Effects of solvent—solvent fractionation on the total terpenoid content and in vitro anti‐inflammatory activity of Serevenia buxifolia bark extract

Abstract Severinia buxifolia (Rutaceae) is often used as a traditional medical plant. The present study was carried out to estimate the effects of solvents (petroleum ether and hexane: ethyl acetate) used in liquid–liquid extraction to total terpenoid content (TTC) and in vitro anti‐inflammatory activity of the extracts obtained from S. buxifolia bark. The results showed that solvent fractionation increased the TTC compared with crude extracts. The hexane: ethyl acetate bark extract fraction (HEF) had the highest TTC (731.48 µg/ml) in comparison with the petroleum ether bark extract fraction (PEF) (564.81 µg/ml) and the crude extract (CE) (184.26 µg/ml). In addition, one of composition of terpenoid of S. buxifolia, namely ursolic acid, was determined by HPLC method from the crude CE and the fractions PEF and HEF: 2.44 μg/g DW, 3.56 μg/g DW and 5.04 μg/g DW, respectively. The samples had an in vitro anti‐inflammatory activity comparable with that of two reference standards (aspirin and indomethacin). Particularly, the HEF fraction had the highest in vitro anti‐inflammatory activity (i.e., albumin denaturation: IC50 = 147.91 μg/mL, heat‐induced hemolysis: IC50 = 159.91 μg/mL, proteinase inhibition: IC50 = 117.72 μg/mL, and lipoxygenase activity: IC50 = 90.45 μg/mL). Besides, the preliminary experiments of this study were conducted to determine the influences of maceration factors (solvent type, temperature, and time) for S. buxifolia bark extract. The TTC ranged from 453.70 to 842.59 mg linalool/g DW, and the extraction yield from 2.40% to 5.120% in all extracts. Based on TTC and EY, the hexane: acetone mixture is recommended as the optimal solvent to obtain the crude bark extract (CE) at 46°C for 24 hr of maceration. Extracts of S. buxifolia bark are a promising source for the treatment of inflammatory diseases.


| INTRODUC TI ON
Among the 20 species of the genus Atalantia (within the family Rutaceae) (Roskov et al., 2016), Severinia buxifolia (Rutaceae) or Atalantia buxifolia is one of the plants of most interest (Safaa et al., 2018). It is frequently used in traditional medicine for the treatment of cough, snakebites, malaria, chronic rheumatism, influenza, and pain (Chang et al., 2018;Truong et al., 2019;Yang et al., 2012). The health benefits of this species are associated with phytochemical components that play a role in human physiology (Wu et al., 2001). Terpenoids (tetraterpenoids and sesquiterpenoids), acridones, and coumarins are the most extensively studied phytochemicals from the branches and roots of this plant (Chang et al., 2018;Shi et al., 2014;Truong et al., 2019;Wu et al., 2001;Yang et al., 2012). For example, Shi et al. (2014) detected a new triterpenoid with an aoptirucallane skeleton in an ethanolic extract of A. buxifolia roots. In our previous study, we isolated the acridone alkaloid from S. buxifolia branches (Truong et al., 2019). The terpenoid group, mainly containing the volatile organic compounds (VOCs), was found as the main chemical components from the bark of S. buxifolia compositions (Vivaldo et al., 2017;Wu et al., 2001). For example, Wu et al. (2001) detected two new tertranortriterpenoids (7-isovaleroylcycloseverinolide (1) and 7-isovaleroylcycloepiatalantin (2)) from the root bark of S. buxifolia. Generally, terpenoids are an important class of bioactive phytochemicals, but to our knowledge, respective information is limited, especially regarding the influence of solvent fractionation on the content of the components.
In the food and pharmacy industry, separation techniques such as adsorption column chromatography, liquid-liquid extraction, solid-liquid extraction, membrane filtration, or gel filtration chromatography are often applied to obtain pure natural products from complex extracts (Phan et al., 2020;Zhang et al., 2018). Within these methods, various polarities of solvents were applied in the liquidliquid technique to extract the phytochemical classes from plants, representing a simple and cost-effective approach (Gharaati, 2019;Sivanandham, 2015). Generally, the selection of a two-phase solvent system is based on the polarity of components in plant extracts. For example, Canbay (2017) used chloroform, dichloromethane, n-hexane, and ethyl acetate to obtain target compounds (focusing on volatile compounds) from rose aromatic water. The results showed that n-hexane was the least suitable liquid-liquid extraction (LLE) solvent. Different solvents for LLE have also been investigated regarding their influence on terpenoid compounds from plant extracts (Pichersky & Raguso, 2018); these authors showed that dichloromethane was the highest optimum extraction solvent for the LLE of rose water terpenoids.
Inflammation is a protective physiological response of living tissues to injury (Chou, 1997;Nurtamin et al., 2018;Rajendran et al., 2013). The characteristics of inflammation are heat, swelling, redness, and pain (George, 2006;Prakash, 2017). The inflammatory response includes the induction of cytokine release, the activity of several enzymes (oxygenases, nitric oxide synthase, and peroxidases), and the expression of cellular adhesion molecules (Gomes et al., 2008;Prakash, 2017). Ruiz-Ruiz et al. (2017) observed that protein denaturation is also considered as a marker of inflammation. The importance of the inflammatory response is the stabilization of the lysosomal membrane (Chippada et al., 2011).
Thus, to evaluate in vitro anti-inflammatory of plant extracts, stabilization of the human red blood cell membrane (RBC) by a hypo tonicity-induced membrane can be used as one of the parameters.
Based on total terpenoid content (TTC) determination and in vitro anti-inflammatory activity (i.e., inhibition of protein (albumin) denaturation, heat-induced hemolysis, proteinase activity, and LOX assay), as well as investigation of the extracts obtained from S. buxifolia bark, we estimated the effects of solvents (petroleum ether and hexane: ethylacetate) used in LLE in comparison with those of the crude bark extract. TLC and HPLC were also applied to detect the presence of natural oleanolic acid (OA) and ursolic acid (UA) in the bark extracts as compared to pure compound (OA and UA, Sigma, Singapore). In addition, in the preliminary experiments, single-factor extraction was used to optimize the recovery of terpenoid content and extraction yield from S. buxifolia bark by studying the influence of four factors, namely solvent type, maceration temperature, and time.

| Plant collection and treatment
The bark was collected from S. buxifolia trees in Phu Loc district (Thua Thien Hue Province, Vietnam). The plants were taxonomically identified by the Botany Research and Development Group of Vietnam (Vietnam). Bark was obtained by removing the leaves, trunks, and thorns, then washed with distilled water to remove debris, cut into small pieces, and dried at 40°C for 10 days. The dried bark was then ground into powder using a mill (Jehmlich, Germany) and kept in a dry, airtight container for further usage.

| Influence of extraction parameters on total terpenoid content and extraction yield
Single-factor extraction was used to optimize the recovery of terpenoid compounds and extraction yield from S. buxifolia bark by studying the influence of four factors, namely extraction solvent, extraction temperature, and maceration time. First, 400 ml of extracting solvent was added to 20 g of S. buxifolia bark powder (1:20, w/v) in a 1,000-ml beaker. The mixture was then macerated in a water bath (LABEC, Marrickville, Australia) at the respective temperature and time. Subsequently, the mixture was homogenized at the respective temperature by constant shaking for 4 hr, using a homogenizer (IKA, Germany). The filtrate was removed from the residue by filtration using Whatman No. 1 filter paper. This process was repeated three times to exhaustively isolate the plant material, and the obtained extracts were combined. The extract solutions were concentrated using a rotary evaporator (Pollab, India) at 40°C and drying at room temperature. Dried extract samples were stored in an airtight container at 4°C. All experiments were performed in triplicate. The yield of S. buxifolia dried extract was calculated according to the following equation: where W 1 is the weight of the extract after evaporation of the solvent and W 2 is the dry weight of the bark sample.

| Maceration temperature
By using the optimal extraction solvent as determined in first step, samples were macerated at different temperatures, namely 6, 26, 46, 66, and 86°C by fixing the maceration time constant at 24 hr.
The best maceration temperature was selected based on the values of TTC and EY of extracts.

| Maceration time
By using the optimal extraction solvent and maceration temperature as determined in first step and second step, respectively, samples were macerated for 0, 12, 24, 36, and 48 hr. The best maceration time was selected according to the values of TTC and EY of extracts.

| Fractionation by liquid-liquid extraction
The crude extract of S. buxifolia bark was obtained at optimum extracting conditions (solvent, temperature, and time), determined in previous experiments (2.3). The crude bark extract (CE) was then dissolved in steriled distilled water in a separating funnel, equilibrated, and successively extracted with petroleum ether, hexane: ethyl acetate (85:15; v/v) (Sigma-Aldrich, Singapore) to obtain fractions of (1) Extraction yield (EY) (\% ) = ( W 1 × 100 ) ∕W 2 , F I G U R E 1 Protocol for the solvent-solvent fractionation of extract components of Severinia buxifolia bark various polarities. The fractionation protocol is shown in Figure 1; fractionation was performed in triplicate. The organic fractions (petroleum ether or hexane: ethyl acetate) were concentrated by a rotary evaporator at 40°C, followed by drying at room temperature.

| Determination of chemical components by TLC method
The TLC procedure described by Gambhava et al. (2013)  where S 1 is the distance travelled by the solute and S 2 is the distance travelled by solvent front on the TLC plates.

| High-performance liquid chromatography (HPLC) technique analysis
We used a high-performance liquid chromatography (HPLC-2160, Agilent, USA) equipped with a UV detector (Agilent Series 1,100) and an Eclipse Plus C18 column (4.6 × 250 mm) to identify bioactive terpenoid compounds in the extracts of S. buxifolia bark.
The program was set up as described by Zhang et al. (2013) with some modifications. The mobile phase was composed of acetonitrile (solvent A) and water containing 0.1% phosphoric acid (solvent B): 0-25 min at 22%-23% (solvent A) and 1.0-1.5 ml/min, 25-40 min at 23% (solvent A) and 1.5-1.0 ml/min, and 40-60 min at 23%-90% (solvent A) and 1.0 ml/min. The column temperature was maintained at 25°C; 20 µl of the sample dissolved in MeOH (20-100 ppm) was injected into the column, and detection was attained at 210 nm. Ursolic acid (Sigma, Singapore) at 0.5 mg/ml was applied as the reference standard. The peaks were identified on the basis of retention time (RT) in comparison with the RT of standard ursolic acid. The detected peaks from the extracts were quantified based on peak area.

| Determination of total terpenoid content
The total terpenoid content (TTC) of the extracts of S. buxifolia bark was determined by the method of Ghorai et al. (2012). To 1 ml of the extracts, we added 2 ml of chloroform. The sample mixture was then vortexed thoroughly before being left for 3 min.
Subsequently, 200 μl of concentrated sulfuric acid (H 2 SO 4 ) was poured into the mixture, followed by incubation at room temperature for 1.5-2 hr in the dark. A reddish-brown precipitate was formed in the mixture during incubation. After that, the supernatant was carefully decanted without disturbing the precipitation, and 3 ml of absolute methanol was added and vortexed well until complete dissolving of the precipitation in methanol. Absorbance was read at 538 nm using a visible spectrometer (V-730 UV-Vis Spectrophotometer, Jasco, USA). The TTC of the extracts was calculated as mg of linalool per gram of extract (dry weight, DW). The equation of the standard curve was y = 0.0036x − 0.001, where R 2 = 0.9927.

| Determination of in vitro antiinflammatory activity
The in vitro anti-inflammatory activity of the S. buxifolia extracts was determined via assessment of the inhibition of albumin denaturation, the membrane stabilization test (heat-induced hemolytic and protein inhibitory action), and antilipoxygenase (anti-LOX) activity, as described by previous studies with minor modifications (Eshwarappa et al., 2016;Govindappa et al., 2011;Leelaprakash & Mohan, 2011;Shaikh et al., 2018;Truong et al., 2019). The extracts of S. buxifolia bark (CE, PEF, and HEF) were serially diluted in dimethyl sulfoxide (DMSO) from 25 to 200 μg/ml. Aspirin and indomethacin (Sigma-Aldrich, Singapore), reference standard anti-inflammatory drugs, were used as positive controls.

| Inhibition of albumin denaturation
We followed the methods of Govindappa et al. (2011), with slight modifications. The reaction mixture consisted of the test extract (1 ml) and 1% aqueous solution of bovine albumin fraction (1 ml).
The pH of the reaction mixture was adjusted to 6.3 using 0.1 N HCl at 37°C. The sample extracts were incubated at 37°C for 20 min and then heated to 51°C for 20 min. After cooling to room temperature, turbidity was measured spectrophotometrically at 660 nm. The DMSO was used as control. Percent inhibition of albumin denaturation was calculated as follows: where A1 = absorption of the control and A2 = absorption of the test sample mixture.

Preparation of red blood cell (RBC) suspension
The RBC suspension was prepared according to the method described in Gunathilake et al. (2018) with some modifications. The blood sample was collected from a healthy human volunteer who had not used any nonsteroidal anti-inflammatory drugs for 2 weeks prior to the experiment. The blood cells were centrifuged at 1008 g for 10 min in heparinized centrifuge tubes and washed with an equal volume of normal saline (0.9% NaCl) (three times). After centrifugation, the volume of blood was measured and reconstituted as a 10% (v/v) suspension with normal saline.

Heat-induced hemolysis
This test was carried out as described by Gunathilake et al. (2018), with some modifications as described in Truong et al. (2019). Briefly, 1 ml of blood cell suspension was mixed with 1 ml of test extract of S. buxifolia bark; instead of the test sample, only saline was added to the control test tube. All centrifuge tubes containing reaction mixture (2 ml) were incubated in a shaking water bath at 56°C for 30 min. After incubation, the mixture was cooled down rapidly to room temperature and centrifuged at 700 g for 5 min to obtain the supernatant. The absorbance of the supernatant was measured at 560 nm using a spectrophotometer (V-730 UV-Vis Spectrophotometer, Jasco, USA). The level of hemolysis was calculated using the following equation: where A1 = absorption of the control and A2 = absorption of the test sample mixture.

Proteinase inhibitory activity
Proteinase inhibitory activity of the S. buxifolia bark extracts was determined according to the method of Leelaprakash and Mohan (2011), modified by Truong et al. (2019). Briefly, the reaction solution consisting of 1 ml of 20 mM Tris-HCl buffer (pH 7.4) and 0.06 mg of trypsin was mixed with 1 ml of the tested extract. The mixture was then incubated (37°C for 5 min) before adding 1 ml of 0.7% (w/v) casein, followed by further incubation for an additional 20 min. Subsequently, 1 ml of 70% perchloric acid (HClO 4 ) was added to terminate the reaction. The reaction mixture was then centrifuged at 4°C (3,000 rpm, 10 min), and the absorbance of the supernatant was measured at 210 nm against buffer as the blank. As control, phosphate-buffered solution was used. The percentage inhibition of protein denaturation was calculated using the following equation: where A1 = absorption of the control and A2 = absorption of the test sample mixture.

Lipoxygenase (LOX) inhibition assay
Lipoxygenase (LOX) inhibition activity of the extracts of S. buxifolia bark was assayed according to the method of Eshwarappa et al. (2016), with some minor modifications. Briefly, we used linoleic acid as substrate and lipoxidase as enzyme, both purchased from Sigma, Singapore. A mixture of a solution of sodium borate buffer (0.8 ml, 0.1 M, pH 8.8) and lipoxygenase (0.08 ml, final concentration 20,000 U/ml) was incubated with 0.8 ml of the tested extract in a 2-mL cuvette for 5 min at 25°C. After incubation, 0.08 ml of linoleic acid solution (0.6 mM) was added and mixed well, and absorbance was measured at 234 nm.
Indomethacin was used as reference standard, whereas phosphatebuffered solution was used as the control; the percentage inhibition of lipoxygenase was calculated using the following equation: where A1 = absorption of the control and A2 = absorption of the test sample mixture.
The results of in vitro anti-inflammatory activity of S. buxifolia extracts were also reported as IC 50 values. The IC 50 is defined as the concentration sufficient to obtain 50% of a maximum scavenging capacity. All tests and analyses were run in triplicate, and the obtained values were averaged.

| Statistical analysis
All experiments were performed in triplicate, and the results are given as the mean ± standard deviation (±SD) for total terpenoid content as well as all in vitro assay tested. Statistical comparisons were carried out by analysis of variance (ANOVA) using the Minitab 15 software and Tukey's multiple comparison test; p values < .05 were considered significant.
To reveal the patterns of variation and clustering among treatments, the obtained data were analyzed by means of multivariate analysis, employing hierarchical cluster analysis (HCA), and principal components analysis (PCA). Square Euclidean distances were used to calculate the distances between tested extracts of S. buxifolia bark. Each calculated principal component was validated using "full cross-validation," with 95% confidence level on parameters. For HCA, the method used was complete linkage. The dendrogram similarity scales generated by the Minitab program ranged from zero (greater similarity) to 5.64 (lower similarity). For PCA, transformed values of variables (average zero and standard deviation 1), called Z scores, were used.

| Effect of extraction solvent on TTC and EY of Severinia buxifolia bark
Based on the ANOVA results, there were significant differences between bark extracts (p < .001) (Figure 3a) (518.51 mg linalool/g DW) was not significantly higher than that of the hexane extract; conversely, the TTC of the acetone extract was significantly higher (648.15 mg linalool/g DW) than that of the hexane extract. Also, the TTC of the hexane: EtOAc extract was not significantly lower (601.85 mg linalool/g DW) than that of the acetone extract. Crude extract yield ranged from 3.43% for hexane extract to 5.23% for hexan: acetone extract (Figure 3a). The yield of the acetone extract (5.08%) was only slightly lower than that of the hexane: acetone extract, whereas the yield of hexane: diethyl ether extract (4.11%) and of the hexane: ethyl acetate extract (4.33%) were significantly higher than that of the hexane extract. Based on the results of TTC and EY, the best extracting solvent for S. buxifolia bark was hexane: acetone (1:1, v/v).

| Effect of maceration temperature on TTC and EY of Severinia buxifolia bark
The influences of maceration temperature (from 6 to 86°C) on the TTC and EY values of S. buxifolia extracts are shown in Figure 3b.
The TTC ranged from 453.70 mg linalool/g DW for the bark extract at 6°C to 842.59 mg linalool/g DW for the extract at 66°C. The TTC of S.buxifolia extract at 46°C (815.52 mg linalool/g DW) was not significantly lower than that of the extract at 66°C. Conversely, the TTC of the bark extract at 26°C (731.48 mg linalool/g DW) was significantly higher than that of the extract at 6°C (453.70 mg linalool/g DW). Extract yield ranged from 3.00% for the bark extract at 86°C to 4.83% for the extract at 46°C. Yield at 66°C (4.07%) was significantly lower than that at 46°C. Both parameters had no significant influence on the extraction yield at 6°C (3.20%) and 26°C (3.63%).
Generally, based on the TTC and EY results, the most suitable maceration temperature for extracting S. buxifolia bark is 46°C.

| Fractionation by liquid-liquid extraction
Based on the results for TTC and EY of the preliminary experiments (3.1), the mixture solvent (hexane: acetone (1:1, v/v) was used to obtain the crude extract of S. buxifolia bark at 46°C for 24 hr of maceration. Organic solvents such as petroleum ether and a mixture of hexane: ethyl acetate (85:15, v/v) were used to partition the crude extract via liquid-liquid extraction.

| Changes in the TTC of Severinia buxifolia bark extracts
The TTC levels of crude extract and fractions are presented in Figure 5. The TTC value significantly increased from the crude extract to the fractions (p < .001). The highest value was found for the HEF fraction (731.48 mg linalool/g DW), whereas the crude CE had the lowest TTC value (184.26 mg linalool/g DW). The TTC value of PEF was significantly higher than that of the crude CE (564.81 mg linalool/g DW) (p < .001).
F I G U R E 4 Scores (a) and Loading (b) plots of principal component analysis (PCA) calculated using means of total terpenoid content (TTC) and percentage of inhibition (%) of in vitro anti-inflammatory activity (inhibition of albumin denaturation, heat-induced hemolysis, proteinase activity, and lipoxygenase (LOX) assay). Symbols of Scores (a): •: crude bark extract (CE); ■: hexane:ethyl acetate bark extract fraction (HEF); and ▲: petroleum ether bark extract fraction (PEF) F I G U R E 5 Total terpenoid content (dot bar) of the crude bark extract (CE) and the bark extract fractions (petroleum ether (PEF) and hexane:ethyl acetate (HEF)) obtained by liquid-liquid extraction. Means within bars with different letters significantly differ by Tukey's test at p < .05

| TLC and HPLC profiling
We performed TLC and HPLC analyses as a part of the quality control of the S. buxifolia bark extracts. First, the TLC of crude extract and fractions was prepared using oleanolic acid and ursolic acid as markers of terpenoid compounds (Figure 6a). The retention factor (Rf) values of crude extract and fractions are shown in Table 1

Proteinase inhibitory activity
The different S. buxifolia bark extracts (crude CE, fractions PEF and HEF) at various concentrations showed significant antiproteinase activity (p < .001) (

Lipoxygenase inhibition assay
The results of the in vitro LOX inhibitory assay of the bark extracts and the standard (indomethacin) are shown in  (Table 3).

TA B L E 2
In vitro anti-inflammatory activity of the crude extract (CE) and the extract fractions (petrolium ether-PEF and hexane:ethyl acetate-HEF) of Severinia buxifolia bark

| D ISCUSS I ON
The present study demonstrated that solvent-solvent fractionation of the S. buxofolia bark extracts can enhance the total terpenoid content as well as the anti-inflammatory properties in vitro in some models, such as inhibition of protein (albumin) denaturation, heatinduced hemolysis, proteinase activity, and LOX assay. In addition, the findings of some preliminary experiments on the effects of some factors of the maceration technique (solvent type, temperature, and time) on TTC and EY showed that hexane: acetone can be considered as the optimal solvent for extracting terpenoids from S. buxifolia bark at 46°C for 24 hr.
By applying the colorimetric assay with linalool as the standard reagent, the preliminary experiments of this study showed changes in the TTC of S. buxifolia extracts due to different maceration factors (solvent type, temperature, and time) (Figures 2   and 3). In all S. buxifolia extracts, the content of total terpenoids ranged from 453.70 to 842.59 mg linalool/g DW, and the yield varied from 2.40% to 5.10%. Single-step extraction using hexane: acetone (1:1, v/v) was the optimal solvent for extracting the maximum amount of terpenoid components, and the extraction should be carried out over 24 hr at 46°C. The extraction technique is one of the most important steps to increase the contents of the desired bioactive compounds from plants (Azmir et al., 2013;Monton & Luprasong, 2019). One solvent is recommended as a good and optimal solvent for extraction based on its capacity in conserving the stability of the chemical structure of the targeted compounds (Do et al., 2014;Monton & Luprasong, 2019). For instance, the results of Thouri et al. (2017) revealed that when using Tunisian date seeds, the phenolic, flavonoid, and condensed tannin contents of the water and methanol extracts were higher than those of the acetone extracts. These authors showed that the polar solvent exhibited the highest amounts of bioactive compounds.
In the present study, the TTC with regard to different solvents Inflammation is a complex process in the human body and caused by physical injury and various chemicals (Chen et al., 2017;Manna & Jain, 2015;Parameswari et al., 2019). Nonsteroidal anti-inflammatory drugs (NSAIDs) are most commonly used for the control of inflammatory conditions, based on the production of prostaglandin (Djuichou Nguemnang et al., 2019;Manna & Jain, 2015). However, they have various side effects, especially the formation of gastric ulcers caused by gastric irritation (Oguntibeju, 2018). In recent years, plant compounds have been considered as novel compounds with effective anti-inflammatory activities (Parameswari et al., 2019;Shaikh et al., 2015;Upadhyay, 2020). Hoang et al. (2015)  TA B L E 3 IC 50 value of in vitro antiinflammatory activity of the crude extract (CE) and the extract fractions (petrolium ether-PEF and hexane:ethyl acetate-HEF) of Severinia buxifolia bark S. buxifolia were performed in this study, but with a focus on the fractionation of plant bark, which contains the terpenoid class as the main compound.
Protein denaturation leads to the destruction of the tertiary and secondary structure of proteins and is associated with the occurrence of inflammatory responses (Anoop & Bindu, 2015). Some anti-inflammatory drugs, such as salicylic acid, phenylbutazone, and flufenamic acid, have a dose-dependent ability to inhibit thermally induced protein denaturation (Djuichou Nguemnang et al., 2019).
The IC 50 values of crude CE as well as PEF and HEF fractions of S.
Cell vitality is a factor of membrane integrity, focusing on injuries of the RBCs' substances (Manoj et al., 2009). Chemical components with membrane-stabilizing properties are expected to provide effective protection for cell membranes against injuries (Liu et al., 1992;Manoj et al., 2009;Shinde et al., 1999). Hence, we investigated membrane stabilization to further establish evidence for the anti-inflammatory action of the plant components. The results showed that the crude extract and fractions have significant membrane-stabilizing properties (  During the inflammatory responses, the leukocyte proteinases play an important role in the development of tissue damage (Gunathilake et al., 2018). In recent studies, many terpenoids contributed significantly to in vitro anti-inflammatory activities of various plant extracts (Gallily et al., 2018;Marques et al., 2018;Namdar et al., 2019;Prakash, 2017). Thus, the presence of this component class in the plant extract maybe contributes to its anti-inflammatory properties. Indeed, the results of the current study showed that the proteinase inhibition activity of S. buxifolia extracts increased according to the TTC content. The IC 50 values of these extracts also significantly varied (p < .05) (Table 3) Lipoxygenases are one of the key enzymes in the biosynthesis of leukotrienes and play an important role in the response of some inflammatory diseases such as asthma, cancer, and arthritis (Chen et al., 2017;Gunathilake et al., 2018). In the anti-inflammatory mechanism, arachidonic acid plays an important role in the responsive properties (Tallima & Ridi, 2017). According to Gardner (1991), the LOX pathway in plants is equivalent to the "arachidonic acid cascades" in animals. Thus, the inhibition of LOX in vitro can provide good evidence for the screening of plant extracts with anti-inflammatory ability (Leelaprakash & Mohan, 2011). The observations of several previous studies showed that some medical plants have high LOX inhibitory activity (Gunathilake et al., 2018;Khasawneh et al., 2011;Leelaprakash & Mohan, 2011;Rackova et al., 2007). For example, Gymnema lactiferum extracts have a potentially high anti-inflammatory effect based on the LOX assay (Gunathilake et al., 2018).
According to Yoon and Baek (2005), plant polyphenols can block or interfere with the cascade process of arachidonic acid metabolism via LOX inhibitory activity. In the present study, the results of the LOX inhibitory differed significantly between the crude extract and the fractions ( proportional to the inhibition increase. Ursolic acid may exhibit potent anti-inflammatory effects (Checker et al., 2012); these authors also recommended the application of ursolic acid in the treatment of inflammatory disorders. Ursolic acid is considered as the major bioactive component of several plant species and has a wide range of biological functions, such as anti-inflammatory, anticancer, and antioxidative activities (Ikeda et al., 2008;Tsai & Yin, 2008). The anti-inflammatory, anticarcinogenic, and proapoptotic activities of ursolic acid could be correlated to its potential to inhibit the immunoregulatory transcription factor NFkB in carcinogens and inflammatory responses (Checker et al., 2012;Shishodia et al., 2003).

| CON CLUS IONS
From the S. buxifolia bark, the preliminary experiments of the present study were determined the optimal conditions of maceration technique (solvent type, temperature, and time) for extracting the terpenoid from S. buxifolia bark. The results showed that the mixture of hexane:acetone (1:1, v/v) was considered as the optimal solvent to obtain the S. buxifolia crude extract at 46°C for 24 hr of maceration. Different organic solvents like petroleum ether and mixture of hexane:ethyl acetate (85:15, v/v) were used to fractionate the phytochemical groups. Solvent fractionation was found to increase the TTC compared with crude extracts based on colorimetric assay using linalool as standard reagent. The samples were indicated to have in vitro anti-inflammatory comparable with two reference standards (aspirin and indomethacin) based on the inhibition of albumin denaturation, proteinase activity, heat-induced hemolysis, and lipoxygenase assay. The HEF fraction was mostly showed the highest in vitro anti-inflammatory assays in comparison with the PEF fraction and the crude CE. In addition, one of composition of terpenoid, namely ursolic acid, was determined by HPLC method from the crude CE and the fractions PEF and HEF: 2.44 μg/g DW, 3.56 μg/g DW, and 5.04 μg/g DW, respectively. Further studies would be conducted in the next research to purify and isolate more components from fractions as well as crude extracts. Generally, this study investigated that solvent-solvent fractionation significantly influenced on TTC as well as in vitro anti-inflammatory activity of plant extracts.

ACK N OWLED G M ENTS
The authors are grateful to Ton Duc Thang University for supporting with laboratory to conduct this project.

CO N FLI C T O F I NTE R E S T
The authors declare no financial or commercial conflict of interest.
The authors confirm that this manuscript is an honest, accurate, and transparent account of the study being reported.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data used to support the findings of this study are available from the corresponding author upon request.