In Vitro Hair Dermal Papilla Cells Induction by Fagraea berteroana, a Tree of the Marquesan Cosmetopoeia (French Polynesia)

Abstract

Fagraea berteroana is a tree used in traditional medicine in various islands of the South Pacific. Here, we studied its hair growth-inducing properties as suggested by one of its Marquesan ethno-uses in haircare. The ethyl acetate extract of the fruits of F. berteroana (FEAE) and four resulting fractions (FEAE-F0, FEAE-F1, FEAE-F2, and FEAE-F3) were tested on hair follicle dermal papilla cells to determine their cell proliferative activity. Furthermore, RT-qPCR analysis enabled gene modulation analysis, while immunostaining of the β-catenin protein was used to follow protein regulation. We found that the plant extracts induced a controlled, dose-dependent cell proliferation. FEAE-F0 simultaneously down-regulated Bone Morphogenetic Protein 2 (BMP2) mRNA expression and upregulated Cyclin-D1 (CCND1) gene expression, which suggests an involvement in the regulation of the Wnt and Transforming Growth Factor beta (TGFβ) pathways that control the hair cycle. FEAE-F0 exhibited a 1.34-fold increase of nuclear β-catenin protein. This is indicative of an active hair growth state. Thus, we conclude that FEAE-F0 could be an innovative candidate in hair care, which opens interesting leads to promote the Marquesan cosmetopoeia.

1. Introduction

Overall body care is an essential routine in Polynesian customs, as depicted by the confection and daily use of monoi and other natural preparations for embellishment [1,2,3]. Moreover, hair is an important physical feature in French Polynesia’s society; it is the second most cited application area of cosmetics after the skin, in the Marquesas Islands [3]. As such, the imagery of Polynesians with long, luxurious, and silky dark hair is widespread, even to this day (Figure 1, [4,5]).

An effort to study French Polynesia’s cosmetopoeia [6] led to discovering one plant with an atypical traditional application. This scarcely known ethnocosmetic use was recorded in the Marquesas Islands where, during the embalming process, the crushed fruits of Fagraea berteroana were rubbed on cadavers head to prevent hair from shedding [7].

Fagraea berteroana A.Gray ex Benth. (Gentianaceae), sometimes wrongly spelled Fagraea berteriana, is a tree of moderate height, naturally distributed throughout the South Pacific, except for Hawaii where it has been introduced. It is known under several names such as pualulu in Samoa, pua kenikeni in Hawaii, pua ‘enana or ka’upe in the Marquesas Islands [3], or simply pua in Tahiti [8]. It produces fragrant tubular white flowers that later turn yellow and seasonally yield 5 cm-long and ellipsoid, orange fruits [9] (Figure 1).

A study focused on the potential of the flowers in the fragrance industry revealed volatile compounds benzyl acetate, methyl benzoate, benzyl benzoate, myristic acid, benzyl salicylate, (E)-β-ocimene, methyl salicylate, acetoin, acetic acid, nerolidol, and heneicosane as main components of the essential oil and flower headspace [10]. The fragrant flowers are traditionally well known throughout Polynesia. In Hawaii, they are used to make ceremonial leis, while in French Polynesia they are macerated in coconut oil to obtain scented monoi [11,12]. Thus, further demonstrating the importance of F. berteroana in Polynesian cosmetopoeia.

Literature shows that little is known concerning the chemical composition of this species or its biological activities. Studies on other species of the genus Fagraea revealed diverse compound families such as alkaloids, secoiridoid glycosides, coumarins, flavanols, other phenols, and lignans in the stem bark, roots, and leaves [13,14,15,16,17,18,19,20,21]. The only chemistry-related study on the fruits is of F. fragrans Roxb. (now named Cyrtophyllum fragrans (Roxb.) DC.) and focused on the amidation of ursolic and oleanolic acids, two components naturally present in this plant part [22].

The peculiar ethnobotanical use specific to the Marquesan Islands and the lack of phytochemistry studies on this plant made it an interesting candidate for research into its hair growth activity through modulation of the hair cycle. The latter consists of three periodic phases, the anagen phase during which the hair elongates, the catagen phase when the epithelial cells of the hair undergo apoptosis, and telogen phase when the hair rests then sheds [23,24]. Several pathways have been reported to intervene during the hair cycle, namely Wnt/β-catenin, Sonic hedgehog (Shh), Transforming Growth Factor beta (TGFB), Phosphoinositide 3-kinase PI3K/AKT, Bone Morphogenetic Protein (BMP), and Fibroblast Growth Factor (FGF) [24,25,26,27,28]. Among these, Wnt, TGFB, and BMP are most discussed during the telogen-to-anagen and anagen-to-catagen transitions, for their role in signaling interactions between mesenchymal and epithelial cells [24]. Activation of the canonical Wnt pathway is essential for transitioning to and maintaining the anagen phase [26]. BMPs [29] and TGFβs [25] are paracrine effectors of the dermal papilla cells that target epithelial cells and promote the telogen and catagen phases, respectively.

In this study, we focused on Cyclin-D1 (CCND1) a downstream gene target of the Wnt pathway and its crucial regulatory protein, β-catenin [30]. We also assessed Bone Morphogenetic Protein 2 (BMP2) and Transforming growth factor beta 1 (TGFB1) levels.

We hypothesized that the extract and fractions would stimulate hair dermal papilla cell induction via regulation of the studied factors of the Wnt, TGFB, and BMP pathways. This would suppose a potential promotion of the anagen phase. To do so, cell proliferation activity of the ethyl acetate extract of the fruits of F. berteroana (FEAE) including its four resulting fractions FEAE-F0, FEAE-F1, FEAE-F2, FEAE-F3 of differing polarity obtained after open column fractioning was determined by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay on hair follicle dermal papilla cells. RT-qPCR was performed to select the most active fraction according to its regulation of the three gene targets of the hair cycle. FEAE-F0 was thus selected and subjected to immunostaining to follow its regulation of β-catenin protein. Finally, UHPLC-ESI-MS/MS analysis of FEAE-F0 was performed to tentatively assess its chemical composition. These results are the first cellular assays focused on F. berteroana fractions and extracts.

2. Materials and Methods

2.1. Plant Material Collection, Extraction, and Fractionation

The fruits of F. berteroana A. Gray ex Benth. were collected in April 2018 in Tahiti (17°34.646′ S 149°36.363′ W), French Polynesia. The tree was identified by the botanist Jean-François Butaud. A voucher specimen (K HUGHES 4) was deposited at the herbarium of French Polynesia (PAP). The fruits were oven-dried at 40 °C for 2 days then ground to a 2 mm powder. The powder was macerated in ethyl acetate during 12 h, under agitation. The liquid ethyl acetate extract was evaporated until a viscous extract was obtained. The ethyl acetate F. berteroana extract (FEAE) was further fractioned (10.89 g) using open column chromatography with silica gel 60 Å. The solvents chosen were cyclohexane, ethyl acetate (EtOAc), and methanol in a gradually polar system. The elution started with two column volumes of cyclohexane/EtOAc (80/20) followed by one column volume of cyclohexane/EtOAc (60/40) which led to the first fraction FEAE-F0. The second fraction FEAE-F1 was obtained by recovering the 2nd volume of cyclohexane/EtOAc (60/40) and the first volume of cyclohexane/EtOAc (70/30). FEAE-F2 is composed of the 2nd volume of cyclohexane/EtOAc (70/30) and 100% EtOAc, and finally, FEAE-F3 is the fraction eluted with 100% methanol. All fractions were evaporated till dry. A fraction of the dried fractions was later dissolved in dimethylsufoxide (DMSO) to obtain a 1% solution for the cellular assays.

2.2. Cell Viability via the MTT Assay

Human hair follicle dermal papilla cells (HDPCs) were purchased from Promocell (Heidelberg, Germany) and grown in follicle dermal papilla cell growth medium (Promocell, Sickingenstr, Heidelberg, Germany) supplemented with 100 U·mL⁻¹ penicillin and 100 μg·mL⁻¹ streptomycin (Gibco). The cells were cultured in a humidified atmosphere at 37 °C and 5% CO₂. The Cell Proliferation kit I MTT assay (Roche, Manheim, Germany) was performed to assess the influence of the extracts on the viability of HDPCs. Cells were seeded (10⁴ and 8 × 10³ for 24- and 48-h experiments, respectively) into 96-well plates. They were treated with 200 μL of increasing concentrations of the extracts, ranging from 0.1 μg·mL⁻¹ to 50 μg·mL⁻¹. The control group consisted of DMSO diluted in medium to concentrations ranging from 0.001% to 0.5%, similarly to those of the extracts. The plates were incubated at 37 °C and 5% CO₂ for 24 h and 48 h independently. The supernatant was then discarded and 100 μL of 10% MTT in fresh medium was added to the wells and incubated for 4 h before adding 100 μL of solubilization solution. The plates were incubated overnight, and the absorbance was read on a Multiskan GO spectrophotometer (Thermoscientific, Inc., Vantaa, Finland) at 570 and 690 nm. All tests were done at least in triplicate. The absorbance of each extract concentration was normalized to its corresponding control. Statistical analyses were performed on results as described in Section 2.5 Statistical analysis.

2.3. Quantitative Real-Time Polymerase Chain Reaction (qRT-PCR)

Cells were seeded in 6-well plates and treated with DMSO 0.5% in medium or the extracts diluted in medium. Total RNA was extracted from cells using TRIzol^® reagent (Thermo Fisher Scientific, Inc., Carlsbad, CA, USA). After quantification on a Nanodrop 1000 spectrophotometer (Thermo Fisher Scientific, Inc., Wilmington, DE, USA), 750 ng of RNA were reversed transcribed to cDNA with Superscript IV (ThermoFisher Scientific, Inc., Vilnius, Lithuania). The qRT-PCR was performed to detect the expression of genes of interest using SYBR-Green PCR Master Mix (Roche, Mannheim, Germany) with a Lightcycler 480 II Roche system (Roche, Manheim, Germany). The thermocycling conditions were as follows: initial denaturation at 95 °C for 2 min, then denaturation at 95 °C for 15 s, annealing at 56 °C for 15 s and extension at 68 °C for 20 s, for a total of 40 cycles. The gene expression was normalized to Calmodulin 2 (CALM2), Glyceraldehyde 3-phosphate dehydrogenase (GAPDH), and Pumilio homolog 1 (PUM1) expressions. The expression level of the genes of interest was calculated using the 2^−ΔΔCq method. Primers were designed and obtained from Eurogentec (

Table 1. List of designed primer sequences used for qRT-PCR analysis.

Gene Names	Forward Primer	Reverse Primer
BMP2	5′-CCC-ACT-TGG-AGG-AGA-AAC-AA-3′	5′-GCT-GTT-TGT-GTT-TGG-CTT-GA-3′
CALM2	5′-GGG-AAC-ATC-TGG-GTT-ATG-CC-3′	5′-GAC-TGT-CCA-TAG-TCC-ACG-CA-3′
CCND1	5′-AAC-TAC-CTG-GAC-CGC-TTC-CT-3′	5′-CCA-CTT-GAG-CTT-GTT-CAC-CA-3′
GAPDH	5′-CCA-GCA-AGA-GCA-CAA-GAG-GA-3′	5′-TGG-TTG-AGC-ACA-GGG-TAC-TT-3′
PUM1	5′-GGT-GCC-CTT-GTA-GTG-AAT-GC-3′	5′-TGT-TGT-TCC-AGC-AAG-ACC-AC-3′
TGFB1	5′-CTG-GCG-ATA-CCT-CAG-CAA-CC-3′	5′-CGG-TAG-TGA-ACC-CGT-TGA-TGT-C-3′

). The expression level obtained after the 2^−ΔΔCq of each treatment for a given gene was finally normalized to the replicate mean of its corresponding DMSO control. Statistical analyses were performed on gene expression values as described in Section 2.5 Statistical analysis.

2.4. Direct Immunofluorescence Staining

The cells were seeded on 24-well plates (Corning, Durham, NC, USA) and treated with either medium alone or FEAE-F0 for 72 h. The HDPCs were then fixed with 4% paraformaldehyde (Thermoscientific, Rockford, IL, USA) during 15 min at room temperature, then permeabilized with Triton 0.1X and saturated for 30 min in 10% fetal bovine serum (FBS). The cells were then directly stained with mouse β-catenin antibody (1:25) conjugated with AlexaFluor 488 (Santa Cruz, Dallas, TX, USA) in 1% FBS overnight at 4 °C.

The nuclei were counterstained with Hoechst 33342 Solution (Thermoscientific, Illinois, USA) the following day. The images were obtained immediately after counterstaining on the microscope AxioObserver 7 (Zeiss, Jena, Germany) and analyzed with the software Fiji coupled to ImageJ version 2.0. According to the protocol set by The University of Queensland [31] with slight modifications.

The analysis of the images obtained from the microscope was carried out as follows.

The surface of each cell nuclei was selected via a macro command on the image obtained for the β-catenin green fluorescence. The selected regions were measured for the area and integrated density (ID). All selections with an area below 640 were discarded as these did not correspond to cells but specks of fluorescence or other artefacts. A mean background value (MBGV) was also measured by selecting a dark region with no cells. The corrected total cell fluorescence (CTCF) was obtained using Equation (1).

CTCF = ID of specific cell − (Area of cell × MBGV)

(1)

The same steps were also done in blue fluorescence for the same cells selected in the green channel. A normalized CTCF value was obtained by dividing the CTCF obtained for the green channel of each cell to the corresponding CTCF in the blue channel. The results obtained were averaged to obtain a mean normalized CTCF equated to nuclear β-catenin expression levels.

2.5. Statistical Analysis

Data on cell-based tests were expressed as mean ± SE (standard error) of at least three independent experiments. The statistical tests were performed on Stata14. A F-test was performed to determine equality of the variances between the control group mean and each treatment group mean, followed by an unpaired Student’s t-test. Welch’s t-test was performed if variances were unequal. A p-value ≤ 0.05 was considered significant.

2.6. UHPLC-MS/MS Analysis of F. berteroana Fraction FEAE-F0

The chemical analyses were performed on a UHPLC system (Dionex Ultimate 3000, Thermo Scientific^® equipped with a photodiode array detector: 254, 280, 340, and 450 nm) coupled to a high-resolution mass spectrometer (HRMS QqToF Impact II equipped with an electrospray ionization source, Bruker Daltonics, Germany) in positive and negative ionization modes (20 eV and 40 eV). The extracts were prepared by solubilizing 1 mg of dry extract in 1 mL of methanol then filtered with a 0.2 µm syringe filter. The separations were carried out on an Acclaim RSLC C18 column (2.1 mm × 150 mm, 2.2 µm, DIONEX Thermo Scientific^®, Sunnyvale, CA, USA) at 40 °C by injecting 2 µL of the prepared solution.

The chromatographic analytical method was developed to obtain the best chromatographic separation, with H₂O + 0.1% formic acid (solvent A) and acetonitrile + 0.1% formic acid (solvent B). The program consisted of 2 min at 5% B followed by a linear gradient up to 100% B for 8 min, then 100% B for 3 min and ended by a 3 min re-equilibration at 5% B (same flow rate). The injection of a formate acetate solution in basic media forming clusters on the studied mass range was used for mass calibration before each analysis.

Mass spectra were acquired ranging from 50 to 1200 m/z at 2Hz. The nebulizer pressure was set at 50.8 psi, the capillary voltage at 3000 V, the dry gas flow rate at 12 L·min^−1, and dry temperature at 200 °C.

The MS/MS spectral data obtained were compared to entries in repositories such as MassBank of North America [32] and MassBank [33], the METlin library, and literature data when available (Table S1).

3. Results and Discussion

3.1. Hair Follicle Dermal Papilla Cells (HDPCs) Proliferation Dynamic after Treatment with Extracts and Target Genes Regulation

The HDPCs were treated with the extracts during both 24 and 48 h to study cell viability and proliferation dynamics as seen in Figure 2A–E.

After 24 h of treatment, FEAE and FEAE-F0 induced a dose-dependent proliferation of HDPCs (Figure 2A,E). It resulted in the highest cell proliferation after 24 h of treatment at 50 μg·mL⁻¹ amongst all the other fractions, with a 1.43-fold increase compared to the control. FEAE-F1 and FEAE-F2, also stimulated cell proliferation compared to the control after 24 h of treatment between 6.25 and 25 μg·mL⁻¹, although not dose-dependent. FEAE-F3 also promoted cell proliferation in cells between 12.5 and 50 μg·mL⁻¹. Furthermore, after 48 h of treatment, FEAE-F0 also induced a dose-dependent proliferation of HDPCs, along with FEAE-F3. FEAE-F3 induced a 1.24-fold increase in cell proliferation compared to the control. On the other hand, FEAE showed no significant effect on cell proliferation after 48 h of treatment.

The RNA experiments were conducted by treating the cells with FEAE, FEAE-F0, and FEAE-F3 at 50 μg·mL⁻¹, FEAE-F1 at 25 μg·mL⁻¹ and the 0.5% DMSO control during 48 h, according to the results of the cell proliferation assay. Further testing on FEAE-F2 at 25 μg·mL⁻¹ was excluded because we observed a persistent adverse effect on the physiological state of the cells when seeded in 6-well plates. Three gene targets were studied, Cyclin D1 (CCND1), Transforming growth factor beta 1 (TGFB1), and Bone Morphogenetic Protein 2 (BMP2) (Figure 2F). BMP2 is an inhibitor of the anagen, thus follicle growth, phase. Its levels are highest during the refractory telogen phase and rapidly decrease to allow entry into a new anagen phase [22]. Similarly, TGFB1 is a paracrine effector that facilitates catagen entry while Wnt activation decreases [25,34]. The extract FEAE significantly decreased TGFB1 expression 1.22-fold (p = 0.04) when compared to the control using a Student t-test. This modulation of the TGFB1 gene expression was not found in any of the fractions. Nonetheless, FEAE-F0 induced a significant 2-fold (p = 0.005) decrease of BMP2 while simultaneously increasing CCND1 1.5-fold (p = 0.002). FEAE-F1 also enabled a significant 1.7-fold (p = 0.007) increase in CCND1 but had no significant effect on the expression levels of BMP2. FEAE-F3 showed no significant regulation of any of the studied genes.

Reported studies have shown crosstalk between Wnt and TGFβ signaling [35,36]. Inhibitors of the Wnt pathway show a subsequent increase in TGFβ signaling in non-balding cells, as well as an increased TGFβ1 expression [36]. Likewise, BMP signaling prohibits activation of the Wnt pathway during the telogen phase. As the signaling environment of the cells changes, BMP levels lower to allow an upregulation of Wnt signaling and thus entry into anagen phase [24]. FEAE-F0 stood out as the most active fraction by displaying a similar regulation. Indeed, it not only increased activation of the Wnt pathway, as seen by the upregulation of CCND1, but it also downregulated BMP2. By lowering BMP signaling, FEAE-F0 facilitates transitioning to the follicle growth (anagen) phase. This can be observed by an increase in Wnt signaling [24]. CCND1 is a downstream target of the Wnt pathway and a rise in its expression levels suggested a stimulated Wnt state that we aimed to confirm by following β-catenin expression in the presence of FEAE-F0.

3.2. Modulation of β-Catenin Production by FEAE-F0

As β-catenin is a key protein of the Wnt pathway that intervenes during the growth phase of hair follicles [37], the effect of FEAE-F0 on its expression was studied (Figure 3). Indeed, when the Wnt/β-catenin pathway is inactive, β-catenin in the cytoplasm is phosphorylated by a complex composed of glycogen synthase kinase 3β, Axin, Adenomatous polyposis coli, and casein kinase 1a. The phosphorylated β-catenin is then ubiquitinated and recognized by a proteasome that degrades it before it enters the nucleus [38,39]. Conversely, when the canonical Wnt pathway is activated, the phosphorylating complex is inhibited which allows β-catenin to stabilize, accumulate in the cytoplasm, and successfully translocate to the nucleus [30]. Once in the nucleus, β-catenin serves as a co-activator of several downstream targets such as cyclin-D1 [40]. According to our experiments in Figure 3, FEAE-F0 showed a significant (p < 0.0001) 1.34-fold increase of nuclear β-catenin after treatment of HDPCs for 72 h compared to the control. Increased β-catenin levels in the nucleus are indicative of an active Wnt state allowing signaling to downstream targets [37]. Furthermore, it is important to note that this effect is done in a controlled manner. Indeed, while activation of the Wnt pathway in hair follicle cells is paramount for induction of the anagen phase and in fine hair growth, overexpression of the Wnt pathway could lead to developing cancerous cells [41]. FEAE-F0 showed a controlled proliferation and upregulation of the Wnt pathway which further crystallizes its efficiency in promoting hair growth.

3.3. Chemical Composition of F. berteroana FEAE-F0

An identification of the main compounds present in FEAE-F0 was attempted by UHPLC-MS/MS analysis. The MS/MS data analyzed was that of 20 eV and 40 eV collision energy in positive and negative ionization mode. Thirteen main peaks were identified on the UV chromatogram in Figure 4. The compilation and tentative identification of the peaks by MS² correspondence were reported in

Table 2. Corresponding MS data compilation of peaks present in FEAE-F0 chromatogram.

		Positive Mode			Negative Mode		UV (nm)	Database Search (21/01/2021)
Peak No.	Rt (min)	[M + H]+	[M + Na]+	Product Ions (MS/MS, 20 and 40 eV)	[M − H]−	Product Ions (MS/MS)		Metlin	MetFrag (vs. Pubchem)	Molecular Formula	Compound Annotation
1	7.67	171.1019	193.084	20 eV: 57 (100, C4H9+), 83 (10, C5H7O+), 111 (19, C7H11O+), 125 (4, C8H13O+) 40 eV: 55 (21, C4H7+), 57 (100, C4H9+), 59 (10, C3H7O+), 97 (17, -), 145 (10, -)	-	-	224	3 entries, no experimental MS2 spectra	5240 candidates	C9H14O3	boonein
2	7.82	181.0861	203.0681	20 eV: 57 (100, C3H5O+), 117 (6, C9H9+), 131 (4, -), 151 (3, C8H7O3+) 40 eV: 57 (100, C3H5O+), 91 (57, C7H7+), 109 (36, -), 115 (50, C9H7+), 127 (21, -), 152 (20, -)	-	-	225, 274	21 entries, 7 experimental MS2 spectra, no correspondence	3909 candidates	C10H12O3	-
3	8.22	153.0915	175.0734	20 eV: 67 (46, C5H7+), 79 (85, C6H7+), 91 (74, C7H7+), 105 (25, C8H9+), 107 (100, C8H11+) 40 eV: 53 (28, C4H5+), 55 (22, C4H7+), 65 (71, C5H5+), 67 (27, C5H7+), 79 (44, C6H7+), 91 (100, C7H7+)	-	-	226, 279	13 entries, 2 experimental MS2 spectra, no correspondence	3668 candidates	C9H12O2	-
4	8.57	213.1454	-	20 eV: - 40 eV: 55 (39, C4H7+), 57 (28, -), 91 (67, -), 91 (64, C7H7+), 95 (41, C6H7O+), 145 (100, -)	-	-	226, 279	13 entries, 2 experimental MS2 spectra, no correspondence	4703 candidates	C12H20O3	-
5	8.83	-	-	-	-	-	226, 279			-	-
6	9.00	181.1223	203.1044	20 eV: 79 (21, C6H7+), 81 (29, C6H9+), 93 (64, C7H9+), 95 (56, C7H11+), 107 (93, C8H11+), 121 (42, C9H13+), 135 (84, C10H15+), 145 (22, C11H13+), 163 (58, C11H15O+), 181 (100, C11H17O2+) 40 eV: 53 (33, C4H5+), 55 (40, C4H7+), 67 (31, C5H7+), 69 (17, C5H9+), 79 (69, C6H7+), 91 (100, C7H7+), 93 (47, C7H9+), 105 (47, C8H9+), 107 (24, C8H11+)	-	-	226, 284	12 entries, 1 experimental MS2 spectrum hypothetical match Metlin ID 90552	6090 candidates	C11H16O2	-
7	9.36	-	-	-	217.0874	20 eV: 117.0346 (33, C8H5O−), 119.0503 (100, C8H7O−), 145.0296 (51, C9H5O2−), 163.0402 (28, C9H7O3−), 217.0873 (9, C13H13O3−) 40 eV: 89 (3, C7H5−), 93 (5, C6H5O−), 117 (100, C8H5O−), 119 (38, C8H7O−), 145 (5, C9H5O2−)	227	7 entries, 1 experimental MS2 spectrum, no correspondence	5081 candidates	C13H14O3	-
8	9.45	213.1484	235.1306	20 eV: 57 (100, C4H9+), 81 (4, C6H9+), 83 (5, C6H11+), 109 (5, C7H9O+), 127 (48, C7H11O2+), 139 (6, C8H11O2+), 195 (10, C12H19O2+) 40 eV: 57 (100, C4H9+), 79 (8, C6H7+), 81 (14, C6H9+), 109 (3, C7H9O+), 127 (8, C7H11O2+)	211.1344 (low)	-	227	13 entries, 2 experimental MS2 spectra, no correspondence	4703 candidates	C12H20O3	-
9	9.52	223.0965	247.0745	20 eV: 121 (2, C7H5O2+), 149 (100, C8H5O3+), 177 (7, C10H9O3+) 40 eV: 65 (80, C5H5+), 93 (19, C6H5O+), 111 (10, C6H7O2+), 121 (72, C7H5O2+), 149 (100, C8H5O3+), 167 (4, C8H7O4+)	-	-	228, 277	13 entries, 1 experimental MS2 spectrum, no correspondence	6105 candidates	C12H14O4	-
10	9.80	-	-	-	195.1393	20 eV: 109 (8, C7H9O−), 111 (6, C7H11O−), 125 (3, C8H13O−), 138 (12, C8H10O2−), 167 (16, C11H19O−), 179 (24, C11H15O2−), 195 (100, C12H19O2−) 40 eV: 53 (100, C3HO−), 109 (11, C7H9O−), 122 (14, C7H6O2−), 138 (14, C8H10O2−)	227	59 entries, 1 experimental MS2 spectrum, no correspondence	6334 candidates	C12H20O2	-
11	10.50	181.1224	203.1045	20 eV: 93 (53, C7H9+), 95 (28, C6H7O+), 97 (25, C6H9O+), 107 (32, C8H11+), 111 (17, C7H11O+), 125 (100, C7H9O2+), 135 (37, C10H15+), 138 (22, C8H10O2+), 163 (28, C11H15O+), 181 (94, C11H17O2+) 40 eV: 53 (50, C4H5+), 55 (40, C4H7+), 65 (20, C5H5+), 67 (44, C5H7+), 77 (19, C6H5+), 79 (38, C6H7+), 81 (12, C6H9+), 91 (100, C7H7+), 93 (42, C7H9+), 95 (70, C6H7O+), 97 (22, C6H9O+), 105 (21, C8H9+), 109 (20, C7H9O+), 123 (11, C7H7O2+), 125 (12, C7H9O2+)	-	-	228, 276	12 entries, 1 experimental MS2 spectrum no close match	6090 candidates	C11H16O2	-
12	10.98	209.1538	231.1357	20 eV: - 40 eV: 53 (10, C4H5+), 55 (36, C4H7+), 57 (100, C4H9+), 59 (30, C3H7O+), 65 (7, C5H5+), 67 (56, C5H7+), 69 (15, C5H9+), 77 (5, C6H5+), 79 (42, C6H7+), 81 (31, C6H9+), 83 (4, C5H7O+, 5, C6H11+), 91 (49, C7H7+), 93 (31, C7H9+), 95 (9, C6H7O+, 15, C7H11+), 97 (12, C6H9O+), 103 (6, C8H7+), 105 (46, C8H9+), 107 (37, C8H11+), 109 (8, C7H9O+), 111 (4, C7H11O+), 119 (18, C9H11+), 121 (17, C9H13+), 123 (3, C7H7O2+, C8H11O+, C9H15+), 133 (10, C10H13+), 135 (8, C10H15+), 151 (6, C10H15O+), 179 (11, C11H15O2+), 194 (4, C12H18O2+)	-	-	228, 292	13 entries, no experimental MS2 spectra	6905 candidates	C13H20O2	-
13	11.45	279.1594	301.1414	20 eV: 57 (10, C4H9+), 109 (2, C8H13+), 149 (100, C8H5O3+), 167 (5, C8H7O4+), 209 (2, C10H9O5+) 40 eV: 57 (19, C4H9+), 121 (23, C7H5O2+), 149 (100, C8H5O3+), 167 (5, C8H7O4+)	-	-	227, 279	4 entries, 1 experimental MS2 spectrum, no correspondence	5392 candidates	C16H22O4	-

Rt = retention time.

.

Our previous work on FEAE revealed the presence of loganic and coumaric acids [42]. The present paper mainly focused on the minor bioactive fraction FEAE-F0, derived from FEAE the crude extract previously studied [42]. An exhaustive search of all compounds found in the literature for the genus Fagraea and their available MS² was compiled in Table S1 (Supplementary Material). We compared the obtained data for the thirteen peaks to that of Table S1. Only one iridoid was successfully annotated, boonein. Indeed, boonein and several other iridoids such as (±)-gentiolactone, naucledal, angelone, swermacrolactone C amongst others (Table S1) were found in the stem bark of F. fragrans (syn. of Cyrtophyllum fragrans) [17,19,20,43]. No biological activities linked to hair care have been specified for boonein in literature. Nevertheless, iridoids are known for their anti-inflammatory and cardiovascular activity [44,45]. This could suggest an implication in hair growth by enabling proper vascularization and thus maintaining nutrient support for the efficient proliferation of dermal papilla cells, notably during the anagen phase [46]. Indeed, minoxidil, one of the only FDA-approved treatments to promote hair growth, is known to mediate its effects via different mechanisms such as promoting vascularization around HDPCs [47]. Moreover, minoxidil has been reported to prolong anagen in hair follicles [26,48].

Overall, the remaining compounds that have not yet been identified, showed carbon atoms ranging between 9 and 16, suggesting medium-sized molecules. Their molecular formulas led to many structural propositions amongst known compounds in Pubchem (Table 2), thus uncovering a plethora of structural possibilities.

A few structural hypotheses could nevertheless be proposed.

According to mass and molecular formulas, FEAE-F0 included several isomers. Compounds 6 (m/z = 181.1223) and 11 (m/z = 181.1224), as well as compounds 4 (m/z = 213.1454) and 8 (m/z = 213.1484) only differed in fragmentation pattern. The MS² experimental data at 20 eV for compound 90552 in Metlin is remarkably similar to that of compound 6; the 40 eV spectrum fragmentation did not match as closely. Entry 90552 in Metlin corresponds to dihydroactinidiolide (DHA), which raises the hypothesis of the potential presence of DHA in our fraction, or of closely related terpenoid structures. Such a proposition may be plausible as DHA is a terpene obtained by cleavage of β-carotene [49]. As previously stated, the fruits of F. berteroana are reddish-orange when ripe. This latter point could explain a putative presence of β-carotene derivatives in our fraction. However, compound 11 had less fragments that matched with compound 90552. It could potentially correspond to an isomer of DHA. The resulting fragments of compound 11 could suggest either a linear structure with the loss of oxygen or a substituted phenyl ring (m/z = 91), followed by a successive loss of carbon atoms. Such a pattern could potentially correspond to shorter chained derivatives of the diterpene phytol, or potentially terpenoids containing an aromatic ring core. Interestingly, several terpenes have been the object of patents for their efficiency in preventing hair loss and promoting hair growth [50,51]. Nevertheless, the above propositions remain hypotheses.

4. Conclusions

Our results showed that there is a positive correlation between the ethnocosmetic use of the fruits of F. berteroana and observed molecular activities. Indeed, an ethyl acetate extract of the fruits and fractions FEAE-F0 through FEAE-F3 could potentiate hair growth via stimulation of hair follicle dermal papilla cells proliferation, shown herein by in vitro induction of HDPCs. Furthermore, FEAE-F0 was selected for its potential positive effect on the anagen phase of the hair cycle by upregulating the Wnt pathway, alongside down-regulating the BMP pathway. The in vitro activity observed for FEAE-F0 can be attributed to its chemical content, comprising of iridoids and potentially terpenoids. The fraction stands out as an interesting natural active for applications in hair growth. This could further be cemented with in and ex vivo studies geared towards the prevention of hair loss due to genetic factors such as androgenetic alopecia, or environmental stresses.

Lastly, the actual natural resource is limited in low- to mid-altitude in French Polynesia, but its cultivation is not unreasonable as the tree is often grown in marae, a large and clear area devoted to traditional or communal practices in Polynesian societies, because of its spiritual symbolism or gardens. Hence, an effort to cultivate it could be successful and interesting to obtain a sustainable resource for further development of this plant material as a potential natural cosmetic ingredient.