Skip to main content
NCBI home page
The Journal of Clinical Endocrinology and Metabolism logoLink to The Journal of Clinical Endocrinology and Metabolism
. 2009 Jan 13;94(4):1288–1294. doi: 10.1210/jc.2008-1053

Keratinocyte Growth Inhibition through the Modification of Wnt Signaling by Androgen in Balding Dermal Papilla Cells

Tomoko Kitagawa 1, Ken-Ichi Matsuda 1, Shigeki Inui 1, Hideya Takenaka 1, Norito Katoh 1, Satoshi Itami 1, Saburo Kishimoto 1, Mitsuhiro Kawata 1
PMCID: PMC2682470  PMID: 19141591

Abstract

Context/Objective: Androgen induces androgenetic alopecia (AGA), which has a regressive effect on hair growth from the frontal region of the scalp. Conversely, Wnt proteins are known to positively affect mammalian hair growth. We hypothesized that androgen reduces hair growth via an interaction with the Wnt signaling system. The objective of this study was to investigate the effect of androgen on Wnt signaling in dermal papilla (DP) cells.

Design: The effect of androgen and Wnt3a on keratinocyte proliferation was measured by use of a coculture system consisting of DP cells and keratinocytes. The molecular mechanisms of androgen and Wnt pathway interactions in DP cells were examined by analyzing the expression, intracellular localization, and activity of the androgen receptor (AR) and also downstream Wnt signaling molecules.

Results: Wnt3a-dependent keratinocyte growth was suppressed by the addition of dihydrotestosterone in coculture with DP cells that were derived from AGA patients, but growth was not suppressed in coculture with DP cells from non-AGA males. Whereas DP cells from both scalp regions expressed AR protein, the expression levels of AR and cotranslocation with β-catenin, a downstream Wnt signaling molecule, were higher in DP cells of AGA patients than in DP cells from non-AGA males. In addition, significant suppression of Wnt signal-mediated transcription in response to dihydrotestosterone treatment was observed only in DP cells from AGA patients.

Conclusion: These results suggest that Wnt signaling in DP cells is regulated by androgen and this regulation plays a pivotal role in androgen’s action on hair growth.

Wnt signaling in DP cells is regulated by androgen, and this regulation plays a pivotal role in androgen’s action on hair growth.

Androgen has region-specific oppositional functions for hair growth: activation in beard, axilla, and pubic area vs. regression in frontal to vertex scalp regions of individuals with an appropriate genetic background, which induces androgenetic alopecia (AGA) (1). Androgen receptor (AR) is expressed in dermal papilla (DP) cells but not in follicular epithelial (FE) cells except for the skin of the pubic area (2,3,4). AR is a member of the nuclear receptor superfamily and functions as a ligand-dependent transcription factor. Androgen binds to AR, and then this ligand-receptor complex translocates to the nucleus to regulate the transcription of downstream molecules (5). When FE cells were cocultured with DP cells of frontal AGA scalp, androgen increased the secretion of TGFβ and subsequently reduced the growth of FE cells (6), whereas this growth reduction was not observed in a coculture with DP cells of non-AGA patients. Conversely, androgen increased the secretion of IGF-I from DP cells (7) and up-regulated the number of FE cells cocultured with beard DP cells. However, hair epithelial-mesenchymal interactions, including the region-specific paradoxical effect of androgen and its molecular mechanism of AGA, are still not fully understood.

Wnt proteins are secreted signal molecules, and their signaling pathway plays a key role in various biological phenomena including morphogenesis and cell proliferation (8,9,10). It has been found that hair follicle development and regeneration are regulated by Wnt (11,12,13). As a canonical Wnt-β-catenin signaling pathway, Wnt proteins bind to their membrane receptors, Frizzled proteins, and initiate a cascade that results in the accumulation of cytoplasmic β-catenin, followed by the translocation of β-catenin to the nucleus in which it interacts with the N terminus of a member of the lymphoid enhancer factor (Lef)/T cell factor (Tcf) family of transcription factors (14,15). Wnt3a is secreted from hair epithelium and acts in autocrine and paracrine fashion, and it has been demonstrated that Wnt3a maintains anagen gene expression in DP cells (16) and mediates hair-inductive activity in an organ culture (11). This Wnt3a-mediated hair growth might depend on the canonical Wnt signaling pathway because deletion of β-catenin (8) or the Lef1 (17) gene resulted in hair loss in mice.

Recently interaction of AR and β-catenin, and resulting activation or inhibition of AR- or Lef/Tcf-mediated transcription was demonstrated in several cell lines, such as neuronal cells and prostate cancer cells (18,19,20). In addition, interaction between β-catenin and coactivators of AR has also been reported (20,21). These findings strongly indicate that there is functional cross talk between AR and Wnt signaling pathways in target tissues. Thus, we hypothesized that androgen regulates hair growth through the modulation of the Wnt signaling.

The present study shows the effect of androgen on Wnt-mediated growth of keratinocytes in a coculture system of a keratinocyte cell line and human DP cells derived from males with AGA and males without AGA. In addition, we investigated the effect of androgen on the AR and downstream molecules of the Wnt3a signal pathway in DP cells.

Subjects and Methods

Isolation and culture of human DP cells

Skin specimens were obtained at plastic surgery and dermatological surgery, and DP cells were isolated from the skin samples of 69- and 70-yr-old men with AGA in frontal scalp, and six men without AGA in occipital scalp, aged 27–70 yr (average 40 yr). The medical ethics commission of the Kyoto Prefectural University of Medicine Graduate School of Medical Science, Japan, approved the described studies. Informed written consent was obtained from the patients. The mid-lower parts of hair follicles were cut off from the side of subcutis with scissors. These hair follicles were mixed with digestive solution (collagenase 1 mg/ml; Wako, Osaka, Japan), and incubated at 37C for 5–6 h (22). DMEM (Invitrogen, Gaithersburg, MD) supplemented with 10% fetal bovine serum (FBS), penicillin (100 U/ml), streptomycin (100 μg/ml), and amphotericin-B (100 μg/ml) were added. Sample was sucked and blown with a pipette until the DP cells emerged freely. DP cells were cultured in the medium at 37 C under a humidified atmosphere of 95% O2-5% CO2. Cells were subcultured after 0.025% trypsin treatment (usually 2 wk after beginning primary cell cultures). For the experiment, DP cells in the third to sixth passage of subculture were used.

Cell lines

A human immortalized DP cell line, AGA40M, which is a Simian virus 40-transformed DP cell clone from the frontal scalp of a 40-yr-old man with AGA was maintained in the same medium as the primary culture of DP cells (23). An immortalized human keratinocyte cell line, HaCaT (provided from Dr. Fusenigh), was cultured at 37 C in DMEM supplemented with 10% FBS. It has been reported that HaCaT cells exhibit altered and unlimited growth potential, yet similarly to normal keratinocytes, HaCaT cells do reform an orderly structured and differentiated epidermal tissue when transplanted onto nude mice (24).

Growth analysis of DP cells in the coculture system

DP cells (1 × 104 cells/cm2) were seeded in the upper compartment of Transwell culture dishes (12 mm diameter, 0.4 μm pore size; Corning Inc., Corning, NY) and incubated for 24 h. Then the cells were cocultured with HaCaT cells, which were seeded (8 × 103 cells/cm2) in the lower compartment of the Transwell culture dishes on the day before the coculture in FBS-free DMEM. After 2 d, the medium was changed and 10−9 m dihydrotestosterone (DHT) (Wako) and/or 20 ng/100 μl recombinant mouse Wnt3a (R&D Systems, Minneapolis, MN) were added to the culture. After 4 d, we counted the number of HaCaT cells. The statistical significance of differences in mean values was assessed using one-way ANOVA followed with Tukey-Kramer test.

Transcription assay

The mammalian expression system Tcf-binding luciferase reporter plasmid, pGL3-OT, was kindly provided by Dr. B. Vogelstein (Johns Hopkins University School of Medicine, Baltimore, MD). This reporter plasmid contains a trimerized Tcf optimal promoter sequence upstream of the luciferase gene (25). The mouse mammary tumor virus (MMTV)-luciferase reporter plasmid contains the MMTV-long terminal repeat coupled to a luciferase reporter gene. A β-actin promoter-driven β-galactosidase expression construct was used as an internal control. DP cells were transfected in 35-mm dishes with 1 μg of the reporter plasmid DNA using FuGENE6 (6). After 24 h, the medium was changed to OPTI-MEM (Invitrogen). Cells were treated with 10−9 m DHT and/or 20 ng/100 μl Wnt3a for 48 h. Luciferase activities in the cell lysates were assayed using a luciferase assay system Pica Gene (Toyo Inki, Tokyo, Japan). β-Galactosidase activities were assayed using the Luminescent β-galactosidase detection kit (BD Biosciences, Palo Alto, CA). Average luciferase activity values from triplicate experiments were normalized relative to β-galactosidase activities from the same supernatant preparations.

Double-labeling immunocytochemical detection

DP cells were plated on a poly-l-lysine-coated, 35-mm glass-bottom dish (Matsunami Glass, Inc., Ltd., Kishiwada, Japan), and cultured in OPTI-MEM for 24 h. Cells were untreated or treated with 10−9 m DHT and/or 20 ng/100 μl Wnt3a for 10 min, 1 h, or 2 h. Cells were fixed with 4% paraformaldehyde in PBS, permeabilized with 2 ml of 2% BSA in 0.3% Triton X-100 in PBS for 60 min at room temperature, and then incubated together with rabbit polyclonal anti-AR IgG (1:200 dilution; Santa Cruz Biotechnology, Santa Cruz, CA) and mouse monoclonal anti-β-catenin IgG antibodies (1:200 dilution; Zymed, Laboratories Inc., South San Francisco, CA) overnight at 4 C. Cells were washed with PBS and incubated together with Alexa Fluor 488-labeled antirabbit IgG (1:1000 dilution; Invitrogen, Eugene, OR) and Alexa Fluor 546-labeled antimouse IgG (1:1000 dilution; Invitrogen) for 2 h at room temperature. Images were collected with a confocal laser microscope, LSM510META (Carl Zeiss Co. Ltd., Jena, Germany). Images were analyzed using the confocal laser microscope software. To estimate the amount of nuclear colocalization of AR and β-catenin, the ratio of the dot number of the immunofluorescence signal of nuclear AR, which overlapped with the immunofluorescence signal of β-catenin, to the dot number of the entire immunofluorescence signal of AR in the cell was calculated.

Western blot

DP cells were solubilized in sodium dodecyl sulfate (SDS) sample buffer. These cell lysates were separated by 7.5% SDS-PAGE and transferred to polyvinylidene difluoride membranes (Immunobilon-P; Millipore Corp., Bedford, MA) using a semidry blotting apparatus (Transblot-SD; Bio-Rad Laboratories, Inc., Hercules, CA). The blotted membrane was blocked with 3% albumin in 0.02% Tris-HCl (pH 8.0), 0.05% Tween 20, and 150 mm HCl for 1 h. Immunoblotting was per-formed with rabbit anti-AR (1:200 dilution; Santa Cruz Biotechnology), rabbit anti-Tcf-4 (1:200 dilution; Santa Cruz Biotechnology), and mouse anti-β-catenin (1:200 dilution; Zymed). Detection of immunoreactive bands was achieved with alkaline phosphatase-conjugated goat antirabbit IgG secondary antibody (1:1000 dilution; Chemicon, Temecula, CA) or alkaline phosphatase-conjugated antimouse IgG secondary antibody (1:1000 dilution; Chemicon). The membranes were incubated with stripping buffer [100 nm 2-mercaptoethanol, 2% SDS, 62.5 mm Tris-HCl (pH 6.7)] for 30 min at room temperature and blotted again with mouse anti-glyceraldehyde-3-phosphate dehydrogenase antibody (1:5000 dilution; Chemicon) as internal control. The blot was scanned with a densitometer and peak areas were integrated by the computer software program, ImageJ (Wayne Rasband, National Institutes of Health, Bethesda, MD).

Immunoprecipitation

DP cells were treated with 10−9 m DHT and/or 20 ng per 100 μl Wnt3a for 2 h, and whole cell extracts of the cells were prepared using a lysis buffer containing 1% Tris-buffered saline, 1% Triton X-100, and 1% protease inhibitor cocktail (Nacalai Tesque, Kyoto, Japan). Cell lysates were passed several times through a 28-gauge needle to disrupt the nuclei. The lysates (500 μl) were incubated with anti-AR antibody (Santa Cruz Biotechnology) for 5 h at 4 C and then incubated with protein G-Sepharose (GE Healthcare, Uppsala, Sweden) overnight. After washing with 0.5 ml of lysis buffer four times, the pellets were resuspended in SDS sample buffer and boiled for 5 min. SDS-PAGE and immunoblotting were performed as described above for Western blot.

Results

Effect of DHT on Wnt3a-dependent growth of keratinocytes in coculture system

To see the effect of DHT on Wnt3a-mediated hair growth, we counted the number of HaCaT cells after coculturing with DP cells in the absence or presence of DHT and/or Wnt3a. This coculture system is illustrated in Fig. 1A.

Figure 1.

Figure 1

Open in a new tab

Androgen effect on Wnt3a-dependent growth of keratinocytes in a coculture system. A, Illustration of the coculture system. DP and HaCaT cells were seeded in upper and lower compartments of Transwell culture dishes, respectively. HaCaT cells were cocultured with DP cells of AGA40M (B), non-AGA males (n = 5) (C) in the absence of both DHT and Wnt3a (−), in the presence of DHT (10−9 m), Wnt3a (20 ng/μl) (W), or both DHT and Wnt3a (WD) for 4 d. The graph represents the relative cell number compared with −. D, The number of HaCaT cells cultured alone in the absence and presence of Wnt3a and presence of both Wnt3a and DHT was counted. Statistical significance of the differences in mean values was assessed using one-way ANOVA followed by Tukey-Kramer test. *, P < 0.05 vs. −; **, P < 0.01 vs. −; §, P < 0.05 vs. W.

When HaCaT cells were cocultured with an AGA patient-derived DP cell line, AGA40M, the number of HaCaT cells was not influenced by the addition of DHT. But the number of HaCaT cells was significantly increased in the presence of Wnt-3a, and this growth was significantly suppressed by the addition of DHT (Fig. 1B). A similar effect of DHT was detected in the coculture of the HaCaT cells with DP cells from other AGA patients (data not shown). In the coculture with DP cells of non-AGA males, the number of HaCaT cells was also increased by the addition of Wnt3a, but significant growth suppression due to DHT was not detected (Fig. 1C).

Wnt3a treatment significantly accelerated the growth of HaCaT cells in the single cell culture, but the addition of DHT did not affect growth (Fig 1D). On the other hand, the growth of DP cells was not affected by Wnt3a in single cell culture conditions (data not shown). It has been reported that androgen does not influence growth of DP cells in a single cell culture condition (26,27). These results indicate that Wnt3a-dependent keratinocyte proliferation is reduced by DHT in the coculture with AGA DP cells. This suggests that this DHT effect is not a direct action on keratinocytes but is mediated through DP cells.

Subcellular localization of AR and β-catenin in DP cells

It was reported that in neuronal cells, AR and β-catenin were cotranslocated to the nucleus with the addition of DHT (18). We used an immunofluorescence method to investigate time courses of the translocation of AR and β-catenin in DP cells that were supplemented with DHT. In both AGA and non-AGA DP cells, AR and β-catenin translocation to the nucleus was time dependent and reached plateau in 1 h (Fig. 2A); thus, we decided to carry out image analysis at 2 h after the treatment with ligand(s) when the translocation would be complete. β-Catenin was also observed in the head-cap area, which presumably corresponds to the Golgi apparatus, in approximately half of both AGA40M and non-AGA-derived DP cells.

Figure 2.

Figure 2

Open in a new tab

Localization of AR and β-catenin in DP cells. A, DP cells of AGA40M and non-AGA male were treated with DHT (10−9 m) for various times (10 min, 1 h, 2 h). The cells were fixed and labeled with anti-AR and anti-β-catenin antibodies, followed by fluorescent second antibodies. All images were captured using confocal laser-scanning microscopy. DP cells of AGA40M (B) and from other AGA males (C) and non-AGA male (D) were cultured in the absence of both DHT and Wnt3a (−), presence of DHT, presence of Wnt3a (20 ng/μl) (W), or presence of both DHT and Wnt3a (WD) for 2 h. The cells were visualized as in A. Fluorescent images shown in green are AR, whereas the images in red are β-catenin. Merged images are shown in the right panels. Bar, 10 μm.

To examine whether the differences in HaCaT cell proliferation in coculture with AGA and non-AGA male DP cells were mediated by the action of AR and β-catenin, we investigated the subcellular distribution of AR and β-catenin in the cultured DP cells after treatment with DHT and/or Wnt3a. In DP cells of AGA40M (Fig. 2B), AR was diffusely distributed in cytoplasm, and β-catenin was present in pericellular cytoplasm in the absence of DHT and Wnt3a. In the presence of DHT, AR was significantly translocated to the nucleus, whereas cytoplasmic distribution of β-catenin was not so changed. When the DP cells were treated with Wnt3a, AR was present in the cytoplasm. In contrast, β-catenin was strongly enhanced and exhibited filamentous appearance. With the addition both of DHT and Wnt3a, AR and β-catenin were distributed in the nucleus in which they were colocalized. Similar distribution patterns of AR and β-catenin were observed in DP cells from other AGA males (Fig. 2C).

Next we show that there is intracellular localization of AR and β-catenin in a ligand-free condition in DP cells of non-AGA males (Fig. 2D). Addition of either DHT or Wnt3a yielded this same intracellular localization in AGA40M or other male AGA cells. When DHT and Wnt3a were added together, colocalization of AR and β-catenin in the nucleus was sparse.

Nuclear localization of AR and β-catenin in DP cells

By using image analysis software, we examined the degree of nuclear colocalization of AR and β-catenin semiquantitatively using the ratio of merged dots in the nucleus/AR dots in cells treated with both Wnt3a and DHT (Fig. 3A). The ratios in DP cells from AGA males was higher by 2.5-fold than in DP cells of non-AGA males. These results indicate that AR and β-catenin in the nucleus was more highly colocalized in DP cells of AGA40M and other AGA males than in that of non-AGA males.

Figure 3.

Figure 3

Open in a new tab

Expression of AR in DP cells and colocalization with β-catenin. A, Subcellular distribution of AR and β-catenin in each DP cell treated with both DHT and Wnt3a was detected by immunofluorescence as shown in Fig. 2. The ratio of the dot number of immunofluorescence signal of nuclear AR that overlapped with immunofluorescence signal of β-catenin to the dot number of entire immunofluorescence signal of AR in the cell was measured by image analysis software. The graph represents means of 11–16 independent cells. Bar, Average of each group. Statistical significance of the difference in mean values was assessed using one-way ANOVA followed by Tukey-Kramer test. **, P < 0.01 vs. AGA. B, Total cell lysates of DP cells were analyzed by immunoblotting using anti-AR antibody. The blot was scanned with a densitometer and peak areas were integrated by the computer software program, ImageJ. Expression of AR was normalized to the internal control, glyceraldehyde-3-phosphate dehydrogenase. Relative expression levels were plotted. Statistical significance of the differences in mean values was assessed using one-way ANOVA followed by Tukey-Kramer test. *, P < 0.05. C, AGA40M cells and non-AGA DP cells were treated with either DHT and/or Wnt3a for 2 h. Immunoprecipitation (IP) with anti-AR antibody was carried out using each cell lysate. Precipitates were analyzed by immunoblotting with each antibody. The upper panel shows the detection of β-catenin after immunoprecipitation with anti-AR antibody and the lower panel shows the detection of AR on the same membrane. D, Presence of DHT; W, presence of Wnt3a; WD, presence of both DHT and Wnt3a; WB, western blot.

Expression of AR protein in DP cells

Immunofluorescence observation exhibited similar immunoreactivity of AR in DP cells of AGA and non-AGA in the presence of DHT. To examine whether there are quantitative differences in AR protein level in DP cells of AGA and non-AGA, we investigated expression of AR in the DP cells by immunoblotting. Specific bands for AR were detected in all DP cells, and the expression level was higher in DP cells of AGA males by 3.1-fold than in DP cells with non-AGA males (Fig. 3B).

Direct binding of β-catenin with AR in DP cells

Furthermore, to investigate whether AR and β-catenin was bound directly in the nucleus, we performed immunoprecipitation experiments (Fig. 3C) using cell lysates of DP cells of AGA40M and non-AGA male. A direct interaction of AR and β-catenin was detected even in both untreated AGA and non-AGA DP cells. In addition, the amount of AR-β-catenin complex was increased by the treatment with Wnt3a in AGA40M cells but not the non-AGA male cells. This result is consistent with the result of the subcellular localization analysis.

Change of Lef/Tcf-mediated transcriptional activity by DHT in DP cells of AGA

Because differences in nuclear translocation of β-catenin between DP cells of AGA and non-AGA may result in differential Lef/Tcf-mediated transcriptional activity, we performed luciferase reporter assay using the cell lysates of each DP cell.

DP cells were transfected with a reporter plasmid in which the luciferase gene was connected downstream to the Lef/Tcf-binding promoter. Lef/Tcf-mediated transcriptional activity was induced by treatment with Wnt3a in AGA40M cells, but this induction was significantly suppressed by the addition of DHT (Fig. 4A). The same result was observed in DP cells of other AGA male. In contrast, in DP cells of non-AGA males, the transcriptional activity was not suppressed by DHT. These results indicate that the Wnt-β-catenin signaling pathway is negatively influenced by ligand-activated AR in DP cells of AGA but not DP cells of non-AGA males.

Figure 4.

Figure 4

Open in a new tab

Lef/Tcf- and AR-mediated transcriptional activity in DP cells. A Lef/Tcf-responsive reporter (pGL3-OT) (A) or an AR-responsive reporter (MMTV-Luc) (B) and a plasmid for an internal standard, the mammalian-positive control vector (pAct-βGal), were each separately cotransfected in to DP cells. The cells were cultured in the absence of both DHT and Wnt3a (−), presence of DHT (10−9 m) (D), presence of Wnt3a (20 ng /μl) (W), or presence of both DHT and Wnt3a (WD). Fold inductions shown were calculated from luciferase activities of each sample relative to − after normalization of the transfection efficiencies with β-galactosidase activities. Data are expressed as mean ± sd from three independent experiments as relative activity normalized to β-actin-gal. Statistical significance of the differences in mean values was assessed using one-way ANOVA followed by Tukey-Kramer test. *, P < 0.01 vs. −; #, P < 0.01 vs. W. C, Expression of Tcf-4 in each DP cell was detected by immunoblotting using anti-Tcf-4 antibody.

We also examined AR-mediated transcriptional regulation. DP cells were transfected with a reporter plasmid with luciferase as a reporter gene downstream to the MMTV promoter, which is responsive to AR as well as glucocorticoid and progesterone receptors. No significant effect of Wnt3a on AR-mediated transcriptional activity was detected in both AGA and non-AGA DP cells (Fig. 4B).

It has been reported that Tcf-4 is a downstream target of AR in skeletal muscle (28); therefore, we investigated expression of Tcf-4 protein in the DP cells by immunoblotting. Both AGA and non-AGA DP cells expressed Tcf-4 protein (Fig. 4C).

Discussion

Wnt3a is secreted from hair epithelium and mediates hair growth through the canonical Wnt signaling pathway. Because functional cross talk of AR and Wnt signaling in several target tissues has been reported (15,18,19,20), we developed an in vitro coculture system to simulate in vivo conditions of simultaneous hair follicle-signaling effects of both androgen and Wnt. In the present coculture system, AGA and non-AGA DP cells showed distinct effects of androgen on growth of keratinocytes induced by Wnt3a, which correlated well with physiological and pathological conditions of human hair growth. The result suggests that modification of Wnt signaling by androgen in DP cells may be a central component of the androgen effect for hair growth including AGA.

It has been reported that type II5α-reductase catalyzes the conversion of testosterone to the more potent androgen DHT. Type II 5α-reductase is specifically localized in the beard and top and front of the scalp DP cells (29). Region-specific expression of type II 5α-reductase and AR in DP cells is speculated to be involved in beard growth at puberty and to genetically determine balding of the frontal scalp. 5α-Reductase expression is detectable in freshly microdissected DP cells. Because 5α-reductase expression is not maintained in culture (30), it is difficult to investigate the influence of 5α-reductase under in vitro conditions. Therefore, we used DHT in this experiment.

It has been proposed that one of the reasons for the region specificity of the androgen effect is the difference of AR protein expression levels in each area (2,31). Several studies demonstrated that AR protein levels in DP cells of vertex were higher than that of occipital scalp (32). On the other hand, there have been conflicting results with regard to whether AR protein levels are different in AGA and non-AGA DP cells (32). In this study, AR protein levels were higher in DP cells isolated from the frontal scalp of men with AGA than in DP cells of the occipital scalp isolated from non-AGA males. This was despite the fact that the average age of the men from whom the AGA DP cells were isolated was higher than that of the non-AGA men. Taken together, our results suggest that AR expression level is a crucial factor for androgen sensitivity of DP cells for the hair growth.

Specific reduction of Wnt3a-induced keratinocyte proliferation in the coculture with DP cells of AGA in the presence of DHT indicates that Lef/Tcf-mediated transcriptional activity is modified by liganded AR, specifically in DP cells of AGA. Pawlowski et al. (18) reported that Lef/Tcf-related transcriptional activity was suppressed by androgen in neuronal cells and that AR directly interacted with β-catenin in the cells, although the extent of AR/β-catenin interaction was increased on AR-ligand binding. The binding of AR to β-catenin appears to suppress Lef/Tcf-mediated transcription in an androgen-dependent manner (33). The same results were reported for other cell lineages such as prostate cancer cells, embryonic kidney cell, and colon adenocarcinoma (34).

Consistent with these previous results, Wnt3a-induced transcriptional activation of Lef/Tcf was influenced by treatment with DHT, and interestingly, reduction of transcription with DHT was found only in DP cells of AGA but not non-AGA males. AR-dependent transcription was not affected by Wnt3a. These results indicate again that modulation of Wnt signaling by liganded-AR is a central mechanism for androgen-mediated hair growth regulation. Thus, the identification of target genes of Lef/Tcf, which are involved in hair growth regulation, will lead to further understanding of the mechanisms related to how AGA develops.

It has been reported that liganded AR enhances nuclear translocation of β-catenin in neuronal cells and preadipocytes (18,20). Subsequent formation of AR-β-catenin-Tcf-4 binding complex and resulting inhibition of adipogenic differentiation was demonstrated in a preadipocyte cell line (20). It has also been shown in neuronal cells that AR is the only steroid hormone receptor with the ability to cotranslocate β-catenin to the nucleus (18).

Thus, we assumed that AGA-derived DP cells specifically inhibited the growth of keratinocyte and the potential of Lef/Tcf mediated transcription might be due to differences in translocation of AR and β-catenin to the nucleus. As expected, in the presence of DHT and Wnt3a, a significantly higher ratio of AR to β-catenin was cotranslocated to the nucleus in DP cells of AGA than in DP cells of non-AGA males. Considering this result together with analysis of Lef/Tcf mediated transcription, it is conceivable that nuclear interaction of liganded AR with β-catenin inhibits transcriptional activity in DP cells of AGA. In fact, direct binding of AR and β-catenin was more frequent in AGA DP cells than non-AGA DP cells as determined by the present immunoprecipitation analysis.

The reason for the differences in the ratio of the cotranslocation between DP cells of AGA and other DP cells is not clear. As mentioned above, expression levels of AR may explain these differences. Because β-catenin was reported to possess the ability to migrate into the nucleus without requiring general Ran-importin-mediated nuclear transport machinery (35), it is also possible that unknown mechanisms underlie the difference.

Acknowledgments

The authors thank Dr. N. Kaku, Dr. Y. Hirahara, Dr. K. Fujikawa, and Ms. N. Oumi for their technical assistance.

Footnotes

This work was supported by Grants-in-Aid for Scientific Research from the Ministry of Education, Science, Sports, Culture, and Technology, Japan.

Disclosure Summary: T.K., K.-I.M., S.In., H.T., N.K., S.It., S.K., and M.K. have nothing to declare.

First Published Online January 13, 2009

Abbreviations: AGA, Androgenetic alopecia; AR, androgen receptor; DHT, dihydrotestosterone; DP, dermal papilla; FBS, fetal bovine serum; FE, follicular epithelial; Lef, lymphoid enhancer factor; MMTV, mouse mammary tumor virus; SDS, sodium dodecyl sulfate; Tcf, T cell factor.

References

  1. Hamilton JB 1942 Male hormone stimulation is prerequisite and incitant in common baldness. Am J Anat 71:451–480 [Google Scholar]
  2. Choudhry R, Hodgins MB, Van der Kwast TH, Brinkmann AO, Boersma WJ 1992 Localization of androgen receptors in human skin by immunohistochemistry: implications for the hormonal regulation of hair growth, sebaceous glands and sweat glands. J Endocrinol 133:467–475 [DOI] [PubMed] [Google Scholar]
  3. Thornton MJ, Taylor AH, Mulligan K, AI-Azzawi F, Lyon CC, O'Driscoll J, Messenger AG 2003 The distribution of estrogen receptor β is distinct to that of estrogen receptor α and the androgen receptor in human skin and the pilosebaceous unit. J Investig Dermatol Symp Proc 8:100–103 [DOI] [PubMed] [Google Scholar]
  4. Inui S, Itami S, Huei-Ju Pan, Chawnshang Chang 2000 Lack of androgen receptor transcriptional activity in human keratinocytes. J Dermatol Sci 23:87–92 [DOI] [PubMed] [Google Scholar]
  5. Mainwaring WIP 1977 The mechanism of action of androgen. Monographs on endocrinology. Berlin: Springer-Verlage; 10 [PubMed] [Google Scholar]
  6. Inui S, Fukuzato Y, Nakajima T, Yoshikawa K, Itami S 2002 Androgen-inducible TGF-β1 from balding dermal papilla cells inhibits epithelial cell growth: a clue to understand paradoxical effects of androgen on human hair growth. FASEB J 16:1967–1969 [DOI] [PubMed] [Google Scholar]
  7. Itami S, Kurata S, Takayasu S 1995 Androgen induction of follicular epithelial cell growth is mediated via insulin-like growth factor-I from dermal papilla cells. Biochem Biophys Res Commun 212:988–994 [DOI] [PubMed] [Google Scholar]
  8. Huelsken J, Vogal R, Erdmann B, Cotsarelis G, Birchmeir W 2001 β-Catenin controls hair follicle morphogenesis and stem cell differentiation in the skin. Cell 105:533–545 [DOI] [PubMed] [Google Scholar]
  9. Reddy S, Andl T, Bagasra A, Lu MM, Epstein DJ, Morrisey EE, Millar SE 2001 Characterization of Wnt gene expression in developing and postnatal hair follicles and identification of Wnt5a as a target of Sonic hedgehog in hair follicle morphogenesis. Mech Dev 107:69–82 [DOI] [PubMed] [Google Scholar]
  10. Cadigan KM, Nusse R 1997 Wnt signaling: a common theme in animal development. Genes Dev 11:3286–3305 [DOI] [PubMed] [Google Scholar]
  11. Kishimoto J, Burgeson RE, Morgan BA 2000 Wnt signaling maintains the hair-inducing activity of the dermal papilla. Genes Dev 14:1181–1185 [PMC free article] [PubMed] [Google Scholar]
  12. Andl T, Reddy ST, Gaddapara T, Millar SE 2002 WNT signaling are required for the initiation of hair follicle development. Dev Cell 2:643–653 [DOI] [PubMed] [Google Scholar]
  13. Ito M, Yang Z, Andl T, Cui C, Kim N, Millar SE, Cotsarelis G 2007 Wnt-dependent de novo hair follicle regeneration in adult mouse skin after wounding. Nature 447:316–320 [DOI] [PubMed] [Google Scholar]
  14. Wodarz A, Nusse R 1998 Mechanisms of Wnt signaling in development. Annu Rev Cell Dev Bio 14:59–88 [DOI] [PubMed] [Google Scholar]
  15. Mulholland DJ, Dedhar S, Coetzee GA, Nelson CC 2005 Interaction of nuclear receptors with the Wnt/b-catenin/Tcf signaling axis: Wnt you like to know? Endocr Rev 26:898–915 [DOI] [PubMed] [Google Scholar]
  16. Shimizu H, Morgan BA 2004 Wnt signaling through the β-catenin pathway is sufficient to maintain, but not restore, anagen-phase characteristics of dermal papilla cells. J Invest Dermatol 122:239–245 [DOI] [PubMed] [Google Scholar]
  17. van Genderen C, Okamura RM, Farinas I, Quo RG, Parslow TG, Bruhn L, Grosschedl R 1994 Development of several organs that require inductive epithelial-mesenchymal interactions is impaired in LEF-1-deficient mice. Genes Dev 8:2691–2703 [DOI] [PubMed] [Google Scholar]
  18. Pawlowski JE, Ertel JR, Allen MP, Xu M, Butler C, Wilson EM, Wierman ME 2002 Liganded androgen receptor interaction with β-catenin: nuclear co-localization and modulation of transcriptional activity in neuronal cells. J Bio Chem 277:20702–20710 [DOI] [PubMed] [Google Scholar]
  19. Verras M, Brown J, Li X, Nusse R, Sun Z 2004 Wnt3a growth factor induces androgen receptor-mediated transcription and enhances cell growth in human prostate cancer cells. Cancer Res 64:8860–8866 [DOI] [PubMed] [Google Scholar]
  20. Singh R, Artaza JN, Taylor WE, Braga M, Yuan X, Gonzalez-Cadavid NF, Bhasin S 2006 Testosterone inhibits adipogenic differentiation in 3T3–L1 cells: nuclear translocation of androgen receptor complex with β-catenin and T-cell factor 4 may bypass canonical Wnt signaling to down-regulate adipogenic transcription factors. Endocrinology 147:141–154 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Inui S, Fukuzato Y, Nakajima T, Kurata S, Itami S 2007 Androgen receptor co-activator Hic-5/ARA55 as a molecular regulator of androgen sensitivity in dermal papilla cells of human hair follicles. J Invest Dermatol 127:2302–2306 [DOI] [PubMed] [Google Scholar]
  22. Li Y, Li GQ, Lin CM, Cai XN 2005 One-step collagenase 1 treatment: an efficient way for isolation and cultivation of human scalp dermal papilla cells. J Dermatol Sci 37:58–60 [DOI] [PubMed] [Google Scholar]
  23. Tabata H, Nakano T, Inui S, Itami S 2005 Hair-growing activity in the extract of Ruta graveolens L. Fragrance J 12:58–62 [Google Scholar]
  24. Boukamp P, Petrussevska RT, Breitkreutz D, Hornung J, Markham A, Fusenig NE 1988 Normal keratinization in a spontaneously immortalized aneuploid human keratinocyte cell line. Cell Biol 106:761–771 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. He TC, Sparks AB, Rago C, Hermeking H, Zawel L, da Costa LT, Morin PJ, Vogelstein B, Kinzler KW 1998 Identification of c-MYC as a target of the APC pathway. Science 281:1509–1512 [DOI] [PubMed] [Google Scholar]
  26. Itami S Kurata S, Sonoda T, Takayasy S 1991 Mechanism of action of androgen in dermal papilla cells. Ann NY Acad Sci 642:385–395 [DOI] [PubMed] [Google Scholar]
  27. Randall VA, Thornton MJ, Hamada K, Redfern CPF, Nutbrown M, Ebling FJG, Messenger AG 1991 Androgen and the hair follicle: cultured human dermal papilla cells as a model system. Ann NY Acad Sci 642:355–375 [PubMed] [Google Scholar]
  28. Bhasin S, Calof OM, Storer TW, Lee ML, Mazer NA, Jasuja R, Montori VM, Gao W, Dalton JT 2006 Drug insight: testosterone and selective androgen receptor modulators as anabolic therapies for chronic illness and aging. Nat Clin Pract Endocrinol Metab 2:146–159 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Itami S, Inui S 2005 Role of androgen in mesenchymal epithelial interactions in human hair follicle. J Investig Dermatol Symp Proc 10:209–211 [DOI] [PubMed] [Google Scholar]
  30. Asada Y, Sonoda T, Ojiro M, Kurata S, Sato T, Ezaki T, Takayasu S 2001 5α-Reductase type 2 is constitutively expressed in the dermal papilla and connective tissue sheath of the hair follicle in vivo but not during culture in vitro. J Clin Endocrinol Metab 86 6:2875–2880 [DOI] [PubMed] [Google Scholar]
  31. Itami S, Kurata S, Sonoda T, Takayasu S 1995 Interaction between dermal papilla cells and follicular epithelial cells in vitro: effect of androgen. Br J Dermatol 132 4:527–532 [PubMed] [Google Scholar]
  32. Hibberts NA, Howell AE, Randall VA 1998 Balding hair follicle dermal papilla cells contain higher levels of androgen receptors than those from non-balding scalp. J Endocrinol 156 1:59–65 [DOI] [PubMed] [Google Scholar]
  33. Cullen DA, Killick R, Leigh PN, Gallo JM 2004 The effect of polyglutamine expansion in the human androgen receptor on its ability to suppress β-catenin-Tcf/Lef dependent transcription. Neurosci Lett 354:54–58 [DOI] [PubMed] [Google Scholar]
  34. Chesire DR, Isaacs WB 2002 Ligand-dependent inhibition of β-catenin/TCF signaling by androgen receptor. Oncogene 21:8453–8469 [DOI] [PubMed] [Google Scholar]
  35. Yokoya F, Imamoto N, Tachibana T, Yoneda Y 1999 β-Catenin can be transported into the nucleus in a Ran-unassisted manner. Mol Biol cell 10:1119–1131 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from The Journal of Clinical Endocrinology and Metabolism are provided here courtesy of The Endocrine Society

RESOURCES