Response surface methodology (RSM) was employed to optimize the medium composition and culture conditions for the production of alkaline proteases by Bacillus mojavensis A21 on uncommon substrates: chickpea (CF) and faba bean (FF) flours. A significant positive influence of temperature, CF, FF, incubation time and inoculums size on the protease production was evaluated by Plackett Burman Design. Among these, CF was the most influential factor. The enhancement of protease to 9127 U/ml was achieved with the optimization procedure on the medium composed of (g/l): CF, 40; FF 30, NaCl 2.0; KH2PO4 1; K2HPO4 1; CaCl2, 0.1; MgSO4 0.1. The cultures were conducted for 72 hours with an IS of 2%, at 30°C, an agitation speed of 150 rpm and an initial pH of 8.0. More interestingly, the optimization was accomplished using two cheap and local fermentation substrates, CF and FF, which could result in a significant reduction in the cost of medium constituents. The maximum alkaline protease production was 9127 U/ml after 72 h of incubation and showed 5-fold increase in protease production over the initial level.
Microorganisms are essential sources for enzyme production for industry. Among these enzymes, proteases account for nearly 60% of the total industrial enzyme market [
The use of alkaline proteases has increased remarkably in many industrial processes, including the production of detergents, food processing, in animal feed and for silver regeneration of X-ray films [
In general, no defined medium could be carried out for the production of alkaline proteases from different microorganisms; each strain has its specific required conditions for maximum enzyme production [
Optimization of media compounds by the traditional “one-variable at-a-time” strategy is the most frequently used method in biotechnology [
Bacillus mojavensis A21 has been recently identified as a producer of extracellular bleaching-stable alkaline proteases. Haddar et al., 2009 [
In view of the promising applicability of the alkaline proteases as a builder for detergents, there is an interest in producing these enzymes in the highest yields with the lowest cost fermentation media. Herein, we report a statistical optimization of culture conditions for the production of proteases by B. mojavensis A21, on uncommon fermentation substrates such as faba bean and chickpea flours.
B. mojavensis A21 producing bleach-stable alkaline proteases was isolated from a marine water sample. It was identified on the basis of 16 S rRNA gene sequencing [
Seeds of chickpea (Cicer arietinum), faba bean (Vicia faba), lens (Lens culinaris), corn (Zea mays), pea (Pisum sativum) and mil (Pennisetum glaucum) were purchased in local market and minced in our laboratory to obtain the corresponding flours: CF, FF, LF, CNF, PF and MF. Sardinella peptone (SP) [
B. mojavensis A21 was maintained on Luria-Bertani (LB) medium [
Protease activity was measured according to the method of Kembhavi et al. (1993) [
A Plackett-Burman design is used for rapid screening multifactor to find the most significant independent factors [
Box-Behnken design of RSM (Response Surface Methodology) [
The protease production was fitted using a second-order polynomial equation and a multiple regression of the data was carried out for obtaining an empirical model related to the factors. The general form of the second-order polynomial equation is:
where Y is the predicted response,
Variables | Units | Symbol code | Lower (−1) | Higher (+1) |
---|---|---|---|---|
CF | g/l | X1 | 10 | 40 |
FF | g/l | X2 | 10 | 30 |
CaCl2 | g/l | X3 | 0.5 | 2 |
NaCl | g/l | X4 | 0.5 | 2 |
KH2PO4 | g/l | X5 | 0.1 | 1 |
K2HPO4 | g/l | X6 | 0.1 | 1 |
MgSO4 | g/l | X7 | 0.1 | 1 |
Temperature | ˚C | X8 | 30 | 37 |
Speed of agitation | rpm | X9 | 150 | 250 |
Time | H | X10 | 24 | 72 |
IS | % | X11 | 2 | 10 |
CF: chickpea flour; FF: faba bean flour and IS: inoculums size.
Run | X1 | X2 | X3 | X4 | X5 | X6 | X7 | X8 | X9 | X10 | X11 | Proteolytic activity (U/ml) |
---|---|---|---|---|---|---|---|---|---|---|---|---|
1 | 1 | 1 | −1 | 1 | 1 | 1 | −1 | −1 | −1 | 1 | −1 | 9127 |
2 | −1 | 1 | 1 | −1 | 1 | 1 | 1 | −1 | −1 | −1 | 1 | 463 |
3 | 1 | −1 | 1 | 1 | −1 | 1 | 1 | 1 | −1 | −1 | −1 | 663 |
4 | −1 | 1 | −1 | 1 | 1 | −1 | 1 | 1 | 1 | −1 | −1 | 1227 |
5 | −1 | −1 | 1 | −1 | 1 | 1 | −1 | 1 | 1 | 1 | −1 | 236 |
6 | −1 | −1 | −1 | 1 | −1 | 1 | 1 | −1 | 1 | 1 | 1 | 1190 |
7 | 1 | −1 | −1 | −1 | 1 | −1 | 1 | 1 | −1 | 1 | 1 | 1163 |
8 | 1 | 1 | −1 | −1 | −1 | 1 | −1 | 1 | 1 | −1 | 1 | 1581 |
9 | 1 | 1 | 1 | −1 | −1 | −1 | 1 | −1 | 1 | 1 | −1 | 8481 |
10 | −1 | 1 | 1 | 1 | −1 | −1 | −1 | 1 | −1 | 1 | 1 | 2045 |
11 | 1 | −1 | 1 | 1 | 1 | −1 | −1 | −1 | 1 | −1 | 1 | 2281 |
12 | −1 | −1 | −1 | −1 | −1 | −1 | −1 | −1 | −1 | −1 | −1 | 636 |
13 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2581 |
14 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2490 |
15 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2527 |
16 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2493 |
Variables X1 to X11 are described in
Factor | Units | Levels | ||
---|---|---|---|---|
−1 | 0 | 1 | ||
CF | g/l | 10 | 25 | 40 |
FF | g/l | 10 | 20 | 30 |
Time | H | 24 | 48 | 72 |
Temperature | ˚C | 30 | 33,5 | 37 |
IS | % | 2 | 6 | 10 |
coefficient,
Design-expert, version 7.0 (STAT-EASE Inc., Minneapolis, USA) was used for the experimental designs and statistical analysis of the experimental data. The analysis of variance (ANOVA) was used to estimate the statistical parameters.
Dry weights of chickpea and faba bean flours were determined after heating samples at 105˚C to constant weight, and then ash content was determined after heating dried samples at 600˚C for 2 h. Total nitrogen was determined using the Kjeldahl method and then protein content was estimated. Crude fat was determined gravimetrically after Soxhlet extraction of dried samples with diethyl ether. Starch and cellulose were estimated as described by Ewers [
Run | CF | FF | Time | Temperature | IS | Proteolytic activity (U/ml) | |
---|---|---|---|---|---|---|---|
Experimental | Predicted | ||||||
1 | 10 | 10 | 48 | 33.5 | 6 | 672 | 624 |
2 | 40 | 10 | 48 | 33.5 | 6 | 2372 | 2071 |
3 | 10 | 30 | 48 | 33.5 | 6 | 1218 | 1451 |
4 | 40 | 30 | 48 | 33.5 | 6 | 3745 | 2898 |
5 | 25 | 20 | 24 | 30 | 6 | 1290 | 1608 |
6 | 25 | 20 | 72 | 30 | 6 | 2736 | 2671 |
7 | 25 | 20 | 24 | 37 | 6 | 1554 | 1601 |
8 | 25 | 20 | 72 | 37 | 6 | 1500 | 1164 |
9 | 25 | 10 | 48 | 33.5 | 2 | 1390 | 1348 |
10 | 25 | 30 | 48 | 33.5 | 2 | 1936 | 2174 |
11 | 25 | 10 | 48 | 33.5 | 10 | 1300 | 1348 |
12 | 25 | 30 | 48 | 33.5 | 10 | 2290 | 2174 |
13 | 10 | 20 | 24 | 33.5 | 6 | 1854 | 1347 |
14 | 40 | 20 | 24 | 33.5 | 6 | 1654 | 1862 |
15 | 10 | 20 | 72 | 33.5 | 6 | 1090 | 728 |
16 | 40 | 20 | 72 | 33.5 | 6 | 2754 | 3107 |
17 | 25 | 20 | 48 | 30 | 2 | 1909 | 2139 |
18 | 25 | 20 | 48 | 37 | 2 | 1354 | 1383 |
19 | 25 | 20 | 48 | 30 | 10 | 2081 | 2139 |
20 | 25 | 20 | 48 | 37 | 10 | 918 | 1383 |
21 | 25 | 10 | 24 | 33.5 | 6 | 954 | 1191 |
22 | 25 | 30 | 24 | 33.5 | 6 | 1654 | 2018 |
23 | 25 | 10 | 72 | 33.5 | 6 | 1272 | 1504 |
24 | 25 | 30 | 72 | 33.5 | 6 | 2500 | 2331 |
25 | 10 | 20 | 48 | 30 | 6 | 1190 | 1416 |
26 | 40 | 20 | 48 | 30 | 6 | 2500 | 2863 |
27 | 10 | 20 | 48 | 37 | 6 | 900 | 659 |
28 | 40 | 20 | 48 | 37 | 6 | 2509 | 2106 |
29 | 25 | 20 | 24 | 33.5 | 2 | 2290 | 1605 |
30 | 25 | 20 | 72 | 33.5 | 2 | 2018 | 1917 |
31 | 25 | 20 | 24 | 33.5 | 10 | 1845 | 1605 |
32 | 25 | 20 | 72 | 33.5 | 10 | 1727 | 1917 |
33 | 10 | 20 | 48 | 33.5 | 2 | 1036 | 1038 |
34 | 40 | 20 | 48 | 33.5 | 2 | 3127 | 2484 |
35 | 10 | 20 | 48 | 33.5 | 10 | 1263 | 1038 |
36 | 40 | 20 | 48 | 33.5 | 10 | 2136 | 2484 |
37 | 25 | 10 | 48 | 30 | 6 | 2054 | 1726 |
38 | 25 | 30 | 48 | 30 | 6 | 2918 | 2553 |
39 | 25 | 10 | 48 | 37 | 6 | 763 | 970 |
40 | 25 | 30 | 48 | 37 | 6 | 1127 | 1796 |
41 | 25 | 20 | 48 | 33.5 | 6 | 1681 | 1761 |
42 | 25 | 20 | 48 | 33.5 | 6 | 2000 | 1761 |
43 | 25 | 20 | 48 | 33.5 | 6 | 1563 | 1761 |
44 | 25 | 20 | 48 | 33.5 | 6 | 1254 | 1761 |
45 | 25 | 20 | 48 | 33.5 | 6 | 1672 | 1761 |
46 | 25 | 20 | 48 | 33.5 | 6 | 1436 | 1761 |
B. mojavensis A21 strain was cultivated on initial basal medium containing YE at 1 g/l as nitrogen source and different complex sources of carbon (30 g/l) such as chickpea, faba bean, lens, corn, pea and mil. As shown in
Our results are in concordance with other reports regarding proteases production, in which it has been observed that complex carbon sources such hulled grain of wheat [
In a second experiment, protease production by B. mojavensis A21 was checked in the basal medium containing CF at 30 g/l and supplemented with 1 g/1 of various nitrogen sources YE: yeast extract, SP: sardinella peptone, CHVSP: combined heads and viscera sardinella powder, HF: hydrolyzed feathers. As shown in
In order to substitute YE by a low cost and available nitrogen source, FF was assayed at different concentrations from 10 to 40 g/l in basal medium containing 40 g/l of CF as carbon source. Protease production was enhanced with the increase of FF concentration up to 30 g/l (
In this study, two kinds of legume seeds were shown to be excellent carbon and nitrogen sources for B. mojavensis A21 proteases production which was maximized in the presence of CF. This can be explained by the fact that the two respective flours contain high amounts of proteins.
The Plackett-Burman design is a powerful method for screening significant factors. Sixteen runs were carried out to analyze the effect of 11 variables on protease production and the results are demonstrated in
A response surface design is further applied when the optimal region for running the process has been identified. Based on the Plackett-Burman design, RSM using Box-Behnken design was applied to determine the optimal levels of the five selected variables (CF, FF, temperature, incubation time and IS) which significantly influenced the protease production. The respective low and high levels with the coded levels for the five variables are defined in
Y = 1761(±52.93) + 723.375(±89.75) CF + 413.1875(±89.75) FF + 156.375(±89.75) Time − 378.3125(±89.75) Temperature + 466(±179.50) CF × Time − 375(±179.50) Time × Temperature
where Y the protease production by B. mojavensis A21.
The statistical significance of the model equation was evaluated by the F-test for ANOVA. The model F-value of 19.68 implies the model is significant. There was only a 0.01% chance that the model F-value could occur due to noise. The lack of fit F-value of 2.16 implies that was not significant relative to the pure error and that was a 19.83 % chance that the lack of fit F-value could occur due to noise. The adequate precision measures the
Source | Sum of squares | Df | Mean square | F-value | p-value |
---|---|---|---|---|---|
Model | 15,216,225 | 6 | 2536037.4 | 19.68 | <0.0001* |
CF | 8,372,342 | 1 | 8372342.3 | 64.96 | <0.0001* |
FF | 2,731,583 | 1 | 2731582.6 | 21.20 | <0.0001* |
Time | 391,250 | 1 | 391250.3 | 3.04 | 0.0893 |
Temperature | 2,289,926 | 1 | 2289925.6 | 17.77 | 0.0001* |
CF × Time | 868,624 | 1 | 868624.0 | 6.74 | 0.0132* |
Time × Temperature | 562,500 | 1 | 562500.0 | 4.36 | 0.0433* |
Residual | 5,026,193 | 39 | 128876.8 | ||
Lack of fit | 4,706,473 | 34 | 138425.7 | 2.16 | 0.1983 |
Pure error | 319,720 | 5 | 63944.0 | ||
Cor Total | 20,242,418 | 45 |
Standard Deviation = 358.99; R2 = 0.752, adjust R2 = 0.714; *Statistically significant at 95% of confidence level.
signal to noise ratio. A ratio greater than 4 is desirable. Adequate precision which was calculated to be 17.73 indicates an adequate signal. This model can be used to navigate the design space.
According to the model, we can notice that the increasing of the temperature causes to a decrease in the protease production.
In a first attempt to optimize B. mojavensis A21 proteases production by statistical design on wheat bran flour and sardinella peptone, Haddar [
Using cost-effective media formulation and optimizing this media to determine its minimum requirements for maximum enzyme production is extremely important in industrial-scale protease production for economic reasons. Therefore, using common and local low cost substrates such as chickpea and faba bean, will serve as a potential example for the applications in industrial microbial fermentations. The simple expedient of replacing YE in the medium with CF or FF could significantly lower the production cost.
Maximum proteases production of 9127 U/ml was achieved at the optimized culture conditions: 40 g/l CF, 30 g/l FF, NaCl 2 g/l; KH2PO4 1 g/l; K2HPO4 1 g/l; CaCl2, 0.5 g/l; MgSO4 0.1 g/l and IS 2% (v/v). In addition other factors were taken at suitable levels, as shown in the Plackett-Burman design matrix: a temperature of 30˚C, a speed of agitation at 150 rpm, incubation time of 72 h and an initial pH of 8.0.
The time courses of protease activity and the growth of B. mojavensis A21 under the optimized conditions are shown in
Components % | FF | CF |
---|---|---|
Proteins | 29.08 ± 1.55 | 23.64 ± 1.07 |
Fat | 2.18 ± 0.26 | 6.48 ± 0.42 |
Humidity | 13.04 ± 0.77 | 15.35 ± 0.80 |
Starch | 42.24 ± 1.28 | 46.91 ± 0.64 |
Cellulose | 9.61 ± 0.84 | 3.90 ± 1.27 |
Ash | 3.85 ± 0.21 | 3.72 ± 0.50 |
The chemical compositions of CF and FF were determined (
The two pulses contained important quantities of proteins and starch which are higher than in hulled grain of wheat (HGW) 8% - 12% and 50% - 60% respectively [
Thus, the use of CF and FF as carbon and nitrogen sources may result in a more cost-effective process, and so the production of detergent proteases by the A21 strain will be effectively economic.
Due to the increasing economic relevance of alkaline proteases, this study was conducted in an attempt to optimize a variety of fermentation parameters, including medium compositions and culture conditions, for maximal alkaline protease production. Eleven variables were tested using the Plackett-Burman design, and five variables (CF, FF, incubation time, temperature and IS) were selected as the most influential factors on enzyme production. The maximum alkaline protease production was amplified by 5 folds over the initial level and reached 9127 U/ml after 72 h of incubation. The optimized medium established in this work might result in a significant reduction in the cost of medium constituents and would thus offer advantages for large-scale fermentations.
The authors are thankful to Mrs Imen Khouni from the Centre de Biotechnologie Borj Cédria (Tunisia) for starch and cellulose determination.
The present work was supported by the Ministère de l’Enseignement Supérieur et de la Recherche Scienti- fique, Tunisia.