Conference proceedings of Biotechnology for Green Solutions and Sustainable Environment: 19-25, 2010

Transcription

1 OPTIMIZATION CAROTENOIDS PRODUCTION BY RHODOTORULA GLUTINIS GBDOU USING DESIGN OF PLACKETT BURMAN MATRIX AND CENTRAL COMPOSITE DESIGN - RESPONSE SURFACE METHODOLOGY Bui Dinh Hanh Dung 1, Trinh Thi Bich Huyen, Bui Hong Quan 3,* 1 Ho Chi Minh University of Agriculture and Forestry, Viet Nam Faculty of Environment, Ho Chi Minh City University of Technology, Viet Nam. 3 Institute of Biotechnology and Food Technology, Ho Chi Minh University of Industry, Viet Nam. Tel: buihongquan@hui.edu.vn/buihongquan@gbd.edu.vn ABSTRACT The yeast Rhodotorula glutinis GBDOU was optimized for maximum carotenoids production. The design of optimum multifactorial experiments Plackett-Burman was employed to estimate level effect of physical and chemical factors on carotenoids production. As the result, saccharose (%), yeast extract (%) and cooking oil (%) were identified as significant factors (p<0.05). After screening, these factors were subsequently optimized using the response surface methodology (RSM) - Central Composite Design (CCD). These optimal levels were found out, viz. saccharose (1.5 %w/v), yeast extract (4.69 %w/v) and cooking oil (0.5 %v/v) in which the carotenoids yield was the highest. The regression equations (model) obtained and predicted maximum carotenoids yield (µg/g DW). Key words: Carotenoid, Rhodotorula glutinis GBDOU, Plackett -Burman, Response Surface Methodology- Central Composite Design. INTRODUCTION Carotenoids are the largest organic pigments that are naturally found in many plants and some microorganisms. According to Sutter and Whitaker (1981), carotenoids are divided into two main groups of pigments: (1) the group of hydrocarbon carotene such as β-carotene, torulen; () oxidized xanthophyll as astaxanthin and torularhodin. The yeast R.glutinis is one of the very few types of yeast which is able to synthesize a large number of carotenoid pigments, which are mainly β-carotene, torulene, torularhodin (Frengova & Beshkova, 009; Hayman, 1974). The main carotenoids were studied from the same yeast Rhodotorula sp., including both groups, in which torulen, β-carotene is in the first group, and torularhodin belongs to the group of xanthophyll (Bong, Byoung, Hee & Eui, 004). All over the world, R.glutinis is studied in certain countries such as India, Canada, France, Russia (Lampila, 1985). In studies, Park and his colleagues (005) cultured R.glutinis on sugarcane molasses using response surface methodology (Park, Cho, Kim & Chu, 005). Cells mixed with DMSO, acetone, petroleum ether, 0% NaCl solution was used to extract carotenoids. This result is the highest carotenoids amount of 3.46 mg/l in liquid medium (Park et al., 005). To increase the production of β-carotene, Bhosale & Grade (001a) the mutant from NCIM 3353 R.glutinis treated with UV light was cultured in sea water. By this method, carotenoids obtained were 86 mg/l. In distilled water, meanwhile, it was only 70 mg/l. This resulted in 3-fold increase in β-carotene and.3-fold decrease in torulen. This result is interesting because β-carotene is more useful than torulen. Sqina and her colleagues optimized carotenoids from R.glutinis in a different culturing medium. The medium was supplemented with sugarcane juice as a carbon source in order to gain the highest amount of total carotenoids of 497. µg/g dry cell biomass (Sqina, Yamashita, Pereira & Mercadante, 00). Buzzini (001) also optimized carotenoids production from R.glutinis 3853 DBVPG yeast in grape juice as carbon source. By this way, the carotenoids yield reached the highest value of 6.9 mg/l and the amount of β-carotene was 1100 µg/l (Buzzini, 001). Ho Chi Minh City University of Industry, Viet Nam 19

2 Bui Dinh Hanh Dung et al. Optimizing the fermentation process to build a model to obtain greater yields and increased scale of production has been playing a crucial role in applying basic research in industrial carotenoids synthesis. Plackett-Burman method is implementing effect, low cost, allowing the study of the interaction and simultaneously predicting the optimal values of the elements-optimal experimental design. This method has been introduced (Plackett & Burman, 1946; Dennis, 1995). The matrix has been widely using to screen medium components in shaking flask culture conditions. After the initial screening step, the optimal experimental Response surface methodology central composite design (RSM-CCD) is used to optimize the value of the factors being studied. In present study, carotenoids have been optimized physical and chemical factors in the process of culturing by using the Plackett-Burman matrix and RSM-CCD to obtain the maximum carotenoids production from culture yeast R.glutinis GBDOU. MATERIALS AND METHODS Microorganism and culture conditions The yeast R.glutinis GBDOU was obtained from Global Biotechnology Development Company Limited (GBD Co., Ltd.) both on Sabouraund slant agar at 13 o C and in glycerol at -0/-80 o C. The yeast was cultured in sabouraund broth for inoculums at room temperature (RT). The culture was maintained in both on Sabouraund slant agar at 13 o C and in glycerol at -0/-80 o C. Extraction of carotenoids Solvents used to extract carotenoids are a mixture acetonitrile, -propanol, ethyl acetate (40:40:0, v/v) (Bhosale & Grade, 001a). Quantification of carotenoids The amount of carotenoids was determined as described previously (Vijayalakshmi, Shobha, Vanajakshi, Divakar & Manohar, 001). Convertible extract color is measured on the spectrophotometer at 460 nm. Carotenoids content determined by the formula: Production of carotenoids (µg/g) = A 460 X / Y Where A 460 : the value of the sample optical absorption at 460 nm wavelength, X (ml): volume of the sample, Y (g): dry weight biomass. Plackett-Burman design and RSM-CCD method To determine the factors and estimate level effect to biosynthesis of carotenoids R.glutinis GBDOU yeast, 11 factors were selected as glucose, saccharose, yeast extract, peptone, urea, NaCl, KH PO4, MgSO 4, cooking oil, ph, and inoculum s size for experiments. The experiment was designed according to Plackett-Burman matrix (Plackett & Burman, 1946; Dennis, 1995) with 11 elements in 1 experiments (Table ) for screening of important factors affecting the production of carotenoids (µg/g). Low (-1) and high (+1) of the 11 factors was listed in Table 1. Three major factors determining the RSM-CCD was optimized and the value was studied in five levels (- α, -1, 0, +1, + α) (Table 3) in the CCD 0 experiments (Table 4) (Castillo, 007). Response function was chosen as the carotenoids yield (Y, µg carotenoids/g dry weight biomass). Modeling is represented by equation : Y = b o + b 1 x 1 + b x + b 3 x 3 + b 11 x 1 + b x + b 33 x 3 + b 1 x 1 x + b 3 x x 3 + b 13 x 1 x 3 where Y was the predict response; b 0 intercept; b 1, b, b 3 linear coefficients; b 1, b 3, b 13 interaction coefficients; b 11, b, b 33 squared coefficients; x 1, x, x 3, x 11, x, x 33, x 1, x 3, x 13 are levels of the factors. The Ho Chi Minh City University of Industry, Viet Nam 0

3 Design expert gave the response curves and they were used for determining the optimum level of the factors for maximal carotenoids production. Table 1. Ranges of the factors studied in the Plackett-Burman and effect estimates for carotenoids production from the results of the Placket Burman design Coded factors Name Lower level (-1) Higher level (+1) Effect Prob > F X 1 Saccharose (%w/v) (a) X Glucose (%w/v) (b) X 3 Yeast extract (%w/v) (a) X 4 Peptone (%w/v) (b) X 5 Urea (%w/v) 0,1 0, (b) X 6 KH PO 4 (%w/v) 0,1 1, (a) 0.07 X 7 MgSO 4 (%w/v) 0,01 0, (b) X 8 Inoculum s size (%v/v) (b) X 9 ph (b) X 10 Cooking oil (%v/v) 0 0, (a) X 11 NaCl (%w/v) 0 0, (b) (a): significant (p<0,05); (b): not significant (p>0,05) Table. The Plackett Burman experimental design matrix for screening medium compositions of carotenoids production by R.glutinis GBDOU Run Factor Carotenoids (µg/g DW) X 1 X X 3 X 4 X 5 X 6 X 7 X 8 X 9 X 10 X 11 Observed Predicted Table 3 Concentrations of three factors used in RSM CCD Factor and name Range studied Levels -α α x 1 : Sucrose (%w/v) x : Yeast extract (%w/v) x 3: Cooking oil (%v/v) Ho Chi Minh City University of Industry, Viet Nam 1

4 Bui Dinh Hanh Dung et al. Table 4 Experimental plan for optimization of carotenoids production using response surface methodology Run Coded value Carotenoids (µg/g DW) x 1 x x 3 Observed Predicted α α α α α α RESULTS Yeast was grown in 100 ml in center point environmental Sabouraund 50 ml at room temperature with shaking mode. Screening of important factors affecting the production of carotenoids R.glutinis GBDOU Plackett-Burman matrix derived carotenoids production from 74.5 to µg/g dry weight biomass (Table ). The value of each factor influencing carotenoids production was calculated by Design Expert software (Table 1). What elements of value and large positive effects will affect carotenoids production R.glutinis GBDOU. Three factors affect the value and positive large affect carotenoids production at the 95% confidence level (p<0.05) are: saccharose, yeast extract, cooking oil. Ho Chi Minh City University of Industry, Viet Nam

5 Fig. 1: Response surface carotenoid yield obtained by a RMS- CCD with two variables (saccharose and yeast extract) Optimizing the value of factors for maximum carotenoids production After screening the main factors affecting the production of carotenoids, experimental planning is carried out by RSM-CCD. The result was analyzed by Design expert The response value functions of the experiments and model predictions were presented in Table 4. After analysis of variance (ANOVA), regression equations were used as a model to predict carotenoids yield obtained. Production of carotenoids can be predicted from the model: y = x x 34.75x 1 x Where y is the production of carotenoids (µg/g DW); x 1 : saccharose, x : yeast extract. Regression coefficient (R ) was calculated as This represents 64.15% of data that is compatible with experimental data in the model predictions. The three dimensional response surface curve was plotted by statistically significant model to understand the interaction of the medium components and the optimum concentration of each component required for optimum carotenoids production. Analysis of variance (ANOVA) showed that the factor x 3 (cooking oil) was insignificance and x 1 (sucrose), x (yeast extract) and x 1 x were significant model terms. Therefore, the cooking oil was taken at higher level (0.5 %v/v). The interactive effect of two variables (saccharose and yeast extract) at constant cooking oil was depicted in Fig.1. Response surface (Fig. 1) production of carotenoids showing the interaction of each factor and from this graph we can determine the optimal value of each factor makes to reach maximal response levels. From the results obtained from experiments with ANOVA analysis found out the regression equation. We carried out to solve the problem by optimization tab of Design Expert to find out the optimal value of the factors as first aim. The program DE brought out 8 solutions. Solution No. 1 to No. 11 from the input data for research within the carotenoids yield reached the maximum value ( µg/g DW). The solutions to achieve maximum carotenoids output corresponded with the high ratio yeast extract (5%) but relatively high costs for raw materials. The nd solution was selected base on low yeast extract rate and the possible highest carotenoids. The rate of sucrose, yeast extract, cooking oil was 1.5%, 4.69%, 0.5%, respectively. This leads maximum predicted carotenoids production is µg/g DW. Ho Chi Minh City University of Industry, Viet Nam 3

6 Bui Dinh Hanh Dung et al. Validation of the model A validation of the model and regression equation was done by taking saccharose (1.5%), the rate of yeast extract (4.69%), the rate cooking oil (0.5%) in the experiment. The response for carotenoids production was µg/g dry weights. This proves the validity of the model. DISCUSSION In this study, carotenoids yield obtained is lower compared to others research on R.glutinis. Park et al. (005) used sugar cane molasses medium and achieved 455 µg/g and Squina et al. (00) was 497. µg/g. The volumetric carotenoid production could be increased to 19 +/- mg x 1(-1) in a medium containing (g x l(-1)) yeast extract 11.74, glucose 46 and threonine 18 along with other micronutrients, wherein, beta-carotene yield was 10 +/- mg x l(-1), accounting for 80% of the total carotenoids. (Bhosale & Grade, 001c, p.1) Although R.glutinis GBDOU can biosynthesis maximum carotenoids production of yeast extract, the cost is quite high compared to other nitrogen sources. Therefore, alternative materials need to be examined in order to minimize input costs for the actual application. This can made a favorable impression on production and enhances favorable economic values. Yeast with the ability R.glutinis carotenoids biosynthesis; with colored compounds have a wide range of applications in industries such as the food industry, pharmaceuticals and cosmetics as well as food supplements for aquaculture industry seafood and oviparous poultry. We will continue research on purification properties of yeast R.glutinis GBDOU to apply to the facts. CONCLUSION The evaluation of the medium components for carotenoids production was done by using the Plackett- Burman statistical method. From the original 11 factors were studied and amongst them saccharose, yeast extract, cooking oil were found to be the significant variables for carotenoids production by R.glutinis GBDOU as the percentage confidence level was more than 95%. RSM-CCD was carried out. As the result, saccharose (1.5%), yeast extract (4.69%), cooking oil (0.5%) was determined to be yield carotenoids maximum from R.glutinis GBDOU ( µg/g DW). Using the method of response surface analysis, it was possible to determine optimal operating conditions to obtain a high carotenoids yield. The validity of the model was proved by fitting the values of the variables in the model equation and by actually carrying out the experiment at those values of the variables. The methodology as a whole proved to be quite adequate for the design and optimization of the bioprocess. Thus, a multifactorial statistical approach that considers interaction of independent variables provided a basis for the model to search for a non-linear nature of the response in a short term experiment. Acknowledgements We gratefully acknowledges the technical and financial assistance from Global biotechnology development Company limited (GBD Co., Ltd.), Viet Nam during the course of this investigation. The authors thank executive committee of Institute of Biotechnology and Food technology Ho Chi Minh City University of industry create favorable conditions for this investigation. Mr.Quan especially thanks to valuable constructive criticisms of Mr. H. H. Tan Tai for his help editing the English usage and APA style of this article. REFERENCES Bhosale P and Grade RV, (001a). Production of carotenoids by a mutant of Rhodotorula glutinis in sea water medium. Biosource Technology 76(1): Ho Chi Minh City University of Industry, Viet Nam 4

7 Bhosale P and Grade RV, (001b). β-carotene production in sugarcane molasses by a Rhodotorula glutinis mutant. Journal of Industrial Microbiology and Biotechnology 6(6): Bhosale P and Grade RV, (001c). Optimization of carotenoid production from hyper-producing Rhodotorula glutinis mutant 3 by a factorial approach. Journal of Applied Microbiology 33(1): Bhosale P, Jogdant VV and Grade RV, (003). Stability of β-carotene in spray dried preparation of Rhodotorula glutinis mutant 3. Journal of Applied Microbiology 95(3): Bong KK, Byoung KP, Hee JC and Eui Y K, (004). Effect of phenol on β-carotene content in total carotenoids production in cultivation of Rhodotorula glutinis. Korean Journal of Chemical Engineering 1(3): Buzzini P, (001). An optimization study of carotenoids production by Rhodotorula glutinis DBVPG 3853 from substrates containing concentrated rectified grape must as the sole carbohydrate source. Journal of Industrial Microbiology and Biotechnology 4(1): Castillo E Del, (007). Process Optimization A Statistical Approach. Springer Science. New York, USA: p Dennis KJL (1995) Screening properties of certain two-level Design. Metrika 4(1): Frengova IG and Beshkova DM, (009). Carotenoids from Rhodotorula and Phaffia: yeast of biotechnological importance. J Ind Microbiol Biotechnol 36(): Hayman P, (1974). Carotenoid biosynthesis in Rhodotorula glutinis. Journal of bacterionlogy. American Society for Mirobiology 10(3): Lampia LE, Wallen SE and Bullrman LB, (1985). A review of factors affecting biosynthesis of carotenoid by oder Mucorales. Mycopathologia 90(): Park PK, Cho DH, Kim EY and Chu KH, (005). Optimization of carotenoid production by Rhodotorula glutinis using statistical experimental design. World Journal of Microbiology and Biotechnology 1(4): Plackett RL and Burman JP, (1946). The design of optimum multifactorial experiments. Biometrika 33(4): Sqina FM, Yamashita F, Pereira JL and Mercadante AZ, (00). Production of carotenoids by Rhodotorula rubra and Rhodotorula glutinis in culture medium supplemented with sugar cane juice. Food Biotechnology 16(3): Sutter RP and Whitaker JP, (1981). Sex pheromone metabolism in Blakeslea trispora. Naturwissenschaften 68(3): Vijayalakshmi G, Shobha B, Vanajakshi V, Divakar S and Manohar B, (001). Response surface methodology optimization of growth parameters for the production of carotenoids by a mutant strain Rhodotorula gracilis. Eur Food Rs Technol 13(3): Ho Chi Minh City University of Industry, Viet Nam 5