Reprogramming of three-dimensional microenvironments for in vitro hair follicle induction

During embryonic development, reciprocal interactions between epidermal and mesenchymal layers trigger hair follicle morphogenesis. This study revealed that microenvironmental reprogramming via control over these interactions enabled hair follicle induction in vitro. A key approach is to modulate spatial distributions of epithelial and mesenchymal cells in their spontaneous organization. The de novo hair follicles with typical morphological features emerged in aggregates of the two cell types, termed hair follicloids, and hair shafts sprouted with near 100% efficiency in vitro. The hair shaft length reached ~3 mm in culture. Typical trichogenic signaling pathways were up-regulated in hair follicloids. Owing to replication of hair follicle morphogenesis in vitro, melanosome production and transportation were also monitored in the hair bulb region. This in vitro hair follicle model might be valuable for better understanding hair follicle induction, evaluating hair growth and inhibition of hair growth by drugs, and modeling gray hairs in a well-defined environment.


Supplementary Text
Long-term culture of hair follicle germs To prepare hair follicle germs (HFGs), we suspended epithelial and mesenchymal cells (10 × 10 3 cells/0.2 mL/well at 1:1 mesenchymal:epithelial cell ratio) in the DMEM/KG2 medium, which is a mixture containing DMEM and KG2 media (1:1 ratio), 10% FBS, and 1% penicillinstreptomycin. Suspensions containing the two cell types were mixed and seeded into the wells of a non-cell-adhesive, round-bottomed, 96-well or lab-made microarray plates. After 6, 15, and 23 days of culture, morphological changes in cell aggregates were examined using a phase contrast microscope (DP-71, Olympus, Tokyo, Japan), and the length of a generated hair shaft obtained from a cell aggregate surface was quantified from phase contrast images using ImageJ software (National Institutes of Health, Bethesda, MD, USA). The morphological features of generated hair shafts, including those of the hair cortex and microfibrils, were observed using a transmission electron microscope. To evaluate the effects of FGF-2 on hair shaft generation, epithelial and mesenchymal cells were suspended in a culture medium containing 10 ng/mL of FGF-2 and seeded into a 96-well plate. After 10 days of culture, the number of hair strands generated in vitro was counted using a phase contrast microscope, and the relative expression of genes associated with hair follicle morphogenesis was evaluated via real-time reverse transcription polymerase chain reaction (RT-PCR) analysis.

Optimization of hair follicloid culture
To investigate the effects of Matrigel addition time on hair shaft generation, we supplemented the culture solution with Matrigel at different time points (after 0, 0.5, 3, 6, 15, and 22 h of culture) to ensure a 1:1 epithelial:mesenchymal cell ratio, 1×10 4 of total cells, and 1% (v/v) of Matrigel. To investigate the effects of the ratio of two cell types on hair shaft generation, we prepared five solutions containing hair follicloids, with a 9:1, 4:1, 1:1, 1:4, or 1:9 epithelial: mesenchymal cell ratio, comprising a total of 1×10 4 cells and 2% (v/v) of Matrigel. To investigate the effects of the total cell number of the two cell types on hair shaft generation, we prepared different solutions containing hair follicloids under eight different conditions. Thus, we prepared solutions with densities of 0.3, 0.6, 1.2, 2.5, 5, 10, 20, 40 ×10 4 cells/0.2 mL, which included a 1:1 epithelial:mesenchymal cell ratio and 2% (v/v) of Matrigel. To investigate the effects of Matrigel concentrations on hair shaft generation, we prepared six Matrigel solutions containing hair follicloids with concentrations of 0%, 1%, 2%, 3%, 5%, and 10% (v/v) of Matrigel, comprising 1×10 4 cells and 1:1 epithelial:mesenchymal cell ratio. To identify components of Matrigel that were effective for achieving hair shaft generation, we supplemented six components (1% laminin, 10% laminin, 1% laminin/entactin complex, 1 % collagen IV (all in v/v, from Corning [Corning, NY, USA]), 1% laminin/entactin complex + 1% collagen IV, and 1% growth factor reduced Matrigel) in DMEM/F-12 medium as a replacement for Matrigel. To prepare hair follicloids using type I collagen, we supplemented 2.4 mg/mL porcine type I collagen gel (Nitta gelatin, Morrisville, NC, USA) into 5% (v/v) in DMEM/F-12 medium as a replacement for Matrigel.

Large-scale preparation of hair follicloids
We fabricated microdevices via soft lithography processes. Briefly, microwell array configurations (diameter, 1 mm; pitch, 1.3 mm; depth, 1 mm; well number, 19 well) were designed and a mold was produced accordingly. A PDMS solution (consisting of a 10:1 mix of a pre-polymer solution and curing agent, Shin-Etsu Silicone, Tokyo, Japan) was poured onto the mold and cured in an oven at 80 °C. The thickness of the PDMS substrate at the floor of the microwells was set to 1 mm by adjusting the volume of utilized PDMS solution. The surface of the PDMS substrate, including the microwells, was modified via exposure (6-18 h) to a prevelex® solution (Nissan Chemical Corporation, Tokyo, Japan) to make it a non-cell-adhesive surface. To prepare multiple hair follicloids, cell suspensions (200 µL) containing epithelial and mesenchymal cells (total density 9.5 × 10 5 cells/mL, 1:1 ratio) were poured onto the chips. The chips were cooled at 4 °C in a freezer for at least 30 min and then incubated at 37 °C in an incubator containing 5% CO2. Hair follicloids were observed after 9 days using an inverted microscope (CKX53, Olympus, Tokyo, Japan).

Transplantation of hair follicloids
Five-week-old ICR nu/nu mice were purchased from Charles River, Japan. The animal study was approved by the committee on animal care and use at Yokohama National University (Permit numbers: 2019-04 and 2019-06). Care and handling of mice conformed to the requirements of the above-mentioned committee. The mice used in the experiments were raised under specific pathogen-free conditions and had access to chow and water ad libitum. The hair follicloids cultured for 6 days were transplanted individually into a shallow stab wound (~5 mm) prepared using a 20G ophthalmic V-lance surgical knife (Alcon, Japan) onto the ICR-nude mice that were under isoflurane anesthesia. The transplantation sites were then rubbed with Vaseline and observed every 2-3 days, and images of the generated hairs were captured using a digital camera (STYLUS TG-3 Tough, Olympus, Japan) to assess the term of hair cycles. TGFβ2, evaluated after 10 days of culture. Error bars represent the standard error calculated from three independent experiments. Numerical variables were evaluated using the Student's t-test; *p < 0.05. (E) Effects of FGF2 on hair generation. The average amount of hair generated from FGF2-treated and non-treated ssHFGs was compared after 10 days of the seeding process. Error bars represent the standard error calculated from 300 independent experiments. Numerical variables were evaluated using the Student's t-test; † p < 0.1.

Fig. S2.
Hair shafts generated by transplantation of hair follicloids. Single hair follicloids were transplanted into a shallow stab wound (~5 mm) generated on the back skin of the ICR-nude mice using a 20 G ophthalmic V-lance (Alcon, Japan). Hair shafts appeared 3 weeks after transplantation, and hair regrowth and hair loss were repeated at 3-4-week intervals for at least 10 months.  Effects of a depigmentation drug on hair follicloids. (A) Schematic representation of the methods used for screening of the depigmentation drug Nexinhib20. (B) Hair shafts generated from hair follicloids cultured with Nexinhib20 after 6 days of culture. Left panels (i) show a magnified view of the boxed areas in the right panels (ii). (C) Image of a hair follicloid cultured with Nexinhib20, obtained with a digital camera. (D) Melanin volumes in hair follicloids cultured with/without α-MSH after 6 days. Error bars represent the standard error calculated from three independent experiments. Numerical variables were evaluated using the Student's t-test; *p < 0.05. (E) Changes in gene expression-related hair pigmentation in cells from hair follicloids cultured with/without Nexinhib20. Error bars represent the standard error calculated from three independent experiments. Numerical variables were evaluated using the Student's t-test; *p < 0.05.

Fig. S5.
Effects of major/minor Matrigel components on hair follicle sprouting. (A) Hair follicle sprouting observed in components of Matrigel. Upon culturing cells in a laminin-supplemented medium, dumbbell-shaped cell structures were formed; no hair shaft generation occurred after 8 days of culture. In the presence of collagen IV and the laminin/entactin and laminin/entactin/collagen IV complexes, hair follicloids efficiently generated hair shafts. Replacement with growth factor-reduced Matrigel showed results comparable to those observed with Matrigel, suggesting that growth factors might not be crucial for hair follicle sprouting. The culture of hair follicloids with only minor constituents of Matrigel including collagen I and fibronectin efficiently resulted in hair follicle sprouting. Stereomicroscopic images were obtained after 8 days of culture. The arrows indicate hair follicle sprouting. (B) The efficiency of hair follicle sprouting. (C) A representative long sprouting hair follicle generated from hair follicloids after an extended period of culture. The hair follicloids prepared using collagen Isupplemented medium at 4 days of culture were embedded in 2.4 mg/mL collagen I in gel form.  Effects of methyl cellulose on hair follicle sprouting. Upon culturing cells in a methyl cellulosesupplemented medium, dumbbell-shaped cell structures were formed; no hair shaft generation occurred after 7 days of culture.

Fig. S8.
Effects of core-shell aggregate formation through two-step seeding on hair follicle sprouting. A core of epithelial cells was formed followed by the formation of a shell of mesenchymal cells. No hair follicle sprouting occurred with this approach, neither with nor without low concentrations of collagen.  Sprouting of pigmented hair shafts generated from hair follicloids.

Movie S4.
Sprouting of pigmented hair shafts generated from hair follicloids cultured with/without α-MSH.