Media Optimization for Laccase Production by Trichoderma harzianum ZF-2 Using Response Surface Methodology

Abstract

Trichoderma harzianum ZF-2 producing laccase was isolated from decaying samples from Shandong, China, and showed dye decolorization activities. The objective of this study was to optimize its culture conditions using a statistical analysis of its laccase production. The interactions between different fermentation parameters for laccase production were characterized using a Plackett-Burman design and the response surface methodology. The different media components were initially optimized using the conventional one-factor-at-a-time method and an orthogonal test design, and a Plackett-Burman experiment was then performed to evaluate the effects on laccase production. Wheat straw powder, soybean meal, and $CuSO_4$ were all found to have a significant influence on laccase production, and the optimal concentrations of these three factors were then sequentially investigated using the response surface methodology with a central composite design. The resulting optimal medium components for laccase production were determined as follows: wheat straw powder 7.63 g/l, soybean meal 23.07 g/l, $(NH_4)_2SO_4$ 1 g/l, $CuSO_4$ 0.51 g/l, Tween-20 1 g/l, $MgSO_4$ 1 g/l, and $KH_2PO_4$ 0.6 g/l. Using this optimized fermentation method, the yield of laccase was increased 59.68 times to 67.258 U/ml compared with the laccase production with an unoptimized medium. This is the first report on the statistical optimization of laccase production by Trichoderma harzianum ZF-2.

Introduction

Laccases (benzenediol: oxygen oxidoreductase, EC 1.10.3.2) are oxidoreductases. They belong to the phenol-oxidase family and can catalyze the oxidation of both phenolic and aromatic amines by using molecular oxygen as the electron acceptor. The broad substrate spectrum of laccases has resulted in a large number of biotechnological applications. In particular, their ability to mineralize a wide range of synthetic dyes makes this biocatalyst suitable for application in several bioprocesses, including bioremediation, biobleaching, biopulping, and the treatment of industrial wastewater [1,10,18,35]. Laccases have also been used in enzyme immunoassays as a marker enzyme [34], in the design of various biosensors [12], and in energy transformation systems [5].

Fungi have been shown as the most promising sources for laccase production [3,7,8,19,28], especially basidiomycete fungi, including Trametes versicolor [16], Phlebia radiata [15], and Penicillium chrysogenum [26]. Whereas there have only been a few reports on ascomycete and deuteromycete laccases, there have been many studies of Trichoderma species, from the moniliales subdivision of deuteromycotina, as potential sources of cellulase enzymes for the commercial hydrolysis of cellulosic materials and as active participants in the biodegradation of cellulose in nature. Assavanig et al. [2] described extracellular laccases in 14 different natural isolates of Trichoderma sp. Moreover, extracellular laccases have also been isolated from Trichoderma atroviride, Trichoderma harzianum, and Trichoderma longibrachiatum [11,13]. Some white-rot fungi increase laccase production when mixing the culture with certain Trichoderma sp. [27,33]. Thus, it is of interest to determine whether Trichoderma strains with laccase activity are capable of more extensive natural substrate hydrolysis than strains without laccase activity.

Laccases can be either constitutive or inducible enzymes, plus their production occurs during secondary metabolism and is subject to complex regulation by nutrients (C, N, inducers, and copper) [6,17,20,29]. These regulators affect the transcription levels of laccase and other genes [23,25,30]. Many efforts have already been made to enhance laccase production by optimizing the nutrient media compositions in fungal fermentation [14,21,24,31,32].

In our previous studies, a Trichoderma sp. strain producing laccase, named ZF-2, was isolated from decaying samples from Shandong, China. The taxonomy of the fungus was confirmed as Trichoderma harzianum based on morphological and 5.8S rDNA/ITS analyses (Accession No. HM05119). Brilliant Blue KN-R, Basic Violet, and Congo Red were all decolorized by its laccase. Accordingly, to maximize its laccase production, this study optimized the different media components using the conventional one-factor-at-a-time method, and then further optimized the critical parameters using the response surface methodology (RSM).

 

Materials and Methods

Microorganism

Trichoderma harzianum ZF-2 was isolated from decaying samples from Shandong, China and maintained on slants of Potato Dextrose Agar (PDA ) at 4℃. Subculturing was performed every fifteenth day to maintain its viability.

Culture Conditions and Enzyme Extraction

The basal medium contained 20 g glucose, 1 g (NH4)2SO4, 1 g MgSO4, 0.6 g KH2PO4, and distilled water to make 1 liter. The ZF-2 strain was inoculated into the liquid culture medium and mixed thoroughly by keeping the flasks on a rotary shaker at 150 rpm for 36 h at 28℃. The culture liquid was then centrifuged at 10,000 rpm for 20 min at 4℃, and the supernatant collected and used for the enzyme assay.

Laccase Activity Assay

The laccase activity was measured spectrophotometrically using guaiacol as the substrate according to Gao et al. [9]. Briefly, an aliquot of the enzyme solution was incubated in 4.0 ml of a 50 mM sodium acetate buffer (pH4.5) containing 1.0 mM guaiacol at 30℃. The changes in absorbance due to the oxidation of guaiacol in the reaction mixture were monitored at 465 nm (ε465 = 12,100 M-1 cm-1). One unit of enzyme activity was defined as the amount of enzyme that oxidized 1 μmol of guaiacol per minute under the above conditions.

Medium Optimization Using One-Factor-at-a-Time

The production of laccase was optimized based on varying the carbon sources, nitrogen sources, carbon/nitrogen ratios, surfactants, and concentrations of copper as part of the culture conditions. Eight carbon sources (glucose, maltose, lactose, sucrose, soluble starch, wheat straw powder, corn flour, and bran), nine nitrogen sources (organic nitrogen sources: urea, soybean meal, yeast extract, beaf extract, and peptone; inorganic nitrogen sources: ammonium chloride, sodium nitrate, ammonium sulfate, and ammonium nitrate), five surfactants (Tween-80, Tween-20, resveratrol, guaiacol, and gallic acid), and five gradient concentrations of copper (0.01, 0.05, 0.1, 0.2, and 0.4 g/l) were used.

Plackett-Burman Experimental Design

Based on the results of a preliminary study on the production of laccase by strain ZF-2, a Plackett-Burman (PB) experiment was performed to identify the significant variables affecting the enzyme production. A set of 12 experiments was conducted. Each variable was set at two levels, high and low, denoted by (+1) and (-1), respectively, where the low level was the optimal condition and the high level was 1.5 times the low level. The response value represented the laccase activity, and three replicates were performed per treatment.

Central Composite Design and Response Surface Methodology

A central composite design was adopted to optimize the effect of the major variables (wheat straw powder, soybean meal, and CuSO4) on laccase production. The effect of each variable was studied at five different levels, namely -α, 1, 0, +1, +α, and the set of 20 experiments was performed in triplicate. The central coded value for all the variables was set at zero. The data obtained from the RSM on laccase production were subjected to analysis of variance (ANOVA). The results of the RSM were used to fit a second-order polynomial, Eq. (1), to represent the behavior of the system:

where Y is the response variable representing laccase activity, b0 is the intercept, b1, b2, b3 are the linear coefficients, b1,1, b2,2, b3,3 are the squared coefficients, b1,2, b1,3, b2,3 are the interaction coefficients, and X1, X2, X3, X12, X22, X32, X1X2, X1X3, X2X3 are the levels of the independent variables. The data analysis and generation of the response surface graphs were conducted using the statistical software Design Expert (Design-Expert 7.0).

 

Results

Effects of Carbon Source, Nitrogen Source, Copper, and Surfactant on Laccase Production

The effects of seven carbon sources on laccase production were investigated by replacing the glucose in the basal medium. Among the carbon sources screened, the organic carbon sources all improved the laccase activity (Table 1). However, corn flour and bran showed a lower improvement than wheat straw powder, whereas maltose and soluble starch produced the lowest laccase levels. When the glucose in the basal medium was replaced with wheat straw powder, the laccase activity increased to 5.242 U/ml. Thus, wheat straw powder was used as the carbon source for the next fermentation.

Table 1.Effects of different compounds on laccase production by strain ZF-2.

(NH4)2SO4 was the nitrogen source in the basal medium for laccase production, and it was replaced by various organic and inorganic nitrogen sources. Among the organic nitrogen sources tested, soybean meal had a good effect on laccase production, but peptone, beef extract, and yeast extract had an even better effect. Conversely, urea produced a small decrease in laccase activity (Table 1). Although the inorganic nitrogen sources were not as effective as the organic sources, (NH4)2SO4 was shown to be better than the others for the production of laccase (Table 1).

Table 2Analysis of variance for orthogonal test.

To determine whether the copper concentration in the basal mineral medium would affect the laccase activity, the basal medium was supplemented with various concentrations of CuSO4. Whereas the lower and higher concentrations both partially inhibited laccase activity, the highest laccase activity of 27.883 U/ml was achieved with a CuSO4 concentration of 0.2 g/l (Table 1).

The effect of a surfactant on laccase production was investigated by supplementing the basal medium with Tween, resveratrol, gallic acid, and guaiacol. The results showed that only Tween-20 significantly increased the laccase activity up to 34.452 U/ml. Meanwhile, Tween-80, resveratrol, and guaiacol had no obvious effect on laccase production, and gallic acid even had a negative effect (Table 1).

Effect of Carbon/Nitrogen (C/N) Ratio on Laccase Production

An L16 (43) orthogonal test design was applied to optimize the levels of wheat straw powder, (NH4)2SO4, and soybean meal for laccase production. The results showed that the maximum laccase activity of 12.49 U/ml was achieved with 1% wheat straw powder, 0.8% soybean meal, and 0.4% (NH4)2SO4. The analysis of variance also showed that the wheat straw powder and soybean meal had a more significant effect on laccase production (P < 0.01), while the effect of (NH4)2SO4 was less significant (P > 0.05). Thus, the best C/N ratio was determined as 1% wheat straw powder, 0.8% soybean meal, and 0.1% (NH4)2SO4, which was used for the PB testing (Table 2).

Identification of Significant Variables Using Plackett- Burman Design

The first optimization step identified the significant factors for laccase production using a 12-run Plackett-Burman design. Variations ranging from 20.606 to 48.371 U/ml in the production of laccase were observed (Table 3), which indicated that optimization is important to attain higher productivity. Statistical analyses of the responses were performed and the results are presented in Table 4. A value of Pr < 0.05 indicates that the model term is significant. As a result, three factors were found to have a significant effect on laccase production: wheat straw powder, soybean meal, and CuSO4. Wheat straw powder showed the highest significance and a negative effect, and thus its level was reduced in the subsequent experiments. Meanwhile, soybean meal and CuSO4 had a positive effect on laccase production, where soybean meal had a stronger effect than CuSO4. Therefore, the levels of soybean meal and CuSO4 were both increased in the subsequent experiments. As Tween-20 had no significant effect on laccase production, it was kept at a low level. The observed responses were used to compute the model coefficients using the least-square method. The model (Eq. (2)) for laccase production was calculated as

where X1, Wheat straw powder; X2, Soybean meal; X3, CuSO4; and X4, Tween-20. Since the statistical model was able to explain 96.8% of the variability in the response, it was concluded that the model fitted well with the measured data.

Table 3.Experimental conditions and response values of Plackett-Burman test for laccase production by strain ZF-2.

Table 4.aStatistically significant at 95% of confidence level (Pr < 0.05).

Statistical Optimization Using Central Composite Design

Based on the PB tests, statistical optimization was used to determine the optimal concentrations of wheat straw powder, soybean meal, and CuSO4 for laccase production. The effect of each variable on enzyme production was characterized at five different levels, namely -1.68, 1, 0, +1, +1.68. Thus, a set of three factors with five levels and a total of 20 treatments were performed. The results of the response surface experiments to determine the effects of wheat straw power, soybean meal, and CuSO4 are presented in Table 5.

The determination coefficient (R2) value of 0.9645 indicates that the statistical model was able to explain 96.45% of the variability in the response. In addition, the value of the adjusted determination coefficient (adjusted R2 = 0.9326) was also very high, indicating the high significance of the model.

Table 5.Experimental plan for central composite design performed for selected parameters.

The significance of each term in the model is presented in Table 6. The ANOVA for the selected quadratic model showed that the model was significant with a Model F = 30.22 and P > F-value < 0.0001. The results indicated that the linear effects of wheat straw powder (X1) and the interaction effect of X1X3 were significant (P < 0.05). Moreover, the linear effects of soybean meal (X2) and CuSO4 (X3), the quadratic effects of wheat straw powder (X12), soybean meal (X22), and CuSO4 (X32), and the interaction effect of X2X3 were more significant than the other factors (P < 0.01). Thus, the response of laccase production (Y) by ZF-2 could be expressed in terms of the following regression equation:

where X1 is wheat straw powder, X2 is soybean meal, and X3 is CuSO4.

Table 6.ANOVA for response surface quadratic model.

A 3D response surface was drawn based on the model equation to investigate the interaction among the variables and determine the optimum concentration of each factor for maximum laccase production by ZF-2 (Figs. 1A-1C). The model predicted that the maximum laccase production was up to 65.457 U/ml when the medium was wheat straw powder 7.63 g/l, soybean meal 23.07 g/l, and CuSO4 0.51 g/l.

Validation of Model

To confirm the accuracy of the model, a validation experiment was performed using three different solutions (wheat straw powder, soybean meal, and CuSO4), as suggested by the software results (Table 7). The results showed a good agreement between the predicted and experimental values, thereby validating the model. The maximum laccase production observed was 67.258 U/ml and the optimal medium was wheat straw powder 7.63 g/l, soybean meal 23.07 g/l, (NH4)2SO4 1 g/l, CuSO4 0.51 g/l, Tween-20 1 g/l, MgSO4 1 g/l, and KH2PO4 0.6 g/l. As a result of optimizing the culture conditions, a 59.68-fold increase in laccase activity was achieved compared with that (1.127 U/ml) previously obtained under basal conditions.

Fig. 1.Three-dimensional response surface plots of laccase production by ZF-2 under conditions optimized using RSM. (A) Effect of interaction of soybean meal and wheat straw; (B) effect of interaction of copper sulfate and wheat straw; (C) effect of interaction of copper sulfate and soybean meal.

Table 7.Experimental verification of combined effect of optimized medium on laccase production.

 

Discussion

Traditional optimization methods change one independent variable, while keeping the other variables fixed at a certain level. However, this single-dimensional search is laborious, time-consuming, and incapable of reaching a true optimum owing to the interactions among the variables. The response surface methodology (RSM), first described by Box and Wilson in 1951 [4], is an effective strategy for seeking the optimum conditions for a multivariable system. Whereas this statistical method for media optimization has already been successfully utilized to improve laccase production from white-rot fungi, there are no reports on the optimization of laccase production from Trichoderma harzianum. Therefore, this study applied RSM to optimize the medium conditions for laccase production by Trichoderma harzianum ZF-2. The obvious important factors (wheat straw power, soybean meal, and CuSO4) that influenced the production of laccase were obtained using a Plackett- Burman experimental design, and the optimal concentrations of these three factors were then sequentially investigated using the response surface methodology with a central composite design. The final optimal medium components included wheat straw powder 7.63 g/l, soybean meal 23.07 g/l, (NH4)2SO4 1 g/l, CuSO4 0.51 g/l, Tween-20 1 g/l, MgSO4 1 g/l, and KH2PO4 0.6 g/l. When using this medium, the yield of laccase was increased 59.68 times (67.258 U/ml) compared with the laccase production in an unoptimized medium (1.127 U/ml).

Among the important factors affecting laccase production, wheat straw, an inexpensive and easily available agroresidue, was the most suitable carbon source for laccase production and showed higher significant effect than the other substrates. In agreement with this result, Patel et al. [22] previously reported that wheat straw was the best substrate for the production of laccase by the basidiomycete fungal isolate Pleurotus ostreatus HP-1. Soybean meal is also an inexpensive organic nitrogen source and had a positive effect on laccase production. Therefore, this study provides useful information regarding the use of cheaper carbon and nitrogen sources for maximizing the production of laccase. Furthermore, CuSO4 supplementation at 0.051% was effective in improving the laccase production by ZF-2. This expressive effect is also supported by the fact that Cu2+ is a laccase cofactor, where every four Cu2+ are associated with one single polypeptide chain.

In conclusion, this research is the first systematic medium optimization for laccase production by Trichoderma harzianum ZF-2 and showed the effect of each component. Although some scale-up experiments are yet to be performed, the optimum medium will be helpful for further studies of laccase production by Trichoderma harzianum ZF-2 using large-scale batch fermentation. Further studies will also explore the role of laccase in the delignification and biodegradation of cellulose in nature by ZF-2.