Dermal papilla-induced hair differentiation of adult epithelial stem cells from human skin
Cecilia Roh,
Qingfeng Tao and
Stephen Lyle
Department of Pathology, Beth Israel Deaconess Medical Center, Boston, Massachusetts 02215
 |
ABSTRACT
|
---|
The epithelial-mesenchymal interactions between keratinocyte stem cells and dermal papilla (DP) cells are crucial for normal development of the hair follicle as well as during hair cycling. During the cyclical regrowth of a new lower follicle, the multipotent hair follicle stem cells are stimulated to proliferate and differentiate through interactions with the underlying mesenchymal DP cells. To characterize the events occurring during the process of epithelial stem cell fate determination, we utilized a coculture system by incubating human hair follicle keratinocyte stem cells with DP cells. Using GeneChip microarrays, we analyzed changes in gene expression within the stem cells upon coculture with the DP over a 5-day time course. A number of important signaling pathways and growth factors were regulated. The hair-specific keratin 6hf (K6hf) gene proved a particularly good marker of hair differentiation, with a 7.9-fold increase in mRNA and resulting increased protein levels. The high expression of K6hf was unique to DP-induced keratinocyte differentiation, since expression of K6hf was not induced by high calcium. Since the ß-catenin signaling pathway has been implicated in hair follicle development, we examined the role of ß-catenin in our system and demonstrated that ß-catenin/lef-1 signaling is required for DP-induced hair differentiation.
microarray; Lef-1; keratin 6hf
 |
INTRODUCTION
|
---|
ADULT STEM CELLS ARE RESPONSIBLE for the regeneration of the tissues in which they reside, during normal physiological turnover as well as during times of stress (30, 33, 41). These stem cells are also considered to have multipotentiality and thus can give rise to multiple cell types within the tissue (34). When a stem cell divides, it gives rise to another stem cell (self-renewal) and generates a daughter transit-amplifying (TA) cell, which will continue to proliferate and differentiate to repopulate the tissue (16). The hair follicle is one such regenerating system, which physiologically undergoes cycles of growth (anagen), regression (catagen), and rest (telogen) for numerous times in adult life (7). It has become evident that the portion of the hair follicle outer root sheath, called the bulge, contains a pool of epithelial stem cells in rodents and humans (8, 18, 20, 21). These epithelial stem cells appear to be multipotent, giving rise to a range of differentiated cell types in skin: epidermis, sebaceous glands, and eight different epithelial cell types of the hair follicle (8, 25, 29, 35).
During development of the hair follicle, cell signaling is interactively coordinated between epithelial keratinocytes and underlying specialized mesenchymal dermal papilla (DP) cells (6). These changes during hair morphogenesis are recapitulated in adult follicles during hair cycling. Since the lower follicle regresses in the catagen stage, a new lower follicle must be regenerated from the stem cells in the bulge region of the telogen club structure and the underlying DP cells, a process described as the "bulge-activation hypothesis" (8, 21). At this time of anagen onset, cell signaling between DP cells and keratinocyte stem cells is important for inducing stem cell proliferation and initiating the cascade of cell differentiation into one or all of the hair follicle cell lineages. Many molecules have been identified within the keratinocytes and DP cells at different stages of the hair cycle, and some signaling pathways have been implicated (2, 4, 11, 37); however, the study of stem cell activation and hair differentiation has been somewhat hindered by a lack of an in vitro system for study. In addition, little molecular information is available for human keratinocyte stem cells and DP cells. To identify signaling pathways and effector molecules that initiate the differentiation of human hair follicle stem cells, we developed a coculture system and used expression array analysis to characterize molecular changes, as well to test the significance of specific signals. Since it has been previously shown that the inductive capacity of the DP can be retained in culture (15, 39), we utilized our ability to isolate an enriched population of human keratinocyte stem cells to model the communication between DP cells and stem cells that occurs at the onset of anagen.
Our data show that the coculture system recapitulates the epithelial-mesenchymal interactions that induce hair differentiation of keratinocyte stem cells. Expression profiling reveals that the onset of hair follicle differentiation is characterized by a unique set of genetic changes within stem cells. Hair-specific keratin 6hf (K6hf) is a particularly good marker of DP-induced differentiation of stem cells, which is specific for DP induction, unlike the more general calcium-induced differentiation. We also demonstrate that ß-catenin/Lef-1 signaling is directly involved in human DP-induced hair differentiation and is required for K6hf expression. These results define a number of targets for manipulating stem cell fate determination and provide a system that can be used to study the function of specific molecules and pathways involved the process of hair follicle differentiation.
 |
MATERIALS AND METHODS
|
---|
Isolation and culture of human follicular keratinocytes and DP cells.
Excess skin from facelift procedures was collected with IRB approval. Human hair follicles were isolated by microdissection. To isolate matrix cells and DP cells, the skin was cut at the dermo-subcutaneous interface with a razor blade, hair follicles at anagen phase were selected, and matrix cell-containing tissue fragments or DP cells were dissected out under the dissecting microscope. To isolate epithelial stem cells, the skin was incubated in dispase (Sigma, 12.5 mg/ml in DMEM) overnight, each hair was pulled out of the skin, and hair follicles at telogen phase were selected and cut at the bulge region. Isolated tissue fragments were incubated in a mixture of 0.05% trypsin-EDTA (GIBCO) and Versene (0.53 mM EDTA·4Na, GIBCO) for 30 min at room temperature and spun down for 5 min at 800 rpm. The supernatant was removed, and isolated cells were plated on mitomycin C-treated (Sigma, 1.5 mg/ml DMEM for 2 h) 3T3-J2 cells (31) in keratinocyte medium [KCM: DMEM and Hams F12 (GIBCO, 4:1), adenine (Sigma, 180 mM), 10% fetal bovine serum (GIBCO), cholera toxin (ICN, 0.1 nM), penicillin/streptomycin (GIBCO, 100 U/ml and 100 mg/ml, respectively), hydrocortisone (Sigma, 0.4 mg/ml, 1.1 mM), T/T3 (transferrin, GIBCO, 5 µg/ml, 649 nM; and triiodo-l-thyronine, Sigma, 2 nM), insulin (Sigma, 5 mg/ml, 862 nM), and EGF (Sigma, 10 ng/ml, 1.6 nM), pH 7.2] without EGF and then changed to EGF-containing KCM. Cells were grown at 37°C in a humid atmosphere containing 5% CO2. All keratinocytes were fed with KCM containing EGF every 2 days and grown for 1420 days. Isolated DP cells were seeded and grown without the feeder layer in Chang medium C (Irvine scientific) supplemented with 10% FBS and penicillin/streptomycin. The cells maintained expression of smooth muscle actin throughout the culture process, indicating their preservation of the DP phenotype (data not shown).
Retroviral infection of keratinocytes.
Once colonies were initially established at 2030 cells, the media containing retroviruses of either an empty vector control or myc-tagged
Nlef-1 (kindly provided by Dr. F. Watt, Cancer Research UK) (27) were added with the same volume of KCM and a final concentration of 2 mg/ml Polybrene (Sigma). After 24 h, virus-containing media were changed to KCM. All cells were infected twice and treated in the same way as for the uninfected keratinocytes.
Coculture of keratinocytes with DP cells.
Colonies of primary keratinocytes were harvested by first removing the feeder layers with Versene and then treating the keratinocytes with 0.1% trypsin-EDTA (GIBCO). Cells were plated at a density of 300,000 cells per one 6-well plate or 100,000 cells per one 12-well plate. Prior to use, all plates were coated with 10 mg of collagen type IV (Sigma) per 1 ml of PBSABC (PBS supplemented with 1 mM CaCl2 and 1 mM MgCl2) overnight at 4°C and then with 0.05 mg/ml BSA in purified water for 30 min at 37°C. Cells attached overnight in KCM without EGF, and then this was changed to fresh KCM containing EGF along with mitomycin-treated 3T3-J2 inserts. To stabilize the newly plated keratinocytes to a normal growth condition, these were preincubated with 3T3-J2 cells overnight. After overnight incubation, half of the 3T3-J2 inserts were replaced with untreated DP cell inserts and all cells were fed with fresh media. Transwell inserts (Corning) for 6-well plates were prepared with 50,000 mitomycin C-treated 3T3-J2 cells or untreated DP cells/insert, and Transwell inserts for 12-well plates and 12-mm Millicell inserts (Millipore) were prepared with 20,000 cells/insert. DP cells were plated on the inserts and cultured for 3 days prior to coculture with keratinocytes, to ensure normalization and stability of cells. On the last day of each experimental setup, both 3T3-J2 and DP inserts were fixed in methanol for 10 min, stained with 0.2% crystal violet for 10 min, washed with water, and viewed for the cell number to confirm that DP cells were not overconfluent and to check the presence of proper number of viable 3T3-J2 cells during the coculture.
Differentiation of keratinocytes with calcium.
Plated cells as described above attached overnight in KCM without EGF and then were changed to low-calcium medium, KGM (keratinocyte growth medium; CELLnTEC, Bern, Switzerland) supplemented with 10% FBS and penicillin/streptomycin. After overnight incubation, keratinocytes were fed with fresh KGM supplemented with additional 1.5 mM CaCl2 and incubated for the indicated times.
Immunofluorescence staining of hair follicles and of
Nlef-1-infected stem cells.
Tissue sections of hair follicles were prepared from fresh human skin after treatment with dispase for 1 h at 37°C, followed by plucking and then freezing in OCT compound (Sakura Finetek). Frozen follicles were cut at 5 µm thick section (Leica) and fixed in acetone for 10 min. All sections were permeabilized in 0.25% Triton X-100 in PBS and blocked for 1 h with 5% goat serum in PBS. The follicle sections were immunofluorescently stained with anti-K6hf antibody (1:100; Progen, Heidelberg, Germany) followed by FITC-conjugated anti-guinea pig antibody (1:100, Dako) or for keratin 15 (K15) with anti-C8/144B (1:100, Dako) antibody followed by Texas Red-conjugated anti-mouse antibody (1:100, Dako). All follicle sections were mounted with Vectashield (Vector) and imaged using a fluorescence microscope (Leica).
Nlef-1-infected stem cells were plated on the glass coverslip and stained for myc-tag with anti-myc (1:100, Cell Signaling) antibody followed by Texas Red-conjugated anti-mouse antibody (1:100).
Isolation of total RNA from cultured keratinocytes and GeneChip microarrays.
Isolated epithelial stem cells were cocultured by plating primary keratinocyte stem cells in 75-mm Transwell inserts and incubating with either 3T3-J2 fibroblasts or DP cells present in the bottom of 100-mm plates. Total RNA was isolated at days 0, 1, 2, and 5 by using the RNeasy kit (Qiagen), according to the manufacturers instructions. RNA samples were subjected to GeneChip microarray analysis (Affymetrix) for monitoring the relative abundance of mRNA transcripts hybridizing to 12,625 probes on the GeneChips in the BIDMC Genomics Center core facility.
Real-time RT-PCR.
PCR primers (5'-TCATCGACAAGGTGAGGTTCTTGG-3' and 5'-GTTGATCTCCTCGGGCAGAGATTT-3') and a fluorogenic probe (5'-ACAAGCGCACAGCTGCTGAGAATGAA) containing a reporter dye (5' 6-FAM) and a quencher dye (3' Black Hole Quencher) for K6hf were designed using the PrimerQuest program at the Integrated DNA Technologies (IDT) web site and were purchased from IDT. TaqMan GAPDH control reagents (human) were purchased from PE Applied Biosystems and used as an endogenous reference. Using TaqMan RT-PCR master mix and isolated total RNA from the DP cocultured stem cells, RT-PCR runs were performed at 48°C for 30 min, 95°C for 10 min, 95°C for 15 s, and 60°C for 1 min for 40 cycles on an ABI Prism 7700 sequence detection system (PE Applied Biosystems). Relative gene induction values were calculated following the manufacturers recommendations.
Preparation of whole, nuclear, and cytoplasmic cell extracts and Western blot analysis.
Whole cell lysates were prepared by using RIPA buffer (1% Nonidet P-40, 0.5% deoxycholate, 0.1% SDS in PBS, pH 7.4) containing 1 mM phenylmethylsulfonyl fluoride (PMSF) and protease and phosphatase inhibitor cocktails (Sigma). Keratinocytes were first washed with cold PBS, scraped with RIPA buffer on ice, and incubated on ice for 20 min. Lysates were microcentrifuged at the maximum speed (16,100 x g) for 10 min at 4°C, and the supernatant was removed and used for the whole cell extracts. For cytoplasmic and nuclei extracts, cells were lysed in lysis buffer (10 mM Tris, pH 7.6, 10 mM NaCl, 3 mM MgCl2, 0.5% Nonidet P-40) containing protease and phosphatase inhibitor cocktails and incubated on ice for 15 min. After a low-speed centrifugation, the supernatant was removed and used for the cytoplasmic extracts. The resulting pellet was washed once with lysis buffer and resuspended in nuclear extraction buffer (20 mM HEPES, pH 7.9, 350 mM NaCl, 1.5 mM MgCl2, 0.2 mM EDTA, pH 8, 25% glycerol) containing protease and phosphatase inhibitor cocktails, incubated on ice for 20 min, and then centrifuged at the maximum speed at 4°C for 10 min. The resulting supernatant was removed and used as nuclear extracts. Protein concentration was measured by using a BCA kit (Pierce). Protein samples were resuspended in SDS sample buffer (Pierce) supplemented with 1% (vol/vol) ß-mercaptoethanol, boiled for 10 min at 95°C, and subjected to 10% SDS-polyacrylamide gel electrophoresis (Bio-Rad). Proteins were transferred to Hybond ECL nitrocellulose membrane (Amersham) in 25 mM Tris and 192 mM glycine, pH 8.3, followed by blocking with 5% nonfat dry milk in PBST (0.05% Tween 20) for 1 h at room temperature. After overnight incubation at 4°C with indicated specific primary antibodies, proteins were detected with horseradish peroxidase-conjugated secondary antibodies (Santa Cruz and Dako), followed by enhanced chemiluminescence (Amersham) or the SuperSignal West Femto Maximum Sensitivity substrate (Pierce). The antibodies used were guinea pig polyclonal anti-K6hf (1:1,000) from Progen; rabbit polyclonal anti-ß-catenin (1:1,000) from Upstate; monoclonal anti-ß-actin (1:5,000) from Sigma; and monoclonal anti-proliferation cell nuclear antigen (anti-PCNA, 1:500) from Cymbus Biotechnology. Densitometry was performed by using the scientific image processing software, IPLab (Scanalytic).
Statistical analysis.
The statistical analysis was performed using ANOVA (analysis of variance between groups) post hoc tests using the StatView program.
 |
RESULTS
|
---|
Of 12,625 genes screened, 269 genes within stem cells are highly regulated upon coculture with DP cells.
To identify genes activated during the initiation of stem cell differentiation, we carried out microarray analysis of epithelial stem cells induced by coculture with DP cells. Clonogenic keratinocyte stem cells were cultured from the bulge area of human telogen follicles and combined with DP cells in a Transwell system (Fig. 1). RNA was isolated from stem cells grown with DP cells for different periods of time. Isolated RNA was used to generate labeled probes, which were then hybridized to U95Av2 human genome arrays, followed by the scanning and data analysis (Affymetrix). Of the 12,625 probes on the arrays, 613, 622, and 357 initial genes showed a change from day 0 (no DP coculture) to 1, 2, and 5 days of coculture with DP cells, respectively. We further filtered and sorted the genes whose expression was either up- or downregulated greater than threefold. The resulting 269 genes were grouped into 10 categories, according to their functions as reported in the literature, and are partially listed in Table 1 (a complete list is in Supplemental Table S1, available at the Physiological Genomics web site).1
Most of the genetic changes were observed between days 1 and 2 and showed a correlation at these two time points, supporting the validity of the data. For a time course over 5 days, it is not unexpected that some genes will show rapid upregulation and then downregulation. For instance, BRCA1-associated protein 1 (BAP1), shows strong upregulation at day 1, followed by lower levels on day 2 and a decrease from starting levels on day 5. This likely reflects an early response to induced expression, which is then rapidly shut off as terminal differentiation progresses.

View larger version (98K):
[in this window]
[in a new window]
|
Fig. 1. Transwell coculture system. Keratinocyte stem cells isolated from plucked human telogen follicles form colonies (A), which are trypsinized and then combined with cultured dermal papilla (DP) cells from anagen follicles (B), in the Transwell plates (C).
|
|
As DP induction of stem cells initiates the differentiation program, it is not surprising that more than 50% of the highly regulated genes were seen in groups, such as cell structure/motility/extracellular matrix, growth response genes, signal transduction, and cell metabolism. The hair keratin K6hf showed a significant increase, while markers of epidermal differentiation such as keratin 1 and profillagrin were dramatically downregulated. Genes involved in extracellular remodeling such as stromelysin-2 and type IV collagenase were highly regulated and would be expected to mediate the morphogenesis occurring when the stem cells regenerate a new lower follicle. Growth factors and cytokines such as heparin-binding EGF-like growth factor and platelet-derived endothelial cell growth factor could correlate with the proliferation and migration of the new TA keratinocytes as well other cell types involved in regrowth of a follicle, which requires angiogenesis and dermal remodeling. Several cytokine/chemokines such as IL-8, colony-stimulating factor-1 (CSF-1), melanoma growth stimulatory activity (MGSA), and MDA-7 (IL-24) were positively regulated, suggesting that stem cells are actively signaling to their immediate environment. Last, several unique and largely uncharacterized transcription factors such as ZFM1 and ZFP2 may be important in the early response to differentiation signals.
Two recent reports have profiled stem-cell-specific gene differences in murine bulge cells compared with non-bulge cells (25, 36). When our results were compared with the results of the mouse keratinocyte stem-cell-specific gene expression profiles, we observed several findings. Of homologous genes that were present on the human arrays, many of the stem-cell-specific genes were present at low levels or were below detection in our time zero keratinocytes. This could represent differences between mouse and human bulge cells but also could be due to changes in the stem cells upon isolation and culturing. A few, such as Gas1 (growth arrest specific), showed significant correlation. Gas1 expression levels were between 2.6- and 3.8-fold higher in mouse stem cells than in non-bulge basal keratinocytes, and in our system, the stem cells showed a 4- to 5-fold reduction in Gas1 expression with DP-induced differentiation. As well, the K15 stem cell marker showed a
2.5-fold decrease in our DP-induced stem cells at day 1. These findings reflect the transition from the stem cell phenotype to more committed and differentiated fate that is induced by the DP.
K6hf, hair-follicle-specific keratin, gene expression is highly upregulated.
The most highly regulated response was scored in the category of cell structure/motility/extracellular matrix, which also includes many hair-follicle-specific genes. Of all the genes, we decided to more thoroughly examine K6hf expression, since keratins have traditionally been used as well-accepted markers of differentiation and because many of the other highly regulated genes are not known to be specific in the hair follicle. Also, well-characterized antibodies to K6hf were available for use in Western blotting and immunofluorescence. K6hf was stimulated up to 7.9-fold within 1 day of coculture with DP cells and sustained the level until the second day (Table 1). K6hf is the type II keratin, exclusively expressed in the companion layer (Fig. 2, A and C) (40), which consists of a single row of flattened cells arranged end to end, adjacent to the outer root sheath and in contact with the outermost Henles layer of the inner root sheet (32, 40). K6hf has also recently been shown to be expressed in the upper matrix cells within the hair bulb and in the medulla of the hair shaft in the anagen phase human hair (38). In our immunofluorescence staining of whole-mounted follicle, we detected K6hf only at the companion layer. Since our starting cells were isolated from the K15-positive undifferentiated stem cell region of the outer root sheath (21), we compared the pattern of K15 immunofluorescence to K6hf within the hair follicle (Fig. 2) and showed that they are clearly two distinct populations of cells. Wang et al. (38) also suggested that K17 might pair with K6hf by showing K17-null hair mice having a reduced expression of K6hf, as well as the in vitro experiments that K6hf and K17 coassemble into keratin filaments (13). Our results are consistent with these findings since K17 protein levels also increased with DP induction, although the difference was not statistically significant due to a high expression in undifferentiated keratinocytes.

View larger version (18K):
[in this window]
[in a new window]
|
Fig. 2. Keratin 15 (K15) and hair-specific keratin 6hf (K6hf) are differentially expressed within hair follicles by immunofluorescence. Plucked follicles were simultaneously stained with anti-K15 (red) and anti-K6hf antibody (green). K6hf is present in the companion layer of the lower follicle (A), which extends along the innermost layer of the outer root sheath up to the bulge area of the infundibulum (C). The K15 stem cell marker is absent in the lower follicle (B), and stains only the basal layer of the outer root sheath the bulge (D). The DP and surrounding matrix area of the lower follicle is designated by the asterisk.
|
|
To confirm the increase of K6hf mRNA by DP induction, quantitative real-time RT-PCR was performed on stem cell RNA after coculture with DP cells. As seen in Fig. 3, an approximate fourfold increase in K6hf mRNA was detected after 1 day of coculture with DP cells and the induction rose to almost sevenfold after 2 days.

View larger version (24K):
[in this window]
[in a new window]
|
Fig. 3. K6hf mRNA expression measured by using real-time RT-PCR. Epithelial stem cells were cocultured with DP cells for the indicated days, and RNAs were isolated and subject to real-time RT-PCR for K6hf mRNA expression. A: graphs represent the standard curves of GAPDH and day 0 K6hf. B: values were normalized by GAPDH mRNA levels and to day 0 (D0) K6hf mRNA.
|
|
To determine whether the DP-induced increase in K6hf mRNA correlates to an increase in K6hf protein, we performed Western blot analysis with the whole cell lysates from stem cells after coculture with DP cells for indicated times. As seen in Fig. 4, we detected a significant upregulation of K6hf protein expression. However, time of the protein expression lagged behind the increase in mRNA, showing the highest expression at days 3 and 4 rather than days 1 and 2 seen for mRNA.

View larger version (26K):
[in this window]
[in a new window]
|
Fig. 4. Expression of K6hf protein in epithelial stem cells cocultured with DP cells. Epithelial stem cells were harvested and lysed, and equal amounts of protein were subjected to 10% PAGE. A: representative blots for expression of K6hf, ß-actin, and PCNA analyzed by Western blotting. B: values are means ± SE from at least 4 independent experiments. J2, 3T3-J2 cells. Comparisons by two-factor ANOVA demonstrate that DP significantly induces the K6hf expression. Further analysis of the samples by a Fisher PLSD test shows that DP induction was significantly different from control (*P < 0.01) and from calcium-induced cells (**P < 0.05).
|
|
K6hf expression is specific to DP-induced differentiation of keratinocytes.
To verify whether K6hf expression is a general marker of keratinocyte differentiation or specific to DP induction, we induced the epithelial stem cells with 1.5 mM calcium after growing cells in low-calcium medium and then examined the cells for the expression of K6hf protein. The expression of K6hf was markedly increased only with the DP coculture (Figs. 4 and 5). High calcium is well-known to induce terminal differentiation of keratinocytes (12, 43) and did show a marked decrease in the PCNA proliferation marker (Fig. 4) but showed no increase in K6hf. High calcium does upregulate markers of epidermal terminal differentiation, keratin 1, and involucrin, in our stem cells (data not shown). Thus K6hf expression appears specific for DP-induced hair follicle differentiation, whereas calcium treatment produces a more general epidermal-type differentiation.

View larger version (28K):
[in this window]
[in a new window]
|
Fig. 5. Expression of K6hf protein in matrix cells cocultured with DP cells. Matrix cells were harvested and lysed, and equal amounts of protein were subjected to 10% PAGE. A: representative blots for expression of K6hf and ß-actin analyzed by Western blotting. B: values are means ± SE from at least 3 independent experiments. Comparisons by two-factor ANOVA demonstrate that DP significantly induces the K6hf expression. Further analysis of the samples by a Fisher PLSD test shows that DP induction was significantly different from control (*P < 0.005) and from calcium-induced cells (**P < 0.0001).
|
|
Matrix (TA daughter) cells are more responsive to DP induction.
Hair follicles are complex structures with multiple different keratinocyte subtypes. One functional distinction lies among stem cells, proliferating TA cells, and terminally differentiated cells. Since matrix cells in the hair bulb represent the TA cell population, which is more committed toward hair differentiation, we wanted to examine the differentiation state and inducibility of these cells. Thus we isolated both epithelial stem cells and matrix cells and induced differentiation by coculturing with DP cells or by raising intracellular calcium concentration. The matrix TA cells appear more responsive to the DP stimulus, exhibiting increased K6hf protein levels within 12 days (Fig. 5) compared with 34 days seen in the stem cells. Calcium induction did not induce K6hf expression of even the more committed matrix cells, again supporting the evidence that specific mesenchymal signals are required for follicular differentiation.
ß-Catenin signaling pathway regulates DP-induced stem cell differentiation and is required for K6hf expression.
Wnt/ß-catenin signaling has been shown to function in hair follicle development and differentiation of murine follicles (2, 9, 14). To investigate whether ß-catenin signaling pathway is involved in DP-induced differentiation in human epithelial stem cells, we fractionated stem cells into cytoplasmic and nuclear extracts and analyzed the change in ß-catenin levels by Western blotting. As demonstrated in Fig. 6, coculturing with DP cells induced translocation of ß-catenin protein into the nucleus of keratinocytes stem cells with a concomitant increase in K6hf protein levels in cytoplasmic fractions. Total ß-catenin protein appeared unchanged, and this correlated with a lack of significant changes in mRNA levels seen by expression array analysis.

View larger version (59K):
[in this window]
[in a new window]
|
Fig. 6. Involvement of ß-catenin signaling pathway in DP-induced stem cell differentiation. Stem cells cocultured with DP cells were fractionated into nuclear and cytoplasmic extracts. Equal amounts of protein were subject to 10% PAGE. Nuclear localization of ß-catenin increased with DP induction and corresponded to increased K6hf expression. Cytoplasmic ß-catenin showed no significant changes. Representative blots are shown for expression of ß-catenin, K6hf, and ß-actin analyzed by Western blotting from 3 independent experiments.
|
|
To further examine the functional significance of ß-catenin signaling in the stem cell differentiation, we infected cells with a retroviral myc-
Nlef-1 construct and cocultured them with DP cells. This construct is missing the ß-catenin binding site and acts as a dominant negative to inhibit ß-catenin signaling (26). As seen in Fig. 7A,
Nlef-1-infected cells showed nuclear staining with anti-myc antibody. Compared with the control cells infected with an empty vector, DP-induced K6hf expression was not detected in
Nlef-1-infected cells (Fig. 7B), indicating that K6hf gene is regulated by the ß-catenin/lef-1 signaling pathway upon DP induction of epithelial stem cells.
 |
DISCUSSION
|
---|
Epithelial-mesenchymal interactions play a pivotal role in the transition from the undifferentiated state of epithelial stem cells in telogen follicles toward the rapidly proliferating and differentiating cells of the hair matrix in anagen follicles. This is evident from our findings that DP cells induce numerous changes in gene expression levels within hair follicle stem cells. Examination of the changes in expression patterns reveals several new opportunities for advancing our knowledge of epithelial stem cells, as well as for understanding normal and diseased hair biology. It is clear that, in addition to changes that directly affect the stem cells themselves, a number of regulated genes appear to be important for modulating and signaling to the immediate stem cell environment or "niche." Although the exact nature of the stem cell niche is not well characterized, it is known that melanocyte precursors (28) and Langerhans cells (24) are present within the hair follicle bulge region and thus may be targets of factors secreted during epithelial stem cell activation. For instance, MGSA is a chemokine known to influence melanocyte proliferation (19), while CSF-1 plays a role in Langerhans cell growth and differentiation (23, 42). As well, platelet-derived endothelial growth factor may modulate the behavior of stromal fibroblasts, endothelial cells, and even the DP cells. It thus appears likely that there is a coordinated process of signaling between different cell types during hair cycling.
The epithelial stem cells present within the hair follicle bulge are generally undifferentiated, both biochemically and ultrastructurally (1, 8). They express K15 and other basal keratins including keratin 19 (1, 21, 22). Since keratins are expressed in a tissue- and differentiation-specific manner, they serve as robust markers of differentiation (17). For instance, the expression of keratins 5 and 14 in the epidermal basal layer give way to keratins 1 and 10 in the spinous layer and subsequently keratin 2 and 10 in the upper spinous and granular layers. Most processes of terminal keratinocyte epidermal differentiation are controlled by the intracellular calcium concentration and are marked by the expression of loricrin, involucrin, and profilaggrin (3). Our results show that stem cells require particular signals from the DP to initiate hair differentiation, whereas the lack of hair induction signals leads to a default epidermal-type of terminal differentiation as seen with high calcium alone. For instance, the keratin 1 and profilaggrin epidermal differentiation markers are significantly downregulated with DP induction. In addition, desmocollin 1 synthesis has previously been shown to be abruptly shut down in terminal differentiation of the hair follicle companion layer (10), correlating with the dramatic downregulation in desmocollin 1 RNA levels seen in our experiments. Our finding of prominent K6hf gene upregulation in DP-induced keratinocyte differentiation is significant as being a novel marker for hair-follicle-specific differentiation of stem cells. The K6hf gene was recently described as being regulated in a hair-cycle-dependent manner, which is missing in the telogen phase of the hair cycle (38). In agreement with this, our stem cell isolates from human telogen follicles express very low levels of K6hf, which then dramatically increase with the DP-induced transition from telogen to anagen. Our findings also provide new insights in hair differentiation. It is well-known that epithelial stem cells first proliferate and generate matrix cells, which then produce inner hair layers such as all seven different type of cells (medulla, cortex, hair cuticle, cuticle of inner root sheath, Huxleys and Henles layers, the companion layer). The complex processes of how these cell layers are positioned and what kind of lateral communications are exchanged remain unknown; however, our results suggest that the K6hf-containing companion layer may be the first layer to be induced and possibly followed by signals that would differentially regulate the next epithelial layer. The sequence of events and exact molecules involved in the development of the complex anagen bulb structure will require more investigation.
We were surprised to see that TA cells derived from the anagen matrix are not preprogrammed to differentiate into the hair lineages, since they also did not show K6hf upregulation with high calcium. Although they represent the more committed TA progeny of the stem cells, they also require specific DP signals for K6hf expression. However, they did show an earlier and stronger response to DP induction, thus indicating that they have undergone some intermediate gene regulation to become primed for receiving the signals. Gene profiling of this compartment could provide additional clues to the earliest differences between stem cells and their progeny.
In addition to differentiation, cell migration is another important process during hair regeneration at the onset of anagen, as the daughter TA cells must migrate through the dermis to their final place in the subcutaneous fat. A number of differentially regulated genes appear to be active in this process. These include upregulated metalloproteinases to degrade the extracellular matrix and downregulated lamin B receptor homolog to help dissociate migrating cells from the basement membrane. Migration also plays a significant role in wound healing, as is evident by data showing stem cell activation during wounding (35), and thus our results could also provide useful information for developing targets for better wound healing.
To understand stem-cell maintenance and lineage determination, we need to characterize the significance of cell signaling pathways that are involved in mesenchymal-epithelial interactions during stem cell growth and differentiation. Recent studies have implicated wnt/ß-catenin/lef-1 as one of the signaling pathways in hair follicle biology as well as pilosebaceous tumorigenesis (5, 14). Our results also demonstrated that nuclear ß-catenin levels increase in DP-induced stem cell differentiation. Endogenous ß-catenin is rapidly degraded, and only a small pool of ß-catenin remains available for signaling. However, because no significant change was detected in ß-catenin mRNA levels by microarray or cytoplasmic extracts by Western blotting, we can reason that DP causes the stabilization of ß-catenin in the cytoplasm, rescuing it from degradation and subsequently leading to nuclear translocation. This is in agreement with Van Mater et al. (37), who demonstrated transient activation of ß-catenin signaling results in activation of a normal transition from telogen to anagen, whereas chronic activation of ß-catenin results in proliferation of the outer root sheath and other epithelial components of the hair follicle. Such stabilization would give a faster and more transient response than producing new RNA and protein in signaling. Using
Nlef-1-infected stem cells, we also verified that K6hf is one of the genes regulated by ß-catenin/lef-1 signaling. Since it is widely believed that most of keratin genes including K6hf contain the consensus binding sites for lef-1/Tcf in the proximal region of the 5' upstream sequence (38), it is evident that K6hf is indeed activated by transcriptionally active ß-catenin/lef-1 complexes. It is also interesting that ß-catenin and lef-1 appear to play a central role in cutaneous tumorigenesis (5, 26, 27). Over-activity of ß-catenin can lead to pilomatricomas, a tumor of the hair matrix cells, whereas blockage of ß-catenin with a dominant-negative lef-1 can lead to sebaceous neoplasms at the expense of hair differentiation. The coculture assay can rapidly be used to characterize the members of the wnt/ß-catenin/lef-1 pathway and resulting regulated genes, as well as other pathways, involved in hair differentiation and hair tumor formation.
Many important intracellular and cell-cell signaling pathways are regulated by the epithelial-mesenchymal interactions that occur when epithelial stem cells are induced to differentiate. The coculturing of skin epithelial stem cells with DP cells recapitulates the interactions occurring at the onset of hair follicle regeneration, which can be assayed by upregulation of the hair-specific keratin, K6hf. It will be interesting to see whether other highly regulated genes such as NNMT, ZFM1, and uncharacterized gene products are specific to hair differentiation. ß-Catenin/lef-1 signaling appears to be one critical pathway in human hair differentiation; however, specific upstream signals have yet to be characterized. The ability to model the epithelial-mesenchymal interactions of stem cell activation and differentiation using an in vitro coculture system should lead to advances in our knowledge of human epithelial stem cell fate determination as it applies to hair biology, wound healing, and carcinogenesis.
 |
GRANTS
|
---|
This work supported by National Institutes of Health Grants AR-42689 and AR-02179 (to S. Lyle) and HL-07893 (to C. Roh) and by the Ellison Medical Foundation.
 |
ACKNOWLEDGMENTS
|
---|
We thank Dr. Fiona Watt, Cancer Research UK, London, UK, for the generous gift of the
Nlef-1 and control vectors.
 |
FOOTNOTES
|
---|
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).
Address for reprint requests and other correspondence: S. Lyle, Beth Israel Deaconess Medical Center, Dept. of Pathology, 330 Brookline Ave., Boston, MA 02215 (E-mail: slyle{at}bidmc.harvard.edu).
The Supplementary Material for this article (Supplemental Table S1) is available online at http://physiolgenomics.physiology.org/cgi/content/full/00134.2004/DC1. 
 |
REFERENCES
|
---|
- Akiyama M, Dale BA, Sun TT, and Holbrook KA. Characterization of hair follicle bulge in human fetal skin: the human fetal bulge is a pool of undifferentiated keratinocytes. J Invest Dermatol 105: 844850, 1995.[Abstract]
- Andl T, Reddy ST, Gaddapara T, and Millar SE. WNT signals are required for the initiation of hair follicle development. Dev Cell 2: 643653, 2002.[ISI][Medline]
- Bikle DD and Pillai S. Vitamin D, calcium, and epidermal differentiation. Endocr Rev 14: 319, 1993.[ISI][Medline]
- Botchkarev VA and Kishimoto J. Molecular control of epithelial-mesenchymal interactions during hair follicle cycling. J Invest Dermatol Symp Proc 8: 4655, 2003.[CrossRef]
- Chan EF, Gat U, McNiff JM, and Fuchs E. A common human skin tumour is caused by activating mutations in beta-catenin. Nat Genet 21: 410413, 1999.[CrossRef][ISI][Medline]
- Chuong CME. Molecular Basis of Epithelial Appendage Morphogenesis. Georgetown, TX: Landes, 1998.
- Cotsarelis G. The hair follicle: dying for attention. Am J Pathol 151: 15051509, 1997.[ISI][Medline]
- Cotsarelis G, Sun TT, and Lavker RM. Label-retaining cells reside in the bulge area of pilosebaceous unit: implications for follicular stem cells, hair cycle, and skin carcinogenesis. Cell 61: 13291337, 1990.[ISI][Medline]
- DasGupta R, Rhee H, and Fuchs E. A developmental conundrum: a stabilized form of ß-catenin lacking the transcriptional activation domain triggers features of hair cell fate in epidermal cells and epidermal cell fate in hair follicle cells. J Cell Biol 158: 331344, 2002.[Abstract/Free Full Text]
- Donetti E, Boschini E, Cerini A, Selleri S, Rumio C, and Barajon I. Desmocollin 1 expression and desmosomal remodeling during terminal differentiation of human anagen hair follicle: an electron microscopic study. Exp Dermatol 13: 289297, 2004.[CrossRef][ISI][Medline]
- Fuchs E, Merrill BJ, Jamora C, and DasGupta R. At the roots of a never-ending cycle. Dev Cell 1: 1325, 2001.[ISI][Medline]
- Hennings H, Michael D, Cheng C, Steinert P, Holbrook K, and Yuspa SH. Calcium regulation of growth and differentiation of mouse epidermal cells in culture. Cell 19: 245254, 1980.[ISI][Medline]
- Hofmann I, Winter H, Mucke N, Langowski J, and Schweizer J. The in vitro assembly of hair follicle keratins: comparison of cortex and companion layer keratins. Biol Chem 383: 13731381, 2002.[ISI][Medline]
- Huelsken J, Vogel R, Erdmann B, Cotsarelis G, and Birchmeier W. ß-Catenin controls hair follicle morphogenesis and stem cell differentiation in the skin. Cell 105: 533545, 2001.[CrossRef][ISI][Medline]
- Jahoda CA, Oliver RF, Reynolds AJ, Forrester JC, and Horne KA. Human hair follicle regeneration following amputation and grafting into the nude mouse. J Invest Dermatol 107: 804807, 1996.[Abstract]
- Jones PH and Watt FM. Separation of human epidermal stem cells from transit amplifying cells on the basis of differences in integrin function and expression. Cell 73: 713724, 1993.[ISI][Medline]
- Kanitakis J. Immunohistochemistry of normal human skin. Eur J Dermatol 8: 539547, 1998.[ISI][Medline]
- Kobayashi K, Rochat A, and Barrandon Y. Segregation of keratinocyte colony-forming cells in the bulge of the rat vibrissa. Proc Natl Acad Sci USA 90: 73917395, 1993.[Abstract]
- Krasagakis K, Garbe C, Zouboulis CC, and Orfanos CE. Growth control of melanoma cells and melanocytes by cytokines. Recent Results Cancer Res 139: 169182, 1995.[Medline]
- Liu Y, Lyle S, Yang Z, and Cotsarelis G. Keratin 15 promoter targets putative epithelial stem cells in the hair follicle bulge. J Invest Dermatol 121: 963968, 2003.[Abstract/Free Full Text]
- Lyle S, Christofidou-Solomidou M, Liu Y, Elder DE, Albelda S, and Cotsarelis G. The C8/144B monoclonal antibody recognizes cytokeratin 15 and defines the location of human hair follicle stem cells. J Cell Sci 111: 31793188, 1998.[Abstract/Free Full Text]
- Michel M, Torok N, Godbout MJ, Lussier M, Gaudreau P, Royal A, and Germain L. Keratin 19 as a biochemical marker of skin stem cells in vivo and in vitro: keratin 19 expressing cells are differentially localized in function of anatomic sites, and their number varies with donor age and culture stage. J Cell Sci 109: 10171028, 1996.[Abstract/Free Full Text]
- Mollah ZU, Aiba S, Nakagawa S, Hara M, Manome H, Mizuashi M, Ohtani T, Yoshino Y, and Tagami H. Macrophage colony-stimulating factor in cooperation with transforming growth factor-ß1 induces the differentiation of CD34+ hematopoietic progenitor cells into Langerhans cells under serum-free conditions without granulocyte-macrophage colony-stimulating factor. J Invest Dermatol 120: 256265, 2003.[Abstract/Free Full Text]
- Moresi JM and Horn TD. Distribution of Langerhans cells in human hair follicle. J Cutan Pathol 24: 636640, 1997.[ISI][Medline]
- Morris RJ, Liu Y, Marles L, Yang Z, Trempus C, Li S, Lin JS, Sawicki JA, and Cotsarelis G. Capturing and profiling adult hair follicle stem cells. Nat Biotechnol 22: 411417, 2004.[CrossRef][Medline]
- Niemann C, Owens DM, Hulsken J, Birchmeier W, and Watt F. Expression of
NLef1 in mouse epidermis results in differentiation of hair follicles into squamous epidermal cysts and formation of skin tumours. Development 129: 95109, 2002.[Abstract/Free Full Text]
- Niemann C, Unden AB, Lyle S, Zouboulis ChC, Toftgard R, and Watt F. Indian hedgehog and beta-catenin signaling: role in the sebaceous lineage of normal and neoplastic mammalian epidermis. Proc Natl Acad Sci USA 100: 1187311880, 2003.[Abstract/Free Full Text]
- Nishimura EK, Jordan SA, Oshima H, Yoshida H, Osawa M, Moriyama M, Jackson IJ, Barrandon Y, Miyachi Y, and Nishikawa S. Dominant role of the niche in melanocyte stem-cell fate determination. Nature 416: 854860, 2002.[CrossRef][ISI][Medline]
- Oshima H, Rochat A, Kedzia C, Kobayashi K, and Barrandon Y. Morphogenesis and renewal of hair follicles from adult multipotent stem cells. Cell 104: 233245, 2001.[CrossRef][ISI][Medline]
- Quesenberry P and Levitt L. Hematopoietic stem cells. N Engl J Med 301: 819823, 1979.[ISI][Medline]
- Rheinwald JG and Green H. Serial cultivation of strains of human epidermal keratinocytes: the formation of keratinizing colonies from single cells. Cell 6: 331343, 1975.[ISI][Medline]
- Rothnagel JA and Roop DR. Hair follicle companion layer: reacquainting an old friend. J Invest Dermatol 104: 42S43S, 1995.[Medline]
- Slack JM. Stem cells in epithelial tissues. Science 287: 14311433, 2000.[Abstract/Free Full Text]
- Spradling A, Drummond-Barbosa D, and Kai T. Stem cells find their niche. Nature 414: 98104, 2001.[CrossRef][ISI][Medline]
- Taylor G, Lehrer MS, Jensen PJ, Sun TT, and Lavker RM. Involvement of follicular stem cells in forming not only the follicle but also the epidermis. Cell 102: 451461, 2000.[ISI][Medline]
- Tumbar T, Guasch G, Greco V, Blanpain C, Lowry WE, Rendl M, and Fuchs E. Defining the epithelial stem cell niche in skin. Science 303: 359363, 2004.[Abstract/Free Full Text]
- Van Mater D, Kolligs FT, Dlugosz AA, and Fearon ER. Transient activation of ß-catenin signaling in cutaneous keratinocytes is sufficient to trigger the active growth phase of the hair cycle in mice. Genes Dev 17: 12191224, 2003.[Abstract/Free Full Text]
- Wang Z, Wong P, Langbein L, Schweizer J, and Coulombe PA. Type II epithelial keratin 6hf (K6hf) is expressed in the companion layer, matrix, and medulla in anagen-stage hair follicles. J Invest Dermatol 121: 12761282, 2003.[Abstract/Free Full Text]
- Warren R, Chestnut MH, Wong TK, Otte TE, Lammers KM, and Meili ML. Improved method for the isolation and cultivation of human scalp dermal papilla cells. J Invest Dermatol 98: 693699, 1992.[Abstract]
- Winter H, Langbein L, Praetzel S, Jacobs M, Rogers MA, Leigh IM, Tidman N, and Schweizer J. A novel human type II cytokeratin, K6hf, specifically expressed in the companion layer of the hair follicle. J Invest Dermatol 111: 955962, 1998.[Abstract]
- Wright N and Alison M. The Biology of Epithelial Cell Populations. Oxford: Clarendon Press, vol. 1, 1984.
- Xu S, Ariizumi K, Edelbaum D, Bergstresser PR, and Takashima A. Cytokine-dependent regulation of growth and maturation in murine epidermal dendritic cell lines. Eur J Immunol 25: 10181024, 1995.[ISI][Medline]
- Yuspa SH, Kilkenny AE, Steinert PM, and Roop DR. Expression of murine epidermal differentiation markers is tightly regulated by restricted extracellular calcium concentrations in vitro. J Cell Biol 109: 12071217, 1989.[Abstract]