Extraction optimization of watermelon seed protein using response surface methodology

Extraction optimization of watermelon seed protein using response surface methodology

ARTICLE IN PRESS LWT 41 (2008) 1514–1520 www.elsevier.com/locate/lwt Extraction optimization of watermelon seed protein using response surface metho...

419KB Sizes 0 Downloads 34 Views

ARTICLE IN PRESS

LWT 41 (2008) 1514–1520 www.elsevier.com/locate/lwt

Extraction optimization of watermelon seed protein using response surface methodology Ali Abas Wani, Devinder Kaur, Idrees Ahmed, D.S. Sogi Department of Food Science and Technology, Guru Nanak Dev University, Amritsar 143005, India Received 13 October 2006; received in revised form 28 September 2007; accepted 8 October 2007

Abstract Extraction conditions for maximum protein recovery from watermelon (Citrullus vulgaris Cv Sugar baby) seed meal were investigated using response surface methodology. A central composite design with four independent variables: temperature (40, 45, 50, 55 and 60 1C); NaOH concentration (0.03, 0.06, 0.09, 0.12 and 0.15 g/L); extraction time (5, 10, 15, 20 and 25 min) and solvent/meal ratio (30:1, 40:1, 50:1, 60:1 and 70:1, v/w) was used to study the response variable (protein yield). The experimental values of protein yield ranged between 72.03 and 81.52 g/100 g seed meal. A second-degree equation for independent and response variables was computed and used to create the contour plots graphs. The predicted values obtained for protein yield revealed that coefficient of determination and standard error was 0.80 and 0.906, respectively. Optimum protein extraction was obtained with 0.12 g/L alkali concentration, 15 min extraction time and 70:1 (v/w) solvent/meal ratio at 50 1C. Confirmatory studies revealed that the protein yield under optimum conditions was 80.71 g/100 g seed meal. r 2007 Swiss Society of Food Science and Technology. Published by Elsevier Ltd. All rights reserved. Keywords: Watermelon seed meal; Protein extraction; Contour plots; Process optimization

1. Introduction Food industry produces large volumes of solids and liquid wastes, resulting from the production, preparation and consumption. These wastes pose potential disposal and pollution problems along with loss of valuable biomass and nutrients. There is a potential for conversion of food processing wastes into useful products or even as raw material for other industries. The utilization of wastes of fruit and vegetable processing as a source of functional ingredients is a promising field (Schieber, Stintzing, & Carle, 2001). Morever, due to stringent national international regulations, the waste management has become inevitable. Food industry is going to face great challenges, such as to minimize wastes arising from processing operations and to utilize the by-products and to treat and dispose the wastes along with sustainable production. Corresponding author. Tel.: +91 183 2258802–08; fax: +91 183 2258820. E-mail address: [email protected] (D.S. Sogi).

Bioconversion of food processing residues is receiving increased attention regarding the fact that these residual matters represent a possible and utilizable resource for conversion into useful products (Martin, 1998). Watermelon is an important crop grown in the warmer regions of the world. Watermelon is utilized for the production of juices, nectars and fruit cocktails, etc. (Ahmed, 1996; Bawa & Bains, 1977; Hour, Ahmed, & Carter, 1980) whereas major by-product ‘‘the rind’’ is utilized for the products like pickle, preserve, pectin, etc. (Godawa & Jalali, 1995; Hasan, 1993). Watermelon seed have been reported to be high in protein (Kamel, Dawson, & Kakuda, 1985; Lasztity, Samei, & Shafei, 1986; Sharma, Lal, Madaan, & Chattarjee, 1986; Teotia & Ramakrishna, 1984) and lipid contents (El-Adaway & Taha, 2001; Lazos, 1986). Arginine, glutamic acid, aspartic acid and leucine were found to be predominant amino acids in the proteins (El-Adaway & Taha, 2001; Lasztity et al., 1986). However, watermelon proteins are the major components of the deoiled meal (El-Adaway & Taha, 2001), remains underutilized, as procedures for its extraction are

0023-6438/$34.00 r 2007 Swiss Society of Food Science and Technology. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.lwt.2007.10.001

ARTICLE IN PRESS A.A. Wani et al. / LWT - Food Science and Technology 41 (2008) 1514–1520

complicated and not commercially viable. Factors including pH, temperature, ionic strength, solvent type, extraction time and solid–liquid ratio have an effect on melon seed meal, linseed meal, flaxseed meal and pigeon pea (Dev, Quensel, & Hansen, 1986; Khalil, 1998; Liu, 1997; Oomah, Mazza, & Cui, 1994). Based on end use requirements, various extractions, isolation and fractionation procedures are followed. Generally, the extraction of protein rich material in alkaline solution followed by isoelectric precipitation is commonly followed for food applications (Sathe, Deshpande, & Salunkhe, 1984). Response surface methodology (RSM) is a statistical technique that helps us in getting information with less cost and short time. This technique relates input and output parameters (Montgomery, 1984). Its use leads to rapid and efficient development of new/improved products or processes. The objectives were to maximize protein yield of watermelon seed meal proteins under practical operating conditions of pH, meal/solvent ratio, alkali concentration and to predict optimum conditions for maximum solubility and extraction of proteins from watermelon seed meal. 2. Material and methods 2.1. Material Certified watermelon seeds of cultivar Sugar baby were procured from Punjab Agriculture University, Ludhiana, India.

1515

The model proposed for response (Y) was Y ¼ b0 þ

4 X n¼1

bn X n þ

4 X n¼1

bnn X 2n þ

4 X

bnm X n Xm ,

(1)

nhm

where b0 is the value for the fixed response at the central point of the experiment, bn, bm, bnn and bnm are the linear, quadratic and cross product coefficients, respectively. 2.5. Protein extraction Watermelon seed meal (10 g) was extracted for protein with different levels of independent variables (Temperature 40–60 1C, NaOH 0.03–0.15 g/L, time 5–25 min and solvent/ meal ratio 30:1–70:1, v/w). The protein extraction was carried out with alkali solution in water-jacketed bottles by connecting it to a water bath (TC 500; Brook Field, USA). The solution was continuously stirred with a magnetic stirrer for a selected period. The solution was immediately centrifuged in a cooling centrifuge at 10000  g for 15 min at 4 1C. The supernatant was filtered through Whatman filter paper # 1 and the soluble protein content was determined (Lowry, Rosebrough, Farr, & Randal, 1951). Transformation of coded variable (X1) level to uncoded variable (x1) level could be obtained from x1 ¼ 5(X1)+50, x2 ¼ 0.3(X2)+0.9, x3 ¼ 5(X3)+15, x4 ¼ 10(X4)+50. The duplicate experiments were repeated twice. 2.6. Statistical analysis

Seeds were dehulled, ground using hammer mill (M/S Narang Scientific Works, India), extracted with n-hexane three times (n-hexane ratio: flour 10:1, v/w), desolventized and ground again to pass through 72 mesh sieve to obtain fine powder, termed as deoiled meal and was stored at 20 1C till use. The experiment was carried out in duplicate.

Average values and standard deviation was computed for proximate composition. A second-order polynomial was fitted to the mean data values to obtain regression equations and statistical significance of its terms was examined using a commercial statistical package Mini Tab11.12 (Mini Tab Inc., USA). The experimental and computed values were analyzed for coefficient of determination (R2), standard error and scattered plot. Extractions conditions were optimized using contour plots for two independent parameters while fixing remaining two at coded zero levels.

2.3. Proximate analysis

3. Results and discussion

Moisture, crude fat, crude protein, crude fibre and ash content of watermelon seed, kernel and meal were determined following standard method of analysis (AOAC, 1990). Carbohydrate was determined by subtracting all constituents from hundred.

3.1. Proximate analysis

2.2. Preparation of defatted seed meal

2.4. Experimental design The effect of four independent variables X1 (temperature), X2 (NaOH concentration), X3 (extraction time) and X4 (solvent/meal ratio) at five levels on protein yield (dependent variable) were investigated using central composite design and RSM.

The proximate analysis of seeds, kernels and deoiled meal of watermelon revealed that seeds, kernels and meal contained 16.34, 40.46 and 56.15 g/100 g crude protein, respectively (Table 1). Seeds showed very high content of fibre due to fibrous seed coat. Removal of seed coat resulted in increase in protein and lipid content of kernels. Earlier studies show that seeds contained 4.85–9.2 g/100 g moisture, 22.2–28.2 g/100 g fat, 18.98–31.02 g/100 g, 32.9–62.4 g/100 g crude fibre, 2.3–4.2 g/100 g ash and 12.3–27.6 g/100 g carbohydrate (by difference) whereas kernels contained 2.7–9.5 g/ 100 g moisture, 44.4–56.0 g/100 g fat, 18.98–51.5 g/100 g,

ARTICLE IN PRESS A.A. Wani et al. / LWT - Food Science and Technology 41 (2008) 1514–1520

1516

1.7–7.9 g/100 g crude fibre, 0.8–4.99 g/100 g ash and 2.6–5.6 g/100 g carbohydrate (by difference) (Akpapunam & Markakis, 1981; El-Adaway & Taha, 2001; Lasztity et al., 1986; Sharma et al., 1986; Teotia & Ramakrishna, 1984). Present studies showed that proximate composition of seeds and kernels was within the range of the earlier reported values. It indicated that the watermelon seeds used

Table 1 Proximate composition of watermelon (Citrullus vulgaris Cv Sugar baby) seed, kernel and meal (n ¼ 3). Parameters (g/100 g)

Seed

Kernel

Meal

Moisture Crude fat Crude protein (N  6.25) Crude fiber Ash Carbohydrate (by difference)

7.6770.23 21.9370.61 16.3470.59 22.2170.37 2.4870.40 26.31

4.0270.25 38.88723.75 40.4670.14 2.6270.18 3.7170.07 10.31

5.1671.13 0.9370.35 56.1371.32 4.5170.72 3.7670.57 29.51

for this study are comparable with those used in other studies. 3.2. Protein extraction Mean experimental protein yield values from 31 selected combinations of the independent variables, varied from 72.03 to 81.52 g/100 g seed meal (Table 2). Independent and dependent variable were analyzed to get regression equation that could predict the response under the given range. The regression equation obtained for protein yield (Y) was as follows: Y ¼ 177:140  3:250X 1 þ 3:120X 2 þ 0:106X 3  0:974X 4 þ 0:029X 21 þ 4:516X 22  0:029X 23 þ 0:006X 24  0:119X 1 X 2 þ 0:00X 1 X 3 þ 0:009X 1 X 4 þ 0:169X 2 X 3  0:102X 2 X 4 þ 0:006X 3 X 4 .

ð2Þ

Table 2 Central composite arrangement for independent variables X1 (temperature, 1C), X2 (NaOH, g/L), X3 (time, min), X4 (solvent/meal ratio, v/w) and their response (protein yield, g/100 g) (n ¼ 3) Run

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

Variables levels (uncoded)

Protein yield (g/100 g)

X1

X2

X3

X4

Experimental

Temperature (1C)

NaOH (g/L)

Time (min)

Solvent/meal ratio (v/w)

P1

P2

P3

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 0 0 0 0 0 0 0 0 0 0 0 0 0

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 2 2 0 0 0 0 0 0 0 0 0 0 0

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 2 2 0 0 0 0 0 0 0 0 0

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 2 2 0 0 0 0 0 0 0

77.96 76.97 77.52 75.98 78.37 78.00 78.95 75.82 77.41 72.88 78.12 74.01 77.48 76.11 78.69 77.35 79.45 78.31 77.76 77.30 72.92 71.76 81.13 75.34 75.39 76.17 76.95 76.03 75.13 75.87 74.72

78.41 77.77 77.80 76.38 78.82 78.20 79.40 76.20 77.53 73.35 78.60 74.35 77.91 76.23 78.91 77.41 79.58 78.47 77.97 77.43 73.67 72.21 81.75 75.80 75.97 76.77 77.83 76.60 75.73 76.57 75.40

78.56 77.67 78.44 76.20 79.42 78.25 79.70 76.23 77.92 73.19 78.54 75.21 78.72 77.13 79.68 77.92 79.92 79.13 78.66 78.15 73.69 72.12 81.68 75.72 76.13 76.85 77.60 76.75 75.91 76.52 75.51

(55) (55) (45) (45) (55) (55) (45) (45) (55) (55) (45) (45) (55) (55) (45) (45) (60) (40) (50) (50) (50) (50) (50) (50) (50) (50) (50) (50) (50) (50) (50)

(0.12) (0.06) (0.12) (0.06) (0.12) (0.06) (0.12) (0.06) (0.12) (0.06) (0.12) (0.06) (0.12) (0.06) (0.12) (0.06) (0.09) (0.09) (0.15) (0.03) (0.09) (0.09) (0.09) (0.09) (0.09) (0.09) (0.09) (0.09) (0.09) (0.09) (0.09)

(20) (20) (20) (20) (10) (10) (10) (10) (20) (20) (20) (20) (10) (10) (0) (10) (15) (15) (15) (15) (25) (05) (15) (15) (15) (15) (15) (15) (15) (15) (15)

P1, P2 and P3 are replicates of experimental protein yield.

(60) (60) (60) (60) (60) (60) (60) (60) (40) (40) (40) (40) (40) (40) (40) (40) (50) (50) (50) (50) (50) (50) (70) (30) (50) (50) (50) (50) (50) (50) (50)

Predicted

78.91 77.81 78.23 76.41 79.04 78.91 78.31 77.53 76.35 74.03 77.57 74.52 77.71 76.36 78.93 76.87 79.18 79.02 79.42 76.25 72.68 72.68 80.13 76.92 76.22 76.21 76.21 76.21 76.21 76.21 76.21

ARTICLE IN PRESS A.A. Wani et al. / LWT - Food Science and Technology 41 (2008) 1514–1520

0.150

Table 3 Analysis of variance of independent variables for extraction optimization of watermelon seed protein Degree of freedom

F-value

Probability

Regression Linear Square Interaction Lack of fit Pure error Total error

14 4 4 6 9 7 30

5.95 4.29 6.68 1.08 2.83

0.001 0.015 0.002 0.414 0.092

The predicted values of protein yields were calculated using regression model and compared with experimental values. Coefficient of determination (R2) was 83.9% and the standard error was 0.906, which indicates the adequacy of the applied model. Earlier studies have reported R2 ranging from 71.00% to 95.20% (Mizubuti, Junior, Souza, Silva, & Ida, 2000; Sogi, Arora, Garg, & Bawa, 2003; Wanasundara & Shahidi, 1996) for flaxseed, pigeon pea and tomato seed. The statistical analysis of the coefficients of the model revealed that linear and quadratic terms were significant while the interaction coefficients were nonsignificant (Table 3). It indicated that independent variables individually affected the response variable. The analysis of variance also showed that there was a nonsignificant lack of fit that further validates the model.

81.0

81.0 78.0

0.125 79.5 NaOH (g/L)

Source

0.100

0.075

0.050 76.5 40

45

25

55

74.2

60

73.0 76.6

76.6

Mixing time (mins)

The optimum conditions could be selected using surface graphs, contour plots, steepest ascent techniques (Liadakis, Tiza, Oreopoulou, & Thomopoulos, 1995; Thakur & Saxena, 2000; Wanasundara & Shahidi, 1996; Wani, Sogi, Grover, & Saxena, 2006). However, in the current study contour plots were employed and effect of two independent variables out of the four on protein recovery was plotted while remaining two were held at zero level. Variation in temperature and alkali concentration revealed that maximum protein extraction was obtained when NaOH was 0.15 g/L and temperature was 40 1C while the extraction time and solvent/meal ratio were 15 min and 50:1 (v/w), respectively (Fig. 1). Increase in temperature did not show significant effect on extracted protein while increase in alkali concentration resulted in increase in protein recovery. Previous reports on production of melon seed kernel protein isolates revealed that increase in alkali concentration from 0.025 to 0.1 mol equivalent/L NaOH significantly (Pp0.05) increased protein solubility from 59.8 to 83.1 g/100 g seed meal (Khalil, 1998). Similar reports are available on extraction optimization of tomato seed meal, winged bean protein (Sogi et al., 2003). Variation in temperature and extraction time revealed that protein yield was maximum when temperature was 40 1C and extraction time was 15 min while alkali concentration and solvent/meal ratio were kept 0.09 g/L and 50:1 (v/w),

50 Temperature (°C)

Fig. 1. Effect of temperature and NaOH concentration on protein yield extracted from watermelon seed meal having 50:1 (v/w) solvent/meal ratio and 15 min extraction time.

20

3.3. Process optimization

1517

75.4

15

10

5

77.8

77.8 74.2

75.4 40

45

50

55

60

Temperature (°C) Fig. 2. Effect of temperature and extraction time on protein yield extracted from watermelon seed meal having 50:1 (v/w) solvent/meal ratio and 0.9% NaOH concentration.

respectively (Fig. 2). Previous studies on extraction of melon seed kernel protein isolates (Khalil, 1998), revealed a significant increase in protein yield up to 1:30 ratio which is lower than the present study; however, studies on protein extraction of flaxseed and deoiled tomato seed meal are in agreement with the present study (Oomah et al., 1994; Sogi et al., 2003). The difference in the reported results may be due to differences in the plant material, type of equipments used and other agricultural practices during

ARTICLE IN PRESS A.A. Wani et al. / LWT - Food Science and Technology 41 (2008) 1514–1520

1518

70

25

81.0

73.5

72.0

76.5

78.0

60

20

Mixing time (mins)

Solvent/meal ratio (v/w)

82.5

79.5 50

40

30

75.0

15

78.0

10

76.5

79.5 40

45

50

55

75.0

5

60

0.050

0.075

Temperature (°C)

growing conditions. Maximum protein extraction was when temperature was at 40 1C and the solvent/meal ratio was 70:1 (v/w) while NaOH concentration and extraction time were kept at 0.09 g/L and 15 min, respectively (Fig. 3). Khalil (1998) also reported a significant (pp0.05) increase in protein yield from 59.8 to 83.4 g/100 g seed meal from 0.025 to 0.1 mol equivalent/L, further increase in alkali concentration did not significantly (pp0.05) increased protein yield. Variation in alkali concentration and extraction time revealed that maximum protein was obtained when extraction time was 15 min and alkali concentration was 0.12 g/L while temperature and solvent/meal ratio were kept at 50 1C and 50:1 (v/w), respectively (Fig. 4). Extraction time did not seem to affect protein extraction in selected range. Similar reports on effect of extraction time on protein yield have been reported for deoiled tomato seed meal, chickpea, flaxseed and pigeon pea protein extraction (Mizubuti et al., 2000; Oomah et al., 1994; Sogi et al., 2003). A maximum protein could be extracted at 70:1 solvent/meal ratio and 0.03 g/L NaOH concentration, while temperature and extraction time were kept at 50 1C and 15 min, respectively (Fig. 5). The results indicated that with the increase in NaOH concentration and solvent/ meal ratio protein recovery from seed meal was decreased significantly. Maximum protein extraction was obtained at 70:1 solvent/meal ratio and 15 min extraction time, when the temperature and NaOH concentration were fixed at 50 1C and 0.09 g/L, respectively (Fig. 6). Considering all the responses, it is evident that alkali concentration and solvent/meal ratio have significant effect while temperature and extraction time have slight effect on protein yield. Studies on the extraction of watermelon seed meal protein could not be traced; however, similar studies using

0.125

0.150

Fig. 4. Effect of NaOH concentration and extraction time on protein yield extracted from watermelon seed meal at 50 1C and having 50:1 (v/w) solvent/meal ratio.

70

80.8

79.6 80.8

Solvent/meal ratio (v/w)

Fig. 3. Effect of temperature and solvent meal ratio on protein yield extracted from watermelon seed meal having 0.9% NaOH concentration and 15 min extraction time.

0.100

NaOH (g/L)

60

77.2

50

40

30

78.4

76.0 0.050

0.075 0.100 NaOH (g/L)

0.125

0.150

Fig. 5. Effect of NaOH concentration and solvent/meal ratio on protein yield extracted from watermelon seed meal at 501C and 15 min extraction time.

other plant material have been reported. Rustom, LopezLeiva, and Nair (1991) worked on protein extraction from peanuts (Arachis hypogea L.) with water using RSM and found significant effects of time, temperature, pH and liquid/solid ratio and concluded that optimum extraction conditions were pH 8.0, time 30 min, temperature 50 1C and liquid/solid ratio 8:1. Sogi et al. (2003) studied extraction of tomato seed protein using RSM and concluded that maximum yield was obtained at NaOH 0.12 g/L, 70-solvent/meal ratio at 60 1C for 5 min. Other studies on winged bean, tomato seeds, flax seeds and pigeon pea proteins are in agreement with the present study that with the increase in solvent/meal ratio and pH resulted

ARTICLE IN PRESS A.A. Wani et al. / LWT - Food Science and Technology 41 (2008) 1514–1520

70

79.5

76.5

78.0

Solvent/meal ratio (v/w)

60

73.5 50

40 75.0

76.5 75.0

30 5

10

15

20

25

Mixing time (mins) Fig. 6. Effect of extraction time and solvent/meal ratio on protein yield extracted from watermelon seed meal at 501C and having 0.9% NaOH concentration.

in higher protein yield (Liadakis et al., 1995; Mizubuti et al., 2000; Wanasundara & Shahidi, 1996). 3.4. Confirmatory studies The experiment was run at the optimum conditions obtained from the above study. The experimental protein yield at the optimum level was 80.71 g/100 g seed meal. The calculated amount of protein yield with these parameters using regression model was 81.52 g/100 g seed meal. It confirmed that these conditions were optimal for protein extraction. It was observed that experimental optimal value was lower than computed value by the regression model. Earlier studies on protein extraction have also demonstrated such pattern (Wani et al., 2006). 4. Conclusions Experimental protein yield was 72.03–81.52 g/100 g seed meal following 31 selected combinations of temperature, NaOH concentration, extraction time and solvent/meal ratio. The model developed for protein yield exhibited nonsignificant lack of fit and an R2 value of 80.7%. The surface graphs indicated that maximum protein yield could be obtained by extracting seed meal with 0.12 g NaOH/L, 70:1 (v/w) solvent/meal ratio, 15 min extraction time at 40 1C. Under optimal condition, experiment yield was lower than predicted by second-order model. References Ahmed, J. (1996). Studies on watermelon products. Indian Food Packer, 50, 15–20.

1519

Akpapunam, M. A., & Markakis, P. (1981). Protein supplementation of cowpeas with sesame and watermelon seeds. Journal of Food Science, 46, 960–961. AOAC. (1990). Official methods of analysis of the association of official analytical chemists. Virginia, USA: AOAC Inc. Bawa, A. S., & Bains, G. S. (1977). Integrated processing of watermelons for juice and seed. Indian Food Packer, 31(6), 12. Dev, D. K., Quensel, E., & Hansen, R. (1986). Nitrogen extractability and buffer capacity of defatted linseed meal (Linum usitatissum L.) flour. Journal of the Science of Food and Agriculture, 19, 1999. El-Adaway, T. A., & Taha, K. M. (2001). Characteristics and composition of different seed oils and flours. Food Chemistry, 74, 47–54. Godawa, I. N. D., & Jalali, M. (1995). Studies on juice making from watermelon fruits. Indian Food Packer, 49(3), 33–41. Hasan, F. (1993). Research on the possibility of using watermelon juice in the fruit juice cocktails. Gida, 18(6), 369–371. Hour, S. S., Ahmed, E. M., & Carter, R. D. (1980). Concentration of watermelon juice. Journal of Food Science, 45, 718–719. Kamel, B. S., Dawson, H., & Kakuda, Y. (1985). Characteristics and composition of melon and grape seed oils and cakes. Journal of American Oil Chemical Society, 62(5), 881–883. Khalil, M. M. (1998). Factors affecting production of meon seed kernel protein: Yield, composition and protein isolates quality. Die Nahrung, 42(5), 295–297. Lasztity, R., Samei, M. B. A. E., & Shafei, M. E. (1986). Biochemical studies on some non-conventional sources of protein. Part 2. Cucurbutaceae seeds. Die Nahrung, 30(6), 621–627. Lazos, E. (1986). Nutritional, fatty acid and oil characteristics of pumpkin and melon seeds. Journal of Food Science, 51, 1382–1383. Liadakis, G. N., Tiza, C., Oreopoulou, V., & Thomopoulos, C. D. (1995). Protein isolation from tomato seed meal-extraction optimisation. Journal of Food Science, 60, 477–482. Liu, K. (1997). Soybeans: Chemistry, technology and utilisation. New York: International Thompson Publisher. Lowry, O. H., Rosebrough, N. J., Farr, A. L., & Randal, R. J. (1951). Protein analysis with folin-phenol reagent. Journal of Biological Chemistry, 193, 265–275. Martin, A. M. (1998). Bioconversion of waste materials to industrial products (2nd ed.). London: Blackie Academic and Professional. Mizubuti, I. Y., Junior, O. B., Souza, L. W. O., Silva, R. S. S. F., & Ida, E. L. (2000). Response surface methodology for extraction optimization of pigeon pea protein. Food Chemistry, 70, 259–265. Montgomery, D. C. (1984). Design and analysis of experiments. Singapore: Wiley. Oomah, B. D., Mazza, G., & Cui, W. (1994). Optimization of protein extraction from flaxseed meal. Food Research International, 27, 355–361. Rustom, I. Y. S., Lopez-Leiva, M. H., & Nair, B. M. (1991). Optimisation of extraction of peanut proteins with water by response surface methodology. Journal of Food Science, 56(6), 1660–1663. Sathe, S. K., Deshpande, S. S., & Salunkhe, D. K. (1984). Dry beans of Phaseolus. A review. Part 1. Chemical composition: Protein. Critical Reviews in Food Science and Nutrition, 20(1), 1–46. Schieber, A., Stintzing, F. C., & Carle, R. (2001). By-products of plant processing as a source of functional compounds—Recent developments. Trends in Food Science and Technology, 12(11), 401–413. Sharma, P., Lal, B. M., Madaan, T. R., & Chattarjee, S. R. M. (1986). Studies on the nutritional quality of some cucurbit kernel proteins. Journal of the Science of Food and Agriculture, 37, 418–420. Sogi, D. S., Arora, M. S., Garg, S. K., & Bawa, A. S. (2003). Response surface methodology for the optimisation of tomato seed protein. Journal of Food Science and Technology, 40(3), 267–271. Teotia, M. S., & Ramakrishna, P. (1984). Chemistry and technology of melon seeds. Journal of Food Science and Technology, 21, 332–340.

ARTICLE IN PRESS 1520

A.A. Wani et al. / LWT - Food Science and Technology 41 (2008) 1514–1520

Thakur, S., & Saxena, D. C. (2000). Formulation of extruded snack food (gum-based-cereal-pulse blend): Optimisation of ingredients levels using response surface methodology. Lebensm-Wiss.u-Technology, 33, 354–361. Wanasundara, P. K. J. P. D., & Shahidi, F. (1996). Optimisation of hexametaphosphate assisted extraction of flaxseed proteins

using response surface methodology. Journal of Food Science, 61, 604–607. Wani, A. A., Sogi, D. S., Grover, L., & Saxena, D. C. (2006). Effect of temperature, alkali concentration, mixing time and meal solvent ratio on the extraction of watermelon seed proteins—Response surface approach. Biosystems Engineering, 94(1), 67–73.