Improvement of Production and Stability of Silk Degumming ... · Keywords: optimization, silk...

15
Chiang Mai J. Sci. 2015; 42(3) : 599-613 http://epg.science.cmu.ac.th/ejournal/ Contributed Paper Improvement of Production and Stability of Silk Degumming Protease by Bacillus sp. C4 SS-2013 Nisa Romsomsa [a,c], Patoomporn Chim-anage*[a,c,d], Sarote Sirisansaneeyakul [b,c] [a] Department of Microbiology, Faculty of Science, Kasetsart University, Bangkok 10900, Thailand. [b] Department of Biotechnology, Faculty of Agro-Industry, Kasetsart University, Chatuchak, Bangkok 10900, Thailand. [c] Center for Advanced Studies in Tropical Natural Resources, NRU-KU, Kasetsart University, Chatuchak, Bangkok, 10900, Thailand. [d] Office of the Higher Education Commission and Center of Excellence for Silk, Kamphang Saen, Nakorn Pathom, 73140, Thailand. *Author for Correspondence; e-mail: [email protected] Received: 23 December 2013 Accepted: 7 May 2014 ABSTRACT Optimization of silk degumming protease production from isolated Bacillus sp. C4 SS-2013 was performed by batch culture in a stirred tank bioreactor. Central composite design (CCD) was applied to optimize aeration and agitation rates and predicted by response surface methodology (RSM). The maximum protease production (P) of 1,890 U/mL and a specific production rate (q p ) of 21,412 U/g/h were obtained at the agitation and aeration rates of 400 rpm and 2 vvm which corresponded to a K L a value of 182.16 h -1 . These experimental values were coincident with the predicted values and the models were proved to be adequate with the coefficient of determination (R 2 ) of P, q p , and K L a equal to 0.950, 0.931 and 0.881, respectively. The pH-stat fed-batch culture could increase the silk degumming protease production to 4,437 U/mL (2.35-fold) and q p to 36,820 U/g/h (1.72-fold). The addition of 10 mM CaCl 2 into the crude protease and storage at 4 O C for 8 weeks could maintain its activity at about 52 % of the initial activity. Both the agitation and aeration rates significantly affected the protease production, specific protease production rate and K L a. The improvement of production and stabilization during storage of silk degumming protease were achieved in our study which has potential for industrial applications. Keywords: optimization, silk degumming, protease production, agitation and aeration rate INTRODUCTION Raw silk from silkworm cocoons consists of double-stranded filaments of protein called fibroin (70-78%), held together by gum-like protein termed sericin (22–30%), water, and mineral salt [1-3]. Degumming is a key process during which sericin is removed by either chemical or enzymatic methods to improve the texture, luster, and dye-binding capacity of the fibroin fiber. Enzyme-degummed silk fabric not only displayed a higher quality than the traditional degummed fabric using boiling alkaline solution but also saved water, energy

Transcript of Improvement of Production and Stability of Silk Degumming ... · Keywords: optimization, silk...

  • Chiang Mai J. Sci. 2015; 42(3) : 599-613http://epg.science.cmu.ac.th/ejournal/Contributed Paper

    Improvement of Production and Stability of Silk Degumming Protease by Bacillus sp. C4 SS-2013Nisa Romsomsa [a,c], Patoomporn Chim-anage*[a,c,d], Sarote Sirisansaneeyakul [b,c][a] Department of Microbiology, Faculty of Science, Kasetsart University, Bangkok 10900, Thailand. [b] Department of Biotechnology, Faculty of Agro-Industry, Kasetsart University, Chatuchak, Bangkok

    10900, Thailand. [c] Center for Advanced Studies in Tropical Natural Resources, NRU-KU, Kasetsart University,

    Chatuchak, Bangkok, 10900, Thailand.[d] Office of the Higher Education Commission and Center of Excellence for Silk, Kamphang Saen,

    Nakorn Pathom, 73140, Thailand.*Author for Correspondence; e-mail: [email protected]

    Received: 23 December 2013 Accepted: 7 May 2014

    ABSTRACT Optimization of silk degumming protease production from isolated Bacillus sp. C4

    SS-2013 was performed by batch culture in a stirred tank bioreactor. Central composite design (CCD) was applied to optimize aeration and agitation rates and predicted by response surface methodology (RSM). The maximum protease production (P) of 1,890 U/mL and a specific production rate (qp) of 21,412 U/g/h were obtained at the agitation and aeration rates of 400 rpm and 2 vvm which corresponded to a KLa value of 182.16 h

    -1. These experimental values were coincident with the predicted values and the models were proved to be adequate with the coefficient of determination (R2) of P, qp, and KLa equal to 0.950, 0.931 and 0.881, respectively. The pH-stat fed-batch culture could increase the silk degumming protease production to 4,437 U/mL (2.35-fold) and qp to 36,820 U/g/h (1.72-fold). The addition of 10 mM CaCl2 into the crude protease and storage at 4OC for 8 weeks could maintain its activity at about 52 % of the initial activity. Both the agitation and aeration rates significantly affected the protease production, specific protease production rate and KLa. The improvement of production and stabilization during storage of silk degumming protease were achieved in our study which has potential for industrial applications.

    Keywords: optimization, silk degumming, protease production, agitation and aeration rate

    INTRODUCTIONRaw silk from silkworm cocoons consists

    of double-stranded filaments of protein called fibroin (70-78%), held together by gum-like protein termed sericin (22–30%), water, and mineral salt [1-3]. Degumming is a key process during which sericin is removed by either

    chemical or enzymatic methods to improve the texture, luster, and dye-binding capacity of the fibroin fiber. Enzyme-degummed silk fabric not only displayed a higher quality than the traditional degummed fabric using boiling alkaline solution but also saved water, energy

  • Chiang Mai J. Sci. 2015; 42(3)600

    and chemicals. Moreover it reduced the chemical waste treatment and sericin-recovery steps for further useful applications [4]. However, the major commercial proteases that are used for degumming of raw silk destroy the silk fiber to some extent [5] and thus a specific protease needs to be developed.

    The crude protease in this experiment was produced from Bacillus sp. C4 SS-2013 which was isolated in our laboratory from the wastewater of a Thai silk factory. It is capable of removing sericin very well, that is, about 26.5 % of the total dry weight of the raw silk yarn which entails 94% of the total sericin after incubation under mild conditions at pH 8 and room temperature (30 to 37°C) for 2 h [6]. The morphology analysis of silk fibroin investigated under an electron microscope after degumming by the active protease revealed it to be clean and smooth without any damage to the filaments [7]. The optimization of medium composition and the culture conditions for the protease production in shake flasks by central composite design (CCD) and response surface methodology (RSM) was investigated using cheap and abundant raw materials in Thailand, for example, cassava starch and soy flour in order to reduce the cost of production [8]. However, hydrolyzed cassava starch and soy flour were also tested also in a Newtonian system for process scale up [9]. Hence, the agitation and aeration rates are the most critical parameters and they play significant roles in determining the productivity of the process. Aeration could be beneficial to the growth and performance of microbial cells by improving the mass transfer characteristics with respect to substrate, product and oxygen [10]. Agitation is also an important parameter for adequate mixing, mass, and heat transfer. It also creates shear forces, causing morphological changes, variation in growth and product formation, and also damages to the cell structure [11].

    The optimization process involves performing the statistically designed experiments, analysis, and predicting the response, and then checking the adequacy of the model. Response surface methodology (RSM) is now being routinely used for optimization studies in several biotechnological and industrial processes including enzyme production. It helps in evaluating the effective factors and building models to study the interactions and select the optimum conditions of variables for a desirable response [12].

    The current study is aimed to optimize the possible scale up of the silk degumming protease production using Bacillus sp. C4 SS-2013 in a 2 L stirred tank bioreactor. In this context, the optimization of the agitation and aeration rates using a CCD experiment and RSM analysis were studied. The effect of these parameters on the volumetric mass-transfer coefficient (KLa), the specific protease production rate (qp), and the maximum protease production were considered. The improvement in protease production by pH-stat fed-batch culture, stabilization of the enzymes, and some kinetic values concerned are also reported.

    2. MATERIALS AND METHODS 2.1 Microorganism and Culture Media

    The Bacillus sp. C4 SS-2013 used in this study was selected from our laboratory (previous name was Bacillus subtilis C4) as it is a promising strain for the production of protease for the silk degumming process [7]. The partial 16S rDNA sequence (955 bp) showed 100% identity with Bacillus subtilis DSM10 (type strain). Based on morphological, biochemical, physiological characteristics and 16S rDNA sequence, this strain tends to be Bacillus subtilis (in preparing for publication) and was given the accession number of AB841263. The culture was maintained at -20oC in modified Davis minimal medium [13] with raw silk [7] containing 20% glycerol.

  • Chiang Mai J. Sci. 2015; 42(3) 601

    The pre-culture medium was made up of hydrolyzed cassava starch (total sugar 20 g/L), soy flour (20 g/L), K2HPO4 (1 g/L), MgSO4.7H2O (0.2 g/L), and skim milk (1 g/L) adjusted pH to 7.5 with 5 N HCl/NaOH before sterilization (110oC, 15 min). The batch culture medium was optimized according to Romsomsa et al. [8] using hydrolyzed cassava starch (total sugar 20 g/L); yeast extract (2.5 g/L); soy flour (20 g/L); K2HPO4, (1 g/L); MgSO4·7H2O, (0.2 g/L) and skim milk (1 g/L). The feed medium for fed-batch culture consisted of hydrolyzed cassava starch (total sugar 20 g/L), yeast extract (5 g/L), soy flour (40 g/L) and skim milk (2 g/L) and was sterilized at 121°C for 15 min. The soy flour and skim milk were sterilized separately at 110°C for 15 min.

    To prepare the hydrolyzed cassava starch, cassava slurry (250 g/L) was prior liquefied by adding Termamyl® (120L type LS, 120 KNU/g, Novo Nordisk, Denmark) at 95°C for 2 h and was then saccharified by dextrozyme (225 AUG/mL, 75 PUN/mL, Novo Nordisk, Denmark) at 60°C for 24 h to obtain 70-75 % dextrose equivalent (DE).

    2.2 Inoculum Preparation Inoculum was prepared by transferring

    preserved culture into 100 mL pre-culture medium in a 500 mL Erlenmeyer flask and incubated on a shaker at 280 rpm and 30oC for 12 h and the process was then repeated once more. The starter absorbance was adjusted to A660 = 0.3 before use.

    2.3 Optimization of Agitation and Aeration Rates on Protease Production in a Fermenter

    All the fermentation runs to optimize the agitation and aeration rates in silk degumming protease production were conducted by batch mode in a 2 L fully controlled fermenter (Biostat B, B. Braun Biotech. Inc.) containing 1.5 L of the batch culture medium. The fermentation was conducted using 2% (A660 = 0.3) inoculum

    and a controlled temperature of 30oC and pH 7.5±0.05 by 10 N NaOH. The dissolved oxygen (DO) concentration was measured using a sterilizable polarographic electrode (In pro® 6800 series O2 sensors, Mettler-Toledo, Switzerland). The volumetric oxygen transfer coefficient (KLa, h

    -1) of cultures was measured according to the static method of gassing out described by Wise, 1951 [14].

    Experimental design analysis was carried out using the Statistica 5.0 software (Statsoft, USA). A central composite design was used to verify the statistical significance of the agitation rate and the aeration rate on the silk degumming protease production. In total of 11 sets of experiments including three central points were employed to determine the significant factors affecting the protease production. The actual values and coded values in the central composite design are listed in Table 1. In the statistical model, Y is the measured response and denotes units of protease activity (U/mL), the specific production rate for protease (U/g/h), and the volumetric oxygen transfer coefficient (h-1). The quadratic model for predicting the optimal points was expressed as follows:

    2112

    2222

    211122110 xxxxxxY

    (1)where Y represents the response variable, β0 is the interception coefficient, β1 and β2 are the linear terms, β11and β22 are the quadratic terms, and X1 and X2 represent the studied variables. The statistical analysis of the model was performed using analysis of variance (ANOVA). The significance of the regression coefficients and the associated probabilities (p), were determined by Student’s t-test; the second-order model equation was determined by Fischer’s test. The variance explained by the model was given by the coefficient of determination (R2).

    2.4 pH-Stat Fed-Batch CultureThe pH-stat fed-batch culture was carried

  • Chiang Mai J. Sci. 2015; 42(3)602

    out in a 2 L jar fermenter by using the 1.5 L batch culture medium and under the following conditions: temperature, 30°C; agitation rate, 400 rpm; aeration rate, 2 vvm; and pH 7.5. Dissolved oxygen concentration was maintained at a level greater than 30% of air saturation by mixing with pure oxygen when necessary. The cells were initially grown in batch mode until the sugar was almost completely consumed (about 11 h) and addition of feed medium consisting of hydrolyzed cassava starch (total sugar 36 g/L), yeast extract (5 g/L), soy flour (40 g/L) and skim milk (2 g/L) was started and continued till the pH of the culture broth was above 7.5. The culture broth of every sample was centrifuged at 9,200 g, for 10 min and the supernatant was used to assay the protease activity and for sugar determination.

    2.5 Enzyme Stability and Storage ConditionThe supernatant of the culture broth of

    Bacillus sp. C4 SS-2013 fermented for 24 h was stored in various chemical solutions to obtain the final concentration as follows: 10 mM

    CaCl2, 0.02 % (w/v) sodium azide, 25 % (v/v) glycerol, 10% (v/v) polyethylene glycol (PEG 400) and 10 mM CaCl2 combined with 10% (v/v) PEG 400. The enzyme mixture was adjusted to pH 8 with 50 mM phosphate buffer and kept at room temperature (30 to 37°C), 4°C and -20°C, respectively. The residual enzyme activity was measured after storage for 0, 1 and 3 weeks to determine the most suitable chemical and storage temperature. The selected chemical for stabilizing the crude protease was consecutively determined for the optimal concentration at pH 8 and 4°C for 3 weeks and the residual protease was assayed. All experiments were carried out in triplicate and the results were presented as mean values.

    2.6 Detection of Protease Activity by ZymographyTo ensure the crude enzyme obtained in

    the culture broth was protease, Native polyacrylamide gel electrophoresis (native PAGE) impregnated with a protein substrate

    [15] was performed.

    Table 1. Actual values and coded values in central composite design for silk degumming protease production from Bacillus sp. C4 SS-2013.

    TreatmentsActual values Coded values

    Agitation rate (rpm)

    Aeration rate (vvm) X1 X2

    1 200 1 -1 -1

    2 600 1 1 -1

    3 200 3 -1 1

    4 600 3 1 1

    5 400 2 0 0

    6 400 2 0 0

    7 400 2 0 0

    8 118 2 -1.414 0

    9 400 0.6 0 -1.414

    10 683 2 1.414 0

    11 400 3.4 0 1.414

  • Chiang Mai J. Sci. 2015; 42(3) 603

    2.7 AnalysisTotal sugar was determined by phenol–

    sulfuric acid method [16] and reducing sugar by a modified dinitrosalicylic acid (DNS) method [17].

    Cell growth was monitored by using the viable cell count (log CFU/mL) and the cell dry weight by the correlation between log CFU/mL and dry cell weight. Spore yields were determined by heating the culture broth at 80°C in a water bath for 20 min and the numbers of endospores were determined by the spread plate method.

    The protease activity was assayed by a modified method of Ferrero et al. [18] with incubation at pH 8 and 60°C for 10 min. One unit of alkaline protease activity was defined as 1 μg tyrosine liberated mL-1 min-1.

    Calculations of the fermentation parametersAll fermentation kinetic parameters and

    the volumetric rates (m, qp, Qp, YP/S, and YX/S) were calculated as follows:Specific growth rate (m; h-1)

    12

    12

    ln1XX

    tt (2)

    where X1 and X2 are the dry cell weight (g/L) at time t1 and t2 (h) during the exponential phase.Specific production rate (qp; U/g/h)

    ∆∆

    =tP

    Xq

    avgp

    1 (3)

    where Xavg is the average dry cell weight during the exponential phase and tP ∆∆ / is the slope of the regression line for the plot between protease activity (P, U/L) and time (t).Productivity (Qp; kU/L/h)

    12

    12

    ttPPQP −

    −= (4)

    Yield coefficients (YP/S, YX/S)

    21

    12/ SS

    PPY SP −−

    = (5)

    21

    12/ SS

    XXY SX −−

    = (6)

    where X (g/L), P (kU/L), S (g/L), t (h) are the dry cell weight, the protease activity, total sugar concentration and fermentation time, respectively. The subscripts 1 and 2 are denoted as the beginning and the end of the fermentation time, respectively.

    3. RESULTS AND DISCUSSION3.1 Optimization of Agitation and Aeration Rates on Silk Degumming Protease Production

    Since oxygen transfer rate (OTR) is the most important parameter dependent on agitation and aeration rate in aerobic fermentation, and plays a significant role in determining the productivity of fermentation process [19]. The preliminary study, the optimization of medium was carried out by using 32 factorial designs experiments in shake flasks [8]. A clear correlation effect between hydrolyzed cassava starch and the shaker speed on silk degumming protease production was found, with increased hydrolyzed cassava starch in the medium beyond 8 g/L at 280 rpm leading to a decrease in protease production (data not shown). These results emphasize the necessity of oxygen supplementation which is related to the aeration and agitation rates in the fermenter. Table 2 shows the results obtained by using the central composite experiment to study the effects of these two parameters on silk degumming protease production in the fermenter. The maximum protease production (1,880 U/mL), maximum specific protease production rate (27,637 U/g/h) and maximum specific growth rate (0.086 h-1) were achieved by agitation at 400 rpm and an aeration rate of 2 vvm (Treatment 5, 6, and 7), where the average KLa was 171.48 h

    -1. Regression analysis (Table 3) of the

    experimental data showed the results of using the probability (p) values as a tool to check the

  • Chiang Mai J. Sci. 2015; 42(3)604

    significance of each variable. The smaller the magnitude of the p value, the more significant was the correlation with the corresponding coefficient and the very low probability value (p value @0) demonstrated a very high significance for the regression model, which was found in all responses. Among the test variables used in this study, the agitation rate and aeration rate had a strong positive linear effect (X1, X2) but there was also a negative effect of X1X1 and X2X2 (except X1X1 of KLa). In this study, the quadratic terms were significant, which suggested considerable curvature in the model. However the model terms X1X2 of maximum protease production and specific protease production rate (qp) were not significant while the term X1X1 of KLa was not significant either.

    The models fitting the above response variables were as followed: Protease production (Y1)

    2122

    21211 66.0078.7440171.009.702,225.1393.701,3 XXXXXXY

    212221211 66.0078.7440171.009.702,225.1393.701,3 XXXXXXY (7)

    Specific protease production rate (Y2)

    2122

    2122 6.66.175,122.01.198,464.1846.123,60 1 XXXXXXY

    21222122 6.66.175,122.01.198,464.1846.123,60 1 XXXXXXY

    (8)

    Volumetric oxygen transfer coefficient (Y3)

    2122213 24.098.5182.30399.085.399 XXXXXY

    2122213 24.098.5182.30399.085.399 XXXXXY (9)

    where X1 is the agitation rate and X2 is the aeration rate.

    ANOVA to fit the response function of protease production, the specific protease production rate and KLa with the experimental data was performed and the resultant R2 values were 0.9503, 0.9306 and 0.8810, respectively which indicated that the model showed a good fit and explained the variability of these values very well.

    Two-dimensional response surfaces were plotted on the basis of the model equation in

    Table 2. Experimental design showing the optimization of agitation rate and aeration rate for protease production, qp and KLa.

    TreatmentFactors Maximum protease activity (U/mL)

    Specific production rate (qp; U/g/h)

    KLa (h-1)

    Agitation rate (rpm)

    Aeration rate (vvm) Observed* Predicted Observed Predicted Observed Predicted

    1 200 1 296.47 ± 14.11 353.00 3,154 2,952 25.56 6.50

    2 600 1 178.12 ± 1.17 438.97 1,564 6,214 257.76 228.64

    3 200 3 290.00 ± 16.67 67.25 3,687 564 122.40 89.85

    4 600 3 697.04 ± 2.23 678.62 7,339 9,069 163.80 134.22

    5 400 2 1,880.00 ± 4.71 1813.18 27,637 26,016 145.80 171.50

    6 400 2 1,787.06 ± 4.12 1813.18 27,389 26,016 183.24 171.50

    7 400 2 1,776.64± 10.93 1813.18 23,022 26,016 185.40 171.50

    8 118 2 70.32 ± 0.19 195.69 885 3,552 23.76 56.65

    9 400 0.6 558.33 ± 11.66 341.77 4,336 1,507 36.36 66.87

    10 683 2 852.5 ± 10.00 689.06 16,069 11,877 225.72 254.42

    11 400 3.4 130.74 ± 0.84 309.18 535 1,837 37.08 68.25

    *The experiments were performed in duplicate and the results were expressed as mean ± SD

  • Chiang Mai J. Sci. 2015; 42(3) 605

    order to investigate the interaction between the two variables (the aeration rate and agitation rates) and to determine the optimum level of each factor for maximum protease production by Bacillus sp. C4 SS-2013 (Figure 1). The effect of agitation and aeration rates on protease production (Y1) is represented in Figure 1a. There was an increase in the protease production with the increase of agitation and aeration rates. The maximum protease production was obtained with an agitation rate in the range 320 to 520 rpm and an aeration rate from 1.5 to 2.5 vvm. A further increase in the agitation rate over 520 rpm and the aeration rate over 2.5 vvm resulted in decreased protease production. Figure 1b clearly shows that the maximum specific protease production rate can be achieved

    with an increase in the agitation rate from 310 to 480 rpm and in the aeration rate from 1.6 to 2.5 vvm. However, increasing the agitation and aeration rates beyond these values resulted in a decrease in the specific protease production rate. The values of KLa often serve to compare the efficiency of bioreactors and mixing devices. It is also an important scale-up factor to indicate the oxygen demand of a culture. Thus, the aeration and agitation rates are often selected to achieve the desired KLa, since this is the controlling parameter in most fermentation systems [20]. As shown in Figure 1c, increasing the agitation rate clearly increased the KLa value and the maximum KLa occurred with agitation greater than 630 rpm, while increasing the aeration rate to more than 2 vvm, led to a

    Table 3. Estimated regression coefficients for protease production, specific protease production rate and KLa along with their significance levels.

    Model Coefficient Standard error t-value p-value

    Protease production (U/mL)

    Intercept -3,701.93 395.21 -9.37 0.000

    X1 13.25 1.24 10.66 0.000

    X2 2,702.09 248.73 10.86 0.000

    X1X1 -0.0171 0.00 -13.03 0.000

    X2X2 -744.078 52.61 -14.14 0.000

    X1X2 0.66 0.31 2.10 0.052

    Specific protease production rate (qp)

    Intercept -60,123.6 7,169.74 -8.39 0.000

    X1 184.4 22.54 8.18 0.000

    X2 46,198.1 4,512.46 10.24 0.000

    X1X1 -0.2 0.02 -9.59 0.000

    X2X2 -12,175.6 954.52 -12.78 0.000

    X1X2 6.6 5.66 1.16 0.265

    Volumetric oxygen transfer coefficient (KLa )

    Intercept -399.85 73.36 -5.45 0.000

    X1 0.99 0.23 4.27 0.001

    X2 303.82 46.17 6.58 0.000

    X1X1 0.00 0.00 -0.82 0.426

    X2X2 -51.98 9.77 -5.32 0.000

    X1X2 -0.24 0.06 -4.11 0.001

  • Chiang Mai J. Sci. 2015; 42(3)606

    (1a) Volumetric protease production (P)

    (1b) Specific protease production rate (qp)

    (1c) Volumetric oxygen transfer coefficient (KLa)

    Figure 1. Contour and response surface plot for the effects of agitation and aeration rates in a fermenter for silk degumming protease production (a) protease production, (b) qp, (c) KLa.

  • Chiang Mai J. Sci. 2015; 42(3) 607

    decrease in KLa. However, both the enzyme production and biomass dropped at the maximum KLa.

    According to these results, it was clear that agitation and aeration had significant effects on protease production, the specific protease production rate and KLa (Table 2). at a low aeration and agitation rates, incomplete mixing and/or oxygen transfer resistance probably caused insufficient oxygen supply for the bacterial growth and subsequent enzyme production [11]. Therefore, increasing the agitation rate and aeration rate to appropriate values led to an improvement in the cell growth and protease production due to the better air supply to the cells. The increasing rate also provided good mixing of the culture broth, thus helping to maintain a concentration gradient between the interior and the exterior of the cells and thus affects the nutrient availability to microorganisms [21]. This is especially important for the high biomass which is an aerobic microorganism. Moreover, the higher agitation rate may inhibit the growth and activity of microorganisms due to a higher shear stress and may also block the transfer of nutrients and their consumption by disturbing the cells

    [20] and influencing both oxygen transfers and cell activity in the system [22]. Furthermore, the higher aeration rates inhibited cell growth by creating an air dispersion problem in the fermentation system which in turn affected the biomass concentration in the fermenter [11]. Verification experiments under optimal conditions (400 rpm, 2 vvm) showed that the predicted value of protease production (1,831.94 U/mL) was in accordance with the observed value (1,890 U/mL) in batch fermentation as shown in Figure 2. Such evidence was obtained for the observed values of KLa and the specific production rate of protease namely, 182.16 h-1 and 21,412 U/g/h, respectively. The correlation of growth, enzyme production and sporulation of Bacillus sp. C4 SS-2013 was also investigated. The bacterial growth reached its maximum value after 24 h of cultivation and the protease production substantially increased with the maximum protease activity being obtained at 27 h during the stationary phase which revealed that the enzyme is partially associated with growth. At this period, the depletion of glucose and oxygen limitation was observed and made it possible to synchronize the sporulation [23].

    Figure 2. Batch culture of silk degumming protease production in the optimized medium at 400 rpm and 2 vvm in a fermenter. Symbols: cell growth ( ), protease activity ( ), spore ( ), glucose concentration ( ) and dissolved oxygen ( ).

  • Chiang Mai J. Sci. 2015; 42(3)608

    Furthermore, in summarizing the growth kinetic data on protease production by in batch fermentation under different cultivation conditions, it is interesting to notice that variation of the agitation and aeration rates also affected the substrate consumption and product formation rates as it directly affected the cell growth and enzyme formation. The maximum yield of protease production from substrate (Yp/s; 83.48 kU/g total sugar) was achieved under optimized agitation and aeration rates in the fermenter (400 rpm 2 vvm), which was an improvement of 1.15-fold over the un-optimized medium in a shake flask (Table 4). However, lower and higher agitation or aeration rates resulted in a reduction in alkaline protease productivity which coincided with those values reported from the mutant of Bacillus licheniformis [11]. The production of protease in optimized medium was confirmed by native-PAGE activity gels which were obtained from fermentation broth samples taken at different fermentation time intervals (Figure 3). No protease activity was detected at the beginning of fermentation (Lane 1). However, crude enzyme from the 21, 22, 24 h culture broth showed clear zones of protease activity after staining with Coomassie blue.

    3.2 pH-Stat Fed-Fatch CultureTo improve the protease production by

    supplying the main necessary substrates to overcome nutrient starvation and to prolong the protease production, a fed-batch culture was carried out using pH-stat feeding strategy. It is known that the respiration rate of aerobic microorganism is independent of DO above a certain critical level. However, below that level, a small change in DO concentration may cause a physiological alteration in cell metabolism. Therefore, a supply of oxygen to the growing cells is the rate limiting step in many aerobic fermentation processes [19]. According to the batch culture (Figure 2), the cell growth as well as protease tended to increase at lower rate when DO value was less than 20% air saturation. Thus the fed-batch fermentation was conducted at controlled conditions of agitation 400 rpm and aeration at 2 vvm and the DO concentration was controlled to be greater than 30% of air saturation throughout the fed-batch fermentation. Figure 4 depicts the time courses of fed-batch culture data. After 11 h of batch culture, feed medium was added to the culture broth, the pH of culture broth increased with nutrient consumption and the feed medium was added when pH rose above 7.5. The cell concentration increased up to 5.31 g/L at 42 h. A specific

    Table 4. Comparison of parameters of silk degumming protease production under different cultivation conditions.

    Treatment P m qp Yx/s Yp/s Reference

    Unoptimized medium 729 0.074 3,037.5 0.560 72.90 Romsomsa et al. [8]

    Optimized medium in shake flask 1,537 0.071 16,604 0.076 82.19 Romsomsa et al. [8]

    Batch culture (400 rpm 2 vvm) 1,890 0.084 21,412 0.073 83.48 This study

    pH-stat fed batch culture 4,437 0.012 36,820 0.125 140.22 This study

    P: maximum protease production (U/mL); m: Specific growth rate (h-1); qp: Specific protease production rate (U/g/h); Yx/s: yield of cell ( g. cell/g. total sugar consumed); Yp/s: yield of protease (kU/g. total sugar consumed).

  • Chiang Mai J. Sci. 2015; 42(3) 609

    growth rate of 0.012 h−1 was obtained and remained almost constant throughout the fermentation. The protease production increased with feeding and reached maximum value of 4,437.34 U/mL at 45 h, resulting in a 2.35-fold increase compared to the batch culture. The protease yield based on total sugar (Yp/s) registered a significant improvement from 72.90 to 140.22 kU/g total sugar (Table 4) and the productivity (Qp) from 70.00 to 98.61 kU/L/h. Therefore, pH-stat feeding strategy is effective for silk degumming protease production by Bacillus sp. C4 SS-2013 and these results were in agreement with other reports by which pH-stat feeding strategy was performed [24, 25].

    In the present study, Bacillus sp. C4 SS-2013 provided higher protease production in batch fermentation when compared with other researches as shown in Table 5 [11, 26-31]. The cessation in protease production was observed during batch process due to autoproteolysis and degradation by some proteolytic activity on the cell surface of nitrogen starved cells have been proposed [32]. However, the protease activity and product yield of fed-batch culture in this experiment resulted in higher values as compared to other report [33]. It would be expected to gain even the higher production if repeated fed-batch culture was employed consecutively. Moreover the optimized medium

    used here consisted of cheap raw materials (cassava and soybean flour) which are abundantly available in Thailand to lower the cost of production. Recently we improved the protease production approximately 6-fold compared with un-optimized medium (BMSM) in a shake flask (729 U/mL) and it would worthwhile to scale up to the industrial level.

    3.3 Enzyme Stability and Storage ConditionThough Bacillus sp. C4 SS-2013 can produce

    high protease in the optimized medium, it is not stable during storage. To improve the stability of crude protease, a preliminary study was undertaken on the effect of several additives such as CaCl2, sodium azide, glycerol, and polyethylene glycols (PEG), which were evaluated at room temperature (30 to 35°C), 4°C, and -20°C. As shown in Figure 5, the highest relative activity was obtained in the presence of 10 mM CaCl2 combined with 10% (v/v) PEG 400 at 4°C which remained at 67.05% of the initial activity after storage for 3 weeks. In the presence of 10 mM CaCl2, 25% glycerol, 0.02% sodium azide and 10% PEG 400, the activity remained at 66.27%, 47.47%, 42.11% and 55.28%, respectively, after storage at 4°C for 3 weeks. The crude protease stored at -20OC showed a very low activity (10.38%) and addition of 25% glycerol could not much improve the shelf-life (18.46%) (Figure 5). Glycerol acts as

    Figure 3. Analysis of native-PAGE (left 4 lanes) and gelatin zymogram (right 4 lanes) showing bands of different fermentation times. Lanes 1-4 (left to right): 0, 21, 22 and 24 h of crude protease, respectively.

  • Chiang Mai J. Sci. 2015; 42(3)610

    Table 5. Comparison of the batch fermentation data for protease production by various microorganisms.

    Microorganism Protease production (U/mL) Reference

    B. licheniformis UV-9 1,270 Nadeem et al. [11]

    B. subtilis 9.6 ul-Haq and Mukhtar [26]

    Bacillus sp. 896.5 Irfan et al. [27]

    B. megaterium IBRL MS 8.2 155.38 Darah et al. [28]

    Bacillus sp. 3.56 Prasad et al. [29]

    Pseudomonas putida SKG-1 882 Singh et al.[ 30]

    Rhizopus oryzae CH4 258.2 M’hir et al. [31]

    Bacillus sp. C4 SS-2013. 1,890 This study

    Figure 4. Fed-batch cultures for silk degumming protease production using pH-stat feeding strategy in a fermenter. Symbols: cell growth ( ), protease activity ( ), total volume of culture broth ( ), total sugar ( ).

    Figure 5. Relative activity of protease in the presence of various additives during storage for 3 weeks.

  • Chiang Mai J. Sci. 2015; 42(3) 611

    a cryoprotectant which prevents the formation of ice crystal during cold storage. However, the addition of 25% glycerol may not be sufficient to stabilize the protein structure and the final concentration of 25 to 50% will be more desirable. While storing at room temperature resulted in a substantial loss in its activity after the first week (data not shown). This result indicated that the suitable storage temperature was 4°C and either CaCl2 or CaCl2 with PEG 400 could be good stabilizers for the storage of crude protease. Ma et al. [34] also reported that CaCl2 is a good stabilizer for storage of enzymes. The improvement in protease stability against inactivation in the presence of Ca2+ may be explained by the strengthening of interactions inside protein molecules and by the binding of Ca2+ to autolysis sites [35] whereas PEG was reported to inhibit protein unfolding during the freezing step of lyophilization [36]. When the storage experiment of crude protease at 4°C with 10 mM CaCl2, 10% PEG 400 or their mixture was repeated for a longer time (8 weeks), the residual crude protease activity of these treatments were about 52%, 42% and 57%, respectively indicating that CaCl2 could maintain the shelf life of the crude protease as well as the mixture of CaCl2 and PEG 400. Hence, we decided to add only CaCl2 into the crude enzyme. We finally found the suitable concentration of CaCl2 should be 10 mM. Lower concentrations (1 mM) of CaCl2 had little effect on the stability, while increasing the level to 100 mM CaCl2 did not significantly improve the retention activity.

    CONCLUSIONSOptimization of the agitation and aeration

    rates to improve the growth and alkaline protease production of Bacillus sp. C4 SS-2013 by batch culture was performed in a fermenter using CCD and RSM. The generated model satisfied all the necessary arguments for its use. By fitting the experimental data to a second order

    polynomial equation, the optimum levels of agitation and aeration rates were determined as 400 rpm and 2 vvm, respectively, which provided the maximum enzyme activity of 1,890 U/mL and was characterized as partially growth associated. The pH-stat fed-batch culture showed a small increase in the cell growth. Protease production was improved to 4,437 U/mL which was a 2.35-fold increase compared to batch culture. The addition of 10 mM CaCl2 and storage at 4°C for 8 weeks could maintain the protease activity at about 52%.

    ACKNOWLEDGEMENTSThe authors would like to thank the Higher

    Education Research Promotion and National Research University Project of Thailand, Office of the Higher Education Commission and Center of Excellence for Silk, Kasetsart University, Thailand for providing the necessary financial support to complete this research.

    REFERENCES

    [1] Rajasekhar A., Ravi V., Reddy M.N. and Rao K.R.S.S., Thermostable bacterial protease-a new way for quality silk production, Int. J. Biosci. Biotechnol., 2011; 3(4): 43-58.

    [2] Yuksek M., Kocak D., Beyit A. and Merdan N., Effect of degumming performed with different type natural soaps and through ultrasonic method on the properties of silk fiber, Adv. Environ. Biol., 2012; 6(2): 801-808.

    [3] The National Bureau of Agricultural Commodity and Food Standards (ACFS), Thai silk yarn, Thai agricultural standard TAS 8000-2005; Available at: http://www.centrallabthai.com/web/uploadfiles/pdf/MRLs/Thailand/SILK/silk_eng.pdf. Accessed 15 March 2014.

    [4] Freddi G., Mossotti R. and Innocenti R., Degumming of silk fabric with several proteases, J. Biotechnol., 2003; 106: 101-112. DOI 10.1016/j.jbiotec.2003.09.006.

  • Chiang Mai J. Sci. 2015; 42(3)612

    [5] Arami M., Rahimi S., Mivehie L., Mazaheri F. and Mahmoodi N.M., Degumming of Persian silk with mixed proteolytic enzymes, J. Appl. Polym. Sci., 2007; 106: 267-275. DOI 10.1002/app.26492.

    [6] Suwannaphan S., Partial Purification and Some Characterization of Sericin Degrading Protease from Bacillus subtilis C4, Master Thesis, Kasetsart University, Thailand, 2010.

    [7] Fufeungsombut E., Chim-anage P., Promboon A. and Suwannaphan S., Isolation and Selection of Silk Degumming Protease Producing Bacteria from Thailand, Proceedings of the 47th Kasetsart University Annual Conference, Bangkok, Thailand, 17-20 March 2009; 456-463.

    [8] Romsomsa N., Chim-anage P. and Jangchud A., Optimization of silk degumming protease production from Bacillus subtilis C4 using Plackett-Burman design and response surface methodology, ScienceAsia, 2010; 36: 118-124. DOI 10.2306/scienceasia1513-1874.2010.36.118.

    [9] Converti A., Zilli M., Arni S., Di Felice R. and Del Borghi M., Estimation of viscosity of highly viscous fermentation media containing one or more solutes, Biochem. Eng. J., 1999; 4: 81-85.

    [10] Mantzouridou F., Roukas T. and Kotzekidou P., Effect of the aeration rate and agitation speed on b-carotene production and morphology of Blakeslea trispora in a stirred tank reactor: Mathematical modeling, Biochem. Eng. J., 2002; 10: 123–135. DOI 10.1016/S1369-703X(01)00166-8.

    [11] Nadeem M., Qazi J.I. and Baig S., Effect of aeration and agitation rates on alkaline protease production by Bacillus licheniformis UV-9 mutant, Turk. J. Biochem., 2009; 34(2): 89-96.

    [12] Reddy L.V.A., Wee Y.J., Yun J.S. and Ryu H.W., Optimization of alkaline protease by batch culture of Bacillus sp. RKY3 through Placket-Burman and response surface methodological

    approaches, Bioresour. Technol., 2008; 99: 2242-2249. DOI 10.1016/j.biortech.2007.05.006.

    [13] Horikoshi K., Microorganisms in Alkaline Environments, Kodansha, Tokyo, 1991.

    [14] Wise W.S., The measurement of aeration of culture media, J. Gen. Microbiol., 1951; 5: 167-177.

    [15] García-Carreño F.L., Dimes L.E. and Haard N., Substrate-gel electrophoresis for composition and molecular weight of proteinases or proteinaceous proteinase inhibitor, Anal. Biochem., 1993; 214: 65-69.

    [16] Dubois M., Gilles K.A., Hamilton J.K., Rebers P.A. and Smith F., Colorimetric method for determination of sugar and related substance, Anal. Chem., 1956; 28: 350-356. DOI 10.1021/ac60111a017.

    [17] Prakasham R.S., Rao C.S., Rao R.S. and Sarma P.N., Enhancement of acid amylase production by an isolated Aspergillus awamori, J. Appl. Microbiol., 2007; 102: 204-211. DOI 10.1111/j.1365-2672.2006.03058.x.

    [18] Ferrero M.A., Cartro G.R., Abate C.M., Baigori M.D. and Sineriz F., Thermostable alkaline protease of Bacillus licheniformis MIR 29: Isolation, production and characterization, Appl. Microbiol. Biotechnol., 1996; 45 :327-332.

    [19] Thiry M. and Cingolani D., Optimizing scale up fermentation process, Trends Biotechnol., 2002; 20: 103-105. DOI 10.1016/S0167-7799(02)01913-3.

    [20] Bandaiphet C. and Prasertsan P., Effect of aeration and agitation rates and scale-up on oxygen transfer coefficient, kLa in exopolysaccharide production from Enterobacter cloacae WD7, Carbohydr. Polym., 2006; 66: 216-228. DOI 10.1016/j.carbpol.2006.03.004.

    [21] Rosma A., and Ooi K.I., Production of Candida utilis biomass and intracellular protein content: Effect of agitation speed and aeration rate, Mal. J. Microbiol., 2006; 2(2): 15-18.

  • Chiang Mai J. Sci. 2015; 42(3) 613

    [22] Gibbs P.A. and Seviour R.J., Does the agitation rate and/or oxygen saturation influence exopolysaccharide production by Aureobasdium pullulans in batch culture?, Appl. Microbiol. Biotechnol., 1996; 46: 503-510.

    [23] Park Y.S., Kai K., Lijima S. and Kobayashi T., Enhanced b-galactosidase production by high cell-density culture of recombinant Bacillus subtilis with glucose concentration control, Biotechnol. Bioeng., 1992; 40: 686-696. DOI 10.1002/bit.260400607.

    [24] Kuo C.C., Lin C., Chen J.Y., Lin M.T. and Duan K.J., Production of cyclodextrin glucanotransferase from an alkalophilic Bacillus sp. by pH-stat fed-batch fermentation, Biotechnol. Lett., 2009; 31: 1723–1727. DOI 10.1007/s10529-009-0071-9

    [25] Cho Y.H., Song J.Y., Kim K.M., Kim M.K., Lee I.Y., Kim S.B., Kim H.S., Han N.S., Lee B.H. and Kim B.S., Production of nattokinase by batch and fed-batch culture of Bacillus subtilis, New Biotechnol., 2010; 27(4): 341-346. DOI 10.1016/j.nbt.2010.06.003.

    [26] ul-Haq I. and Mukhtar H., Fuzzy logic control of bioreactor for enhanced biosynthesis of alkaline protease by an alkalophilic strain of Bacillus subtilis, Curr. Microbiol., 2006; 52(2): 149-152.

    [27] Irfan M., Nadeem M., Syed Q., Nawaz W., Baig S. and Qureshi A. M., Relationship of process parameters for the production of alkaline protease by Bacillus sp., Int. J. Agro Vet. Med. Sci., 2010; 4(4): 114-120.

    [28] Darah I., Choo M.L. and Lim S.H., Influence of physical and chemical parameters on protease production by Bacillus megaterium IBRL MS 8.2 in submerged fermentation, Res. J. Biol. Sci., 2013; 8(2): 54-61. DOI 10.3923/rjbsci.2013.54.61.

    [29] Prasad R., Abraham T.K. and Nair A.J., Scale up of production in a bioreactor of a halotolerant protease from moderately halophilic Bacillus sp. isolated from soil, Braz. Arch. Biol. Technol.,

    2014; 57(3): 448-455. DOI 10.1590/s1516-89132014005000014.

    [30] Singh S.K., Singh S.K., Tripathi V.R., Khare S.K. and Garg S.K., Comparative one-factor-at-a-time, response surface (statistical) and bench-scale bioreactor level optimization of thermoalkaline protease production from a psychrotrophic Pseudomonas putida SKG-1 isolate, Microb. Cell Fact., 2011; 10: 114. DOI 10.1186/1475-2859-10-114.

    [31] M’hir S., Mejri A., Sifaoui I., Slama M.B., Mejri M., Thonart P.and Hamdi. M., Improvement of protease production by Rhizopus oryzae CH4 grown on wheat gluten using response surface methodology and its scale-up in a bioreactor, Arch. Appl. Sci. Res., 2012; 4(2): 1110-1116.

    [32] Beg Q.K., Sahai V. and Gupta R., Statistical media optimization and alkaline protease production from Bacillus mojavensis in a bioreactor, Process Biochem., 2003; 39(2): 203-209. DOI 10.1016/S0032- 9592(03)00064-5.

    [33] Tabandeh F., Hosseinian Moghaddam H.R., Yakhchali B., Shariati P., Hamed Mousavian M.T., and Ghasemi F., Fed-batch fermentation of Bacillus clausii for efficient production of alkaline protease using different feeding strategies, Chem. Eng. Comm., 2011; 198: 1063–1074. DOI 10.1080/00986445.2011.552024.

    [34] Ma X., Wang Z., Li S., Shen Q. and Yuan Q., Effect of CaCl2 as activity stabilizer on purification of heparinase I from Flavobacterium heparinum, J. Chromatogr. B. Anal. Technol. Biomed. Life Sci., 2006; 843(2): 209-215. DOI 10.1016/j.jchromb.2006.06.015.

    [35] Purich D.L., Advances in Enzymology, John Wiley and Sons. Inc., New York, 2000.

    [36] Taravati A., Shokradeh M., Ebadi A.G., Valipour P., Hassan A.T.M., and Farrokhi F., Various effects of sugar and polyols on the protein structure and function: Role as osmolyte on protein stability, World Appl. Sci. J., 2007; 2(2): 353-362.