Langerhans cells (LC) are the dendritic APC population of the epidermis, where they reside for long periods and are self-replicating. The molecular signals underlying these characteristics are unknown. The TNF superfamily member receptor activator of NF-κB ligand (RANKL, TNFSF11) has been shown to sustain viability of blood dendritic cells in addition to its role in promoting proliferation and differentiation of several cell types, notably osteoclasts. In this study, we have studied expression of the RANKL system in skin and have defined a key role for this molecule in LC homeostasis. In vitro and in vivo, human KC expressed RANKL and epidermal LC expressed cell surface RANK. In vitro, RANKL sustained CD34+ progenitor-derived LC viability following 72-h cultures in cytokine-free medium (79.5 ± 1% vs 55.2 ± 5.7% live cells, respectively; n = 4; p < 0.05). In vivo, RANKL-deficient mice displayed a marked reduction in epidermal LC density (507.1 ± 77.2 vs 873.6 ± 41.6 LC per mm2; n = 9; p < 0.05) and their proliferation was impaired without a detectable effect on apoptosis. These data indicate a key role for the RANKL system in the regulation of LC survival within the skin and suggest a regulatory role for KC in the maintenance of epidermal LC homeostasis.

Langerhans cells (LC)4 represent a subset of dendritic cells (DC) found in the epithelia of skin and mucosae. They express the common DC marker CD1a but can be distinguished from other DC by the presence of intracellular Birbeck granules, containing the C-type lectin Langerin (CD207) (1), and by their expression of E-cadherin, enabling tight interaction with epithelial cells (2). LC are long-lived in comparison to other DC subtypes, and recent studies have highlighted the ability of LC-precursors to proliferate in situ (3). The molecular signals responsible for these characteristics are unclear.

Recent data have suggested that the TNF superfamily molecule receptor activator of NF-κB (RANK, TNFRSF11a) and its ligand, RANKL (TNFSF11), play a key role in the regulation of several DC functions including enhancement of their capacity to activate T cells (4), increased production of proinflammatory cytokines (5), and prolonged cell survival (6, 7). Interestingly, it was found that though RANK/RANKL interactions are important factors in the regulation of DC functions, lack of either molecule did not affect the development and distribution of peripheral DC (8, 9). Normal ratios of various splenic DC subpopulations were observed as well as normal lymphocyte, platelet, granulocyte, and monocyte counts in peripheral blood. This was despite the osteopetrotic phenotype and may be explained by the extramedullary hematopoiesis observed in the liver of RANKL−/− mice (8). In the skin, RANKL overexpression has been shown to promote LC activation and to control regulatory T cell homeostasis (10). RANK has been shown to be up-regulated upon DC maturation and RANKL signaling from activated T lymphocytes provides a survival signal to some mature DC subtypes ensuring prolonged cellular interaction and induction of a sufficient immune response (11). RANKL activity in vivo is regulated by a decoy receptor, osteoprotegerin (OPG, TNFRSF11b). Signaling through RANK engages TRAF-1, -2, -3, -5, and -6 and leads to the activation of several transcription factors such as NF-κB, AP-1, ERK1/2, and NFAT. It has been associated with numerous physiological processes including organogenesis of lymph nodes and mammary glands (8, 9, 12), bone homeostasis (13, 14), and regulation of DC function (15, 16).

Both RANKL and RANK have recently been shown to be expressed in breast (17) and thymic (18) epithelia and we, therefore, sought to determine their expression and function in human skin epidermis. We hypothesized that RANKL signaling might be of relevance to the extended longevity of LC in the skin and, being tightly intercalated between keratinocytes (KC), that KC may be a good source of such a LC sustaining signal.

Skin biopsies were obtained after informed consent from patients undergoing breast reduction surgery or abdominoplasty. Epidermal sheets and epidermal single-cell suspensions were obtained as previously described (19, 20). Cord blood was obtained following informed consent. CD34+ progenitors were immunomagnetically isolated from umbilical cord blood and differentiated into LC-like DC (LCL-DC) as previously described (21).

Expression of specific markers was determined using the following Abs from R&D Systems: RANKL (mAb6262) and RANK (mAb683), both detected by an Alexa-488-goat anti-mouse IgG (Invitrogen), mouse anti-human CD1a-APC (BD Pharmingen), and rat anti-mouse IA-IE PE (BD Pharmingen). For intracytoplasmic staining, cells were fixed in formaldehyde (2%) for 20 min before saponin (0.5%)-induced permeabilization. Apoptosis was detected by Annexin V-APC and propidium iodide (PI) staining (BD Pharmingen) or TUNEL (Roche) as indicated in the manufacturers’ instructions. Specific binding was measured by flow cytometry on a FACSCalibur (BD Pharmingen) and analysis was performed using the WinMDI 2.8 software.

Expression of specific markers was determined using the following Abs from R&D Systems: RANKL (mAb6262) and RANK (mAb683) detected by an Alexa-488-goat anti-mouse IgG (Invitrogen), mouse anti-human CD1a-PE (BD Pharmingen), and rat anti-mouse IA-IE PE (BD Pharmingen). Specific binding was detected by epifluorescent or confocal microscopy as indicated and images were created using Photoshop 7.0 software (Adobe).

Following differentiation, LCL-DC were washed thoroughly and cultured for 48 h in 10% FCS RPMI 1640 supplemented as indicated with RANKL (50 ng/ml; PeproTech) and OPG (1 μg/ml; PeproTech). Alternatively, cells were cultured for 7 days in GM-CSF (50 ng/ml; Schering-Plough), TGF-β1 (4 ng/ml; R&D Systems), and TNF-α (10 ng/ml; R&D Systems) supplemented medium in the presence or absence of RANKL (50 ng/ml; PeproTech). Cells were collected and apoptosis determined by flow cytometry according to the manufacturer’s protocol (Annexin V APC Apoptosis Detection kit; BD Pharmingen).

RANKL knockout (KO) mice (22) and littermate controls were kindly provided by Y. W. Choi (University of Philadelphia, Medical School, Philadelphia, PA) and housed at the Swiss Institute for Experimental Cancer Research (ISREC) and at the Institut de Biologie Moléculaire et Cellulaire (IBMC), Strasbourg, where all procedures were performed according to institutional guidelines.

Ears were collected from RANKL KO and normal wild-type (WT) littermate controls. Epidermal sheets were obtained as previously described (19) and were labeled with rat anti-mouse IA-IE-PE Ab (BD Pharmingen) (1/100 dilution) to detect LC. Samples were visualized by epifluorescent microscopy and images of six randomly chosen microscopic fields were acquired. LC were later enumerated in each image and mean values generated for each sheet.

RANKL KO and WT littermate controls underwent in vivo BrdU labeling as previously described (23). The 18-day-old mice were injected with 1 mg BrdU in saline per 20-g body weight i.p. and the litter given drinking water containing 0.8 mg/ml BrdU for 7 days. Mice were then sacrificed and whole body skin collected. The hair was removed using a thioglycolate-based hair removal cream (Veet) and the skin incubated overnight at 4°C in KC-adapted medium (CnT-57; CellnTec) containing 5 mg/ml dispase II (Roche). The epithelium was removed and incubated in trypsin (TrypLE Select; Invitrogen) for 20 min at 37°C. Cells were liberated by pipetting, filtered to remove epidermal fragments, washed in saline containing 0.2 mM EDTA and 2% FCS, and labeled for IA-IE (PE-conjugated mAb; eBioscience). Cells were then processed for BrdU detection according to the manufacturer’s protocol (Flow BrdU detection kit; BD Pharmingen) with the modification that the Alexa 488-conjugated anti-Langerin mAb (clone 929F3; Dendritics) was added together with the APC-coupled anti-BrdU mAb to the cells.

Epidermal cell suspensions, generated by trypsin (Invitrogen) treatment of epidermal sheets obtained from ears as previously described (19), were subjected to immunostaining for IA-IE expression and TUNEL according to the manufacturer’s protocol (TUNEL Flow detection kit; ROCHE) and analyzed by flow cytometry.

RANKL and RANK expression was assessed by flow cytometry in epidermal single-cell suspensions obtained from normal human skin biopsies. Using this technique, basal KC can be distinguished from suprabasal KC according to their size and light scatter parameters (24, 25). Consistent with this, small low scatter cells express the basal-specific keratin 5, whereas larger cells associated with greater scatter were keratin 5 negative and keratin 10 positive, typical of suprabasal KC (data not shown). RANKL was clearly expressed by all KC (Fig. 1 A). Both basal and suprabasal KC expressed RANKL intracellularly, whereas suprabasal KC additionally expressed membrane bound RANKL. These findings were confirmed by immunofluorescent labeling of normal skin sections, which demonstrated RANKL expression throughout the epidermis with highest levels of expression in the suprabasal population.

FIGURE 1.

RANKL and RANK are expressed in normal human skin. A, Intracellular and membrane-bound expression of RANKL in KC was assessed by flow cytometry. Basal and suprabasal KC populations were distinguished according to forward and side scatter characteristics, identified, respectively, as regions R1 and R2. Membrane-bound RANKL increased upon differentiation, whereas intracellular expression was detected in both subpopulations (specific binding (red histograms) is shown compared with isotype control staining (clear histograms)). B, RANKL expression in epidermis was assessed by immunofluorescence (green) and counterstained with nuclear dye TO-PRO-3 (blue). Expression was maximal in the suprabasal layers. C, Immunofluorescence for CD1a/RANK in epidermal sections and sheets. All CD1a+ LC expressed RANK. D, Membrane expression of RANK was assessed on epidermal LC by flow cytometry. Epidermal single-cell suspensions were subjected to Ficoll flotation to enrich for LC. RANK was expressed by all CD1a+ LC (R4, upper right quadrant). Each experiment is representative of more than three samples.

FIGURE 1.

RANKL and RANK are expressed in normal human skin. A, Intracellular and membrane-bound expression of RANKL in KC was assessed by flow cytometry. Basal and suprabasal KC populations were distinguished according to forward and side scatter characteristics, identified, respectively, as regions R1 and R2. Membrane-bound RANKL increased upon differentiation, whereas intracellular expression was detected in both subpopulations (specific binding (red histograms) is shown compared with isotype control staining (clear histograms)). B, RANKL expression in epidermis was assessed by immunofluorescence (green) and counterstained with nuclear dye TO-PRO-3 (blue). Expression was maximal in the suprabasal layers. C, Immunofluorescence for CD1a/RANK in epidermal sections and sheets. All CD1a+ LC expressed RANK. D, Membrane expression of RANK was assessed on epidermal LC by flow cytometry. Epidermal single-cell suspensions were subjected to Ficoll flotation to enrich for LC. RANK was expressed by all CD1a+ LC (R4, upper right quadrant). Each experiment is representative of more than three samples.

Close modal

Labeling of skin sections and epidermal sheet preparations with anti-RANK Abs demonstrated constitutive high level expression of RANK by LC in situ (Fig. 1,C). We additionally assessed LC surface expression of RANK by flow cytometry and observed that the vast majority (95.6 ± 2.3%; n = 3) of the CD1a+ (R2) population expressed membrane-bound RANK. RANK was moreover detected on the surface of a KC subpopulation (Fig. 1 D).

RANKL signaling in mature DC has been associated with their survival. We, therefore, assessed whether RANKL/RANK interactions could sustain viability of immature LC obtained in vitro. Birbeck granule-positive, CD1a+ LCL-DC were generated from CD34+ cord blood progenitors in the presence of GM-CSF and TGF-β1 as previously described (21). RANK expression by the LCL-DC population was confirmed (Fig. 2,A). LCL-DC were cultured for an additional 48 h in complete medium deprived of cytokines and supplemented or not with RANKL in the presence or absence of OPG. Cellular viability was assessed after 48 h of culture by flow cytometry following Annexin V/PI staining (Fig. 2,B). Control LCL-DC cultured in cytokine free medium displayed rapid apoptosis with only half the cells alive after 72 h (55.2 ± 5.7%; n = 4). We found that RANKL stimulation induced a significant increase in survival (79.5 ± 1% live cells; n = 4; p < 0.05). Addition of OPG was able to significantly antagonize the action of RANKL (51.4 ± 6.1% live cells; n = 3; p < 0.05), whereas OPG alone was without effect (60.4 ± 2.3% live cells; n = 4; p > 0.05), suggesting that RANKL is not active in this system at a basal level (Fig. 2,C). Used in conjunction with other DC survival factors (GM-CSF, TGF-β1, and TNF-α), RANKL stimulation was also found to significantly increase cell survival (p < 0.05) in 7 days cultures, above and over levels observed in the control, cytokine free, and supplemented (CS) media (cytokine free = 19.5 ± 4.6%; CS = 42.5 ± 5.9%; CS plus RANKL = 61.9 ± 5.3%; Mean viable cells; n = 3) (Fig. 2 D).

FIGURE 2.

RANKL sustains LCL-DC survival in vitro. A, LCL-DC express RANK. LCL-DC were generated as described (21 ) and cell membrane expression of RANK was assessed by flow cytometry. A total of 72% LCL-DC expressed membrane RANK. B, Cell survival was assessed in RANKL stimulated LCL-DC. Cells were washed of differentiating medium and cultured for 48 h in the presence of RANKL (50 ng/ml) and/or OPG (1 μg/ml) as indicated. The cells were labeled with Annexin V and PI and specific labeling was detected by flow cytometry (insets show control staining). C, RANKL sustains LCL-DC survival. Percentage of live (lower left quadrant), apoptotic (lower right), and dead (upper right) cells as visualized in B were calculated. RANKL stimulation induced a significant increase in the percentage of live cells and decrease in the percentage of dead and apoptotic cells. Results are expressed as mean % ± SE; (*) p < 0.05; n = 4; two-way ANOVA with Bonferroni post tests. D, RANKL in conjunction with GM-CSF, TGF-β1, and TNF-α promoted significantly increased cell survival in 7 days cultures (results expressed as percentage of viable cells ± SE; (*) p < 0.05 and (**) p < 0.001; n = 3; one-way ANOVA with Newman-Keuls post tests).

FIGURE 2.

RANKL sustains LCL-DC survival in vitro. A, LCL-DC express RANK. LCL-DC were generated as described (21 ) and cell membrane expression of RANK was assessed by flow cytometry. A total of 72% LCL-DC expressed membrane RANK. B, Cell survival was assessed in RANKL stimulated LCL-DC. Cells were washed of differentiating medium and cultured for 48 h in the presence of RANKL (50 ng/ml) and/or OPG (1 μg/ml) as indicated. The cells were labeled with Annexin V and PI and specific labeling was detected by flow cytometry (insets show control staining). C, RANKL sustains LCL-DC survival. Percentage of live (lower left quadrant), apoptotic (lower right), and dead (upper right) cells as visualized in B were calculated. RANKL stimulation induced a significant increase in the percentage of live cells and decrease in the percentage of dead and apoptotic cells. Results are expressed as mean % ± SE; (*) p < 0.05; n = 4; two-way ANOVA with Bonferroni post tests. D, RANKL in conjunction with GM-CSF, TGF-β1, and TNF-α promoted significantly increased cell survival in 7 days cultures (results expressed as percentage of viable cells ± SE; (*) p < 0.05 and (**) p < 0.001; n = 3; one-way ANOVA with Newman-Keuls post tests).

Close modal

To explore the role of RANKL in vivo, LC numbers in RANKL KO mice (22) and WT controls were compared. Epidermal sheets were obtained from mice deficient in RANKL and littermate controls and their LC enumerated in epidermal sheet preparations. Counts were taken from six randomly chosen sites in each ear and means calculated. RANKL KO mice displayed a 42% (n = 9; p = 0.0007) decrease in LC numbers (mean = 507.1 ± 77.2 per mm2; n = 9) compared with WT controls (mean = 873.6 ± 41.6 per mm2; n = 9) (Fig. 3,A). The morphology of the remaining LC appeared normal (Fig. 3 B).

FIGURE 3.

Epidermal LC numbers are decreased in RANKL KO mice. LC were enumerated in epidermal sheets derived from RANKL KO and littermate controls. A, RANKL KO mice had significantly fewer LC (507.1 ± 77.2 per mm2; n = 9) when compared with controls (873.6 ± 41.6 per mm2; n = 9). B, Immunofluorescent image of one representative sample from each group. We found LC morphology to be preserved in RANKL KO mice.

FIGURE 3.

Epidermal LC numbers are decreased in RANKL KO mice. LC were enumerated in epidermal sheets derived from RANKL KO and littermate controls. A, RANKL KO mice had significantly fewer LC (507.1 ± 77.2 per mm2; n = 9) when compared with controls (873.6 ± 41.6 per mm2; n = 9). B, Immunofluorescent image of one representative sample from each group. We found LC morphology to be preserved in RANKL KO mice.

Close modal

To determine the mechanism responsible for the reduction in LC numbers in RANKL-deficient mice, KO and control mice were exposed to the nucleotide analog BrdU and its incorporation in epidermal populations was analyzed by flow cytometry. To prevent the results from being influenced by the difference in BrdU incorporation observed in IA-IELow/− KC, the studied population was restricted to IA-IEHigh/Langerin+ cells (Fig. 4,A, gate R1). In control mice, 38.26 ± 4.6% (n = 5) of LC had incorporated BrdU at 7 days and had, therefore, divided, in keeping with previous findings (26). However, only 26.62 ± 2.6% (n = 4) of LC from KO animals were positive for BrdU, indicating significantly (p < 0.02) decreased turnover (Fig. 4 B). Moreover, and in keeping with the observed decrease in LC numbers, the overall ratio of LC within epidermal suspensions was assessed and was found reduced in KO samples (WT = 0.56 ± 1.2% and KO = 0.29 ± 0.03%; n = 4). Though the percentages of IA-IEBright LC represents less than the usual 1–3% of LC found in epidermis, which may be explained by the lengthy treatment required to label for surface markers and nuclear BrdU simultaneously, both sample sets were treated identically, the comparison being, therefore, valid.

FIGURE 4.

LC turnover is decreased in RANKL KO mice. A, RANKL KO mice and WT littermate controls were labeled in vivo with BrdU as described. LC were isolated from trunkal skin and identified by flow cytometry as IA-IEHi/Langerin+ cells (gate R1). The percentage of BrdU+ LC (gate R2) was measured for each animal. B, Comparison of BrdU incorporation in LC from WT controls and RANKL KO revealed significantly decreased LC turnover in KO mice (mean % ± SE are 38.26 ± 4.6% and 26.62 ± 2.6%, respectively; p < 0.02; n = 5 and n = 4; paired two-tailed t test).

FIGURE 4.

LC turnover is decreased in RANKL KO mice. A, RANKL KO mice and WT littermate controls were labeled in vivo with BrdU as described. LC were isolated from trunkal skin and identified by flow cytometry as IA-IEHi/Langerin+ cells (gate R1). The percentage of BrdU+ LC (gate R2) was measured for each animal. B, Comparison of BrdU incorporation in LC from WT controls and RANKL KO revealed significantly decreased LC turnover in KO mice (mean % ± SE are 38.26 ± 4.6% and 26.62 ± 2.6%, respectively; p < 0.02; n = 5 and n = 4; paired two-tailed t test).

Close modal

A decrease in LC numbers could also result from increased LC death in situ. We, therefore, assessed LC apoptosis by the TUNEL technique in freshly isolated epidermal cells. No significant difference was observed between WT controls and KO animals (Fig. 5).

FIGURE 5.

LC from RANKL KO mice do not display increased in situ apoptosis. A, Epidermal cell suspensions from WT littermate controls and RANKL KO mice were analyzed by flow cytometry following IA-IE and TUNEL labeling. Only few IA-IE+/TUNEL+ cells (R2) were observed in WT controls and KO animals, though TUNEL staining was clearly detected in Dnase-treated WT positive controls. B, No significant difference in the percentage of TUNEL+ LC was observed between WT controls (11.0 ± 5.0%) and RANKL KO mice (10.5 ± 5.5%). Results are expressed as mean % ± SE.

FIGURE 5.

LC from RANKL KO mice do not display increased in situ apoptosis. A, Epidermal cell suspensions from WT littermate controls and RANKL KO mice were analyzed by flow cytometry following IA-IE and TUNEL labeling. Only few IA-IE+/TUNEL+ cells (R2) were observed in WT controls and KO animals, though TUNEL staining was clearly detected in Dnase-treated WT positive controls. B, No significant difference in the percentage of TUNEL+ LC was observed between WT controls (11.0 ± 5.0%) and RANKL KO mice (10.5 ± 5.5%). Results are expressed as mean % ± SE.

Close modal

This study shows for the first time that the TNF superfamily molecule RANKL is a key regulator of epidermal LC proliferation and survival. There is close spatial apposition of cells expressing RANK and RANKL within epidermis, with KC expressing RANKL in a differentiation-dependent fashion and cell surface expression of RANK by all epidermal LC. Functionally, RANKL was able to sustain LC survival in vitro and mice deficient in RANKL had markedly reduced epidermal LC numbers as a consequence of a proliferation defect. It has previously been show that CD11c+ splenic DC populations did not suffer such disruptions in RANK- and RANKL-deficient mice (8, 9). In this context, our data, therefore, highlight differences between the mechanisms and molecular signals that regulate the ontogeny and homeostasis of LC and other peripheral DC subtypes.

Our study extends recent work that described expression of RANK and RANKL in human epidermis (10). Our data clearly indicate that expression of RANKL is not uniform throughout the epidermis, but that maximal levels are present in the suprabasal compartment, precisely where LC reside. This echoes observations made in prostate epithelium where RANKL expression was found increased in luminal cells compared with basal cells (27). The mechanism underlying this regulated expression is unidentified. However, the pattern of expression of RANKL closely mirrors that of the epidermal calcium gradient, a key determinant of RANKL expression (28, 29, 30) and epidermal differentiation (31).

Similarly to the effect observed on other DC subtypes, RANKL sustained the viability of immature LC in a short term in vitro model of apoptosis (induced by cytokine removal) and enhanced long term survival when added to CS medium (6). Consistent with this, RANKL-deficient mice displayed a marked decrease in epidermal LC density, which appeared to be due to a defect in LC proliferation. Though no increase in LC apoptosis could be detected by TUNEL in RANKL-deficient epidermis, increased cell death cannot be excluded as cells may migrate out of epidermis before undergoing apoptosis. Interestingly, transgenic overexpression of RANKL expression in epidermis using a keratin 14 promoter system did not affect epidermal LC numbers (10). This paradox may be explained by the supposition that the consequences of RANKL over-activity are carefully controlled in the epidermal microenvironment, either by regulation of cell surface expression of RANK, by increased production of OPG, or by changes in downstream signaling pathways. Moreover, we have recently shown that in other K14 transgenic mice, there is complex regulation of the consequences of transgene-driven secreted cytokine release (32).

LC in the steady state have been shown to be long lived (23, 26, 33) in the epidermis and to renew in situ (26, 34), though the molecular mechanisms involved have not yet been described. Our data suggest that KC-derived RANKL is a likely candidate for this role, although other factors cannot be ruled out. Such crosstalk between KC and LC has long been suspected, perhaps most notably in the regulation by IL-1, TNF, and IL-18 of LC migration from epidermis following contact with Ag (35). In the dynamic context of inflammation, LC depletion, and epidermis repopulation, the mechanisms involved have been shown to differ from those involved in the steady-state maintenance of epidermal LC (34). Though the investigation of a potential role for RANKL in this context would be of prime interest, such a study would need to overcome the technical challenges inherent to such long term experiments in short lived animals such as RANKL−/− mice.

Pharmacological manipulation of RANKL/RANK interactions using recombinant OPG or peptide inhibitors is currently undergoing evaluation in the management of conditions such as osteoporosis (36). Our data suggest that such manipulations may be accompanied by an effect on immunocytes in the skin, possibly with the consequence of decreasing epidermal LC longevity. Such an effect may well have useful therapeutic consequences in the treatment of inflammatory skin diseases. Unfortunately, the current resources of our laboratory did not allow us to pursue this further, and neither the reconstitution of RANKL in our system nor a pharmacological manipulation of RANKL levels in healthy animals could be attempted.

Our findings add to the increasing number of roles for RANKL in epithelial and DC biology. Although RANK expression is generally induced in DC following CD40L stimulation and maturation (4), immature CD14+ dermal DC precursors have been shown to coexpress RANK and RANKL (7). More recently, KC-derived RANKL was shown to regulate LC activation and UV-induced immunosuppression via an effect on regulatory T cell numbers (10). Moreover, RANKL expression has been described in other epithelia, including mammary glands, lymph nodes, thymus, and thyroid (18, 37, 38, 39), and its role in regulating DC homeostasis and function in these organs should be investigated.

We thank Wing-Hong Kwan for technical help and the Bourg-La-Reine maternity ward for providing umbilical cord blood.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by the Chelsea and Westminster Hospital Special Trustees and Department of Health via the National Institute for Health Research comprehensive Biomedical Research Centre award to Guy’s and St. Thomas’ National Health Service Foundation Trust in partnership with King’s College London. Financial support was also provided by Institut National de la Santé et de la Recherche Médicale, Centre National de la Recherche Scientifique, and Association pour la Recherche sur le Cancer.

2

J.-B.O.B. designed and performed the research and drafted the manuscript, M.B. performed the research, C.B. provided reagents, C.G.M. designed the research and drafted the manuscript, and R.W.G. designed the research and drafted the manuscript.

4

Abbreviations used in this paper: LC, Langerhans cell; DC, dendritic cell; KC, keratinocyte; LCL-DC, LC-like DC; OPG, Osteoprotegerin; RANK, receptor activator of NF-κB; KO, knockout; PI, propidium iodide; CS, cytokine supplemented; WT, wild type.

1
Valladeau, J., O. Ravel, C. Dezutter-Dambuyant, K. Moore, M. Kleijmeer, Y. Liu, V. Duvert-Frances, C. Vincent, D. Schmitt, J. Davoust.
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
-81.
2
Tang, A., M. Amagai, L. G. Granger, J. R. Stanley, M. C. Udey.
1993
. Adhesion of epidermal Langerhans cells to keratinocytes mediated by E-cadherin.
Nature
361
:
82
-85.
3
Ginhoux, F., F. Tacke, V. Angeli, M. Bogunovic, M. Loubeau, X. M. Dai, E. R. Stanley, G. J. Randolph, M. Merad.
2006
. Langerhans cells arise from monocytes in vivo.
Nat. Immunol.
7
:
265
-273.
4
Anderson, D. M., E. Maraskovsky, W. L. Billingsley, W. C. Dougall, M. E. Tometsko, E. R. Roux, M. C. Teepe, R. F. DuBose, D. Cosman, L. Galibert.
1997
. A homologue of the TNF receptor and its ligand enhance T-cell growth and dendritic-cell function.
Nature
390
:
175
-179.
5
Josien, R., B. R. Wong, H. L. Li, R. M. Steinman, Y. Choi.
1999
. TRANCE, a TNF family member, is differentially expressed on T cell subsets and induces cytokine production in dendritic cells.
J. Immunol.
162
:
2562
-2568.
6
Wong, B. R., R. Josien, S. Y. Lee, B. Sauter, H. L. Li, R. M. Steinman, Y. Choi.
1997
. TRANCE (tumor necrosis factor [TNF]-related activation-induced cytokine), a new TNF family member predominantly expressed in T cells, is a dendritic cell-specific survival factor.
J. Exp. Med.
186
:
2075
-2080.
7
Cremer, I., M. C. Dieu-Nosjean, S. Maréchal, C. Dezutter-Dambuyant, S. Goddard, D. Adams, N. Winetr, C. Menetrier-Caux, C. Sautès-Fridman, W. H. Fridman, C. G. Mueller.
2002
. Long-lived immature dendritic cells mediated by TRANCE-RANK interaction.
Blood
100
:
3646
-3655.
8
Kong, Y. Y., H. Yoshida, I. Sarosi, H. L. Tan, E. Timms, C. Capparelli, S. Morony, A. J. Oliveira-dos-Santos, G. Van, A. Itie, et al
1999
. OPGL is a key regulator of osteoclastogenesis, lymphocyte development and lymph-node organogenesis.
Nature
397
:
315
-323.
9
Dougall, W. C., M. Glaccum, K. Charrier, K. Rohrbach, K. Brasel, T. De Smedt, E. Daro, J. Smith, M. E. Tometsko, C. R. Maliszewski, et al
1999
. RANK is essential for osteoclast and lymph node development.
Genes Dev.
13
:
2412
-2424.
10
Loser, K., A. Mehling, S. Loeser, J. Apelt, A. Kuhn, S. Grabbe, T. Schwarz, J. M. Penninger, S. Beissert.
2007
. Epidermal RANKL controls regulatory T-cell numbers via activation of dendritic cells.
Nat. Med.
12
:
1372
-1379.
11
Bachmann, M. F., B. R. Wong, R. Josien, R. M. Steinman, A. Oxenius, Y. Choi.
1999
. TRANCE, a tumor necrosis factor family member critical for CD40 ligand-independent T helper cell activation.
J. Exp. Med.
189
:
1025
-1031.
12
Fata, J. E., Y. Y. Kong, J. Li, T. Sasaki, J. Irie-Sasaki, R. A. Moorehead, R. Elliott, S. Scully, E. B. Voura, D. L. Lacey.
2000
. The osteoclast differentiation factor osteoprotegerin-ligand is essential for mammary gland development.
Cell
103
:
41
-50.
13
Simonet, W. S., D. L. Lacey, C. R. Dunstan, M. Kelley, M.-S. Chang, R. Luthy, H. Q. Nguyen, S. Wooden, L. Bennett, T. Boone.
1997
. Osteoprotegerin: a novel secreted protein involved in the regulation of bone density.
Cell
89
:
309
-319.
14
Bucay, N., I. Sarosi, C. R. Dunstan, S. Morony, J. Tarpley, C. Capparelli, S. Scully, H. L. Tan, W. Xu, D. L. Lacey, et al
1998
. osteoprotegerin-deficient mice develop early onset osteoporosis and arterial calcification.
Genes Dev.
12
:
1260
-1268.
15
Josien, R., H. L. Li, E. Ingulli, S. Sarma, B. Wong, M. Vologodskaia, R. M. Steinman, Y. Choi.
2000
. TRANCE, a tumor necrosis factor family member, enhances the longevity and adjuvant properties of dendritic cells in vivo.
J. Exp. Med.
3
:
495
-501.
16
Wong, B., R. Josien, Y. Choi.
1999
. TRANCE is a TNF family member that regulates dendritic cell and osteoclast function.
J. Leukocyte Biol.
65
:
715
-724.
17
Kim, N. S., H. J. Kim, B. K. Koo, M. C. Kwon, Y. W. Kim, Y. Cho, Y. Yokota, J. M. Penninger, Y. Y. Kong.
2006
. Receptor activator of NF-κB ligand regulates the proliferation of mammary epithelial cells via Id2.
Mol. Cell. Biol.
26
:
1002
-1013.
18
Lee, H. W., B. S. Kim, H. J. Kim, C. W. Lee, H. J. Yoo, J. B. Kim, S. Yoon.
2005
. Upregulation of receptor activator of nuclear factor-κB ligand expression in the thymic subcapsular, paraseptal, perivascular, and medullary epithelial cells during thymus regeneration.
Histochem. Cell Biol.
123
:
491
-500.
19
Antonopoulos, C., M. Cumberbatch, R. J. Dearman, R. J. Daniel, I. Kimber, R. W. Groves.
2001
. Functional caspase-1 is required for Langerhans cell migration and optimal contact sensitization in mice.
J. Immunol.
166
:
3672
-3677.
20
Mee, J. B., Y. Alam, R. W. Groves.
2000
. Human keratinocytes constitutively produce but do not process interleukin-18.
Br. J. Dermatol.
143
:
330
-336.
21
Barbaroux, J. B., W. H. Kwan, J. P. Allam, N. Novak, T. Bieber, W. H. Fridman, R. Groves, C. G. Mueller.
2006
. Tumor necrosis factor-α- and IL-4-independent development of Langerhans cell-like dendritic cells from M-CSF-conditioned precursors.
J. Invest. Dermatol.
126
:
114
-120.
22
Kim, N., P. R. Odgren, D. K. Kim, S. C. Marks, Jr, Y. Choi.
2000
. Diverse roles of the tumor necrosis factor family member TRANCE in skeletal physiology revealed by TRANCE deficiency and partial rescue by a lymphocyte-expressed TRANCE transgene.
Proc. Natl. Acad. Sci. USA
97
:
10905
-10910.
23
Kamath, A. T., S. Henri, F. Battye, D. F. Tough, K. Shortman.
2002
. Developmental kinetics and lifespan of dendritic cells in mouse lymphoid organs.
Blood
100
:
1734
-1741.
24
Jones, P. H., F. M. Watt.
1993
. Separation of human epidermal stem cells from transit amplifying cells on the basis of differences in integrin function and expression.
Cell
73
:
713
-724.
25
Zanet, J., S. Pibre, C. Jacquet, A. Ramirez, I. M. de Alboran, A. Gandarillas.
2005
. Endogenous Myc controls mammalian epidermal cell size, hyperproliferation, endoreplication and stem cell amplification.
J. Cell Sci.
118
:
1693
-1704.
26
Czernielewski, J. M., M. Demarchez.
1987
. Further evidence for the self-reproducing capacity of Langerhans cells in human skin.
J. Invest. Dermatol.
88
:
17
-20.
27
Brown, J. M., E. Corey, Z. D. Lee, L. D. True, T. J. Yun, M. Tondravi, R. L. Vessella.
2001
. Osteoprotegerin and rank ligand expression in prostate cancer.
Urology
57
:
611
-616.
28
Takami, M., N. Takahashi, N. Udagawa, C. Miyaura, K. Suda, J. T. Woo, T. J. Martin, K. Nagai, T. Suda.
2000
. Intracellular calcium and protein kinase C mediate expression of receptor activator of nuclear factor-κB ligand and osteoprotegerin in osteoblasts.
Endocrinology
141
:
4711
-4719.
29
Takahashi, E., H. Mukohyama, K. Aoki, W. R. Duarte, U. H. Lerner, K. Ohya, K. Omura, S. Kasugai.
2002
. High extracellular calcium affects osteoclastogenesis in mouse bone marrow cell culture.
J. Med. Dent. Sci.
49
:
109
-120.
30
Hikita, A., Y. Kadono, H. Chikuda, A. Fukuda, H. Wakeyama, H. Yasuda, K. Nakamura, H. Oda, T. Miyazaki, S. Tanaka.
2005
. Identification of an alternatively spliced variant of Ca2+-promoted Ras inactivator as a possible regulator of RANKL shedding.
J. Biol. Chem.
280
:
41700
-41706.
31
Bikle, D. D., D. Ng, C. L. Tu, Y. Oda, Z. Xie.
2001
. Calcium- and vitamin D-regulated keratinocyte differentiation.
Mol. Cell. Endocrinol.
177
:
161
-171.
32
Mee, J. B., C. Antonopoulos, S. Poole, T. S. Kupper, R. W. Groves.
2005
. Counter-regulation of interleukin-1α (IL-1α) and IL-1 receptor antagonist in murine keratinocytes.
J. Invest. Dermatol.
124
:
1267
-1274.
33
Holt, P. G., S. Haining, D. J. Nelson, J. D. Sedgwick.
1994
. Origin and steady-state turnover of class II MHC-bearing dendritic cells in the epithelium of the conducting airways.
J. Immunol.
153
:
256
-261.
34
Merad, M., M. G. Manz, H. Karsunky, A. Wagers, W. Peters, I. Charo, I. L. Weissman, J. G. Cyster, E. G. Engleman.
2002
. Langerhans cells renew in the skin throughout life under steady-state conditions.
Nat. Immunol.
3
:
1135
-1141.
35
Griffiths, C. E. M., R. J. Dearman, M. Cumberbatch, I. Kimber.
2005
. Cytokines and Langerhans cell mobilisation in mouse and man.
Cytokine
32
:
67
-70.
36
Lewiecki, E. M..
2006
. RANK ligand inhibition with denosumab for the management of osteoporosis.
Expert. Opin. Biol. Ther.
6
:
1041
-1050.
37
Kim, D., R. E. Mebius, J. D. MacMicking, S. Jung, T. Cupedo, Y. Castellanos, J. Rho, B. R. Wong, R. Josien, N. Kim, et al
2000
. Regulation of peripheral lymph node genesis by the tumor necrosis factor family member TRANCE.
J. Exp. Med.
192
:
1467
-1478.
38
Hofbauer, L., S. Kluger, C. Kühne, C. Dunstan, A. Burchet, M. Schoppet, A. Zielke, A. Heufelder.
2002
. Detection and characterization of RANK ligand and osteoprotegerin in the thyroid gland.
J. Cell. Biochem.
86
:
642
-650.
39
Van Poznak, C., S. S. Cross, M. Saggese, C. Hudis, K. S. Panageas, L. Norton, R. E. Coleman, I. Holen.
2006
. Expression of osteoprotegerin (OPG), TNF related apoptosis inducing ligand (TRAIL), and receptor activator of nuclear factor κB ligand (RANKL) in human breast tumours.
J. Clin. Pathol.
59
:
56
-63.