- Email: [email protected]
Characterization of the CC Chemokine Receptor 3 on Human Keratinocytes
Characterization of the CC Chemokine Receptor 3 on Human Keratinocytes Holger Petering, Christoph Kluthe, Yasmin Dulkys, Peter Kiehl, Paul D. Ponath,* Alexander Kapp, and JoÈrn Elsner
Hannover Medical University, Department of Dermatology and Allergology, Hannover, Germany; *LeukoSite Inc., Cambridge, Massachusetts, U.S.A.
cytes than for eosinophils. In situ hybridization techniques exhibited that basal cell layers of the epidermis were stained homogeneously for CC chemokine receptor 3 mRNA with a decreasing signal to the upper epidermis showing that differentiating and proliferating keratinocytes did express mRNA speci®c for CC chemokine receptor 3. Immunohistochemical studies con®rmed low expression of CC chemokine receptor 3 protein on epidermal keratinocytes compared to the high level observed on in®ltrating eosinophils. Furthermore, stimulation of cultured keratinocytes with eotaxin resulted in an increased [3H]thymidine incorporation indicating a role of CC chemokine receptor 3 in epidermal proliferation and differentiation. These data demonstrate that CC chemokine receptor 3 is expressed not only on hematopoetic cells but also on keratinocytes as nonhematopoetic cells with ectodermal origin. Therefore, the identi®cation of CC chemokine receptor 3 on epidermal keratinocytes may indicate a role for CC chemokine receptor 3 and its ligands in skin physiology and pathophysiology. Key words: chemokine receptor/in¯ammation/keratinocytes. J Invest Dermatol 116:549±555, 2001
CC chemokine receptors are expressed on hematopoetic cells, and these may impart selective homing of monocyte, leukocyte, and lymphocyte subsets to sites of in¯ammation. CC chemokine receptor 3 is the major receptor on eosinophils and is also expressed on other in¯ammatory cells suggesting its important role for allergic diseases such as atopic dermatitis and bronchial asthma. Eotaxin, eotaxin-2 and eotaxin-3 have been identi®ed as ligands that only activate CC chemokine receptor 3. CC chemokine receptor 3 is also activated by other promiscuous ligands, however, such as RANTES and monocyte chemotactic protein 4. To date, CC chemokine receptor 3 has not been reported to be expressed on nonhematopoetic cells. In this study, we investigated whether keratinocytes possess autocrine and paracrine mechanisms for CC chemokine secretion and receptor expression as reported for the expression of interleukin 8 and its receptors. Reverse transcriptase polymerase chain reaction analysis demonstrated that CC chemokine receptor 3 mRNA is expressed constitutively in cultured keratinocytes. The signal quantities of the CC chemokine receptor 3 amplicons showed lower intensities for keratino-
C
The CC chemokines eotaxin, eotaxin-2, and eotaxin-3 have been identi®ed as ligands that exclusively activate the CCR3 and seem to be monospeci®c (Elsner et al, 1996a; 1998; Ponath et al, 1996b; Forssmann et al, 1997; White et al, 1997). Other chemokines recognize more than one chemokine receptor. CCR3 is also activated by other promiscuous ligands, however, such as RANTES, monocyte chemotactic protein 2 (MCP-2), MCP-3, and MCP-4 (Alam et al, 1993; Dahinden et al, 1994; Kapp et al, 1994; Petering et al, 1998). Recent studies have shown that eotaxin, eotaxin-2, and eotaxin3 not only are able to induce chemotaxis but also stimulate the release of reactive oxygen species and induce changes in relative Factin contents of human eosinophils as an indicator for locomotoryassociated cellular responses (Elsner et al, 1996a, 1998). Therefore, eotaxin, eotaxin-2, and eotaxin-3 represent speci®c chemoattractants for certain hematopoetic cells. Furthermore, CXC chemokines, e.g., interleukin-8 (IL-8), do not induce functional activity even on activated human eosinophils indicating that CXCR1 and CXCR2 are not expressed on the cell surface (Petering et al, 1999). These ®ndings may suggest an important role for CCR3 and its ligands in chronic in¯ammatory disorders.
hemokine receptors represent Gi-protein-coupled seven transmembrane receptors that are divided into four subgroups (C, CC, CXC, CX3C) according to their ligands (Baggiolini, 1998; Luster, 1998). Each member is a receptor for a unique set of one or more CC chemokine agonists. Various CC chemokine receptors are expressed on hematopoetic cells, and these may impart selective homing of monocytes, leukocytes, and lymphocytes to sites of in¯ammation. CCR3 is the major receptor on eosinophils (Kitaura et al, 1996; Ponath et al, 1996a) and is also expressed on basophils (Heath et al, 1997; Uguccioni et al, 1997) suggesting an important role for allergic diseases such as atopic dermatitis and bronchial asthma. Furthermore, CCR3 expression was observed on macrophages (Hariharan et al, 1999) and neural microglia cells (Albright et al, 1999). Manuscript received July 7, 1999; revised December 22, 2000; accepted for publication December 29, 2000 Reprint requests to: Prof. JoÈrn Elsner M.D., Hannover Medical University, Department of Dermatology and Allergology, Ricklinger Str. 5, D-30449 Hannover, Germany. Email: [email protected] 0022-202X/01/$15.00
´ Copyright # 2001 by The Society for Investigative Dermatology, Inc. 549
550
PETERING ET AL
The ®nding that CCR3 is the principal receptor for eosinophil responses to CC chemokines has led to the development of monoclonal antibody (MoAb) 7B11 blocking CCR3 ligand binding (Daugherty et al, 1996a; Heath et al, 1997; Elsner et al, 1998; Kitayama et al, 1998). In addition, CC chemokine receptor directed antagonists such as Met-RANTES could be identi®ed. Met-RANTES is principally described as a CCR1 antagonist, but has also weak inhibitory properties on CCR3 (Proudfoot et al, 1996; Wells et al, 1996; Elsner et al, 1997). To date, CCR3 has not been reported to be expressed on nonhematopoetic cells, whereas CXC chemokine receptors exhibit a broad distribution not only on hematopoetic cells but also on endothelial cells, ®broblasts, and melanocytes (Holmes et al, 1991; Murphy and Tiffany, 1991). Primary cultured human keratinocytes are known to express mRNA speci®c for IL-8 as well as for the CC chemokine eotaxin (Hein et al, 1997). Human keratinocytes are known to synthesize and secret CXC chemokines, e.g., IL-8, and can be stimulated by their own cytokine products. To investigate whether human keratinocytes possess similar autocrine and paracrine mechanisms for CC chemokine secretion and CC chemokine receptor expression, reverse transcriptase polymerase chain reaction (RT-PCR) analysis, in situ hybridization, and receptor protein expression analysis were performed with cultured human keratinocytes and skin biopsies. Tissue sections were obtained from healthy individuals, patients with chronic in¯ammatory diseases, and patients suffering from bullous pemphigoid. This subepidermal blistering disease is characterized by a TH2-type cytokine pattern as well as peripheral and tissue eosinophilia. Therefore, in®ltrating eosinophils could be stained for CCR3 as a positive control. In this study, we report that human keratinocytes contain mRNA speci®c for the eotaxin receptor CCR3 and express CCR3 protein. Functional data revealed that CCR3 may play a role in epidermal proliferation and differentiation. MATERIALS AND METHODS Cell preparation Human keratinocyte culture Normal human keratinocytes were obtained from fresh trunk biopsies. Epidermal cells were disaggregated by trypsinization (4°C, 12 h) in Hank's balanced salt solution without Ca2+ and Mg2+ containing 0.25% trypsin. The cells were plated out at 107 cells per 50 ml culture ¯ask in endotoxin-free keratinocyte growth medium (KGM) containing 1.0 mM Ca2+ and incubated with 10% CO2 (37°C). Normal keratinocytes were derived from the second to fourth passages grown as a monolayer to 80% con¯uence. Isolation of human eosinophils Human eosinophils were isolated from heparin-anticoagulated venous blood from normal nonatopic healthy donors. The isolation was performed using Ficoll (Pharmacia) density gradient centrifugation and eosinophils were puri®ed by negative selection with anti-CD16 MoAb (clone 3G8, Immunotech, Hamburg, Germany) coated Dynabeads M-450 (Dynal, Hamburg) as described previously (Elsner et al, 1994). The resulting eosinophil purity was > 99.5% as determined by ¯ow cytometric analysis (FACScan, Becton Dickinson, Heidelberg, Germany) using phycoerythrinconjugated anti-CD16 MoAb (clone 3G8, Immunotech). Chemokines, cytokines, and monoclonal antibodies Eotaxin was purchased from Peprotech (London, U.K.). Bradykinin, epidermal growth factor (EGF), and C5a were obtained from Sigma (Deisenhofen, Germany). The anti-CCR3 MoAb 7B11 was derived from C57BL6 mice immunized with CCR3 transfectants and prepared as described previously (Daugherty et al, 1996; Heath et al, 1997). MoAb EG1 against eosinophilic cationic protein (ECP) was received from Pharmacia (Uppsala). The human IgG1 isotype control was obtained from Sigma. CCR3 mRNA expression Total RNA was isolated from human keratinocytes and eosinophils using the TRIzol kit (Gibco Life Technologies, Karlsruhe, Germany) according to the manufacturer's instructions based on the guanidine isothiocyanate method. First-strand cDNA synthesis and PCR reaction were performed as described previously (Elsner et al, 1996b; Petering et al, 1996, 1999). Primers for the ampli®cation of CCR3 (sense, 5¢-GCCATTTCGGACCTGCTCTT-3¢; antisense, 5¢-TCCGCTCACAGTCATTTCCA-3¢) and b-actin
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
(sense, 5¢-GAGCGGGAAATCGTGCGTGACATT-3¢; antisense, 5¢GAAGGTAGTTTCGTGGATGCC-3¢) were designed according to the published sequences (Accession No.: CCR3, AF026535; b-actin, AB004047). PCR samples were run on a 1.8% agarose gel stained with 0.2 mg per ml ethidium bromide, and the PCR products were visualized with ultraviolet light and photographed. Automated sequencing of PCR products For automated sequencing of PCR products, an Applied Biosystems Model 373 A (Forster City, CA) was used. Nucleotide sequence data were analyzed with Lasergene sequence analysis software (DNASTAR, Madison, WI). CCR3 riboprobes and in situ hybridization RNA probes for CCR3 were designed from a plasmid constructed by cloning a 578 bp RTPCR fragment ampli®ed from human eosinophils into a polylinker site of transcription vector pBluescript II SK± containing T3 and T7 polymerase recognition elements as described previously (Petering et al, 1996). After blunt-end ligation, the orientation and DNA sequence of both insert strains were determined by cycle sequencing as described above. Antisense (T3) and sense (T7) riboprobes were transcribed enzymatically and digoxygenin-labeled according to the manufacturer's instructions (Boehringer Mannheim, Germany). Probes were washed twice by ethanol precipitation to remove unincorporated nucleotides and stored in 10 nM dithiothreitol at ±80°C. For in situ hybridization, tissue sections were ®xed with 4% paraformaldehyde and washed with 1% phosphate-buffered saline (PBS). Tissues were permeabilized with 50 mg per ml protease K (37°C, 30 min), washed once with 1% PBS, and incubated for 5 min with 2 3 standard saline citrate (SSC). The sections were dehydrated through graded alcohol and air-dried. After prehybridization (55°C, 60 min) and hybridization (55°C, 12 h) with standard hybridization solution (10% dextran sulfate, 40% formamide, 4 3 SSC, 1 3 Denhard's, 10 mM dithiothreitol, 100 mg per ml tRNA) slides were washed with SSC under stringent conditions. Staining for CCR3 was carried out with antidigoxygenin F(ab)¢ fragment conjugated to alkaline phosphatase at a ®nal dilution of 1:1000 according to the manufacturer's instructions (Boehringer Mannheim). Slides stained with sense and antisense riboprobes were read blindly by two dermatopathologists. Immunohistochemistry with human keratinocytes and eosinophils Tissue sections were cut to 5 mm and adhered to silanized slides. Immunostaining was performed (room temperature, 60 min) and was developed by the avidin±biotin-peroxidase complex technique with the chromogen fuchsin according to the manufacturer's recommendations (DAKO). Staining for CCR3 protein was performed with the anti-CCR3 MoAb 7b11 (750 mg per ml) at a ®nal dilution of 1:100 in 1% bovine serum albumin/PBS. Staining for eosinophil matrix granular proteins was performed with MoAb EG1 speci®c for ECP (Kiehl and Kapp, 1998). Non-reactive mouse antihuman antibodies of IgG1 isotype were used as negative controls. Immuno¯uorescence of cultured keratinocytes Immuno¯uorescence of cultured human trunk keratinocytes between the second and fourth passages was performed using standard techniques. In brief, keratinocytes were adjusted to a density of 1 3 107 cells per ml. Aliquots (20 ml) containing 2 3 105 cells were incubated at 4°C for 30 min with the indicated antibody or isotype control (IgG2). Thereafter, cells were washed twice with 4°C PBS. For indirect immuno¯uorescence, cells were stained in a second step by a ¯uorescein isothiocyanate (FITC) conjugated mouse antihuman antibody (Immunotech) and subsequently washed twice. Thereafter, cells were analyzed by ¯ow cytometry (FACScan). The samples were excited at 488 nm and emission was measured at 530 nm (FITC-labeled antibodies, ¯uorescence 1, green ¯uorescence). The CCR3 receptor expression was analyzed after preincubation of cultured keratinocytes for 30 min at 37°C with the CCR3 ligand eotaxin or medium alone. The mean channel ¯uorescence of seven experiments was determined. Results were expressed as the median ¯uorescence intensity after preincubation for 30 min versus the median ¯uorescence intensity after incubation with medium. Incubation of cells for longer than 30 min did not change the results. Keratinocyte proliferation assay Keratinocytes were plated at 5 3 103 cells per well of a 96-well microtiter plate and cultured in KGM for 5 d. The medium in each well was then replaced with KGM supplemented with either EGF, IL-8, or eotaxin at the indicated concentrations or KGM alone. Proliferation was measured by labeling six replicate cultures with 0.5 mCi ml [3H]thymidine (Amersham, Braunschweig, Germany) during a further 48 h of incubation. After labeling, medium was removed, each well was washed with PBS, and trichloroacetic acid (TCA) was added. The TCA-precipitated nucleic
VOL. 116, NO. 4 APRIL 2001
CCR3 EXPRESSION ON HUMAN KERATINOCYTES
Figure 1. CCR3 mRNA is expressed in cultured human keratinocytes. mRNA from cultured human keratinocytes and puri®ed eosinophils was isolated and ®rst-strand cDNA synthesis was performed. PCR was carried out with primer pairs speci®c for CCR3 (expected size 578 bp) and b-actin (expected size 221 bp) as an internal standard. The amplicons were separated by electrophoresis in 1.8% agarose gel and stained by ethidium bromide. Lane 1, 50 bp DNA size marker; lane 2, human eosinophil mRNA; lane 3, human keratinocyte mRNA. One representative experiment out of four is shown.
acid was solubilized in 100 ml of 0.1 M sodium hydroxide, 1% sodium dodecyl sulfate, and 2% sodium carbonate for 30 min and the b decay was counted in a liquid scintillation counter. The stimulation index was de®ned by the ratio of mean counts per minute of stimulated to unstimulated cultures.
RESULTS Cultured human keratinocytes express mRNA speci®c for CCR3 RT-PCR analysis was performed to investigate whether human keratinocytes express mRNA speci®c for CCR3. For this purpose, human trunk keratinocytes were cultured and passaged two to four times prior to total RNA isolation. Using the primer pairs indicated, DNA fragments with an expected size of 578 bp speci®c for human CCR3 could be detected in cultured human keratinocytes (Fig 1, lane 3). To clarify whether the ampli®ed PCR products were speci®c for CCR3 and did not represent other chemokine receptors, DNA sequence analysis was performed. PCR products were sequenced automatically with vector-derived primer pairs. Computer-aided nucleotide sequence analysis revealed that all PCR products represent DNA sequences speci®c for CCR3 mRNA. As expected, highly puri®ed human eosinophils did constitutively express CCR3 mRNA (Fig 1, lane 2). To demonstrate that all mRNA preparations were highly puri®ed, RT-PCR was performed simultaneously using primer pairs for bactin as standard (221 bp) (Fig 1). The signal quantities of the CCR3 amplicons showed lower signal intensities for cultured keratinocytes than for isolated human eosinophils. The results indicate that CCR3 is expressed at mRNA level in cultured human keratinocytes. In situ hybridization studies reveal that CCR3 mRNA is expressed in epidermal keratinocytes To investigate whether CCR3 mRNA is not only expressed in vitro in cultured keratinocytes but can also be detected in vivo in epidermal keratinocytes, in situ hybridization studies were performed. Tissue sections from ®ve patients suffering from bullous pemphigoid prior to immunosuppressive therapy were studied for the expression of CCR3 mRNA. Bullous pemphigoid represents a chronic in¯ammatory autoimmune disease characterized by a local accumulation of eosinophils with subsequent subepidermal blister formation (Lever, 1979). Therefore, the colocalization of keratinocytes and eosinophils in®ltrating the upper dermis and epidermis allowed us to compare CCR3 signal speci®city and intensity between different cells. Specimens were obtained from perilesional and lesional skin and CCR3 mRNA expression was analyzed by in situ hybridization. Strong CCR3 mRNA expression was detected in in¯ammatory cells in®ltrating the upper dermis
551
(Fig 2A). Immunohistochemical staining with MoAb EG1 recognizing matrix granular proteins identi®ed these cells as eosinophil granulocytes known to express CCR3 (Fig 2C). Like eosinophils, basal cell layers of the epidermis were stained homogeneously for CCR3 mRNA with a decreasing signal to the upper epidermis showing that differentiating and proliferating keratinocytes did express mRNA speci®c for CCR3 (Fig 2A). Negative controls hybridized with sense RNA probes only showed background signals (data not shown). To investigate whether the expression of CCR3 mRNA in keratinocytes is induced in the state of disease, specimens from healthy volunteers were analyzed by in situ hybridization. As seen in Fig 2(E), basal keratinocytes from normal individuals also showed staining for CCR3 mRNA. A similar staining pattern could be observed in tissue sections from chronic in¯ammatory diseases. Figure 2(G) shows the expression of CCR3 mRNA in atopic dermatitis, a T-cell-mediated disease affecting the upper dermis and epidermis. Figure 2(H) demonstrates the expression of CCR3 mRNA in psoriasis vulgaris, a chronic in¯ammatory disease characterized by epidermal hyperplasia and altered epidermal differentiation. These data reveal that CCR3 mRNA is expressed constitutively in human keratinocytes. The expression may not depend on the state of activation of basal keratinocytes, e.g., in disease affecting the basement membrane zone. Immunohistochemical studies demonstrate that CCR3 is expressed on epidermal keratinocytes To demonstrate that the transcription of CCR3 mRNA is followed by the expression of the receptor protein on the cell surface, bullous pemphigoid tissue sections adjacent to those used for in situ hybridization were stained immunohistochemically for CCR3. The colocalization of keratinocytes and eosinophils allowed us again to compare the signal quantities. Expression of CCR3 protein was observed on epidermal keratinocytes with a marginally increased staining of the basal cell layers (Fig 2B). Immunohistochemical staining of specimens taken from healthy volunteers and patients suffering from atopic dermatitis and psoriasis vulgaris exhibited a similar staining pattern (data not shown). These results indicate that proliferating and differentiating keratinocytes of the epidermis are able to express CCR3 protein. In comparison to human eosinophils in®ltrating the upper dermis and epidermis, however, keratinocytes showed only a weak CCR3 expression in terms of staining intensity indicating fewer binding sites per cell (Fig 2B, D). Positive CCR3 staining of eosinophils without immunohistochemical reaction on other in®ltrating in¯ammatory cells, such as lymphocyte subpopulations, demonstrated that anti-CCR3 MoAb 7B11 is also a speci®c marker for CCR3 in immunohistochemical studies (Fig 2D). Flow cytometry analysis reveals that CCR3 is expressed heterogeneously on cultured keratinocytes To con®rm the immunohistochemical data, the expression of CCR3 on cultured human trunk keratinocytes was investigated by ¯ow cytometry. Cells between the second and fourth passage were stained with the anti-CCR3 MoAb 7B11 (Fig 3A). Surprisingly, the expression of CCR3 on cultured keratinocytes was donor dependent and differed between the healthy volunteers. Cultured keratinocytes from two donors were negative or showed only a weak staining for CCR3 in a small subpopulation of keratinocytes (Fig 3B). Preincubation of cultured keratinocytes for 30 min with TH2-type cytokines (IL-4, IL-5) did not alter CCR3 expression (data not shown). To demonstrate whether the antiCCR3 MoAb 7B11 binds speci®cally to CCR3, cultured human keratinocytes were incubated for 30 min with eotaxin at a concentration of 125.0 nM or medium. As expected, preincubation of cultured keratinocytes with eotaxin induced downregulation of CCR3 on the cell surface, indicating that the CCR3 ligand eotaxin was able to bind speci®cally to its receptor expressed on keratinocytes (Fig 3A).
552
PETERING ET AL
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
Figure 2. In situ hybridization and immunohistochemical analysis for CCR3 mRNA and protein expression on human keratinocytes. (A) In tissue sections from patients with bullous pemphigoid CCR3 mRNA expression was analyzed by in situ hybridization. Basal cell layers of the epidermis were homogeneously stained for CCR3 mRNA with a decreasing intensity to the upper epidermis. Strong CCR3 mRNA expression was detected in in¯ammatory cells in®ltrating the upper dermis. (B) Immunohistochemistry with anti-CCR3 MoAb 7B11 revealed that CCR3 protein is expressed on epidermal keratinocytes. (C) Immunohistochemical staining with MoAb EG1 recognizing ECP identi®ed in¯ammatory cells as human eosinophils. (D) Human eosinophils in®ltrating the upper dermis showed more intense immunohistochemical staining for CCR3 than epidermal keratinocytes indicating more binding sites per cell. (E) In situ hybridization revealed that basal keratinocytes from normal individuals also showed staining for CCR3 mRNA. A similar staining pattern could be observed in chronic in¯ammatory diseases: (F) expression of CCR3 mRNA in bullous pemphigoid; (G) expression of CCR3 mRNA in atopic dermatitis; (H) expression of CCR3 mRNA in psoriasis vulgaris. Scale bars: (A) 100 mm; (B±D) 50 mm.
Proliferation of cultured human keratinocytes after stimulation with eotaxin can be inhibited by antiCCR3 Previous studies have shown that eotaxin is expressed in tissue specimens obtained from chronic in¯ammatory lesions (Rothenberg et al, 1995). These ®ndings prompted us to investigate whether eotaxin may have any functional effects on epidermal keratinocytes, e.g., cell proliferation and differentiation. Cultured human epidermal keratinocytes were stimulated with eotaxin (10±1000 nM) and the incorporation of [3H]thymidine was detected as a marker for cell proliferation. As seen in Fig 4, eotaxin
induced dose-dependently a proliferation of human keratinocytes. In this experiment, eotaxin was as effective as IL-8 with a maximum of cell proliferation at a concentration of 100 nM (Fig 4). In addition, proliferation of cultured human keratinocytes was also induced by other CCR3 ligands such as RANTES and MCP-4 representing two other potent activators of CCR3. At equivalent concentrations eotaxin seemed to be more effective than RANTES or MCP-4, however (Fig 4). EGF used as a positive control was more potent than eotaxin at concentrations above 1 mM. Moreover, whereas EGF induced a dose-related curve all
VOL. 116, NO. 4 APRIL 2001
CCR3 EXPRESSION ON HUMAN KERATINOCYTES
553
Figure 4. Proliferation of cultured human keratinocytes after stimulation with eotaxin can be inhibited by anti-CCR3 MoAb 7B11. Cultured human keratinocytes were stimulated with CC chemokines eotaxin, MCP-4, RANTES, and IL-8 as well as EGF in different concentrations (10±1000 nM). The incorporation of [3H]thymidine was detected as a marker for cell proliferation. Data are expressed as proliferation index (stimuli-induced [3H]thymidine incorporation/medium-induced [3H]thymidine incorporation). In one experiment keratinocytes were preincubated with the anti-CCR3 MoAb 7B11 and subsequently stimulated with eotaxin (10±7 M). One representative experiment out of four is shown indicating that eotaxin induces proliferation of human keratinocytes.
that the proliferation of keratinocytes after stimulation with eotaxin can be inhibited by antagonizing CCR3-mediated effects. DISCUSSION
Figure 3. Flow cytometry analysis reveals that CCR3 is expressed heterogeneously on cultured keratinocytes. Overlay histogram for the expression of CCR3 on cultured human trunk keratinocytes. Cells between the second and fourth passage were stained for CCR3 with the anti-CCR3 MoAb 7B11. (A) The bold line represents cultured keratinocytes expressing CCR3. The dotted line indicates cultured keratinocytes preincubated with 125.0 nM eotaxin for 30 min showing a downregulation of the CCR3. The thin line represents the IgG2a isotype control. These data demonstrate that eotaxin was able to bind speci®cally to CCR3 expressed on human keratinocytes. (B) The expression of CCR3 was donor dependent and differed between healthy volunteers. In two donors cultured keratinocytes appeared negative or showed only a weak staining for CCR3 in a small subpopulation of keratinocytes as demonstrated by the bold line.
chemokines tested induced a bell-shaped curve with respect to incorporation of [3H]thymidine in epidermal keratinocytes. To investigate whether the eotaxin-induced proliferation of keratinocytes was mediated via CCR3, cultured keratinocytes were preincubated with anti-CCR3 MoAb 7B11. As demonstrated previously, this antibody is able to block eotaxin binding to CCR3 completely and inhibits CCR3-speci®c cell activation as demonstrated previously (Daugherty et al, 1996a; Heath et al, 1997; Elsner et al, 1998). As seen in Fig 4 preincubation of keratinocytes with the anti-CCR3 MoAb 7B11 inhibited the proliferation of keratinocytes induced by eotaxin. This experiment demonstrates
Chemokines are proin¯ammatory cytokines known to be involved in chronic in¯ammatory diseases such as atopic dermatitis and allergic asthma (Baggiolini, 1998; Luster, 1998). Recent studies have shown that the CC chemokine eotaxin is elevated in the bronchial mucosa of asthmatic patients and is expressed by epithelial and endothelial cells (Ying et al, 1997). CCR3 was identi®ed independently by two different groups as the eotaxin receptor on human eosinophils (Daugherty et al, 1996; Ponath et al, 1996a) and further studies showed the expression of CCR3 also on basophils suggesting that this chemokine receptor may play a central role in chronic in¯ammatory disorders (Kitaura et al, 1996; Ponath et al, 1996b; Uguccioni et al, 1997). So far, CCR3 has only been identi®ed on hematopoetic cells whereas CXC chemokine receptors are also expressed on nonhematopoetic cells such as ®broblasts, melanocytes, and keratinocytes (Holmes et al, 1991; Murphy et al, 1991; Norgauer et al, 1996). In psoriatic skin, characterized by epidermal hyperplasia and altered epidermal differentiation, IL-8 receptors were found to be expressed at high levels (Schulz et al, 1993; Kulke et al, 1998). In vitro studies con®rmed that keratinocyte migration and growth can be stimulated by its ligand IL-8 (Tuschil et al, 1992). As endothelial and epithelial cells of the lung and also epidermal keratinocytes are known to produce CXC chemokines such as IL-8 (Sica et al, 1990; Gillitzer et al, 1991; Huber et al, 1991), autocrine and paracrine mechanisms seem to be responsible for proliferation and differentiation of nonhematopoetic cells. Recently, increased levels of CC chemokine eotaxin have been found in respiratory epithelium and endothelium as well as in the intestinal mucosa (Garcia-Zepeda et al, 1996; Matthews et al, 1998). Hein et al detected eotaxin mRNA at low levels in stimulated primary keratinocytes (Hein et al, 1997). Therefore, it is tempting to speculate that human keratinocytes also possess autocrine and paracrine mechanisms for CC chemokine secretion and CC
554
PETERING ET AL
chemokine receptor expression similar to their CXC counterpart. This study describes the functional analysis and expression of the predominant eosinophil CC chemokine receptor CCR3 on human keratinocytes in vitro in comparison to isolated eosinophils and in vivo to eosinophils in®ltrating the upper dermis and epidermis. In the ®rst set of experiments, RT-PCR analysis was performed demonstrating that cultured human keratinocytes constitutively express CCR3 mRNA. Signal quantities of CCR3 amplicons showed lower signal intensities for keratinocytes than for isolated eosinophils. To rule out that the RT-PCR products were due to a false-positive ampli®cation of other chemokine receptors, e.g., CXCR1 or CXCR2, nucleotide sequence analysis of the amplicons was performed and revealed a DNA sequence homolog to published CCR3 mRNA sequences. To con®rm that CCR3 mRNA expression was also observed in vivo, skin biopsies were analyzed for CCR3 mRNA by in situ hybridization techniques. Again, CCR3 mRNA could be detected and was predominantly expressed in basal keratinocytes whereas in the upper cell layers the receptor expression was decreased. To investigate whether the expression of CCR3 mRNA is induced in the state of disease, skin biopsies from healthy volunteers and patients suffering from different in¯ammatory diseases (atopic dermatitis, psoriasis vulgaris, and bullous pemphigoid) were analyzed. Our data revealed that CCR3 mRNA is constitutively expressed on epidermal keratinocytes and may not depend on the state of cell activation. The ®nding that CCR3 mRNA is constitutively expressed on keratinocytes obtained from different in¯ammatory diseases affecting the epidermis (psoriasis vulgaris), basal membrane (bullous pemphigoid), and epidermis as well as the upper dermis (atopic dermatitis) suggests that the low level of CCR3 mRNA expression may not have a pathophysiologic role in these disorders. Our results indicate that in particular proliferating and differentiating keratinocytes of the basal stratum express CCR3 mRNA. This study provides evidence that CCR3 mRNA is not only expressed on in¯ammatory cells derived from hematopoetic precursors but can also be expressed on human keratinocytes as nonhematopoetic cells with ectodermal origin. The expression of CCR3 protein on keratinocytes was investigated by immunohistochemical staining for CCR3 with tissue sections adjacent to those used for in situ hybridization. The experiments indicated that CCR3 protein is expressed in human epidermis. Immunohistochemistry was performed with tissue sections from patients with bullous pemphigoid, a blistering autoimmune disease characterized by an excessive synthesis of proin¯ammatory cytokines. TH2-type cytokines such as IL-4 and IL-5 but also CC chemokines, e.g., RANTES, have been identi®ed in the skin and the blister ¯uid of these patients and contribute to the epidermal and dermal in®ltration of eosinophils (D'Auria et al, 1998; Inaoki and Takehara, 1998). IL-4 and IL-5 may upregulate the expression of CCR3 on in¯ammatory cells and may also increase CCR3 protein synthesis in epidermal keratinocytes (Tiffany et al, 1998; Jinquan et al, 1999). Therefore, it was obvious to speculate that cytokine priming of keratinocytes in vivo may contribute to the expression of CCR3 protein. In our in vitro experiments, however, no difference in CCR3 expression could be observed after preincubation of cultured keratinocytes with TH2type cytokines. The colocalization of keratinocytes and eosinophils in®ltrating the upper dermis and epidermis in skin biopsies from bullous pemphigoid allowed us to compare CCR3 signal speci®city and intensity between different cells indicating that human keratinocytes possess fewer binding sites per cell than human eosinophils. To con®rm these immunohistochemical ®ndings ¯ow cytometry analysis was performed demonstrating an inhomogeneous expression of CCR3 on cultured keratinocytes, whereas CCR3 mRNA was expressed constitutively indicating a receptor regulation at the post-transcriptional level. The CCR3 expression on the cell surface was donor dependent and differed between our healthy volunteers. Moreover, we could clearly demonstrate that preincubation of CCR3+ keratinocytes with eotaxin resulted in a downregulation of
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
CCR3, which has recently been shown in eosinophils (Elsner et al, 2000). Preincubation of keratinocytes with TH2-type cytokines did not alter CCR3 receptor expression. Recent studies have shown that the CXC chemokine IL-8 exerts a direct in¯uence on several keratinocyte functions including epidermal proliferation and differentiation (Kapp, 1993). Furthermore, in psoriatic skin, IL-8 and the growth-related oncogen a, another ligand for CXCR2, are produced by lesional keratinocytes especially from the upper epidermal cell layers (Larsen et al, 1989) as part of an autocrine and paracrine cytokine network. The CC chemokine eotaxin can also be synthesized and released by keratinocytes and therefore not only may contribute to the local accumulation of eosinophils and TH2-type lymphocytes but also may modulate the differentiation and proliferation of keratinocytes similar to IL-8 (Hein et al, 1997). To investigate whether such autocrine and paracrine mechanisms exist for CC chemokines, growth-promoting activities of eotaxin on cultured keratinocytes were analyzed. Our data suggest that eotaxin represents a weak stimulus for proliferation of human keratinocytes. The observation that keratinocyte proliferation and differentiation was also induced by RANTES and MCP-4, two other potent activators of CCR3, suggests that the effect of eotaxin may be due to CCR3 binding and signaling. Interestingly, all of the CCR3 ligands tested induced a bell-shaped curve with regard to keratinocyte proliferation. As we could clearly demonstrate that eotaxin was able to induce a downregulation of CCR3 in CCR3+ keratinocytes it could be speculated that CCR3 ligands downregulate CCR3 at a concentration of 10±7 M. Thus, the reduced proliferation index of keratinocytes in response to a higher concentration (10±6 M) of CCR3 ligands could be explained by a downregulation of CCR3. To further prove this hypothesis, cultured keratinocytes were preincubated with anti-CCR3 MoAb 7B11 known to speci®cally block the activation of human eosinophils by CCR3 ligands (Heath et al, 1997; Elsner et al, 1998; Kitayama et al, 1998). The antiCCR3 MoAb 7B11 was able to inhibit eotaxin-induced proliferation of keratinocytes indicating that binding to CCR3 is the initial step of the observed keratinocyte differentiation and proliferation. In summary, these data provide evidence that CCR3 is expressed not only on hematopoetic cells but also on human keratinocytes as nonhematopoetic cells with ectodermal origin. The level of receptor expression is donor dependent. In addition to its involvement in chemotaxis and activation of the respiratory burst of human eosinophils eotaxin and other CCR3 ligands may contribute to local wound repair and healing mechanisms integrating different ways of tissue reconstitution. Therefore, the identi®cation of CCR3 on epidermal keratinocytes suggests that CCR3 and its ligands, especially eotaxin, may play an important role in skin physiology and pathophysiology. This work was supported by a grant from the Deutsche Forschungsgemeinschaft (EL 160/6±1) and by the Hochschulinterne LeistungsfoÈrderung of the Medical University Hannover.
REFERENCES Alam R, Stafford S, Forythe P, Harrison R, Faubion D, Lett-Brown MA, Grant JA: RANTES is a chemotactic and activating factor for human eosinophils. J Immunol 150:3442±3447, 1993 Albright AV, Shieh JT, Itoh T, et al: Microglia express CCR5, CXCR4, and CCR3, but of these, CCR5 is the principal coreceptor for human immunode®ciency virus type 1 dementia isolates. J Virol 73:205±213, 1999 Baggiolini M: Chemokines and leukocyte traf®c. Nature 392:565±568, 1998 Dahinden CA, Geiser T, Brunner T, et al: Monocyte chemotactic protein 3 is a most effective basophil- and eosinophil-activating chemokine. J Exp Med 179:751± 756, 1994 Daugherty BL, Siciliano SJ, DeMartino JA, Malkowitz L, Sirotina A, Springer MS: Cloning, expression, and characterization of the human eosinophil eotaxin receptor. J Exp Med 183:2349±2354, 1996a D'Auria L, Pietravalle M, Mastroianni A, et al: IL-5 levels in the serum and blister ¯uid of patients with bullous pemphigoid: correlations with eosinophil cationic
VOL. 116, NO. 4 APRIL 2001
protein, RANTES, IgE and disease severity. Arch Dermatol Res 290:25±27, 1998 Elsner J, Oppermann M, Czech W, Dobos G, SchoÈpf E, Norgauer J, Kapp A: C3a activates reactive oxygen radical species production and intracellular calcium transients in human eosinophils. Eur J Immunol 24:518±522, 1994 Elsner J, HoÈchstetter R, Kimmig D, Kapp A: Human eotaxin represents a potent activator of the respiratory burst of human eosinophils. Eur J Immunol 26:1919± 1925, 1996a Elsner J, HoÈchstetter R, Spiekermann K, Kapp A: Surface and mRNA expression of the CD52 antigen by human eosinophils but not by neutrophils. Blood 88:4684±4693, 1996b Elsner J, Petering H, HoÈchstetter R, Kimmig D, Wells TN, Kapp A, Proudfoot AE: The CC chemokine antagonist Met-RANTES inhibits eosinophil effector functions through the chemokine receptors CCR1 and CCR3. Eur J Immunol 27:2892±2898, 1997 Elsner J, Petering H, Kluthe C, Kimmig D, Smolarski R, Ponath P, Kapp A: Eotaxin-2 activates chemotaxis-related events and release of reactive oxygen species via pertussis toxin-sensitive G proteins in human eosinophils. Eur J Immunol 28:2152±2158, 1998 Elsner J, Mack M, Bruhl H, et al: Differential activation of CC chemokine receptors by AOP-RANTES. J Biol Chem 275:7787±7794, 2000 Forssmann U, Uguccioni M, Loetscher P, Dahinden CA, Langen H, Thelen M, Baggiolini M: Eotaxin-2, a novel CC chemokine that is selective for the chemokine receptor CCR3, and acts like eotaxin on human eosinophil and basophil leukocytes [In Process Citation]. J Exp Med 185:2171±2176, 1997 Garcia-Zepeda EA, Rothenberg ME, Ownbey RT, Celestin J, Leder P, Luster AD: Human eotaxin is a speci®c chemoattractant for eosinophil cells and provides a new mechanism to explain tissue eosinophilia. Nat Med 2:449±456, 1996 Gillitzer R, Berger R, Mielke V, Muller C, Wolff K, Stingl G: Upper keratinocytes of psoriatic skin lesions express high levels of NAP-1/IL-8 mRNA in situ. J Invest Dermatol 97:73±79, 1991 Hariharan D, Douglas SD, Lee B, Lai JP, Campbell DE, Ho WZ: Interferon-gamma upregulates CCR5 expression in cord and adult blood mononuclear phagocytes. Blood 93:1137±1144, 1999 Heath H, Qin S, Rao P, et al: Chemokine receptor usage by human eosinophils. The importance of CCR3 demonstrated using an antagonistic monoclonal antibody. J Clin Invest 99:178±184, 1997 Hein H, Schluter C, Kulke R, Christophers E, Schroder JM, Bartels J: Genomic organization, sequence, and transcriptional regulation of the human eotaxin gene. Biochem Biophys Res Commun 237:537±542, 1997 Holmes WE, Lee J, Kuang WJ, Rice GC, Wood WI: Structure and functional expression of a human interleukin-8 receptor. Science 253:1278±1280, 1991 Huber AR, Kunkel SL, Todd RF3, Weiss SJ: Regulation of transendothelial neutrophil migration by endogenous interleukin-8 [published errata appear in Science 1991 November 1, 254:631 and 1991 December 6, 254:1435]. Science 254:99±102, 1991 Inaoki M, Takehara K: Increased serum levels of interleukin (IL) -5, IL-6 and IL-8 in bullous pemphigoid. J Dermatol Sci 16:152±157, 1998 Jinquan T, Quan S, Feili G, Larsen CG, Thestrup-Pedersen K: Eotaxin activates T cells to chemotaxis and adhesion only if induced to express CCR3 by IL-2 together with IL-4. J Immunol 162:4285±4292, 1999 Kapp A: The role of cytokines in the psoriatic in¯ammation. J Dermatol Sci 5:133± 142, 1993 Kapp A, Zeck Kapp G, Czech W, SchoÈpf E: The chemokine RANTES is more than a chemoattractant: characterization of its effect on human eosinophil oxidative metabolism and morphology in comparison with IL-5 and GM-CSF. J Invest Dermatol 102:906±914, 1994 Kiehl P, Kapp A: Tissue eosinophilia and local deposition of eosinophil-speci®c granule proteins. Regulation and signi®cance for in¯ammatory response in atopic dermatitis and other in¯ammatory dermatoses. Hautarzt 49:176±183, 1998 Kitaura M, Nakajima T, Imai T, et al: Molecular cloning of human eotaxin, an eosinophil-selective CC chemokine, and identi®cation of a speci®c eosinophil eotaxin receptor, CC chemokine receptor 3. J Biol Chem 271:7725±7730, 1996 Kitayama J, Mackay CR, Ponath PD, Springer TA: The C-C chemokine receptor
CCR3 EXPRESSION ON HUMAN KERATINOCYTES
555
CCR3 participates in stimulation of eosinophil arrest on in¯ammatory endothelium in shear ¯ow. J Clin Invest 101:2017±2024, 1998 Kulke R, Bornscheuer E, SchluÈter C, Bartels J, Rowert J, Sticherling M, Christophers E: The CXC receptor 2 is overexpressed in psoriatic epidermis. J Invest Dermatol 110:90±94, 1998 Larsen CG, Anderson AO, Oppenheim JJ, Matsushima K: Production of interleukin8 by human dermal ®broblasts and keratinocytes in response to interleukin-1 or tumour necrosis factor. Immunology 68:31±36, 1989 Lever WF: Pemphigus and pemphigoid. A review of the advances made since 1964. J Am Acad Dermatol 1:2±31, 1979 Luster AD: Chemokines ± chemotactic cytokines that mediate in¯ammation. N Engl J Med 338:436±445, 1998 Matthews AN, Friend DS, Zimmermann N, et al: Eotaxin is required for the baseline level of tissue eosinophils. Proc Natl Acad Sci USA 95:6273±6278, 1998 Murphy PM, Tiffany HL: Cloning of complementary DNA encoding a functional human interleukin-8 receptor. Science 253:1280±1283, 1991 Norgauer J, Metzner B, Schraufstatter I: Expression and growth-promoting function of the IL-8 receptor beta in human melanoma cells. J Immunol 156:1132±1137, 1996 Petering H, Hammerschmidt S, Frosch M, van Putten JP, Ison CA, Robertson BD: Genes associated with meningococcal capsule complex are also found in Neisseria gonorrhoeae. J Bacteriol 178:3342±3345, 1996 Petering H, HoÈchstetter R, Kimmig D, Smolarski R, Kapp A, Elsner J: Detection of MCP-4 in dermal ®broblasts and its activation of the respiratory burst in human eosinophils. J Immunol 160:555±558, 1998 Petering H, GoÈtze O, Kimmig D, Smolarski R, Kapp A, Elsner J: The biologic role of interleukin-8: functional analysis and expression of CXCR1 and CXCR2 on human eosinophils. Blood 93:694±702, 1999 Ponath PD, Qin S, Post TW, et al: Molecular cloning and characterization of a human eotaxin receptor expressed selectively on eosinophils. J Exp Med 183:2437±2448, 1996a Ponath PD, Qin S, Ringler DJ, et al: Cloning of the human eosinophil chemoattractant, eotaxin. Expression, receptor binding, and functional properties suggest a mechanism for the selective recruitment of eosinophils. J Clin Invest 97:604±612, 1996b Proudfoot AE, Power CA, Hoogewerf AJ, Montjovent MO, Borlat F, Offord RE, Wells TN: Extension of recombinant human RANTES by the retention of the initiating methionine produces a potent antagonist. J Biol Chem 271:2599± 2603, 1996 Rothenberg ME, Luster AD, Lilly CM, Drazen JM, Leder P: Constitutive and allergen-induced expression of eotaxin mRNA in the guinea pig lung. J Exp Med 181:1211±1216, 1995 Schulz BS, Michel G, Wagner S, et al: Increased expression of epidermal IL-8 receptor in psoriasis. Down-regulation by FK-506 in vitro. J Immunol 151:4399± 4406, 1993 Sica A, Matsushima K, Van Damme J, et al: IL-1 transcriptionally activates the neutrophil chemotactic factor/IL-8 gene in endothelial cells. Immunology 69:548±553, 1990 Tiffany HL, Alkhatib G, Combadiere C, Berger EA, Murphy PM: CC chemokine receptors 1 and 3 are differentially regulated by IL-5 during maturation of eosinophilic HL-60 cells. J Immunol 160:1385±1392, 1998 Tuschil A, Lam C, Haslberger A, Lindley I: Interleukin-8 stimulates calcium transients and promotes epidermal cell proliferation. J Invest Dermatol 99:294± 298, 1992 Uguccioni M, Mackay CR, Ochensberger B, et al: High expression of the chemokine receptor CCR3 in human blood basophils. Role in activation by eotaxin, MCP-4, and other chemokines. J Clin Invest 100:1137±1143, 1997 Wells TNC, Power CA, Lusti-Narasimhan M, et al: Selectivity and antagonism of chemokine receptors. J Leukoc Biol 59:53±60, 1996 White JR, Imburgia C, Dul E, et al: Cloning and functional characterization of a novel human CC chemokine that binds to the CCR3 receptor and activates human eosinophils. J Leukoc Biol 62:667±675, 1997 Ying S, Robinson DS, Meng Q, et al: Enhanced expression of eotaxin and CCR3 mRNA and protein in atopic asthma. Association with airway hyperresponsiveness and predominant co-localization of eotaxin mRNA to bronchial epithelial and endothelial cells. Eur J Immunol 27:3507±3516, 1997
Hannover Medical University, Department of Dermatology and Allergology, Hannover, Germany; *LeukoSite Inc., Cambridge, Massachusetts, U.S.A.
cytes than for eosinophils. In situ hybridization techniques exhibited that basal cell layers of the epidermis were stained homogeneously for CC chemokine receptor 3 mRNA with a decreasing signal to the upper epidermis showing that differentiating and proliferating keratinocytes did express mRNA speci®c for CC chemokine receptor 3. Immunohistochemical studies con®rmed low expression of CC chemokine receptor 3 protein on epidermal keratinocytes compared to the high level observed on in®ltrating eosinophils. Furthermore, stimulation of cultured keratinocytes with eotaxin resulted in an increased [3H]thymidine incorporation indicating a role of CC chemokine receptor 3 in epidermal proliferation and differentiation. These data demonstrate that CC chemokine receptor 3 is expressed not only on hematopoetic cells but also on keratinocytes as nonhematopoetic cells with ectodermal origin. Therefore, the identi®cation of CC chemokine receptor 3 on epidermal keratinocytes may indicate a role for CC chemokine receptor 3 and its ligands in skin physiology and pathophysiology. Key words: chemokine receptor/in¯ammation/keratinocytes. J Invest Dermatol 116:549±555, 2001
CC chemokine receptors are expressed on hematopoetic cells, and these may impart selective homing of monocyte, leukocyte, and lymphocyte subsets to sites of in¯ammation. CC chemokine receptor 3 is the major receptor on eosinophils and is also expressed on other in¯ammatory cells suggesting its important role for allergic diseases such as atopic dermatitis and bronchial asthma. Eotaxin, eotaxin-2 and eotaxin-3 have been identi®ed as ligands that only activate CC chemokine receptor 3. CC chemokine receptor 3 is also activated by other promiscuous ligands, however, such as RANTES and monocyte chemotactic protein 4. To date, CC chemokine receptor 3 has not been reported to be expressed on nonhematopoetic cells. In this study, we investigated whether keratinocytes possess autocrine and paracrine mechanisms for CC chemokine secretion and receptor expression as reported for the expression of interleukin 8 and its receptors. Reverse transcriptase polymerase chain reaction analysis demonstrated that CC chemokine receptor 3 mRNA is expressed constitutively in cultured keratinocytes. The signal quantities of the CC chemokine receptor 3 amplicons showed lower intensities for keratino-
C
The CC chemokines eotaxin, eotaxin-2, and eotaxin-3 have been identi®ed as ligands that exclusively activate the CCR3 and seem to be monospeci®c (Elsner et al, 1996a; 1998; Ponath et al, 1996b; Forssmann et al, 1997; White et al, 1997). Other chemokines recognize more than one chemokine receptor. CCR3 is also activated by other promiscuous ligands, however, such as RANTES, monocyte chemotactic protein 2 (MCP-2), MCP-3, and MCP-4 (Alam et al, 1993; Dahinden et al, 1994; Kapp et al, 1994; Petering et al, 1998). Recent studies have shown that eotaxin, eotaxin-2, and eotaxin3 not only are able to induce chemotaxis but also stimulate the release of reactive oxygen species and induce changes in relative Factin contents of human eosinophils as an indicator for locomotoryassociated cellular responses (Elsner et al, 1996a, 1998). Therefore, eotaxin, eotaxin-2, and eotaxin-3 represent speci®c chemoattractants for certain hematopoetic cells. Furthermore, CXC chemokines, e.g., interleukin-8 (IL-8), do not induce functional activity even on activated human eosinophils indicating that CXCR1 and CXCR2 are not expressed on the cell surface (Petering et al, 1999). These ®ndings may suggest an important role for CCR3 and its ligands in chronic in¯ammatory disorders.
hemokine receptors represent Gi-protein-coupled seven transmembrane receptors that are divided into four subgroups (C, CC, CXC, CX3C) according to their ligands (Baggiolini, 1998; Luster, 1998). Each member is a receptor for a unique set of one or more CC chemokine agonists. Various CC chemokine receptors are expressed on hematopoetic cells, and these may impart selective homing of monocytes, leukocytes, and lymphocytes to sites of in¯ammation. CCR3 is the major receptor on eosinophils (Kitaura et al, 1996; Ponath et al, 1996a) and is also expressed on basophils (Heath et al, 1997; Uguccioni et al, 1997) suggesting an important role for allergic diseases such as atopic dermatitis and bronchial asthma. Furthermore, CCR3 expression was observed on macrophages (Hariharan et al, 1999) and neural microglia cells (Albright et al, 1999). Manuscript received July 7, 1999; revised December 22, 2000; accepted for publication December 29, 2000 Reprint requests to: Prof. JoÈrn Elsner M.D., Hannover Medical University, Department of Dermatology and Allergology, Ricklinger Str. 5, D-30449 Hannover, Germany. Email: [email protected] 0022-202X/01/$15.00
´ Copyright # 2001 by The Society for Investigative Dermatology, Inc. 549
550
PETERING ET AL
The ®nding that CCR3 is the principal receptor for eosinophil responses to CC chemokines has led to the development of monoclonal antibody (MoAb) 7B11 blocking CCR3 ligand binding (Daugherty et al, 1996a; Heath et al, 1997; Elsner et al, 1998; Kitayama et al, 1998). In addition, CC chemokine receptor directed antagonists such as Met-RANTES could be identi®ed. Met-RANTES is principally described as a CCR1 antagonist, but has also weak inhibitory properties on CCR3 (Proudfoot et al, 1996; Wells et al, 1996; Elsner et al, 1997). To date, CCR3 has not been reported to be expressed on nonhematopoetic cells, whereas CXC chemokine receptors exhibit a broad distribution not only on hematopoetic cells but also on endothelial cells, ®broblasts, and melanocytes (Holmes et al, 1991; Murphy and Tiffany, 1991). Primary cultured human keratinocytes are known to express mRNA speci®c for IL-8 as well as for the CC chemokine eotaxin (Hein et al, 1997). Human keratinocytes are known to synthesize and secret CXC chemokines, e.g., IL-8, and can be stimulated by their own cytokine products. To investigate whether human keratinocytes possess similar autocrine and paracrine mechanisms for CC chemokine secretion and CC chemokine receptor expression, reverse transcriptase polymerase chain reaction (RT-PCR) analysis, in situ hybridization, and receptor protein expression analysis were performed with cultured human keratinocytes and skin biopsies. Tissue sections were obtained from healthy individuals, patients with chronic in¯ammatory diseases, and patients suffering from bullous pemphigoid. This subepidermal blistering disease is characterized by a TH2-type cytokine pattern as well as peripheral and tissue eosinophilia. Therefore, in®ltrating eosinophils could be stained for CCR3 as a positive control. In this study, we report that human keratinocytes contain mRNA speci®c for the eotaxin receptor CCR3 and express CCR3 protein. Functional data revealed that CCR3 may play a role in epidermal proliferation and differentiation. MATERIALS AND METHODS Cell preparation Human keratinocyte culture Normal human keratinocytes were obtained from fresh trunk biopsies. Epidermal cells were disaggregated by trypsinization (4°C, 12 h) in Hank's balanced salt solution without Ca2+ and Mg2+ containing 0.25% trypsin. The cells were plated out at 107 cells per 50 ml culture ¯ask in endotoxin-free keratinocyte growth medium (KGM) containing 1.0 mM Ca2+ and incubated with 10% CO2 (37°C). Normal keratinocytes were derived from the second to fourth passages grown as a monolayer to 80% con¯uence. Isolation of human eosinophils Human eosinophils were isolated from heparin-anticoagulated venous blood from normal nonatopic healthy donors. The isolation was performed using Ficoll (Pharmacia) density gradient centrifugation and eosinophils were puri®ed by negative selection with anti-CD16 MoAb (clone 3G8, Immunotech, Hamburg, Germany) coated Dynabeads M-450 (Dynal, Hamburg) as described previously (Elsner et al, 1994). The resulting eosinophil purity was > 99.5% as determined by ¯ow cytometric analysis (FACScan, Becton Dickinson, Heidelberg, Germany) using phycoerythrinconjugated anti-CD16 MoAb (clone 3G8, Immunotech). Chemokines, cytokines, and monoclonal antibodies Eotaxin was purchased from Peprotech (London, U.K.). Bradykinin, epidermal growth factor (EGF), and C5a were obtained from Sigma (Deisenhofen, Germany). The anti-CCR3 MoAb 7B11 was derived from C57BL6 mice immunized with CCR3 transfectants and prepared as described previously (Daugherty et al, 1996; Heath et al, 1997). MoAb EG1 against eosinophilic cationic protein (ECP) was received from Pharmacia (Uppsala). The human IgG1 isotype control was obtained from Sigma. CCR3 mRNA expression Total RNA was isolated from human keratinocytes and eosinophils using the TRIzol kit (Gibco Life Technologies, Karlsruhe, Germany) according to the manufacturer's instructions based on the guanidine isothiocyanate method. First-strand cDNA synthesis and PCR reaction were performed as described previously (Elsner et al, 1996b; Petering et al, 1996, 1999). Primers for the ampli®cation of CCR3 (sense, 5¢-GCCATTTCGGACCTGCTCTT-3¢; antisense, 5¢-TCCGCTCACAGTCATTTCCA-3¢) and b-actin
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
(sense, 5¢-GAGCGGGAAATCGTGCGTGACATT-3¢; antisense, 5¢GAAGGTAGTTTCGTGGATGCC-3¢) were designed according to the published sequences (Accession No.: CCR3, AF026535; b-actin, AB004047). PCR samples were run on a 1.8% agarose gel stained with 0.2 mg per ml ethidium bromide, and the PCR products were visualized with ultraviolet light and photographed. Automated sequencing of PCR products For automated sequencing of PCR products, an Applied Biosystems Model 373 A (Forster City, CA) was used. Nucleotide sequence data were analyzed with Lasergene sequence analysis software (DNASTAR, Madison, WI). CCR3 riboprobes and in situ hybridization RNA probes for CCR3 were designed from a plasmid constructed by cloning a 578 bp RTPCR fragment ampli®ed from human eosinophils into a polylinker site of transcription vector pBluescript II SK± containing T3 and T7 polymerase recognition elements as described previously (Petering et al, 1996). After blunt-end ligation, the orientation and DNA sequence of both insert strains were determined by cycle sequencing as described above. Antisense (T3) and sense (T7) riboprobes were transcribed enzymatically and digoxygenin-labeled according to the manufacturer's instructions (Boehringer Mannheim, Germany). Probes were washed twice by ethanol precipitation to remove unincorporated nucleotides and stored in 10 nM dithiothreitol at ±80°C. For in situ hybridization, tissue sections were ®xed with 4% paraformaldehyde and washed with 1% phosphate-buffered saline (PBS). Tissues were permeabilized with 50 mg per ml protease K (37°C, 30 min), washed once with 1% PBS, and incubated for 5 min with 2 3 standard saline citrate (SSC). The sections were dehydrated through graded alcohol and air-dried. After prehybridization (55°C, 60 min) and hybridization (55°C, 12 h) with standard hybridization solution (10% dextran sulfate, 40% formamide, 4 3 SSC, 1 3 Denhard's, 10 mM dithiothreitol, 100 mg per ml tRNA) slides were washed with SSC under stringent conditions. Staining for CCR3 was carried out with antidigoxygenin F(ab)¢ fragment conjugated to alkaline phosphatase at a ®nal dilution of 1:1000 according to the manufacturer's instructions (Boehringer Mannheim). Slides stained with sense and antisense riboprobes were read blindly by two dermatopathologists. Immunohistochemistry with human keratinocytes and eosinophils Tissue sections were cut to 5 mm and adhered to silanized slides. Immunostaining was performed (room temperature, 60 min) and was developed by the avidin±biotin-peroxidase complex technique with the chromogen fuchsin according to the manufacturer's recommendations (DAKO). Staining for CCR3 protein was performed with the anti-CCR3 MoAb 7b11 (750 mg per ml) at a ®nal dilution of 1:100 in 1% bovine serum albumin/PBS. Staining for eosinophil matrix granular proteins was performed with MoAb EG1 speci®c for ECP (Kiehl and Kapp, 1998). Non-reactive mouse antihuman antibodies of IgG1 isotype were used as negative controls. Immuno¯uorescence of cultured keratinocytes Immuno¯uorescence of cultured human trunk keratinocytes between the second and fourth passages was performed using standard techniques. In brief, keratinocytes were adjusted to a density of 1 3 107 cells per ml. Aliquots (20 ml) containing 2 3 105 cells were incubated at 4°C for 30 min with the indicated antibody or isotype control (IgG2). Thereafter, cells were washed twice with 4°C PBS. For indirect immuno¯uorescence, cells were stained in a second step by a ¯uorescein isothiocyanate (FITC) conjugated mouse antihuman antibody (Immunotech) and subsequently washed twice. Thereafter, cells were analyzed by ¯ow cytometry (FACScan). The samples were excited at 488 nm and emission was measured at 530 nm (FITC-labeled antibodies, ¯uorescence 1, green ¯uorescence). The CCR3 receptor expression was analyzed after preincubation of cultured keratinocytes for 30 min at 37°C with the CCR3 ligand eotaxin or medium alone. The mean channel ¯uorescence of seven experiments was determined. Results were expressed as the median ¯uorescence intensity after preincubation for 30 min versus the median ¯uorescence intensity after incubation with medium. Incubation of cells for longer than 30 min did not change the results. Keratinocyte proliferation assay Keratinocytes were plated at 5 3 103 cells per well of a 96-well microtiter plate and cultured in KGM for 5 d. The medium in each well was then replaced with KGM supplemented with either EGF, IL-8, or eotaxin at the indicated concentrations or KGM alone. Proliferation was measured by labeling six replicate cultures with 0.5 mCi ml [3H]thymidine (Amersham, Braunschweig, Germany) during a further 48 h of incubation. After labeling, medium was removed, each well was washed with PBS, and trichloroacetic acid (TCA) was added. The TCA-precipitated nucleic
VOL. 116, NO. 4 APRIL 2001
CCR3 EXPRESSION ON HUMAN KERATINOCYTES
Figure 1. CCR3 mRNA is expressed in cultured human keratinocytes. mRNA from cultured human keratinocytes and puri®ed eosinophils was isolated and ®rst-strand cDNA synthesis was performed. PCR was carried out with primer pairs speci®c for CCR3 (expected size 578 bp) and b-actin (expected size 221 bp) as an internal standard. The amplicons were separated by electrophoresis in 1.8% agarose gel and stained by ethidium bromide. Lane 1, 50 bp DNA size marker; lane 2, human eosinophil mRNA; lane 3, human keratinocyte mRNA. One representative experiment out of four is shown.
acid was solubilized in 100 ml of 0.1 M sodium hydroxide, 1% sodium dodecyl sulfate, and 2% sodium carbonate for 30 min and the b decay was counted in a liquid scintillation counter. The stimulation index was de®ned by the ratio of mean counts per minute of stimulated to unstimulated cultures.
RESULTS Cultured human keratinocytes express mRNA speci®c for CCR3 RT-PCR analysis was performed to investigate whether human keratinocytes express mRNA speci®c for CCR3. For this purpose, human trunk keratinocytes were cultured and passaged two to four times prior to total RNA isolation. Using the primer pairs indicated, DNA fragments with an expected size of 578 bp speci®c for human CCR3 could be detected in cultured human keratinocytes (Fig 1, lane 3). To clarify whether the ampli®ed PCR products were speci®c for CCR3 and did not represent other chemokine receptors, DNA sequence analysis was performed. PCR products were sequenced automatically with vector-derived primer pairs. Computer-aided nucleotide sequence analysis revealed that all PCR products represent DNA sequences speci®c for CCR3 mRNA. As expected, highly puri®ed human eosinophils did constitutively express CCR3 mRNA (Fig 1, lane 2). To demonstrate that all mRNA preparations were highly puri®ed, RT-PCR was performed simultaneously using primer pairs for bactin as standard (221 bp) (Fig 1). The signal quantities of the CCR3 amplicons showed lower signal intensities for cultured keratinocytes than for isolated human eosinophils. The results indicate that CCR3 is expressed at mRNA level in cultured human keratinocytes. In situ hybridization studies reveal that CCR3 mRNA is expressed in epidermal keratinocytes To investigate whether CCR3 mRNA is not only expressed in vitro in cultured keratinocytes but can also be detected in vivo in epidermal keratinocytes, in situ hybridization studies were performed. Tissue sections from ®ve patients suffering from bullous pemphigoid prior to immunosuppressive therapy were studied for the expression of CCR3 mRNA. Bullous pemphigoid represents a chronic in¯ammatory autoimmune disease characterized by a local accumulation of eosinophils with subsequent subepidermal blister formation (Lever, 1979). Therefore, the colocalization of keratinocytes and eosinophils in®ltrating the upper dermis and epidermis allowed us to compare CCR3 signal speci®city and intensity between different cells. Specimens were obtained from perilesional and lesional skin and CCR3 mRNA expression was analyzed by in situ hybridization. Strong CCR3 mRNA expression was detected in in¯ammatory cells in®ltrating the upper dermis
551
(Fig 2A). Immunohistochemical staining with MoAb EG1 recognizing matrix granular proteins identi®ed these cells as eosinophil granulocytes known to express CCR3 (Fig 2C). Like eosinophils, basal cell layers of the epidermis were stained homogeneously for CCR3 mRNA with a decreasing signal to the upper epidermis showing that differentiating and proliferating keratinocytes did express mRNA speci®c for CCR3 (Fig 2A). Negative controls hybridized with sense RNA probes only showed background signals (data not shown). To investigate whether the expression of CCR3 mRNA in keratinocytes is induced in the state of disease, specimens from healthy volunteers were analyzed by in situ hybridization. As seen in Fig 2(E), basal keratinocytes from normal individuals also showed staining for CCR3 mRNA. A similar staining pattern could be observed in tissue sections from chronic in¯ammatory diseases. Figure 2(G) shows the expression of CCR3 mRNA in atopic dermatitis, a T-cell-mediated disease affecting the upper dermis and epidermis. Figure 2(H) demonstrates the expression of CCR3 mRNA in psoriasis vulgaris, a chronic in¯ammatory disease characterized by epidermal hyperplasia and altered epidermal differentiation. These data reveal that CCR3 mRNA is expressed constitutively in human keratinocytes. The expression may not depend on the state of activation of basal keratinocytes, e.g., in disease affecting the basement membrane zone. Immunohistochemical studies demonstrate that CCR3 is expressed on epidermal keratinocytes To demonstrate that the transcription of CCR3 mRNA is followed by the expression of the receptor protein on the cell surface, bullous pemphigoid tissue sections adjacent to those used for in situ hybridization were stained immunohistochemically for CCR3. The colocalization of keratinocytes and eosinophils allowed us again to compare the signal quantities. Expression of CCR3 protein was observed on epidermal keratinocytes with a marginally increased staining of the basal cell layers (Fig 2B). Immunohistochemical staining of specimens taken from healthy volunteers and patients suffering from atopic dermatitis and psoriasis vulgaris exhibited a similar staining pattern (data not shown). These results indicate that proliferating and differentiating keratinocytes of the epidermis are able to express CCR3 protein. In comparison to human eosinophils in®ltrating the upper dermis and epidermis, however, keratinocytes showed only a weak CCR3 expression in terms of staining intensity indicating fewer binding sites per cell (Fig 2B, D). Positive CCR3 staining of eosinophils without immunohistochemical reaction on other in®ltrating in¯ammatory cells, such as lymphocyte subpopulations, demonstrated that anti-CCR3 MoAb 7B11 is also a speci®c marker for CCR3 in immunohistochemical studies (Fig 2D). Flow cytometry analysis reveals that CCR3 is expressed heterogeneously on cultured keratinocytes To con®rm the immunohistochemical data, the expression of CCR3 on cultured human trunk keratinocytes was investigated by ¯ow cytometry. Cells between the second and fourth passage were stained with the anti-CCR3 MoAb 7B11 (Fig 3A). Surprisingly, the expression of CCR3 on cultured keratinocytes was donor dependent and differed between the healthy volunteers. Cultured keratinocytes from two donors were negative or showed only a weak staining for CCR3 in a small subpopulation of keratinocytes (Fig 3B). Preincubation of cultured keratinocytes for 30 min with TH2-type cytokines (IL-4, IL-5) did not alter CCR3 expression (data not shown). To demonstrate whether the antiCCR3 MoAb 7B11 binds speci®cally to CCR3, cultured human keratinocytes were incubated for 30 min with eotaxin at a concentration of 125.0 nM or medium. As expected, preincubation of cultured keratinocytes with eotaxin induced downregulation of CCR3 on the cell surface, indicating that the CCR3 ligand eotaxin was able to bind speci®cally to its receptor expressed on keratinocytes (Fig 3A).
552
PETERING ET AL
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
Figure 2. In situ hybridization and immunohistochemical analysis for CCR3 mRNA and protein expression on human keratinocytes. (A) In tissue sections from patients with bullous pemphigoid CCR3 mRNA expression was analyzed by in situ hybridization. Basal cell layers of the epidermis were homogeneously stained for CCR3 mRNA with a decreasing intensity to the upper epidermis. Strong CCR3 mRNA expression was detected in in¯ammatory cells in®ltrating the upper dermis. (B) Immunohistochemistry with anti-CCR3 MoAb 7B11 revealed that CCR3 protein is expressed on epidermal keratinocytes. (C) Immunohistochemical staining with MoAb EG1 recognizing ECP identi®ed in¯ammatory cells as human eosinophils. (D) Human eosinophils in®ltrating the upper dermis showed more intense immunohistochemical staining for CCR3 than epidermal keratinocytes indicating more binding sites per cell. (E) In situ hybridization revealed that basal keratinocytes from normal individuals also showed staining for CCR3 mRNA. A similar staining pattern could be observed in chronic in¯ammatory diseases: (F) expression of CCR3 mRNA in bullous pemphigoid; (G) expression of CCR3 mRNA in atopic dermatitis; (H) expression of CCR3 mRNA in psoriasis vulgaris. Scale bars: (A) 100 mm; (B±D) 50 mm.
Proliferation of cultured human keratinocytes after stimulation with eotaxin can be inhibited by antiCCR3 Previous studies have shown that eotaxin is expressed in tissue specimens obtained from chronic in¯ammatory lesions (Rothenberg et al, 1995). These ®ndings prompted us to investigate whether eotaxin may have any functional effects on epidermal keratinocytes, e.g., cell proliferation and differentiation. Cultured human epidermal keratinocytes were stimulated with eotaxin (10±1000 nM) and the incorporation of [3H]thymidine was detected as a marker for cell proliferation. As seen in Fig 4, eotaxin
induced dose-dependently a proliferation of human keratinocytes. In this experiment, eotaxin was as effective as IL-8 with a maximum of cell proliferation at a concentration of 100 nM (Fig 4). In addition, proliferation of cultured human keratinocytes was also induced by other CCR3 ligands such as RANTES and MCP-4 representing two other potent activators of CCR3. At equivalent concentrations eotaxin seemed to be more effective than RANTES or MCP-4, however (Fig 4). EGF used as a positive control was more potent than eotaxin at concentrations above 1 mM. Moreover, whereas EGF induced a dose-related curve all
VOL. 116, NO. 4 APRIL 2001
CCR3 EXPRESSION ON HUMAN KERATINOCYTES
553
Figure 4. Proliferation of cultured human keratinocytes after stimulation with eotaxin can be inhibited by anti-CCR3 MoAb 7B11. Cultured human keratinocytes were stimulated with CC chemokines eotaxin, MCP-4, RANTES, and IL-8 as well as EGF in different concentrations (10±1000 nM). The incorporation of [3H]thymidine was detected as a marker for cell proliferation. Data are expressed as proliferation index (stimuli-induced [3H]thymidine incorporation/medium-induced [3H]thymidine incorporation). In one experiment keratinocytes were preincubated with the anti-CCR3 MoAb 7B11 and subsequently stimulated with eotaxin (10±7 M). One representative experiment out of four is shown indicating that eotaxin induces proliferation of human keratinocytes.
that the proliferation of keratinocytes after stimulation with eotaxin can be inhibited by antagonizing CCR3-mediated effects. DISCUSSION
Figure 3. Flow cytometry analysis reveals that CCR3 is expressed heterogeneously on cultured keratinocytes. Overlay histogram for the expression of CCR3 on cultured human trunk keratinocytes. Cells between the second and fourth passage were stained for CCR3 with the anti-CCR3 MoAb 7B11. (A) The bold line represents cultured keratinocytes expressing CCR3. The dotted line indicates cultured keratinocytes preincubated with 125.0 nM eotaxin for 30 min showing a downregulation of the CCR3. The thin line represents the IgG2a isotype control. These data demonstrate that eotaxin was able to bind speci®cally to CCR3 expressed on human keratinocytes. (B) The expression of CCR3 was donor dependent and differed between healthy volunteers. In two donors cultured keratinocytes appeared negative or showed only a weak staining for CCR3 in a small subpopulation of keratinocytes as demonstrated by the bold line.
chemokines tested induced a bell-shaped curve with respect to incorporation of [3H]thymidine in epidermal keratinocytes. To investigate whether the eotaxin-induced proliferation of keratinocytes was mediated via CCR3, cultured keratinocytes were preincubated with anti-CCR3 MoAb 7B11. As demonstrated previously, this antibody is able to block eotaxin binding to CCR3 completely and inhibits CCR3-speci®c cell activation as demonstrated previously (Daugherty et al, 1996a; Heath et al, 1997; Elsner et al, 1998). As seen in Fig 4 preincubation of keratinocytes with the anti-CCR3 MoAb 7B11 inhibited the proliferation of keratinocytes induced by eotaxin. This experiment demonstrates
Chemokines are proin¯ammatory cytokines known to be involved in chronic in¯ammatory diseases such as atopic dermatitis and allergic asthma (Baggiolini, 1998; Luster, 1998). Recent studies have shown that the CC chemokine eotaxin is elevated in the bronchial mucosa of asthmatic patients and is expressed by epithelial and endothelial cells (Ying et al, 1997). CCR3 was identi®ed independently by two different groups as the eotaxin receptor on human eosinophils (Daugherty et al, 1996; Ponath et al, 1996a) and further studies showed the expression of CCR3 also on basophils suggesting that this chemokine receptor may play a central role in chronic in¯ammatory disorders (Kitaura et al, 1996; Ponath et al, 1996b; Uguccioni et al, 1997). So far, CCR3 has only been identi®ed on hematopoetic cells whereas CXC chemokine receptors are also expressed on nonhematopoetic cells such as ®broblasts, melanocytes, and keratinocytes (Holmes et al, 1991; Murphy et al, 1991; Norgauer et al, 1996). In psoriatic skin, characterized by epidermal hyperplasia and altered epidermal differentiation, IL-8 receptors were found to be expressed at high levels (Schulz et al, 1993; Kulke et al, 1998). In vitro studies con®rmed that keratinocyte migration and growth can be stimulated by its ligand IL-8 (Tuschil et al, 1992). As endothelial and epithelial cells of the lung and also epidermal keratinocytes are known to produce CXC chemokines such as IL-8 (Sica et al, 1990; Gillitzer et al, 1991; Huber et al, 1991), autocrine and paracrine mechanisms seem to be responsible for proliferation and differentiation of nonhematopoetic cells. Recently, increased levels of CC chemokine eotaxin have been found in respiratory epithelium and endothelium as well as in the intestinal mucosa (Garcia-Zepeda et al, 1996; Matthews et al, 1998). Hein et al detected eotaxin mRNA at low levels in stimulated primary keratinocytes (Hein et al, 1997). Therefore, it is tempting to speculate that human keratinocytes also possess autocrine and paracrine mechanisms for CC chemokine secretion and CC
554
PETERING ET AL
chemokine receptor expression similar to their CXC counterpart. This study describes the functional analysis and expression of the predominant eosinophil CC chemokine receptor CCR3 on human keratinocytes in vitro in comparison to isolated eosinophils and in vivo to eosinophils in®ltrating the upper dermis and epidermis. In the ®rst set of experiments, RT-PCR analysis was performed demonstrating that cultured human keratinocytes constitutively express CCR3 mRNA. Signal quantities of CCR3 amplicons showed lower signal intensities for keratinocytes than for isolated eosinophils. To rule out that the RT-PCR products were due to a false-positive ampli®cation of other chemokine receptors, e.g., CXCR1 or CXCR2, nucleotide sequence analysis of the amplicons was performed and revealed a DNA sequence homolog to published CCR3 mRNA sequences. To con®rm that CCR3 mRNA expression was also observed in vivo, skin biopsies were analyzed for CCR3 mRNA by in situ hybridization techniques. Again, CCR3 mRNA could be detected and was predominantly expressed in basal keratinocytes whereas in the upper cell layers the receptor expression was decreased. To investigate whether the expression of CCR3 mRNA is induced in the state of disease, skin biopsies from healthy volunteers and patients suffering from different in¯ammatory diseases (atopic dermatitis, psoriasis vulgaris, and bullous pemphigoid) were analyzed. Our data revealed that CCR3 mRNA is constitutively expressed on epidermal keratinocytes and may not depend on the state of cell activation. The ®nding that CCR3 mRNA is constitutively expressed on keratinocytes obtained from different in¯ammatory diseases affecting the epidermis (psoriasis vulgaris), basal membrane (bullous pemphigoid), and epidermis as well as the upper dermis (atopic dermatitis) suggests that the low level of CCR3 mRNA expression may not have a pathophysiologic role in these disorders. Our results indicate that in particular proliferating and differentiating keratinocytes of the basal stratum express CCR3 mRNA. This study provides evidence that CCR3 mRNA is not only expressed on in¯ammatory cells derived from hematopoetic precursors but can also be expressed on human keratinocytes as nonhematopoetic cells with ectodermal origin. The expression of CCR3 protein on keratinocytes was investigated by immunohistochemical staining for CCR3 with tissue sections adjacent to those used for in situ hybridization. The experiments indicated that CCR3 protein is expressed in human epidermis. Immunohistochemistry was performed with tissue sections from patients with bullous pemphigoid, a blistering autoimmune disease characterized by an excessive synthesis of proin¯ammatory cytokines. TH2-type cytokines such as IL-4 and IL-5 but also CC chemokines, e.g., RANTES, have been identi®ed in the skin and the blister ¯uid of these patients and contribute to the epidermal and dermal in®ltration of eosinophils (D'Auria et al, 1998; Inaoki and Takehara, 1998). IL-4 and IL-5 may upregulate the expression of CCR3 on in¯ammatory cells and may also increase CCR3 protein synthesis in epidermal keratinocytes (Tiffany et al, 1998; Jinquan et al, 1999). Therefore, it was obvious to speculate that cytokine priming of keratinocytes in vivo may contribute to the expression of CCR3 protein. In our in vitro experiments, however, no difference in CCR3 expression could be observed after preincubation of cultured keratinocytes with TH2type cytokines. The colocalization of keratinocytes and eosinophils in®ltrating the upper dermis and epidermis in skin biopsies from bullous pemphigoid allowed us to compare CCR3 signal speci®city and intensity between different cells indicating that human keratinocytes possess fewer binding sites per cell than human eosinophils. To con®rm these immunohistochemical ®ndings ¯ow cytometry analysis was performed demonstrating an inhomogeneous expression of CCR3 on cultured keratinocytes, whereas CCR3 mRNA was expressed constitutively indicating a receptor regulation at the post-transcriptional level. The CCR3 expression on the cell surface was donor dependent and differed between our healthy volunteers. Moreover, we could clearly demonstrate that preincubation of CCR3+ keratinocytes with eotaxin resulted in a downregulation of
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
CCR3, which has recently been shown in eosinophils (Elsner et al, 2000). Preincubation of keratinocytes with TH2-type cytokines did not alter CCR3 receptor expression. Recent studies have shown that the CXC chemokine IL-8 exerts a direct in¯uence on several keratinocyte functions including epidermal proliferation and differentiation (Kapp, 1993). Furthermore, in psoriatic skin, IL-8 and the growth-related oncogen a, another ligand for CXCR2, are produced by lesional keratinocytes especially from the upper epidermal cell layers (Larsen et al, 1989) as part of an autocrine and paracrine cytokine network. The CC chemokine eotaxin can also be synthesized and released by keratinocytes and therefore not only may contribute to the local accumulation of eosinophils and TH2-type lymphocytes but also may modulate the differentiation and proliferation of keratinocytes similar to IL-8 (Hein et al, 1997). To investigate whether such autocrine and paracrine mechanisms exist for CC chemokines, growth-promoting activities of eotaxin on cultured keratinocytes were analyzed. Our data suggest that eotaxin represents a weak stimulus for proliferation of human keratinocytes. The observation that keratinocyte proliferation and differentiation was also induced by RANTES and MCP-4, two other potent activators of CCR3, suggests that the effect of eotaxin may be due to CCR3 binding and signaling. Interestingly, all of the CCR3 ligands tested induced a bell-shaped curve with regard to keratinocyte proliferation. As we could clearly demonstrate that eotaxin was able to induce a downregulation of CCR3 in CCR3+ keratinocytes it could be speculated that CCR3 ligands downregulate CCR3 at a concentration of 10±7 M. Thus, the reduced proliferation index of keratinocytes in response to a higher concentration (10±6 M) of CCR3 ligands could be explained by a downregulation of CCR3. To further prove this hypothesis, cultured keratinocytes were preincubated with anti-CCR3 MoAb 7B11 known to speci®cally block the activation of human eosinophils by CCR3 ligands (Heath et al, 1997; Elsner et al, 1998; Kitayama et al, 1998). The antiCCR3 MoAb 7B11 was able to inhibit eotaxin-induced proliferation of keratinocytes indicating that binding to CCR3 is the initial step of the observed keratinocyte differentiation and proliferation. In summary, these data provide evidence that CCR3 is expressed not only on hematopoetic cells but also on human keratinocytes as nonhematopoetic cells with ectodermal origin. The level of receptor expression is donor dependent. In addition to its involvement in chemotaxis and activation of the respiratory burst of human eosinophils eotaxin and other CCR3 ligands may contribute to local wound repair and healing mechanisms integrating different ways of tissue reconstitution. Therefore, the identi®cation of CCR3 on epidermal keratinocytes suggests that CCR3 and its ligands, especially eotaxin, may play an important role in skin physiology and pathophysiology. This work was supported by a grant from the Deutsche Forschungsgemeinschaft (EL 160/6±1) and by the Hochschulinterne LeistungsfoÈrderung of the Medical University Hannover.
REFERENCES Alam R, Stafford S, Forythe P, Harrison R, Faubion D, Lett-Brown MA, Grant JA: RANTES is a chemotactic and activating factor for human eosinophils. J Immunol 150:3442±3447, 1993 Albright AV, Shieh JT, Itoh T, et al: Microglia express CCR5, CXCR4, and CCR3, but of these, CCR5 is the principal coreceptor for human immunode®ciency virus type 1 dementia isolates. J Virol 73:205±213, 1999 Baggiolini M: Chemokines and leukocyte traf®c. Nature 392:565±568, 1998 Dahinden CA, Geiser T, Brunner T, et al: Monocyte chemotactic protein 3 is a most effective basophil- and eosinophil-activating chemokine. J Exp Med 179:751± 756, 1994 Daugherty BL, Siciliano SJ, DeMartino JA, Malkowitz L, Sirotina A, Springer MS: Cloning, expression, and characterization of the human eosinophil eotaxin receptor. J Exp Med 183:2349±2354, 1996a D'Auria L, Pietravalle M, Mastroianni A, et al: IL-5 levels in the serum and blister ¯uid of patients with bullous pemphigoid: correlations with eosinophil cationic
VOL. 116, NO. 4 APRIL 2001
protein, RANTES, IgE and disease severity. Arch Dermatol Res 290:25±27, 1998 Elsner J, Oppermann M, Czech W, Dobos G, SchoÈpf E, Norgauer J, Kapp A: C3a activates reactive oxygen radical species production and intracellular calcium transients in human eosinophils. Eur J Immunol 24:518±522, 1994 Elsner J, HoÈchstetter R, Kimmig D, Kapp A: Human eotaxin represents a potent activator of the respiratory burst of human eosinophils. Eur J Immunol 26:1919± 1925, 1996a Elsner J, HoÈchstetter R, Spiekermann K, Kapp A: Surface and mRNA expression of the CD52 antigen by human eosinophils but not by neutrophils. Blood 88:4684±4693, 1996b Elsner J, Petering H, HoÈchstetter R, Kimmig D, Wells TN, Kapp A, Proudfoot AE: The CC chemokine antagonist Met-RANTES inhibits eosinophil effector functions through the chemokine receptors CCR1 and CCR3. Eur J Immunol 27:2892±2898, 1997 Elsner J, Petering H, Kluthe C, Kimmig D, Smolarski R, Ponath P, Kapp A: Eotaxin-2 activates chemotaxis-related events and release of reactive oxygen species via pertussis toxin-sensitive G proteins in human eosinophils. Eur J Immunol 28:2152±2158, 1998 Elsner J, Mack M, Bruhl H, et al: Differential activation of CC chemokine receptors by AOP-RANTES. J Biol Chem 275:7787±7794, 2000 Forssmann U, Uguccioni M, Loetscher P, Dahinden CA, Langen H, Thelen M, Baggiolini M: Eotaxin-2, a novel CC chemokine that is selective for the chemokine receptor CCR3, and acts like eotaxin on human eosinophil and basophil leukocytes [In Process Citation]. J Exp Med 185:2171±2176, 1997 Garcia-Zepeda EA, Rothenberg ME, Ownbey RT, Celestin J, Leder P, Luster AD: Human eotaxin is a speci®c chemoattractant for eosinophil cells and provides a new mechanism to explain tissue eosinophilia. Nat Med 2:449±456, 1996 Gillitzer R, Berger R, Mielke V, Muller C, Wolff K, Stingl G: Upper keratinocytes of psoriatic skin lesions express high levels of NAP-1/IL-8 mRNA in situ. J Invest Dermatol 97:73±79, 1991 Hariharan D, Douglas SD, Lee B, Lai JP, Campbell DE, Ho WZ: Interferon-gamma upregulates CCR5 expression in cord and adult blood mononuclear phagocytes. Blood 93:1137±1144, 1999 Heath H, Qin S, Rao P, et al: Chemokine receptor usage by human eosinophils. The importance of CCR3 demonstrated using an antagonistic monoclonal antibody. J Clin Invest 99:178±184, 1997 Hein H, Schluter C, Kulke R, Christophers E, Schroder JM, Bartels J: Genomic organization, sequence, and transcriptional regulation of the human eotaxin gene. Biochem Biophys Res Commun 237:537±542, 1997 Holmes WE, Lee J, Kuang WJ, Rice GC, Wood WI: Structure and functional expression of a human interleukin-8 receptor. Science 253:1278±1280, 1991 Huber AR, Kunkel SL, Todd RF3, Weiss SJ: Regulation of transendothelial neutrophil migration by endogenous interleukin-8 [published errata appear in Science 1991 November 1, 254:631 and 1991 December 6, 254:1435]. Science 254:99±102, 1991 Inaoki M, Takehara K: Increased serum levels of interleukin (IL) -5, IL-6 and IL-8 in bullous pemphigoid. J Dermatol Sci 16:152±157, 1998 Jinquan T, Quan S, Feili G, Larsen CG, Thestrup-Pedersen K: Eotaxin activates T cells to chemotaxis and adhesion only if induced to express CCR3 by IL-2 together with IL-4. J Immunol 162:4285±4292, 1999 Kapp A: The role of cytokines in the psoriatic in¯ammation. J Dermatol Sci 5:133± 142, 1993 Kapp A, Zeck Kapp G, Czech W, SchoÈpf E: The chemokine RANTES is more than a chemoattractant: characterization of its effect on human eosinophil oxidative metabolism and morphology in comparison with IL-5 and GM-CSF. J Invest Dermatol 102:906±914, 1994 Kiehl P, Kapp A: Tissue eosinophilia and local deposition of eosinophil-speci®c granule proteins. Regulation and signi®cance for in¯ammatory response in atopic dermatitis and other in¯ammatory dermatoses. Hautarzt 49:176±183, 1998 Kitaura M, Nakajima T, Imai T, et al: Molecular cloning of human eotaxin, an eosinophil-selective CC chemokine, and identi®cation of a speci®c eosinophil eotaxin receptor, CC chemokine receptor 3. J Biol Chem 271:7725±7730, 1996 Kitayama J, Mackay CR, Ponath PD, Springer TA: The C-C chemokine receptor
CCR3 EXPRESSION ON HUMAN KERATINOCYTES
555
CCR3 participates in stimulation of eosinophil arrest on in¯ammatory endothelium in shear ¯ow. J Clin Invest 101:2017±2024, 1998 Kulke R, Bornscheuer E, SchluÈter C, Bartels J, Rowert J, Sticherling M, Christophers E: The CXC receptor 2 is overexpressed in psoriatic epidermis. J Invest Dermatol 110:90±94, 1998 Larsen CG, Anderson AO, Oppenheim JJ, Matsushima K: Production of interleukin8 by human dermal ®broblasts and keratinocytes in response to interleukin-1 or tumour necrosis factor. Immunology 68:31±36, 1989 Lever WF: Pemphigus and pemphigoid. A review of the advances made since 1964. J Am Acad Dermatol 1:2±31, 1979 Luster AD: Chemokines ± chemotactic cytokines that mediate in¯ammation. N Engl J Med 338:436±445, 1998 Matthews AN, Friend DS, Zimmermann N, et al: Eotaxin is required for the baseline level of tissue eosinophils. Proc Natl Acad Sci USA 95:6273±6278, 1998 Murphy PM, Tiffany HL: Cloning of complementary DNA encoding a functional human interleukin-8 receptor. Science 253:1280±1283, 1991 Norgauer J, Metzner B, Schraufstatter I: Expression and growth-promoting function of the IL-8 receptor beta in human melanoma cells. J Immunol 156:1132±1137, 1996 Petering H, Hammerschmidt S, Frosch M, van Putten JP, Ison CA, Robertson BD: Genes associated with meningococcal capsule complex are also found in Neisseria gonorrhoeae. J Bacteriol 178:3342±3345, 1996 Petering H, HoÈchstetter R, Kimmig D, Smolarski R, Kapp A, Elsner J: Detection of MCP-4 in dermal ®broblasts and its activation of the respiratory burst in human eosinophils. J Immunol 160:555±558, 1998 Petering H, GoÈtze O, Kimmig D, Smolarski R, Kapp A, Elsner J: The biologic role of interleukin-8: functional analysis and expression of CXCR1 and CXCR2 on human eosinophils. Blood 93:694±702, 1999 Ponath PD, Qin S, Post TW, et al: Molecular cloning and characterization of a human eotaxin receptor expressed selectively on eosinophils. J Exp Med 183:2437±2448, 1996a Ponath PD, Qin S, Ringler DJ, et al: Cloning of the human eosinophil chemoattractant, eotaxin. Expression, receptor binding, and functional properties suggest a mechanism for the selective recruitment of eosinophils. J Clin Invest 97:604±612, 1996b Proudfoot AE, Power CA, Hoogewerf AJ, Montjovent MO, Borlat F, Offord RE, Wells TN: Extension of recombinant human RANTES by the retention of the initiating methionine produces a potent antagonist. J Biol Chem 271:2599± 2603, 1996 Rothenberg ME, Luster AD, Lilly CM, Drazen JM, Leder P: Constitutive and allergen-induced expression of eotaxin mRNA in the guinea pig lung. J Exp Med 181:1211±1216, 1995 Schulz BS, Michel G, Wagner S, et al: Increased expression of epidermal IL-8 receptor in psoriasis. Down-regulation by FK-506 in vitro. J Immunol 151:4399± 4406, 1993 Sica A, Matsushima K, Van Damme J, et al: IL-1 transcriptionally activates the neutrophil chemotactic factor/IL-8 gene in endothelial cells. Immunology 69:548±553, 1990 Tiffany HL, Alkhatib G, Combadiere C, Berger EA, Murphy PM: CC chemokine receptors 1 and 3 are differentially regulated by IL-5 during maturation of eosinophilic HL-60 cells. J Immunol 160:1385±1392, 1998 Tuschil A, Lam C, Haslberger A, Lindley I: Interleukin-8 stimulates calcium transients and promotes epidermal cell proliferation. J Invest Dermatol 99:294± 298, 1992 Uguccioni M, Mackay CR, Ochensberger B, et al: High expression of the chemokine receptor CCR3 in human blood basophils. Role in activation by eotaxin, MCP-4, and other chemokines. J Clin Invest 100:1137±1143, 1997 Wells TNC, Power CA, Lusti-Narasimhan M, et al: Selectivity and antagonism of chemokine receptors. J Leukoc Biol 59:53±60, 1996 White JR, Imburgia C, Dul E, et al: Cloning and functional characterization of a novel human CC chemokine that binds to the CCR3 receptor and activates human eosinophils. J Leukoc Biol 62:667±675, 1997 Ying S, Robinson DS, Meng Q, et al: Enhanced expression of eotaxin and CCR3 mRNA and protein in atopic asthma. Association with airway hyperresponsiveness and predominant co-localization of eotaxin mRNA to bronchial epithelial and endothelial cells. Eur J Immunol 27:3507±3516, 1997