To produce finasteride-loaded microspheres, an Inventage Lab Precision Particle Fabrication (IVL-PPF) method based on micro-electromechanical systems was used. Briefly, finasteride (Aurobindo Pharma, Ltd., Hyderabad, India) was dispersed in 20 ml dichloromethane (DCM; J.T. Baker, Phillipsburg, NJ, USA) containing the PLGA 7502A (PURAC Asia Pacific Pte. Ltd., Singapore) biodegradable polymer for organic phase solutions, and the surfactant polyvinyl alcohol (PVA; Merck KGaA, Darmstadt, Germany) was dissolved in purified water at a concentration of 0.25% to prepare a 250-ml water phase solution. The organic and water phase solutions were introduced into the microchannels of the IVL-PPF device and the finasteride-loaded microspheres were manufactured and collected in a collecting bath. To obtain the pure finasteride-loaded microspheres, stirring was continued for 3 h at 250 rpm on a magnetic stirrer at temperatures increasing from 17 to 25°C until the DCM had completely evaporated. This was followed by washing in distilled water, freezing and lyophilizing using a freeze-drier.

The prepared microspheres were characterized according to particle morphology, size, drug content and microsphere production yield. The microspheres were visualized using a scanning electron microscope (SEM; SNE-3000MS, SEC Co., Ltd., Suwon, Korea) to examine the morphology and the particle surface structure. A laser-light particle size analyzer (S3550; Microtrac, Inc., Montgomeryville, PA, USA) was used for particle size analysis. With regard to drug content, 0.1 mg/ml of finasteride test solution was prepared by dissolving 50 mg microspheres in acetonitrile (Sigma-Aldrich; Merck KGaA) to obtain 100 ml solution, which was analyzed via high-performance liquid chromatography (HPLC; ULTIMATE-3000, Thermo Fisher Scientific, Inc., Waltham, MA, USA).

Male, 5-week-old C57BL/6 mice were supplied by RaonBio, Inc. (Yongin, Korea). Solid feed (antibiotic-free) and water were sufficiently supplied until the day of the experiment, and the animals that were used for the experiment were acclimatized for 1 week to a temperature of 23±2°C, humidity of 55±10% and 12-h light/dark cycles. All experimental procedures were conducted based on the principles of laboratory animal care of the National Institute of Health (Bethesda, MA, USA) and were approved by the Ethics Committee for Laboratory Animals at Chung-Ang University (Seoul, Korea; no. 201800078).

Treatment with reagents was performed as described previously with modifications (23). The hair on the back of the male mice was removed with an electric clipper. A total of 60 mice were randomly divided into four groups of 15 mice. All mice, with the exception of those in the control group, were subcutaneously injected with 0.1 ml of 5 mg/ml testosterone propionate (TP; Tokyo Chemical Industry, Tokyo, Japan) once daily for 8 weeks. The mice in the microspheres- and finasteride-loaded microspheres-treated (15 mg/ml) groups were subcutaneously injected with 0.1 ml on the first day of the experiment, and mice in the orally-applied finasteride-treated group (0.1 mg/ml) were orally administered with 0.1 ml finasteride once daily for 56 days. Mice in the control group were not treated with TP or finasteride. After 8 weeks, hair growth was observed for a period of 2 weeks. The differences in hair growth in each group were evaluated visually and recorded weekly through image capture (Canon 3000D; Canon, Inc., Tokyo, Japan). The effect of preventing androgenic alopecia was assessed by measuring changes in the color of the mouse back skin, which indicates hair cycle transitions. Mice with completely hair-covered backs were counted as 'total' and partially hairy mice as 'partial'. Histological changes, including the number, length, anagen/telogen ratio and diameter of hair follicles, were also evaluated by hematoxylin and eosin (H&E) staining. The experiment was performed over a total of 10 weeks. After 9 and 10 weeks, the mice were sacrificed, and their skins were harvested.

Total RNA was extracted from isolated skin tissues using Tri-RNA reagent (Favorgen Biotech Corporation, Ping-Tung, Taiwan). Single-stranded cDNA was synthesized from whole RNA templates using Prime Script™ RT Master mix (Takara Bio, Inc., Tokyo, Japan). RT was performed after gently mixing the reaction solution; the solution was then incubated at 37°C for 15 min and 85°C for 5 sec; cDNA was maintained at 4°C until qPCR. The resulting cDNA was subjected to qPCR using qPCR 2X PreMIX SYBR (Enzynomics, Seoul, Korea) with the CFX-96 system (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The PCR used to amplify all genes was performed over 40 cycles (95°C for 10 sec, 60°C for 15 sec and 72°C for 30 sec), following denaturation at 95°C for 10 min. The expression levels were calculated from the quantification cycle (Cq) value using the ΔCq quantification method (24), and expression was normalized to that of 60S acidic ribosomal protein P0 (Rplp0). Amplification was performed using the following primers: Mouse Wnt3a forward, 5′-CTCGCTGGCTACCCAATTTG-3′ and reverse, 5′-CTTCACACCTTCTGCTACGCT-3′; mouse dickkopf WNT signaling pathway inhibitor 1 (Dkk1) forward, 5′-CTCATCAATTCCAACGCGATCA-3′ and reverse, 5′-GCCCTCATAGAGAACTCCCG-3′; mouse alkaline phosphatase (Alpl) forward, 5′-CCAACTCTTTTGTGCCAGAGA-3′ and reverse, 5′-GGCTACATTGGTGTTGAGCTTTT-3′; and mouse Rplp0 forward, 5′-AGATTCGGGATATGCTGTTGGC-3′ and reverse, 5′-TCGGGTCCTAGACCAGTGTTC-3′.

The isolated skin tissues were dissolved in PRO-PREP (Intron Biotechnology, Inc., Seongnam, Korea) and centrifuged at 14,000 × g for 20 min at 4°C. Protein content in the supernatants was quantified using a BCA kit (Thermo Fisher Scientific, Inc.). The proteins in the supernatant (total protein content, 30 µg) were heated for 5 min at 100°C. Each sample was subjected to 12% sodium dodecyl sulfate-poly-acrylamide gel electrophoresis, and the separated proteins were transferred onto a polyvinylidene fluoride (PVDF) membrane (EMD Millipore, Billerica, MA, USA). The PVDF membrane was blocked using 5% skim milk for 1 h at room temperature and incubated with anti-transforming growth factor (TGF)-β2 (1:1,000, cat. no. sc-90; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA), caspase-3 (1:2,500, cat. no. 9662), poly (ADP-ribose) polymerase (PARP; 1:2,500, cat. no. 9532S; both Cell Signaling Technology Inc., Beverly, MA, USA), B-cell lymphoma 2 (Bcl-2; 1:1,000, cat. no. sc-492), Bcl-2-associated X protein (Bax; 1:1,000, cat. no. sc-7480; both Santa Cruz Biotechnology, Inc.), Wnt3a (1:2,500, cat. no. ab28472; Abcam, Cambridge, UK), phospho (p)-glycogen synthase kinase 3β (GSK3β; 1:2,500, cat. no. 5558), GSK3β (1:2,500, cat. no. 9315), p-β-catenin (1:2,500, cat. no. 4176S), β-catenin (1:2,500, cat. no. 8814S; all Cell Signaling Technology, Inc.), and β-actin (1:1,000, cat. no. sc-47778; Santa Cruz Biotechnology, Inc.) antibodies at 4°C for 12 h. After washing with Tris-buffered saline with Tween^®-20 (TBST), horseradish peroxidase (HRP)-conjugated anti-mouse (1:10,000; cat. no. PI-2000) or anti-rabbit (1:10,000; cat. no. PI-1000) secondary antibodies (Vector Laboratories, Inc., Burlingame, CA, USA) specific to the primary antibody was added and incubated at room temperature for 1 h. After washing, the membranes were color-developed using electrochemiluminescence solution (EMD Millipore) and assessed using a ChemiDoc™ XRS+ system (Bio-Rad Laboratories, Inc.).

The tissue slides were prepared according to methods described previously (25), and the tissues were sliced into 6-µm sections. Deparaffinized skin sections were stained with H&E staining (hematoxylin for 5 min and eosin for 30 sec at room temperature). Following staining, the tissues were dehydrated, sealed and examined using an optical microscope (DM750; Leica Microsystems GmbH, Wetzlar, Germany). The follicular length and diameter were evaluated using the longitudinal section parallel to the direction of hair growth. The follicular number, anagen follicles (A), and telogen follicles (T) were measured, and the A/T ratio was calculated by analyzing the transverse section.

Immunohistochemistry was performed using the UltraVision LP Large Volume Detection System HRP Polymer kit (Thermo Fisher Scientific, Inc.). Anti-β-catenin (1:500, cat. no. 8814S; Cell Signaling Technology, Inc.) and anti-TGF-β2 (1:500, cat. no. sc-90; Santa Cruz Biotechnology, Inc.) primary antibodies were used for incubation at 4°C for 18 h in a moist chamber. After washing of the slides, a primary antibody enhancer was added for 10 min, and a HRP polymer solution was added for further incubation at room temperature for 15 min. After washing, the slides were color-developed using 3,3′-diaminobenzidine for 10-30 sec and were dehydrated and sealed for observation following contrast staining with Mayer's hematoxylin. After staining, tissues were observed via optical microscopy (DM750; Leica Microsystems GmbH).

To evaluate the apoptosis of hair follicles, TUNEL staining was performed using a DeadEnd™ Fluorometric TUNEL system (Promega Corporation, Madison, WI, USA). The tissue slides were deparaffinized and rehydrated, and the tissue sections were fixed in 4% methanol-free formaldehyde solution in PBS for 15 min at room temperature. After washing twice in PBS, each slide was incubated with 20 µg/ml Proteinase K for 10 min at room temperature. The tissue sections were then fixed in 4% methanol-free formaldehyde for 5 min at room temperature. After washing in PBS, the slides were incubated with a nucleotide mix and recombinant terminal deoxynucleotidyl transferase for 1 h at 37°C, avoiding exposure to light. The reaction was terminated by the addition of a stop solution for 15 min at room temperature. After washing in PBS, the tissues were mounted in VECTASHEILD mounting medium with 4′,6-diamidino-2-phenylindole (DAPI; cat. no. H-1200; Vector Laboratories, Inc.), and green fluorescence of the dead cells (fluorescein-12-dUTP) was detected. Fluorescence images were captured via confocal microscopy (LSM 700; Zeiss GmbH, Jena, Germany).

The blood samples (1.5 ml) collected from mice were centrifuged at 4°C and 1,000 × g for 30 min. Following centrifugation, the supernatant was stored at -70°C until further analysis. Serum testosterone (cat. no. CSB-E05101m) and DHT (cat. no. CSB-E07880m) were measured via enzyme linked immunosorbent assays (ELISA; Cusabio, Wuhan, China). Each sample (50 µl each in duplicate) was placed in a 96-well plate, and 50 µl of HRP-conjugate and 50 µl of antibodies (testosterone or DHT) were added to each well, followed by incubation for 1 h at 37°C. After washing three times with wash buffer, 50 µl of substrate A and 50 µl of substrate B were added and incubated for 15 min at 37°C for color development. The reaction was then terminated by the addition of 50 µl stop solution, and the absorbance was measured at 450 nm using a Spectra Max i3x Multi-Mode detection platform (Molecular Devices, LLC, Sunnyvale, CA, USA).

All statistical analyses were performed using the Statistical Package for the Social Sciences (IBM SPSS statistics version 25, IBM Corp., Armonk, NY, USA). The experiment results are expressed as the mean ± standard error of the mean, and statistical significance was evaluated via one-way or two-way analysis of variance for the levels of serum testosterone and DHT. Tukey's honest significance test was performed post hoc, and P<0.05 was considered to indicate a statistically significant difference.

The morphology and size of microspheres containing finasteride were analyzed. The SEM analysis of the prepared finasteride-loaded microspheres revealed smooth and perfectly spherical microspheres and the absence of collapsed spheres (Fig. 1A). Finasteride was encapsulated in the PLGA polymer and homogeneously prepared, and the micrographs did not exhibit any pores on the microspheres. The diameter of the prepared finasteride-loaded microspheres was ~30-50 µm as determined via SEM analysis. The particle size and distribution of the finasteride-loaded microspheres were measured using a laser diffraction particle size analyzer. The particle diameter ranged between 30 and 50 µm with a median diameter of 40 µm and a size distribution width of 9.29 µm. That indicates the majority of particles were similar in size, close to the median diameter, as shown in Fig. 1B; a narrower size distribution of particles generally improves drug release characteristics.

The contents of the prepared finasteride-loaded micro-spheres were confirmed via HPLC analysis, and the unique finasteride standard peaks (retention time, 3.57 min; peak area, 24.40) and the drug peaks (retention time, 3.52 min; peak area, 24.20) were examined (Fig. 1C-E). The drug content measured in the microspheres was 99.2%.

To investigate the effect of finasteride-loaded microspheres on AGA in the murine model, hair loss was induced in the dorsal skin of C57BL/6 mice via testosterone treatment (Fig. 2A). In the control mice, hair growth was observed from week 4 and in 80% of the mice (total hair growth, 46.6%; partial hair growth, 33.3%). However, hair began to grow from week 6 in the orally-applied finasteride and the finasteride-loaded micro-spheres treatment groups. At week 10, hair growth was observed in 86.7% of mice (total, 60%; partial, 26.7%) in the orally-applied finasteride-treated group, and 93.3% of mice (total, 60%; partial, 33.3%) in the finasteride-loaded microspheres-treated group. In the microspheres-treated group, hair grew in 33.3% of the mice (total, 6.7%; partial, 26.7%) (Figs. 2B and S1).

The levels of serum testosterone and DHT in the TP-induced mice were significantly higher than those in the control group in the first 8 weeks (Fig. 2C and D). However, the DHT levels were significantly lower in the orally-applied finasteride- and finasteride-loaded microspheres-treated groups than in the microspheres-treated group after 6 weeks. There was no difference between the orally-applied finasteride- and finaste-ride-loaded microspheres-treated groups. These results suggest that the 5α-reductase inhibitory effect of the finasteride-loaded microspheres may last for ~4-8 weeks following a single injection, and that treatment with finasteride-loaded microspheres exerts similar effects to daily orally-applied finasteride.

The tissues observed in the longitudinal and transverse sections revealed that the hair follicle tissue was in the anagen phase (Figs. 3A and B and S2). The miniaturization of hair follicles was observed in the skin sections of mice treated with TP and the microspheres, however, the number and length of hair follicles was markedly increased in the orally-applied finasteride- and finasteride-loaded microspheres-treated groups (Figs. 3C and D, S2C and D). At weeks 9 and 10, the orally-applied finasteride- and finasteride-loaded microspheres-treated mice exhibited significantly higher numbers of anagen hair follicles (A/T ratio), and the hair bulb diameters were larger than those in the microspheres-treated mice (Figs. 3C and S2C). In addition, the expression of Wnt3a increased, the protein kinase B (Akt)/GSK-3β/β-catenin signaling pathway was activated and the expression of β-catenin was increased in the newly developed hair follicles (Fig. 4A-C). Taken together, these results indicate that a single injection of finasteride-loaded micro-spheres results in marked hair growth-promoting activity and catagen-induction preventative effects, similar to those of orally-applied finasteride.

DHT leads hair follicles into the catagen phase, which leads to increased levels of TGF-β2 and the apoptosis of hair follicle cells (26). The increase in TGF-β2 alters the regulation of Bcl-2 and caspase-3, thereby affecting the apoptosis of hair cells. The expression of Bcl-2 in the finasteride-loaded microspheres-treated group was increased, whereas the expression levels of cleaved-PARP, cleaved-caspase-3, TGF-β2 and Bax were lower than those in the microspheres-treated group (Fig. 5A and B). In addition, TGF-β2 was expressed and TUNEL staining, which indicated DNA damage by apoptosis, was observed in the hair follicles of the testosterone-treated mice, in which the catagen and telogen phases were maintained. However, the expression of TGF-β2 and TUNEL staining were reduced in the orally-applied finasteride- and finasteride-loaded microspheres-treated groups, in which hair follicle growth was induced in the anagen phase (Fig. 5C and D). These results suggest that a single injection of finasteride-loaded micro-spheres interferes with the expression of apoptotic factors in testosterone-exposed hair follicles.