Phytochemical Profiling of the Halophyte Artemisia fukudo Makino and Its In Vitro Effects on Androgen‐Related Pathways Associated With Benign Prostatic Hyperplasia and Alopecia

TL;DR

This study explores the plant Artemisia fukudo and finds that its compounds may help treat prostate enlargement and hair loss.

Contribution

The study identifies jaceidin and eupatilin as novel compounds with dual therapeutic potential for BPH and alopecia.

Findings

Jaceidin inhibits 5α-reductase type 2 and downregulates BPH-related proteins in prostate cells.

Eupatilin modulates Wnt/β-catenin signaling and upregulates VEGF and IGF-1, promoting hair growth.

Artemisia fukudo contains bioactive compounds like scopoletin, jaceosidin, eupatilin, and jaceidin.

Abstract

Artemisia fukudo Makino is a biennial halophyte from the Asteraceae family, known for its tolerance to high‐salinity soils. Based on the excellent activity of A. fukudo extract in regulating the 5α‐reductase type 2 (5αR2) enzyme, this study aimed to investigate the phytochemical properties and pharmaceutical potential of A. fukudo for treating benign prostatic hyperplasia (BPH) and alopecia. Eight compounds including sesquiterpenes (1
–4), coumarin (5), and flavonoids (6
–8) were isolated from 90% MeOH fraction of A. fukudo, and their structures were determined using NMR and MS experiments. Among the isolated compounds, compounds 5–7 were detected in HPLC‐DAD, and contents of compounds 5
–8 were successfully quantified using ESI‐MS. It was found that 90% MeOH fraction of A. fukudo contained scopoletin (5), jaceosidin (6), eupatilin (7), and jaceidin (8) at the concentration of 1.44, 35.71, 6.44, and 11.19 mg/g, respectively. Compound 8 effectively inhibited the expressions of BPH‐related proteins, AR, PCNA, PSA, and 5αR2 in LNCaP and RWPE‐1 cells. Meanwhile, compound 7 exhibited potent activity in regulating Wnt/β‐catenin signaling and induced the expression of VEGF and IGF‐1 in HaCaT and HUVECs. These findings suggest that A. fukudo and its bioactive constituents represent novel natural substances for treatment or improvement of BPH and alopecia.

This study investigated the phytochemical constituents of the halophyte Artemisia fukudo, leading to the isolation and identification of eight compounds characterized by HPLC–DAD and ESI–MS. Importantly, we demonstrate for the first time that jaceidin inhibits 5α‐reductase type 2 and downregulates BPH‐associated proteins (AR, PSA, and PCNA) in prostate cells, while eupatilin modulates Wnt/β‐catenin signaling and upregulates VEGF and IGF‐1, indicating hair growth–promoting activity. These findings reveal A. fukudo as a novel natural resource with dual therapeutic potential against benign prostatic hyperplasia and hair loss.© 2026 WILEY‐VCH GmbH

Figures

Structures of 1 –8 isolated from A. fukudo.

HPLC chromatograms of the 90% MeOH fraction of A. fukudo (A) and the mixture of isolated compounds 5 –7 (B) and UV spectrum of each compound (C).

Product ion mass spectra of scopoletin (5) (A), jaceosidin (6) (B), eupatilin (7) (C), and jaceidin (8) (D) isolated from A. fukudo.

Extracted ion chromatograms of analytes quantified scopoletin (5) (A), jaceosidin (6) (B), eupatilin (7) (C), and jaceidin (8) (D) isolated from A. fukudo in MRM mode of ESIMS.

The effects of 90% MeOH fraction of A. fukudo (AFE) on the expressions of androgen receptor (AR), proliferating cell nuclear antigen (PCNA), and 5‐alpha reductase type 2 (5αR2) in RWPE‐1 cells activated with testosterone propionate (TP). Cells were treated with TP (0.5 μM) and then treated with AFE (6.25, 12.5, and 25 μg/mL) or finasteride (20 μM) for 24 h. The expression levels of AR, 5αR2, and PCNA in cells were analyzed by Western blotting. Band intensities were quantified by densitometry, normalized to α‐tubulin, and expressed relative to the TP‐treated group (set to 1). Data are presented as the mean ± SD (error bars) from three independent experiments (n = 3). Statistical analysis was performed by one‐way ANOVA followed by Duncan's multiple range test; *p < 0.05, p < 0.01, and *p < 0.001 compared to TP‐treated cells. Fina, finasteride; NC, nontreated control; TP, testosterone propionate.

The effects of 90% MeOH fraction of A. fukudo (AFE) on the expressions of androgen receptor (AR), prostate‐specific antigen (PSA), and 5‐alpha reductase type 2 (5αR2) in LNCaP cells activated with TP (testosterone propionate). Cells were treated with TP (0.5 μM) and then treated with AFE (6.25, 12.5, and 25 μg/mL) or finasteride (10 μM) for 72 h. The expression levels of AR, 5αR2, and PSA in cells were analyzed by Western blotting. Band intensities were quantified by densitometry, normalized to α‐tubulin, and expressed relative to the TP‐treated group (set to 1). Data are presented as the mean ± SD (error bars) from three independent experiments (n = 3). Statistical analysis was performed by one‐way ANOVA followed by Duncan's multiple range test; *p < 0.05, p < 0.01, and *p < 0.001 compared to TP‐treated cells. Fina, finasteride; NC, nontreated control; TP, testosterone propionate.

The effects of 90% MeOH fraction of A. fukudo (AFE) on cell viability of HaCaT cells (A) and human umbilical vein endothelial cells (HUVECs) (B). Cells were treated with AFE (6.25, 12.5, and 25 μg/mL) for 24 h, and cell viability was determined by CCK‐8 assay. Data are presented as mean ± SD (error bars) from three independent experiments (n = 3); p < 0.01 and *p < 0.001 compared to nontreated control (NC).

The effects of 90% MeOH fraction of A. fukudo (AFE) on the expressions of 5‐alpha reductase type 2 (5αR2), β‐catenin, and phosphorylation‐GSK (p‐GSK‐3β) in HaCaT cells (A) and the expressions of vascular endothelial growth factor (VEGF) and insulin‐like growth factor (IGF)‐1 in HUVECs (B). Cells were treated with AFE (6.25, 12.5, and 25 μg/mL) or MNX (10 μM, a positive control) for 24 h. The expression levels of each protein in cells were analyzed by Western blotting. Band intensities were quantified by densitometry and normalized to α‐tubulin. Data are presented as the mean ± SD (error bars) from three independent experiments (n = 3). Statistical analysis was performed by one‐way ANOVA followed by Duncan's multiple range test; *p < 0.05, p < 0.01, and *p < 0.001 compared to NC. MNX, minoxidil; NC, nontreated control.

Wild A. fukudo growing along the seashore.

Tables

TABLE 1. Optimized parameters of standards for multiple reaction monitoring (MRM)‐MS analysis.

Compounds	Precursor ion (m/z)	Product ion (m/z)	Optimized parameters
			DP	EP	CE	CEP	CXP
Scopoletin (5)	190.8	177.0	−30	−5.0	−12	−14	−6.0
Jaceosidin (6)	329.7	314.3	−20	−3.5	−16	−28	−6.0
Eupatilin (7)	343.3	328.3	−60	−5.0	−16	−14	−6.0
Jaceidin (8)	359.4	344.6	−40	−8.5	−16	−20	−6.0

TABLE 2. Parameters of the developed method for quantification of standards in LC‐ESI MS/MS analysis.

Compounds	Regression equation	R 2	Linearity range, ng/mL	LOD, ng/mL	LOQ, ng/mL
Scopoletin (5)	y = 591x + 635	0.9995	3.125–50	0.006	0.02
Jaceosidin (6)	y = 97x – 2.81e + 004	0.9974	300–1500	3.158	9.571
Eupatilin (7)	y = 47.3x + 612	0.9981	25.0–200	0.69	2.08
Jaceidin (8)	y = 96.7x + 1.06e + 003	0.9971	25.0–200	0.44	1.34

TABLE 3. Inhibitory effects of compounds 1–8 on the expression of AR, PCNA, and 5αR2 in RWPE‐1 cells activated with TP. Values are presented as the mean ± SD of triplicate experiments.

RWPE‐1 cells		AR	PCNA	5αR2
NC		0.59 ± 0.05	0.71 ± 0.05	0.47 ± 0.02
TP (0.5 μM)		1.00	1.00	1.00
Fina (20 μM)		0.83 ± 0.05*	0.94 ± 0.04	0.83 ± 0.03**
Compounds (10 μM)	1	1.31 ± 0.09	0.92 ± 0.05	0.73 ± 0.04***
	2	0.89 ± 0.06	0.93 ± 0.05	0.70 ± 0.05***
	3	0.96 ± 0.04	1.18 ± 0.02	0.75 ± 0.05***
	4	0.91 ± 0.06	0.96 ± 0.05	0.84 ± 0.04**
	5	1.21 ± 0.08	1.04 ± 0.05	0.98 ± 0.05
	6	1.19 ± 0.04	1.04 ± 0.05	0.92 ± 0.06
	7	1.07 ± 0.04	0.96 ± 0.05	0.91 ± 0.05
	8	0.55 ± 0.08***	0.72 ± 0.03***	0.49 ± 0.05***

TABLE 4. Inhibitory effects of compounds 1–8 on the expression of AR, PSA, and 5αR2 in LNCaP cells activated with TP. Values are presented as the mean ± SD of triplicate experiments.

LNCaP cells		AR	PSA	5αR2
NC		0.72 ± 0.04	0.21 ± 0.05	0.94 ± 0.03
TP (0.5 μM)		1.00	1.00	1.00
Fina (10 μM)		0.89 ± 0.07	0.83 ± 0.06	0.81 ± 0.03**
Compounds (10 μM)	1	0.74 ± 0.07**	0.69 ± 0.13**	0.51 ± 0.05***
	2	0.82 ± 0.10*	0.81 ± 0.12*	0.90 ± 0.03
	3	0.73 ± 0.04**	0.66 ± 0.13**	0.75 ± 0.08***
	4	0.79 ± 0.04*	0.71 ± 0.13	0.91 ± 0.06
	5	0.76 ± 0.08**	0.64 ± 0.13**	0.69 ± 0.06***
	6	0.96 ± 0.09	0.92 ± 0.07	0.92 ± 0.09
	7	0.68 ± 0.04***	0.72 ± 0.11*	0.85 ± 0.06*
	8	0.66 ± 0.13***	0.29 ± 0.06***	0.51 ± 0.09***

TABLE 5. Inhibitory effect of compounds 1–8 on the expression of 5αR2 and regulatory effect of Wnt/β‐catenin signaling of compounds 1–8 on the expression of p‐GSK‐3β and β‐catenin in HaCaT cells. Values are presented as the mean± SD of triplicate experiments.

HaCaT cells		5αR2	p‐GSK‐3β	β‐Catenin
NC		1.00	1.00	1.00
MNX (10 μM)		0.82 ± 0.02***	1.47 ± 0.06***	1.65 ± 0.07***
Compounds (10 μM)	1	0.84 ± 0.02	1.02 ± 0.03	1.22 ± 0.07
	2	0.75 ± 0.02***	1.08 ± 0.06	1.15 ± 0.04
	3	0.62 ± 0.02***	1.25 ± 0.05	1.23 ± 0.06
	4	0.82 ± 0.04***	1.09 ± 0.07	1.31 ± 0.05
	5	0.73 ± 0.01***	1.08 ± 0.05	1.58 ± 0.08
	6	0.69 ± 0.03***	1.27 ± 0.04	1.56 ± 0.11
	7	0.57 ± 0.03***	1.33 ± 0.04	1.91 ± 0.12***
	8	0.72 ± 0.03***	1.23 ± 0.05	1.02 ± 0.10

TABLE 6. The effects of compounds 1–8 on the expression of VEGF and IGF‐1 in HUVEC cells. Values are presented as the mean ± SD of triplicate experiments.

HUVEC cells		VEGF	IGF‐1
NC		1.00	1.00
MNX (10 μM)		2.22 ± 0.13***	3.39 ± 0.14***
Compounds (10 μM)	1	2.35 ± 0.11***	2.71 ± 0.17
	2	2.20 ± 0.08	3.53 ± 0.19***
	3	2.07 ± 0.09	3.74 ± 0.19***
	4	2.36 ± 0.13***	3.68 ± 0.21***
	5	2.34 ± 0.10***	3.90 ± 0.27***
	6	1.76 ± 0.27	3.97 ± 0.25***
	7	1.64 ± 0.08	4.76 ± 0.22***
	8	1.26 ± 0.12	4.16 ± 0.25***

Full text

Introduction

1

Halophytes are salt‐tolerant plants capable of surviving and thriving in high‐salinity water and soils (≥200 mM NaCl equivalent; sensu) [1]. Although halophytes account for about 1–2% of the entire plants, they are widely distributed across various saline habitats such as coastal wetlands, mangrove swamps, and salt deserts [2, 3]. Halophytes have evolved a variety of morphological, anatomical, and physiological adaptations to cope with the osmotic and ionic stress imposed by salinity conditions [4, 5]. To survive and thrive under such stressful conditions, halophytes have evolved a range of morphological and physiological adaptations that enable them to maintain cellular homeostasis and water uptake under salt stress. These adaptations are closely linked to the production of diverse secondary metabolites, including terpenoids, phenolics, and glycosides [2, 6]. It is well known that these metabolites with strong activity to scavenge ROS contribute to the pharmacological potential of halophytes [7, 8]. Recent studies highlighted that phenolic compounds and flavonoids from halophytes play a key role in regulation inflammatory responses and scavenging reactive oxygen species, further supporting their value as promising sources of novel bioactive agents [9, 10, 11].

Artemisia is a genus of small, aromatic herbs belonging to the family Asteraceae, comprising over 500 species distributed widely across Asia, Europe, and North America [12]. The Artemisia genus is notable for its high content of terpenoids and sesquiterpenoids, which contribute to its characteristic pungent smell and bitter taste [13]. Various species of Artemisia have been traditionally used for medicinal purposes, including the treatment of malaria, cancer, inflammation, and infections caused by fungi, bacteria, and viruses [14]. Among the Artemisia genus, Artemisia annua, Artemisia capillaris, and Artemisia argyi are prominent medicinal plants in Asia.

A. annua L., native to China, has a long history in traditional Chinese medicine for treating fever, malaria, and jaundice [15]. The representative bioactive compound contained in A. annua is artemisinin, a sesquiterpene lactone renowned for its efficacy in treating malaria [16]. Additionally, artemisinin exhibits various physiological activities, including antibacterial, antitumor, anti‐inflammatory, antioxidant, and antileishmanial properties [17, 18, 19]. A. argyi Thunb, found in Asia, Europe, and the Americas, particularly in East Asia (Korea, Japan, China, and Mongolia), has been traditionally used to treat gastrointestinal diseases and symptoms such as stomachache, gastritis, and gastric ulcers [20, 21]. A. argyi is particularly rich in polymethoxy flavones, with notable compounds including eupatilin and jaceosidin [22, 23]. Eupatilin is the predominant flavonoid in the fresh leaves of A. argyi, present in the concentrations ranging from 3.5 to 10.16 µg/g [22]. This compound has been extensively studied for its significant anti‐inflammatory, antiobesity, and gastroprotective properties [24, 25, 26]. A. capillaris has a long history of use in traditional medicine for treating conditions such as pyrexia, pain, hepatotoxicity, inflammation, cholestasis, and jaundice [27]. The main constituents of A. capillaris include coumarins, flavonoids, polyacetylenes, and chromones. Among these, scoparone (coumarin) and capillarisin (chromone) are particularly noteworthy for their hepatoprotective and bile secretion effects [28, 29].

Artemisia fukudo Makino is a biennial halophyte typically found near seamarks, reclaimed land, and along the seashores of Jeju Island and the southern part of Korea, Japan, and Taiwan [30]. Yoon et al. [30] demonstrated that the essential oil obtained from A. fukudo inhibited inflammation by blocking the activation of NF‐κB and MAPK. Though biological activities of A. fukudo including anticancer and antimelanogenic activities [31, 32] have been reported, phytochemical studies on A. fukudo remain limited.

To meet the growing market demand for benign prostatic hyperplasia (BPH) and alopecia, we attempted to explore new natural substances capable of regulating the 5‐alpha reductase type 2 (5αR2) enzyme that acts as a common biomarker in both BPH and alopecia. It was found that the extract of A. fukudo exhibited excellent inhibition on 5αR2 activity in RWPE‐1 and HaCaT cells. Inflammation has been recognized as one of the important factors in the development and progression of BPH [33]. Chronic inflammation in the prostate can induce an abnormal tissue repair response [34], leading to excessive cell proliferation and the formation of BPH nodules by promoting cell growth or limiting cell death. It acts in conjunction with androgen signaling pathways such as dihydrotestosterone (DHT) and the enzyme 5αR2 [35, 36]. Recent studies have reported that the inflammatory cell infiltration and proinflammatory cytokine secretion in the prostate are closely associated with increased prostate volume and elevated serum prostate specific antigen (PSA) levels [37, 38, 39]. Moreover, Xu et al. [40] demonstrated that inflammation and androgen pathways interact to promote prostate cell proliferation and prostate tissue remodeling. These findings suggest that targeting both mechanisms may provide a more effective therapeutic strategy for BPH and related diseases.

Considering the diverse pharmacological activities of Artemisia species and the potent inhibitory effects of A. fukudo on inflammation and 5αR2, the present study aimed to identify the bioactive constituents of A. fukudo and evaluate their potential to modulate markers associated with BPH and alopecia. Eight compounds were identified using nuclear magnetic resonance (NMR) spectroscopy. In addition, four of these compounds were quantitatively analyzed in the extract using LC‐ESI/MS. In testosterone propionate (TP)‐activated prostate cells (RWPE‐1 human prostatic epithelial cells and LNCaP human prostate adenocarcinoma), the inhibitory effects of A. fukudo extract (AFE) and its derived compounds on BPH‐related proteins including androgen receptor (AR), proliferating cell nuclear antigen (PCNA), PSA, and 5αR2 were measured. Additionally, their ability to regulate alopecia‐associated pathways, Wnt/β‐catenin signaling, and insulin‐like growth factor (IGF) and vascular endothelial growth factor (VEGF) expressions were evaluated in HaCaT (human keratinocyte) and HUVEC (human umbilical vein endothelial) cells.

Results

2

Isolation of Compounds 1–8 From A. fukudo

2.1

The 90% methanol fraction of AFE (see Materials and Methods) was subjected to repeated column chromatography to give eight compounds (1–8). The structures of 1 –8 were determined by MS and 1D and 2D NMR experiments. The eight compounds were identified as anolide (1) [41], 2α‐methoxyguaia‐3, 10(14), 11(13)‐trien‐12, 6α‐olide (2) [42], reynosin (3) [43], baynol C (4) [44], scopoletin (5) [45], jaceosidin (6) [46], eupatilin (7) [46], and jaceidin (8) [41] from their spectroscopic data by comparison with values reported in the literature. All compounds were reported for the first time from A. fukudo (Figure 1). Several of the compounds isolated in this study have been previously reported to exhibit diverse biological activities, although their relevance to BPH or hair loss has not been well explored. Reynosin (3) has been shown to inhibit 3T3‐L1 adipocyte differentiation through suppression of the MAPK/PPARγ signaling pathway [47]. Scopoletin (5) is a well‐characterized coumarin with documented anti‐inflammatory, hepatoprotective, antidiabetic, and neuroprotective activities [48]. In addition, the flavones jaceosidin (6), eupatilin (7), and jaceidin (8) are recognized as bioactive Artemisia‐derived metabolites with a broad spectrum of activities, including antioxidant, anti‐inflammatory, antiallergic, and anticancer effects, and have also been investigated for their interactions with cytochrome P450 enzymes [49, 50]. Baynol C (4) has been reported to possess antioxidant activity along with moderate acetylcholinesterase inhibitory effects [51]. In contrast, pharmacological information on anolide (1) and 2α‐methoxyguaia‐3, 10(14), 11(13)‐trien‐12, 6α‐olide (2), both belonging to the guaianolide‐type sesquiterpene lactones, remains limited. However, guaianolide‐type sesquiterpene lactones as a class are known to exhibit diverse bioactivities, including anti‐inflammatory and antiproliferative effects [52, 53], suggesting their potential relevance to the biological activities observed in the present study.

Structures of 1 –8 isolated from A. fukudo.

HPLC Chromatograms of 90% MeOH Fraction of A. fukudo

2.2

The qualitative analysis of AFE was conducted using the isolated compounds 1 –8 via high‐pressure liquid chromatography (HPLC)‐diode array detector (DAD) detected at 365 nm. Major peaks detected in chromatogram were compared to retention time and UV spectrum of each compound and identified as follows: Scopoletin (5), jaceosidin (6), and eupatilin (7) were detected with retention time of 13.57, 29.87, and 34.24 min, respectively (Figure 2).

HPLC chromatograms of the 90% MeOH fraction of A. fukudo (A) and the mixture of isolated compounds 5 –7 (B) and UV spectrum of each compound (C).

Quantitative Determination of Compounds 5–8 in 90% MeOH Fraction of AFE Using LC‐ESIMS

2.3

The content of the isolated compounds from A. fukudo was analyzed using the LC‐ESIMS in multiple reaction monitoring (MRM) mode. Prior to the analysis, the mass parameters for each compound were optimized to enhance detection sensitivity, selecting the most appropriate ion transitions: m/z 190.8 → 176.0 for scopoletin (5), m/z 329.7 → 314.3 for jaceosidin (6), m/z 343.3 → 328.3 for eupatilin (7), and m/z 359.4 → 344.6 for jaceidin (8) (Figures 3 and 4). The optimized parameters for MRM mode analysis of each compound are presented in Table 1. Calibration curves were constructed for each compound by plotting the peak area (y) against compound concentration (x) at various concentration levels. The linearity of each calibration curve was assessed using the regression coefficient (R ^2^). The calibration curves for compounds 5–8 showed excellent linearity with regression coefficients (R ^2^) higher than 0.9971 (Table 2). Using the optimized parameters, the content of 5 –8 in AFE was quantified. The results showed that scopoletin, jaceosidin, eupatilin, and jaceidin contained 1.44 ± 0.05, 35.71 ± 0.42, 6.44 ± 0.26, and 11.19 ± 0.18 mg/g, respectively.

Product ion mass spectra of scopoletin (5) (A), jaceosidin (6) (B), eupatilin (7) (C), and jaceidin (8) (D) isolated from A. fukudo.

Extracted ion chromatograms of analytes quantified scopoletin (5) (A), jaceosidin (6) (B), eupatilin (7) (C), and jaceidin (8) (D) isolated from A. fukudo in MRM mode of ESIMS.

The Effects of 90% MeOH Fraction of AFE on the Expression of BPH‐Related Proteins in TP‐Activated RWPE‐1 and LNCaP Prostate Cells

2.4

To evaluate the effects of A. fukudo to prevent BPH, the regulatory effects of AFE on the expressions of AR, PCNA, PSA, and 5αR2 were measured in prostate cell lines, RWPE‐1 and LNCaP cells. RWPE‐1 cells retain key characteristics of normal prostate epithelium and are widely used to investigate AR signaling and prostate cell proliferation under nonmalignant conditions. Therefore, in RWPE‐1 cells, AR, PCNA, and 5αR2 were analyzed to assess androgen signaling, cellular proliferation, and DHT‐producing enzyme regulation relevant to BPH progression. LNCaP cells represents an androgen‐dependent prostate model that expresses functional AR and produces PSA in response to androgen stimulation. As PSA is a well‐established downstream target of AR signaling, LNCaP cells were used to evaluate androgen‐regulated protein expression in a disease‐relevant context. Accordingly, PSA, AR, and 5αR2 were analyzed in LNCaP cells to further validate the effects of AFE on androgen signaling pathways associated with BPH. The combined use of RWPE‐1 and LNCaP cells allowed assessment of both normal epithelial and androgen‐responsive pathological conditions.

Prior to evaluating the effects of AFE, the cytotoxicity of AFE was measured using a CCK‐8 assay. The treatment of AFE showed no cytotoxicity against RWPE‐1 and LNCaP cells at the concentration range from 6.25 to 25 μg/mL (data not shown). Based on CCK‐8 assay, the inhibitory effects of AFE on the expressions of the enzymes in TP‐activated RWPE‐1 cells and LNCaP cells were measured. In both cells, finasteride was used as a positive control. Finasteride is approved for the first time clinically as a 5‐alpha reductase inhibitor for the treatment of BPH and androgenetic alopecia [54]. However, persistent sexual dysfunction and semen parameter changes are reported [55, 56, 57]. RWPE‐1 cells and LNCaP cells were activated with TP (0.5 μM) for 30 min and then treated with AFE (6.25, 12.5, and 25 μg/mL) or finasteride (20 or 10 μM). After 24‐ or 72‐h incubation, the expression levels of AR, PCNA, PSA, and 5αR2 proteins in cells were measured using Western blot. In both cells, the expressions of AR, PCNA, PSA, and 5αR2 were increased by treatment with TP. In RWPE‐1 cells, the increased expressions of the enzymes were effectively suppressed by AFE, while no significant change was found in finasteride‐treated cells. The treatment of 25 µg/mL of AFE decreased the expressions of AR, PCNA, and 5αR2 by 18%, 27%, and 23%, respectively, compared to TP‐treated cells (Figure 5).

*The effects of 90% MeOH fraction of A. fukudo (AFE) on the expressions of androgen receptor (AR), proliferating cell nuclear antigen (PCNA), and 5‐alpha reductase type 2 (5αR2) in RWPE‐1 cells activated with testosterone propionate (TP). Cells were treated with TP (0.5 μM) and then treated with AFE (6.25, 12.5, and 25 μg/mL) or finasteride (20 μM) for 24 h. The expression levels of AR, 5αR2, and PCNA in cells were analyzed by Western blotting. Band intensities were quantified by densitometry, normalized to α‐tubulin, and expressed relative to the TP‐treated group (set to 1). Data are presented as the mean ± SD (error bars) from three independent experiments (n = 3). Statistical analysis was performed by one‐way ANOVA followed by Duncan's multiple range test; *p < 0.05, **p < 0.01, and **p < 0.001 compared to TP‐treated cells. Fina, finasteride; NC, nontreated control; TP, testosterone propionate.

In LNCaP cells, it was observed that the expressions of AR, PSA, and 5αR2 were increased by TP. The increased expressions of AR, PSA, and 5αR2 were further reduced by AFE compared to finasteride‐treated cells. The expressions of AR, PSA, and 5αR2 induced by TP were decreased by 32%, 67%, and 27%. Among them, the expression of PSA was observed a significant concentration‐dependent decrease by AFE (Figure 6).

*The effects of 90% MeOH fraction of A. fukudo (AFE) on the expressions of androgen receptor (AR), prostate‐specific antigen (PSA), and 5‐alpha reductase type 2 (5αR2) in LNCaP cells activated with TP (testosterone propionate). Cells were treated with TP (0.5 μM) and then treated with AFE (6.25, 12.5, and 25 μg/mL) or finasteride (10 μM) for 72 h. The expression levels of AR, 5αR2, and PSA in cells were analyzed by Western blotting. Band intensities were quantified by densitometry, normalized to α‐tubulin, and expressed relative to the TP‐treated group (set to 1). Data are presented as the mean ± SD (error bars) from three independent experiments (n = 3). Statistical analysis was performed by one‐way ANOVA followed by Duncan's multiple range test; *p < 0.05, **p < 0.01, and **p < 0.001 compared to TP‐treated cells. Fina, finasteride; NC, nontreated control; TP, testosterone propionate.

The Effects of Compounds 1–8 Isolated From A. fukudo on the Expression of BPH‐Related Proteins in TP‐Activated RWPE‐1 and LNCaP Prostate Cells

2.5

Based on the therapeutic potential of AFE to improve BPH in TP‐activated RWPE‐1 and LNCaP cells, the effects of compounds 1–8 isolated from A. fukudo to reduce the expressions of AR, PCNA, PSA, and 5αR2 were further evaluated. In CCK‐8 assay, no significant toxicity of all compounds was observed in RWPE‐1 cells and LNCaP cells treated within the concentration range from 1 to 10 µM (data not shown). The inhibitory effects of compounds 1–8 on the expressions of the enzymes in TP‐activated RWPE‐1 cells and LNCaP cells were measured by Western blot. RWPE‐1 cells and LNCaP cells were treated with TP (0.5 µM) and then treated with each compound (10 µM) or finasteride (10 or 20 µM). After 24 or 72‐h incubation, the expression levels of AR, PCNA, PSA, and 5αR2 proteins in cells were measured using Western blot. In both cells, finasteride was used as positive control. In RWPE‐1 cells, the treatment of TP increased the expression levels of AR, PCNA, and 5αR2 by 41%, 29%, and 53%, respectively, compared to nontreated control (NC) (Table 3). The expression of 5αR2 was suppressed in cells treated with compounds 1–4 and 8 by 27%, 30%, 25%, 16%, and 51% of TP‐only‐treated cells, respectively, in which the activities of 1–3 and 8 on 5αR2 were more potent than finasteride, a positive control. The reduction of AR and PCNA expression was only observed in cells treated with 8.

In LNCaP cells, the treatment with TP induced the expression levels of AR, PSA, and 5αR2 by 28%, 79%, and 6%, respectively, compared to NC (Table 4). The expression of AR was reduced by all compounds except for 6, and the expression of PSA was inhibited by all compounds except for 2 and 6. Among them, 8 showed excellent activity to inhibit the expressions of AR, PSA, and 5αR2 by 34%, 71%, and 49%, respectively. The activity of 8 was more potent than finasteride (11%, 17%, and 19%). Based on above the results, it was shown that 8 was most potent for inhibition of AR, PCNA, and 5αR2 in RWPE‐1 cells and for AR, PSA, and 5αR2 in LNCaP cells.

Based on the results, the most significant reduction in the expressions of AR, PSA, and 5αR2 was observed in LNCaP cells treated with 1 and 8.

The Effects of 90% MeOH Fraction of AFE on the Expression of Alopecia‐Related Proteins in HaCaT and HUVEC Cells

2.6

Based on the effect of AFE on BPH‐related proteins including 5αR2, the therapeutic potential of AFE on regulating alopecia was further evaluated in HaCaT and HUVEC cells. For alopecia‐related experiments, HaCaT human keratinocytes and human umbilical vein endothelial cells (HUVECs) were selected to reflect epithelial and vascular components of hair biology, respectively. HaCaT cells are commonly used as a surrogate model for hair follicle‐associated epithelial signaling and androgen‐related responses in skin research. In HaCaT cells, 5αR2, phosphorylated glycogen synthase kinase‐3β (p‐GSK‐3β), and β‐catenin were analyzed, as these markers are closely associated with androgen metabolism and Wnt/β‐catenin signaling pathways that regulate hair follicle growth and cycling. HUVECs were included to assess the effects of AFE on growth factor‐mediated signaling relevant to hair follicle vascularization and nourishment. Adequate blood supply and angiogenic signaling are critical for hair follicle development and maintenance. Therefore, vascular endothelial growth factor (VEGF) and insulin‐like growth factor‐1 (IGF‐1) were evaluated in HUVECs as representative markers of endothelial‐derived growth factors involved in hair biology and follicular support. As shown in Figure 7, the treatment of AFE exhibited no toxicity against HaCaT and HUVEC cells. Additionally, the viability of HaCaT cells was slightly increased upon AFE treatment. In both cells, minoxidil (MNX), a widely used drug for prevention and treatment of alopecia, was used as a positive control. HaCaT cells and HUVEC cells were treated with AFE (6.25, 12.5, and 25 µg/mL) or MNX (10 µM) for 24 h; then, expression levels of 5αR2, p‐GSK‐3β, β‐catenin, VEGF, and IGF‐1 proteins in cells were measured using Western blot. In HaCaT cells, treatment with AFE increased the levels of phosphorylated GSK‐3β and β‐catenin. Furthermore, AFE significantly inhibited the expression of 5αR2 at a concentration of 12.5 µg/mL, resulting in a 33% reduction compared to the NC (Figure 8A). In HUVEC cells, the expressions of VEGF and IGF‐1 were significantly upregulated by AFE compared to the NC (Figure 8B).

*The effects of 90% MeOH fraction of A. fukudo (AFE) on cell viability of HaCaT cells (A) and human umbilical vein endothelial cells (HUVECs) (B). Cells were treated with AFE (6.25, 12.5, and 25 μg/mL) for 24 h, and cell viability was determined by CCK‐8 assay. Data are presented as mean ± SD (error bars) from three independent experiments (n = 3); **p < 0.01 and **p < 0.001 compared to nontreated control (NC).

*The effects of 90% MeOH fraction of A. fukudo (AFE) on the expressions of 5‐alpha reductase type 2 (5αR2), β‐catenin, and phosphorylation‐GSK (p‐GSK‐3β) in HaCaT cells (A) and the expressions of vascular endothelial growth factor (VEGF) and insulin‐like growth factor (IGF)‐1 in HUVECs (B). Cells were treated with AFE (6.25, 12.5, and 25 μg/mL) or MNX (10 μM, a positive control) for 24 h. The expression levels of each protein in cells were analyzed by Western blotting. Band intensities were quantified by densitometry and normalized to α‐tubulin. Data are presented as the mean ± SD (error bars) from three independent experiments (n = 3). Statistical analysis was performed by one‐way ANOVA followed by Duncan's multiple range test; *p < 0.05, **p < 0.01, and **p < 0.001 compared to NC. MNX, minoxidil; NC, nontreated control.

The Effects of Compounds 1–8 Isolated From A. fukudo on the Expression of Alopecia‐Related Proteins in HaCaT and HUVEC Cells

2.7

Based on the regulatory effects of AFE on alopecia in HaCaT and HUVEC cells, the effects of compounds 1–8 isolated from A. fukudo to regulate the expressions of p‐GSK‐3β, β‐catenin, 5αR2, VEGF, and IGF‐1 were further evaluated. In CCK‐8 assay, all compounds exhibited no significant toxicity in HaCaT cells and HUVEC cells within the concentration range from 0.1 to 10 µM (data not shown). HaCaT and HUVEC cells were treated with each compound (10 µM) or MNX (10 µM). After 24‐h incubation, the expression levels of 5αR2, p‐GSK‐3β, β‐catenin, VEGF, and IGF‐1 proteins in cells were measured. In HaCaT cells, all compounds except 1 significantly suppressed the expression of 5αR2, with reduction ranging from 18% to 43% compared to the NC. Additionally, β‐catenin expression was significantly induced by 7 (Table 5).

In HUVEC cells, treatment with 1, 4, and 5 increased VEGF expression by 134%–136%. All compounds except 1 significantly induced the IGF‐1 expression, with increases ranging from 253% to 376%, with remarkable induction in cells treated with 7. These results indicate that compound 7 effectively regulates Wnt/β‐catenin signaling and induces VEGF and IGF‐1 expression, suggesting its potential to prevent or treat hair loss (Table 6).

Materials and Methods

3

Instruments and Reagents

3.1

High‐resolution‐electrospray ionization (ESI) mass spectra were measured on a SCIEX X500R mass spectrometer (Sciex, Framingham, MA, United States). NMR spectra were recorded on a Varian VNMRS 500 NMR spectrometer (Varian, Palo Alto, CA, United States) operating at 500 MHz (^1^H) and 125 MHz (^13^C), with chemical shifts of the proton, and carbon spectra measured in methanol‐d4 solution were reported in reference to residual solvent peaks at 3.30 and 49.0 ppm, respectively. Medium‐pressure liquid chromatography used an Isolera (Biotage, Uppsala, Sweden) equipped with a UV detector at 254 and 365 nm. HPLC system was performed using UltiMate 3000 (Dionex, Thermo Scientific, Germany) equipped with a pump (HPG‐3200SD), autosampler (WPS‐3000TSL analytical), temperature controller (TCC‐3000SD), and DAD (DAD‐3000). HPLC system was equipped with pump (Waters 515 HPLC pump) and RI detector (Spectra System RI‐150, Thermo Electron Corporation). LC‐ESI‐MS/MS analysis was performed using an Agilent HPLC 1100 series (Santa Clara, CA, United States) and an ABSciex API 3200 mass spectrometer (Applied Biosystems, MDS Sciex, Concord, Canada) equipped with an ESI‐Source. Analyst software (version 1.6.3) was used to process the data obtained using HPLC and MS/MS. The column was used YMC‐Triart C_18_ (50 × 2.1 mm I.D. S‐3 μm, 12 nm, Japan). Column chromatography was performed with a silica gel 60 (Merck Art: 9385, Darmstadt, Germany) and ODS gel (COSMOSIL 140C18‐OPN). Compounds were isolated and purified using HPLC with Thermo Acclaim Polar Advantage II C_18_ (5 μm, 4.6 × 250 mm) and HECTOR‐M C_18_ (5 μm, 4.6 × 250 mm). Methanol, n‐hexane, and CH_2_Cl_2_ solvents were purchased from Daejung Chemical & Metals Co. Ltd. (Siheung‐si, Republic of Korea). HPLC‐grade methanol, acetonitrile, and water were purchased from Honeywell Burdick and Jackson (Muskegon, MI, United States). Formic acid (HPLC grade) was purchased from FUJIFILM (Osaka, Japan).

Plant Materials

3.2

The aerial parts of A. fukudo were collected in Sinan‐gun, Jeollanam‐do, South Korea, in July 2018. Plant identification was authenticated by Prof. Min Hye Yang, Pusan National University, Korea. A voucher specimen (voucher number GNTEX‐23) has been deposited in the Laboratory of Pharmacognosy, Gyeongsang National University, Korea. The dried aerial parts of A. fukudo were sealed in tightly closed vials and stored at −20°C (Figure 9).

Wild A. fukudo growing along the seashore.

Extraction and Isolation

3.3

The freeze‐dried aerial parts of A. fukudo (7.39 kg) were ground and then extracted three times for 3 h with 80% methanol using ultrasonic apparatus. After removal of the solvent in vacuo, 1.35 kg of the crude extract (yield: 18.27%, w/w, relative to the freeze‐dried plant material) was obtained. The methanolic extract was suspended in dH_2_O and partitioned with CH_2_Cl_2_ (312.0 g; 4.22%, w/w, relative to the freeze‐dried plant material). CH_2_Cl_2_ fraction was suspended in 90% methanol and partitioned with n‐hexane. The n‐hexane and 90% methanol fraction (AFE) were concentrated in vacuo and yielded 103.9 g (1.41%, w/w, relative to the freeze‐dried plant material) and 208.1 g (2.81%, w/w, relative to the freeze‐dried plant material), respectively. Then, the extract and each fraction were stored at −20°C before use. Among the four fractions, 90% MeOH soluble fraction showed the most potent inhibition on 5αR2 activity (data not shown). The 90% MeOH fraction was subjected to silica gel column chromatography using mixtures of n‐hexane‐EtOAc to EtOAc‐MeOH of increasing polarity as eluents to give 64 fractions (M1 ~ 64). M35 fraction was subjected to ODS gel chromatography with MeOH‐Water gradient (20% MeOH to 100% MeOH) to obtain six fractions (M35–1~6). Compounds 3 (8.7 mg) and 8 (2.0 mg) were isolated from M35‐4 by additional C_18_ reverse‐phase HPLC (Hector‐M, 4.6 × 250 mm, MeOH‐Water 60:40, 1.0 mL/min). M35–4–8 fraction was subjected to repeated C_18_ reverse‐phase HPLC (Thermo Acclaim Polar Advantage II, 4.6 × 250 mm, MeOH‐Water 60:40, 1.0 mL/min) to get compounds 2 (5.2 mg) and 7 (1.3 mg). ODS gel chromatography of fraction M37 with a gradient elution of MeOH‐Water (20% MeOH to 100% MeOH) afforded six fractions (M37‐1–6). Compounds 1 (6.9 mg) and 5 (23.7 mg) were isolated from M37‐2 fraction by C_18_ reverse‐phase HPLC (Hector‐M, 4.6 × 250 mm, MeOH‐Water 37:63, 1.0 mL/min). Compound 4 (0.7 mg) was obtained from M37‐2–15 fraction by additional purification on C_18_ reverse‐phase HPLC (Hector‐M, 4.6 × 250 mm, MeOH‐Water 40:60, 1.0 mL/min). M37‐4 fraction was further purified by C_18_ reverse‐phase HPLC (Hector‐M, 4.6 × 250 mm, MeOH‐Water 60:40, 1.0 mL/min) to yield compound 6 (5.2 mg).

HPLC‐DAD Analytical Conditions

3.4

The 90% MeOH fraction of AFE or the mixture of isolated compounds was dissolved in methanol. Each solution was filtered using a 0.45 μm PVDF membrane filter (Hyundai Micro, PVDF, 25 mm) and analyzed using HPLC‐DAD. Chromatographic separation of the 90% MeOH fraction of A. fukudo and the mixture of isolated compounds were performed on a YMC‐Triart C_18_ (4.6 × 150 mm) using elution with water containing 0.1% formic acid (A) and methanol (B) as mobile phase under the following gradient conditions: 0–15 min linear from 20% to 50% B and 15–45 min linear from 50% to 75% B. The flow rate was 0.8 mL/min at 30°C and the eluent was detected at 365 nm.

LC‐ESIMS Analytical Conditions

3.5

LC‐ESIMS equipment consisted of Agilent 1100 series (Santa Clara, CA, United States). A YMC‐Triart C18 (2.1 × 50 mm, S‐3 μm, 12 nm) was used and column temperature was maintained at 35°C. The mobile phase was composed of water (A) and acetonitrile (B) containing 0.1% formic acid. Flow rate was 0.3 mL/min with a gradient elution: 0–2 min, 10% B; 2–8 min, 10%–90% B; and 8–10 min, 100% B. The injection volume was 1.0 μL. The mass spectra of MS/MS analysis were performed through MRM mode after optimization using the isolated compounds. Ionization was performed using ESI with Z‐spray, argon gas was used as collision gas, and nitrogen gas was used as the nebulizer gas. The mass spectrometer was operated in negative ion mode. The mass parameters were as follows: ion spray voltage, −4500 V; ion source temperature, 500°C; nebulizer gas flow, 50 L/min; curtain gas flow, 20 L/min; and auxiliary gas flow, 55 L/min. Data acquisition and analysis were controlled using Analyst software (version 1.6.3). The optimized parameters for MRM mode analysis of each compound are presented in Table 6.

Preparation of Samples for LC‐ESIMS Analysis

3.6

Twenty milligrams of AFE were dissolved in 1 mL of methanol to prepare a stock solution (20 mg/mL). The stock solution was further diluted with methanol to obtain a working solution of 20 μg/mL prior to LC analysis. Each compound isolated from A. fukudo was prepared by dissolving in methanol to a stock concentration of 1 mg/mL. The stock solutions were further diluted with methanol to various concentrations for calibration curve preparation: scopoletin (5), 3.125–50 ng/mL; jaceosidin (6), 300–1500 ng/mL; eupatilin (7); and jaceidin (8), 12.5–200 ng/mL. Sample solution was filtered through a 0.2 μm nylon filter prior to injection for quantitative analysis and stored at 4°C before analysis.

The Limits of Detection and Quantification (LOD and LOQ)

3.7

The limit of detection (LOD) and limit of quantification (LOQ) were determined from calibration standard solutions and are expressed as solution‐based concentrations (ng/mL). In contrast, compound contents in samples are reported on a dry AFE weight basis (mg/g); therefore, these values are presented using different reporting bases. The LOD value was determined using the formula 3.3σ/S, where σ represents the standard deviation (SD) of the response and S is the slope of the calibration curve. The LOQ value was calculated based on the expression 10σ/S.

Cell Cultures

3.8

RWPE‐1 cell (human prostatic epithelial cell line; RRID: CVCL‐3791) and LNCaP cell (human prostate adenocarcinoma cell line; RRID: CVCL‐0395) were purchased from American Type Culture Collection (ATCC, Manassas, VA, United States; CRL‐11609D and CRL‐1740). RWPE‐1 cells were cultured in keratinocyte serum‐free (K‐SFM, Gibco BRL, Grand Island, NY, United States) supplemented with 0.05 mg/mL bovine pituitary extract, 5 ng/mL epidermal growth factor, and 1% (v/v) antibiotics (PS, 100 U mL^−1^ penicillin and 100 μg mL^−1^ streptomycin, Sigma–Aldrich Inc. P4333). LNCaP cells were cultured in RPMI 1640 medium (Gibco BRL, Grand Island, NY, United States) supplemented with 10% (v/v) fetal bovine serum (FBS) and 1% (v/v) PS. HaCaT cell (human keratinocyte cell line) were obtained from KB cosmetic (Jinju‐si, Republic of Korea) and were cultured in Dulbecco's modified Eagle's medium supplemented with 10% (v/v) FBS and 1% (v/v) PS. HUVEC (human umbilical vein endothelial cells; RRID: CVCL‐2959) were purchased from Thermo Fisher Scientific (United States; C0035C) and were cultured in Endothelial Cell Basal Medium‐2 (EBM‐2, Lonza, Walkersville, United States) supplemented with EGM‐2 Single Quots supplement (Lonza, Walkersville, United States). The cells were incubated at 37°C and 5% CO_2_ saturation.

Estimation of Cell Viability

3.9

Cell viability was measured using the CCK‐8, Cell Counting Kit‐8 (Dojinjo Molecular Biology, Inc., Kumamoto, Japan), according to the manufacturer's instructions. Briefly, cells were seeded at a density of 1 × 10^4^ cells/well in 96‐well plates and incubated for 24 h. Then, the cells were treated with each compound at various concentrations for 24 h. After incubation, 10 μL of CCK‐8 solution was added to each well and incubated for 3 h at 37°C. The absorbance was measured using a microplate reader at 450 nm. The experiments were performed in triplicate.

Western Blotting

3.10

Total cell lysates were isolated as described above. RWPE‐1, LNCaP, HaCaT, and HUVEC cells were seeded at a density of 6 × 10^5^, 4 × 10^5^, 3 × 10^5^, and 3 × 10^5^ cells/well, respectively, in 6‐well plates and incubated overnight. The cells were treated with minoxidil (10 μM); TP (0.5 μM); finasteride (10 or 20 μM); 6.25, 12.5, and 25 µg/mL of AFE; and 10 μM of each compound for a further 24 or 72 h. The cells were washed three times with cold phosphate‐buffered saline, and cell lysates were extracted with a lysis buffer (M‐PER Mammalian Protein Extraction Reagent 78501, Thermo Scientific, Waltham, MA, United States) containing a protease inhibitor cocktail (Thermo Scientific, Waltham, MA, United States) on ice for 1 h. Protein extracts were centrifuged at 13,000 rpm for 20 min at 4°C. The protein contents of cell lysates were quantified by Bradford assay. Twenty micrograms of harvested proteins was separated using 10% SDS‐PAGE at 100 V and transferred to polyvinylidene fluoride (PVDF) membranes. The membranes were blocked with 5% skim milk for 1 h in room temperature. Then, membranes were incubated with 1:1000 diluted primary antibodies against AR, PCNA, PSA, GSK‐3β, and p‐GSK‐3β (Cell Signaling Technology, Inc., Danvers, MA, United States), 5αR2, β‐actin, VEGF, IGF‐1, β‐catenin, and α‐tubulin (Santa Cruz Biotechnology, Santa Cruz, CA, United States) at 4°C overnight. After washing three times with Tris‐buffered saline with Tween, the immunoreactive bands were visualized by using immuno‐pure peroxidase‐conjugated mouse antirabbit IgG and goat antimouse IgG (1:10,000 dilution; Santa Cruz Biotechnology, Santa Cruz, CA, United States). Membranes were incubated with secondary antibodies for 1 h at room temperature. Protein bands were visualized using ECL solution (Bio‐Rad Clarity Max Western ECL Substrate, Bio‐Rad, Hercules, CA, United States) and calibrated using the Chemidoc Imaging System (Fusion FX5, Vilber Lourmat, France). The density level of the protein bands was normalized to α‐tubulin.

Statistical Analysis

3.11

Data were analyzed using Prism version 6 (GraphPad Software, San Diego, CA, United States). Significant differences between the two groups were analyzed using one‐way ANOVA followed by Duncan's multiple range test for multiple comparisons, with p < 0.05 considered statistically significant. Data from three independent experiments are expressed as the mean ± SD.

Discussion and Conclusions

4

BPH, defined as histological overgrowth of prostate epithelial tissue [58], has a prevalence of approximately 60% in men over 60 years of age [59]. In BPH patients, the proliferation of stromal and epithelial prostate tissue surrounding the urethra leads to urethral constriction, causing lower urinary tract symptoms (LUTS), such as frequent urination, urgency, weak urine flow, and urinary retention [60]. It is well known that LUTSs are commonly associated with BPH [61]. In prostate cells, testosterone is converted to the more active DHT by the enzyme 5αR2. DHT binds to the AR, activating the transcription of target genes and contributing to the development of BPH [62]. Additionally, in the prostate tissues of patients with BPH or prostate cancer, a proliferation marker PCNA which is associated with DNA damage and repair, is significantly increased [63]. PSA, one of the most widely used markers for prostate cancer, is also correlated with BPH [64]. PSA levels have been reported to increase in men with BPH as the prostate enlarges. In this study, using TP‐activated RWPE‐1 and LNCaP cells, the inhibitory effects of AFE on BPH‐associated proteins, including PSA, AR, 5αR2, and PCNA were assessed. As a result, AFE significantly inhibited the expression of these proteins which were comparable or even better than positive control, finasteride. Among the isolated compounds, 8 showed the strongest inhibitory activity on AR and 5αR2 in RWPE‐1 cells. Additionally, 1 –4 and 8 significantly decreased PSA expression, with 8 outperforming finasteride. In LNCaP cells, 8 also demonstrated superior suppression of AR, PSA, and 5αR2 compared to finasteride. These in vitro findings indicate that A. fukudo and its bioactive compounds, particularly compound 8 (jaceidin), warrant further investigation for their potential relevance to BPH. Although the quantified levels of compounds 5–8 were in the ng/g range, such low‐abundance metabolites can still be biologically relevant, as many flavonoids and sesquiterpene lactones are known to exert cellular effects at low micromolar or submicromolar concentrations. The high sensitivity of the ESIMS–based quantitative method employed in this study was therefore essential to reliably detect and compare these bioactive constituents. Importantly, the quantitative data do not imply direct in vivo exposure levels but rather provide a comparative chemical basis to support the observed in vitro biomarker modulation and to guide future pharmacological investigations.

While hair loss (alopecia) does not typically affect general health, it can significantly impact self‐confidence. Individuals with alopecia often experience more frequent psychiatric disorders such as depression, anxiety, or social phobia compared to the general population [65]. Preventing the death of hair follicle cells and regulating the hair growth cycle are crucial aspects of hair loss therapy [66, 67]. The cycle of hair growth consists of three stages including anagen, catagen, and telogen stages [68]. Hair growth factors such as VEGF, IGF‐1, epidermal growth factor (EGF), fibroblast growth factor (FGF‐5), and FGF‐7 play important roles in regulating hair growth [69, 70, 71, 72], and dysregulation of the growth cycle is known to contribute to hair loss. Yano et al. [73] reported that VEGF is an important regulator of hair follicle growth and circulation, promoting hair growth and increasing hair follicle and hair size through improved follicular vascularization. In addition, the blood vessels around hair follicles have been shown to play an important role in hair growth due to the activation of VEGF, which is found in hair follicles, sebaceous glands, dermal papilla cells, keratinocytes, and other supporting cells [74]. IGF‐1 plays a crucial role in regulating the hair cycle and hair shaft differentiation during hair follicle development [75]. Knockout mice lacking the IGF‐1 receptor exhibit a reduced absolute number of hair follicles, abnormal hair follicle patterns, and impaired hair differentiation [76]. Especially, the Wnt/β‐catenin signaling pathway is recognized for its critical role in hair follicle development, regeneration, and growth [77, 78, 79]. In the canonical Wnt/β‐catenin signaling pathway, Wnt proteins bind to the Frizzled receptor and the coreceptor low‐density lipoprotein receptor‐related protein, leading to the inactivation of glycogen synthase kinase‐3β (GSK‐3β). GSK‐3β functions as an enzyme that phosphorylates β‐catenin, leading to its subsequent ubiquitination and degradation. Wnt‐mediated inactivation of GSK‐3β results in the stabilization of β‐catenin in the cytoplasm. This phosphorylation prevents GSK‐3β from targeting cytosolic β‐catenin for ubiquitin‐dependent degradation, allowing β‐catenin to accumulate in the cytoplasm and translocate into the nucleus [80]. Previous studies have shown that hair follicle neogenesis depends crucially on the activation of Wnt/β‐catenin signaling in DP cells and keratinocytes, along with activation of hair follicle stem cells [78, 81, 82, 83]. Previous studies have revealed that androgens exert varying effects on hair follicles depending on the region, with appropriate concentrations promoting hair growth and excessive concentrations inhibiting hair growth [84]. Additionally, dermal papilla cells (DPCs) have been shown to induce hair follicle stem cells (HFSCs) to differentiate into hair follicles, while DHT inhibits HFSCs differentiation by disrupting the Wnt pathway in a coculture model involving human DPC and HFSCs. Also, it was reported that DHT can reduce the expression of p‐GSK‐3β and β‐catenin in the hair follicles of patients with AGA [85, 86]. In this study, we investigated the potential of AFE and its isolated compounds to prevent hair loss by inhibiting 5αR2, and regulating Wnt/β‐catenin signaling in HaCaT cells. Additionally, the expressions of VEGF and IGF‐1, growth factors related to hair growth, were measured in HUVEC cells. As shown in Results, AFE upregulated the Wnt/β‐catenin signaling in HaCaT cells and markedly increased the expressions of VEGF and IGF‐1 in HUVEC cells compared to the positive control, MNX. Among the isolated compounds, 7 exhibited the strongest inhibitory activity on 5αR2 and effectively upregulated Wnt/β‐catenin signaling in HaCaT cells, outperforming MNX. In HUVEC cells, VEGF expression was increased by the treatment with 1, 4, and 5, with greater efficacy than MNX. All compounds, except 1 and 4, significantly increased IGF‐1 expression. Although the regulatory activity of the compounds varied across the two hair loss‐related cell lines and protein expressions, 7 exhibited exceptional efficacy compared to MNX. These findings suggest that A. fukudo and its bioactive compounds, particularly compound 7 (eupatilin), may be of interest for further investigation in relation to hair loss–associated mechanisms; however, the current results are based on in vitro experiments and should be regarded as preliminary, requiring validation in appropriate in vivo models and additional mechanistic studies.

In conclusion, the present study provides preliminary in vitro evidence that AFE and its derived bioactive compounds may be relevant to biological pathways associated with both BPH and hair loss. These findings should be interpreted with caution, and further in vivo and mechanistic studies are necessary to determine their potential relevance for therapeutic development.

Supporting Information

Additional supporting information can be found online in the Supporting Information section. Supporting Fig. S1 : ^1^H‐NMR spectrum for compound 1. Supporting Fig. S2 : ^13^C‐NMR spectrum for compound 1. Supporting Fig. S3 : ^1^H‐NMR spectrum for compound 2. Supporting Fig. S4 : ^13^C‐NMR spectrum for compound 2. Supporting Fig. S5 : ^1^H‐NMR spectrum for compound 3. Supporting Fig. S6 : ^13^C‐NMR spectrum for compound 3. Supporting Fig. S7 : ^1^H‐NMR spectrum for compound 4. Supporting Fig. S8 : ^13^C‐NMR spectrum for compound 4. Supporting Fig. S9 : ^1^H‐NMR spectrum for compound 5. Supporting Fig. S10 : ^13^C‐NMR spectrum for compound 5. Supporting Fig. S11 : ^1^H‐NMR spectrum for compound 6. Supporting Fig. S12 : ^13^C‐NMR spectrum for compound 6. Supporting Fig. S13 : ^1^H‐NMR spectrum for compound 7. Supporting Fig. S14 : ^13^C‐NMR spectrum for compound 7. Supporting Fig. S15 : ^1^H‐NMR spectrum for compound 8 . Supporting Fig. S16 : ^13^C‐NMR spectrum for compound 8.

Funding

This study was supported by Gyeongsang National University (Research Resurgence under the Glocal University 30 Project at Gyeongsang National University in 2024).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supplementary Material