The study design was reviewed and approved by the Institutional Review Board of Seoul National University Hospital (approval number C-1312-078-541), and included written informed consents from all human volunteers. All experimental procedures that used human materials were conducted in compliance with the ethical principles of the Helsinki Declaration. All animal experiments were performed in accordance with the national and institutional guidelines for animal care, at the animal facility of Seoul National University Hospital.

Lyophilized AA was procured from Cayman Chemicals (Ann Arbor, MI, USA), and dissolved in ethanol to create stock solutions of 100, 10, 5 mM, and 1 mM (1 mM=1 mmol/L). The stock solutions were stored at -20℃ until their use. For the animal experiments, 3% minoxidil (MNX) was obtained from Hyundai Pharm. Ind., Co., Ltd. (Cheonan, Korea).

Skin biopsy specimens (1.5×1.0 cm scalp tissue samples from the occipital region) were obtained from healthy male volunteers (age, 20~50 years) who did not have any current or prior scalp disease. The HFs were isolated under a dissecting stereomicroscope (Olympus, Tokyo, Japan), and the hDPCs were isolated from individual HFs that were determined to be in the anagen stage as described by a standardized technique in previous studies1415.

As previously described1516, The hDPCs were cultured at 37℃ (5% CO₂) in Dulbecco's modified Eagle's medium (DMEM; Welgene, Daegu, Korea), supplemented with 10% fetal bovine serum (Welgene), 10 ng/ml of basic fibroblast growth factor (bFGF; R&D Systems, Minneapolis, MN, USA), and antibiotic/antimycotic 1X solution (100 mg/ml streptomycin and 100 U/ml penicillin).

Viability of the hDPCs was determined by using the MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide) assay. Briefly, hDPCs (1×10⁴ cells/well) were seeded into 96-well plates, cultured for 24 hours in serum-free DMEM, and then treated for 24 hours with the vehicle (ethanol at a diluted to 1:1,000 in serum-free DMEM) or with AA (1~50 µM [final concentration], diluted to 1:1,000 in serum-free DMEM). Subsequently, 20 µl of MTT solution (5 mg/ml) was added to each well, and the plate was incubated for 3 hours at 37℃. The supernatant containing the MTT was removed, and the formazan crystals in the wells were dissolved by adding 200 µl of dimethyl sulfoxide. The samples were finally incubated for 30 minutes at room temperature before a plate reader enabled quantitative analysis for viable cells by optical density measurement at 570 nm.

Human scalp HFs were isolated and cultured as previously described17. Each HF was cut at the level of the sebaceous duct, and then cultured for 12 days at 37℃ (5% CO₂) in Williams' E medium (Gibco-BRL laboratories, Grand Island, NY, USA), supplemented with 10 ng/ml hydrocortisone, 10 µg/ml insulin, 2 mM L-glutamine, and antibiotic/antimycotic 1X solution (penicillin and streptomycin, Gibco-BRL laboratories). AA was added to culture medium at final concentrations of 1, 2, 5 µM, or 10 µM. On every third day, the culture medium was changed and the elongation of the hair shaft in each HF was measured by using a stereomicroscope (Olympus). A total of 400 HFs from 5 different volunteers (80 HFs per volunteer) were analyzed for each growth condition (based on the concentration of AA solution).

Immunohistochemistry and immunofluorescence staining were performed with 5 µm paraffin sections of human HFs, as previously described18. The antibodies used were anti-FGF-7 and anti-FGF-10 (Santa Cruz Biotechnology, Santa Cruz, CA, USA). Anti-Ki-67 (DAKO, Carpentaria, CA, USA) was used as an indicator of cell proliferation, and DAPI (4',6-diamidino-2-phenylindole) mounting media kit (Vector Laboratories, Burlingame, CA, USA) was used to counterstain the nuclei.

Total RNA was isolated from the hDPCs by using RNAiso Plus (Takara Bio Inc., Otsu, Japan) and was treated with DNase I (Roche Pharmaceuticals, Welwyn Garden City, UK) to remove the genomic DNA. Subsequently 1~2 µg of total RNA was used for the cDNA synthesis reaction, performed by using the First Strand cDNA Synthesis Kit (Fermentas, St. Leon-Rot, Germany) according to the manufacturer's instructions. Quantitative estimation of mRNA expression was performed by real-time polymerase chain reaction by using a 7500 Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) with SYBR Premix Ex Taq (Takara Bio Inc.), according to the manufacturers' instructions. All experiments with SYBR Green were performed in triplicate and were independently repeated thrice.

The total protein from the hDPCs was extracted by using RIPA lysis buffer (Millipore, Billerica, MA, USA), according to the manufacturer's instructions. Proteins were separated by using 10% and 15% sodium dodecyl sulfate polyacrylamide gel electrophoresis, and were transferred to a polyvinylidene fluoride membrane (Amersham, Buckinghamshire, UK) by using a wet transfer system. The blotted membranes were incubated at 4℃ with the appropriate antibodies: anti-total extracellular signal-related kinase (ERK), anti-phosphorylated ERK, anti-total protein kinase B (Akt), anti-phosphorylated AKT, anti-B-cell lymphoma 2 (Bcl-2), anti-total cyclic AMP response element-binding protein (CREB), anti-phosphorylated CREB (Cell Signaling Technology, Beverly, MA, USA), and anti-β-actin (Santa Cruz Biotechnology). The membranes were then probed with anti-mouse, anti-rabbit, or anti-goat IgG-horseradish peroxidase conjugates (Santa Cruz Biotechnology) for 1 hour at room temperature. Antibody-antigen complexes were then detected by using the enhanced chemiluminescence system (Amersham Pharmacia Biotech, Little Chalfont, UK).

The methods that we used for the anagen induction assay has been described in previous studies1920. Briefly, 8-week-old female C57BL/6 mice (5 mice/group) that were in the telogen phase were selected. On their shaved backs, 200 µl of AA (2%), the vehicle (negative control), or 3% MNX (positive control) were topically applied every day for a 4-week period, as in the previously reported in vivo studies with PUFA2122. The skin over the back was observed every day and photographed every week. Skin biopsies were performed from the treated area on day 28 and assessed for anagen hairs by histological examination with hematoxylin and eosin staining.

The back skin of 8-week-old female C57BL/6 mice in the telogen phase was depilated by waxing, as described in the previously reported studies2324. All HFs in the depilated area entered the anagen VI stage after day 10. The mice were divided into three groups of five mice each. Each mouse received a topical application of either AA (2%), vehicle, or MNX (3%) on its back skin every day for the next 10 days. Biopsies were performed on day 21, and stained with hematoxylin and eosin. The hair cycle score was calculated by assigning the following arbitrary scores, as described in previous studies24: HFs in anagen VI, 100; HFs in early catagen (catagen I~III), 200; HFs in mid-catagen (catagen IV~V), 300; and HFs in late catagen (catagen VI~VIII), 400. Fifty HFs were identified for each mouse and graded on the biopsy specimens. The Image J program (National Institutes of Health, Bethesda, MD, USA) was used for quantitative analysis.

Statistical significance was determined by using Student's t-test. Paired t-test was used to compare the HF culture results. All tests were two-tailed, and differences with a p-value of <0.05 were considered statistically significant. The statistical analyses were done by using the IBM SPSS Statistics 21.0 software package (IBM Co., Armonk, NY, USA).

MTT assay was used to evaluate the effect of AA on survival of hDPCs. In this experiment, AA (1~5 µM) significantly enhanced viability of hDPCs compared to the vehicle-treated controls (Fig. 1A). However, the viability of hDPCs decreased at concentrations of AA greater than 10 µM.

In addition, hair shaft elongation by treating culturing human HFs with AA (1~10 µM) for 12 days was analyzed. In this experiment, AA, at concentrations of 1, 2 µM, and 5 µM significantly enhanced hair shaft elongation after 6, 9, and 12 days, compared to the vehicle-treated controls (Fig. 1B).

To analyze the proliferation of human HFs, immunofluorescence staining was performed for Ki-67 (a proliferation marker) after HFs were cultured with AA for 3 days. The number of cells among the follicular matrix keratinocytes that were positive for Ki-67 were counted and this number was normalized by using the number of DAPI positive cells. Treatment with AA (1~5 µM) significantly increased the number keratinocytes positive for Ki-67 (Fig. 1C and D).

After treatment with AA (1, 2 µM, and 5 µM) for 24 hours, the levels of mRNA expression for FGF-7 and FGF-10, which are known to stimulate hair growth252627, exhibited a marked increase in a dose-dependent manner. In addition, the expression of hepatocyte growth factor (HGF) mRNA, another potent inducer of hair growth2829, also significantly increased (Fig. 2A). The expression of insulin-like growth factor-1 and vascular endothelial growth factor mRNA were unaffected (data not presented).

Immunohistochemical staining for FGF-7 and FGF-10 was performed following treatment of HFs with AA for 3 days. The expression of FGF-7 increased in the DP cytoplasm of 1, 2 µM, and 5 µM solutions of AA. The expression of FGF-10 increased in DP treated with 2 µM and 5 µM solutions of AA (Fig. 2B).

In addition, Western blotting was performed to analyze the expression of various signaling molecules that are related to cell proliferation and survival, such as mitogen-activated protein kinase ERK, CREB, AKT, and Bcl-2233031. For Western blotting, hDPCs on passage 4~5 under subculture were used. In this experiment, AA treatment enhanced phosphorylation of ERK, CREB, and AKT, and increased the expression of Bcl-2 (Fig. 2C and D). These results indicate that AA affects intracellular signaling cascades and prevents cell death by increasing the phosphorylation of ERK and promoting cell survival by up-regulation of AKT phosphorylation and Bcl-2 expression.

To investigate effect of AA on hair growth, 8-week-old female C57BL/6 mice were treated with one among AA, MNX, or the vehicle. It should be noted here that the hair of C57BL/6 mice enters the telogen stage at postnatal week 7, which is the optimal time to examine anagen induction without endogenous hair growth32. In this experiment, both, 2% AA and 3% MNX were able to produce a significantly higher level of hair growth at day 28, compared to the controls (Fig. 3A and B). In addition, the change in the color of the back skin and hair re-growth percent, after four weeks, were significantly greater than controls when treated with 2% AA or 3% MNX (Fig. 3B and C). These results indicate that AA and MNX are potent anagen inducers.

The back skin mice were synchronized to the anagen stage by depilation, as all HFs in the depilated skin areas enter the anagen VI stage 10 days after depilation2432. At day 21, after 10 days of treatment (Fig. 4A), histological examination revealed an increase in the size, depth, and length of the HFs that were treated with 2% AA and 3% MNX (Fig. 4B). In addition, the hair cycle scoring revealed markedly prolonged anagen hair growth in the HFs that were treated with 2% AA. However, no significant difference in hair thickness was observed between the three groups (Fig. 4C).