Introduction
Hair loss or alopecia is one of the biggest concerns of modern people regardless of gender. To treat and prevent the hair loss, the researches about the transplantation of hair follicles (HFs) [1,2] or inhibition of testosterone pathway that mediates the transformation of terminal hair into vellus hair [3] have been actively conducted. However, it can be the management not only to inhibit hair loss but also to autonomously induce hair regrowth [4].
The epidermis consists of keratinocytes, melanocytes, Langerhans cells, Merkel cells, and inflammatory cells [5], and is divided into 4 layers; basal, spinous, granular, and cornified layer [6]. Of those cells, keratinocytes are made up of 90% of the epidermis, and play a key role in forming HFs and growing hair [7,8]. To make a new HF, keratinocytes become ingrowing toward dermis by migration with active proliferating [9]. The resulting new HF has dermal papilla cells (DPCs) surrounded by keratinocytes, which exist in the bulb region of the HF, in the bottom [10]. The interaction between bulge-keratinocytes and DPCs is necessary for the proper morphogenesis of HFs and the growth of the hair shaft [11].
During the processes of HF formation and hair growth, keratinocytes are regulated by several signaling pathways such as the hedgehog (Hh), Wnt, and bone morphogenetic protein signaling pathway [12]. In particular, the previous studies have proven that the control of the Hh signaling pathway is related to the differentiation, proliferation, and migration of cancer cells, embryos, and skin-derived precursor cells [13-15].
The Hh signaling pathway is well conserved from Drosophila to human [16,17]. There exist three Hh ligands; sonic hedgehog (SHH), desert hedgehog, and Indian hedgehog. In the absence of the Hh ligands, the twelve transmembrane protein Patched1 (PTCH1) inhibits the seven transmembrane spanning protein Smoothened (SMO), thereby hampering the downstream signaling transduction cascades [18,19]. Consequently, transcription factor Glioma-associated oncogene homolog (GLI) as a repressor form moves into the nucleus and represses the expression of Hh target genes [20]. Hh ligands bind to PTCH1, weakening the inhibitory effect of PTCH1 on SMO [21]. This leads to activating downstream signaling transduction cascades of Hh pathway, and GLI as an activator form activates the expression of Hh target genes [22].
The researches on the effect of the decrease in the Hh pathway on the HF development and formation have been actively progressed. Abe and Tanaka [12] suggested that the mice lacking SHH have a reduced number of HFs and smaller DPCs. However, there have been few studies to confirm the change when the Hh signaling pathway is increased in keratinocytes and DPCs in vitro. Therefore, the present study was conducted to investigate the effect of the up-regulated Hh signaling on the cells related to hair formation.
Materials and Methods
1. Cell culture
Normal human epidermal keratinocytes (NHEK) and human follicle DPCs (HFDPCs) were purchased from PromoCell (Heidelberg, Germany). NHEK was cultured in the keratinocyte growth medium 2 (PromoCell) and HFDPCs was cultured in the follicle DPC growth medium (PromoCell) at 37℃ in an incubator containing humidified atmosphere of 5% CO2. The treatment concentration of the Hh regulators, cyclopamine (Cayman Chemical, Ann Arbor, MI, USA), an antagonist of the Hh pathway, and purmorphamine (Merck, Kenilworth, NJ, USA), an agonist of the Hh pathway, was determined with reference to the previous studies [23-25].
2. Crystal violet and alkaline phosphatase staining
The morphology of both NHEK and HFDPC was examined using crystal violet (CV) solution (Sigma-Aldrich, St. Louis, MO, USA). The cells were washed with phosphate buffered saline twice, stained with 0.25% CV solution, and placed in room temperature. After 30 minutes, the solution was washed thoroughly using tap water followed by drying the remaining water. The alkaline phosphatase (ALP) activity was confirmed by ALP staining kit II (Stemgent, Cambridge, MA, USA) as the manufacturer’s instruction. The stained cells were observed by the EVOS CL Core microscope (Life Technologies, Carlsbad, CA, USA) and counted for evaluating the changes in the cell morphology.
3. Cell viability assay
NHEKs and HFDPCs were seeded at a density of 2.0 × 103 cells per well in 96-well plate for viability assay. After 24 hours of seeding, the medium was replaced either with cyclopamine or purmorphamine at various concentrations and maintained for 72 hours. Then, the cell viability was measured by a WST-1 cell viability assay kit (Dongin LS, Seoul, Korea) as the manufacturer’s instruction.
4. Senescence assay
NHEKs were seeded at a density of 2.0 × 104 cells per well in 24-well plate for senescence assay. After 24 hours of seeding, the medium was replaced either with cyclopamine or purmorphamine at various concentrations and maintained for 72 hours. Then, the senescence of NHEK was measured by Senescence β-Galactosidase staining kit (Cell Signaling Technology, Beverly, MA, USA). Briefly, the medium was removed, and the cells were washed three times with phosphatebuffered saline (PBS). The cells were then fixed for 15 minutes with fixative solution at room temperature and rinsed three times with PBS, followed by adding β-Galactosidase staining solution. The plate was sealed with parafilm and incubated overnight at 37℃ in a dry incubator with no CO2.
5. RNA isolation and polymerase chain reaction
Total RNA was isolated using PureLink RNA Mini Kit (Invitrogen, Camarillo, CA, USA) as the manufacturer’s instruction. The quality of the isolated RNA was assessed by UV absorbance at 260/280 nm using Nanodrop ND-2000 (Thermo Fisher Scientific, Waltham, MA, USA). For reverse transcription polymerase chain reaction (RT-PCR), the extracted RNA samples were then reverse transcribed using M-MLV Reverse Transcriptase (Promega, Madison, WI, USA) according to the manufacturer’s protocol. Subsequently, 2 μL of 100 ng/μL cDNA, 2 μL of 10 μM the appropriate primer pair (Table 1), and 16 μL of RNase-free water were added into AccuPower® PCR PreMix (Bioneer, Daejeon, Korea). The PCR was initiated at 95℃ for 5 minutes, followed by 35 cycles of denaturation at 95℃ for 20 seconds, annealing at the appropriate temperature for each primer pair for 20 seconds and extension at 72℃ for 30 seconds, and finished at 72℃ for 3 minutes. For quantitative PCR (qPCR), reactions performed in triplicate with TB GreenTM Premix Ex TaqTM II (TaKaRa Bio Inc., Shiga, Japan) as the manufacturer’s protocol using StepOnePlus (Applied Biosystems, Foster City, CA, USA). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an endogenous control and for normalization of the differences in individual samples. For the gene expression analysis in NHEK, the treatment of Hh regulators was limited to 24 hours before RNA extraction due to the remarkable reduction of cell viability when the cells were exposed to Hh regulators over 24 hours.
6. Statistics
All experiments in the present study were statistically analyzed using a one-way ANOVA by Prism software 5.0 (Graph- Pad Software, San Diego, CA, USA). Differences were considered statistically significant at p-values ＜ 0.05.
Results and Discussion
1. The up-regulation of Hh pathway makes the appearance of NHEK into fibroblast-like shape
To validate the effect of Hh regulators on NHEK, the various concentration of cyclopamine or purmorphamine was treated to NHEK. The morphological change of NHEK was observed in medium and after staining using CV solution (Fig. 1A–1C), and quantitatively analyzed for each group (Fig. 1D) after 72 hour treatment. While the cells treated with cyclopamine did not show any morphological change, the cells treated with purmorphamine showed significant changes in their shape depending on the concentration of purmorphamine. Although there was no morphological change in the group treated with 0.2 μM of purmorphamine, the change of the cellular shape began to emerge from the group treated with 0.5 μM of purmorphamine. More than six times of cells, treated with 1 μM of purmorphamine, transformed into fibroblast-like cells when compared to no treatment control. The viability of NHEK after the treatment of Hh regulators was then determined by WST-1 assay (Fig. 1E). Both Hh agonist and antagonist reduced the viability of NHEK in a dose-dependent manner. In addition, purmorphamine-treated cells were less viable than cyclopamine-treated counterpart. The effect of Hh regulators on the senescence of the cells was investigated and we found that the senescence was not induced by the treatment of both regulators (Fig. 1F). Epithelium-mesenchymal transition (EMT) is a process by which the cells attached to one place break their adhesion molecules and gain migratory property. A visible characteristic of EMT is that the cells undergoing EMT lose their own cellular shape and get elongated [26]. Hh signaling is essential for the proper formation of HFs and hair growth [27,28]. The previous studies showed that the change in the morphology of NHEK may be due to EMT by the up-regulation of Hh pathway [26-28]. The process of hair growth consists of three stages; anagen (growth stage), catagen (regression state), and telogen (resting stage). Epidermal keratinocytes are invaginated toward dermis to make HF structures through migration from early to late anagen. The active proliferation of keratinocytes is also important in hair bulb formation and hair shaft elongation during anagen [9]. Senescent keratinocytes could induce hair thinning and even hair loss [29]. The present study confirms that the viability of NHEK after the treatment of Hh regulators was decreased and this was not from cell senescence. In particular, SHH plays a key role in hair growth because it can induce the transition from telogen to anagen [30,31]. Inhibition through the treatment of anti-SHH antibody and exogenously administrated supplementation of SHH led to blocking of anagen phase and stimulation of hair regrowth [32,33]. Therefore, although the aberrant control of Hh signaling may have an adverse effect on hair shaft growth in terms of proliferation of NHEK, the up-regulation of Hh signaling can help the epidermis to form HFs.
2. Hh regulators changed gene expression related to EMT in NHEK
As the phenotypic change was observed in NHEK after the treatment of Hh regulators, expression level of EMT-related genes in NHEK treated with Hh regulators was investigated. After 24 hours of Hh regulator treatment, the expression level of vimentin (Vim ), α-smooth muscle actin (α-SMA ), and fibronectin (Fn), known as EMT-related genes, were analyzed by qPCR. In the cyclopamine-treated groups, while the mRNA expression level of Fn was significantly decreased in a dosedependent manner, there was no significant change in the levels of Vim and α-SMA (Fig. 2A). In the purmorphaminetreated groups, the mRNA expression level of α-SMA was not changed significantly while the level of Fn was decreased. In addition, the level of Vim was increased 1.4 times in 1 μM purmorphamine treatment group (Fig. 2B). Fn is an extracellular matrix glycoprotein coding gene, involving in cell adhesion, migration, and even proliferation [34]. The decrease of Fn in all experiment groups may be associated with the reduction of viability by Hh regulators. Vim is a gene encoding an intermediate filament protein upregulating in the EMT-undergoing cells, serving as a positive regulator of EMT [35]. Previous studies showed that the expression of Vim was higher in the anagen stage than catagen or telogen stage [36,37]. Therefore, the result may indicate that purmorphamine induces EMT in NHEK and leads the cell to the anagen stage.
3. The morphology and the viability of HFDPC are not affected by Hh regulators
The effect of Hh regulators on HFDPCs was also investigated as with NHEK. The morphology and the viability of HFDPCs were observed after 72 hours of the treatment of both Hh regulators (Fig. 3A–3D). However, unlike the results from NHEK, any change by Hh regulators was not observed in HFDPCs. It has been known that the inhibition of Hh signaling in dermis region prevents the dermal condensation by causing loss of dermal papilla precursors, and hampers the growth of hair shaft [38] and that DPCs play a major role in the formation of HFs and growth of hair shafts. The number of DPC determines the size, shape of the hair, and even hair growth cycle. Also, the reduction of DPCs decreased in the number of follicles and associated with miniaturization of HFs [10]. The expression level of EMT-related genes was consistent regardless of the treatment of Hh regulators (Fig. 3E). Together with the result of the present study, the up- or down-regulation of Hh signaling in HFDPCs may not affect their proliferation, and also the interaction between keratinocytes and DPCs, neither.
4. The ALP activity of HFDPC is maintained consistently despite the change of Hh pathway
ALP, the marker of DPC and indicator for hair inductive ability, staining was performed in consideration of the possibility of other changes while the change in morphology or viability was not found (Fig. 4). However, the expression of ALP in HFDPCs was consistent regardless of up- or down-regulation of Hh pathway by the regulators. A previous study showed that ALP activity in DPCs exhibited the maximum level in early anagen, but declined as hair cycles progress to mid-anagen [39]. Based on this result, it is assumed that the control of Hh signaling in DPCs does not change the identity of DPCs.
Taken together, both Hh regulators in NHEK reduced the viability and, in particular, purmorphamine changed the morphology and gene expression level by EMT induction. However, because the deregulated control of Hh pathway in NHEK can induce basal cell carcinoma, the regulation of this signaling pathway should be controlled carefully [12]. Although there is a previous report that the reduction of Hh pathway through knockout of SMO in vivo failed to maintain DPC identity and hair morphogenesis [38], the regulators did not make any change in shape, viability, ALP activity, and even gene expression level of HFDPCs in culture in the present study. The results of the present study indicate that even though the alteration of the Hh signaling could support NHEK to form HFs causing migration, it did not affect the interaction between keratinocytes and DPCs to make hair shafts.
