Hypericin and berberine promote elongation of hair peg-like sprouting in hair follicle organoids via oxytocin signaling activation

Abstract

Oxytocin is a hormone produced during childbirth and lactation. Moreover, it is expressed in both men and women during physical contact with family members and pets. Oxytocin influences various organs and tissues, including the heart, digestive system, and reproductive organs. We recently demonstrated that oxytocin acts on hair follicles to promote hair growth. However, owing to its large molecular weight and poor skin permeability, small-molecule alternatives are required for its practical application as a hair growth–promoting agent. In this study, we comprehensively investigated molecules that increase oxytocin receptor expression and those that act as receptor agonists to identify compounds with oxytocin-like activity. Using in vitro 2D culture of dermal papilla cells and in silico screening based on binding affinity for the oxytocin receptor, we identified hypericin as a compound that enhances oxytocin receptor expression and berberine as an oxytocin receptor agonist. The additive effects of hypericin and berberine on hair growth promotion were further demonstrated using hair follicle organoid-based assays. These findings highlight the potential of hypericin and berberine to act as small-molecule alternatives to oxytocin and should contribute to the development of novel therapeutic strategies for hair-loss diseases such as alopecia.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-55544-3.

Introduction

Hairstyles play a crucial role in shaping the overall appearance of individuals. There is a growing demand for effective alopecia treatments across all age groups and sexes, and pharmacotherapy remains the primary approach for its management. Although several drugs have been developed to target specific types and stages of alopecia, their efficacy and potential side effects remain significant limitations¹. Recent research has therefore focused on identifying novel and safer drug candidates for the treatment of alopecia.

The skin is a hormone-producing and hormone-responding organ, and it plays homeostatic roles in the cutaneous endocrine system, for example, by providing protective responses against environmental insults^2,3. These hormones also act on skin appendages. Dermal papilla cells (DPCs) function as signaling centers that regulate hair growth cycles within hair follicles. Various hormones act on DPCs to promote hair growth or induce hair loss. For instance, the male hormone dihydrotestosterone shortens the anagen (growth) phase of hair follicles, leading to hair loss⁴. Cortisol also inhibits hair growth by reducing the expression of the hair growth promoter growth arrest specific 6 (Gas6) in DPCs⁵. Conversely, melatonin promotes hair regeneration by activating Wnt signaling in DPCs⁶, whereas serotonin enhances hair growth by upregulating the expression of trichogenic gene in DPCs⁷. Based on these findings, researchers and pharmaceutical companies have sought to develop hair growth–promoting drugs that target these hormonal pathways.

We recently demonstrated that oxytocin (OXT) possesses hair growth-promoting potential⁸. DPCs express oxytocin receptors (OXTR) and produce hair growth–related factors, such as vascular endothelial growth factor (VEGF) in response to OXT stimulation. Although OXT is a promising candidate for hair growth promotion, it has several limitations—including rapid degradation in the blood⁹, a relatively high molecular weight that prevents skin penetration (molecular weight (Mw) > 500)¹⁰, and side effects related to labor induction¹¹. Therefore, smaller, more stable compounds that can activate the OXT signaling pathway through topical application are required. We recently found that increasing OXTR expression in DPCs and using OXTR agonists can activate OXT signaling, thereby yield hair growth–promoting effects comparable to those of OXT treatment^12,13.

In this study, we aimed to identify compounds that increase OXTR expression and discover potential OXTR agonists through comprehensive in vitro and in silico primary screening of a library containing pharmaceutical active ingredients, quasi-drug active ingredients, and quasi-drug additives applicable to hair growth products. We then evaluated the effects of these candidates on the expression of hair growth–related genes in DPCs. For compounds showing efficacy in secondary screening, we conducted an in vitro hair peg–like sprouting assay using hair follicle organoids to confirm their hair growth-promoting effects. These findings provide new insights into OXT pathway activation and may facilitate the development of novel strategies for alopecia treatment.

Results

In vitro screening of OXTR activators

To identify compounds that promote OXTR expression, we screened 300 natural product-derived compounds in vitro (Fig. 1). DPCs were treated with each compound at three different concentrations (0.01, 0.1, and 1 µM) and OXTR expression levels were compared, which was followed by immunostaining for OXTR (Fig. 2a). Treatment with hypericin significantly upregulated OXTR expression (Fig. 2b). Among these, hypericin at 1 µM increased OXTR expression more than two-fold compared with that in the no treatment control (Fig. 2b,c). Therefore, hypericin was selected for further experiments to enhance OXTR expression in DPCs.

Fig. 1.

Schematic representation of the experimental procedure. Human dermal papilla cells (DPCs) were seeded in 24-well plates and cultured in medium supplemented with hypericin (oxytocin receptor (OXTR) activator) and berberine (oxytocin (OXT) agonist). Treatment with hypericin and berberine enhanced OXT signaling and increased hair inductivity. Both compounds promoted the elongation of hair-peg like sprouts, indicating that hypericin and berberine stimulated hair growth.

Fig. 2.

Oxytocin receptor (OXTR) expression in dermal papilla cells (DPCs) treated with hypericin. (a) Experimental procedure for identifying OXTR activators using a 2D DPC culture system. DPCs were seeded in 96-well plates and treated with OXTR activator candidates from 1 day to day 4 of culture. (b) Relative OXTR expression after 3 days of treatment. Error bars represent the standard error of the mean (SEM). Statistical analyses were performed using the Tukey’s test. * indicates p < 0.05. (c) Immunostaining of OXTR in DPCs treated with 1 µM hypericin.

In silico screening of OXTR agonists

To identify OXTR agonists, we performed in silico screening using the following databases: pharmaceutical active ingredients and quasi-drug active ingredients for hair growth products (24 compounds + 4 extracts), quasi-drug additives (small molecules, 320 compounds), and quasi-drug additives (extracts, 423 extracts)¹⁴. First, binding affinities for OXTR were calculated, and substances with biding affinities ≤  − 7 kcal/mol were selected. Next, we calculated the molecular access system (MACCS) Dice coefficient¹⁵ relative to known agonists LIT-001¹⁶ and WAY267464¹⁷. Compounds with a MACCS Dice coefficient > 0.6 (compared to either LIT-001 or WAY267464) were designated as OXTR agonist candidates (Fig. 3, Table 1). Among the 23 candidates identified, four compounds—cepharanthine, oxymatrine, N-feruloyl serotonin, and berberine were advanced for further validation.

Fig. 3.

In silico screening for candidate oxytocin receptor (OXTR) agonists. Binding affinity for the 7RYC site of OXTR was calculated, and compounds with a binding affinity <  − 7 kcal/mol were selected. Compounds with a molecular access system (MACCS) Dice coefficient (vs. LIT001 or WAY267464) > 0.6 were further identified as candidate OXTR agonists.

Table 1.

Binding affinity and molecular access system (MACCS) Dice coefficient.

	Binding affinity	(A) and MACCS Dice coefficient	(A) and MACCS Dice coefficient
	<  − 7 kcal/mol (A)	(vs. LIT-001) > 0.6	(vs. WAY267464) > 0.6
1. Quasi-pharmaceutical products (compounds)	6 Compounds	2 Compounds	2 Compounds
Quasi-pharmaceutical products (extracts)	7 Compounds (3 extracts)	1 Compound (1 extract)	2 Compounds (1 extract)
2. Quasi-drug excipients (compounds)	23 Compounds	3 Compounds	3 Compounds
3. Quasi-drug excipients (extracts)	169 Compounds (171 extracts)	8 Compounds (6 extracts)	14 Compounds (13 extracts)

Investigation of hair growth–promoting effects of OXTR agonists

DPCs were treated with the four OXTR agonist candidates (cepharanthine, oxymatrine, N-feruloyl serotonin, and berberine) for six days. Real-time reverse transcription polymerase chain reaction (RT-PCR) analysis was performed to measure expression of the hair growth–related genes NOG and VEGFA, which are upregulated by OXT treatment (Fig. 4a). The expression of these genes was increased by cepharanthine, oxymatrine, and berberine treatment compared with that in untreated controls (Fig. 4b), with berberine showing significantly higher upregulation than OXT treatment (Fig. 4b). RNA sequencing (RNA-seq) was then performed on DPCs cultured for six days under six conditions: cepharanthine, oxymatrine, N-feruloyl serotonin, berberine, OXT, and no-treatment control. Berberine treatment increased the expression of previously identified DPC signature genes (Fig. 4c). Kyoto encyclopedia of genes and genomes (KEGG) pathway enrichment analysis comparing berberine-treated and untreated controls revealed activation of pathways enhanced by OXT signaling, such as neuroactive ligand–receptor and cytokine–cytokine receptor interactions (Fig. 4d).

Fig. 4.

Effects of oxytocin receptor (OXTR) agonists on dermal papilla cells (DPCs). (a) Experimental procedure for OXTR agonist treatment of DPCs. (b) Expression of hair growth-related marker genes in DPCs treated with OXTR agonists. Error bars represent the standard error of the mean (SEM) from three independent experiments. Statistical analyses were performed using the Tukey’s test. p < 0.05. (c) Heatmap showing expression of trichogenic marker genes based on RNA-sequencing data. (d) Top 15 Kyoto encyclopedia of genes and genomes (KEGG) pathways enriched among upregulated differentially expressed genes following OXTR agonist treatment.

To elucidate the relationship between OXTR and hair growth–related genes, DPCs were cultured in berberine-supplemented medium containing the OXTR antagonist L-371,257 (Fig. 5a). NOG expression decreased following L-371,257 treatment (Fig. 5b), whereas VEGFA expression remained unchanged (Fig. 5b). These findings suggest that hair growth–promoting effects are likely mediated via multiple pathways. Previous studies have shown that berberine promotes hair growth by activating the Wnt/β-catenin signaling pathway¹⁸. The increase in VEGFA expression might be mediated by OXT signaling–independent mechanisms, including Wnt/β-catenin signaling. We further investigated the additive effects of hypericin and berberine (Fig. 5a). Compared with berberine alone, co-treatment enhanced the expression of hair growth–related genes in DPCs (Fig. 5c). Statistical analysis using a two-way ANOVA revealed a trend toward an interaction between hypericin and berberine. These results indicating that combining an OXTR expression enhancer (hypericin) with an OXTR agonist (berberine) amplifies OXT signaling, thereby potentiating hair growth–promoting activity.

Fig. 5.

Effects of oxytocin receptor (OXTR) antagonist/activator and berberine on dermal papilla cells (DPCs). (a) Experimental procedures for OXTR inhibition and activation in DPCs. (b) Effects of OXTR antagonist on hair growth-related marker expressions in DPCs treated with berberine. Error bars represent the standard error of the mean (SEM) from three independent experiments. Statistical analyses were performed using the Tukey’s test. p < 0.05. (c) Effects of the OXTR activator hypericin on hair growth related marker expression in DPCs treated with berberine. Error bars represent the SEM from three independent experiments. Statistical analyses were performed using the Tukey’s test. p < 0.05.

Hair–peg sprouting assay of hypericin and berberine using hair follicloids

We previously developed hair follicle organoids capable of regenerating hair shafts in vitro using mouse-derived cells¹⁹. When generated from human-derived cells, these organoids produce immature hair shafts, or hair peg–like structures, that elongate in vitro²⁰. Briefly, human DPCs and hair follicular keratinocytes were co-cultured on spheroid-forming plates in low-concentration Matrigel-containing medium. These cells initially formed single aggregates, which spontaneously generated hair peg-like sprouts through epithelial–mesenchymal interactions. Treatment with minoxidil significantly increased the length of these sprouts after six days of treatment²⁰. Using this system, we previously identified multiple drug candidates that enhanced hair peg elongation^7,8,12,13.

Here, we applied the hair peg–sprouting assay to evaluate the hair growth–promoting effects of hypericin and berberine (Fig. 6a). The elongation rate of hair peg–like sprouts increased with hypericin and/or berberine treatments, and lengths after two days were significantly greater than those in the control (Fig. 6b,c). Co-treatment with hypericin and berberine at concentrations used in DPC cell culture (Fig. 5c) resulted in no additional elongation compared with that in the berberine-alone treatment (Fig. 6c). Hair peg–sprouting assays with varied hypericin concentrations further confirmed a concentration-dependent increase in sprout length and an additive effect between hypericin and berberine (Fig. 6d). These findings suggest that hypericin and berberine may serve as effective hair growth–promoting agents for alopecia by activating the OXT signaling pathway.

Fig. 6.

Hair growth assessment using hair follicle organoids. (a) Experimental procedures for hypericin and berberine treatment in hair follicle organoids. (b) Microscopic images of hair follicle organoids cultured with or without hypericin and berberine for 10 days. Hair follicle organoids were permeabilized and imaged using a stereomicroscope. (c) Quantification of sprouting structure length with or without compound treatment. The graph shows the relative length on days 6, 8, and 10 compared to that on day 4. Fifty-six organoids per treatment condition were analyzed to calculate mean sprout length. Statistical analyses were performed using the Tukey’s test. p < 0.05. (d) Quantification of sprouting structure length with both hypericin and berberine treatment at different concentrations. The graph shows the relative length on days 6, 8, and 10 compared to that on day 4. Twenty-eight organoids per treatment condition were analyzed to calculate mean sprout length. Statistical analyses were performed using the Tukey’s test. p < 0.05.

Discussion

Hypericin is a compound found in Hypericum perforatum, commonly known as St. John’s Wort. It possesses anticancer²¹, antidepressant²², and analgesic²³ properties; however, its precise effect on hair growth has not been fully elucidated. Berberine is an isoquinoline alkaloid found in extracts of barberry and other traditional medicinal herbs²⁴. It exhibits antioxidant²⁵, anti-inflammatory²⁶, and anticancer²⁷ activities. Within hair follicles, berberine has been reported to promote hair growth by activating the Wnt/β-catenin pathway¹⁸. However, the relationship between berberine-mediated hair growth and OXT signaling remains unclear. In this study, we identified hypericin and berberine as hair growth-promoting factors associated with the activation of OXT signaling. These findings hold strong potential for enhancing the efficacy of hair growth-promoting formulations.

Cepharanthin and oxymatrine also effectively increased the expression of hair growth–related genes; however, the expression levels were not reduced when these compounds were co-administered with an OXTR antagonist (Supplemental Fig. 1), suggesting that these compounds may promote hair growth through pathways other than OXT signaling. Future studies should explore the molecular mechanisms underlying cepharanthine and oxymatrine-mediated hair growth. A combinatorial approach involving these compounds alongside hypericin and berberine could enable a multi-pathway therapeutic strategy to maximize efficacy.

The identification of berberine as a potential OXTR agonist was based primarily on in silico docking and expression analysis of downstream genes. To further assess the functional activity, we performed Ca²⁺ fluorescence imaging as an assay for the second messenger. Upon berberine treatment, DPCs exhibited Ca²⁺ oscillations (Supplemental Fig. 2), indicative of intracellular signaling activation. These findings are consistent with OXTR-mediated signaling. However, definitive confirmation of direct receptor activation by berberine will require additional studies, such as receptor-binding assays.

The concentrations required for the additive effect of hypericin and berberine differed between the 2D DPC cultures and organoid systems, likely reflecting tissue size–dependent differences in diffusion and metabolic activity. In 2D culture models, cells extend excessively on a culture surface, leading to intense proliferative stimulation and a reduction in sensitivity. In 3D culture models, more physiological cuboidal morphology and cell cycle are maintained, but responsiveness is reduced due to limited molecular diffusion within the cell aggregates. This balance may collectively account for the differences in effective concentrations between the 2D and 3D culture models. To determine the optimal conditions for alopecia treatment, further studies should optimize the concentrations of these compounds using human hair follicle organ cultures and animal models. Additionally, investigating their combined effects with existing hair growth-promoting agents such as minoxidil could provide additional benefits.

This study focused primarily on the effects of hypericin and berberine on DPCs. However, hair follicles comprise multiple cell types, including epithelial stem cells, outer root sheath cells, hair matrix cells, dermal sheath cup cells, melanocytes, and melanocyte stem cells²⁸. Evaluating the effects of hypericin and berberine on these cell populations will help determine whether their benefits extend beyond hair growth to include hair pigmentation and antiaging effects.

OXTR is widely expressed across multiple skin cell types, including keratinocytes and dermal fibroblasts²⁹. In addition, keratinocytes have been reported to produce oxytocin³⁰. These findings raise the possibility that activation of oxytocin signaling in skin cells induces the production of secondary factors that may influence hair follicle cell activity. Considering applications in hair growth–promoting products, evaluating their efficacy with regard to potential paracrine effects mediated by oxytocin agonists may be important.

A key limitation of this study is the incomplete characterization of the potential side effects associated with berberine and hypericin. Oral administration of berberine can cause some common side effects, such as stomach upset and diarrhea²⁵. Topical administration of hypericin can cause increased sensitivity to sunlight, a condition known as photosensitivity. Nevertheless, both compounds are naturally derived bioactive molecules that have been widely used in dietary supplements and cosmetic formulations, which indicates their relatively favorable safety profile when used at appropriate concentrations. In preliminary experiments, cytotoxicity assessment based on Hoechst nuclear staining revealed that more than 90% of the cells were viable following treatment with either hypericin or berberine under the experimental conditions used. These findings suggest that the concentrations employed by us exert minimal cytotoxic effects in vitro. However, given that cytotoxicity was evaluated using a single assay, further studies incorporating additional viability and functional assays, as well as in vivo evaluation, will be necessary to more comprehensively assess the safety profile of these compounds.

In conclusion, we identified hypericin and berberine as functional alternatives to OXT through in vitro and in silico screenings. Both compounds increased the expression of hair growth–related genes in DPCs and accelerated the elongation of hair peg–like structures in hair follicle organoids. These results indicate that hypericin and berberine are promising small-molecule agents that activate OXT signaling and may serve as effective therapeutic candidates for alopecia. Notably, these findings represent an early stage discovery, as our experiments were conducted in vitro and in organoid systems. Although these models are valuable, they do not fully recapitulate in vivo hair follicle biology. Therefore, further validation in ex vivo and in vivo models will be necessary to assess clinical applicability. Overall, the present findings provide a basis for the future development of topical therapies targeting OXT-related signaling pathways.

Materials and methods

Preparation of human DPCs and human hair follicular keratinocytes

DPCs were obtained from Promocell (Heidelberg, Germany), subcultured up to passage four using Dermal Papilla Cell Growth Medium (DPCGM; Promocell), and used for 2D culture to investigate OXTR agonists. Additional human DPCs were purchased from Cell applications (Sandiego, CA, USA), cultured up to passage four in Dulbecco’s modified eagle medium (DMEM) containing 10% fetal bovine serum (FBS) and used for 2D culture to investigate OXTR activators. Human hair follicular keratinocytes were obtained from Scientific cell research laboratories. DPCs obtained from Promocell at passage four and human hair follicular keratinocytes at passage two were used for hair follicle organoid experiments. Incubator conditions were maintained at 21% O₂ and 5% CO₂ at 37 °C.

Effects of OXTR activator alternatives in DPCs

DPCs (4 × 10³ cells) were suspended in 0.1 mL DMEM, supplemented with 10% FBS, and seeded into 96-well plates. After 24 h of culture, the medium was replaced with DMEM containing one of the following OXTR activator alternatives (300 natural product-derived compounds). After 3 days of treatment, OXTR expression in DPCs was evaluated using immunostaining.

Effects of OXTR agonist alternatives in DPCs

DPCs (1 × 10⁴ cells) were suspended in 0.5 mL R-STEM medium (Rohto, Japan), supplemented with 10 µM oxytocin or 2.5 µM OXTR agonist alternatives (cepharanthine, oxymatrine, berberine, or feruloyl-serotonin), and seeded into 24-well plates (Corning, Corning, NY, USA). The medium was refreshed every three days. Gene expression was assessed using RT-PCR after six days of culture.

Treatment with berberine and OXTR antagonist

DPCs (1 × 10⁴ cells) were suspended in 0.5 mL R-STEM medium, supplemented with 2.5 µM berberine and/or 10 µM L-371,257 (Medchemexpress, Monmouth Junction, NJ, USA), and seeded into 24-well plates. Gene expression in DPCs was analyzed using RT-PCR after six days of culture.

Treatment with hypericin and berberine

DPCs (1 × 10⁴ cells) were suspended in 0.5 mL R-STEM medium, supplemented with 2.5 µM berberine and/or 1 µM hypericin, and seeded into 24-well cell plates. Gene expressions were analyzed using RT-PCR after six days of culture.

Hypericin and berberine treatment of hair follicle organoids

To investigate the effects of hypericin and berberin on hair growth, DPCs (5 × 10³ cells) and human hair follicular keratinocytes (5 × 10³ cells) were suspended in 0.2 mL advanced DMEM/F-12 (Thermo fisher scientific, Waltham, MA, USA) containing 2% (v/v) Matrigel (Corning) and seeded into non-cell-adhesive, round-bottom 96-well plate (Primesurface 96U plate; Sumitomo bakelite, Japan). The medium was supplemented with 2.5 µM berberine and/or 1, 5, 10 µM hypericin for 4–10 days after seeding and 0.1 mL of the medium was replaced with fresh medium every two days. Hair sprout length was observed using an all-in-one fluorescence microscope (BZ-X810; Keyence, Tokyo, Japan).

In silico screening

Docking simulations were performed using PyRx version 0.8 with the AutoDock Vina engine³¹. The OXTR structure (PDB ID: 7RYC) was obtained from the Protein Data Bank and prepared using PyMOL version 3.1.3 by removing the co-crystallized ligand. The cleaned receptor structure was saved in PDB format. Ligands in SDF format were energy-minimized using the Universal Force Field (UFF) via Open Babel within PyRx and subsequently converted to PDBQT format for docking. The docking grid box was defined with center coordinates (x, y, z) = (140.06, 120.73, 106.30) and dimensions (size_x, size_y, size_z) = (46.93, 49.20, 47.25) Å. Exhaustiveness was set to 8, whereas all other parameters were kept at PyRx/Vina default settings. Docking was performed once for each ligand. Binding affinities and top-ranked poses were obtained through the PyRx GUI.

Compounds with binding affinities of − 7.0 kcal/mol or lower (i.e., ≤ − 7.0 kcal/mol) were selected for further analysis. Structural similarity between OXTR agonists (LIT-001 and WAY-267464) and the screened compounds were assessed in the KNIME Analytics Platform (version 5.3.3) using RDKit nodes. Similarity was calculated using Morgan and MACCS fingerprints with Tanimoto and Dice coefficients. Compounds with MACCS Dice coefficient > 0.6 relative to that of either of the known agonists, LIT-001 or WAY-267464, were selected as candidate OXTR agonists. These thresholds were not intended to define biological activity but were aimed at reducing the screening space to a manageable number of candidates.

Docking scores depend on the target structure and scoring function, and no universally accepted cutoff for biological activity has been established. Binding free energy values below − 6.0 kcal/mol, and in some cases below − 8.0 kcal/mol, are commonly used for candidate prioritization; however, consensus on activity thresholds is lacking³². Based on this range, − 7.0 kcal/mol was selected as a balanced threshold for initial screening.

Structural similarity was assessed using the MACCS Dice coefficient, with a threshold of 0.6, to retain compounds with moderate similarity to known OXTR ligands while maintaining a tractable number of candidates. Direct similarity filtering against oxytocin was not applied because oxytocin is a peptide, whereas the screened compounds are predominantly small molecules, which could introduce size- and chemotype-dependent bias. The MACCS Dice coefficients for oxytocin relative to the known OXTR ligands LIT-001 and WAY-267464 were 0.687 and 0.732, respectively, and were used only as internal reference points.

For extract-based entries, compounds were not treated as undefined mixtures. Instead, representative major constituents were selected based on reported phytochemical compositions and evaluated as chemically defined entities in both docking and similarity analyses.

Gene expression analysis

Total RNA was extracted from samples using the RNeasy Mini Kit (Qiagen, Hilden, Germany). Complementary DNA was synthesized using the ReverTra Ace RT-qPCR Kit (Toyobo, Osaka, Japan) following the manufacturer’s instructions. Quantitative RT-PCR was performed using the StepOnePlus system (Applied biosystems, Waltham, MA, USA) with SYBR Premix Ex Taq II (Takara bio) and primers specific for human NOG, VEGFA, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Primer sequences are listed in Table 2. Gene expression levels were normalized to those of GAPDH, and relative expression was calculated using the 2^−∆∆Ct method. Data are presented as mean ± standard error from three independent experiments. Statistical analyses were performed using the Tukey’s test, with p < 0.05 considered significant. To clarify the combined effects of hypericin and berberine, we performed a formal interaction analysis using two-way ANOVA.

Table 2.

PCR primer sequences.

Genes	Forward (5′–3′)	Reverse (5′–3′)
NOG	CTGGTGGACCTCATCGAACA	CGTCTCGTTCAGATCCTTTTCCT
VEGFA	ACTTCTGGGCTGTTCTCG	TCCTCTTCCTTCTCTTCTTC
GAPDH	TGGAAGGACTCATGACCACAG	GGATGATGTTCTGGAGAGCCC

Immunohistochemical staining

Samples were fixed in 4% formaldehyde for 10 min at 4 °C, blocked in PBS containing 3% goat serum and 0.1% Tween-20 for 1 h at 25 °C, and incubated for 1 h with rabbit anti-OXTR antibody (1:100; Proteintech, Rosemont, IL, USA) at 25 °C. After washing, the samples were incubated with Alexa 488-conjugated goat anti-rabbit IgG secondary antibody (1:500; Thermo fisher scientific) in blocking solution for 1 h at 25 °C and finally with Hoechst (1:1000; Dojindo laboratories, Kumamoto, Japan) in PBS for 10 min.

RNA-seq analysis

Total RNA was extracted from DPCs, with or without 10 µM oxytocin or 2.5 µM OXTR agonist alternatives (cepharanthine, oxymatrine, berberine, or feruloyl-serotonin), for six days using the RNeasy Mini Kit (Qiagen). RNA-Seq was performed by Takara bio. Genes significantly upregulated in the berberine-treated DPCs were analyzed using gene ontology and KEGG pathway analyses via the Database for Annotation, Visualization, and Integrated Discovery (DAVID; http://david.abcc.ncifcrf.gov/)^33,34.

Author contributions

T.K., K.N., O.K., S.M., and J.F. designed the experiments. K.N. conducted in vitro screening for OXTR activators. O.K. designed and carried out the methodological and computational aspects of the in silico screening for OXTR agonists. S.M. curated and organized the compound data to support the screening process. S.M. conducted in silico screening to investigate OXTR agonists. T.K. conducted gene expression analysis and hair follicle organoid experiments. Y.I. conducted Ca²⁺ imaging analysis. The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

Data availability

The datasets generated and analyzed during the current study are available in the NCBI repository, GSE315061.

Declarations

Competing interests

T.K. and J.F. are co-founders of TrichoSeeds, a company that provides hair regeneration medicine. The other authors have no competing financial interests to declare.