Polyphenol Characterization and Skin-Preserving Properties of Hydroalcoholic Flower Extract from Himantoglossum robertianum (Orchidaceae)

Himantoglossum robertianum (Loisel.) P. Delforge is a Mediterranean orchid whose propagation in vitro has been achieved, making it eligible as a source of bioactive substances. Flowers were analyzed by light and SEM microscopy and used to obtain a polyphenol-rich, hydroalcoholic flower extract (HFE). HFE was characterized for total phenols, flavonoids and proanthocyanidins, and for polyphenol profile by RP-LC-DAD. Antioxidant assays, in vitro collagenase and elastase inhibition, and MTT and cell motility assays on HaCaT keratinocytes were done. Microscopy showed epidermal cells containing anthocyanins in the flower labellum. Flavonoids (flavones and flavan-3-ols) represented the most abundant compounds (42.91%), followed by scopoletin (33.79%), and phenolic acids (23.3%). Antioxidant assays showed strong activities, rating ORAC > FRAP > TEAC > β-carotene bleaching > DPPH > iron-chelation. Biological assays showed elastase and collagenase inhibition (up to 42% and 78%, respectively), improvement of HaCaT cell viability after treatment with 500 μM H2O2 (from 30% to 84% of control), and stimulation of cell migration rate up to 210% of control. In summary, HFE counteracted different free radicals, while protective properties were shown by cell-free and cell-based bioassays, suggesting the possible use of H. robertianum flowers for skin-preserving, repair, and anti-aging applications.


Introduction
Orchids have been widely used in the folk tradition to treat several ailments, including tuberculosis, inflammation, hepatitis, wounds and sores, tumor, asthma, malaria, and menstrual disorders [1]. The uses of Anacamptis morio L. and Epipactis helleborine (L.) Crantz for wound healing have been reported [2], while a number of species are known as sources of bioactive compounds [3][4][5].
Different studies have started to explore the biological activities of orchid extracts, especially with reference to skin care applications. A cosmetic serum containing 5% orchid extracts, including the Marcella Koss intergeneric hybrid of Brassocattleya, has been tested in vivo for its skin-whitening properties on melasma and lentigo senilis in Japanese women [6]. In a study about the potential of

Flower Morphological Characterization
The strongly scented flowers are arranged in dense spikes with 25-40 large flowers each, growing up to 15-30 cm ( Figure 1A,B). Flowers are variable in color, from pink-purple to greenish-white, show an upper hood made up of 3 petals, and a large lower lip with a lighter spotted center and darker lip margins. The lip measures about 2.0 cm, is divided into three lobes, and the central lobe is further subdivided. Considering color variability, we collected flowers showing a rather homogeneous purplish color in the labellum and sepals.
Flower samples observed under stereomicroscope showed three sepals and petals covering the fertile gynostemium (Figure 2A). The central petal, or labellum, differs from the lateral ones in color, shape, and size. It is composed by a median central region and two lateral zones (arms), showing epidermal cells with different colors and shapes (Figure 2A, Figure 3A). The labellum forms the spur Flower samples observed under stereomicroscope showed three sepals and petals covering the fertile gynostemium ( Figure 2A). The central petal, or labellum, differs from the lateral ones in color, shape, and size. It is composed by a median central region and two lateral zones (arms), showing epidermal cells with different colors and shapes (Figure 2A, Figure 3A). The labellum forms the spur that is probably involved in deceptive pollination, acting as a resting place for pollinators and as an osmophore involved in the emission of attractive odorants.
In sections observed under light microscopy, epidermal cells of the central region of the labellum appeared purple-red due to the presence of anthocyanins ( Figure 2B). In the lateral region of the labellum, several short papillose cells, brown-purple in color, were found ( Figure 2C) among more flattened, sub-polygonal cells covered by waxes ( Figure 2C, Figure 3C). The sub-stigmatic zone of the central labellum showed outgrowths made of long papillose cells ( Figure 2D, Figure 3B), possibly facilitating the emission of volatile compounds. Raphides and stomata were observed both in the sepals and in the labellum ( Figure 2E).      In sections observed under light microscopy, epidermal cells of the central region of the labellum appeared purple-red due to the presence of anthocyanins ( Figure 2B). In the lateral region of the labellum, several short papillose cells, brown-purple in color, were found ( Figure 2C) among more flattened, sub-polygonal cells covered by waxes ( Figure 2C, Figure 3C). The sub-stigmatic zone of the central labellum showed outgrowths made of long papillose cells ( Figure 2D, Figure 3B), possibly facilitating the emission of volatile compounds. Raphides and stomata were observed both in the sepals and in the labellum ( Figure 2E).

Phytochemical Characterization
By ultrasound-assisted method, we obtained a hydroalcoholic flower extract (HFE) with a high extraction yield (8.28%). A preliminary phytochemical screening (Table 1) revealed high total phenol content (243.7 mg GAE/100 g fresh weight, FW), with flavonoids representing the most abundant compounds (398.1 ± 9.8 mg QE/100 g FW), and a lower presence of anthocyanins (4.89 mg ChE/100 g FW).

Antioxidant and Free-Radical Scavenging Activity
In order to evaluate the antioxidant and free-radical scavenging activity of HFE, various in vitro cell-free assays, based on different environments and reaction mechanisms, were carried out. All assays showed dose-dependent (R2 ≥ 0.99) antioxidant and free-radical scavenging activities, with the following order of potency ORAC > FRAP > TEAC > β-carotene bleaching > DPPH > Iron-chelating activity (Table 3). These data indicate that HFE can counteract different charged free-radical types, especially peroxyl ones.

Effects of HFE on Keratinocytes and Skin Enzymes
Effects of HFE on cell viability were preventively tested by the MTT assay, finding IC 05 > 500 and IC 50 > 1000 µg/mL. Therefore, in order to keep HFE at subtoxic levels, all experiments were conducted by using 500 µg/mL as the highest concentration.
In order to evaluate the antioxidant potential of HFE directly on cells, we pre-exposed HaCaT cells to HFE for 24 h, and then to HFE combined with H 2 O 2 for a further 24 h. Controls were run by incubating cells with HFE alone for 48 h, or with H 2 O 2 alone for 24 h. At the end of incubations, cell  Figure 5A). The scratch wound assay, conducted for 48 h on HaCaT cells, showed a significant increase of cell migration after incubation with 50 and 500 μg/mL HFE ( Figure 5B).
Finally, we performed cell-free assays to verify the ability of HFE to inhibit two major enzymes involved in skin extracellular matrix degradation, viz. elastase and collagenase. In the elastase test, a significant average inhibition of 42% was found with 500 μg/mL HFE ( Figure 5C), while in the collagenase test significant average inhibitions of 70% and 78% were obtained with 50 and 500 μg/mL HFE, respectively ( Figure 5D).

Discussion
The polyphenol characterization of HFE showed that the extract is very rich in flavonoids, scopoletin, and phenolic acids. It is well known that the antioxidant and free-radical scavenging properties of polyphenols are directly correlated with the number of hydroxyl groups linked to the The scratch wound assay, conducted for 48 h on HaCaT cells, showed a significant increase of cell migration after incubation with 50 and 500 µg/mL HFE ( Figure 5B).
Finally, we performed cell-free assays to verify the ability of HFE to inhibit two major enzymes involved in skin extracellular matrix degradation, viz. elastase and collagenase. In the elastase test, a significant average inhibition of 42% was found with 500 µg/mL HFE ( Figure 5C), while in the collagenase test significant average inhibitions of 70% and 78% were obtained with 50 and 500 µg/mL HFE, respectively ( Figure 5D).

Discussion
The polyphenol characterization of HFE showed that the extract is very rich in flavonoids, scopoletin, and phenolic acids. It is well known that the antioxidant and free-radical scavenging properties of polyphenols are directly correlated with the number of hydroxyl groups linked to the phenolic structure, and indirectly correlated with their glycosylation degree [19]. Flavones are the most abundant sub-class of flavonoids in HFE, but they are mostly glycosylated, and therefore, they should only marginally contribute to the observed antioxidant and free-radical scavenging activities. In contrast, flavan-3-ols like catechin and epicatechin, also having catechol structure, could play a primary role in antioxidant activities, together with major polyhydroxylated phenolic acids like protocatecuic, chlorogenic, and caffeic acids.
Scopoletin, as previously demonstrated, showed a good scavenging ability against ABTS •+ , but was not effective against other charged radicals, while it showed only a weak antioxidant activity. Also, in coumarins the catechol group markedly contributes to the antioxidant activity, while glycosylation adversely affects it. In addition, the α-pyrone ring makes coumarins more hydrophobic than phenolic acids, and therefore more reactive with respect to lipid peroxidation [20]. In conclusion, the complex of phytochemicals suggests that HFE can act as a powerful scavenger of different charged radicals.
In parallel with phytochemical characterization and antioxidant evaluation, skin-preserving properties have been clearly suggested by the complex of our bioassays. Significant protection of keratinocytes against injurious effects has been found after double exposure to HFE and H 2 O 2 as the oxidant agent, but only at high HFE doses. The lack of a pan-antioxidant protection could reflect the known tendency of some polyphenols, e.g., flavonoids, to elicit pro-oxidant activities under specific redox conditions [21,22]. However, polyphenols are also widely known to exert antioxidant effects, as fully confirmed in this study by the panel of specific assays. Hence, the polyphenols of H. robertianum, used at proper doses, could support skin redox balance processes, thus contributing to prevent or counteract oxidative-related dysfunctions, especially those leading to skin ageing. This kind of activity should be further potentiated by stimulation of keratinocyte mobilization, revealed by our scratch wound assay, suggesting a support to skin repair following injury or degenerative processes.
The elastase inhibition induced by HFE rates at average levels among plant extracts, while collagenase inhibition approaches topmost values obtained with polyphenol-rich white tea extract [23]. Collagen and elastin are the most abundant protein constituents of the dermal extracellular matrix, being primarily affected by photoaging and other oxidative processes that cause wrinkles and reduction in skin thickness [24]. Hence, the results obtained with HFE on matrix degrading enzymes make this orchid-derived product a very promising ingredient for skin aging remedies.
Orchids are in general protected plant species, but the development of protocols for in vitro propagation is making these plants exploitable for industrial purposes without affecting wild populations [25]. These techniques have been applied also to H. robertianum, through protoplast isolation and in vitro propagation from asymbiotic seed germination [26,27]. Hence, the giant orchid H. robertianum shows remarkable biological properties of its flower extract and can be exploited in environment-friendly mode, suggesting its possible use for post-traumatic and post-inflammatory skin repair, antiaging treatments, and skin preserving applications.

Plant Materials
Flowers of H. robertianum were gathered by LC and MB from wild populations growing at Taggia (43 • 52'05.2" N, 7 • 50'14.6" E), and Carpasio (43 • 57'24.8" N, 7 • 50'38.7" E) (Imperia, Liguria, Italy) from February to May 2018. Sampling of the species, protected under law, was allowed by the Ligurian Region Government with act n. 363/29-01-2018. Only flowers were carefully sampled from resident plants, in order to minimize damage to populations. The species has a quite distinctive habitus and size, and it was determined in the field by LC and MB, and two voucher specimens (one from each sampling site) were deposited at the Herbarium of DISTAV, University of Genova, Genova, Italy (number: GE 1038).

Reagents and Cells
All reagents were purchased from Sigma Aldrich (Milan, Italy), unless otherwise indicated. The HaCaT human keratinocyte cell line was purchased from the Biological Bank of the Azienda Ospedaliera Universitaria San Martino-IST, Genova, Italy. Cells were cultured at 37 • C, in a 5% CO 2 , humidified atmosphere, using Dulbecco's Modified Eagle Medium (DMEM, EuroClone, Milan, Italy) enriched with 10% (v/v) FBS, 1% glutamine, and 1% antibiotic.

Light and Scanning Electron Microscopy Analyses
Flower portions were observed by a Leica M205C stereomicroscope, coupled with EC3 camera and LAS EZ V1.6.0 image analysis software. Epidermal peels of the fresh flowers were mounted on glass slides and observed by a Leica DM 2000 transmission-light microscope coupled with a computerdriven DFC 320 camera (Leica Microsystems, Wetzlar, Germany).

Hydroalcoholic Flower Extract (HFE) Preparation
Flowers were gently grounded by a blade mill (IKA ® A11 basic analytical mill) under liquid nitrogen. One hundred milliliters of 70% ethanol were added to 10 g of powdered sample mixing for 3 min. The extraction was carried out three times by sonication in an ice-cold bath for 5 min using a 3 mm titanium probe, set to 200W power and 30% amplitude signal (Vibra CellTM Sonics Materials inc., Danbury, CT, USA). Thereafter, the sample was centrifuged at 3000 rpm for 15 min (NEYA 10R, REMI, Carpi, Italy) and the supernatant was collected and evaporated by rotary evaporator (BUCHI R-205, Cornaredo, Italy). Dry HFE was dissolved in 70% ethanol in order to obtain a 50 mg/mL stock solution, which was properly diluted to carry out polyphenol characterization, antioxidant assays, cell-free bioassays, and cell treatments.

Total Phenols
Total phenols content was established by Folin-Ciocalteu assay as previously reported [29]. The absorbance was recorded at 786 nm using an UV-Vis spectrophotometer (Shimadzu UV-1601, Kyoto, Japan). Results were expressed as mg of gallic acid equivalents (GAE)/100 g of sample fresh weight (FW).

Vanillin Index
The proanthocyanidin and anthocyanin contents were determined as previously reported [30], recording the absorbance of sample by an UV-Vis spectrophotometer (Shimadzu UV-1601, Kyoto, Japan) at 500 nm against a blank. Catechin was used as reference compound (0-500 µg/mL) and results were expressed as mg of catechin equivalent (CatE)/100 mg of sample FW.

β-Carotene Bleaching Assay
The β-carotene bleaching assay was performed as previously described [19]. Aliquots of the fresh β-carotene emulsion (8.0 mL) were mixed with 320µL of HFE solutions (12.50-200 µg/mL). An emulsion without β-carotene was used as negative control. The absorbance was recorded at 470 nm at the starting time (t = 0), and then incubated at 50 • C in a water bath for 120 min, recording the absorbance every 20 min. Butylated-hydroxytoluene (BHT) 1 mg/mL was used as positive control. The antioxidant activity was expressed as inhibition (%) of the β-carotene bleaching calculating the half-maximal inhibitory concentration (IC 50 ) with the respective C.L. at 95%.

Iron-Chelating Activity
The iron-chelating activity of HFE was evaluated according to a previous study [34], with some modifications. Briefly, 50 µL of FeCl 2 ·4H 2 O solution (2.0 mM) was added to 100 µL of HFE (62.5-100 µg/mL) incubating at RT for 5 min. After that, 100 µL of ferrozine solution (5 mM) was added to the reaction mixture and the sample solution diluted to 3 mL with deionized water, mixed and incubated for 10 min at RT. The absorbance was read at 562 nm using an UV-VIS Spectrophotometer (Shimadzu UV-1601, Kyoto, Japan). Results were expressed as inhibition (%) of the Fe 2+ chelating capacity calculating the half-maximal inhibitory concentration (IC 50 ) with the respective C.L. at 95%.

Collagenase and Elastase Inhibition Assays
The elastase and collagenase assays were performed according to a previously-reported method [35].
For the assay of HFE cytoprotective properties, HaCaT were settled as above in 96-wells plates, preincubated for 24 h with 5, 50, or 500 µg/mL HFE, and then co-incubated for further 24 h with the above HFE doses combined with 500, 750, or 1000 µM H 2 O 2 . Cell viability was then determined by MTT test. Results were expressed as the percentage of viable or dead cells (ratio of unstained or stained cells to the total number of cells, respectively).

In Vitro Wound Healing Assay
The migration rate of HaCat cells exposed to HFE was assessed by the scratch wound assay method as reported by Muniandy et al. [36]. Briefly, cells were seeded on 12-well plates (2 × 10 5 cells/well), grown to monolayer, and wounded with a 100 µL pipette tip. One well was immediately fixed in Finefix working solution and stained with TBO 0.1%, to represent the T0 sample. Negative control was exposed to medium without serum, and control to complete medium. The remaining wells were exposed for 48 h to 5, 50, or 500 µg/mL HFE. After incubation, cells were washed with PBS, fixed, and stained as above. Images of wounds were taken with a Leica M205 C stereomicroscope coupled to a Leica EZ 2.1.5 camera, and wound width was measured by image analysis with ImageJ software (https://imagej.nih.gov/ij/). Cell migration was expressed as the percentage of wound closure, calculated with the following formula: Wound closure (%) = 100 × ((wound width at 0 h) − (wound width at 48 h))/(wound width at 0 h). (1)

Statistical Analysis
The statistical analysis used in this study was based on one-way analysis of variance (ANOVA) and Bonferroni post-hoc test for multiple pairwise mean comparisons. The statistically significant was considered when p < 0.01. Data were showed as mean ± S.D.

Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.