05 Wibowo.cdr A conventional pulping process involve application of chlorine in bleaching process of the pulp. The used of chlorine in the bleaching process not only affects the waste water released into the environment but also the final paper products. Indonesia is one of the top ten pulp producer countries using chemical pulp production method (Popp et al. 2011). Chemical processes in the pulp and paper industries are generally done at high temperature and pH, especially in the kraft pulping process, the process of sulfate and chlorine bleaching process. Recently, demand of elemental chlorine free (ECF) or totally chlorine free (TCF) in the world market of paper product has dramatically increased in the last decade. In order to fulfill the market demand, utilization of biotechnology might can be done by the application biobleaching (Bajpai et al. 1999). Biobleaching is the process of bleaching the pulp using xylanase (Buchert et al. 1994; Da Silva et al. 1994). The use of xylanase can give a good impact on the environment which replaces or reduces the amount of chlorine used in the bleaching process of pulp. Enzymes are biocatalysts that can speed up a chemical reaction without changing its structure. It is also widely used in industry as an alternative to the use of chemicals that pollute the environment. Application of thermoalkalophilic xylanase in pulp and paper process has been studied extensively since the last two decades (Buchert et al. 1994; Kulkarni et al 1996; Zheng et al. 2000; Oakley et al. 2002; Ibarra et al. 2010; Paes et al. 2012). Xylanase is an extracellular enzyme that can hydrolyze xylan (hemicellulose) into short chain xylooligosaccharide (Kulkarni et al.1999). According to Kulkarni et al. (1999) in the process of bleaching paper (pulp) requires a temperature of 80 °C with a pH Vol.10, No.3, September 2016, p 112-117 DOI: 10.5454/mi.10.3.5 Application of Response Surface Method in Optimization of Medium Composition for Xylanase Production by Bacillus halodurans CM1 in Submerged Fermentation 1 2 3 2* SARA GUSTIA WIBOWO , IS HELIANTI , ANI SURYANI , AND BUDIASIH WAHYUNTARI 1 Post Graduate Program, Biotechnology, Institut Pertanian Bogor, Jalan Raya Dramaga, Bogor 16680, Indonesia; 2 Center for Technology of Bioindustry, Agency for the Assessment and Aplication of Technology (BPPT), LAPTIAB 1, PUSPIPTEK#611, Serpong, South Tangerang 15314, Indonesia; 3 Department of Agroindustrial Technology, Faculty of Agricultural Engineering and Technology, Institut Pertanian Bogor, Jalan Raya Dramaga, Bogor 16680, Indonesia A two level factorial design was performed to optimize xylanase production by alkalothermophilic Bacillus halodurans CM1 using response surface method. The variables involved in this experiment were carbon (X ), 1 nitrogen source (X ) concentration, and pH (X ), corn cob and fish powder were used as carbon and nitrogen 2 3 source respectively. Statistical analysis of the experimental results in the range studied, only carbon source gave significant effect on xylanase production. A second-order model was proposed to represent the enzyme activity as a function of xylan concentration (X ) and pH (X ). The optimum corn cobs concentration was 4.37% (w/v), 1 3 and the fish powder (with high P content) concentration was 1.75% (w/v) and pH 9. These conditions were tested and validated experimentally since the maximum growth rate achieved with these parameters, and the highest -1 xylanase activity observed was 522 U mL after 24 h fermentation. Key words: Bacillus halodurans CM1, production optimization, RSM, xylanase Disain faktorial dua tingkat digunakan untuk optimasi produksi enzim xylanase alkalitermofilik dari Bacillus halodurans CM1 dengan metode statistik respons permukaan. Variabel yang terlibat dalam penelitian ini adalah konsentrasi sumber xilan tongkol jagung (X ), konsentrasi tepung ikan dengan kandungan fosfor (P) (X ), dan pH 1 2 (X ). Analisis statistik hasil CCD menunjukkan bahwa, dalam kisaran yang diamati, konsentrasi tepung ikan 3 tidak memberi pengaruh yang signifikan terhadap produksi xilanase. Sebuah model orde kedua diusulkan untuk mewakili aktivitas enzim sebagai fungsi konsentrasi xilan (X ) dan pH (X ). Konsentrasi Xilan tongkol jagung 1 3 dan tepung ikan P optimum masing-masing adalah 4,37% (b/v) dan 1,75% (b/v) pada pH 9. Kondisi ini diuji dan divalidasi secara eksperimental karena laju pertumbuhan maksimum dicapai dengan parameter ini. Hasil -1 pengamatan aktivitas xilanase terbaik adalah 522 U mL setelah fermentasi berlangsung 24 jam. Kata kunci: Bacillus halodurans CM1, optimasi produksi, RSM, xylanase MICROBIOLOGY INDONESIA Available online at http://jurnal.permi.or.id/index.php/mionline ISSN 1978-3477, eISSN 2087-8575 *Corresponding author: Phone/Fax: +62-21-7560536; Email: budiasih.solichin@gmail.com Volume 10, 2016 Microbiol Indones 113 range of 6.0-8.0. Therefore, the xylanase needed is the alkalothermophilic one. Bacillus halodurans CM1 is a xylanase-producing bacterium and has been isolated by Ulfah et al. (2011) of hot springs Cimanggu, West Java. B. halodurans CM1 is a gram-positive bacterium that produces extracellular xylanase in submerged fermentation system. However, the optimum conditions for the enzymes production include pH, source concentration of carbon and nitrogen has not been known yet. Based on preliminary experiment, B. halodurans CM1 produced xylanase in the xylan containing medium (Ulfah et al. 2011). Therefore, the production of xylanase used xylan carbon source. This study would be planned to be continued through to the industrial scale, so inducer xylan must be substituted with a carbon source (xylan) from xylan-containing agricultural waste (corn cobs and sugarcane bagasse), as well as the source of nitrogen substitution with fish powder. Experiment optimization of xylanase production conditions implemented using response surface optimization techniques methodology (RSM). The aims of this study was to obtain the optimum condition of pH, carbon, and nitrogen source concentration for B. halondurans CM1 xylanase production using response surface method. MATERIALS AND METHODS Microorganism. Bacillus halodurans CM1 (culture collection of Laboratory of Bioindustry, BPPT), isolated from Cimanggu hot pond, West Java was used in this study. Stock cultures were maintained on culture medium, which was used for the submerged culture, refreshed at 55 °C in modified Luria Broth medium containing 0.5 % (w/v) bactopeptone, 0.25 % (w/v) yeast extract, 0.25 % (w/v) sodium chloride, and 1 % (w/v) beech wood xylan. Cultivation Medium. The medium composition used in the experiment was modification of medium according to Mamo et al. (2006). In 100 mL medium consisted of 0.5 g corn cobs xylan, 0.5 g fish powder rich in P, 0.29 g KH PO , 1 g Na CO , 0.2 g NaCI, 0.01 g 2 4 2 3 MgSO , 0.01 g CaCI . For the optimization 4 2 experiments, the cultivation medium was composed of different concentration of corn cob, fish powder, and pH according to the experimental design. Cultivation. The optimization experiments were carried out in 500 mL Erlenmeyer flasks containing 100 mL of cultivation medium. Shake flasks were seeded with inoculum, at initial concentration of 2.2 x 8 10 cells mL, which were incubated for 24 h at 50 °C 120 rpm. Enzyme Assays. Xylanase activity was determined by dinitrosalicylic acid (DNS) method, according to the Miller (1959) for quantifying reducing sugar. 50 µL of crude enzyme extract at appropriate dilutions in tris-base buffer was mix with 450 µL of 0,5 % (w/v) beechwood xylan in 50 mM of buffer at the indicated pH 9. The mixture was incubated for 5 min in the thermomixer at the optimum temperature for enzyme activity of 70 °C. Subsequently, 750 µL DNS reagent (1% dinitrosalicylic acid, 0.2% phenol, 0.05% sodium sulfite, and 1% sodium hydroxide, 20% (w/v) potassium sodium tartrate) was added to stop the reaction. Then the mixture was boiled at 100 °C for 5 min, and kept at room temperature. Afterward 250 µL of distilled water was added. Reducing sugar released during incubation was measured as xylose equivalents at 540 nm. As blank we used the same mixture as in the above sample; however, enzymes were added following addition of DNS into the reaction mixture. One unit (U) of enzyme activity is defined as the amount of enzyme releasing 1 µmol of reducing sugar equivalent per minute under the assay conditions. 3 Experimental Design. A 2 Central Composite Design (CCD) was performed in order to determine the optimal conditions for producing xylanase by B. halodurans CM1. The dependent variable selected for -1 this study was the enzyme activity, expressed in U mL , and the independent variables chosen were corn cob concentration, fish powder (rich in P) concentration, and medium pH. The range and the levels of these variables are given in Table 1. Statistical Analysis. Statistics, Software Design Expert 7 version 7.0.0, stat-ease Corporation Minneapolis, USA, was used for regression and graphical analysis. RESULTS Based on earlier studies the media selected was modified medium with corn cob was selected as a carbon source and fish powder was selected as nitrogen source. The variables involved in this study were corn cob concentration (X ), fish powder concentration (X ), 1 2 and pH (X ). The following experiment was to 3 determine which factors affected the enzyme production. A two level factorial design was defined to determine factors that affected the enzyme production (data not shown). The experimental results showed that the model of corn cob concentration, fish powder concentration and pH value was significant so was the curvature F value. In this work, these variables were statistically optimized with the help of central composite design (CCD) using Response Surface Method (Oh et al. 1995; Wang et al. 2004). The experimental design and the results are shown in Table -1 2. The highest xylanase activity (522.512 U mL ) was observed at run number 8, where the factors corn cob concentration and fish powder concentration was 4.37% (b/v) and (1.75% (b/v), respectively. The medium pH was 9.0 and the incubation time was 24 h. This activity was 15.7 fold higher than that of observed at run number 11, where the related factors were used at highest levels for X , and lowest for X and X . Table 3 2 1 3 shows the regression analysis for this experiment, presenting the estimates and hypothesis test to the coefficients of regression. At the 5% probability level the linier and quadratic coefficient of X (corn cob 1 concentration), and the coefficient of the interaction X X (corn cob concentration and pH) were found to be 1 3 significant. The mathematical model representing the xylanase activity (ŷ) in the experimental region studied can be expressed by Eq.(1) ŷ = -18.89893 + 2.07786X + 4.33193X – 0.19490 1 3 2 X X – 0.22459 X (1)1 3 3 The independent variables, corn cobs, pH and interaction between corn cobs and pH had a significant effect on xylanase production (P > 0.05) (Table 3). The statistical significance of a second-order model equation was evaluated by the F-test analysis of variance (ANOVA). The computed F-value (19.61) indicated that the model was significant at high confidence level. The probability P-value was very low Independent variables Symbol Levels -1 0 1 Xylan X1 0.63 1.31 2.00 Fish powder P X2 0.54 0.99 1.44 pH X3 8 9 10 1 Table 1 Values of independent variables a different levels of the CCD Run number Coded levels Xylanase activity (U mL) X1 X2 X3 Observed Predicted 1 0 0 0 345.18 505.01 2 0 0 0 97.089 90.2609 3 1 1 -1 196.26 266.453 4 1 -1 1 138.47 168.974 5 -1 -1 -1 42.496 51.88 6 -1 1 1 103.85 112.534 7 0 0 0 197.37 212.716 8 0 0 0 522.51 521.328 9 -1 -1 1 99.26 112.534 10 1 1 1 243.19 169.974 11 -1 1 -1 33.19 51.88 12 1 -1 -1 272.19 266.453 13 0 0 0 276.38 212.71 14 0 -1682 0 153.85 212.71 15 0 0 -1.682 254.61 212.71 16 -1682 0 0 67.49 43.09 17 0 0 0 198.58 212.71 18 1.682 0 0 50.34 56.74 19 0 0 1.682 234.75 212.71 20 0 1.682 0 187.32 212.71 1 Table 2 Experimental design and result of the CCD 114 WIBOWO ET AL. Microbiol Indones *Statistically significant at 99% of confidence level **Statistically significant at 95% of confidence level Table 3 Results of regression analysis of the CCD (P=<0.0001) reflecting the significance of the model (Liu et al. 2003). The model show lack of fit not significant and presented a fairly high determination 2 coefficient (R = 0.80), explaining 80% of the variability in the response and indicates correlation between the experimental observed and predicted values (Table 4). In general, a regression model having 2 R value higher than 0.9 considered to have a very high correlation (Haaland 1989). The value of the 2 determination coefficient (R = 80%) was also high enough to indicate the significance of the model. The response surface described by the model equation (ŷ) to estimate xylanase activity over independent variables xylan concentration (X ) and pH 1 (X ) is shown in Fig 1. That illustrated the relationship 3 between the response and experimental data. There is a rather broad plateau region over which the enzyme activity changes relatively litle when xylan concentration is varied. As observed, the fish powder P concentration in the range studied (0.23-1.75% (b/v)), did not affect xylanase production therefore the fish powder concentration was maintained at 1.75 % (b/v). The validation steps of the models using solution from regression model (Eq. 1). Based on the model obtained, the optimal working conditions were defined to attain high xylanase activity minimizing the xylan concentration and cultivation time. Thus, the point assigned as optimum corresponded to 4.37% (b/v) of corn cob concentration and 1.75% (b/v) of fish powder P concentration, pH 9.0 and 24 h cultivation time. Under this condition, the model predicted a xylanase -1 activity of 521.238 U mL was slightly different with -1 the predicted value of 519.5-523.1 U mL . As a result, the models developed were considered to be accurate and reliable for predicting the production of xylanase by alkalophilic B. halodurans CM1. DISCUSSION Although previous research regarding alkalo thermophilic xylanase production has been reported, little information on the optimization of its production is available (Heck et al. 2005). Some of these studies with bacteria have not been reported using statistical designs (Brodel et al. 1990; Subramanian et al. 1998). In this paper, experimental method of 2 level factorial design, and RSM design demonstrated that the CCD and regression analysis methods were effective to find the optimized corn cobs, fish powder concentration and pH for the production of alkalo-thermostable xylanase from alkalophilic B. halodurans CM1. The bacterium a newly isolated strain growing on some economical and Term Coefficient Standard error P-value Intercept 2.33 0.067 0.0060 X1 0.22 0.045 0.0011* X1X1 -0.0049 0.044 0.2939 X2 0.053 0.045 0.2679 X2X2 0.037 0.044 0.4178 X3 0.033 0.045 0.4811 X3X3 -0.23 0.044 0.0008* X1X2 0.024 0.058 0.6938 X1X3 -0.13 0.058 0.0518** X2X3 0.064 0.058 0.3079 1 Source SS DF MS F-ratio P-value Model 1.55 3 0.52 19.61 <0.0001 Residual 0.37 14 0.026 Lack of Fit 0.35 11 0.032 6.44 0.0759 Total (corr.) 1.93 19 1 2 R = 0.80;SS, sum of squares; DF, degrees of freedom; MS,mean square Table 4 Analysis of variance (ANOVA) for the model regression representing xylanase activity Volume 10, 2016 Microbiol Indones 115 abundant substrate as corn cobs replaces the expensive carbon source-xylan and fish powder replaces the expensive nitrogen source-pepton. Using the RSM analysis it was found that maximum production xylanase activity was obtained at 4.37% (w/v) corncobs (w/v), 1.75% (w/v) fish powder P , pH 9.0 after 24 h incubation the activity was increased nearly twice compared to the original medium. According to Lin et al. (2007) experiment alkalophilic Bacillus sp under high pH conditions around 9.0-10.0 the cell growth and the production of xylanase were comparatively desirable, it corresponds to B. halodurans optimum pH 9.0. Statistical method for optimizing fermentation process could overcome the limitations of classic empirical methods and was proved to be a powerful tool to optimize the production of alkalothermostable xylanase from alkaliphilic B. halodurans CM1. In this work, RSM model was proposed to study the combined effects of culture media composition. Under optimal composition, the predicted production of xylanase after 24 h incubation -1 was 52 U mL . Optimization study of xylanase production by Bacillus circulans D1 using RSM by Bocchini et al. (2002), showed that optimum xylanase activity of 19 -1 mL was gained at the optimum xylan concentration -1 and cultivation time of 5 g L and 48 h respectively. In our work the xylanase production was much higher and the highest xylanse activity was reached in a shorter period of time than that of Bocchini et al. (2002) report. Response surface methods were also used to optimizing a culture medium with regards to xylanase production by a Schizophyllum commune strain (Haltrich et al. 1993), the optimum concentration of -1 -1 -1 73.4 g L avicel, 55.5 g L yeast extract, and 1.38 g L NH NO were determined by a central composite 4 3 -1 design. The highest xylanase production (5.74 U mL ), resulted in an increase of 330% compared initial medium and was reached within 11 days fermentation. The cultural conditions for thermostable xylanase production of Thermomyces lanuginosus was also studied by Purkarthofer et al. (1993), an experiment using CCD was performed to optimize the medium component concentrations. When the fungus was -1 grown in the optimized medium 31.2 g L corn cob, -1 -1 30.2 g L yeast extract, and 5.0 g L KH PO , the 2 4 highest xylanase production observed within 7 days -1 was 2.70 U mL . Validation experiments were also carried out to verify the availability and the accuracy of the models, and the result showed that the predicted values were well agreed with the experimental values. It strongly suggested that the production of alkalo-thermostable xylanase by wild-strain alkaliphilic B. halodurans CM1 could improve in a large-scale fermentation process. The result of this study also provides some useful information for other extremozymes fermentation processes. These facts are important in making the whole process economically more feasible, in view of the high cost of pure xylan and the difficulty of its extraction from agricultural waste material. The alkalophilic pH of the fermentation medium 9.0 and the Fig 1 Response surface described by the model ŷ, which represents xylanase activity (U mL) as a function of xylan concentration and pH. 116 WIBOWO ET AL. Microbiol Indones high temperature used 50 °C also reduced the chances of contamination by opportunist microorganisms. REFERENCES Bajpai P. 1999.Application of enzymes in the Pulp and Paper Industry. Biotechnol Prog. 15:147-157. Bocchini DA, Alves-Prado HF, Baida LC, Roberto IC, Gomes E, Da-Silva R. 2002. Optimization of xylanase production by Bacillus circulans D1 in submerged fermentation using response surface methodology. Process Biochem. 38:727–31. Brodel B, Samain E, Debeire P. 1990. Regulation and optimization of xylanase production in Clostridium thermolacticum. Biotechnol Lett. 12(1):65-70. Buchert J, Tenkanen M, Kantelinen A, Viikari L. 1994. Application of xylanases in the pulp and paper industry. Bioresour Technol. 50:65-72. Chen C, Chen JL, Lin TY. 1997. Purification and characterization of a xylanase from Trichoderma longibrachiatum for xylooligosaccharide production. Enzyme Microbiol Technol. 21:91-6. Da Silva R, Yim DK, Park YK. 1994. Application of thermostable xylanases from Humicola sp for pulp improvement. J Ferment Bioeng. 77(1):109-11. Gupta S, Bhusnan B, Hoondal GS. 1999. Enhanced production of Xylanase from Staphylococcus sp. SG-13 using amino acids. World J Microbiol Biotechnol. 15:511-512. Haaland, P.D. (Ed.), 1989. Separating signals from the noise. in: experimental design in biotechnology. Marcel Dekker, New York, p.61-83. Haltrich D, Preiss M, Steiner W. 1993. Optimization of a culture medium for increased xylanase production by a wild strain of Schizophyllum commune. Enzyme Microbiol Technol. 15:854-60. Heck JX, De Barros Soares LH, Ayub MAZ (2005) Optimization of xylanase and mannanase production by Bacillus circulans strain BL53 on solid-state cultivation. Enzyme Microbiol Technol. 37:417–423. Katapodis P, Christakopoulou V, Chistakopoulos P. 2006. Optimization of xylanase production by Sporotrichum thermophile using corn cobs and response surface methodology. Eng Life Sci. 2006. 6(4): 410-415. doi: 10.1002/elsc.200520134. Kubata BK, Suzuki T, Horitsu H, Kawal K,Takamizawa K. 1994. Purification and characterization of Aeromonas caviae ME-1 xylanases V, which produces exclusively xylobiose from xylan. Appl Environ Microbiol. 60: 531-535. Kulkarni, N., A. Shendye & M. Rao. 1999. Molecular and biotechnological aspects of xylanases. FEMS Microbiol Rev. 23: 411-456. Lin, S.,Dou W,Xu H, Li H, Xu Z, Ma Y.2007. Optimization of medium composition for the production of alkaline β-mannanase by alkaliphilic Bacillus sp. N16-5 using response surface methodology. Appl Microbiol Biotechnol. 75:1015-1022. doi 10.1007/s00253-007- 0907-y. Liu JZ, Weng LP, Zhang QL, Xu H, Ji LN. 2003. Optimization of glucose oxidase production by Aspergillus niger in a benchtop bioreactor using response surface methodology. World J Microbiol Biotechnol. 19:317–323. Mamo G, Rajni HK, Mattiason B. 2006. A thermostable alkaline active endo-β1-4 xylanase from Bacillus halodurans S7: purification and characterization. Enzyme Microbiol Technol. 39(7):1492. Miller GL. 1959. Use of dinitrosalycylic acid as reagent for the determination of reducing sugars. Anal Chem. 31(3):208-218. doi:10.10.1021/ac60147a030. Oh S, Rheem S, Sim J, Kim S, Back Y. 1995. Optimizing conditionsfor the growth of Lactobacillus casei YIT9018 in tyrptone-yeast extract-glucose medium by using response surface methodology. Appl Environ Microbiol. 61:3809–3814. Purkarthofer H, Sinner M, Steiner W. 1993. Cellulase-free x y l a n a s e f r o m T h e r m o m y c e s l a n u g i n o s u s : optimization of production in submerged and solid- state culture. Enzyme Microbiol Technol. 15:677-82. Subramaniyan S, Prema P. 2002. Biotechnology of microbial xilanases: enzymology, molecular biology and application. Critical Rev Biotechnol. 22(1):33-46. Ulfah M, Helianti I , Wahyuntari B, Nurhayati N. 2011. Characterization of a new thermoalkalophilic xylanase- producing bacterial strain isolated from Cimanggu hot spring,West Java, Indonesia . J Microbiol Indonesia. 5(3): 139-143. doi: 10.5454/mi.5.3.7. Viikari L, Kantelinen A, SundquistJ, Linko M. 1994. Xylanases in bleaching: from an idea to the industry. FEMS Microbiol Rev. 13: 335-350. Wang YX, Lv FX, Lu ZX. 2004. Optimization of cultivation medium Clitocybe sp.AS5.112 for the extracellular polysaccharide production and mycelial growth by response surface methodology. J NanJing Agriculture University. 27:89–94. Volume 10, 2016 Microbiol Indones 117 Page 1 Page 2 Page 3 Page 4 Page 5 Page 6