CD1a and CD1c cell sorting yields a homogeneous population of immature human Langerhans cells

Journal of Immunological Methods 279 (2003) 41 – 53 www.elsevier.com/locate/jim

CD1a and CD1c cell sorting yields a homogeneous population of immature human Langerhans cells Matthias Peiser a, Andreas Gru¨tzkau b, Reinhard Wanner c, Gerhard Kolde a,* a

Department of Dermatology and Allergy, Charite´, Humboldt-University of Berlin, Schumannstr. 20/21, D-10117, Berlin, Germany b Deutsches Rheuma-Forschungszentrum Berlin, Berlin, Germany c Department of Molecular Biology and Biochemistry, Free University of Berlin, Berlin, Germany Received 5 February 2003; received in revised form 28 May 2003; accepted 2 June 2003

Abstract There is increasing evidence that ex vivo generated Langerhans cells (LCs) cannot fully substitute for their physiological counterparts in normal epidermis when studying the immunobiology of this prototype of a tissue-residing immature dendritic cell (DC). Here, we present CD1-based magnetic-activated cell-sorting (MACS) protocols for the effective isolation of human epidermal LCs. CD1c selection yielded a homogeneous population of pure and viable HLA-DR+/CD1a+ DCs, with the ultrastructural features, surface antigen expression and cytokine profile, characteristic of epidermis-resident immature LCs. The immature state and functional integrity were established by allogeneic mixed lymphocyte reactions showing a weak stimulatory capacity of freshly isolated cells and upregulation upon stimulation. Characterizing the cells in more detail, we could demonstrate for the first time that normal human LCs express CXCR4, CD40 ligand (CD40L), and Fas and Fas ligand (FasL). The observed constitutive transcription of TGF-h suggests that the viability and immature state of epidermal LCs are maintained not only by the TGF-h production from the microenvironment, but also in an autocrine or paracrine manner. LPS and IFN-N stimulated the expression of the inflammatory cytokines TNF-a and IL-1h, and there was secretion of IL-12p70 after CD40 ligation. Remarkably, the CD1-sorted LCs showed no loss of their Birbeck granules and CD1a expression upon culturing and no spontaneous phenotypic and functional maturation into potent antigen-presenting cells (APCs). We conclude that human epidermal LCs obtained by the CD1c cell-sorting protocol are optimal candidates with which to elucidate the properties and capabilities of immature cells and to develop immunotherapeutic vaccines. D 2003 Elsevier B.V. All rights reserved. Keywords: Human; Dendritic cells; Skin; Cellular differentiation

1. Introduction

Abbreviations: APC, antigen-presenting cell; CD40L, CD40 ligand; DC, dendritic cell; EC, epidermal cell; FasL, Fas ligand; LC, Langerhans cell; MACS, magnetic-activated cell sorting; MLR, mixed lymphocyte reaction; SEA, Staphylococcus enterotoxin A. * Corresponding author. Tel.: +49-30-450-51-80-11; fax: +4930-450-51-89-15. E-mail address: [email protected] (G. Kolde). 0022-1759/$ - see front matter D 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0022-1759(03)00257-6

Dendritic cells (DCs) are professional antigen-presenting cells (APCs) within the immune system that have the capacity to induce different types of T cellmediated immune responses or even tolerance. The lineage and subset of a particular DC, the local maturation affected by the microenvironment, and the activation signals from pathogens and cytokines are crucial

42

M. Peiser et al. / Journal of Immunological Methods 279 (2003) 41–53

for the effector functions (Banchereau et al., 2000; Liu et al., 2001). Langerhans cells (LCs) are a distinct type of DC characterized by their unique tennis racket-like Birbeck granules. LCs reside in an immunologically immature state in the epidermis and in mucosal tissues where they can efficiently take up and process a wide variety of antigens, including contact sensitizers, tumor-associated antigens, and microbial antigens (Banchereau et al., 2000; Teunissen et al., 1997). After capturing antigens, they emigrate from the epidermis and undergo a maturational process into professional antigen-presenting cells (APCs) capable of initiating a primary T cellmediated immune response in the regional lymph nodes. Crucial for understanding the immunobiology of LCs is a precise elucidation of their physiological phenotype and functional capacity (Banchereau et al., 2000; Girolomoni et al., 2002). Due to the difficulties in isolating a population of pure, viable, and functionally intact LCs from the epidermis, most studies on human LCs are now performed on ex vivo generated cells. Human LCs can be generated from either CD34+ precursors or monocytes by the use of diverse differentiation-inducing mediators (Ardavin et al., 2001). There is, however, increasing evidence that the LCs obtained by the various protocols show a more pronounced heterogeneity and plasticity of their differentiation and maturation than their physiological counterparts in normal epidermis (Ardavin et al., 2001; Banchereau et al., 2000; Shortman and Liu, 2002). For example, TGF-hindependent LC generation from CD34+ precursors seems to reflect a defective differentiation or maturation process unrelated to the behaviour of tissueresident LCs (Ardavin et al., 2001; Caux et al., 1996). The LCs generated from monocytes in the presence of either GM-CSF, IL-4, and TGF-h1 (Geissmann and Hermine, 1998), or IL-15 (Mohamadzadeh and Banchereau, 2001), often fail to display Birbeck granules despite expressing the C-type lectin Langerin, a potent inducer of Birbeck granules (Valladeau et al., 2000). These and other findings indicate that ex vivo generated LCs can only partially substitute for their naturally occurring counterparts when studying phenotypic and functional properties. Cell-specific surface antigens can be utilized for sorting procedures. The CD1 genes encode for MHC class I-like transmembrane glycoproteins that are

prominently expressed on cells involved in antigen presentation (Porcelli and Modlin, 1999). Five closely related human CD1 proteins designated CD1a –e are now known to present microbial lipid and glycolipid antigens to a variety of effector T cells. In human epidermis, the CD1a and CD1c molecules are almost exclusively expressed by LCs (Elder et al., 1993) and therefore represent excellent surface markers for their isolation from other epidermal cells (ECs), in particular, from keratinocytes. Various enrichment techniques have indeed shown that the positive selection via CD1a labeling can be used for isolating LCs from human skin (Ashworth et al., 1989; Bjercke et al., 1984; Hanau et al., 1988; Nilsson et al., 1987; Simon et al., 1995). However, all these techniques were shown to have disadvantages untenable for in vitro studies, such as a significant number of dead LCs, an impairment of phenotypic and functional characterization due to surface labeling, and most importantly, an incomplete separation from keratinocytes which is not acceptable due to their generation of numerous immunomodulatory cytokines (Wang et al., 1999) and their ability to proliferate and overgrow cultured LCs. Here, we report a new isolation technique for human epidermal LCs that is based on the positive selection of CD1a- and CD1c-labelled cells. The cellsorting protocols are capable at isolating a homogeneous population of pure, viable, and functionally intact LCs that retain their immature state in normal skin. The characterization of the cells permitted new insights into the phenotypic and functional properties of normal human LCs and into their behaviour upon long-term culturing.

2. Material and methods 2.1. Media and reagents Cells were cultured in RPMI 1640 supplemented with 2mM L-glutamine, 100 U/ml penicillin, 100 Ag/ml streptomycin, and 10% FCS (all from Biochrom, Berlin, Germany) at 37 jC, 5% CO2, and 95% humidity. Viability assays were performed additionally with the serum-free media X-Vivo 10 (Bio Whittaker, Wakersville, USA) and CellGro-DC (Cell Genix, Freiburg, Germany). CD154 transfected fibroblasts (TRAP cells) were kindly provided by Dr. Kroczek (Robert

M. Peiser et al. / Journal of Immunological Methods 279 (2003) 41–53

Koch-Institute, Berlin, Germany). LPS serotype 0128:B12 was from Sigma (Deisenhofen, Germany), Staphylococcus enterotoxin A (SEA) from Toxin Technology (Sarasota, USA), recombinant human IFN-g from Pharmingen (Heidelberg, Germany), and recombinant human IFN-N from Bender (Vienna, Austria). 2.2. Preparation of epidermal cell suspensions Normal human skin was obtained as discarded material from plastic surgery (reduction mammoplasty) after local ethical committee approval. Epidermal cell (EC) suspensions were prepared as previously described (Picut et al., 1987) with slight modifications. Briefly, thin skin stripes were incubated in PBS containing 3 RMB units of dispase I (Roche, Mannheim, Germany) for 18 h at 4 jC. Epidermal sheets were then stripped off the dermal layer using forceps and dissociated in PBS containing 0.25% trypsin (Biochrom) and 0.01% DNase (Roche) for 15 min at 37 jC. Single cell suspensions were obtained by vigorous pipetting and passaging through a 40-Am cell strainer (BD, Heidelberg, Germany). Cell numbers, debris, and clumps of aggregated cells were determined using a CASY 1 cell counter (Schaerfe System, Reutlingen, Germany), and numbers of dead cells were quantified by trypan blue exclusion. To avoid spontaneous migration of dendritic cells from the explants, the skin was kept on ice during transport and prepared immediately.

43

emigrating dendritic cells, epidermal sheets were placed in RPMI 1640 supplemented with 10% FCS at 37 jC for 24 h and the non-adherent cells collected. 2.4. Culture of Langerhans cells The isolated LCs were cultured for up to 15 days in 24-well tissue culture plates (Greiner, Frickenhausen, Germany) using culture media with no cytokines added. Routinely, cell cultures were set up at a density of 106 cells/ml. Every second day, one-third of the medium was changed (cell loss due to change of medium was < 1% per well). The cell numbers, viability, phenotype, and function were assessed at the beginning and after different time points of culturing. 2.5. Flow cytometry The cell surface expression of various proteins was analyzed by three-colour flow cytometry. The monoclonal antibodies used are listed in Table 1. Each 2  105 cells were labeled with F(abV)2 fragments of rabbit anti-mouse Ig (RPE-Cy5 dye 3; recognizing the CD1a-MACS-mAb) and anti-HLA-DR-PE. Other proteins were detected with FITC-coupled monoclonal antibodies gated on CD1a and HLA-DR doublepositive cells using an Epics XL-II flow cytometer (Beckman Coulter, Krefeld, Germany). Dead cells and debris were excluded by scatter gates and propidium iodide staining (1 Ag/ml) (Sigma).

2.3. Isolation of Langerhans cells 2.6. Electron microscopy LCs were isolated from EC suspensions by sorting via CD1a and via CD1c, and by harvesting cells migrating from epidermal sheets. For CD1a sorting, the EC suspensions were incubated for 15 min at 8 jC with a mouse anti-human CD1a antibody directly coupled to super-paramagnetic microbeads (clone HI149, IgG1) (Miltenyi Biotec, Bergisch Gladbach, Germany) in PBS/10mM EDTA/0.5% BSA. For CD1c sorting, the EC suspension was subjected to a biotinylated mouse anti-human CD1c antibody (clone AD58E7, IgG2a; Miltenyi Biotec) and to an anti-biotin antibody conjugated to microbeads (clone Bio3-18E7, IgG1; Miltenyi Biotec). Magnetic-activated cell sorting (MACS) was performed according to the manufacturer’s recommendations by passing the cells over large cell separation columns (Miltenyi Biotec). To harvest

The methods used for routine electron microscopy and the immunoelectron microscopic visualization of the microbeads have been described in detail elsewhere (Kolde and Knop, 1986; Kolde et al., 1992). Briefly, routine electron microscopy was performed on cell preparations fixed in half-strength Karnovsky’s solution (0.1 M cacodylate buffer, pH 7,4) for 30 min at 4 jC. After washing in PBS, the cells were postfixed in 1.33% osmic acid (0.05 M PBS, pH 7.4; Fluka, Buchs, Switzerland) for 2 h, dehydrated in a graded ethanol series and embedded in Araldite (Poersch, Frankfurt, Germany). Ultrathin sections were counterstained with uranyl acetate and lead citrate and were examined with a Zeiss electron microscope (EM 906; Zeiss, Oberkochen, Germany).

44

M. Peiser et al. / Journal of Immunological Methods 279 (2003) 41–53 Table 1 (continued)

Table 1 mAbs used for flow cytometry Specificity

Clone

Isotype

Source

CD1a

HI149

mIgG1

CD1c

AD5-8E7

mIgG2a

CD11c

3.9

mIgG1

CD40

B-B20

mIgG1

CD40L

TRAP1

mIgG1

CD44

F10-44-2

mIgG2a

CD49b

AK7

mIgG1

CD49c

MIKd2

mIgG1

CD49d

BU49

mIgG1

CDw49f

4F10

mIgG2b

CD50

Cal 3.10

mIgG1

CD54

84H10

mIgG1

CD80

L307.4

mIgG1

CD83

HB15e

mIgG1

CD86

FUN-1

mIgG1

CD95

DX2

mIgG1

CD95L

NOK-1

mIgG1

CD120a

16803.161

mIgG1

CD120b

22235.311

mIgG2a

HLA-DR

B8.12.2

mIgG2b

CCR5

45531.111

mIgG2b

CCR7

2H4

mIgM

CXCR4

44717.111

mIgG2b

FcqRIa

29C6

mIgG1

Cntrl

MOPC-21

mIgG1

Cntrl,

CBL601

mIgG2a

BD Pharmingen, Heidelberg, Germany Miltenyi, Bergisch Gladbach, Germany Dianova, Hamburg, Germany Dianova, Hamburg, Germany Dr. Kroczek, Berlin, Germany Dianova, Hamburg, Germany Dianova, Hamburg, Germany Dianova, Hamburg, Germany Dianova, Hamburg, Germany Dianova, Hamburg, Germany Coulter, Krefeld, Germany Coulter, Krefeld, Germany BD Pharmingen, Heidelberg, Germany BD Pharmingen, Heidelberg, Germany BD Pharmingen, Heidelberg, Germany BD Pharmingen, Heidelberg, Germany BD Pharmingen, Heidelberg, Germany R&D, WiesbadenNordenstadt, Germany R&D, WiesbadenNordenstadt, Germany Coulter, Krefeld, Germany R&D, WiesbadenNordenstadt, Germany BD Pharmingen, Heidelberg, Germany R&D, WiesbadenNordenstadt, Germany Dr. Hakimi, Nutley, NJ, USA BD Pharmingen, Heidelberg, Germany Dianova, Hamburg, Germany

Specificity

Clone

Isotype

Source

Cntrl

MOPC-195

mIgG2b

Cntrl

G155-228

mIgM

mIgG

C0090

rF(abV)2

Coulter, Krefeld, Germany BD Pharmingen, Heidelberg, Germany Dako, Hamburg, Germany

For the immunoelectron microscopic microbead visualization, the cell preparations were incubated for 1 h at 37 jC with a gold-conjugated goat antimouse IgG antibody (particle diameter 10 nm; Dianova, Hamburg, Germany) prior to fixation. After washing, the cells were fixed in half-strength Karnovsky’s solution and processed as described above. 2.7. Allogeneic stimulation Naive CD4+ T cells were purified from buffy coats (purity >95%) using a CD4+ T cell isolation kit (Miltenyi Biotec). Freshly isolated LCs and long-term cultured LCs (days 7 and 14 of culture) were irradiated (30 Gy) and added in graded doses as stimulator cells to 105 allogeneic CD4+ T cells in 96-well roundbottom plates and then cultured in RPMI/10% FCS for 5 days. For the last 16 h, cells were pulsed with 1 ACi/well [3H]thymidine (Amersham, Freiburg, Germany). Cells were harvested onto glass fiber filters, and [3H]TdR incorporation was determined in a liquid scintillation counter (LS 6500, Beckman Coulter, Unterschleissheim-Lohhof, Germany). Results were calculated as means for three replicate wells. 2.8. RT-PCR and Multiplex PCR RNAwas isolated using a RNeasy Mini Kit (Qiagen, Hilden, Germany), and was transcribed into cDNA using a cDNA Cycle Kit (Invitrogen, Groningen, The Netherlands). Various probes were titrated to equal amounts of h-actin transcripts by semiquantitative PCR with primers 5V-CCTTCCTGGGCATGGAGTCCT-3V and 5V-AATCTCATCTTGTTTTCTGCG3V (TIB Molbiol, Berlin, Germany). For Multiplex PCR of inflammatory cytokines, primers for TNF-a, IL-1h, IL-6, IL-8, GM-CSF, and TGF-h were from a hINF1G kit (Maxim Biotech, San Francisco, USA), and cycling was allowed to proceed according to the man-

M. Peiser et al. / Journal of Immunological Methods 279 (2003) 41–53

ufacturer’s instructions. For the RT-PCR of IL-12p35, primers 5V-GAGAGAGACACAGAAGGAGA-3Vand 5V-GATTACCCTCAACGGACCGGAG-3V, and for IL12p40, primers 5V-GCAAGATGTGTCACCAGCAGTT-3V and 5V-AAGACCTGCAAAGTGGACGACC-3V(all TIB Molbiol) were used. The products were electrophoretically separated on 2% agarose gels and stained with ethidium bromide. As size standard, a 100-bp ladder with a pronounced 500-bp band (Invitrogen) was used. 2.9. Generation of monocyte-derived Langerhans cells Adherent monocytes from human blood were cultured for 6 days by the addition of GM-CSF (250 ng/ ml), IL-4 (50 ng/ml), and TGF-h1(10 ng/ml, all R&D, Wiesbaden-Nordenstadt, Germany) as described previously [Geissmann and Hermine, 1998]. 2.10. IL-12p70 ELISA Supernatants and cell homogenates from 1  106 LCs/ml cultured on CD154 transfected fibroblasts (TRAP cells, kindly provided by Dr. Kroczek, RKI Berlin) and stimulated with 1 Ag/ml LPS for 48h were collected, and IL-12p70 was quantified by ultrasensitive chemiluminescence ELISA (Quantiglo; R&D).

3. Results 3.1. CD1a isolates a distinct population of pure HLADR+ cells LCs were isolated from EC suspensions by magnetic cell sorting via CD1a labeling. To obtain a sufficient purity of the enriched cells, we optimized the concentration of the CD1a microbeads and the type of separation column. Best results were achieved when using a low microbead dilution (1/300) and a large cell separation column. The contaminating keratinocytes were efficiently removed by placing the fraction of the positively selected cells onto a second separation column. This second column run was consistently capable of isolating a homogeneous population of cells of similar size and granularity in flow-cytometric analysis and of a mean diameter of 11.3 Am as measured by a CASY cell counter. A mean of 93% of the

45

cells was positive for CD1a and also for HLA-DR+ (Fig. 1D and E; Table 2). The double sorting procedure yielded 1  106 CD1a+/HLA-DR+ cells from 25 g of human skin, with donor-dependent variations. As measured by propidium iodide staining, 90% of all isolated cells were viable independent of single or double separation. The increased purity of the cell population yielded by the second column run was, however, associated with a loss of 77% of LCs (not shown). 3.2. Like CD1a, CD1c isolates the same distinct population of CD1a+/HLA-DR+ cells In order to optimize further our LC isolation procedure, we used CD1c as the specific antigen in magnetic cell sorting. Like CD1a, the CD1c double sorting resulted in the isolation of a homogeneous population of LCs, coexpressing CD1a/HLA-DR (Fig. 1). However, sorting via CD1c yielded in higher purity (mean of 99%, Table 2) and also higher numbers of isolated LCs as compared to CD1a sorting. 3.3. CD1a/c-isolated LCs differ from LCs from epidermal sheets Enriched LC populations can be obtained following their emigration from epidermal sheets (Larsen et al., 1990). The phenotypic properties of such emigrated LCs differed from LCs isolated from the same donor via CD1a or CD1c sorting. Emigrated LCs showed a reduced CD1a expression (Fig. 1D) and an increased HLA-DR expression (Fig. 1E) as compared to magnetically sorted cells. 3.4. All of the CD1a/c-isolated cells contain Birbeck granules Transmission electron microscopy of the cell preparations obtained by CD1a or CD1c sorting revealed a homogeneous population of dendritic cells with only few, if any, contaminating keratinocytes (Fig. 2A). The dendritic cells were characterized by a lobulated nucleus, a well-organized cytoplasm and short dendritic processes (Fig. 2B). The cytoplasm of all dendritic cells contained rod- or racket-shaped Birbeck granules identifying them as LCs (Fig. 2C). The immunoelectron microscopic visualization of the CD1a microbeads demonstrated surface and intra-

46

M. Peiser et al. / Journal of Immunological Methods 279 (2003) 41–53

Fig. 1. FS/SS plots and flow-cytometric analyses of CD1-sorted cells. (A – C) FS/SS plots of CD1c sorting showing the cells not retained by the separation column (A) and the retained cells (B). After subjecting these retained cells to a second run in the separation column, there is a distinct population of less-granulated cells (C). (D – E) Flow-cytometric analyses reveal similar CD1a (D) and HLA-DR (E) expression patterns of cells sorted twice via CD1a (CD1a-LC, black line) or via CD1c (CD1c-LC, gray line). In addition, the expression patterns of the negative fraction (neg frac, dotted line) and of cells obtained by their spontaneous emigration from an epidermal sheet (mig-LC, less positive, black line) are shown. All analyses were performed on cells from the same donor. Data are representative of five independent experiments. Isotype controls (ctrl, bold line).

cellular gold labeling (Fig. 2D and E). The intracellular labeling was confined to endosomal and endoplasmic vesicles beneath the cell membrane and to some Birbeck granules (Fig. 2E). Remarkably, there

Table 2 Comparison of yield and purity of CD1a and CD1c isolated LCs CD1a

Total yield (%) CD1a expression (%) LCs/10 g skin

Isolation

CD1c

Isolation

Mean

S.D.

Mean

S.D.

0.4 92.6 4  105

0.2 3.8 2.6  105

0.6 98.8 6.3  105

0.2 0.8 1.7  105

Epidermal cells of one donor were split: fraction 1 was incubated with CD1a mAb, and fraction 2 was incubated with CD1c mAb. After two cycles of MACS isolation, the enriched cells were counted, stained with anti-mIgG F(abV)2 coupled to Cy5 (fraction 1), or stained directly with anti-CD1a FITC (fraction 2). Expression on cells of 15 donors were calculated as mean and standard derivation (S.D.).

was no labeling of other endoplasmic membranes, lysosomes or the Golgi area. 3.5. Phenotype of isolated LCs The multicolour analyses of double-sorted CD1a+/ HLA-DR+ cells revealed one distinct population without any remarkable subpopulation (Fig. 3). The expression of costimulatory and adhesion molecules, chemokine receptors, and death receptors was analyzed by triple-colour flow cytometry. As depicted in Figs. 3 and 6, none of these molecules was expressed at pronounced strong levels. Relatively high expression levels were detected for H-CAM (CD44), ICAM-3 (CD50), and TNF-RII (CD120b). Moderate expression levels were found for the costimulatory molecule CD40 and for the adhesion molecules CD11c and CD49f. The majority of isolated cells were positive for ICAM-1

M. Peiser et al. / Journal of Immunological Methods 279 (2003) 41–53

47

Fig. 2. Electron microscopy of CD1-sorted cells. (A) There is a homogeneous population of DCs with thin cytoplasmic protrusions (original magnification  4600). (B) The DCs show a lobulated nucleus (N), electron-dense lysosomes (L) and large electron-lucent vacuoles (V) (  12,900). (C) Numerous rods of Birbeck granules (BG) and coated vesicles (CV) accumulate near the plasma membrane (  27,800). (D) Immunoelectron microscopic visualization of the CD1a-coupled microbeads showing weak, heterogeneous labeling of the cell membrane. (  46,500). Upper inset shows higher magnification of the immunogold-coupled beads (  100,000). (E) Intracellular uptake of CD1a microbeads, with clustering at coated pits on the cell surface and strong labeling of early endosomes (  36,000).

(CD54), for the a integrins CD49c, CD49d, and CD49e, for CD40 ligand (CD40L), and for the chemokine receptors CXCR4 and CCR5. On a few cells only, CD86, CCR7, the death receptor Fas (CD95), Fas ligand (FasL) (CD95L), TNF-RI (CD120a), FcqRI, and CD49b were detectable. Minimal expression could be detected for CD80 and for the maturation marker CD83. 3.6. Analyses of inflammatory cytokine expression The expression of inflammatory cytokines was evaluated by Multiplex RT-PCR. Isolated LCs cultured for 4 h without stimulation demonstrated constitutive expression of IL-6, IL-8, and unexpectedly, of TGF-h (Fig. 4). Stimulation of LCs from the same donor with LPS, SEA or with the type I interferon-N induced IL-1h and TNF-a transcription (Fig. 4). Stimulation with LPS additionally induced GM-CSF transcription. Stimulation with the type II interferon IFN-g did not result in significant upregulation of the

cytokine expression (Fig. 4). As compared to unstimulated cells of the leukemic human mast cell line HMC-1, isolated LCs expressed IL-8 at higher levels (Fig. 4). IL-12p35 and -p40 transcript levels were determined by semiquantitative RT-PCR. Unexpectedly, LPS did not stimulate the transcription of these IL12 related RNAs (not shown). Only when stimulated by CD40L, cultured LCs secreted bioactive IL-12p70 protein (not shown). Addition of IFN-g upregulated IL-12, but at a lower level as compared to LCs generated from monocytes. 3.7. The stimulatory capacity of isolated LCs resembles functionally intact LCs The capacity of isolated LCs to stimulate naive allogeneic T cells was measured by mixed lymphocyte reaction (MLR) assays. Freshly isolated LCs induced no T cell proliferation. LCs stimulated with LPS, SEA or with IFN-g were able to induce moderate T cell

48

M. Peiser et al. / Journal of Immunological Methods 279 (2003) 41–53

Fig. 3. Flow-cytometric analyses of CD1c-isolated LCs. The expression of adhesion molecules, receptors, and ligands of costimulatory and apoptosis-associated molecules and of adhesion molecules resemble immature DCs. 1  105 CD1c+/HLA-DR+ cells from one donor were measured without any gating. The quadrants show staining with isotype controls, and numbers refer to percentage of positive cells. Results are representative of 10 independent experiments.

proliferation (Fig. 5). IFN-g secretion by T cells was intensified with increased numbers of cocultured isolated LCs (not shown). 3.8. Serum is required to maintain cell viability and the LC phenotype in culture Fig. 4. Multiplex PCR analysis of inflammatory cytokine expression. Size standard (l00 bp), positive control, and the PCR products generated from HMC-1 cells and from CD1c sorted LCs (lanes 4 – 8), were electrophoresed on a 2% agarose gel and stained with ethidium bromide. LCs were left untreated (w/o) or were stimulated for 4 h with 500 ng/ml LPS, 500 ng/ml SEA, 10 ng/ml IFN-g or with 10 ng/ml IFN-N. A representative result of three independent experiments is shown.

To determine the optimal conditions for long-term culturing, freshly isolated LCs were cultured in RPMI, and in X-Vivo 10 and CellGro-DC commonly used for generation of DCs under serum-free conditions. In all the media, a marked decrease of cell viability was observed within the first 2 days of culturing. Thereafter, the remaining viable cells survived only if the

M. Peiser et al. / Journal of Immunological Methods 279 (2003) 41–53

49

were often characterized by irregular ring-shaped and elongated stack formations (not shown). 3.9. In culture, LCs express CD83 and B7 molecules at low levels

Fig. 5. T cell stimulatory capacity of isolated LCs. The LCs from one donor were left untreated or were prestimulated with LPS (500 ng/ml), SEA (500 ng/ml) or with IFN-g (10 ng/ml) for 1 day in RPMI/10% FCS. Naive CD4+ cells (99% CD3+, 94% CD4+ proven by flow cytometry) were isolated from buffy coats using MACS depletion. Graded numbers of LCs were cocultured with 5  104 T cells. Proliferation of alloreactive T cells was measured as triplicates by [3H]TdR incorporation. A representative result of three independent experiments is shown.

culture medium was RPMI supplemented with 10% FCS (not shown). Electron microscopy of these longterm cultured cells showed Birbeck granules which

We analyzed the phenotype of LCs during culturing in RPMI supplemented with 10% FCS. Flow cytometry showed constant CD1a expression over a culture period of 14 days (not shown). The expression of the maturation marker CD83 and of the costimulatory molecules CD80 and CD86 was analyzed during the first 7 days of culturing. After 1 day, about 30% of the cells became positive for CD83 at a weak expression level, which was maintained upon further culturing (Fig. 6). A distinct cell population with high level CD83 expression shifted to lower expression levels after 1 day of cell culturing. The CD86 analyses revealed five different cell populations after 1 day, one showing high level expression for both CD86 and HLA-DR, three populations with low CD86 and either high, moderate or absent HLA-DR expression, and one CD86 negative population (Fig. 6). These different expression patterns disappeared during the subse-

Fig. 6. B7 and CD83 expression by freshly obtained CD1c-isolated LCs (day 0) and LCs cultured for several days in RPMI/10% FCS. Dot plots demonstrate HLA-DR coexpression with B7 molecules and CD83 in cells from one donor. The quadrants show staining with isotype controls, and numbers refer to percentage of positive cells.

50

M. Peiser et al. / Journal of Immunological Methods 279 (2003) 41–53

quent culture, and there was again a homogeneous population, with CD86 and HLA-DR expression in 80% of the cells. Neoexpression of CD80 was found on 30% of the cultured cells. LCs cultured for 1, 3, and 7 days did not show an increased stimulatory capacity in MLR assays as compared to freshly isolated LCs (not shown).

4. Discussion The CD1a and CD1c magnetic-activated cell-sorting protocols reported in this study are effective and reliable methods for isolating epidermal LCs from human skin. Both the CD1a and CD1c cell sorting isolated a homogeneous population of pure and viable HLA-DR+ DCs. The presence of typical Birbeck granules in all of the separated cells excluded contamination with CD1a+ or CD1c+ dermal DCs (Elder et al., 1993). Contaminating keratinocytes which make LC preparations untenable for in vitro studies were successfully removed. The CD1c sorting protocol proved to be an even more effective tool, as the loss of cells observed with CD1a sorting after two cycles of MACS could be markedly reduced. As described for the microbead-based isolation of CD34+ cells (Handgretinger et al., 1998), the phenotypic and functional analyses of the isolated LCs were not impaired by microbead labeling. The immunoelectron microscopic visualization of the CD1a microbeads showed a rather patchy labeling of the cell surface and, in addition, an intracellular uptake of the beads, with positive labeling of only the peripheral endocytotic and endosomal vesicles, and of some Birbeck granules. The analyses of the morphological, phenotypic, and functional features of the isolated LCs demonstrated that the microbead labeling with anti-CD1a or anti-CD1c antibodies, and the separation runs induced neither significant alterations nor an artificial maturation of the cells. Like normal epidermis-resident LCs (Teunissen et al., 1997), the freshly isolated cells were characterized by dendritic morphology with few cytoplasmic veils, strong expression of CD1a and HLADR, and a distinct expression profile of adhesion molecules, with moderate to strong expression of CD11c, CD44, CD49f, and CD50. The low levels of CD49d and CD49e detected on the majority of our

cells fit with the findings in ex vivo generated human LCs (Riedl et al., 2000; Strunk et al., 1997). The isolated LCs showed no expression of the costimulatory molecule CD80 or of the maturation marker CD83. The costimulatory molecule CD86 was weakly expressed, but this finding was recently also observed in other studies on normal human LCs (Companjen et al., 2001; McLellan et al., 1998). Most importantly, unstimulated CD1a and CD1c isolated LCs showed only weak allostimulatory capacity in a primary MLR and were capable of maturation into APCs after appropriate stimulation. The effective isolation of pure, viable, and functionally intact normal human LCs enabled us to get new insights into their phenotypic and functional properties and behaviour upon culturing. With regard to chemokine receptors, the analyses confirmed that epidermis-resident immature LCs express CCR5, but mostly lack CCR7 (Yanagihara et al., 1998; Zaitseva et al., 1997). The CXCR4 protein as yet has only been shown to be expressed intracellularly in fresh human LCs and to be transported to the surface during in vitro culture (Zaitseva et al., 1997; Zoeteweij et al., 1998). In contrast to these findings, we, like Tchou et al. (2001), could clearly detect a surface expression of this receptor that was quite similar to the CCR5 expression. In addition to the well-documented expression of CD40 (Companjen et al., 2001; PeguetNavarro et al., 1995; Rattis et al., 1998), our cells also demonstrated low expression of CD40L. Recently, freshly isolated murine LCs were reported to express mRNA and intracellular but not surface CD40L (Salgado et al., 1999). The expression of death receptors in normal LCs is largely unknown, except for TNF-RII that is known to be strongly expressed in immature cells and to play a crucial role in mediating their migration during maturation (Takayama et al., 1999). The cells obtained by CD1a and CD1c cell sorting showed not only strong TNF-RII, but also weak TNF-RI expression. Moreover, the cells were weakly positive for both Fas and FasL. We are currently aware of only three reports demonstrating that maturate LCs are capable of expressing these death receptors (Kawamura et al., 1999, 2000; Shibaki and Katz, 2001). Due to the difficulties in isolating pure preparations of human epidermal LCs, their production of inflammatory cytokines is still far from clear. Our RT-PCR

M. Peiser et al. / Journal of Immunological Methods 279 (2003) 41–53

analyses demonstrated that unstimulated human LCs are characterized by constitutive expression of IL-6, IL-8, and TGF-h. A spontaneous expression of IL-6 RNA and protein has previously only been reported for murine epidermal LCs (Cumberbatch et al., 1996; Schreiber et al., 1992). The detection of IL-8 transcripts confirms recent findings (Nakagawa et al., 1999) and indicates that unstimulated LCs already contribute to the production of this chemokine in normal human epidermis. TGF-h is critically involved in LC development from CD34+ progenitors and in the preservation of the viability and immature state of epidermis-resident LCs (Borkowski et al., 1996; Geissmann et al., 1999; Strobl et al., 1996). The latter effects are thought to be mediated by the TGF-h production and release from the epidermal microenvironment (Borkowski et al., 1996; Geissmann et al., 1999). The detection of TGF-h mRNA in our cells suggests that epidermis-resident LCs can produce this cytokine by themselves and contribute in the regulation of their own cellular and functional properties in an autocrine or paracrine manner. In LCs, the cytokines IL-1h and TNF-a have been demonstrated to be upregulated during maturation, particularly during the early events of contact hypersensitivity (Enk and Katz, 1992; Heufler et al., 1992; Larrick et al., 1989; Sauder et al., 1984). Most of these studies were, however, performed on only partially enriched LC preparations and therefore could not exclude the possibility that the increased production of IL-1h and TNF-a was due to the contaminating keratinocytes. In our analyses, there were no transcripts for these cytokines in unstimulated LCs. Stimulation with LPS or IFN-N induced transcription of IL-1h, TNF-a, and additionally of GM-CSF that was recently also detected by immunocytochemistry (Lore et al., 1998). The detection of all these proinflammatory cytokines already 4 h after stimulation indicates that immature LCs immediately respond to antigenic stimulation. The absence of IL-12p35 and -p40 transcripts in the freshly isolated LCs even after LPS stimulation is in contrast to the findings in human blood-derived myeloid DCs, which produce this cytokine in response to bacteria, viruses, and protozoa (Reis e Sousa et al., 1999). However, when cross-linked with CD40L isolated LCs also secreted bioactive IL-12p70 protein. The release of only small amounts of IL-12p70 seems not to

51

have been due to the isolation procedure, as LCs generated in vitro from blood monocytes and then subjected to CD1c sorting were found to strongly express IL-12 protein after stimulation (data not shown). Our findings are supported by the failure to measure Il-12 protein in the supernatants of unstimulated LCs or of CD40L- or IFN-g-stimulated human LCs that were obtained by emigration from epidermal sheets (Nakagawa et al., 1999). Moreover, the MLR analyses of our LCs showed that the activation of the cells with LPS, SEA or IFN-g was not able to increase their stimulatory capacity to fully maturate accessory cells (Heufler et al., 1988). It is thus reasonable to assume that physiological epidermis-resident LCs like thymic and splenic DCs (Hochrein et al., 2000; Schulz et al., 2000) require multiple activation signals to become professional APCs in contrast to their in vitro generated counterparts. Since the report by Schuler and Steinman in 1985, it is generally thought that epidermal LCs isolated from normal human and murine skin are committed to spontaneously undergo a continuous maturation into follicular dendritic cells upon culturing. This maturation was found with all isolation techniques used and is associated with profound morphological, phenotypic, and functional changes such as loss of Birbeck granules and CD1a expression, upregulation of the expression of MHC class II molecules, of costimulatory molecules, and of the maturation marker CD83, as well as an increase of the allostimulatory capacity in a primary MLR (Teunissen et al., 1997). The LCs obtained by CD1a and CD1c cell sorting did not show significant changes of their differentiation state upon culturing. Nevertheless, upon the first days of culturing, the isolated cells showed a marked decrease of their viability that is also described for cultured LCs obtained by other enrichment techniques (Liu et al., 1994; Schuler and Steinman, 1985; Teunissen et al., 1990; Witmer-Pack et al., 1987). In conclusion, the CD1a and, in particular, CD1c microbead cell sorting of human LCs ensures the isolation of pure, viable, and functionally intact cells that retain their immature state and do not undergo significant spontaneous maturation upon culturing. The cells thus provide an optimal tool for elucidating the phenotypic and functional properties of epidermisresident LCs and for studying the largely unknown cellular and molecular mechanisms responsible for

52

M. Peiser et al. / Journal of Immunological Methods 279 (2003) 41–53

their differentiation, immunostimulatory capabilities, and migration. Moreover, it is reasonable to assume that the homogeneous and defined population of physiological LCs obtained by our protocol may be suitable for the development of immunotherapeutic vaccines.

Acknowledgements This work was supported by a grant from Deutsche Forschungsgemeinschaft (DFG Ko 1730/1-2).

References Ardavin, C., Martinez-del-Hoyo, G., Martin, F., Anjuere, P., Arias, C.F., Marin, A.R., Ruiz, S., Parrillas, V., Hernandez, H., 2001. Origin and differentiation of dendritic cells. Trends Immunol. 22, 691. Ashworth, J., Kahan, M.C., Breathnach, S.M., 1989. Flow cytometric analysis and sorting of HLA-DR+CD1+ Langerhans cells. Br. J. Dermatol. 121, 11. Banchereau, J., Briere, F., Caux, C., Davoust, J., Lebecque, S., Liu, Y.L., Pulendran, B., Palucka, K., 2000. Immunobiology of dendritic cells. Annu. Rev. Immunol. 18, 767. Bjercke, S., Elgo, J., Braathen, L., Thorsby, E., 1984. Enriched epidermal Langerhans cells are potent antigen-presenting cells for T cells. J. Invest. Dermatol. 83, 286. Borkowski, T.A., Letterio, J.J., Farr, A.G., Udey, M.C., 1996. A role for endogenous transforming growth factor beta 1 in Langerhans cell biology: the skin of transforming growth factor beta 1 null mice is devoid of epidermal Langerhans cells. J. Exp. Med. 184, 2417. Caux, C., Vanbervliet, B., Massacrier, C., Dezutter-Dambuyant, C., de-Saint-Vis, B., Jacquet, C., Yoneda, K., Imamura, S., Schmitt, D., Banchereau, J., 1996. CD34+ hematopoietic progenitors from human cord blood differentiate along two independent dendritic cell pathways in response to GM-CSF + TNF alpha. J. Exp. Med. 184, 695. Companjen, A.R., van der Wel, L.I., Wie, L., Laman, J.D., Prens, E.R., 2001. A modified ex vivo skin organ culture system for functional studies. Arch. Dermatol. Res. 293, 184. Cumberbatch, M., Dearman, R.J., Kimber, I., 1996. Constitutive and inducible expression of interleukin-6 by Langerhans cells and lymph node dendritic cells. Immunology 87, 513. Elder, J.T., Reynolds, N.J., Cooper, K.D., Griffiths, C.E., Hardas, B.D., Bleicher, P.A., 1993. CD1 gene expression in human skin. J. Dermatol. Sci. 6, 206. Enk, A.H., Katz, S.I., 1992. Early molecular events in the induction phase of contact sensitivity. Proc. Natl. Acad. Sci. U. S. A. 89, 1398. Geissmann, F., Hermine, O., 1998. TGF-beta1, in the presence of GM-CSF and IL-4 induces differentiation of human peripheral blood monocytes into dendritic LCs. J. Exp. Med. 187, 961.

Geissmann, F., Revy, P., Regnault, A., Lepelletier, Y., Dy, M., Brousse, N., Amigorena, S., Hermine, O., Durandy, A., 1999. TGF-beta 1 prevents the noncognate maturation of human dendritic Langerhans cells. J. Immunol. 162, 4567. Girolomoni, G., Caux, C., Dezutter-Dambuyant, C., Lebecque, S., Ricciardi-Castagnoli, P., 2002. Langerhans cells: still a fundamental paradigm for studying the immunobiology of dendritic cells. Trends Immunol. 23, 6. Hanau, D., Schmitt, D.A., Fabre, M., Cazenave, J.P., 1988. A method for the rapid isolation of human epidermal Langerhans cells using immunomagnetic microspheres. J. Invest. Dermatol. 91, 274. Handgretinger, R., Lang, P., Schumm, M., Taylor, G., Neu, S., Koscielnak, E., Niethammer, D., Klingebiel, T., 1998. Isolation and transplantation of autologous peripheral CD34+ progenitor cells highly purified by magnetic-activated cell sorting. Bone Marrow Transplant. 21, 987. Heufler, C., Koch, F., Schuler, G., 1988. Granulocyte/macrophage colony-stimulating factor and interleukin 1 mediate the maturation of murine epidermal Langerhans cells into potent immunostimulatory dendritic cells. J. Exp. Med. 167, 700. Heufler, C., Topar, G., Koch, F., Trockenbacher, B., Kampgen, E., Romani, N., Schuler, G., 1992. Cytokine gene expression in murine epidermal cell suspensions: interleukin 1 beta and macrophage inflammatory protein 1 alpha are selectively expressed in Langerhans cells but are differentially regulated in culture. J. Exp. Med. 176, 1221. Hochrein, H., O’Keeffe, M., Luft, T., Vandenabeele, S., Grumont, R.J., Maraskovsky, E., Shortman, K., 2000. Interleukin (IL)-4 is a major regulatory cytokine governing bioactive IL-12 production by mouse and human dendritic cells. J. Exp. Med. 192, 823. Kawamura, T., Azuma, M., Kayagaki, N., Shimada, S., Yagita, H., Okumura, K., 1999. Fas/Fas ligand-mediated elimination of antigen-bearing Langerhans cells in draining lymph nodes. Br. J. Dermatol. 141, 201. Kawamura, T., Azuma, M., Kayagaki, N., Shimada, S., Yagita, H., Okumura, K., 2000. Fas/Fas ligand-mediated apoptosis of murine Langerhans cells. J. Dermatol. Sci. 22, 96. Kolde, G., Knop, J., 1986. Ultrastructural morphometry of epidermal Langerhans cells: introduction of a simple method for a comprehensive quantitative analysis of the cells. Arch. Dermatol. Res. 278, 298. Kolde, G., Schulze-Osthoff, K., Meyer, H., Knop, J., 1992. Immunohistological and immunoelectron microscopical identification of TNFa in normal human and murine epidermis. Arch. Dermatol. Res. 284, 154. Larrick, J.W., Morhenn, V., Chiang, Y.L., Shi, T., 1989. Activated Langerhans cells release tumor necrosis factor. J. Leukoc. Biol. 45, 429. Larsen, C.P., Steinman, R.M., Witmer-Pack, M., Hankins, D.F., Morris, P.J., Austyn, J.M., 1990. Migration and maturation of Langerhans cells in skin transplants and explants. J. Exp. Med. 172, 1483. Liu, H.N., Chang, Y.T., Wong, C.K., 1994. Enrichment of unlabelled human Langerhans cells by modified discontinuous Ficoll-metrizoate density medium and following Langerhans cell culture. Clin. Exp. Dermatol. 19, 113.

M. Peiser et al. / Journal of Immunological Methods 279 (2003) 41–53 Liu, Y.J., Kanzler, H., Soumelis, V., Gilliet, M., 2001. Dendritic cell lineage, plasticity and cross-regulation. Nat. Immunol. 2, 585. Lore, K., Sonnerborg, A., Spetz, A.L., Andersson, U., Andersson, J., 1998. Immunocytochemical detection of cytokines and chemokines in Langerhans cells and in vitro derived dendritic cells. J. Immunol. Methods 218, 173 (Erratum). McLellan, A.D., Heiser, A., Sorg, R.V., Fearnley, D.B., Hart, D.N., 1998. Dermal dendritic cells associated with T lymphocytes in normal human skin display an activated phenotype. J. Invest. Dermatol. 111, 841. Mohamadzadeh, M., Banchereau, J., 2001. Interleukin 15 skews monocyte differentiation into dendritic cells with features of LCs. J. Exp. Med. 194, 1013. Nakagawa, S., Koomen, C.W., Bos, J.D., Teunissen, M.B., 1999. Differential modulation of human epidermal Langerhans cell maturation by ultraviolet B radiation. J. Immunol. 163, 5192. Nilsson, H., Johansson, C., Scheynius, A., 1987. Removal of Langerhans cells from human epidermal cell suspensions by immunomagnetic particles. J. Immunol. Methods 105, 165. Peguet-Navarro, J., Dalbiez-Gauthier, C., Rattis, F.M., Van Kooten, C., Banchereau, J., Schmitt, D., 1995. Functional expression of CD40 antigen on human epidermal Langerhans cells. J. Immunol. 155, 4241. Picut, C.A., Lee, C.S., Dougherty, E.P., Andersen, K.L., Lewis, R.M., 1987. Purification and identification of murine epidermal dendritic cells: flow cytometry and immunogold labeling. J. Histochem. Cytochem. 35, 745. Porcelli, S.A., Modlin, R.T., 1999. The CD1 system: antigen-presenting molecules for T cell recognition of lipids and glycolipids. Annu. Rev. Immunol. 17, 297. Rattis, F.M., Concha, M., Dalbiez-Gauthier, C., Courtellemont, P., Schmitt, D., Peguet-Navarro, J., 1998. Effects of ultraviolet B radiation on human Langerhans cells: functional alteration of CD86 upregulation and induction of apoptotic cell death. J. Invest. Dermatol. 111, 373. Reis e Sousa, C., Sher, A., Kaye, P., 1999. The role of dendritic cells in the induction and regulation of immunity to microbial infection. Curr. Opin. Immunol. 11, 392. Riedl, E., Stockl, J., Majdic, O., Scheinecker, C., Rappersberger, K., Knapp, W., Strobl, H., 2000. Functional involvement of Ecadherin in TGF-beta 1-induced cell cluster formation of in vitro developing human Langerhans-type dendritic cells. J. Immunol. 165, 1381. Salgado, C.G., Nakamura, K., Sugaya, M., Tada, Y., Asahina, A., Koyama, Y., Irie, S., Tamaki, K., 1999. Functional CD40 ligand is expressed on epidermal Langerhans cells. J. Leukoc. Biol. 66, 281. Sauder, D.N., Dinarello, C.A., Morhenn, V.B., 1984. Langerhans cell production of interleukin-1. J. Invest. Dermatol. 82, 605. Schreiber, S., Kilgus, O., Payer, E., Kutil, R., Elbe, A., Mueller, C., Stingl, G., 1992. Cytokine pattern of Langerhans cells isolated from murine epidermal cell cultures. J. Immunol. 149, 3524. Schuler, G., Steinman, R.M., 1985. Murine epidermal Langerhans cells mature into potent immunostimulatory dendritic cells in vitro. J. Exp. Med. 161, 526. Schulz, O., Edwards, A.D., Schito, M., Aliberti, J., Manickasingham, S., Sher, A., Reis e Sousa, C., 2000. CD40 triggering of

53

heterodimeric IL-12 p70 production by dendritic cells in vivo requires a microbial priming signal. Immunity 13, 453. Shibaki, A., Katz, S.I., 2001. Activation through CD40 ligation induces functional Fas ligand expression by Langerhans cells. Eur. J. Immunol. 31, 3006. Shortman, K., Liu, Y.J., 2002. Mouse and human dendritic cell subtypes. Nat. Rev. Immunol. 2, 151. Simon, J.C., Dittmar, H.C., de-Roche, R., Wilting, J., Christ, B., Schopf, E., 1995. Rapid purification of human Langerhans cells using paramagnetic microbeads. Exp. Dermatol. 4, 155. Strobl, H., Riedl, E., Scheinecker, C., Bello-Fernandez, C., Pickl, W.F., Rappersberger, K., Majdic, O., Knapp, W., 1996. TGFbeta 1 promotes in vitro development of dendritic cells from CD34+ hemopoietic progenitors. J. Immunol. 157, 1499. Strunk, D., Egger, C., Leitner, G., Hanau, D., Stingl, G., 1997. A skin homing molecule defines the Langerhans cell progenitor in human peripheral blood. J. Exp. Med. 185, 1131. Takayama, K., Yokozeki, H., Ghoreishi, M., Satoh, T., Katayama, I., Umeda, T., Nishioka, K.J., 1999. IL-4 inhibits the migration of human Langerhans cells through the downregulation of TNF receptor II expression. J. Invest. Dermatol. 113, 541. Tchou, I., Misery, L., Sabido, O., Dezutter-Dambuyant, C., Bourlet, T., Moja, P., Hamzeh, H., Peguet-Navarro, J., Schmitt, D., Genin, C., 2001. Functional HIV CXCR4 coreceptor on human epithelial Langerhans cells and infection by HIV strain X4. J. Leukoc. Biol. 70, 313. Teunissen, M.B., Wormmeester, J., Krieg, S.R., Peters, P.J., Vogels, I.M., Kapsenberg, M.L., Bos, J.D., 1990. Human epidermal Langerhans cells undergo profound morphologic and phenotypical changes during in vitro culture. J. Invest. Dermatol. 94, 166. Teunissen, M.B.M., Kapsenberg, M.L., Bos, J.D., 1997. Langerhans cells and related skin dendritic cells. In: Bos, J.D. (Ed.), Skin Immune System. CRC Press, Boca Raton, p. 59. Valladeau, J., Ravel, O., Dezutter-Dambuyant, C., Moore, K., Kleijmeer, M., Liu, Y., Duvert-Frances, V., Vincent, C., Schmitt, D., Davoust, J., Caux, C., Lebecque, S., Saeland, S., 2000. Langerin, a novel C-type lectin specific to Langerhans cells, is an endocytic receptor that induces the formation of Birbeck granules. Immunity 12, 71. Wang, B., Amerio, P., Sander, D.M., 1999. Role of cytokines in epidermal Langerhans cell migration. J. Leukoc. Biol. 66, 33. Witmer-Pack, M.D., Olivier, W., Valinsky, J., Schuler, G., Steinman, R.M., 1987. Granulocyte/macrophage colony-stimulating factor is essential for the viability and function of cultured murine epidermal Langerhans cells. J. Exp. Med. 166, 1484. Yanagihara, S., Komura, E., Nagafune, J., Watarai, H., Yamaguchi, Y., 1998. EBI1/CCR7 is a new member of dendritic cell chemokine receptor that is up-regulated upon maturation. J. Immunol. 161, 3096. Zaitseva, M., Blauvelt, A., Lee, S., Lapham, C.K., Klaus-Kovtun, V., Mostowski, H., Manischewitz, J., Golding, H., 1997. Expression and function of CCR5 and CXCR4 on human Langerhans cells and macrophages: implications for HIV primary infection. Nat. Med. 3, 1369. Zoeteweij, J.P., Golding, H., Mostowski, H., Blauvelt, A., 1998. Cytokines regulate expression and function of the HIV coreceptor CXCR4 on human mature dendritic cells. J. Immunol. 161, 3219.