Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul

Classification of some and Alpinia extracts and their medicinal products based on chemical composition, antioxidant activity, and concentration of some heavy metals.

Journal: Songklanakarin Journal of Science and Technology ManuscriptFor ID SJST-2019-0229.R1 Review Only Manuscript Type: Original Article

Date Submitted by the 17-Oct-2019 Author:

Complete List of Authors: Thummajitsakul, Sirikul ; Srinakharinwirot University Ongkharak Campus, Department of health promotion, Faculty of Physical Therapy Silprasit, Kun; Srinakharinwirot University, Faculty of Environmental Culture and Ecotourism

Boesenbergia and Alpinia extracts, Massage products, Radical Keyword: scavenging activity, FTIR, Heavy metals

For Proof Read only Page 3 of 36 Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul

1 2 3 1 Classification of some Boesenbergia and Alpinia extracts and their medicinal products 4 5 6 2 based on chemical composition, antioxidant activity, and concentration of some heavy 7 8 3 metals. 9 10 4 Sirikul Thummajitsakul1*, Kun Silprasit2 11 12 13 5 14 15 6 1 Faculty of Physical Therapy, Srinakharinwirot University, Ongkharak, Nakhon-Nayok, 16 17 7 Thailand 18 19 8 2 Faculty of EnvironmentalFor Culture Review and Ecotourism, Srinakharinwirot Only University, Bangkok, 20 21 22 9 Thailand 23 24 10 25 26 11 * Corresponding author. 27 28 29 12 Thummajitsakul, S. 30 31 13 1Faculty of Physical Therapy, Srinakharinwirot University, Ongkharak, Nakhon-Nayok 26120 32 33 14 Thailand 34 35 E-mail: [email protected] Tel (662) 649-5000 ext 27319 36 15 37 38 16 Abstract 39 40 41 17 The result showed total phenolic content, total flavonoid content and radical scavenging activity 42 43 18 in ethanol extracts higher than oil extracts and massage products of Boesenbergia rotunda, 44 45 19 Alpinia conchigera, Alpinia , and Alpinia siamensis. Fourier-transform infrared 46 47 -1 48 20 spectroscopy of the ethanol extracts provided specific peaks in a range of 4000 to 550 cm . 49 50 21 Chemical differentiation between Alpinia sp. and Boesenbergia sp. was indicated by main peaks 51 52 22 that corresponded to aromatic and N-H in amino acids, O-H in phenyl group, and C-O in 53 54 55 23 carbohydrates and glycoprotein. Chemical differentiation among three Alpinia species were 56 57 58 59 60 For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul Page 4 of 36

1 2 3 24 wavenumber ranges of C-O in acid or ester, C-O in carbohydrates and glycoprotein, and C-H in 4 5 6 25 isoprenoids. The cluster and PCA analysis showed good separation of chemical compositions 7 8 26 and antioxidant activity among the ethanol extract, massage solution and oil extract. 9 10 27 Additionally, Ni, and Cd were found in all massage products, especially Ni above its maximum 11 12 permission level. 13 28 14 15 29 16 17 30 Keywords: Boesenbergia and Alpinia extracts, Massage products, Radical scavenging activity, 18 19 31 FTIR, Heavy metals For Review Only 20 21 22 32 23 24 25 33 Introduction 26 27 28 34 Medicinal in the family are widely throughout tropical and 29 30 35 subtropical areas, and are abundant sources of antioxidants, nutrients and fiber that are often used 31 32 33 36 as local food and folk medicine (Rachkeeree et al., 2018). Specially, their are well 34 35 37 known as major ingredient in local food, , medicines and cosmetic due to their high 36 37 38 therapeutic and nutritional values (Saensouk, Saensouk, Pasorn, & Chanshotikul, 2018). 38 39 40 39 Nowadays, healthy food and organic medicinal products have become worldwide increasingly 41 42 40 popular. Several bioactive agents (i.e. flavonoids) can act as antioxidants that are linked to 43 44 41 prevent several diseases, such as anti-oxidative activity and inflammation (Panche, Diwan, & 45 46 47 42 Chandra, 2016). 48 49 43 In Thailand, the Zingiberaceae family have been classified into 30 genera and 300 species 50 51 44 (Larsen & Larsen, 2006). Of these, Boesenbergia rotunda is one of local edible that there 52 53 45 are several medicinal properties (i.e. antioxidant activity and antibacterial activity) and used as 54 55 56 46 natural ingredient in Thai food (Ongwisespaiboon & Jiraungkoorskul, 2017). Moreover, other 57 58 59 60 For Proof Read only Page 5 of 36 Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul

1 2 3 47 edible herbs in Zingiberaceae family, namely Alpinia conchigera, and Alpinia 4 5 6 48 siamensis, are also commonly used as food and spices (Saensouk et al., 2018). Alpinia species 7 8 49 contain several bioactive agents (i.e. flavonoids, terpenoids and kavalactones) with antioxidant 9 10 50 activities involving arterial hypertension and inflammation (Victório, 2011). 11 12 Nowadays, trend of consumer awareness and concern involving safety and quality of 13 51 14 15 52 foods and cosmetics is growing that lead to increase consumer demand for natural foods and 16 17 53 products. To reduce side effects from synthetic chemicals, native is become popular and 18 19 54 applied as medicinal productsFor due Review to its high medical Only value, safety, friendly to human, and 20 21 22 55 cheap. Massage ointment is one of herbal medicinal products that is used as biologically based 23 24 56 therapy and recognized as complementary and alternative medicine. 25 26 57 However, massage products consist of several components (i.e. vegetable oil, essential 27 28 29 58 oil, and phytochemicals) that may show neither synergistic or antagonistic interactions, including 30 31 59 substance (i.e. solvent extract and oil) and artificial ingredients may be contaminated with 32 33 60 some toxic elements, such as pesticides and heavy metals that may effect on skin or health of 34 35 consumers (Andersen, Holmberg, Larsen, Søborg, & Cohr, 2006). Nowadays, heavy metal 36 61 37 38 62 contamination and biological properties of massage products from extracts of B. rotunda, A. 39 40 63 conchigera, A. galanga and A. siamensis is still unknown. Specially, phytochemicals and 41 42 64 medical properties of A. siamensis was not well known. Combination of fourier transform 43 44 45 65 infrared spectra (FTIR) fingerprint and bioactivity assessment of this plant extracts is novel 46 47 66 knowledge and big challenge for future health care, such as development of medicinal products. 48 49 67 Therefore, our major objective was to characterize ethanol extracts, oil extracts, and 50 51 52 68 massage products of B. rotunda, A. conchigera, A. galanga and A. siamensis by total phenlic 53 54 69 contents, total flavonoid contents, antioxidant activity, and fourier-transform infrared (FTIR) 55 56 57 58 59 60 For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul Page 6 of 36

1 2 3 70 spectra, and to monitor heavy metal contamination in massage products from the local herbs. 4 5 6 71 These data was useful for improvement of their pharmaceutical activities, safety and quality as 7 8 72 healthy natural products for consumers, especially children and women. 9 10 73 Method and materials 11 12 Sample collection 13 74 14 15 75 Boesenbergia rotunda, Alpinia conchigera, Alpinia galanga and Alpinia siamensis were 16 17 76 collected from an organic family farm in Nong Saeng sub-district, Pakphli district, Nakhon 18 19 77 Nayok province (latitude For14.197479 Review and longitude 101.303588). Only The external morphology of 20 21 22 78 each plant sample was then identified by comparing with the samples from BGO plant databases 23 24 79 of The Botanical Garden Organization (2011). Some specimens were kept at Faculty of 25 26 80 Environmental Culture and Ecotourism, Srinakharinwirot University. After that, each samples 27 28 29 81 was cleaned, and left for 24 hours at room temperature, followed by cutting and homogenizing 30 31 82 into small pieces. 32 33 83 Sample extraction by coconut oil 34 35 Each ground sample was fried in coconut oil on 1:1 ratio at 80 oC for 1 hour, and left it 36 84 37 38 85 cool at room temperature, then filtrated through a fine sieve, and kept at 4 oC. Each sample was 39 40 86 processed in duplicate. 41 42 87 Sample extraction by solvent extraction 43 44 45 88 Each ground sample was mixed with 95% ethanol solvent in 1:2 ratio, and incubated at 46 47 89 37 oC for 16 hours. After that, the solvent was evaporated by a vacuum evaporator (IKAa RV10) 48 49 90 for 2-3 hours, and kept at -20 oC (Thummajitsakul, Kaewsri, & Deetae, 2016). 50 51 52 91 53 54 92 55 56 57 58 59 60 For Proof Read only Page 7 of 36 Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul

1 2 3 93 Production of massage solution and ointment 4 5 6 94 Massage ointment is a medicinal product that consists of borneol, menthol, camphor, 7 8 95 basic oil, petroleum jelly and wax, but massage solution is a form without petroleum jelly and 9 10 96 wax and suitable for determining biological activities. A stock solution was prepared by mixing 11 12 camphor, borneol and menthol in the ratio of 5: 4.5: 1.25, and melted at approximately 80 oC 13 97 14 15 98 temperature. After letting it to cool down, oil extract (20 ml) from each plant sample was added 16 17 99 and stirred, and poured into a glass bottle. For 95% ethanol extract, each sample extract was 18 19 100 used for 2.5 ml, and coconutFor oil for Review 17.5 ml was added Onlyin the mixed solution. Coconut oil was 20 21 22 101 used as basic oil or non-volatile oils. 23 24 102 For ointment preparation, petroleum jelly and wax were mixed in the ratio of 15.6: 3.4, 25 26 103 and melted at approximately 90 oC temperature. After that, the stock solution was added, and left 27 28 29 104 to cool down. Each extract was added, stirred, poured into a glass bottle, and left at room 30 31 105 temperature. 32 33 106 Total phenolic contents 34 35 The Folin–Ciocalteu colorimetric technique was used to estimate total phenolic contents 36 107 37 38 108 (Deetae, Parichanon, Trakunleewatthana, Chanseetis, & Lertsiri, 2012; Thummajitsakul et al., 39 40 109 2016). Each sample (300 µl) was reacted with Folin-Ciocalteu reagent (1.5 ml) at room 41 42 110 temperature for 5 min, and then reacted with sodium carbonate (7.5% w/v) (1.2 ml) for 30 min at 43 44 45 111 room temperature, followed by measuring the absorbance at 765 nm using a spectrophotometer 46 47 112 (Model T60UV). Each reaction was performed in duplicate. Gallic acid at concentration 0 to 200 48 49 113 µg/ml was used as positive control to produce a calibration curve (y=0.0114x-0.1699, R2 = 50 51 52 114 0.9518). The total phenolic content was measured from the calibration curve and expressed as 53 54 55 56 57 58 59 60 For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul Page 8 of 36

1 2 3 115 mg gallic acid per g extract for the ethanol extract, mg gallic acid per g fresh weight for the oil 4 5 6 116 extract, or mg gallic acid per ml solution for the massage solution. 7 8 117 Total flavonoid contents 9 10 118 Total flavonoid content was measured using aluminium chloride colorimetric method 11 12 (Chang, Yang, Wen, & Chern, 2002). Rutin was prepared in 80% ethanol solvent at 13 119 14 15 120 concentration 25, 50 and 100 µg/ml and was used to generate a calibration curve (y=0.0019x- 16 17 121 0.0078, R2 = 0.9918). Each sample (500 µl) was reacted with 95% ethanol (1.5 ml), 10% 18 19 122 aluminum chloride (0.1 ml),For 1M potassiumReview acetate (0.1 Only ml), and distilled water (2.8 ml). After 20 21 22 123 that, each reaction was incubated for 30 min at room temperature, and absorbance at 415 nm was 23 24 124 determined using a spectrophotometer (Model T60UV). Distilled water was used as blank. Each 25 26 125 sample was performed in duplicate, and the total flavonoid content was measured from the 27 28 29 126 calibration curve and expressed as mg rutin equivalent per g extract for the ethanol extract, mg 30 31 127 rutin equivalent per g fresh weight for the oil extract, or mg rutin equivalent per ml solution for 32 33 128 the massage solution. 34 35 Antioxidant activity 36 129 37 38 130 Antioxidant activity was measured by ABTS [2,2′-azinobis(3-ethylbenzothiazoline-6- 39 40 131 sulfonic acid) diammonium] method (Deetae et al. 2012). Briefly, ABTS radical solution was 41 42 132 prepared from 7 mM ABTS solution (10 ml) and 140 mM potassium persulfate (179 µl) by 43 44 45 133 incubation under dark at room temperature for 12-16 hours. The ABTS radical solution (3.9 ml) 46 47 134 with absorbance 0.700±0.005 at 734 nm was reacted with each extract or massage solution (20 48 49 135 µl) in the dark at room temperature for 6 min, and the absorbance at 734 nm was measured. 50 51 52 136 Each reaction was carried out in duplicate. The percentage of antioxidant capacity was calculated 53 54 137 according to Deetae et al. (2012) and Thummajitsakul et al., (2016). 55 56 57 58 59 60 For Proof Read only Page 9 of 36 Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul

1 2 3 138 The radical scavenging capacity was demonstrated as 50% effective concentration (EC50) 4 5 6 139 that was sample concentration needed to reduce ABTS radicals by 50% and was calculated from 7 8 140 each simple linear regression (R2 =0.97-1.00). 9 10 141 Fourier-transform infrared spectroscopy (FTIR) 11 12 The powdered sample (5 mg) from each ethanol extract of the plant samples was directly 13 142 14 15 143 placed on the center of the crystal plate in FTIR spectroscope (Spectrum Two™, Perkin Elmer, 16 17 144 USA). Each samples was pressed using an identical mechanical pressure to provide FTIR spectra 18 19 For Review Only-1 -1 20 145 in a range of 550 to 4000 cm with a resolution of 4 cm . Duplicate sample was run for five 21 22 146 times. Each FTIR spectra was analyzed with PerkinElmer spectrum IR version 10.6.0, and 23 24 147 compared with the reports of Coates (2006), Caunii, Pribac, Grozea, Gaitin, and Samfira (2012), 25 26 27 148 Cao, Wang, Shang, and Zhao (2017), and Topalăa, Tătarua, and Ducu (2017). 28 29 149 Determination of heavy metals 30 31 150 Each massage ointment (0.5 g) or massage solution (15 ml) were digested with 65% 32 33 o 34 151 HNO3 (10 ml) at 100 C until completely digestion, then added 1% HNO3, and filtrated by 35 36 152 Whatman No. 1 paper. The volume of each digested sample was then adjusted with 1% HNO3 to 37 38 153 50 ml. Each digestion was carried out in duplicate. Heavy metals namely Cu, Pb, Cd, Ni, Hg, Mn 39 40 154 and Zn in the digested samples were determined using an atomic absorption spectrometry 41 42 43 155 (Model 200 Series AA, Agilent Technologies, USA). External standard method was used to 44 45 156 generate a standard curve (R2 = 1), which was used to compare concentration of each heavy 46 47 157 metal (Thummajitsakul et al., 2018). 48 49 50 158 Statistical analysis 51 52 159 Descriptive statistics (i.e. mean, SD and percentage) were used to express heavy metal 53 54 160 concentration, total phenolic content, total flavonoid content, and antioxidant activity by using 55 56 57 58 59 60 For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul Page 10 of 36

1 2 3 161 PSPP program version 0.10.5 (Pfaff et al., 2013). In addition, the principal component analysis 4 5 6 162 (PCA) and unweighted pair group method (UPGMA) were performed by Paleontological 7 8 163 statistic program version 3.16 (Hammer, Harper, & Ryan, 2001). 9 10 164 Results and discussion 11 12 In our result, total phenolic contents in the ethanol extracts and oil extracts were found in 13 165 14 15 166 order of A. galanga > A. siamensis > A. conchigera > B. rotunda ranging from 23.62 to 31.43 16 17 167 mg gallic/ g extract and from 2.39 to 2.74 mg gallic/g fresh weight, respectively. Additionally, 18 19 168 total flavonoid contents inFor the ethanol Review extracts were found Only in order of A. galanga > B. rotunda > 20 21 22 169 A. conchigera > A. siamensis ranging from 19.36 to 77.62 mg rutin equivalent per g extract. 23 24 170 However, the order of total flavonoid content in oil extracts was B. rotunda > A. galanga > A. 25 26 171 conchigera > A. siamensis ranging from 1.73 to 2.75 mg rutin equivalent per g fresh weight. 27 28 29 172 Moreover, radical scavenging activities were found in the ethanol extracts of A. galanga, A. 30 31 173 siamensis, A. conchigera, and B. rotunda that were expressed as 1/EC50 values for 108.70, 32 33 174 96.15, 81.97 and 39.84, respectively. However, the order of radical scavenging activities in the 34 35 oil extracts were B. rotunda > A. conchigera > A. galanga > A. siamensis ranging from 0.12 to 36 175 37 38 176 0.35 (Table 1). 39 40 177 Moreover, the result showed that the order of total phenolic content in the massage 41 42 178 solution from the ethanol extract was A. siamensis> A. conchigera > A. galanga > B. rotunda in 43 44 45 179 ranging of 0.0541 to 0.0657 mg gallic/ ml solution, while the order of total phenolic content of 46 47 180 massage solution from oil extract was A. conchigera > A. siamensis > A. galanga > B. rotunda in 48 49 181 ranging of 0.0711 to 0.854 mg gallic/ ml solution (Table 2). 50 51 52 182 Additionally, our result showed that total flavonoid contents in the massage solution from 53 54 183 ethanol extract of A. conchigera, A. galanga, A. siamensis and B. rotunda were in ranging of 55 56 57 58 59 60 For Proof Read only Page 11 of 36 Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul

1 2 3 184 0.29 to 0.81 mg rutin equivalent per ml solution, while the order of total flavonoid contents of 4 5 6 185 massage solution from oil extract was A. galanga> B. rotunda> A. siamensis > A. conchigera 7 8 186 in ranging of 0.02 to 0.20 mg rutin equivalent per ml solution. However, the radical scavenging 9 10 187 activities were not found in the massage solutions. It is possible that the massage solution 11 12 consists of components that are not associated with radical scavenging activity, or its solute and 13 188 14 15 189 solvent capacities in solubility effect on radical scavenging activity. Complicated solvation and 16 17 190 surface characteristics in a multiphase system may disturb ingredient interaction and antioxidant 18 19 191 capacity (Mcclements & ForDecker, 2000),Review such as antioxidant Only partition among hydrophobic phase, 20 21 22 192 hydrophilic phase and the interfacial region (Yin, Becker, Andersen, Leif, & Skibsted, 2012). 23 24 193 Consequently, FTIR tool was applied in this study to classify A. siamensis, A. 25 26 194 conchigera, A. galanga, and B. rotunda. The FTIR spectra of their ethanol extracts was 27 28 -1 29 195 performed by a wavenumber range of 4000-550 cm (Figure 1). Most peaks corresponded to the 30 31 196 absorbance of C-O, C-H, N-H, C=O and O-H bands, respectively (Table 3) that indicated the 32 33 197 presence of carbohydrates, isoprenoids, phenyl groups, amino acids, fatty acids, ester, alcohols 34 35 and phenols. The FTIR spectra showed some peaks that specific to A. conchigera, A. siamensis, 36 198 37 38 199 A. galanga, and B. rotunda at different wavenumbers that confirmed the presence of different 39 40 200 chemical components in the ethanol extracts (Figure 1). 41 42 201 Additionally, cluster analysis from data of total phenolic content, total flavonoid content, 43 44 45 202 and radical scavenging activity provided good separation among ethanol extract, massage 46 47 203 solution, and oil extract of the studied plants (Figure 2A). The phylogenetic tree classified them 48 49 204 into two clusters that the first cluster was ethanol extracts, and the second cluster consisted of 50 51 52 205 massage solution and oil extracts of each plant species. The phylogenetic tree of A. siamensis, 53 54 206 A. conchigera, and A. galanga showed close relationship in sub-group of the first cluster, 55 56 57 58 59 60 For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul Page 12 of 36

1 2 3 207 specially between A. siamensis and A. conchigera. Corresponding to the result of the FTIR 4 5 6 208 spectra, cluster analysis was also able to recognize the studied plants into two clusters (Figure 7 8 209 2B). The first group consisted of sub-clusters namely A. siamensis, A. conchigera and A. 9 10 210 galanga, and another group consisted of B. rotunda. 11 12 As the result, the composition of ethanol extracts, massage solution, and oil extracts of 13 211 14 15 212 the studied plants was also obvious from PCA analysis. The values of total phenolic content, 16 17 213 total flavonoid content, and radical scavenging activity represented good segregation among 18 19 214 ethanol extracts, massageFor solution, Reviewand oil extracts of theOnly studied plants, and among ethanol 20 21 22 215 extracts of plant species (Figure 3A). Additionally, the PCA result showed that ethanol extracts 23 24 216 of B. rotunda, A. conchigera, A. siamensis, and A. galanga showed more total phenolic content, 25 26 217 total flavonoid content, and radical scavenging activity than those of massage solution and oil 27 28 29 218 extracts (Figure 3A). 30 31 219 Similarly, PCA analysis from absorption values of the FTIR spectra show also good 32 33 34 220 separation among B. rotunda, A. conchigera, A. siamensis, and A. galanga (Figure 3B). The 35 36 221 major peaks discriminating of the ethanol extracts of the three Alpinia sp. from B. rotunda were 37 38 222 vibration 5 (aromatic and N-H in amino acids), vibration 6 (primary or secondary, O-H in-plane 39 40 41 223 bend) and phenol or tertiary alcohol, O-H bend), and vibration 8 (C-O in mono–, oligo– 42 43 224 carbohydrates and glycoprotein). However, the three Alpinia sp. was differentiated by vibration 44 45 225 7 (C-O in acid or ester), 8 (C-O in mono–, oligo– carbohydrates and glycoprotein), and 9 (C-H in 46 47 48 226 isoprenoids). Therefore, the cluster and PCA analysis can help to discriminate types of the plant 49 50 227 samples and their massage products based on data of bioactive activity and the FTIR absorbance. 51 52 228 Interestingly, combination of fourier transform infrared spectra (FTIR) and bioactivity analysis 53 54 229 indicated that A. siamensis had FTIR fingerprint, bioactive contents and biological activity 55 56 57 58 59 60 For Proof Read only Page 13 of 36 Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul

1 2 3 230 similarly to A. conchigera. Corresponding to previous studies, A. siamensis has close 4 5 6 231 relationship with A. conchigera, indicated by a key to the Zingiberaceae genus in Thailand 7 8 232 (Larsen, 1980). 9 10 233 The FTIR technique have been used for classifying several plants, such as Medicago 11 12 sativa and Cucumis sp. (Caunii et al., 2012; Kukula-Koch, Grzybek, Strachecka, Jaworska, & 13 234 14 15 235 Ludwiczuk, 2018). In addition, it has been reported that this technique was used for quality 16 17 236 evaluation and identification of several plants from various locations and plant products such as 18 19 237 Rhodobryum roseum Limpr.For and edible Review vegetable oils (CaoOnly et al., 2017; Jiménez-Carvelo, 20 21 22 238 Osorio, Koidis, González-Casado, & Cuadros-Rodríguez, 2017). 23 24 239 Several studies have been reported that ethanol extract from B. rotunda consists 25 26 240 of three flavanones namely 2’,4’-dihydroxy-6-methoxychalcone, 5-hydroxy-7-methoxyflavanone 27 28 29 241 and 5, 7-dihydroxyflavanone, which relate to antioxidant and antibacterial activities (Atun, 30 31 242 Handayani, & Rakhmawati, 2018). Similarly, the rhizome of A. conchigera has been used as 32 33 243 folk medicine in treatment of fungal or skin infection (Ibrahim et al., 2000). Moreover, its 34 35 extract and bioactive agents (i.e. caryophyllene oxide, chavicol acetate, and 1'S-1'- 36 244 37 38 245 acetoxychavicol acetate) show antimicrobial, anticandidal and antidermatophyte activities (Aziz 39 40 246 et al., 2013). For A. galanga, it possess several bioactive agents that are many useful in treatment 41 42 247 of several diseases, such as antifungal activity and antioxidant activity (Wong, Lim, & Omar, 43 44 45 248 2009). 46 47 249 Although the local herbs have several effective biological properties, the safety concern 48 49 250 of the pharmaceutical products must also be considered. Thus, monitoring of heavy metal 50 51 52 251 contamination in massage products from the local herbs was also carried out in this study. 53 54 252 Recently, several reports show that Cd, Ni and Mn are commonly found in several cosmetics, 55 56 57 58 59 60 For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul Page 14 of 36

1 2 3 253 such as facial cosmetics, moisturizing creams, and skin-lightening creams (Iwegbue et al., 2015; 4 5 6 254 Iwegbue, Bassey, Obi, Tesi, & Martincigh, 2016). In our current study, the seven heavy metals 7 8 255 namely Cu, Pb, Cd, Ni, Hg, Mn and Zn were investigated in all massage products. The result 9 10 256 showed that Cd and Ni were found in all massage products, while Mn was only found for 11 12 2.68±2.27 mg/kg in massage solution from oil extract of A. conchigera (Table 4). Cd was found 13 257 14 15 258 in the massage ointment from oil extract of each plant sample in a range of 2.10±1.06 to 16 17 259 3.02±0.79 mg/kg higher than those of massage solution from 95% ethanol and oil extract 18 19 260 (ranged from 1.30±0.47 toFor 1.92 ± 0.58Review mg/kg). Similarly, Only Ni was found in the massage ointment 20 21 22 261 products from oil extract in a range of 23.14±3.72 to 26.83±4.78 mg/kg higher than those of the 23 24 262 massage solutions (ranged from 15.53±3.63 to 16.78±3.42 mg/kg). Interestingly, massage 25 26 263 ointment without a sample was also contaminated with Cd and Ni for 1.81±0.88 and 26.72±3.77, 27 28 29 264 respectively. 30 31 265 Although, Cd was found in all massage products, it was below its maximum permission 32 33 266 level in comparison with a 3 mg/kg criteria value (Health Canada, 2011), except for massage 34 35 ointment from oil extract of A. conchigera (3.02±0.79). The presence Cd in the samples can 36 267 37 38 268 effect on human body through skin and may lead to damage organ in the body, if cadmium blood 39 40 269 concentration is higher its maximum permissible value (Godt et al., 2006). More importantly, Ni 41 42 270 showed higher mean concentration in all massage products than its safe maximum permissible 43 44 45 271 limit (5 ppm) from the report of Basketter, Angelini, Ingber, Kern, and Menné (2003). It has 46 47 272 been reported that the contamination of Ni is cause of contact dermatitis (Torres, das Graças, 48 49 273 Melo, & Tosti, 2009) that its threshold values are 0.2 µg/cm2/week for direct and prolonged 50 51 2 52 274 contact with human skin, and 0.5 µg/cm /week for short-term contact (Menné et al., 1987; Ma'or 53 54 275 et al., 2015). Additionally, Mn may also effect on human body including brain, liver, and the 55 56 57 58 59 60 For Proof Read only Page 15 of 36 Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul

1 2 3 276 cardiovascular system if they are overexposed (O’Neal & Zheng, 2015), which Mn concentration 4 5 6 277 in human blood should be ranged from 4 to 15 µg/L for normal level (Agency for Toxic 7 8 278 Substances and Disease Registry, 2012). 9 10 279 However, Cd and Ni were higher found in each massage ointment with and without a 11 12 sample than each massage solution product. The massage ointment consisted of petroleum jelly 13 280 14 15 281 as an ointment base to protect moisture loss from skin, and paraffin wax to make soft-solid 16 17 282 texture of the ointment. Petroleum jelly is recognized as a byproduct of the oil from industry, and 18 19 283 commonly used in cosmeticsFor and pharmaceuticalReview products Only (i.e. skin cream, pain plam, lipsticks, 20 21 22 284 and skin ornaments), and it should not content heavy metal over 20 ppm (Unicorn, 2008). Heavy 23 24 285 metals in cosmetics that consist of petroleum jelly as major intergradient, such as lipsticks, 25 26 286 moisturizing creams, and skin-lightening creams, have been reported in several studies (Iwegbue 27 28 29 287 et al., 2015). Paraffin wax is a byproduct that is also generated from oil with saturated long-chain 30 31 288 hydrocarbons (C18 to C60), and is commonly used for several applications, such as cosmetics 32 33 289 and pharmaceuticals (Cottom, 2000). These ingredients from petroleum are frequently 34 35 contaminated with impurities that lead to skin irritation. Therefore, healthy natural ingredients 36 290 37 38 291 (i.e. plant butter and beeswax) are good alternative that should be used to replace petrolatum 39 40 292 ingredients in the pharmaceutical products. This finding can help to more understand of chemical 41 42 293 composition of the plants with radical scavenging activity, to evaluate quality of the 43 44 45 294 pharmaceutical products, to develop the pharmaceutical products, and to avoid heavy metal 46 47 295 hazard through dermal exposure to humans 48 49 296 50 51 52 297 53 54 298 55 56 57 58 59 60 For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul Page 16 of 36

1 2 3 299 Conclusions 4 5 6 300 In this study, the ethanol extract showed the highest total phenolic content, total flavonoid 7 8 301 content, and radical scavenging activity, while radical scavenging activities was not found in the 9 10 302 massage solution. Furthermore, the FTIR result showed specific chemical fingerprint that helped 11 12 to separate different types of plant samples namely B. rotunda, A. conchigera., A. siamensis, and 13 303 14 15 304 A. galanga. Cluster and PCA analysis from the FTIR data corresponded to the result from total 16 17 305 phenolic and flavonoid contents and biological activity of the ethanol extract. This result 18 19 306 confirmed that the FTIRFor method Review can predict components Only in the ethanol extract of the plant 20 21 22 307 samples. However, the result showed that Ni and Cd showed the highest concentration in 23 24 308 massage ointment without any sample extracts, followed by massage ointment from the oil 25 26 309 extracts, massage solution from the oil extracts, and massage solution from the ethanol extracts. 27 28 29 310 Importantly, Ni in all tested samples exceeded its maximum permission level. This research 30 31 311 therefore emphasize monitoring chemical compounds, biological activities, and heavy metals in 32 33 312 massage products commonly used in Thailand. 34 35 36 313 Acknowledgements 37 38 314 This research was financially supported by Srinakharinwirot University, Thailand (Grant 39 40 41 315 number 033/2562). The authors thank Faculty of Physical Therapy and Faculty of Environmental 42 43 316 Culture and Ecotourism, Srinakharinwirot University for research facilities and equipment. The 44 45 317 authors have no conflicts of interest in this research. 46 47 48 318 References 49 50 319 Agency for Toxic Substances and Disease Registry (2012). Toxicological profile for manganese. 51 52 320 Retrieved from http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=102&tid=23 53 54 55 321 Andersen, D. N., Holmberg, R. D., Larsen, J. R., Søborg, I., & Cohr, K. H.(2006). Survey and 56 57 58 59 60 For Proof Read only Page 17 of 36 Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul

1 2 3 322 health assessment of chemical substances in massage oils. Survey of Chemical 4 5 6 323 Substances in Consumer Products (No. 78). Retrieved from 7 8 324 https://www2.mst.dk/udgiv/publications/2006/87-7052-220-0/pdf/87-7052-221-9.pdf 9 10 325 Atun, S., Handayani, S., & Rakhmawati, A. (2018). Potential bioactive compounds isolated from 11 12 Boesenbergia rotunda as antioxidant and antimicrobial agents. Pharmacognosy Journal, 13 326 14 15 327 10(3), 513-518. doi:10.5530/pj.2018.3.84 16 17 328 Aziz, A. N., Ibrahim, H., Rosmy Syamsir, D., Mohtar, M., Vejayan, J., & Awang, K. (2013). 18 19 329 Antimicrobial compoundsFor Review from Alpinia conchigera Only. Journal of ethnopharmacology, 20 21 22 330 145(3), 798-802. doi: 10.1016/j.jep.2012.12.024 23 24 331 Basketter, D. A., Angelini, G., Ingber, A., Kern, P. S., & Menné, T. (2003). Nickel, chromium 25 26 332 and cobalt in consumer products: revisiting safe levels in the new millennium. Contact 27 28 29 333 Dermatitis, 49(1), 1–7. Retrieved from https://doi.org/10.1111/j.0105-1873.2003.00149.x 30 31 334 Cao, Z., Wang, Z., Shang, Z., & Zhao, J. (2017). Classification and identification of Rhodobryum 32 33 335 roseum Limpr. and its adulterants based on fourier-transform infrared spectroscopy 34 35 (FTIR) and chemometrics. PLoS ONE, 12(2), e0172359. Retrieved from 36 336 37 38 337 https://doi.org/10.1371/journal.pone.0172359 39 40 338 Caunii, A., Pribac, G., Grozea, I., Gaitin, D., & Samfira, I. (2012). Design of optimal solvent for 41 42 339 extraction of bioactive ingredients from six varieties of Medicago sativa. Chemistry 43 44 45 340 Central journal, 6, 123. doi: 10.1186/1752-153X-6-123 46 47 341 Chang, C., Yang, M., Wen, H., & Chern, J. (2002). Estimation of total flavonoid content in 48 49 342 propolis by two complementary colorimetric methods. Journal of Food and Drug 50 51 52 343 Analysis, 10, 178-182. Retrieved from 53 54 344 https://www.fda.gov.tw/en/publishjfdalistContent.aspx?id=27 55 56 57 58 59 60 For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul Page 18 of 36

1 2 3 345 Cottom, W. P. (2000). Waxes. Retrieved from 4 5 6 346 https://doi.org/10.1002/0471238961.2301240503152020.a01 7 8 347 Coates, J. (2006). Interpretation of infrared spectra, a practical approach. Retrieved from 9 10 348 https://doi.org/10.1002/9780470027318.a5606 11 12 Deetae, P., Parichanon, P., Trakunleewatthana, P., Chanseetis, C., & Lertsiri, S. (2012). 13 349 14 15 350 Antioxidant and anti-glycation properties of Thai herbal teas in comparison with 16 17 351 conventional teas. Food Chemistry, 133(3), 953–959. Retrieved from 18 19 352 https://doi.org/10.1016/j.foodchem.2012.02.012For Review Only 20 21 22 353 Environmental Defence Canada (2011). Heavy metal hazard—environmental defence: the health 23 354 \ risks of heavy metals in face makeup. Retrieved from 24 25 355 https://d36rd3gki5z3d3.cloudfront.net/wp-content/uploads/2016/01/HeavyMetalHazard- 26 27 356 FINAL.pdf?x84918. Accessed 2 April 2019. 28 29 357 Godt, J., Scheidig, F., Grosse-Siestrup, C., Esche, V., Brandenburg, P., Reich, A., & Groneberg, 30 31 358 D. A. (2006). The toxicity of cadmium and resulting hazards for human health. Journal of 32 33 359 Occupational Medicine and Toxicology, 1, 1-6. doi: 10.1186/1745-6673-1-22 34 35 Hammer, Ø., Harper, D. A. T., & Ryan, P. D. (2001). PAST: Paleontological statistics software 36 360 37 38 361 package for education and data analysis. Palaeontological Association. Retrieved from 39 40 362 http://palaeo-503 electronica.org/2001_1/past/issue1_01.htm 41 42 363 Ibrahim, H., Chooi, O. H., & Hassan, R. (2000). Ethnobotanical survey of the 43 44 45 364 family in selected Malay villages in Peninsular Malaysia. Malaysian Journal of Science, 46 47 365 19, 93–99. Retrieved from https://mjs.um.edu.my/index.php/MJS/article/view/9368 48 49 366 Iwegbue, C. M., Bassey, F. I., Tesi, G. O., Onyeloni, S. O., Obi, G., & Martincigh, B. S. (2015). 50 51 52 53 54 55 56 57 58 59 60 For Proof Read only Page 19 of 36 Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul

1 2 3 367 Safety evaluation of metal exposure from commonly used moisturizing and skin- 4 5 6 368 lightening creams in Nigeria. Regulatory Toxicology and Pharmacology, 71(3), 484-490. 7 8 369 doi: 10.1016/j.yrtph.2015.01.015 9 10 370 Iwegbue, C. M. A., Bassey, F. I., Obi, G., Tesi, G. O., & Martincigh, B. S. (2016). 11 12 Concentrations and exposure risks of some metals in facial cosmetics in Nigeria. 13 371 14 15 372 Toxicology Reports, 3, 464-472. Retrieved from 16 17 373 https://doi.org/10.1016/j.toxrep.2016.04.004 18 19 374 Jiménez-Carvelo, A. M., ForOsorio, M. Review T., Koidis, A., González-Casado, Only A., & Cuadros-Rodríguez, 20 21 22 375 L. (2017). Chemometric classification and quantification of olive oil in blends with any 23 24 376 edible vegetable oils using FTIR-ATR and Raman spectroscopy. LWT - Food Science 25 26 377 and Technology, 86, 174–184. Retrieved from https://doi.org/10.1016/j.lwt.2017.07.050 27 28 29 378 Kukula-Koch, W., Grzybek, M., Strachecka, A., Jaworska, A., & Ludwiczuk, A. (2018). ATR- 30 31 379 FTIR based fingerprinting of some Cucurbitaceae extracts: a preliminary study. Acta 32 33 380 Societatis Botanicorum Poloniae, 87(2), 1-6. Retrieved from 34 35 https://doi.org/10.5586/asbp.3579 36 381 37 38 382 Larsen, K. (1980). Annotated key to the genera of Zingiberaceae of Thailand. Natural History 39 40 383 Bulletin of the Siam Society, 28, 151-169. Retrieved from http://www.siam- 41 42 384 society.org/pub_NHB/nhbss_index_v21-30.html 43 44 45 385 Larsen, K., & Larsen, S. S. (2006). of Thailand. Chiang Mai, Thailand: Queen Sirikit 46 47 386 Botanic Garden. 48 49 387 Ma'or, Z., Halicz, L., Portugal-Cohen, M., Russo, M. Z., Robino, F., Vanhaecke, T., & Rogiers, 50 51 52 53 54 55 56 57 58 59 60 For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul Page 20 of 36

1 2 3 388 V. (2015). Safety evaluation of traces of nickel and chrome in cosmetics: the case of dead 4 5 6 389 sea mud. Regulatory Toxicology and Pharmacology, 73, 797-801. doi: 7 8 390 10.1016/j.yrtph.2015.10.016 9 10 391 Mcclements, D. J., & Decker, E. A. (2000). Lipid oxidation in oil-in-water emulsions: 11 12 impact of molecular environment on chemical reactions in heterogeneous food 13 392 14 15 393 systems. Journal of Food Science, 65(8), 1270–1282. Retrieved from 16 17 394 https://doi.org/10.1111/j.1365-2621.2000.tb10596.x 18 19 395 Menné, T., Brandup, F., Thestrup-Pedersen,For Review K., Veien, Only N. K., Andersen, J. R., Yding, F., & 20 21 22 396 Valeur, G. (1987). Patch test reactivity to nickel alloys. Contact dermatitis, 16(5), 255- 23 24 397 259. Retrieved from https://doi.org/10.1111/j.1600-0536.1987.tb01448.x 25 26 398 O’Neal, S. L., & Zheng, W. (2015). Manganese toxicity upon overexposure: a decade in review. 27 28 29 399 Current Environmental Health Reports, 2(3), 315–328. doi: 10.1007/s40572-015-0056-x 30 31 400 Ongwisespaiboon, O., & Jiraungkoorskul, W. (2017). Fingerroot, Boesenbergia rotunda and its 32 33 401 aphrodisiac activity. Pharmacognosy Reviews, 11(21), 27-30. doi: 34 35 10.4103/phrev.phrev_50_16. 36 402 37 38 403 Panche, A. N., Diwan, A. D., & Chandra, S. R. (2016). Flavonoids: an overview. Journal of 39 40 404 Nutritional Science, 5, 1-15. doi: 10.1017/jns.2016.41 41 42 405 Pfaff, B., Darrington, J., Stover, J., Satman, M. H., Beckmann, F., Williams, J., 43 44 45 406 Kiefte, M., Kobly, P., & van Son, R. (2013). GNU PSPP version. The 46 47 407 GNU Operating System. Retrieved from https://www.gnu.org/software/pspp/ 48 49 408 Rachkeeree, A., Kantadoung, K., Suksathan, R., Puangpradab, R., Page, P. A., & Sommano, 50 51 52 53 54 55 56 57 58 59 60 For Proof Read only Page 21 of 36 Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul

1 2 3 409 S.R. (2018). Nutritional compositions and phytochemical properties of the edible flowers 4 5 6 410 from selected Zingiberaceae found in Thailand. Frontiers in nutrition, 5(3), 1-10. doi: 7 8 411 10.3389/fnut.2018.00003 9 10 412 Saensouk, S., Saensouk, P., Pasorn, P., & Chanshotikul, N. (2018). Diversity and traditional 11 12 uses of Zingiberaceae in Nakhon Phanom province, Thailand. Research & Knowledge, 13 413 14 15 414 4(1), 47-55. Retrieved from 16 17 415 http://rms.msu.ac.th/report/show_research.php?action=show_data&data_id=6200 18 19 416 Thummajitsakul, S., Subsinsungnern,For Review R., Treerassapanich, Only N., Kunsanprasit, N., Puttirat, L., 20 21 22 417 Kroeksakul, P., & Silprasit, K. (2018). Pesticide contamination, heavy metal contents and 23 24 418 potential health risks of some vegetables from a local market and family farm in 25 26 419 Ongkharak district of Nakhon Nayok province . Pertanika Journal of Tropical 27 28 29 420 Agricultural Science, 41 (3), 987 – 1001. Retrieved from 30 31 421 http://www.pertanika.upm.edu.my/regular_issues.php?jtype=1&journal=JTAS-41-3-8 32 33 422 Topalăa, C. M., Tătarua, L. D., & Ducu, C.b. (2017). ATR-FTIR spectra fingerprinting of 34 35 medicinal herbs extracts prepared using microwave extraction. Arabian Journal of 36 423 37 38 424 Medicinal and Aromatic Plants, 3, 1-9. Retrieved from 39 40 425 https://revues.imist.ma/index.php?journal=AJMAP&page=article&op=view&path%5B% 41 42 426 5D=7985 43 44 45 427 Torres, F., das Graças, M., Melo, M.,& Tosti, A. (2009). Management of contact dermatitis due 46 47 428 to nickel allergy: an update. Clinical, Cosmetic and Investigational Dermatology, 2, 39– 48 49 429 48. Retrieved from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3047925/ 50 51 52 430 The Botanical Garden Organization. (2011). BGO plant databases. Retrieved from 53 54 431 http://www.qsbg.org/Database/BOTANIC_Book%20full%20option/ 55 56 57 58 59 60 For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul Page 22 of 36

1 2 3 432 Thummajitsakul, S., Kaewsri, W., & Deetae, P. (2016). Analysis of intraspecific genetic 4 5 6 433 variation, antioxidant and antibacterial activities of Zingiber zerumbet. International 7 8 434 Food Research Journal, 23(4), 1552-1557. Retrieved from 9 10 435 http://www.ifrj.upm.edu.my/volume-23-2016.html 11 12 Unicorn (2008). Petroleum jelly. Retrieved from https://asia.in- 13 436 14 15 437 cosmetics.com/__novadocuments/488833?v=636669133227100000 16 17 438 Victório, C. P. (2011). Therapeutic value of the genus Alpinia, Zingiberaceae. Brazilian Journal 18 19 439 of PharmacognosyFor, 21(1), 194-201.Review Retrieved fromOnly http://dx.doi.org/10.1590/S0102- 20 21 22 440 695X2011005000025 23 24 441 Wong, L. F., Lim, Y. Y., & Omar, M. (2009). Antioxidant and antimicrobial activities of some 25 26 442 Alpina species. Journal of Food Biochemistry, 33(6), 835-851. Retrieved from 27 28 29 443 https://doi.org/10.1111/j.1745-4514.2009.00258.x 30 31 444 Yin, J., Becker, E. M., Andersen, M. L., & Skibsted, L. H. (2012). Green tea extract as food 32 33 445 antioxidant. Synergism and antagonism with a-tocopherol in vegetable oils and their 34 35 colloidal systems. Food Chemistry, 135(4), 2195–2202. doi: 36 446 37 38 447 10.1016/j.foodchem.2012.07.025 39 40 448 41 42 449 43 44 450 45 451 46 47 452 48 49 453 50 51 454 52 455 53 54 456 55 56 457 57 58 59 60 For Proof Read only Page 23 of 36 Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul

1 2 3 458 4 5 459 6 7 460 FIGURE AND TABLES 8 9 461 Table 1. Total phenolic contents, total flavonoid contents, and antioxidant activities of 95% 10 11 462 ethanol extracts and oil extracts of B. rotunda, A. conchigera, A. galanga and A.siamensis. 12 13 14 463 Table 2. Total phenolic contents, total flavonoid contents, and antioxidant activities of massage 15 16 464 solution of B. rotunda, A. conchigera, A. galanga and A.siamensis. 17 18 465 Table 3. FTIR spectral peak values and functional groups of ethanol extracts of each sample 19 For Review Only 20 21 466 Table 4. The concentrations of heavy metals in massage products 22 23 467 Figure 1. The FTIR Spectra at wavenumber 4000-550 cm-1 of the ethanol extracts of B. 24 25 468 rotunda (BRE), A. conchigera (ACE), A. siamensis (ASE) and A. galanga (AGE). Specific 26 27 469 peaks of each sample were indicated at a wavenumber of 1736.81 - 1739.31 cm-1 (a), 1607.53- 28 29 -1 -1 -1 -1 30 470 1608.92 cm (b) , 1513.11-1513.85 cm (c), 1051.12-1052.68 cm (d), 1043.91-1043.78 cm (e), 31 32 471 819.18-819.15 cm-1(f), 1016.54-1016.75 cm-1 (g), 1621.97-1622.15 cm-1 (h), 1339.36-1339.35 33 34 472 cm-1 (i), 1303.03-1301.59 cm-1 (j), and 1030.62-1031.34 cm-1(l). 35 36 37 473 Figure 2. An unweighted pair group method (UPGMA) tree of B. rotunda (BR), A. conchigera 38 39 474 (AC), A. siamensis (AS) and A. galanga (AG). (A) was UPGMA tree of relationship among 40 41 475 ethanol extracts, massage solution from 95% ethanol extracts (BRES, ACES, ASES, and AGES), 42 43 44 476 massage solutions from oil extract (BROS, ACOS, ASOS, and AGOS), and oil extracts (BRO, 45 46 477 ACO, ASO, and AGO) based on total phenolic contents, total flavonoid contents, and 47 48 478 antioxidant activity. (B) was UPGMA tree of relationship among absorption values of the FTIR 49 50 479 spectra from ethanol extracts (BRE, ACE, ASES, and AGE) of the studied plants. 51 52 53 480 Figure 3. Principal component analysis (PCA). (A) PCA of total phenolic contents, total 54 55 481 flavaniod contents, and antioxidant activity of ethanol extracts, massage solution from ethanol 56 57 58 59 60 For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul Page 24 of 36

1 2 3 482 extracts, massage solutions from oil extract, and oil extracts. (B) PCA of absorption values of 4 5 6 483 the FTIR spectra of ethanol extracts. 7 8 9 10 11 12 13 14 15 16 17 18 19 For Review Only 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Proof Read only Page 25 of 36 Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 For Review Only 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul Page 26 of 36

1 2 3 485 4 5 6 7 8 9 10 11 12 13 14 For Review Only 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 For Proof Read only 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 27 of 36 Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul

1 2 3 4 5 6 Plant types Samples Total phenolic Total flavonoid Antioxidant activities 7 contents (mg contents (mg 8 EC50 1/EC 9 gallic/ g rutin 10 11 extract) equivalent/ g 12 13 extract) 14 15 B. rotunda 95% 23.62±0.92 74.87±3.30 0.0251±0.0009 39.84 16 17 ethanol 18 extract 19 20 21 Oil extractFor 2.39±0.36Reviewa 2.75±0.12Onlyb 2.8440±1.6866 0.35 22 23 A. conchigera 95% 24.73±4.23 48.36±1.49 0.0122±0.0006 81.97 24 25 ethanol 26 27 extract 28 29 Oil extract 2.51±0.22a 1.82±0.02b 3.6734±1.1078 0.27 30 31 32 A. galanga 95% 31.43±5.96 77.62±2.38 0.0092±0.0021 108.70 33 ethanol 34 35 extract 36 37 Oil extract 2.74±0.20a 2.21±0.10b 5.6292±4.3209 0.18 38 39 40 A.siamensis 95% 25.99±1.64 19.36±2.13 0.0104±0.0014 96.15 41 42 ethanol 43 extract 44 45 46 Oil extract 2.58±0.25a 1.73±0.08b 8.0980±1.7095 0.12 47 48 aTotal phenolic content was expressed as mg gallic/ g fresh weight 49 50 b Total flavonoid contents was expressed as mg rutin equivalent/ g fresh weight 51 52 53 Table 1. Total phenolic contents, total flavonoid contents, and antioxidant activities of 95% 54 55 ethanol extracts and oil extracts of B. rotunda, A. conchigera, A. galanga and A.siamensis. 56 57 58 59 60

For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul Page 28 of 36

1 2 3 Plant types Samples Total phenolic Total flavonoid contents 4 5 contents (mg gallic/ (mg rutin equivalent/ ml 6 7 ml solution) solution) 8 9 B. rotunda Massage 0.054±0.002 0.29±0.04 10 11 solution 11 12 13 Massage 0.071±0.018 0.15±0.04 14 15 solution 22 16 17 18 A. conchigera Massage 0.059±0.007 0.81±0.34 19 solution 11 20 21 For Review Only 22 Massage 0.085±0.012 0.02±0.01 23 solution 22 24 25 26 A. galanga Massage 0.058±0.003 0.72±0.13 27 1 28 solution 1 29 30 Massage 0.075±0.003 0.20±0.04 31 2 32 solution 2 33 34 A. siamensis Massage 0.066±0.014 0.70±0.49 35 36 solution 11 37 38 Massage 0.080±0.014 0.04±0.03 39 40 solution 22 41 42 43 Massage solution without sample3 - 1.92±0.02 44 45 46 1 Massage solution 1 was produced from 95% ethanol extract. 47 2 48 Massage solution 2 was produced from oil extract. 49 3 Massage solution was produced from coconut oil without sample. 50 51 52 53 54 55 56 57 58 59 60

For Proof Read only Page 29 of 36 Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul

1 2 3 Table 2. Total phenolic contents, total flavonoid contents, and antioxidant activities of 4 5 6 massage solution of B. rotunda, A. conchigera, A. galanga and A.siamensis. 7 8 9 10 Wavenumber range Wavenumber Assignment Function groups References 11 (detected in this range 12 studied, cm-1) (reference, cm- 13 1 14 ) 15 1 16 3281.7 - 3336.71 3000–3600 O-H and N–H water, alcohols, Caunii et al. (2012) stretch phenols, 17 3500-3000 Cao et al. (2017) 18 19 carbohydrates, 20 peroxides 21 For Review Only 22 2923.45 - 2926.352 2920 C-H stretch polysaccharides, Cao et al. (2017) 23 lipids, and 24 carbohydrates 25 26 2853.24 - 2854.33 2800-2900 C-H stretch polysaccharides, Cao et al. (2017) 27 lipids, and 28 2850 Topalăa et al. carbohydrates 29 (2017) 30 31 1607.53 - 1739.814 1600-1760 N-H bending amino acids, fatty Topalăa et al. 32 C=O acids, ester 33 vibrations, (2017) 34 bending vibrations 35 5 36 1513.11- 1513.69 1500–1600 aromatic and N-H amino acids Caunii et al. (2012) bending vibrations 37 Topalăa et al. 38 39 (2017) 40 6 41 1301.59-1443.99 1300-1450 Primary or phenyl groups Coates (2006) 42 secondary O-H Caunii et al. (2012) 43 bending (in-plane), 44 and phenol or 45 46 tertiary alcohol 47 (O-H bend) 48 1153.79-1230.987 1150-1270 C-O stretching acid or ester Caunii et al. (2012) 49 vibrations 50 Topalăa et al. 51 (2017)] 52 53 1016.54-1052.688 997–1130 C-O stretching mono–, oligo– Caunii et al. (2012) 54 vibrations carbohydrates, 55 997-1140 Topalăa et al. oligosaccharides, 56 (2017) 57 glycoprotein 58 59 570.75- 921.659 < 1000 C-H bending Isoprenoids Caunii et al. (2012) 60 Topalăa et al.

For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul Page 30 of 36

1 2 3 Wavenumber range Wavenumber Assignment Function groups References 4 5 (detected in this range 6 studied, cm-1) (reference, cm- 7 1) 8 9 vibrations (2017) 10 11 1-9 codes of wavenumber ranges. 12 13 14 15 Table 3. FTIR spectral peak values and functional groups of ethanol extracts of each sample 16 17 18 19 20 21 For Review Only 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For Proof Read only Page 31 of 36 Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul

1 2 3 4 5 6 7 Plant types Samples Heavy metal content (mg/kg for solid and mg/L for 8 9 liquid) 10 11 12 Cd Ni Mn 13 14 B. rotunda Massage solution 1.42±0.96 15.83±2.95 ND** 15 1 16 1 17 18 Massage solution 1.57±0.56 16.78±3.42 ND** 19 20 22 21 For Review Only 22 Massage ointment 2.62±0.90 25.01±3.79 ND** 23 24 from oil extract 25 26 A. conchigera Massage solution 1.59±0.35 15.68±2.29 ND** 27 28 11 29 30 31 Massage solution 1.60±0.46 15.68±2.68 2.68±2.27 32 22 33 34 35 Massage ointment 3.02±0.79 23.14±3.72 ND** 36 from oil extract 37 38 39 A. galanga Massage solution 1.92±0.58 16.05±2.06 ND** 40 1 41 1 42 43 Massage solution 1.30±0.47 15.53±3.63 ND** 44 2 45 2 46 47 Massage ointment 2.94±0.64 23.69±4.27 ND** 48 49 from oil extract 50 51 A. siamensis Massage solution 1.41±0.60 16.23±2.67 ND** 52 53 11 54 55 56 Massage solution 1.51±0.56 15.53±2.69 ND** 57 22 58 59 60

For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul Page 32 of 36

1 2 3 Massage ointment 2.10±1.06 26.83±4.78 ND** 4 5 from oil extract 6 7 Massage ointment without a sample 1.81±0.88 26.72±3.77 ND** 8 9 10 a b 11 *MPLcosmetics 3 5 NV*** 12 13 14 15 *MPL = Maximum permitted levels (mg/kg for solid samples and mg/L for liquid samples) 16 **ND = Not Detected 17 18 ***NV= No Value 19 1 20 Massage solution 1 wasFor produced Review from 95% ethanol Only extract. 21 2Massage solution 2 was produced from oil extract. 22 23 3 Massage solution was produced from coconut oil without sample. 24 25 a Environmental Defence Canada (2011) 26 b 27 Data from Basketter et al. (2003) 28 29 30 Table 4. The concentrations of heavy metals in massage products 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For Proof Read only Page 33 of 36 Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 For Review Only 20 21 22 23 24 1 25 2 Figure 1. The FTIR Spectra at wavenumber 4000-550 cm-1 of the ethanol extracts of B. 26 27 3 rotunda (BRE), A. conchigera (ACE), A. siamensis (ASE) and A. galanga (AGE). Specific 28 -1 29 4 peaks of each sample were indicated at a wavenumber of 1736.81 - 1739.31 cm (a), 1607.53- 30 5 1608.92 cm-1 (b) , 1513.11-1513.85 cm-1 (c), 1051.12-1052.68 cm-1 (d), 1043.91-1043.78 cm-1(e), 31 32 6 819.18-819.15 cm-1(f), 1016.54-1016.75 cm-1 (g), 1621.97-1622.15 cm-1 (h), 1339.36-1339.35 33 -1 -1 -1 34 7 cm (i), 1303.03-1301.59 cm (j), and 1030.62-1031.34 cm (l). 35 8 36 37 9 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 10 57 58 59 60 For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul Page 34 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 For Review Only 20 21 11 22 12 Figure 2. An unweighted pair group method (UPGMA) tree of B. rotunda (BR), A. conchigera 23 24 25 13 (AC), A. siamensis (AS) and A. galanga (AG). (A) was UPGMA tree of relationship among 26 27 14 ethanol extracts, massage solution from 95% ethanol extracts (BRES, ACES, ASES, and AGES), 28 29 15 massage solutions from oil extract (BROS, ACOS, ASOS, and AGOS), and oil extracts (BRO, 30 31 32 16 ACO, ASO, and AGO) based on total phenolic contents, total flavonoid contents, and 33 34 17 antioxidant activity. (B) was UPGMA tree of relationship among absorption values of the FTIR 35 36 18 spectra from ethanol extracts (BRE, ACE, ASES, and AGE) of the studied plants. 37 38 19 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Proof Read only Page 35 of 36 Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 For Review Only 20 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 21 40 22 41 42 23 Figure 3. Principal component analysis (PCA). (A) PCA of total phenolic contents, total 43 44 24 flavaniod contents, and antioxidant activity of ethanol extracts, massage solution from ethanol 45 46 25 extracts, massage solutions from oil extract, and oil extracts. (B) PCA of absorption values of 47 26 the FTIR spectra of ethanol extracts 48 49 50 51 52 53 54 55 56 57 58 59 60 For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0229.R1 Thummajitsakul Page 36 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 For Review Only 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For Proof Read only