Potential Cosmeceutical Applications and Evaluation of Human Skin Irritation of Tagetes erecta L. Flower Extract

Background: Tagetes erecta L. has been traditionally used for treatment of skin diseases. Objectives: This study aimed to evaluate in vitro skin anti-aging, antioxidant, antibacterial activities, and human skin irritation of T. erecta flower extract for development as a cosmeceutical skin product using aqueous, 50%, and 95% ethanolic extracts of T. erecta flower. Methods: Total phenolic content was defined. Quantitative analysis of syringic acid was determined by using HPLC. Antioxidant activities were performed by measurement of 2,2-diphenyl 1-pichylhydazyl free radical scavenger. Anti-aging assays were monitored inhibitory effects on tyrosinase, elastase, hyaluronidase, and collagenase activities. Antibacterial effects against Staphylococcus aureus and Staphylococcus epidermidis was also investigated. The extract was formulated as a lotion and its stability study was evaluated and human skin irritation tests were performed. Results: The 95% ethanolic extract had significantly higher total phenolic content and syringic acid content than the other extracts. The 95% ethanolic extract also had stronger inhibitory activities on DPPH scavenging and against tyrosinase, elastase and hyaluronidase enzyme activities than those of 50% ethanolic or water extracts. The 95% ethanolic extracts showed a better antibacterial activity against S. aureus and S. epidermidis . The developed lotion containing 1% of T. erecta extract using 95% ethanol was stable. Moreover, skin irritation was not observed in healthy volunteers after applying the T. erecta lotion. Conclusion: This study indicates that T. erecta flower extracts, especially a 95% ethanolic extract, are potential sources for development of cosmeceutical skin products against enzymes/reactions responsible for skin aging and bacterial skin infection. This study focused on the simultaneous efficacy including antiaging properties throughout inhibitory properties on oxidation reaction and antibacterial activity, along with human skin irritation testing of the Tagetes erecta L. flower extract by using safe solvents, ethanol and water, for extraction. The results revealed that ethanolic extract using 95% ethanol was found to contain higher phenolic compounds and syringic acid than the others and had greater inhibitory effects on antioxidant, skin aging-related enzymes and antibacterial activity. The prototype formulation containing 1% T. erecta ethanolic extract was physically stable and non-irritation to the skin in healthy volunteers.


INTRODUCTION
Tagetes erecta L., commonly known as marigold or "Dow Ruang" in Thai, belongs to the Asteraceae family. It is generally a wild growing plant, but it is also cultivated for ceremonial, decorative, and herbalmedicinal remedies. It is rich of phenolic compounds. [1] Phytochemical studies have shown that thiophenes, flavonoids, carotenoids, triterpinoids, methyl-3,5-dihydroxy-4-methoxy benzoate, and ethyl gallate are present in T. erecta flowers. [2] Syringic acid and β-amyrin, which were isolated from T. erecta flowers, were the bioactive compounds due to they possess inhibitory activities on hyaluronidase, elastase, and collagenase enzymes. [3] Traditionally, this plant has been claimed to be effective for treatment of skin diseases including wounds, burns, carbuncles, eczema, as well as piles, muscular pain and muscle relaxant, rheumatic pain, respiratory diseases, and kidney troubles. [4][5][6] The flowers of T. erecta have been reported to contain various pharmacological activities such as antioxidant, [1,7,8] antibacterial (Streptococcus mutans, Pseudomonas aeroginosa), [7] anxiolytic and sedative-like activities, [9] and treatment of ulcerative colitis. [8] Considering cosmeceutical application, methanolic extract of T. erecta flowers was reported to exhibit anti-elastase, anti-hyaluronidase, and anti-collagenase activities. [3] Ethyl acetate extract of T. erecta flower showed antioxidation or anti-tyrosinase effects. [10] Vallisuta et al. also reported that ethyl acetate extract of T. erecta flower possessed a strong elastase inhibitory activity and its ethanolic extract showed anti-tyrosinase effect. [11] Creams containing 30% nanostructured lipid carriers of T. erecta flower ethyl acetate extract reduced skin wrinkles compared with untreated and cream base in healthy volunteers after application for 8 weeks. [12] These reports showed that T. erecta flower extract had the potential to be used as cosmeceutical products.
Nowadays, cosmeceutical products for prevention and treatment for skin aging have received increasing interest. Mechanisms underlying the development of skin aging involve many enzymes and other factors. Elastase is an enzyme that is responsible primarily for cleaving elastin in skin causing loss of skin elasticity and resilience. Hyaluronidase is an enzyme that is capable of cleaving hyaluronic acid or hyaluronan which is mucopolysaccharide providing skin hydration and skin elasticity. Collagen in skin acts as supporter of skin structure providing skin firming and is responsible for the tensile strength of the skin. Collagen can be broken down by collagenase enzyme. [13] Increasing degradation of elastin, hyaluronic acid, and collagen are significant in the aging process. Thereby, the inhibition of these three enzymes activities can prevent loss of skin elasticity, reduce skin wrinkles, delay skin aging, and improve skin hydration. Additionally, skin aging can be caused by UV exposure induced oxidation reactions. The continuous exposure of UV radiation can stimulate physical changes in skin through the formation of lipid peroxides and reactive oxygen species leading to loss of elasticity, skin wrinkles, uneven pigmentation, brown spots, and melanoma. [13] Thus, maintaining antioxidant homeostasis is also a proper strategy to prevent skin aging by quenching reactive oxygen species produced during UV radiation exposure. Additionally, melasma is also a problem of skin aging caused by an increase of melanin resulting from induction of tyrosinase. Tyrosinase is known as an enzyme that plays a role in the melanin synthesis in melanocytes. The activation of tyrosinase enzymes, which mostly results from UV radiation exposure, results in hyperpigmentation in epidermis and dermis. The accumulation of excessive pigmentation results in melasma, age spots, and freckles associated with age. [14] Thus, tyrosinase inhibitors are useful cosmeceutical ingredients as they reduce uneven pigmentation, brown spots, and melanoma on skin. Besides, cosmeceutical products have been also utilized for prevention or treatments of bacterial infections that can cause blister or acne. Acne is the most common disorder of human skin associated with microbial infection. Propionibacterium acnes, Staphylococcus epidermidis, Staphylococcus aureus, and Candida albicans are commonly found to cause acne. [15][16][17] However, antimicrobial agents which are mostly synthetic compounds, can cause drug resistance, adverse effects, and skin allergy/ irritation. [16] There is increasing interest in antimicrobial products for acne treatment from natural sources in order to decrease the incidence of drug resistance. Aforementioned, the previous studies showed that aqueous and ethanolic extracts of T. erecta flowers possessed a good antioxidant activity. Antielastase, anti-hyaluronidase, and anti-collagenase activities were also found in methanol extract of T. erecta flowers. Moreover, ethyl acetate extract of T. erecta flowers possessed anti-elastase activity. These results suggest that T. erecta flower has a potential to be developed as a cosmeceutical product for skin care products. Nevertheless, the residual methanol and ethyl acetate may remain in the products which are able to penetrate into skin and can cause skin toxicity and skin irritation. [18,19] The biocompatible and low toxicity solvents should be utilized for the plant extraction. In general, the use of different solvents for plant extraction may provide different chemical constitutes and demonstrate the different biological effects. To the best of our knowledge, biological effects and benefits on the skin of T. erecta flowers extracted with water or ethanol have been rarely investigated. Therefore, the aim of this study was to investigate the skin efficacy including antioxidant, anti-tyrosinase, anti-elastase, anti-hyaluronidase, anti-collagenase, and antibacterial activity against S. aureus and S. epidermidis of T. erecta flower extracts by using biocompatible and the low toxicity solvents including water and ethanol for the plant extraction. Additionally, a skin product prototype was then developed in order to test the stability of T. erecta flower extract when formulated as a skin care product and to evaluate skin irritation in healthy volunteers after applying the product prototype.

Preparation of plant extracts
T. erecta flowers, 60 days old at flowering stage, were collected from an organic plantation in Nakhonratchasima province, Thailand. A voucher specimen (No. TE2561) has been deposited at the Faculty of Pharmacy, Mahasarakham University, Thailand. The flowers were washed twice with DI water. The yellow parts of the flowers were then separated, and small pieces were dried at 45°C in hot air oven for 24 h. The dried samples were ground into fine powder using a herbal grinder. For preparation of aqueous extract, the dried powder was added to hot water at a controlled temperature of 70-80°C and allowed to infuse for 1 h, filtered through filter paper (Whatman ® Number 1) and the filtrate retained as aqueous extract. Afterwards, the filtrates were lyophilized to provide the aqueous extract. For preparation of 50% and 95% ethanolic extracts, a maceration technique was employed. The dried powder of T. erecta flowers was macerated with 50% or 95% ethanol for 5 consecutively days with stirring 3 times a day. After filtration, the liquid extract was concentrated using a rotary evaporator and then the concentrated solution was dried by using lyophilizer. Yields (%) were calculated by dividing the weight of obtained extract by the dry weight of T. erecta powder. The extracts were kept at -20°C until used.

Determination of total phenolic contents
The Folin-Ciocalteu colorimetric method was used to determine total phenolic contents in the three extracts as described previously. [20] Briefly, aqueous extract was dissolved in water while 50% and 95% ethanolic extracts were dissolved in ethanol to make 0.1-1 mg/mL solutions. The mixture containing the extract or gallic acid (20 µL), 10% folin-ciocalteu reagent (100 µL), and 7.5% Na 2 CO 3 (80 µL), were mixed in 96-well plate and kept in darkness for 2 h. Afterwards, the absorbance was recorded at 765 nm in UV-microplate reader (Spectro ® StarNano, Germany). Total phenolic content was calculated based on the amount of gallic acid equivalent (µg GAE/mg extract). The equivalences were read from the straight line generated by linear regression of absorbances and concentrations of gallic acid.

Quantitative analysis of syringic acid by HPLC
Determination of syringic acid in the extracts was performed by using HPLC following the previous published article with minor modification. [3] HPLC Shimadzu SCL-10AVP (Japan) coupled with DAD detector was employed. A C 18 column (250 x 4.6 mm, 5 µm) from Inversion GL science (Japan) was used. Methanol (20%) and 1.25% acetic acid in water (80%) were mobile phases. Flow rate was set at 1 mL/min. The extracts were dissolved in the mixture of mobile phases. Syringic acid in the extracts was detected at a wavelength of 280 nm at retention time of 31 min.

Antioxidant activity
DPPH assay was employed according to the previous published article. [21] Aqueous extract and the ethanolic extracts solutions in various concentrations were prepared in DI water and ethanol, respectively. Ascorbic acid was used as a positive control by dissolving in DI water. The extract or ascorbic acid solution (20 µL) was mixed with 0.08 mM DPPH solution in ethanol (80 µL) in a 96-well plate. The reaction was allowed to take place at room temperature (25-30°C) for 15 min and then the absorbance of the mixture was measured at 517 nm using a UV-Vis microplate reader (Spectro ® StarNano, Germany). All determinations were performed in triplicate. The percentage of inhibition in each concentration was calculated according to the following formula: The percentage of inhibition = [(Absorbance of the DPPH solution -Absorbance of the mixture of sample and DPPH solution) / Absorbance of the DPPH solution] × 100 Then, half maximal inhibitory concentration or 50% inhibitory concentration (IC 50 ) was obtained from the linear regression of the percentage of inhibition against sample concentration.

Evaluation of biological activities of T. erecta flower extracts
Anti-tyrosinase activity Anti-tyrosinase activity was investigated by the dopachrome method as described previously using L-DOPA as a substrate and mushroom tyrosinase as an enzyme. [22] Kojic acid was selected as a positive control. Different concentrations of aqueous extract and ethanolic extracts solutions were prepared in DI water and methanol, respectively. Tyrosinase enzyme solution was prepared in phosphate buffer solution (PBS) pH 6.8 at the concentration of 0.16 mg/mL. L-DOPA was dissolved in PBS pH 6.8 to make 0.85 µM/mL solution. Then, tyrosinase enzyme solution (20 µL) was mixed with PBS pH 6.8 (140 µL) and the extract or positive control solution (40 µL) in a 96-well plate. After incubation at room temperature (25-30°C) for 10 min, L-DOPA solution (20 µL) was added to the mixture. The absorbance of the mixture was read after incubation for 20 min in room temperature (25-30°C) using UV-Vis microplate reader (Spectro ® StarNano, Germany) at 492 nm. The IC 50 value was further obtained from the linear regression between the percent inhibition and concentration. The percentage of tyrosinase inhibition was calculated using the following formula: Where A: absorbance of the mixture of tyrosinase enzyme, PBS pH 6.8, L-DOPA, and methanol B: absorbance of the mixture of PBS pH 6.8, L-DOPA, and methanol C: absorbance of the mixture of tyrosinase enzyme, PBS pH 6.8, L-DOPA, and extract/positive control D: absorbance of the mixture of PBS pH 6.8, L-DOPA, and extract/positive control

Anti-elastase activity
The assay employed was based on methods from previous literature that monitor the release of p-nitroanilide, product of the enzyme reaction. [23] SANA, which is a substrate, was dissolved in 0.1 mol/L Tris-HCl buffer solution pH 8.0 to make the solution at the concentration of 1 mmol/L. Elastase enzyme solution was prepared to make a 0.03 U/mL solution in 0.1 mol/L Tris-HCl buffer solution pH 8.0. Tris-HCl buffer solution pH 8.0 (0.1 mol/L) and methanol were used to dissolve the aqueous and ethanolic extracts, respectively. Oleanolic acid, a positive control, was dissolved in methanol. The extract solution or oleanolic solution (20 µL) in various concentrations was preincubated with elastase enzyme (40 µL) in a UV transparent 96-well plate at room temperature (25-30°C) for 10 min. Afterwards, SANA solution (200 µL) was added to the mixture. After incubation at room temperature (25-30°C) for 60 min, the absorbance of sample was measured at 410 nm using a UV-Vis microplate reader (Spectro ® StarNano, Germany). The following formula was employed for calculation: Anti-hyaluronidase activity Hyaluronidase inhibitory activity was measured by a modified method from the previous literature. [23] Buffer A was prepared by mixing 82 mg of Na 2 HPO 4 , 51 mg of NaH 2 PO 4 , 0.225 NaCl, 5 mg of bovine serum albumin and dissolving in DI water to make a final volume at 50 mL. Buffer B containing 63 mg of Na 2 HPO 4 and 1.76 g of NaH 2 PO 4 in 50 mL of DI water was prepared. Bovine serum albumin solution was prepared by mixing 50 mg of bovine serum albumin, 98.4 mg of CH 3 COONa, 237 mg of CH 3 COOH, and 40 mL of DI water. The pH value of the bovine serum albumin solution was adjusted to pH 3.75 by adding hydrochloric acid solution and DI water was then added to adjust the final volume to 50 mL. Aqueous extract was dissolved in DI water. Ethanolic extracts and oleanolic acid (positive control) were dissolved in methanol. Hyaluronidase enzyme solution (15 U/mL) was prepared in buffer A. Hyaluronic acid, which is a substrate, was dissolved in buffer B to make a solution at concentration of 0.03%. The extract (50 µL) in various concentrations was pipetted to a microtube to which 100 µL of hyaluronidase enzyme solution was then added. After incubation at 37°C in a hot air oven for 10 min, hyaluronic acid (100 µL) was added and continually incubated at 37°C for 45 min. The decomposition of hyaluronic acid was stopped by adding the bovine serum albumin solution (1 mL) to the mixture and placed in room temperature (25-30°C) for 10 min. The solution was pipetted to a 96-well plate and the absorbance measured of undigested hyaluronic acid at 600 nm using a UV-Vis microplate reader (Spectro ® StarNano, Germany). The percentage of enzyme inhibition was calculated from absorbance of sample reaction divided by absorbance of hyaluronic acid. IC 50 value was obtained from the linear regression of %inhibition against concentration.

Anti-collagenase activity
Collagenase activity measurements employed colorimetric assay kits using FALGPA as substrate in the enzymatic reaction. Aqueous extract and ethanolic extract solutions were prepared in water and ethanol, respectively. The mixture containing 5 µL of the extract, 10 µL of 0.35 U/mL collagenase enzyme, and 85 µL of collagenase assay buffer in a UV transparent 96-well plate was incubated in a hot air oven at 37°C for 10 min. The reaction was started by adding 40 µL of FALGPA and 60 µL of collagenase assay buffer to the mixture. The absorbance at 345 nm was tight cap. Lotion base containing the same ingredients and quantities, but no adding T. erecta flower extract was prepared as a control sample.

Stability test
Heating and cooling cycles were employed for stability testing of T. erecta lotions. The lotion in a glass bottle was stored at 4°C in a refrigerator for 12 h (n=3) followed by 12 h at 45°C with a total of 6 repeated cycles of heating and cooling. Physical appearance including texture, color, phase separation, precipitation, and odor, and pH value were evaluated prior and after heating and cooling cycles.

Skin irritation test in healthy volunteers
The protocol for the skin irritation test in healthy volunteers (n=30) was reviewed and approved by the human research ethical committee, Faculty of Pharmacy, Mahasarakham University, Thailand (Approval No. 007/2560). Participants were entered according to specific inclusion criteria and excluded according to specific exclusion criteria. Inclusion criteria were healthy volunteers between 18 and 60 years of age. Volunteers were excluded if pregnant, breast feeding, in treatment with medicines, experiencing heart, liver, kidney, endocrine or immune systems diseases, experiencing skin allergies from the ingredients in T. erecta lotions, experiencing allergy to pollen, and participation into any clinical study. All participants gave written informed consent. Lotions (0.5 mL) were applied at inside of right upper arm for T. erecta lotion and left upper arm for lotion base and gently spread over an area of 1.5x2 cm 2 . Afterward, the test sites were covered with gauze for 30 min. Clinical assessments of the test sites were performed after removing the gauze consecutively at 24 h and 48 h. Guideline from the International Contact Dermatitis Research Group was applied for evaluating skin irritation based on morphology ( Table 1). The study was to be discontinued if the participant required or if the investigators deemed it necessary for medical reasons.

Statistical analysis
All results were presented as mean and standard deviation (SD). One-way ANOVA with least significant difference (LSD) post hoc test was employed using IBM SPSS Statistics 26. P<0.05 was considered as indicative of statistically significant difference.

Total phenolic content in T. erecta flower extract
The percent yields of extractions are shown in Table 2. The results showed that the highest percent yield was found in 50% ethanolic extract followed by aqueous extract and 95% ethanolic extract, respectively. For total phenolic content, the 95% ethanolic extract possessed the highest total phenolic content at 86.75±0.49 µg GAE/mg extract at p<0.05 followed by 50% ethanolic extract and aqueous extract, respectively as shown in Table 2.

Quantitative analysis of syringic acid in T. erecta flower extract
Syringic acid was used as a chemical marker in the extracts and quantitatively analyzed by using HPLC. The chromatogram of syringic acid and the extracts are presented in Figure 1. The retention time of syringic acid was 31 min. Syringic acid was able to be determined in three extracts. For quantitative analysis, the amount of syringic acid found in the three extracts is shown in Table 2. Among the extracts, 95% ethanolic extract contained the highest amount of syringic acid at p<0.05.

Antioxidant activity
Antioxidant assessment, DPPH assay revealed activity for all extracts and the inhibitory concentration (IC 50 ) of samples toward DPPH free measured immediately after adding the substrate and then continuously for 15 min using a UV-microplate reader (Spectro ® StarNano, Germany) providing the linear regression of absorbance (Y-axis) and time (X-axis). The slope from the linear regression was used to calculate percentage of inhibition, as given below:

% Inhibition = [(A-(B-C)) × 100] / A
Where A: slope obtaining from the linear regression of the mixture of collagenase enzyme, collagenase assay buffer, and FALGPA B: slope obtaining from the linear regression of the mixture of sample solution, collagenase enzyme, collagenase assay buffer, and FALGPA C: slope obtaining from the linear regression of the mixture of sample solution, collagenase assay buffer, and FALGPA Then, IC 50 was obtained from the linear regression between the percentage of inhibition and concentration.

Antibacterial activity
Measurement of antibacterial activity in this study was modified from the previous literature. [24] S. aureus DMST 8013 and S. epidermidis DMST 15505 were purchased from Department of Medical Sciences, Thailand. Bacteria was grown in MHB at 37°C for 18-20 h and then diluted in normal saline solution to approximately 10 6 cfu/mL. The paper disc diffusion method was used to determine inhibition zone. Diluted bacterial suspension was spread on MHA and then a sterile paper disc (6 mm diameter) was placed on the surface of MHA. Extracts were diluted in DMSO and then filtrated using a 0.45 µm sterile membrane filter. The extracts were dropped (20 µL) onto the paper disc at the final amount 2 mg/disc. Clindamycin (2 µg/disc) was used as a reference antibiotic. Then, the plates were incubated at 37°C for 18 h. All tests were performed in triplicate. The diameter of inhibition zones (mm) was measured. Microbroth dilution method was used to determine the minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC). Overnight cultures of bacteria were diluted in MHB to approximately 10 6 cfu/mL and then added (100 µL) into a 96-well plate. Extracts and gentamycin (a reference antibiotic) were diluted in MHB to various concentration and then added (100 µL) into the bacterial suspension in the well plate. After incubation at 37°C for 24 h, bacterial growth was observed from the media turbidity. MIC was determined as the lowest concentration that showed no bacterial growth in the wells. Then culture from concentrations at ≥ MIC were transferred (20 µL) to soyabean casein digest agar. After incubation at 37°C for 24 h, bacterial colonies were counted. MBC was determined as the lowest concentration that no bacterial colony. All tests were performed in triplicate.

Formulation of T. erecta flower extract lotions
In order to determine the stability and skin irritation of T. erecta flower extract in the formulation, T. erecta flower extract lotions were formulated. A beaker method was employed by using polysorbate 20 and sorbitan monooleate as the emulsifiers. Carbomer 934 (0.4%) was dispersed in water and then gel was formed after adding triethanolamine (1.5%). The water phase containing polysorbate 20 (2.4%), polyethylene glycol 400 (0.5%), carbomer 934 gel, and water (adjust to give 100%) was heated up in water bath at 75-78°C. The oil phase included sorbitan monooleate (3.5%), glyceryl monostearate (2.0%), petrolatum (2%), and isopropyl myristate (2.0%), was heated in water bath at 72-75°C. Afterwards, the oil phase was added into the water phase and gently mixed. The extract (1%) was dissolved in ethanol. After the lotion had cooled to 40-50°C, the extract solution was added into the lotion and mixed until the smooth lotion was formed. The developed lotion was kept in glass bottle with radical scavenging activity are shown in Figure 2. The 95% ethanolic extract showed the strongest radical scavenging activities compared with 50% ethanolic and aqueous extracts (p<0.05). The 95% ethanolic extract was comparable to the positive control, ascorbic acid (p>0.05).

Anti-tyrosinase activity
Inhibition of tyrosinase enzyme activity is shown in Figure 3. IC 50 values of three extracts were significantly lower than that of the positive control, kojic acid. The lowest of IC 50 value was possessed by the ethanolic extracts. The 95% ethanolic extract exhibited the best activity followed by 50% ethanolic and aqueous extract, respectively.   Anti-elastase activity The ethanolic extracts exhibited anti-elastase activities that were more potent than that of oleanolic acid, a positive control (p<0.05) (Figure 4). Whereas the aqueous extract showed no anti-elastase activity even when the concentration was as high as 1 mg/ml.

Anti-hyaluronidase activity
Assay of anti-hyaluronidase activity, was based on precipitation of hyaluronic acid which is used for high throughput screening for hyaluronidase inhibitors. [25] Among three extracts, 95% ethanolic extract showed the highest activity in hyaluronidase inhibition followed by 50% ethanolic extract and aqueous extracts, respectively. However, three extracts had weaker inhibitory effect than positive control, oleanolic acid (IC 50 of 17.28±2.62 µg/mL) as shown in Figure 5.

Anti-collagenase activity
Three extracts had activities against collagenase enzyme and were ranked in the order of 50% ethanolic extract > 95% ethanolic extract > aqueous extract as illustrated in Figure 6. The efficacy against collagenase enzyme of the 50% ethanolic extract was comparable to the positive control. However, the IC 50 values of the aqueous and 95% ethanolic extracts were higher than that of positive control, epigallocatechin gallate (p<0.05).

Antibacterial activity
From Table 3, 95% and 50% ethanolic extracts showed antibacterial activity against S. aureus and S. epidermidis whereas aqueous extract did not inhibit bacterial growth. The 95% ethanolic extract inhibited bacterial growth more than the 50% ethanolic extract. In addition, 95% ethanolic

Stability of T. erecta lotions
According to the results of the biological activity studies, T. erecta flower that was extracted using 95% ethanol showed a greater potential to develop as a cosmeceutical product. Therefore, T. erecta lotions using 95% ethanolic extract was developed as a prototype product in order to examine the stability of the extract. T. erecta lotion after being freshly been found in various plants and possesses antioxidant activity, [26][27][28] cyclooxygenase-2 inhibitory, [28] and anticancer effects. [29] In the current study, syringic acid was found to be higher in the 95% ethanolic extract compared with the other extracts. The highest phenolic content was also found in 95% ethanolic extract. These results demonstrated the correlation between ethanol concentration for extraction and syringic acid content and total phenolic contents. A higher ethanol concentration in the extraction process, higher syringic acid and total phenolic contents were obtained. Moreover, the three T. erecta flower extracts in the present study had higher amount of total phenolic contents compared to 95% ethanolic, 50% ethanolic and aqueous extract of T. erecta flower extracts in the previous study (48.27±0.26, 46.36±0.26, and 8.50±0.10 µg GAE/mg extract, respectively). [1] Considering the process of extraction, in the previous study, the extracts were obtained by maceration at 30°C for 24 h [1] while in this study, aqueous extract was obtained by soaking at 70-80°C for 1 h and ethanolic extracts were prepared by maceration at room temperature (25-30°C) for 5 days. These findings showed that temperature in the case of aqueous extract and duration of maceration in the case of ethanolic extracts played crucial roles in determining the phenolic and syringic acid content. For investigation of antioxidant activity, the method was based on a single relatively stable reagent, DPPH which is the most popular method because of it is controllable and simple to set-up becoming a routine tool for the rapid identification of antioxidants in plant extracts. [30] The results in this study revealed a high correlation between antioxidant activities and total phenolic content as well as syringic acid contents. Increasing ethanol concentration for the plant extraction increased total phenolic content, syringic acid concentration, and antioxidant activity. The highest total phenolic content possessed the highest antioxidant activity. From many antioxidative studies, phenolic compounds have been proven to be the best free radical scavengers and inhibition of lipid peroxidation. [31] Previously, the antioxidant capacity of T. erecta flower extracts had been reported to markedly contribute from the amount of phenolic compounds in the extracts. [32] Thereby, the phenolic compounds in T. erecta flower may play a crucial rule in its antioxidant activity. Several studies have been investigated potential activities of phenolic compounds on skin aging. The phenolic compounds have been proposed for aging prevention and contributed to treatment of skin aging problems. [33] Possible aging prevention or anti-aging properties of T. erecta flower, which is rich of phenolic compounds, has been considered. Aforementioned, skin aging is caused by the increased production of reactive oxygen species, elastase, hyaluronidase, collagenase, and tyrosinase activities leading to loss of skin strength, loss of skin elasticity, skin dryness, wrinkles and sagging formation. Approaches that inhibit these enzymes activities, can be applied as useful methods to protect skin aging. This study revealed an anti-aging property of T. erecta flower via inhibition these enzymes activities. The 95% ethanolic extracts of T. erecta flower showed the greatest potential to be an anti-aging agent.
In comparison to previous studies, the 95% ethanolic extract in this study (IC 50 of 5.05±1.41 µg/mL) showed anti-elastase effects similar to the methanol and butanol extracts of T. erecta flower in the previous study (IC 50 of 4.13±0.93 and 4.01±1.37 µg/mL, respectively). [3] A recent study demonstrated that the ethanolic T. erecta flower extracts had the greater anti-elastase and anti-tyrosinase activities compared with ethyl acetate T. erecta flower extracts in the previous study. The ethyl acetate T. erecta flower extracts at concentration of 125 µg/mL showed % inhibition at 5.5% for inhibitory activity against elastase enzyme and showed IC 50 value for anti-tyrosinase activity at 1,467 µg/mL. [11] However, the aqueous extract in the current study at a concentration more than 1 mg/mL showed % inhibition less than 50 % in the anti-elastase study. Therefore, IC 50 values of aqueous extract cannot be determined. These findings prepared, was light-yellow, had smooth texture and was easily applied onto the skin. After passing through heating and cooling cycles, its texture, odor, and color were stable. There was no phase separation or solid precipitation. The pH values of T. erecta lotion containing 1% of the 95% ethanolic extract were stable at pH 6.15±0.03.

Skin irritation of T. erecta lotions in healthy volunteers
A total of 30 healthy volunteers (11 men and 19 women, aged 22.50±1.12 years) were recruited from the Mahasarakham University (Thailand) and the surrounding area. The volunteers were received verbal and written information and gave their informed consent. All participants completed the study without protocol deviation. After application of the developed lotion containing 1% of 95% ethanolic extract or lotion base there were no signs or symptoms of skin irritation in 30 healthy volunteers immediately or after 30 min, 24 h, and 48 h (Table 4).

DISCUSSION
T. erecta flowers not only contribute to ceremonial or decorative purposes but have also been reported to be utilized as an herbal medicine. Antielastase, anti-hyaluronidase, and anti-collagenase properties of T. erecta have been established in previous studies. However, the previous studies investigated the efficacy in T. erecta flower extracted by methanol, butanol, and ethyl acetate, in which residual methanol, butanol, and ethyl acetate may remain in the extracts causing possible toxicity. [18,19] Moreover, there have been no comprehensive studies reporting the biological effects on the skin including antiaging properties throughout inhibitory properties on oxidation reaction and antibacterial activity, along with human skin irritation testing of T. erecta flower extract. Therefore, this study focused on the simultaneous biological effects on the skin and skin irritation studies of T. erecta flower extracts by using biocompatible and low toxicity solvents. Thereby, water, 95% ethanol and 50% ethanol were selected for the plant extraction. Syringic acid, a phenolic compound, was selected as a chemical marker in this study due to it having been reported to have been antiaging activity in previous studies. [3] However, the quantity of syringic acid in T. erecta flower extracts has not been reported. Considering the biological activity of syringic acid, it has   [34] The ethanolic extract from T. erecta showed antibacterial activity against B. cereus, S. aureus, S. epidermidis, S. typhi, Klebsiella pneumoniae, E. coli, P. aeruginosa and P. mirabilis. [35] A mode of action in which the extract caused deformation and lysis of S. aureus cells was also reported. [35] These results indicated the potential effect of T. erecta especially 95% ethanolic extract on bacterial growth inhibition. There is report showing that aged skin is found to be a susceptible skin infection. In neonates, adipose tissue in adipocyte promotes innate immunity via production of antimicrobial peptide (cathelicidin). Ages skin is found the thinning of skin adipose tissue resulting in decrease of cathelicidin followed by impairment of S. aureus clearance and loss of antimicrobial function. [36] Therefore, antibacterial activities of T. erecta flower extracts may be useful for aged skin.
According to the results of antioxidant activity and the biological activities of T. erecta flower extracts, 95% ethanolic extract which contained higher amounts of total phenolic and syringic acid and possessed stronger radical scavenging activities, inhibitory effects against skin aging-related enzymes and against bacterial growth, had a potential to be a natural source of bioactive compounds for development of cosmeceutical skin products. Therefore, 95% ethanolic extract was selected to formulate as a prototype product in order to study the stability of the extract when it is in the formulation as well as skin irritation after application on human skin. The content of 95% ethanolic extract at 1% was preliminarily selected for adding to this prototype product. The results from stability studies revealed that this formulation was stable. In addition, the formulation showed no skin irritation in healthy volunteers after it was applied for 48 h. This study is the first study showing both the biological activities on the skin and the skin safety of T. erecta flower extracts. The results suggest potential of T. erecta flower extracts to be a source of bioactive compounds from a natural source for developing as a skin care product.

CONCLUSION
This study revealed that among three T. erecta flower extracts, ethanolic extract using 95% ethanol was found to contain higher phenolic compounds and syringic acid than the others. It also had greater biological effects on the skin including antioxidant, anti-skin aging-related enzymes and antibacterial activity than the other extracts. Additionally, the lotion containing the ethanolic T. erecta flower extracts was physicochemical stable and was also safe for human skin. These findings highlight the potential of T. erecta flower extracted by 95% ethanol to be utilized in the skin care products as a natural source of phytochemical which it offers excellent biological activities and safe.