-7, and IL-8); colony-stimulating factors [granulocyte-macrophage colony-stimulating factor (GM-CSF), granulocyte colony-stimulating factor (G-CSF) and macrophage colony-stimulating factor (M-CSF)], and growth factors [TNF-α, transforming growth factor; TGF-α, TGF-β and platelet-derived growth factor (PDGF)] (Grone, 2002). Most of these mediators are not produced constitutively; instead, their gene expression is upregulated in response to several stimuli. Previous studies have shown that Malassezia yeasts can induce IL-1α, IL-6, IL-8 and TNF-α secretion from human keratinocytes (Watanabe et al., 2001).

In this study, we investigated the cytokine secretion profiles of human keratinocytes exposed to M. globosa and M. restricta.

Materials and methods

Yeast strains and culture conditions

Malassezia globosa CBS7966, M. restricta CBS7877, M. furfur CBS1878, and M. sympodialis CBS7222 were used in this study. These strains were cultured at 32°C on modified Leeming and Notman agar (LNA; 20 g glucose, 50 g malt extract, 1 g polypeptone and 20 g bile salts; Oxoid, Hampshire, UK) containing 1% Tween 40, 0.2% glycerol and 50 μg mL−1 chloramphenicol (Sankyo, Tokyo, Japan). Prior to use, the yeasts were harvested and suspended in Dulbecco's phosphate-buffered saline (PBS; pH 7.4) at a concentration of 1 × 108 yeast cells mL−1.

Human keratinocytes and their cultivation

The human keratinocyte cell line PHK16-0b (JCRB0141) was obtained from the Japanese Collection of Research Bioresources (JCRB) cell bank and the primary normal human epidermal keratinocyte (NHEK) from PromoCell (Germany). The cells were cultured at 37°C in 5% CO2 in KBM-2 medium (Clonetics, Inc.) supplemented with 0.1 ng mL−1 EGF, 5 μg mL−1 insulin, 30 μg mL−1 bovine pituitary extract, 0.5 μg mL−1 hydrocortisone, 50 μg mL−1 gentamicin and 50 μg mL−1 amphotericin.

Exposure of keratinocytes to Malassezia yeasts

At 18 h prior to infection, PHK16-0b or NHEK cells were seeded into 35-mm tissue culture dishes at a density of 1 × 106 cells dish−1. Prior to yeast challenge, the keratinocyte monolayers that had grown in the tissue culture dishes were washed three times and incubated with antibiotic-free medium for 2 h. The keratinocyte monolayers were cocultivated with 2 × 107 yeast cells dish−1 [multiplicity of infection (MOI) of 20], which was shown by a preliminary experiment to be the optimal MOI for inducing cytokine secretion. Cell culture supernatants were collected after exposure for 24 h at 37°C in 5% CO2. In the trypan blue exclusion assay, we observed that the viabilities of the Malassezia-exposed keratinocytes did not differ from those of the control cells. More than 95% of the cells in all the cultures were viable (data not shown). In a separate experiment, using the SV total RNA isolation system (Promega, Madison, WI) according to the manufacturer's instructions, the total RNA was isolated from the cells that were exposed to Malassezia by cocultivation for 2 and 4 h, and was quantified spectrophotometrically.

Measurement of cytokine release using antibody arrays

Cytokine release from the keratinocytes was analysed using Human Cytokine Array VI and VII (RayBiotech Inc., Norcross, GA) according to the manufacturer's instructions. Briefly, the cytokine array membranes were blocked with 1 × blocking buffer for 30 min and then incubated at 4°C overnight with 1 mL of the samples. After incubation, the membranes were washed three times with 2 mL of 1 × wash buffer I, followed by two washes with 2 mL of 1 × wash buffer II at room temperature with shaking. The membranes were then incubated with 2 mL of 1 : 500 diluted biotin-conjugated antibodies at room temperature for 2 h and washed as described above before incubation with 1 mL of 1 : 40 000 diluted streptavidin-conjugated peroxidase at room temperature for 1 h. After a thorough wash, the membranes were exposed to a peroxidase substrate (detection buffers C and D) for 5 min in the dark before imaging. The membranes were exposed to an X-ray film within 30 min of exposure to the substrate. Signal intensities were quantified using the scanalyze software (Michael Eisen, Lawrence Berkeley National Laboratory, http://www.microarrays.org/software.html). Horseradish peroxidase (HRP)-conjugated antibody that was placed at six spots served as the positive substrate control, and it was also used to identify membrane orientation. For each spot, the net signal intensity was determined by subtracting the background level from the total raw signal intensity.

Enzyme-linked immunosorbant assay (ELISA)

The keratinocyte supernatants were assayed for IL-4 and IL-5 using the Human IL-4 and IL-5 ELISA kits (RayBiotech, Inc.) according to the manufacturer's instructions. The results are given in terms of pg (106 cells mL−1).

cDNA microarray

The relative expression of cytokine mRNA was analysed using GEArray Q Series Human Inflammatory Cytokines & Receptors Gene Array (SuperArray Inc., Bethesda, MD) according to the manufacturer's instructions. In brief, 2 μg total RNA was reverse transcribed into cDNA in the presence of biotin-16-dUTP (Roche, Mannheim, Germany) using the AmpoLabeling-LPR kit (SuperArray Inc.). The obtained biotin-labelled cDNA samples were then hybridized overnight to cytokine and cytokine receptor gene-specific probes that were spotted on the GEArray membranes. After incubation with streptavidin-alkaline phosphatase (AP) conjugate (1 : 12 500), the array image was developed using the CDP-Star chemiluminescent substrate and recorded on an X-ray film. The image was scanned using a scanner, and the obtained raw data was analysed using the scanalyze software and the gearray Analyzer software (http://www.superarray.com). The signal that was obtained on the array from the expression of each gene was normalized to the signal that was derived from an internal cyclophilin A standard present on the same membrane.

Results

Cytokine profile induced in human keratinocytes by M. globosa or M. restricta

We examined the secretion of cytokines from human keratinocytes in response to M. globosa or M. restricta. PHK16-0b cells were exposed to M. globosa or M. restricta at an MOI of 20. After 24 h, the cytokine levels in the culture supernatants were detected and semiquantified on membrane arrays containing 120 different cytokine antibodies (Fig. 1a–c). Untreated human keratinocytes (control) were observed to secrete only trace or undetectable amounts of the cytokines. In this study, the differential cytokine secretion was arbitrarily defined by an intensity change of twofold or greater in secretion. Stimulation with M. globosa induced a pattern of cytokine secretion that was different from that induced by M. restricta, i.e. the exposure of PHK16-0b cells to M. globosa resulted in a marked secretion of IL-3, IL-5, IL-6, IL-7, IL-10 and IL-13, while exposure to M. restricta did not induce the secretion of these cytokines. M. globosa also induced the secretion of remarkable amounts of GM-CSF, IL-8, tissue inhibitor of metalloproteinases (TIMP)-1 and TIMP-2, while M. restricta induced the secretion of these cytokines in much smaller amounts. Similarly, while M. restricta stimulated IL-4, monocyte inhibitory protein (MIP)-3α, leptin, cutaneous T cell attracting chemokine (CTACK) and placental growth factor (PlGF) secretion from PHK16-0b cells, M. globosa did not. G-CSF was the most predominant cytokine that was detected in response to both Malassezia species. Consistent with the antibody array data, ELISA revealed that after exposure to M. globosa, keratinocytes released a significant amount of IL-5 but not IL-4 (Fig. 2a). On the other hand, a significant increase was observed in the IL-4 level but not the IL-5 level in the culture supernatant of M. restricta-exposed keratinocytes (Fig. 2b).

Figure 1.

 Cytokine profile of human keratinocytes exposed to Malassezia globosa or Malassezia restricta. PHK16-0b cells (1 × 106 cells dish−1) were exposed to Malassezia globosa or Malassezia restricta at a multiplicity of infection of 20 for 24 h. Untreated cell cultures were used as the control. Cell culture supernatants were subject to a cytokine antibody array. (a) Each cytokine is represented by duplicate spots in the locations shown. The cytokines exhibited a twofold or greater increase in response to M. globosa and M. restricta are represented in black and grey, respectively. (b) The cytokine array image represents the results of one of two independent experiments in which similar patterns of expression were observed. (c) The average net signal intensity for each pair of cytokine spots is shown. BMP, bone morphogenetic protein; FGF, fibroblast growth factor; GM-CSF, granulocyte-macrophage colony-stimulating factor; IL, interleukin; MIP-3α, macrophage inflammatory protein-3α; CTACK, cutaneous T cell attracting chemokine; G-CSF, granulocyte colony-stimulating factor; PlGF, placental growth factor; TIMP, tissue inhibitor of metalloproteinases.

Figure 2.

 IL-4 (a) and IL-5 (b) release from keratinocytes after exposure to Malassezia globosa and Malassezia restricta. Keratinocyte cultures were exposed to M. globosa or M. restricta at a multiplicity of infection of 20 for 24 h. ELISAs were performed for IL-4 and IL-5 using the culture supernatants. The values represent the means ±SE of four experiments. *, P<0.05 versus control.

The human primary keratinocytes were then used as more relevant cells for determining the cytokine secretion in response to M. globosa or M. restricta. The exposure of NHEK cells to M. globosa induced IL-3, IL-5, IL-6, IL-10 and IL-13 secretion; M. restricta did not have any remarkable effect on the secretion of these cytokines. Malassezia restricta induced high level of IL-4 secretion, whereas exposure to M. globosa did not (Table 1). These results were essentially identical to those obtained with PHK16-0b cells, although NHEK cells showed only a modest increase in secretion of IL-7, leptin and MIP-3α in response to Malassezia yeasts.

Table 1.   Cytokine profile of normal human epidermal keratinocytes (NHEK) exposed to Malassezia globosa or Malassezia restricta
 Fold increase
M. globosaM. restricta
  1. NHEK cells (1 × 106 cells dish−1) were exposed to M. globosa or M. restricta at a multiplicity of infection of 20 for 24 h, and the cell culture supernatants were subjected to the human cytokine antibody array VI. Each experiment was done in duplicate. Data are presented as the fold increase relative to control. The cytokines exhibited a twofold or greater increase in response to Malassezia yeasts are represented in bold.

BMP-42.31.6
BMP-61.91.4
FGF-61.01.4
GM-CSF4.41.3
IL-102.71.3
IL-132.51.3
IL-32.61.4
IL-40.92.0
IL-52.91.3
IL-62.71.4
IL-71.71.1
Leptin0.81.6
MIP-3α0.71.9

Comparison of cytokine profiles of human keratinoyctes in response to various Malassezia species

Previous work has reported that both M. globosa and M. restricta were detected in c. 90% of both AD and healthy subjects, while the other Malassezia species were observed in <40% of AD (Sugita et al., 2001). To assess and compare cytokine secretion changes in response to various Malassezia species, culture supernatants from PHK16-0b cells exposed to M. globosa, M. restricta, M. furfur or M. sympodialis were collected and analysed using human antibody array VI (Table 2). Exposure to M. globosa resulted in a marked increase in IL-5, IL-10 and IL-13 secretion from keratinocytes, whereas M. restricta induced IL-4 secretion. In contrast, no remarkable secretion of Th2-type cytokines was observed when keratinocytes were exposed to M. furfur or M. sympodialis. Only a few cytokines including BMP-6 and IL-6 were secreted from keratinocytes in response to M. sympodialis. This result was in line with the hypothesis that M. globosa and M. restricta play a role in the pathogenesis of AD (Sugita et al., 2001).

Table 2.   Major differences of cytokine secretion in PHK16-0b keratinocytes exposed to Malassezia species
CytokinesFold increase
M. globosaM. restrictaM. furfurM. sympodialis
  1. PHK16-0b cells (1 × 106 cells dish−1) were exposed to Malassezia globosa, M. restricta, M. furfur or M. sympodialis at an multiplicity of infection of 20 for 24 h. Cell culture supernatants were subject to the human cytokine antibody array VI. Each experiment was done in duplicate. Data are presented as the fold increase relative to control. The cytokines exhibited a twofold or greater increase in response to Malassezia yeasts are represented in bold.

BMP-45.20.50.91.7
BMP-62.71.71.62.4
FGF-61.32.71.41.8
GM-CSF8.11.01.00.9
IL-102.41.01.60.9
IL-133.61.01.21.0
IL-32.11.00.91.2
IL-40.83.21.11.0
IL-52.81.01.00.9
IL-62.51.01.02.4
IL-72.21.11.31.1
Leptin0.75.21.20.9
MIP-3α0.72.41.00.9

Characteristics of cDNA microarray analysis

cDNA microarray analysis was performed to confirm the pattern of cytokine expression induced by Malassezia yeasts. PHK16-0b cells were exposed to M. globosa or M. restricta at an MOI of 20. The short exposure periods (2 and 4 h) were chosen to avoid secondary effects such as de novo secretion of other molecules that may alter cytokine expression. After these incubation periods, total RNA was extracted from the keratinocytes and the cytokine gene expression profile was determined by cDNA microarray analysis (Table 3). The level at which M. globosa-exposed cells showed the differential expression of 44 cytokine-related genes (26 genes were upregulated and 18 genes were downregulated) was threefold greater than that in nontreated cells after 2 h incubation. Similarly, M. restricta induced the regulation of 38 genes (27 genes were upregulated and 11 genes were downregulated) at 2 h. Almost similar results were obtained after 4 h incubation. The upregulation of several genes was observed to be specific for the Malassezia species, i.e. M. globosa was capable of inducing the gene transcription of Th2-type cytokines (IL-5, IL-10, and IL-13) in keratinocytes, while M. restricta specifically induced the upregulation of the IL-4 gene. Increased expressions of chemokine receptor genes (CCR2, CCR4, CCR5, XCR1, CX3CR1 and CXCR4) were also observed with both yeast species.

Table 3.   Cytokine gene expression profiles of human keratinocytes exposed to Malassezia globosa and Malassezia restricta
DescriptionGene nameFold regulation in
M. globosaM. restricta
2 h4 h2 h4 h
  1. PHK16-0b monolayers (1 × 106 cells dish−1) were exposed to M. globosa and M. restricta at a multiplicity of infection of 20 for 2 and 4 h. The differential expression of the cytokine transcripts was determined by comparing the calculated fold change in the signal intensity values of the exposed cells and the corresponding control cells. The upregulated genes are represented in black, and the downregulated ones are represented in grey. inline image, up-regulated; inline image, down-regulated.

Homo sapiens Burkitt lymphoma receptor 1, GTP-binding protein (BLR1)CXCR5 (BLR1)2.91.84.54.2
Chemokine (C-C motif) receptor 1CCR12.81.55.33.9
Chemokine (C-C motif) receptor 2CCR214.110.311.69.3
Chemokine (C-C motif) receptor 3CCR32.83.34.55.9
Chemokine (C-C motif) receptor 4CCR414.015.918.623.4
Chemokine (C-C motif) receptor 5CCR510.213.08.411.9
Chemokine (C-C motif) receptor 6CCR61.92.52.13.3
Chemokine (C-C motif) receptor 7CCR7−2.7−1.9−1.0−1.5
Chemokine (C-C motif) receptor 8CCR81.11.34.13.1
Chemokine (C-C motif) receptor 9CCR91.21.61.21.8
Homo sapiens chemokine (C motif) XC receptor 1 (CCXCR1)XCR16.46.78.09.2
Chemokine (C-X3-C) receptor 1CX3CR13.43.16.36.1
Chemokine (C-X-C motif), receptor 4 (fusin)CXCR44.75.54.96.5
Interferon, γIFN-r6.88.14.25.7
Interleukin 10IL-103.43.0−4.9−2.4
interleukin 10 receptor, αIL-10Ra1.32.11.32.3
Interleukin 10 receptor, βIL-10Rb7.14.46.04.4
Interleukin 11IL-11−1.9−1.9−1.3−1.1
Interleukin 11 receptor, αIL-11Ra−1.4−1.3−1.3−1.2
Interleukin 12A, p35IL-12A7.14.418.312.1
Interleukin 12B,p40IL-12B4.92.112.96.6
Interleukin 12 receptor, β 1IL-12Rb11.41.14.12.5
Interleukin 12 receptor, β 2IL-12Rb2−1.1−1.0−1.8−1.4
Interleukin 13IL-133.43.9−1.7−1.1
Interleukin 13 receptor, α 1IL13RA1−3.5−1.51.11.4
Interleukin 13 receptor, α 2IL-13Ra2−11.1−3.53.24.1
Interleukin 15IL-15−5.4−3.4−2.0−2.1
Interleukin 15 receptor, αIL-15Ra20.632.911.220.8
Interleukin 16 (lymphocyte chemoattractant factor)IL-166.622.522.925.7
Interleukin 17 (cytotoxic T-lymphocyte-associated serine esterase 8)IL-171.01.01.82.2
Homo sapiens IL-17 receptor mRNAIL-17 R−1.3−1.1−1.8−1.1
Interleukin 18 (interferon-γ-inducing factor)IL-18−2.81.21.61.8
Interleukin 18 receptor 1IL18R1−5.7−3.7−2.5−2.4
Interleukin 1, αIL-1a−23.4−5.21.92.9
Interleukin 1, βIL-1b−3.6−1.1−2.1−1.3
Interleukin-1 receptor type IIL-1R1−1.5−1.81.92.1
Interleukin-1 receptor type IIIL-1R2−1.0−1.02.82.6
Interleukin 2IL-2−11.1−7.5−1.6−1.5
Interleukin 20IL20−1.2−1.25.310.0
Homo sapiens interleukin 21 (IL21)IL211.11.8−1.2−1.5
Likely ortholog of mouse interleukin 25IL25−16.0−9.2−4.3−2.1
Interleukin 2 receptor, αIL-2 Ra−51.4−27.7−22.7−20.2
Interleukin 2 receptor, βIL-2 Rb8.818.21.72.9
Interleukin 2 receptor, γ (severe combined immunodeficiency)IL-2 Rr−5.6−1.3−1.8−2.6
Interleukin 4IL-41.21.93.76.8
Interleukin 5 (colony-stimulating factor, eosinophil)IL-53.54.52.12.2
Interleukin 5 receptor, αIL-5 Ra3.26.714.033.2
Interleukin 6 (interferon, β 2)IL-65.03.92.62.7
Interleukin 6 receptorIL-6 Ra−1.9−1.7−4.4−3.9
Interleukin 6 signal transducer (gp130, oncostatin M receptor)gp130−8.7−8.0−1.0−1.0
Interleukin 9IL-9/p40−1.61.13.05.1
Interleukin 9 receptorIL-9 Ra−3.6−4.11.0−1.9
Leptin (murine obesity homolog)Leptin−1.41.63.12.1
Lymphotoxin-α (TNF superfamily, member 1)TNF-b/Lta19.521.616.721.2
Lymphotoxine-βLT-b4.48.82.66.0
Homo sapiens lymphotoxin β receptor (TNFR superfamily, member 3(LTBR)LTbR−27.6−19.0−6.0−4.0
Macrophage migration inhibitory factor (glycosylation-inhibiting factor)MIF−2.4−1.5−2.1−1.6
Small inducible cytokine A1 (I-309, homologous to mouse Tca-3)I-309−3.8−2.0−2.2−1.1
Small inducible cytokine subfamily A (Cys-Cys), member 11 (eotaxin)Eotaxin−1.41.1−1.7−1.0
Small inducible cytokine subfamily A (Cys-Cys), member 13MCP-42.22.91.21.7
Small inducible cytokine subfamily A (Cys-Cys), member 14HCC-1−4.8−2.8−5.9−2.8
Small inducible cytokine subfamily A (Cys-Cys), member 15MIP-1 delta−3.1−1.5−4.3−2.0
Small inducible cytokine subfamily A (Cys-Cys), member 16HCC-4−2.0−1.5−2.9−2.0
Small inducible cytokine subfamily A (Cys-Cys), member 17TARC (SCYA17)−1.1−1.1−2.3−2.0
Small inducible cytokine subfamily A (Cys-Cys), member 18, pulmonary and activation-regulatedPARC1.02.41.02.7
Small inducible cytokine subfamily A (Cys-Cys), member 19SCYA192.74.11.21.8
Small inducible cytokine A2 (monocyte chemotactic protein 1, homologous to mouse Sig-je)MCP-1(SCYA2)2.91.0−2.51.0
Small inducible cytokine subfamily A (Cys-Cys), member 20MIP-3a1.82.43.14.2
Small inducible cytokine subfamily A (Cys-Cys), member 21MIP-2 (SCYA21)1.01.6−1.11.5
Small inducible cytokine subfamily A (Cys-Cys), member 22MDC−1.2−1.1−1.4−1.2
Small inducible cytokine subfamily A (Cys-Cys), member 23MPIF-1−1.5−1.2−3.0−1.6
Small inducible cytokine subfamily A (Cys-Cys), member 24MPIF-21.92.51.64.4
Human chemokine (TECK)TECK1.74.01.22.4
Small inducible cytokine A3 (homologous to mouse Mip-1a)MIP-1a1.11.9−1.8−1.0
Small inducible cytokine A4 (homologous to mouse Mip-1b)MIP-1b6.812.04.07.8
Small inducible cytokine A5 (RANTES)SCYA5 (RANTES)4.58.92.42.4
Homo sapiens mRNA for monocyte chemotactic protein-3 (MCP-3)MCP-3−1.7−1.91.71.6
Small inducible cytokine subfamily A (Cys-Cys), member 8 (monocyte chemotactic protein 2)MCP-22.22.32.93.7
γ-interferon inducible early response gene (small inducible cytokine subfamily B (Cys-X-Cys)P10 (IP 10)−2.7−2.0−7.5−3.0
Small inducible cytokine subfamily B (Cys-X-Cys), member 11I-TAC (IP9) (SCYB11)−1.7−1.2−1.5−1.2
Small inducible cytokine B subfamily (Cys-X-Cys motif), member 13 (B-cell chemoattractant)SCYB13−2.9−1.31.72.3
Small inducible cytokine subfamily B (Cys-X-Cys), member 5 (epithelial-derived neutrophil-activatingENA-783.35.6−1.0−2.1
Human chemokine α 3 (CKA-3) mRNAGCP-2−2.3−1.2−2.8−1.3
Small inducible cytokine subfamily C, member 1 (lymphotactin)Lymphotactin2.63.92.13.4
Small inducible cytokine subfamily C, member 2SCYC23.14.72.64.3
Small inducible cytokine subfamily D (Cys-X3-Cys), member 1 (fractalkine, neurotactin)Fractalkine1.41.91.41.8
Small inducible cytokine subfamily E, member 1 (endothelial monocyte-activating)SCYE1−1.4−1.21.51.6
Stromal cell-derived factor 1SDF11.71.51.21.0
Homo sapiens mRNA for SDF2SDF24.16.54.17.1
Transforming growth factor, αTGF-a−2.5−2.5−2.1−2.2
Transforming growth factor, β 1TGFb1−17.1−10.0−13.5−7.3
Transforming growth factor, β 2TGF b21.32.31.31.9
Transforming growth factor, β 3TGF b34.615.85.017.3
Tumor necrosis factor (TNF superfamily, member 2)TNFa−1.01.11.31.3
Tumor necrosis factor receptor superfamily, member 1ATNFR1−6.2−3.7−7.1−4.5
Human tumor necrosis factor receptor 2TNFR2 (TNFSF1B)−1.3−1.5−1.6−1.3

Discussion

The contribution of Malassezia colonization to the pathogenesis of AD has been proposed on the basis of the observation that most AD patients have a high titre of Malassezia-specific serum IgE antibodies (Leung, 2000), and that organisms of the Malassezia species, particularly M. globosa and M. restricta, are identified at high frequencies in AD patients (Sugita et al., 2001). However, the precise mechanisms by which Malassezia colonization induces the immune and inflammatory cascades that lead to AD remain unclear. It is now apparent that cutaneous inflammatory and immune responses involve close interactions between immunocompetent cells and epidermal keratinocytes (Grone, 2002; Esche et al., 2004; Albanesi et al., 2005). By producing cytokines and chemokines (Grone, 2002) and expressing specific adhesion molecules (Banerjee et al., 2004), keratinocytes play an important role in the initiation and maintenance of allergic responses in the skin. In the present study, we examined the cytokine secretion profiles of human keratinocytes exposed to M. globosa and M. restricta, and we demonstrated that the Malassezia-keratinocyte interaction is important to the pathogenesis of AD. Human keratinocytes can, in fact, produce Th2-type cytokines in response to M. globosa and M. restricta.

AD has been described as a Th2-type disease, at least in the initiating phase (Grewe et al., 1998). Of the Th2-type cytokines, IL-4 and IL-5 are known to orchestrate the immune and inflammatory responses, including IgE synthesis, eosinophil recruitment, and the upregulation of eosinophil function, that are characteristic of the inflammatory state observed in AD (Romagnani, 1995; Koning et al., 1997). Using antibody array and ELISA systems we demonstrated that M. globosa induced IL-5 secretion from PHK16-0b cells, while M. restricta induced IL-4 secretion. Similar results were obtained with NHEK cells. These findings were confirmed by cDNA microarray analysis that showed that M. globosa induced the transcription of the IL-5 gene in keratinocytes, while M. restricta induced the upregulation of the IL-4 gene. Our observation that M. globosa induced both the gene transcription and protein secretion of other Th2-type cytokines, such as IL-10 and IL-13, provides additional evidence for a possible relationship between Malassezia colonization and the Th2-type immune response occurring in AD. IL-10 promotes the development of a Th2 response to antigens and is essential for eosinophil infiltration into the skin in allergic dermatitis (Laouini et al., 2003). IL-10 is expressed by keratinocytes as well as by hemopoietic cells including dendritic cells (DCs), macrophages, mast cells and lymphocytes (Enk & Katz, 1992; Moore et al., 1993). The biological effects of IL-13 on B cells, macrophages and monocytes are very similar to those of IL-4, probably because the IL-4 and IL-13 receptors share a common chain (Schnyder et al., 1996). In B cells, IL-13 promotes proliferation, differentiation and Ig heavy chain class switching to IgE and IgG4 (Minty et al., 1993; Zurawski & deVries, 1994). In addition, IL-13 enhances CD23/FcɛRII expression on resting B cells (Defrance et al., 1994). It was also noteworthy that M. globosa induced the secretion of IL-6 from keratinocytes because the role of IL-6 has been demonstrated in IL-4-dependent IgE synthesis (Vercell et al., 1989). Therefore, it is likely that M. globosa and M. restricta play a synergistic role in triggering or exacerbating AD by inducing the secretion of differential Th2-type cytokines from human keratinocytes. As expected, using ELISA we found that mixed exposure to M. globosa and M. restricta induced simultaneous secretion of IL-4 and IL-5 at extents similar to those observed in the single exposures (data not shown). Our study also revealed that other Malassezia species such as M. furfur and M. sympodialis failed to induce secretion of Th2-type cytokines from keratinocytes. These results are in good agreement with the previous finding that both M. globosa and M. restricta are identified in c. 90% of the AD patients, while M. furfur and M. sympodialis are detected less frequently, suggesting that M. globosa and M. restricta play a role in the pathogenesis of AD.

Another important connection between Malassezia colonization and AD is the increased secretion of colony-stimulating factors and chemokines from keratinocytes. We observed that M. globosa induced GM-CSF secretion from keratinocytes. GM-CSF is essential for the differentiation and maturation of Langerhans cells (LCs) and DCs (Witmer-Pack et al., 1987; Caux et al., 1992). It is known that in AD, GM-CSF is overproduced by keratinocytes (Girolomoni, 1997). The current hypothesis is that the lesional abundance of GM-CSF primarily contributes to the maintenance of the chronic inflammatory process in AD by enhancing the antigen-presenting capacity of LCs and DCs (Witmer-Pack et al., 1987). In addition, the mitogenic effect of GM-CSF on keratinocytes (Braunstein et al., 1994) would result in epidermal hypertrophy, which is a characteristic of chronic inflammation. We also observed that M. restricta induced the secretion of CTACK. CTACK is another Th2-specific chemokine that belongs to the CC chemokine family and selectively attracts cutaneous lymphocyte antigen-positive memory T cells to the inflammatory sites (Morales et al., 1999). A recent finding has revealed that CTACK is upregulated in AD patients (Kakinuma et al., 2003). Taken together, these findings suggest that the secretion of GM-CSF and CTACK from keratinocytes in response to M. globosa and M. restricta colonization may also contribute to the pathogenesis of AD. IL-8, a CXC chemokine, is known to be involved in the inflammatory process by stimulating the migration of neutrophils and lymphocytes (Larsen et al., 1995; Keller et al., 2005). Our present finding that M. globosa can induce the secretion of IL-8 from keratinocytes further extends the proinflammatory role of Malassezia colonization in AD pathogenesis.

We observed that human keratinocytes could produce TIMP-1 and TIMP-2 in response to M. globosa. TIMPs block the activities of matrix metalloproteinases (MMPs) that degrade the various components of the connective tissue matrix (Woessner, 1991). The appropriate regulation of MMPs and their TIMPs during the processes of tissue repair is important for tissue remodelling (Edwards et al., 1996). Elevated levels of serum TIMPs have been documented in AD patients (Katoh et al., 2002). The increased production of TIMPs could contribute to the formation of chronic eruptions such as lichenified plaques in AD.

The molecular mechanisms underlying the upregulation of cytokine production are presently unclear. It has been reported that Toll-like receptor 2 (TLR2) mediates intracellular signalling in human keratinocytes in response to M. furfur, leading to the expression of human β-defensin-2 (hBD2) and IL-8 mRNA (Baroni et al., 2006). Since our study revealed the differential cytokine production from keratinocytes in response to M. globosa and M. restricta, it appears likely that several different microbial components and the corresponding keratinocyte receptors may be involved in triggering cytokine secretion. Further studies will be attempted to identify the keratinocyte receptors and signalling pathways involved.

In conclusion, the results of this study demonstrate that M. globosa and M. restricta induce the secretion of distinct Th2-type cytokines from human keratinocytes; M. globosa induces IL-5, IL-10 and IL-13 secretion, while M. restricta induces IL-4 secretion. We hypothesize that M. globosa and M. restricta play a synergistic role in triggering or exacerbating AD by stimulating the Th2 immune response. Elucidation of the role of Malassezia-induced cytokine secretion from keratinocytes will shed light on the pathogenesis of AD and provide clues for the development of new therapeutic strategies.

Acknowledgements

This study was supported in part by research grants from the Ministry of Education, Culture, Sports, Science and Technology of Japan for an Open Research Center Project.

Ancillary