Optimization of total saponin extraction from Tam that roots using the desirablity methodology

JOURNAL OF SCIENCE OF HNUE DOI: 10.18173/2354-1059.2016-0052 Natural Sci. 2016, Vol. 61, No. 9, pp. 26-33 This paper is available online at 26 OPTIMIZATION OF TOTAL SAPONIN EXTRACTION FROM TAM THAT ROOTS USING THE DESIRABLITY METHODOLOGY Nguyen Quang Tuyen and Dang Ngoc Quang Faculty of Chemistry, Hanoi National University of Education Abstract: In Vietnam, Tam that hoang (Panax stipuleanatus Tsai et Feng) belonging to Araliaceous family is a valuable medicinal plant that possesses many biologically active compounds. Among them, flavonoids, polysaccharides, and saponins are the major constituents of this plant. They have anti-oxidant, anti-microbial, anti-cancer, and tonic properties. In this paper, we studied the factors that affect the process of total saponin extraction from Tam that roots originated from Sapa town, Lao Cai province. The total saponin extraction from Tam that roots was optimized using the desirability methodology with the experimental order of Box-Behnken design. The concentrations of ethanol, the ratio of ethanol/material (v/m), the time of ultrasonic extraction with the frequency of 33 - 40 kHz are the three main considered factors. The optimized condition for total saponin extraction was found as follows: ethanol concentration 65%, ethanol/material ratio 55/1(v/m), and extraction time 75 minutes. In this condition, the maximum total saponin content was predicted to be 15.01%. Keywords: Extraction, saponin, Panax stipuleanatus. 1. Introduction Extraction is the first step of great importance for the recovery and purification of active ingredients from plant materials. Traditional extraction techniques usually require many extraction hours and have low efficiency. Recently, there have been numerous reports on the application of high intensity or power ultrasound in the extraction of various phyto-chemicals such as alkaloids, flavonoids, poly-saccharides, proteins and essential oils from various parts of plants and their seeds [1-3]. The ultrasound-assisted extraction can be done at a low temperature to avoid the decomposition of natural compounds and essential oil. Even so there are several advanced extraction technologies, ultrasound-assisted extraction using ethanol solvent is a reasonable choice, since it is economical and can be applied in industry without toxic residues [4-7]. Statistical experimental designs are proved to be more satisfactory and effective than other traditional which only allow us to study one-at-a-time factor or mathematical methods because it can be used to study many variables simultaneously, with less observation. This method has been widely used in different research fields such as biology, agriculture, food, and medicine etc. Received October 10, 2016. Accepted November 25, 2016. Contact Dang Ngoc Quang, e-mail address: quangdn@hnue.edu.vn Optimization of total saponin extraction from Tam that roots using the desirablity methodology 27 Desirability methodology is considered one of the effective methods for multiple objective optimizations of technical factors [8]. In this study, the desirability methodology was employed to optimize total saponin extraction from dried powder of the roots of Panax stipuleanatus using ethanol as solvent. 2. Content 2.1. Experiment 2.1.1. Material Roots of Panax stipuleanatus, were collected in Sapa town, Lao Cai province, northern part of Vietnam in May 2013, with samples identified by Mrs. Pham Thanh Huyen from National Institute of Medicinal Materials. Voucher specimens were deposited at the Faculty of Chemistry, Hanoi National University of Education (N 0 -9754; 9755). Materials were collected, cleaned steamed at 100 0 C for 5 minutes, and sliced, then dried at 60 0 C for 5 days to the moisture of 5 - 6%, finally powdered with the size of particles of 0.1 - 1 mm. Dried samples were kept in two-layer polyethylene bags, placed into glass bottles and stored in a dry and cool place. 2.1.2. Analysis methods Quantitative analysis of the total saponin was carried out using the analytical balance. In addition, the saponin content was determined by HPLC (Hitachi) with standard saponins (notoginsenosid R1, ginsenosid Rg1 and ginsenosid Rb1), equipped with a reversed phase column, phenomenex C18 (5 m × 250 mm × 4.6 mm) using acetonitrile and water as a solvent system. 2.1.3. Research methods Extraction: Dried powder of Tam that roots (approximately 2 g) was placed in a 250 mL flask. The extraction was carried out in a thermo-regulatory ultrasound machine with an ethanol concentration 50 - 90%, ethanol/material ratio (30 - 60/1, v/m), and extraction time 30 - 90 minutes. The extract was filtered and solvents were removed by a rotary evaporator. The residue was finally dried by a vacuum dryer. The total saponin was calculated by a percentage of the sample (% dry substance). Statistical experimental design method: Using the statistical Box-Behnken design with three factors: ethanol concentration (%), ethanol/material ratio (v/m) and ultrasonic extraction times (minutes), (Table 1). This design has 15 trials including 3 ones for center points to response: total saponin (% dry substance), (Table 2). Table 1. The factors and their levels Factors Symbols Units Levels -1 0 +1 Ethanol concentration A % 50 70 90 Ethanol/material B v/m 30 45 60 Ultrasonic times C minutes 30 60 90 Nguyen Quang Tuyen and Dang Ngoc Quang 28 Table 2. Experimental design and total saponin Run Ethanol concentration (%) Ethanol/material (v/m) Ultrasonic times (minutes) Total saponin (%) 1 50.00 30.00 60.00 9.65 2 90.00 30.00 60.00 6.49 3 50.00 60.00 60.00 12.35 4 90.00 60.00 60.00 9.16 5 50.00 45.00 30.00 9.99 6 90.00 45.00 30.00 7.04 7 50.00 45.00 90.00 11.98 8 90.00 45.00 90.00 8.56 9 70.00 30.00 30.00 9.97 10 70.00 60.00 30.00 10.79 11 70.00 30.00 90.00 9.85 12 70.00 60.00 90.00 14.41 13 70.00 45.00 60.00 14.17 14 70.00 45.00 60.00 14.16 15 70.00 45.00 60.00 14.18 2.1.4. Statistical analysis The statistical Design-Expert 7.1 software (Stat-Ease, Inc., Minneapolis, USA) was used for regression analysis of experimental data to plot response surface and to optimize, according to the desirability methodology [9]. ANOVA (Analysis of variance) was used to estimate the statistical parameters. 2.2. Results and discussion 2.2.1. Study on the influence factors on total saponin extraction The total saponin amount depended on many factors such as ethanol concentration (%) (Figure 1), ethanol/material ratio, (v/m) (Figure 2) and ultrasonic times (minutes) (Figure 3) [10, 11]. In our work, each single factor was studied in order to find out its effect on total saponin extraction. Based on those, we will choose a value range of each factor for optimization. For example, the ultrasonic time (60 min) and the ethanol/material ratio (v/m: 50/1) were fixed, but ethanol concentration was varied to find out the value range of ethanol concentration. Optimization of total saponin extraction from Tam that roots using the desirablity methodology 29 Figure 1. Ethanol concentration (%) Figure 2. Ethanol/material ratio (v/m) Figures 1 - 3 show that total saponin extraction depended on all three 3 factors. Based on the experimental data, we decided to choose value ranges of each factor as follow: ethanol concentration 50 - 90%, ethanol/material ratio, 30/1 - 60/1 (v/m), and time of extraction using ultrasound 30 - 90 min. Figure 3. Time of extraction using ultrasound (min) 2.2.2. Model building The results of experimental design are shown in table 2. The response total saponin was fitted with a second-order polynomial equation of the three factors. The significance of the model and the co-efficients are controlled by a regression analysis (Table 3). The F-values of the model are 1.916.10 5 showing that the model is completely significant of 99.99% (p < 0.0001) confidence level. The p-value less than 0.05 indicating that the coefficient is significant, therefore, in the model only interactive coefficient AB (p = 0.099) was insignificant, but it was remained in the model for optimization. F-value for “lack of fit” of the model was 0.25 (p = 0.8576). It showed that the model was fit with the experiment. In addition, the coefficient of determination (R 2 ) of the model 0.999 showed that the model was well-matched with the designed experiment. The total saponin is performed in a second-degree model as follows: Y = 14.17 - 1.59A + 1.34B + 0.88C - 7.5.10 -3 AB - 0.12AC + 0.94BC - 3.31A 2 - 1.45B 2 - 1.47C 2 Nguyen Quang Tuyen and Dang Ngoc Quang 30 Table 3. Regression analysis of total saponin Source Mean square F-value p-value (prob > F) Model A B C AB AC BC A2 B2 C2 Lack of fit 94.85 20.22 14.45 6.14 2.250E-004 0.055 3.50 40.45 7.74 7.95 2.00E-005 1.916E + 005 3.677E + 005 2.626E + 005 1.117E + 005 4.09 1004.09 63580.00 7.355E + 005 1.407E + 005 1.446E + 005 0.25 < 0.0001 < 0.0001 < 0.0001 < 0.0001 0.0990 < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001 0.8576 Considering in turn the effect of each factor on the total saponin extract (while other factors are remained in the central level) (Figure 4), it showed that the ethanol concentration (A) has played a significant role on the total saponin content: meanwhile ethanol/material ratio (B) and ultrasonic time (C, time of extraction) took the same role on the total saponin. This can be shown obviously in the response surface of Y-function (Figures 5a, 5b, 5c). Figure 4. Influence of factors on total saponin extraction Optimization of total saponin extraction from Tam that roots using the desirablity methodology 31 5a 5b 5c Figure 5. Response surface plot for total saponin extraction 2.2.3. Optimization of the total saponin extraction The most important objective of total saponin extraction process from Tam that roots is to get the highest total saponin. To this point, the optimization using algorism of desirability methodology invented by Derringer and Suich was applied [8]. The results after using the Design- Expert 7.1 Software were as follows: ethanol concentration 65%, ethanol/material ratio 55/1 (v/m), and time of extraction using ultrasound 75 min. In this condition, the predicted maximum total saponin extract was 15.01% (Figure 6). Figure 6. Responsible desirability level Nguyen Quang Tuyen and Dang Ngoc Quang 32 2.2.4. Validation of the model The trial experiments were conducted (four times) under optimized process conditions with ethanol concentration 65%, ethanol/material ratio 55/1 (v/m), and time of extraction using ultrasound 75 min. The mean value of the total saponin has reached 15.00 ± 0.02 % (Table 4), which was very close to the predicted total saponin obtained from the regression equation (15.01 %). Thus, the model could be used to predict the total saponin content obtained from the extraction. Table 4. The extraction results under optimal condiction Number Ethanol concentration (%) Ethanol/material ratio (v/m) Time of extraction (min) Total saponin (%) 1 65 55/1 75 15.01 2 65 55/1 75 14.99 3 65 55/1 75 15.01 4 65 55/1 75 14.98 Average 15.00 ± 0.02 3. Conclusion Based on the statistical experimental design using response surface and desirability methodology, the optimal condition for total saponin extraction by ethanol solvent was determined as follows: ethanol concentration 65%, ethanol/material ratio 55/1 (v/m), and time of extraction using ultrasound 75 min. The maximum total saponin content was predicted to be 15.01%. REFERENCES [1] Nguyen Tap, 2006. Herbal medicines research. Science & Technology Publishing House, 2006, pp 33-111 (in Vietnamese). [2] Nguyen Thuong Dong, 2008. Panax vietnamensis Ha et Grushv in the genus Panax L. and Araliaceae in the world. Vietnam second ginseng workshop, pp. 13-25. [3] S.R. Sargenti, W. Vichnewski, 2000. Sonication and liquid chromatography as a rapid technique for extraction and fractionation of plant material. Phytochem. Anal. 11, pp. 69-73. [4] M. Vinatoru, M. Toma, O. Radu, P.I. Filip, D. Lazurca, T.J. Mason, 1997. The use of ultrasound for the extraction of bioactive principles from plant materials. Ultrason. Sonochem. 4, pp. 135-139. [5] L.C. Wang, 1984. Ultrasonic extraction of a heat-labile 7S protein fraction from autoclaved, defatted soybean flakes. J. Food Sci. 49, pp. 551-554. [6] D.H.Vila, F.J.H. Maria, R.B. Lucena, M.A.F. Recamales, 1999. Optimization of an extraction method of aroma compounds in white wine using ultrasound . Talanta, 50, pp. 413-421. Optimization of total saponin extraction from Tam that roots using the desirablity methodology 33 [7] T.J. Mason, 2000. Large scale sonochemical processing: aspiration and actuality. Ultrason. Sonochem. 7, pp. 145-149. [8] G. Derringer, R. Suich, 1980. Simultaneous optimization of several responses variable. Journal of quality technology, 12, pp. 214-219. [9] Design-Expert version 7.1 (2007), Software for design of experiments, Stat-Ease, Inc., Minneapolis, USA. [10] S. Gafner, C. Bergeron, M.M. Mc Collom, L.M. Cooper, K.L. Mephail, W.H. Gerwick, C.K. Angerhofer, 2004. Evaluation of the efficiency of three different solvent systems to extract triterpene saponins from roots of Panax quinquefolius using high-performance liquid chromatography. J. Agric. Food. Chem., 52, pp. 1546-1550. [11] W. Jianyong, L. Lidong, C. Foo-Tim, 2001. Ultrasound-assised extraction of ginseng saponins from ginseng roots and cultured ginseng cells . Ultrasonics Sonochemistry, 8, pp. 347-352.