International Journal of Food Sciences and Nutrition, February 2011; 62(1): 91–96 Optimization conditions for anthocyanin and phenolic content extraction form purple sweet potato using response surface methodology MARUF AHMED1,2, MST. SORIFA AKTER1, & JONG-BANG EUN1 1 Department of Food Science and Technology and Institute of Biotechnology, Chonnam National University, Gwangju, South Korea, and 2Department of Food Processing and Preservation, Hajee Mohammad Danesh Science and Technology University, Dinajpur, Bangladesh Abstract Purple sweet potato ﬂour could be used to enhance the bioactive components such as phenolic compounds and anthocyanin content that might be used as nutraceutical ingredients for formulated foods. Optimization of anthocyanin and phenolic contents of purple sweet potato were investigated using response surface methodology. A face-centered cube design was used to investigate the effects of three independent variables: namely, drying temperature 55 – 658C, citric acid concentration 1– 3% w/v and soaking time 1– 3 min. The optimal conditions for anthocyanin and phenolic contents were 62.918C, 1.38%, 2.53 min and 60.948C, 1.04% and 2.24 min, respectively. However, optimal conditions of anthocyanin content were not apparent. The experimental value of anthocyanin content was 19.78 mg/100 g and total phenolic content was 61.55 mg/g. These data showed that the experimental responses were reasonably close to the predicted responses. Therefore, the results showed that treated ﬂours could be used to enhance the antioxidant activities of functional foods. Keywords: Purple sweet potato, response surface methodology, phenolic compounds, anthocyanin content Introduction Purple-ﬂeshed sweet potatoes have an intense purple ethanol, acetone, water or mixture (Pathirana and color in the storage roots due to the accumulation of Shahidi 2005, Huang et al. 2006). The stability of anthocyanins (Terahara et al. 2004).The anthocyanins anthocyanin and phenolic content were inﬂuenced by in purple sweet potato are mono-acylated or di-acylated several factors (Jiang 2000). Among them, polyphenol forms of cyanidin and peonidin (Yang and Gadi 2008). oxidase plays an important role in the degradation of Sweet potatoes had intermediate antioxidant activity anthocyanin and phenolic content. Citric acid has been among 43 vegetables (Huang et al. 2006). Recently used extensively for the inhibitory activity on poly- natural antioxidants have attracted considerable atten- phenol oxidase and the anti-browning activity in tion due to their positive health beneﬁt (Huang et al. minimally processed fruits and vegetables. Citric acid 2006). Rumbaboa et al. (2009) reported that antho- extracts have a double inhibitory effect by chelating cyanin from purple sweet potato has better radical copper at lower pH (Altunkaya and Gokmen 2009). scavenging activity than that of red cabbage, grape skin, Sweet potatoes can be processed into ﬂour, which elderberry and purple corn. Anthocyanins from purple are less bulky and more stable than the highly sweet potatoes have many biological functions, such as perishable fresh root. Flour can be used as a thickener scavenging free radicals, anti-mutagenicity, anti-carci- in soup, gravy, fabricated snacks and bakery products. nogen activity and antihypertensive effect (Oki et al. It could be used to enhance food products through 2002). Several extraction methods have been used to color, ﬂavor, natural sweetness and nutrients. Sing et al. obtain extracts rich in anthocyanin and phenolic (2003) used potassium metabisulphite, citric acid and content based on different solvents such as methanol, sodium chloride to improve the quality of chips from Correspondence: Jong-Bang Eun, Department of Food Science & Technology, Chonnam National University, 77 Yongbong-ro Buk-gu, Gwangju 500-757, South Korea. Tel: 82 62 530 0255. Fax: 82 62 530 2149. E-mail: firstname.lastname@example.org ISSN 0963-7486 print/ISSN 1465-3478 online q 2011 Informa UK, Ltd. DOI: 10.3109/09637486.2010.511167 92 M. Ahmed et al. sweet potatoes. Response surface methodology (RSM) 6 – 7%) was obtained by milling the dried slices using a has been successfully used to optimize biochemical blender (FM-681C; Hanil, Gwangju, Korea), and and biotechnological process related to food systems sieved through an 80-mesh (Chung gye sang gongsa, (Cacace and Maza 2003, Pathirana and Shahidi Seoul, South Korea) screen. 2005). Therefore, the goal of the present study was to optimize different pretreatments such as drying Experimental design for RSM analysis temperature, citric acid concentration and soaking A three-factor (X1, X2 and X3) and three-level (– 1, 0 time for production of sweet potato ﬂour with and 1) face-centered cube design were employed in high retention of anthocyanin and phenolic content this study, and 15 individual run points were taken for using RSM. analysis (Wanasundara and Shahidi 1999). The actual and corresponding values are presented in Table I. Materials and methods The multiple regression equation was used to ﬁt the Raw materials second-order polynomial equation based on the experimental data as follows: Sweet potato (Ipomoea batatas L. Lam variety, Sinjami) was purchased from a local farm. Roots Y ¼ b0 þ b1 X1 þ b2 X2 þ b3 X3 þ b11 X1 X1 were washed with tap water to remove dirt and soil and allowed to dry at ambient temperature (, 208C). þ b22 X2 X2 þ b33 X3 X3 þ b12 X1 X2 þ b13 X1 X3 The washed sweet potatoes were stored at 148C for þ b23 X2 X3 15 days without curing. where Y is the response variable, b0 is the intercept, b1, Sample preparation and treatment b2, b3, b11, b22, b33 and b12, b13, b23 are linear, Sweet potatoes were peeled with a hand peeler (Han quadratic and interaction coefﬁcients respectively, and Sung 27 stainless; Namdong Industry Park, Incheon, X1, X2 and X3 are the coded independent variables. South Korea). Then peeled samples were cut into slices (1 mm thickness) using a slicing machine (HFS 350G; Veriﬁcation of model Hankook fujee Industries Co. Ltd. Suwon-si, RSM was used to optimize anthocyanin and phenolic Gyeonggi-do, Fujee, South Korea). Various levels of contents from purple sweet potato. The experimental citric acid concentration (1 – 3% w/v) were solubilized and predicted values were compared to conﬁrm the in deionized water at room temperature (20 ^ 18C). validity of the model. After that, peeled slices were dipped in aqueous citric acid solutions (1 –3% w/v) for different soaking times Analysis of anthocyanin contents (1 – 3 min) at room temperature. The content of anthocyanin was determined following the procedures of Proctor (1974) and Huang et al. Preparation of sweet potato ﬂour (2006) The sweet potato ﬂour (1 g) was treated with The slices were dried using a convection drying oven 15 ml HCl – methanol (0.15% HCl:methanol ¼ (Dasol Scientiﬁc Co. Ltd, Seoul, South Korea) at 15:85) for 4 h. The extract was ﬁltered and its different temperatures 558C, 608C, and 658C for absorbance was determined at 530 nm. The antho- 7 – 8 h. The sweet potato ﬂour (moisture content cyanin content was calculated on the basis of the Table I. Three-factor, three-level, face-centered cube design for RSM. Factor X1 Factor X2 Factor X3 Variable Assay number Drying temperature (8C) Concentration (%) Soaking time (min) Anthocyanin (mg/100 g) Total phenolics (mg/g) 1 55 (21) 2(0) 3(þ 1) 40.32 ^ 1.32 56.85 ^ 8.28 2 55(21) 1(21) 2(0) 39.56 ^ 1.03 47.77 ^ 1.53 3 55(21) 2(0) 1(21) 40.79 ^ 0.60 47.18 ^ 5.73 4 55(21) 3(þ 1) 2(0) 37.40 ^ 2.08 47.52 ^ 4.02 5 60(0) 3(þ 1) 3(þ 1) 34.17 ^ 4.72 44.64 ^ 0.34 6 60(0) 1(21) 1(21) 32.53 ^ 2.73 51.69 ^ 0.19 7 60(0) 3(þ 1) 1(21) 24.45 ^ 4.96 56.38 ^ 0.39 8 60(0) 1(21) 3(þ 1) 24.51 ^ 0.37 63.38 ^ 1.02 9 65(þ 1) 2(0) 3(þ 1) 29.16 ^ 3.64 46.46 ^ 1.07 10 65(21) 1(21) 2(0) 26.49 ^ 1.83 51.65 ^ 2.37 11 65(þ 1) 2(0) 1(21) 24.55 ^ 0.29 51.65 ^ 2.37 12 65(þ 1) 3(þ 1) 2(0) 22.45 ^ 0.28 52.03 ^ 2.43 13 60(0) 2(0) 2(0) 23.94 ^ 0.57 45.70 ^ 0.41 14 60(0) 2(0) 2(0) 20.02 ^ 1.18 49.39 ^ 0.81 15 60(0) 2(0) 2(0) 21.63 ^ 0.17 58.91 ^ 1.18 Optimizing anthocyanin and phenolic content of purple sweet potato ﬂour 93 following equation: (a) Anthocyanin content ¼ A £ MW £ DF £ 100=ðe £ WÞ 49.22 where A is the absorbance, MW is the molecular weight of cyanidin-3-glucoside (MW ¼ 449.2), DF is Anthocyanins (mg/100 g) the dilution factor, 1 is the molar absorptivity (34,300), and W is the sample weight (g). 39.39 Analysis of total phenolic contents The total phenolic content was determined using 29.55 Folin – Ciocalteau reagent according to a slightly 3.00 %) modiﬁed method (Swain and Hills 1959). The sample 2.33 n( (0.1 g) was extracted three times with 20 ml of 75% tio 19.72 tra methanol and was ﬁltered through Whatman No. 2 1.67 en 65.00 ﬁlter paper. Extracts were combined and concentrated nc 61.67 Co in a rotary vacuum evaporator (Rikakikai Co. Ltd, Tempe 58.33 1.00 ra ture (°C 55.00 Tokyo, Japan) at 408C; the volume was adjusted to ) 20 ml with 75% methanol. One milliliter of extract, 5 ml distilled water and 2 ml of 10% Folin – Ciocalteau (b) reagent were added into a Falcon tube. After 3 min at room temperature, 2 ml of 7.5% Na2CO3 solution was added and the sample was diluted to 20 ml with distilled water. Each sample was allowed to stand for 49.22 Anthocyanins (mg/100 g) Table II. Regression coefﬁcients of predicted quadratic polynomial 39.39 for the response anthocyanin and phenolic contents. Coefﬁcient Anthocyanin content Phenolic content 29.55 b0 1240.40*** 2176.82 3.00 Linear n) mi b1 236.84*** 6.19 2.33 e( b2 224.72 21.65 tim b3 247.49*** 37.69 19.72 1.67 ing Quadratic 65.00 ak b11 0.28*** 20.04 61.67 So 58.33 1.00 b22 3.12*** 24.76 Tempe 55.00 rature (°C b33 3.92*** 21.95 ) Cross-product b12 0.04 20.16 Figure 1. Response surface plots of the anthocyanin content of b13 0.39 20.30 purple sweet potato ﬂour as affected by temperature, citric acid b23 4.43*** 25.97** concentration and soaking time. (a) Temperature and concentration. R2 0.97 0.80 (b) Temperature and soaking time. R 2, coefﬁcient of multiple determination. ***Signiﬁcant at P # 0.01. **Signiﬁcant at P # 0.05. 1 h at room temperature and absorbance was measured at 760 nm (UV-1201; Shimadzu, Kyoto, Table III. Analysis of variance for the response surface quadratic Japan). The total phenolic content was calculated on model for anthocyanin and phenolic contents. the basis of standard curves of gallic acid, and Degree Sum expressed as milligrams of gallic acid equivalents per Source of freedom of squares Mean square F value gram of sample on a wet weight basis. For anthocyanin Lack of ﬁt 3 7.50 2.50 Statistical analysis Pure error 2 7.76 3.88 0.64 Total error 5 15.26 3.05 All determinations were carried out in triplicate and For total phenolic the experimental results are expressed as means ^ Lack of ﬁt 3 18.75 6.25 Pure error 2 92.91 46.45 0.93 standard deviation. Statistical analysis of the veriﬁca- Total error 5 111.67 22.33 tion results was carried out by analysis of variance and Duncan’s multiple-range tests using SAS (version 94 M. Ahmed et al. (a) 65.0 (a) 62.5 Temperature (°C) 64.46 60.0 Phenolic content (mg/g) 57.5 56.82 55.0 1.0 1.5 2.0 2.5 3.0 49.17 Concentration (%) 3.00 %) 21.20 25.62 30.05 34.47 2.33 n( Anthocyanins (mg/100 g) 38.90 43.32 47.75 tio 41.53 tra 1.67 65.00 en (b) 65.0 61.67 nc 58.33 1.00 Co Tempera 55.00 tu re (°C) Temperature (°C) 62.5 (b) 60.0 57.5 64.46 55.0 Phenolic content (mg/g) 1.0 1.5 2.0 2.5 3.0 Soaking time (min) 56.82 21.20 25.62 30.05 34.47 Anthocyanins (mg/100 g) 38.90 43.32 47.75 Figure 2. Contour plots showing the effects of temperature, citric 49.17 3.00 acid concentration and soaking time on anthocyanin content of n) purple sweet potato ﬂour. (a) Temperature and citric acid mi 2.33 e( concentration. (b) Temperature and soaking time. tim 41.53 1.67 65.00 ing 61.67 ak 9.1). The optimal conditions were estimated through 58.33 1.00 So Tempera 55.00 three-dimensional response surface analyses of the tu re (°C) three independent variables and each dependent variable. Figure 3. Response surface plots of the phenolic contents of purple sweet potato ﬂour as affected by temperature, citric acid concentration and soaking time. (a) Temperature and concentration. (b) Temperature and soaking time. Results and discussion Fitting the models Analysis of variance using SAS was performed to Effect of pretreatment on anthocyanin and phenolic determine the signiﬁcance of the linear, quadratic, contents cross-product (Table II) and the lack of ﬁt (Table III) The anthocyanin content of sweet potato ﬂours ranged of the independent variables on the anthocyanin and from 20.02 to 40.79 mg/100 g wet weight basis phenolic contents. The lack-of-ﬁt test is a measure of (Table I). The contents of anthocyanin were much the failure of a model to represent data in the higher than those of sweet potato puree (Steed and experimental domain at the points that were not Truong 2008) and of steamed or kneaded ﬂours included in the regression (Montgomery 1984). However, the R 2 value of the dependent variables (Huang et al. 2006). The results of multiple regression was approximately 0.80, indicating that a high analysis showed that the anthocyanin contents were proportion of variability was explained by the data signiﬁcantly (P # 0.001) affected by the linear term of (Varnalis et al. 2004). Therefore, the results showed temperature and soaking time, the quadratic of all that the experimental model was adequate due to no terms and the interaction term of concentration and signiﬁcant lack of ﬁt and satisfactory levels of R 2. soaking time (Table II). The predicted model obtained Optimizing anthocyanin and phenolic content of purple sweet potato ﬂour 95 the anthocyanin structure could be varying with pH (a) 65.0 (Cevallos-Casala and Cisneros-Zevallos 2004). The total phenolic content of sweet potato Temperature (°C) 62.5 ﬂours ranged from 44.64 to 64.32 mg/g wet weight basis (Table I). These results were much higher 60.0 than those reported in literature for fresh and steamed sweet potato ﬂours (Yang and Gadi 2008). 57.5 The results of multiple regression showed that the total phenolic content was signiﬁcantly affected by the 55.0 interaction term of concentration and soaking time 1.0 1.5 2.0 2.5 3.0 (X2, X3, P # 0.05). The ﬁnal predictive model for Concentration (%) phenolic content is given below: 42.68 46.12 49.56 53.00 Phenolic content (mg/g) 56.43 59.87 63.31 Y ¼ 2176:82 2 5:97X2 X3 The response surface plots in Figures 3 and 4 show the (b) 65.0 relationship between the phenolic content and drying temperatures, citric acid concentrations and soaking Temperature (°C) 62.5 times. The total phenolic content increased with increasing drying temperatures (Figure 3a,b). This 60.0 might release more bound phenolic compounds from the breakdown of cellular constituents. Huang et al. 57.5 (2006) found that steaming treatment increased the total phenolic content of purple sweet potato ﬂour. 55.0 Dewanto et al. (2002a) also found that the free 1.0 1.5 2.0 2.5 3.0 phenolic content of sweet corn increased with Soaking time (min) increasing heating temperature and time. However, Phenolic content (mg/g) 42.68 46.12 49.56 53.00 thermal processing had no effect on the phenolic 56.43 59.87 63.31 content of tomato (Dewanto et al. 2002b). On the other hand, total phenolic contents decreased with increas- Figure 4. Contour plots showing the effects of temperature, citric ing the concentration and soaking time (Figure 4a,b). acid concentration and soaking time on phenolic contents of purple This was probably because some phenolic compounds sweet potato ﬂour. (a) Temperature and citric acid concentration. (b) Temperature and soaking time. were more hydrolyzed or oxidized because dispersions were prepared in the presence of ambient oxygen. for Y is given below: Y ¼ 1240:40 2 36:84X1 2 47:49X3 þ 0:28X2 1 Optimization of pretreatments and veriﬁcation of models þ 3:12X2 þ 3:92X2 þ 4:43X2 X3 The predicted and experimental results are presented 2 3 in Table IV. For the phenolic content, the predicted Figures 1 and 2 show the response surface plots of the response surface of the stationary point was a saddle relationship between anthocyanin content and drying point. Thus the estimated surface did not have a temperatures, citric acid concentrations and soaking unique optimum. However, for the anthocyanin times. Anthocyanin contents decreased with increas- content, the predicted response surface of the ing drying temperatures (Figure 1a,b). This was as stationary point was a minimum. Therefore, different expected because heating opened the structure of optimum conditions were obtained for both responses. anthocyanin to form chalcones, which was degraded The optimal conditions for anthocyanin and phenolic further to form brown products (Delgado-Vargas et al. contents were 62.918C, 1.38%, 2.53 min, whereas for 2000). However, the anthocyanin content increased total phenolic contents they were drying temperature with increasing soaking time and concentration 60.948C, citric acid concentration 1.04% and soaking (Figure 2a,b). This might be due to interaction time 2.24 min. However, optimal conditions of the between citric acid and anthocyanin. In acidic media, anthocyanin content were not apparent. This is due to Table IV. Comparison of predicted and experimental values for the response of anthocyanin and phenolic contents. Optimum conditions Values Response variable Stationary point Temperature (8C) Soaking time (min) Soaking concentration (%) Experimental Predicted Anthocyanin Minimum 62.91 2.53 1.38 19.78 ^ 0.97 19.71 Total phenolics Saddle 60.64 1.04 2.24 61.55 ^ 2.9 52.89 96 M. Ahmed et al. the fact that the optimization point was a minimum. Fan G, Han Y, Gu Z, Chen D. 2008. Optimization conditions for The optimal value of anthocyanin content was anthocyanins extraction from purple sweet potato using response surface methodology (RSM). LWT Food Sci Technol 41: lower than expected values. This might be related to 155 –160. the anthocyanin extraction conditions by second- Huang YC, Chang YH, Shao YY. 2006. Effects of genotype and order polynomials (Fan et al. 2008). The corres- treatment on the antioxidant activity of sweet potato in Taiwan. ponding experimental responses of anthocyanin and Food Chem 98:529–538. total phenolic contents were 19.78 mg/100 g and Jiang Y. 2000. Role of anthocyanins, polyphenoloxidase and phenols 61.55 mg/g, respectively. These data showed that the in lychee pericarp browning. J Sci Food Agric 80:305–310. experimental responses were reasonably close to the Montgomery DC. 1984. Design and analysis of experiments. 2nd ed. predicted responses. New York: John Wiley and Sons. Oki T, Masuda M, Furuta S, Nishiba Y, Terahara N, Suda I. 2002. Involvement of anthocyanins and other phenolic compounds in Conclusion radical-scavenging activity of purple-ﬂeshed sweet potato The results of anthocyanin and phenolic contents were cultivars. J Food Sci 67:1752–1756. higher than the previous reported values for raw, steam Pathirana LC, Shahidi F. 2005. Optimization of extraction of phenolic compounds from wheat using response surface and kneaded sweet potato ﬂours. Therefore, treated methodology. Food Chem 93:47– 56. ﬂours could be used to make the higher quality products Proctor JTA. 1974. Colour stimulation in attached apples with that would be more attractive to product developers and supplementary light. Can J Plant Sci 54:499–503. consumers. Rumbaboa RGO, Cornago DF, Geronimo IM. 2009. Phenolic content and antioxidant capacity of Philippine sweet potato Declaration of interest: The authors report no (Ipomoea batatas) varieties. Food Chem 113:1133–1138. conﬂicts of interest. The authors alone are responsible Sing S, Raina CS, Bawa AS, Saxena DC. 2003. Optimization of for the content and writing of the paper. processing variables in the preparation of sweet potato chips using response surface methodology. Euro Food Res Technol 217:374–381. References Steed LE, Truong VD. 2008. Anthocyanin content, antioxidant Altunkaya A, Gokmen V. 2009. Effect of various anti-browning activity and selected physical properties of ﬂowable purple- agents on phenolic compounds proﬁle of fresh lettuce (L. Sativa). ﬂeshed sweet potato purees. J Food Sci 73:215–221. Food Chem 117:122–126. Swain T, Hillis WE. 1959. The phenolic constituents of prunus Cacace JE, Mazza G. 2003. Optimization of extraction of domestica. I. The quantitative analysis of phenolic constituents. anthocyanins from black currants with aqueous ethanol. J Sci J Sci Food Agric 10:63– 68. Food Agric 68:240–248. Terahara N, Konczak I, Ono H, Yoshimoto M, Yamakewa O. 2004. Cevallos-Casala BA, Cisneros-Zevallos L. 2004. Stability of Characterization of acylated anthocyanins in callus induced from anthocyanin-based aqueous extracts of Andean purple corn storage root of purple-ﬂeshed sweet potato, Ipomoea batatas L. and red-ﬂeshed sweet potato to synthetic and natural colorants. J Biomed Biotechnol 5:279–286. Food Chem 86:69–77. Varnalis AI, Brenan JG, Macdougall DB, Gilmour SG. 2004. Delgado-Vargas F, Jimenez AR, Paredes-Lopez O. 2000. Natural Optimization of high temperature pufﬁng of potato cubes using pigments: Carotenoids, anthocyanins, and betalains—character- response surface methodology. J Food Eng 61:153–163. istics, biosynthesis, processing and stability. Crit Rev Food Sci Wanasundara UN, Shahidi F. 1999. Concentration of omega Nutr 40:173–289. polyunsaturated fatty acids of seal blubber oil by urea Dewanto V, Xianzhong WU, Adom KK, Liu RH. 2002a. Thermal processing enhances the nutritional value of tomatoes by complexation: Optimization of reaction conditions. Food Chem increasing total antioxidant activity. J Sci Food Agric 50: 65:41–49. 3010–3014. Yang J, Gadi RL. 2008. Effects of steaming and dehydration Dewanto V, Xianzhong WU, Adom KK, Liu RH. 2002b. Processed on anthocyanins, antioxidant activity, total phenols and sweet corn has higher antioxidant activity. J Sci Food Agric 50: color characteristics of purple-ﬂeshed sweet potatoes 4959–4964. (Ipomoea batatas). Am J Food Technol 3:224– 234. Copyright of International Journal of Food Sciences & Nutrition is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.