Thai plants with high antioxidant levels, free radical scavenging activity, anti-tyrosinase and anti-collagenase activity

Background Ultraviolet radiation from sunlight induces overproduction of reactive oxygen species (ROS) resulting in skin photoaging and hyperpigmentation disorders. Novel whitening and anti-wrinkle compounds from natural products have recently become of increasing interest. The purpose of this study was to find products that reduce ROS in 14 Thai plant extracts. Methods To determine total phenolic and flavonoid content, antioxidant activity, anti-tyrosinase activity and anti-collagenase activity, we compared extracts of 14 Thai plants prepared using different solvents (petroleum ether, dichloromethane and ethanol). Antioxidant activities were determined by DPPH and ABTS assays. Results Total phenolic content of the 14 Thai plants extracts was found at the highest levels in ethanol followed by dichloromethane and petroleum ether extracts, respectively, while flavonoid content was normally found in the dichloromethane fraction. Scavenging activity ranged from 7 to 99% scavenging as assessed by DPPH and ABTS assays. The ethanol leaf extract of Ardisia elliptica Thunb. had the highest phenolic content, antioxidant activity and collagenase inhibition, while Cassia alata (L.) Roxb. extract had the richest flavonoid content. Interestingly, three plants extracts, which were the ethanolic fractions of Annona squamosa L., Ardisia elliptica Thunb. and Senna alata (L.) Roxb., had high antioxidant content and activity, and significantly inhibited both tyrosinase and collagenase. Conclusion Our finding show that the ethanol fractions of Annona squamosa L., Ardisia elliptica Thunb. and Senna alata (L.) Roxb. show promise as potential ingredients for cosmetic products such as anti-wrinkle agents and skin whitening products.


Background
Ultraviolet radiation (UVR) from sunlight is the most significant risk factor for nonmelanoma and melanoma skin cancers [1]. Overexposure to sunlight, in particular UVA and UVB, induces the overexpression of reactive oxygen species (ROS) which damage lipids, proteins and deoxyribonucleic acids. Collagen is the major foundation of the extracellular matrix in the dermis layer of the skin. Excessive ROS increases expression of collagenase, a protease that degrades collagen which can result in photoaging and wrinkling of the skin [2]. In addition, UV exposure induces melanin production resulting in hyperpigmentation. Tyrosinase is the key enzyme initiating skin pigmentation. Firstly, L-tyrosine is hydroxylated to form 3,4-dihidroxyphenylalanine (L-DOPA) by tyrosinase. Subsequently, L-DOPA is oxidized to DOPA quinone by tyrosinase. DOPA quinone is further converted to DOPA chrome that can be converted to 5,6-dihydroxyindole (DHI) or 5,6-dihydroxyindole-2-carboxylic acid (DHICA) [3]. The current treatments for skin aging involves hydroxyl acid to peel the epidermal layer, retinoids to reduce rough skin, and skin filler administered by injecting collagen into the skin. However, these treatments have adverse effects, such as hyperpigmentation, inflammation, cytotoxicity, irritation and bacterial infection [4]. The most popular skin whitening agent is hydroquinone, which inhibits tyrosinase, but its side effects include dermatitis, edema, allergic reactions and ochronosis [5]. Recently, researchers have focused on natural products that inhibit UV-induced ROS, suppress enzymes, and reduce melanin formation as alternatives to current treatments. For example, active phytocompounds, such as arbutin, aloesin, gentisic acid, flavonoids, hesperidin, licorice, niacinamide, yeast derivatives, and polyphenols, inhibit melanogenesis without cytotoxicity to melanocytes [6]. Thus, plants may reduce wrinkle formation and hyperpigmentation caused by sunlight exposure.
The aim of this study was to analyze 14 Thai plants extracted with three different solvents for their potential as anti-wrinkle and skin whitening ingredients. The quantity of antioxidant phenols and flavonoids was evaluated for a correlation with free radical scavenging activities, and anti-collagenase and anti-tyrosinase activities. The extracts had antioxidants that scavenged free radicals and inhibited enzymes involved in wrinkle and pigment formation. We identify Ardisia elliptica Thunb., Annona squamosa L. and Senna alata (L.) Roxb as very promising candidates for use in cosmetic products.

Determination of total phenolic content
Total phenolic content of plant extracts was evaluated using the Folin-Ciocalteu method [7]. Briefly, 50 μl of extracts at 1 mg/ml in distilled water was mixed with 50 μl of 10% Folin-Ciocalteu reagent and 50 μl of 0.1 M Gallic acid from 1.56 to 100 μg/ml was used as the standard. Total phenolic content of the extracts is expressed as mg gallic acid equivalents (GAE) per g dry plant material. All samples were analyzed in triplicate.

Flavonoid content determination
Total flavonoid content (TFC) was determined using the aluminium chloride (AlCl 3 ) colorimetric assay [7]. Briefly, 50 μl of the extracts at 1 mg/ml in 80% ethanol was mixed with 50 μl of 2% AlCl 3 solution in the well of a 96 wellplate. The plate was incubated for 15 min at room temperature. The absorbance at 435 nm was measured using a microplate reader. Quercetin from 1.56 to 100 μg/ ml served as a standard. Total flavonoid content is expressed as mg quercetin equivalents (QE) per g dry plant material. Samples were analyzed in triplicate.

DPPH scavenging activity
DPPH scavenging activity assay was performed as described by Yamasaki et al. [8]. DPPH solution was freshly prepared for each assay. Briefly, 100 μg/ml extracts or 1.56 to 100 μg/ml ascorbic acid standard in absolute methanol was mixed with 180 μl of DPPH reagent in a 96 well-plate. The reaction mixture was incubated for 30 min at room temperature in the dark. Then, the absorbance at 517 nm was measured with a microplate reader. The experiments were undertaken in triplicate. The absorbance at 517 nm of DPPH was 0.70 ± 0.02, and decreased absorbance measured scavenging activity. The scavenging ability was calculated as scavenging activity (%) = 100% × [(훥A 517 of control -훥A 517 of sample)/ 훥A 517 of control]. Percentages of DPPH scavenging activity of the extracts were compared with those of ascorbic acid, and are expressed as mg vitamin C equivalent antioxidant capacity (VCEAC) per g dry plant material. IC50 was determined from a graph of percent inhibition against concentration (from 0.78-100 μg/ ml of each extract).

ABTS scavenging activity
ABTS free radical scavenging activity was performed as previously described [9]. The ABTS •+ working reagent was prepared by mixing 7 mM ABTS • and 2.45 mM potassium persulfate at 8:12 volume/volume ratio. The working solution was kept for 16 to 18 h at room temperature in the dark. The ABTS •+ solution was diluted with absolute ethanol to give an absorbance at 734 nm of 0.70 ± 0.02. Then, 100 μg/ml extracts or 1.56 to 100 μg/ ml ascorbic acid standard in absolute ethanol was added to 180 μl of ABTS •+ working reagent in the wells of a 96 well plate. The plate was incubated for 45 min at room temperature, and absorbance was measured at 734 nm.
Experiments were undertaken in triplicate. The scavenging ability was calculated as scavenging activity (%) = 100 × [(훥 A 734 of control -훥A 734 of sample)/ 훥A 734 of control]. The percentages of ABTS scavenging activity of the extracts were compared with those of ascorbic acid, and are presented as mg vitamin C equivalent antioxidant capacity (VCEAC) per g dry plant material. IC50 was determined from a graph of percent inhibition against concentration (from 15.62-1000 μg/ml of each extract).

Determination of mushroom tyrosinase inhibition
The dopachrome method was performed with slight modification [10].

Determination of collagenase inhibition
Collagenase inhibition was determined by a previously described method [11]. Briefly, 40 μl of collagenase from Clostridium histolyticum at 0.25 units/ml in 50 mM Tricine buffer containing 10 mM CaCl 2 and 400 mM NaCl, and 10 μl of 50 mM Tricine buffer were mixed with 10 μl of the extracts or DMSO (as control). Epigallocatechin gallate (EGCG) was used as a positive control. After a 15-min incubation at room temperature, 50 μl of N-[3-(2-furyl)acryloyl]-Leu-Gly-Pro-Ala (FALGPA) was added. The absorbance was measured at 340 nm immediately and continually for 20 min. Enzyme activity was evaluated by decreased absorbance during the time interval. The percent inhibition of collagenase activity was calculated as 100 × [(Activity of control -Activity of sample)/ Activity of control]. Final concentrations of the extracts and epigallocatechin gallate were 1 and 0.1 mg/ml, respectively. IC50 was determined from a graph of percent collagenase inhibition against concentration (from 15.62-1000 μg/ml of each extract).

Statistical analyses
All experiments were carried out in triplicate and results are expressed as mean ± standard error. The correlation coefficiency (R 2 ) between antioxidant contents and antioxidant activities was determined by using SigmaPlot version 12.2 software. Difference between two means was evaluated using Student's t-test. Differences were considered significant when the P-value was less than 0.05.

Results
Extraction yields Total flavonoid content did not correlate with total phenolic content (R 2 = 0.0284, Fig. 1a).

DPPH radical scavenging activity in different extracts from 14 Thai plants
Free radical scavenging activity using DPPH as the indicator is a basic antioxidant assay [12]. As shown in Table 3, scavenging activities of the extracts varied greatly, ranging from 7.11 ± 0.59% to 96.17 ± 0.05%. The Ardisia elliptica Thunb ethanol extract had the highest scavenging activity at 96%. Moreover, the next strongest antioxidant activities (> 90%) were observed in ethanol fractions from Stemona curtisii Hook.f., Annona squamosa L., Phyllanthus acidus (L.) Skeels. and Garcinia mangostana Linn. In terms of the other solvents, Ardisia elliptica Thunb, also had the richest scavenging activity among the petroleum ether fractions, and Garcinia mangostana L. had the highest antioxidant activity in the dichloromethane fractions. The lowest scavenging ability was detected in Croton sublyratus Kurz

ABTS radical scavenging activity in different extracts from 14 Thai plants
Antioxidant activity of aqueous and lipid phases in the plants has also been evaluated by a decolorization assay using ABTS [13]. Again, ascorbic acid served as the standard antioxidant. As with the DPPH assay, scavenging activity in the ABTS assay varied greatly among the plant preparations with a similar broad range from 8.03 ± 0.54% to 99.84 ± 0.07% (Table 4). Furthermore, the next strongest scavenging activities (> 90%) were observed in the same 4 ethanol fractions as shown by the DPPH assay. In addition, no scavenging activity was found in the same 5 petroleum ether extracts. In general, the values obtained with the ABTS assay were higher than the DPPH values. Hence, activity in the ethanol extract from Senna alata (L.) Roxb. was now observed as >90%, and scavenger activity was detected in all dichloromethane extracts, and petroleum ether extracts from Annona squamosa L. and Ipomoea pes-caprae (L.) R.br. which was not detected by the DPPH assay.

Tyrosinase activity inhibition by plant extracts
The ability of compounds from the Thai plants to inhibit mushroom tyrosinase activity was evaluated using an in vitro assay with L-DOPA as the substrate. Kojic acid served as a known inhibitor, and caused maximal enzymatic inhibition of 93.38 ± 1.63%. As shown in Table 5, only ethanol extracts significantly inhibited tyrosinase activity, with Ardisia elliptica Thunb. preparations being the exception. The petroleum ether and dichloromethane fractions of Ardisia elliptica Thunb. inhibited tyrosinase activity by approximately 20%. The ethanol fraction from Rhinacanthus nasutus (L.) Kurz (IC 50 value of 271.50 μg/ml) was the most potent tyrosinase inhibitor, followed by the ethanol extracts from Ardisia elliptica Thunb. and Phyllanthus acidus (L.) Skeels. Other ethanol fractions significantly decreased enzymatic activity by more than 20% (Table 5), whereas the remaining extracts did not have detectable inhibitory activity (data not shown).

Discussion
Solar radiation is a significant environmental factor in skin damage and can induce skin cancer [14]. UV radiation causes a pro-inflammatory response, extracellular matrix degradation and antioxidant depletion [15,16]. UV causes formation of reactive oxygen species (ROS) that induce hyperpigmentation and collagenase expression [17,18]. Our study investigated 14 Thai plants extracted with three different solvents for their potential as anti-wrinkle and skin whitening ingredients. In this study, we used petroleum ether, dichloromethane and ethanol for plant extraction using Soxhlet apparatus. Ardisia elliptica Thunb. had the highest yield in the petroleum ether and ethanol extracts, whereas Garcinia mangostana L. had the highest percent yield from dichloromethane extraction. These solvents are a series of organic solvents with increasing polarities. Variation among the percent yields depended on the plant species, and probably reflected differences in chemical composition of the plants.
Phenolics are the largest group of phytochemicals found in plants and they have various biological activities in animals, including humans [19]. Total phenolic content in the plants was determined by the Folin-Ciocalteu method. Overall, the ethanol fraction had the richest phenolic content, followed by dichloromethane, while petroleum ether with low polarity had the lowest phenolic content compared to the other solvents. In this study, Ardisia elliptica Thunb. had the highest phenolic content in all three types of extracts. In previous studies, dichloromethane leaf extracts of Ardisia elliptica Thunb. have a phenolic content of 101 ± 1.3 mg GAE per g dry plant material, which is more than the content in a twig extract [20]. Moreover, a methanol extract of ripe Ardisia fruit contained 5.64 ± 0.37 g GAE per 100 g extract [21]. Hence, leaves and fruits of Ardisia elliptica Thunb. have a high phenolic content that can be easily extracted with methanol, dichloromethane and ethanol.
Flavonoids are pigments in flowers, leaves, fruits and seeds. These compounds are secondary metabolites of plants and are widely distributed among plant species [22]. Next, the flavonoid content within the Thai plants was evaluated using the aluminium chloride colorimetric assay. Our results showed that the highest flavonoid quantity was found in the ethanol extract from Senna alata (L.) Roxb leaves. In a previous study, high flavonoid content was found in water (4.25 mg QE per 100 g) and methanol fractions (3.97 mg QE per 100 g) of Senna alata (L.) Roxb. [23]. Thus, Senna alata (L.) Roxb preparations have a high flavonoid content when extracted with high polarity solvents including ethanol, methanol and water. Ardisia elliptica Thunb. had the richest flavonoid content in the dichloromethane fraction. Fruit of this plant also has a high flavonoid content 36.91 ± 2.37 mg QE per g extract [24]. Hence, fruit and leaves of Ardisia elliptica Thunb. are rich in flavonoids. Total flavonoid content did not correlate with total phenolic content. However, flavonoids have many biological activities such as UVB protection [25], anti inflammatory [26], anti-hepatotoxicity [27] and anti cancer [28].
Free radical scavenging activity using DPPH and ABTS assay. In the DPPH assay, DPPH receives a hydrogen atom from an antioxidant [29]. We found that Ardisia elliptica Thunb ethanol extract had the highest scavenging activity. Other investigators have also reported that dichloromethane fractions of Ardisia elliptica Thunb. leaves and stems have high levels of antioxidant activity as determined by the DPPH assay, and, hence, this plant is very interesting to investigate further as a herbal treatment [20]. The extracts from the ethanol fraction with high polarity clearly showed better antioxidant activity than fractions with lower polarities containing dichloromethane and petroleum ether. Ethanol extracts contained the highest levels of free radical scavenging activity compared with the other extracts, and all ethanol extracts were Each value is mean ± S.D. of triplicate independent analyses. Significantly different from the control group ( * P < 0.05, ** P < 0.01, *** P < 0.001). NA not available active. In the ABTS assay, ABTS is converted to its radical cation by the addition of potassium persulfate. In the presence of an antioxidant, the reactive ABTS cation (or ABTS • + ) is converted to its colorless natural form [9]. In agreement with the DPPH assay, ethanol extracts contained the highest levels of scavenger activity as compared with the other extracts. Again, the highest scavenging activities in ethanol, dichloromethane and petroleum ether extracts were from the same plants as shown by the DPPH assay.
The results of the DPPH and ABTS assays were highly correlated as expected (Fig. 1f). However, total flavonoid content of the plant extracts did not correlate with free radical scavenger activity as detected by the DPPH assay (Fig. 1c) or by the ABTS assay (Fig. 1e). Our findings of no significant relationship between flavonoid content and scavenger activity using the ABTS assay is consistent with other investigators' results [30]. By contrast, total phenolic content of the plant preparation positively correlated with scavenger activity measured by both assays (Fig. 1b and d) in agreement with a previous study [31]. Noticeably, the scavenging activity depended on total phenolic content and solvents with high polarity, such as ethanol and dichloromethane. These results suggest that the phenolic content is the major constituent with antioxidant activity in the 14 Thai plants.
Melanin, the major pigment of skin and hair color, is synthesized by melanocytes within melanosomes. Overproduction and accumulation of melanin in skin may lead to pigmentary disorders and aesthetic problems. Hyperpigmentation occurs in sun-exposed areas of the skin [32]. In the melanogenesis, tyrosinase is the key enzyme in the ratelimiting step in which L-tyrosine is hydroxylated to L-DOPA, which is further oxidized into DOPAquinone. After that, it is converted into DOPAchrome that is a substrate for melanin synthesis [3]. Downregulation of tyrosinase activity has been proposed to be responsible for decreased melanin production. The development of novel whitening phytochemical compounds from natural products has recently become a growing trend. Our finding showed that the ethanol fraction from Rhinacanthus nasutus (L.) Kurz was the most potent tyrosinase inhibitor, followed by the ethanol extracts from Ardisia elliptica Thunb. and Phyllanthus acidus (L.) Skeels. Obviously, 7 plants from 14 plants had the high phenolic content, especially Ardisia elliptica Thunb. and Annona squamosa L.. Moreover, Senna alata (L.) Roxb. had the richest flavonoid content which can inhibit tyrosinase activity. Active compounds from the plants such as arbutin, aloesin, gentisic acid, flavonoids, hesperidin, licorice, niacinamide, yeast derivatives, and polyphenols, can inhibit melanogenesis without cytotoxicity to melanocytes [6].
Collagenase is a transmembrane zinc peptidase that cleaves the X-Gly bond of collagen. Collagen is an abundant structural protein and extracellular matrix component [33]. Decreased collagen and elastin fibers increases with age and damage from UV radiation inducing wrinkled skin [34]. Collagenase inhibition has been proposed to prevent skin aging. Of those causing inhibition in our study, Ardisia elliptica Thunb. exhibited the highest level of collagenase inhibition, followed by Annona squamosa L., Senna alata (L.) Roxb., and Croton sublyratus Kurz in rank order. In a previous study, cocoa pod extract had phenolic acid and flavonoids that inhibited elastase and collagenase activity [35]. Notably, three ethanol extracts (Ardisia elliptica Thunb., Annona squamosa L. and Senna alata (L.) Roxb. inhibited both tyrosinase and collagenase. These plants also had high phenolic and flavonoid levels, and antioxidant activity. Interestingly, these extracts have possible uses as ingredients for cosmetic products.

Conclusion
Our results demonstrate that extracts of 14 Thai plants had varying degrees of total phenolic and flavonoid content as well as free radical scavenging activities, depending on the extraction solvents. There was a high correlation between total phenolic content and free radical scavenging activity ass assessed by the DPPH and ABTS assays. The ethanol fraction of Ardisia elliptica Thunb. had the highest phenolic content, followed by Annona squamosa L. Both plants significantly inhibited tyrosinase and collagenase activities, while Rhinacanthus nasutus (L.) Kurz showed the highest tyrosinase inhibition. Moreover, Senna alata (L.) Roxb. was richest in flavonoid content, and also exhibited tyrosinase and collagenase inhibitory behavior. The ethanol fraction of three plants, namely Annona squamosa L., Ardisia elliptica Thunb and Senna alata (L.) Roxb., have the potential to be ingredients in cosmetic products for antiwrinkling as well as skin whitening. Further studies are necessary to investigate the active components and safety of these extracts.