Box-Behnken Design for Optimizing Ultrasonic-Assisted Enzymatic Extraction of Soluble Dietary Fiber from Pleurotus citrinopilestus

Abstract

Pleurotus citrinopilestus contains a variety of physiologically and pharmacologically active compounds. A key active component among these compounds is dietary fiber, a polysaccharide that exhibits several biological properties. The objective of this study was to assess how soluble dietary fiber (SDF) from Pleurotus citrinopilestus responded to ultrasonic-assisted enzymatic (UAE) extraction. The response surface method (RSM) combined with the Box-Behnken design method (BBD) was used to optimize the yield of SDF. The effects of the liquid-solid ratio (35–55 mL/g), α-amylase concentration (0.5–2.5%), complex protease concentration (0.4–2.0%), and ultrasonication time (15–55 min) on the yield of SDF were examined. The RSM results revealed the optimal liquid-solid ratio (45 mL/g), α-amylase concentration (1.5%), complex protease concentration (1.2%), and ultrasonic time (35 min). The SDF yield was 10.25%, which is close to the predicted value (10.08%).

1. Introduction

Pleurotus citrinopileatus Sing., an edible mushroom belonging to the genus Pleurotus and the family Basidiomycete (Pleurotaceae), is popular in China, India, and Japan [1], and it frequently grows on the dead wood of broad-leaved trees. P. citrinopilestus is a type of natural fungus that contains significant amounts of lignin, can decompose cellulose, and is utilized as food and medicine [2]. The fungus flesh is white, the stalk is only partially present, and the fungus cap is grass yellow to bright yellow with a smooth funnel form. It has good edible and ornamental value because of its vivid color and rich and distinctive flavor, and it has enormous market potential. In recent years, P. citrinopilestus stands out among the many artificially cultivated edible fungus varieties and has become the preferred edible fungus variety in underforest cultivation, agriculture, tourism, life and leisure experience projects, labor education, and research activities due to its unique edible–ornamental value, which goes against the idea that plant cultivation is the only aspect of the horticultural industry [3,4]. This fungus is rich in protein, amino acids, trace elements, and minerals, and contains a variety of physiologically and pharmacologically active substances. In addition, some researchers performed a detailed analysis of the chemical composition of P. citrinopilestus and found it contains a variety of phenolic substances, such as tannic acid, vanillic acid, gentian acid, and gallic acid. In addition to polyphenols, it also contains an antioxidant, ergothioneine, which can slow the aging process [5,6,7,8]. Sawabe et al. [9,10] used a FAB/MS method to isolate 5 kinds of sphingolipids from P. citrinopilestus and carried out a structural analysis, thus proving that the FAB/MS method is an effective means to identify unstable lipids in natural products. In addition, polysaccharides are the main active ingredient of this fungus, with antitumor, immunomodulatory, antioxidant, anti-inflammatory, antiaging, hypoglycemic, and other biological activities. According to a previous study [11], P. citrinopilestus has a high crude fiber content and is a potentially abundant natural source of green dietary fiber (DF). Research on P. citrinopilestus is currently mostly focused on evaluating germplasm resources and optimizing cultivation formulas; there are limited studies on its DF and other active components [11,12].

DF, which mostly consists of pectin, lignin, resin, pentaglycine, cellulose, and hemicellulose, is defined as multiple units of carbohydrate polymers, a type of polysaccharide that is not digested by human intestinal enzymes [13,14,15]. DFs are found naturally in grains, vegetables, fruits, and nuts. DF is the seventh most abundant nutrient investigated by the scientific community, and its significance for our bodies is clear. Foods with high DF typically have lower fat and calorie counts, are more nutrient-dense, and can increase feelings of fullness [16,17,18]. In addition, DF exhibits many physiological activities and is involved in various functions, such as weight loss and lipid reduction; the prevention of cardiovascular and cerebrovascular diseases, diabetes, and cancer; and improving the gut microbiota [19,20,21,22].

DF can be divided into two categories, soluble dietary fiber (SDF) and insoluble dietary fiber (IDF), based on its solubility. SDF is easily digested by the digestive system via fermentation by the microbial community of the colon. SDF with excellent water, oil, and dilatancy qualities can help control blood sugar levels, reduce cholesterol levels, and moderate the risk of cardiovascular disease. It also has the ability to expand, induce fullness, absorb water, prevent intestinal bacterial infections, and reduce appetite. SDFs play a major role in the physiological function of DFs. Studies have shown that SDF is very beneficial to human bodies, so it is crucial to use acceptable and effective extraction techniques to accelerate SDF extraction and improve the quality and physiological activity of DF [23,24,25,26].

At present, the commonly used methods of SDF extraction include thermal extraction, chemical extraction, enzyme extraction, and emerging technology-assisted extraction [27,28]. However, the products generated by these methods exhibit poor structures and characteristics, and the yield of DF extracted by heat extraction and chemical extraction is minimal. Using cutting-edge technology and enzymes for DF extraction is inexpensive, with high purity of the products and a high extraction efficiency [29,30,31,32]. To improve the yield of DF, these auxiliary extraction methods are widely used. The commonly used new auxiliary extraction methods mainly include ultrasonic waves, microwave waves, osmotic pressure, and high-voltage pulsed electric fields. Among them, ultrasonic extraction is considered the most promising extraction technology for bioactive substances extraction. Zhao et al. [33] used the UAE method to extract polysaccharides from Flammulina velutipes and found that when the ultrasonic power was 680 W, the extraction time was 19.8 min, the solid–liquid ratio was 1:28, the polysaccharide yield was 16.20%, and the extraction rate was much higher than the traditional extraction method. Zhang et al. [34] used an ultrasonic method to extract Lentinan polysaccharide. The optimal technological conditions were as follows: extraction temperature 140 °C, extraction time 40 min, solid-liquid ratio 1:25, ultrasonic power 190 W. Under these conditions, the extraction rate of polysaccharides was 17.34%, and the extraction time was shortened. The ultrasonic-assisted enzyme (UAE) method, as a combined technology, can effectively destroy the structure of the cell wall and contribute to the dissolution of SDF in materials. This method can be employed as the optimal technology for effectively preparing high-quality DF due to its moderate conditions, quick reaction time, and high efficiency [35]. On the other hand, the advantage of response surface methodology (RSM) is saving time and money while determining the best combination, improving the quality and efficiency [36]. UAE and RSM have previously been used to extract polysaccharides from other Pleurotus species and other types of mushrooms. Up to present, there is no research using RSM to optimize the DF extraction rate from P. citrinopilestus by UAE extraction technology. Therefore, the authors utilized an UAE method to extract SDF from P. citrinopilestus. The SDF extraction from P. citrinopilestus was investigated comprehensively by single factor analysis and RSM to improve the yield of SDF. The liquid–solid ratio, α-amylase concentration, complex protease concentration, and ultrasonic time were taken as the investigation factors, and the yield was taken as the evaluation standard. The purpose of this study was to provide some references for future research on DF extraction methods of Pleurotus.

2. Materials and Methods

2.1. Materials

Artificially cultivated P. citrinopilestus was obtained from Shanghai Peng’s Moregood Co., Ltd. (Shanghai, China). Images of the front and side of the fruiting bodies that we collected are shown in Figure 1. Complex protease and α-amylase were purchased from Shanghai Solarbio Bioscience & Technology Co., Ltd. (Shanghai, China). Absolute ethyl alcohol was purchased from Sinophosphoric Chemical Reagent Co., Ltd. (Shanghai, China). Reagents were all of analytical quality. Distilled water was used throughout the experiment. Information about the instruments and equipment used in this experiment can be found in Table S1 in the Supplementary Materials.

2.2. Sample Pretreatment

P. citrinopilestus was harvested, dried at 55 °C for 6–8 h, powdered, and subsequently sieved through a 60-mesh sieve. Until further examination, the mushroom powder was stored at room temperature in a desiccator.

2.3. Extraction of SDF from P. citrinopilestus

A method by Jia et al. was followed [37] with appropriate modifications for the extraction process. Specifically, enzyme extraction with ultrasonic assistance was used to extract SDF from P. citrinopilestus. Two grams of mushroom powder were combined with water, the pH was adjusted to 7.0, and then it was hydrolyzed with α-amylase at 55 °C for two hours. After that, the samples were hydrolyzed at 55 °C for 2 h using a complex protease in an ultrasonic instrument. After hydrolyzing, the enzyme was inactivated at 90 °C for 5 min, and the supernatant was centrifuged at 7000 rpm for 10 min. The supernatant was precipitated with 95% ethanol (4:1 w/v) for 12 h, and centrifuged (5500 rpm, 10 min), washed with anhydrous ethanol and acetone, and centrifuged (5500 rpm, 10 min). The supernatant was discarded, precipitated, and dried at 55 °C to obtain SDF. The extracted DF powder is shown in Figure 2.

2.4. Yield Determination

After the extraction and drying process, the yield (%) of SDF was calculated as follows using Equation (1):

$$\mathrm{Yield}\mathrm{of}\mathrm{soluble}\mathrm{dietary}\mathrm{fibre}\left(\%\right)=\frac{\mathrm{Weight}\mathrm{of}\mathrm{extracted}\mathrm{fibre}}{\mathrm{Sample}\mathrm{weight}}\times 100$$

(1)

2.5. Single-Factor Experiment

The following influential factors were selected: liquid-solid (mL/g), α-amylase concentration (%), complex protease concentration (%), and ultrasonic time (min). There were four experiments, one factor was modified in each experiment, and the other three factors remained constant at the median. The α-amylase concentration was between 0.5 and 2.5%, the complex protease concentration was between 0.4 and 2.0%, the ultrasonic time was between 15 and 55 min, and the liquid-solid ratio was between 35 and 55 mL/g. The ultrasonic power was always maintained at 100 W. Each experiment was repeated three times.

2.6. RSM Design to Optimize the Extraction Yield

The optimal order of SDF yield in P. citrinopilestus was examined using RSM with three-level, four-factor combined with BBD. The liquid-solid ratio (mL/g), α-amylase concentration (%), complex protease concentration (%), ultrasonic time (min), and other parameters were selected according to the single factor experimental results and were recorded as X₁, X₂, X₃, and X₄, respectively. Compared with other methods, a BBD experiment can obtain the optimal response value, save time and money, and improve the experimental efficiency and accuracy.

BBD was used to evaluate the effect of the combination experiment, and the SDF yield of P. citrinopilestus was used as a response value for the interaction of these four variables. Among them, the quadratic formula reflecting the influence of the four independent variables on the yield of SDF is shown as Equation (2):

$$Y={\beta }_0+{\displaystyle \sum }_{i=1}^4{\beta }_i{X}_i+{\displaystyle \sum }_{i=1}^4{\beta }_{ii}{X}_i^2+{\displaystyle \sum }_{i<j=2}^4{\beta }_{ij}{X}_i{X}_j$$

(2)

where Y is the predicted response (extraction yield); β₀ stands for the model intercept coefficient; β_i, β_ii, and β_ij are the linear, quadratic, and interaction coefficients, respectively; and X_i and X_j are the coded independent variables. The quadratic and interaction terms are denoted by X_i² and X_iX_j, respectively.

2.7. Statistical Analysis

By using SPSS software (Statistical Products and Services Solutions, IBM SPSS Statistical version 27, Armonk, New York, NY, USA), one-way analysis of variance (ANOVA) and Duncan’s multiple comparison test, significant differences between groups were assessed. Standard deviations were used to express the data dispersion of the mean. Statistical significance is indicated by a p-value < 0.05.

3. Results and Discussion

3.1. Single-Factor Experiment Analysis

Figure 3a shows the effect of the liquid-solid ratio on the SDF yield. The extraction yield of SDF showed a trend of increasing and then decreasing as the liquid-solid ratio increased, and the yield was the highest when the liquid-solid ratio was 45 mL/g. This may be because the low concentration of raw materials is not conducive to the binding of DF to the active site of enzymes.

The effects of two enzymes on the extraction yield of SDF were studied. Figure 3b shows the effect of different concentrations of α-amylase on the SDF yield. When the enzyme concentration was 0.5–1.5%, the yield increased significantly, but following a further increase of the enzyme concentration, the yield decreased. Figure 3c shows the effect of complex proteases on the SDF yield. The yield of SDF increased and then decreased as the enzyme concentration increased. This may be due to competition between the two different enzymes when the total enzyme content is high, resulting in a weakened catalytic effect of the enzymes and a decreasing yield.

The yield of SDF increases with the extension of ultrasonic time as shown in Figure 3d. The extension of ultrasonic time can mix the mushroom powder with the solvent more fully, and the yield was the highest when the ultrasonic treatment was performed for 35 min. After that, the yield of SDF began to decline, possibly because the molecular structure of SDF was destroyed due to the long ultrasonic treatment time.

The effects of the liquid-solid ratio, α-amylase concentration, complex protease concentration, and ultrasonic time on the yield of SDF extraction were examined. As shown in Figure 3, the extraction yield of SDF was optimal when the liquid-solid ratio was 45 mL/g, the α-amylase concentration was 1.5%, the complex protease concentration was 1.2%, the ultrasonic time was 35 min, and the other variables remained constant. In general, the trends of various variables on the extraction yield of DF were comparable with each other. With the increase of any extraction condition in a single factor experiment, the SDF yield increased first and then decreased.

Before the formal experiment, a pre-experiment was performed. In the absence of UAE, the final extraction yield of SDF was 7.22%. These results demonstrated that ultrasonic treatment might produce a higher yield of DF. UAE extraction of SDF may provide a beneficial effect due to the improved enzymatic extraction by ultrasound irradiation [38]. Ultrasonic irradiation enhances mass transfer, intragranular diffusion, and superficial tissue damage in DF, making the process of water entering the cell walls of mushrooms simpler. Cavitation phenomena, which can generate microjets and shock waves, change the surface of materials and are one of the mechanisms of ultrasonic extraction. However, high ultrasonic power can generate many small bubbles, thereby reducing cavitation performance. Additionally, a radiation force and microflow are caused by the oscillation of the stable cavitation bubble, which can change the space structure of the enzymes, enhancing the activity of α-amylase and complex protease directly [39,40]. Notably, polysaccharides exposed to ultrasound conditions for a long time can be destroyed by the shearing force of ultrasonic waves, leading to degradation [41,42,43]. On the other hand, a specific temperature can improve the molecular migration of the fiber from cells into solution, raising the diffusion coefficient and solubility, and thus increasing the yield of SDF extraction [44,45].

3.2. Statistical Analysis and Model Fitting Using RSM

The yield of DF from P. citrinopilestus could be enhanced by using an UAE extraction. To optimize the extraction procedure, the liquid-solid ratio, α-amylase concentration, complex protease concentration, and ultrasonic time were selected as independent variables. The conditions for SDF extraction were optimized using the RSM approach, and the effects of four independent factors on the yield were assessed. The encoding levels and ranges of the argument variables are shown in

Table 1. Values of the argument variables and the corresponding levels used in the process.

Independent Variables	Symbol	Levels
		−1	0	1
Liquid–solid ratio (mL/g)	X1	35	45	55
α-Amylase concentration (%)	X2	0.5	1.5	2.5
Complex protease concentration (%)	X3	0.4	1.2	2
Ultrasonic time (min)	X4	15	35	55

. Yield (%) was considered as the response value of the design test.

Table 2. The actual yield and predicted yield (in brackets) of SDF from P. citrinopilestus.

	Coded Variable Levels				Yield of SDF (%)
Run	X1	X2	X3	X4
1	45	1.5	0.4	15	6.43% (6.51%)
2	45	2.5	2.0	35	5.80% (5.96%)
3	45	0.5	0.4	35	6.00% (5.96%)
4	45	1.5	2.0	15	6.33% (6.36%)
5	45	2.5	1.2	55	6.22% (6.27%)
6	35	0.5	1.2	35	6.54% (6.32%)
7	35	1.5	2.0	35	6.02% (6.08%)
8	45	2.5	1.2	15	7.34% (7.21%)
9	45	0.5	1.2	55	7.14% (7.32%)
10	55	1.5	0.4	35	6.72% (6.72%)
11	45	0.5	1.2	15	6.29% (6.30%)
12	45	0.5	2.0	35	6.52% (6.50%)
13	55	0.5	1.2	35	7.59% (7.52%)
14	35	2.5	1.2	35	6.86% (6.73%)
15	45	1.5	2.0	55	6.91% (6.63%)
16	55	2.5	1.2	35	7.21% (7.10%)
17	55	1.5	1.2	15	7.42% (7.42%)
18	45	1.5	1.2	35	9.95% (10.09%)
19	55	1.5	1.2	55	7.62% (7.74%)
20	35	1.5	0.4	35	6.45% (6.50%)
21	45	2.5	0.4	35	6.23% (6.36%)
22	45	1.5	1.2	35	9.99% (10.08%)
23	45	1.5	1.2	35	10.25% (10.08%)
24	35	1.5	1.2	15	6.99% (6.98%)
25	35	1.5	1.2	55	6.63% (6.74%)
26	45	1.5	1.2	35	10.06% (10.07%)
27	55	1.5	2.0	35	7.25% (7.28%)
28	45	1.5	0.4	55	6.54% (6.45%)
29	45	1.5	1.2	35	10.16% (10.08%)

lists the complete experimental design, consisting of 29 runs with independent variables at three variable levels.

A total of 29 runs were performed to optimize each parameter in the BBD design in total. The regression equation of the response can be predicted by examining the independent and dependent variables. The mathematical model for predicting the response Y (dietary fiber yield) can be represented as the following quadratic formula by regression analysis of the experimental data:

$$\begin{gathered} Y=10.08+0.36X_1-0.035X_2+0.038X_3+0.022X_4-0.18X_1{X}_2+0.24X_1{X}_3+0.014X_1{X}_4-0.24X_2{X}_3 \\ -0.49X_2{X}_4+0.12X_3{X}_4-1.34X_1{}^2-1.78X_2{}^2-2.10X_3{}^2-1.52X_4{}^2 \end{gathered}$$

(3)

X₁, X₂, X₃, and X₄ are the coding variables for the liquid-solid ratio, α-amylase concentration, complex protease concentration, and ultrasonic time, respectively. Y represents the yield of DF.

The predicted and actual values of SDF production are shown in Figure 4. The regression model is significant and has a good R² = 0.9933. The resulting R² value (0.9933) shows that the fitting model can account for 99.33% of the variation.

Table 3. ANOVA of the response surface quadratic model analysis for the extraction yield.

Source	Sum of Squares	DF	Mean Square	F Value	p Value
Model	52.96	14	3.78	148.05	<0.0001
X1	1.56	1	1.56	60.87	<0.0001
X2	0.015	1	0.015	0.58	0.4607
X3	0.018	1	0.018	0.69	0.4201
X4	5.633 × 10−3	1	5.633 × 10−3	0.22	0.6459
X1X2	0.12	1	0.12	4.79	0.0460
X1X3	0.23	1	0.23	9.02	0.0095
X1X4	0.078	1	0.078	3.07	0.1017
X2X3	0.23	1	0.23	8.83	0.0101
X2X4	0.97	1	0.97	37.97	<0.0001
X3X4	0.055	1	0.055	2.16	0.1636
X12	11.63	1	11.63	455.12	<0.0001
X22	20.64	1	20.64	807.93	<0.0001
X32	28.64	1	28.64	1121.11	<0.0001
X42	14.97	1	14.97	585.72	<0.0001
Residual	0.36	14	0.026	—	—
Lack of Fit	0.30	10	0.030	1.96	0.2704
Pure Error	0.061	4	0.015	—	—
Cor Total	53.32	28	—	—	—

displays the results of the analysis of variance (ANOVA). The F test reveals that the created model is significant and can completely describe the relationship among the parameters because it has a high model F value (F = 148.05) and a very low p value (p < 0.0001). The linear regression coefficients X₁ (p < 0.0001), interaction coefficients X₁X₃ (p = 0.0095), and X₂X₄ and quadratic coefficients X₁², X₂², X₃², and X₄² were all significant (p < 0.0001), indicating that the liquid-solid ratio, α-amylase concentration, complex protease concentration, and ultrasonic time were significantly correlated with the SDF yield. The liquid-solid ratio was shown to be the major factor impacting the extraction yield of SDF, followed by the α-amylase concentration, complex protease concentration, and ultrasonic time, according to a comparison of the F values of X₁, X₂, X₃, and X₄.

3.3. Optimizing the UAE Extraction of SDF

The interactions between the variables were examined by a three-dimensional (3D) response surface plot and a two-dimensional (2D) contour plot. In order to optimize the appropriate extraction conditions for the conventional enzyme extraction method, graphs were created to show the correlation between the response and the level of processing variable, as well as the interactions among the variables. In Figure 5, the various shapes of the contour plots reveal the different interactions among the factors. An elliptic contour and a circular contour are produced when the independent variables interact perfectly [46]. Two variables are presented in Figure 5 and Figure 6, respectively, in a three-dimensional surface plot and a two-dimensional contour plot, while the other variables are left at 0 because the 0 level is the optimal condition value derived from each single factor experimental result. The interaction values of AB, AC, AD, BC, BD, and CD rose prior to reaching the peak value, demonstrating an increasing yield of SDF extraction. The interaction value of each parameter increased after reaching the highest value, indicating a decreased yield of SDF extraction. The greatest yield of SDF was 10.08%, and equation 3 revealed that the ideal extraction conditions for SDF were as follows: liquid-solid ratio 45 mL/g, α-amylase 1.5%, complex protease 1.2%, and ultrasonic time 35 min. Additional experiments were conducted under the aforementioned circumstances in triple repetitions, and the extraction yield of SDF was 10.25%, which is consistent with the theoretical predicted yield. These results demonstrate the applicability and sufficiency of the model.

The ideal extraction procedure of the UAE method, which was the subject of our investigation, provides more benefits than the other extraction techniques. Wang et al. [32] compared ultrasonic-assisted heat extraction with nonassisted heat extraction of DF, and they found that the yield was higher by 16.34%, the temperature was lower by 13.3 °C, and the extraction time was reduced by 37.78 min compared to nonassisted heat extraction. Similarly, some researchers investigated the extraction of polysaccharides from Lycium barbarum and found that using ultrasonic-assisted extraction obtained more polysaccharides, took less time, and completed the extraction process more quickly than traditional hot water treatment [47]. However, UAE extraction has a milder condition and is more efficient compared with ultrasound-assisted heat extraction, because an excessive water temperature may damage the structure of DF. Yu et al. [48] used the RSM method to optimize the UAE extraction process of Flammulina velutipes polysaccharide. The optimal extraction conditions were raw material size of 400 mesh, ultrasonic power of 380 W, and extraction time of 25 min. Under these conditions, the total polysaccharide yield of Flammulina velutipes increased to 20.52%. Benjarat et al. [49] used high pressure pretreatment assisted extraction of crude polysaccharides from mushrooms and found that extraction with the new assisted technology could increase the polysaccharide yield by 2–12%. In addition, Shehzad et al. [50] evaluated the impact of UAE and other green extraction technologies on the SDF yield in sea buckthorn and obtained a maximum SDF yield of 16.08%, which was close to the predicted value of 15.66%. In general, UAE extraction can indeed increase the yield of SDF. These conclusions are consistent with the results of our experiment. Therefore, the UAE technique for DF extraction used in this work has a very high efficiency [51,52,53,54]. These results will provide reference values for research on the polysaccharide extraction methods of P. citrinopilestus in the future.

4. Conclusions

The optimal technological conditions for the extraction of SDF by UAE were obtained: a 45 mL/g liquid-solid ratio, a 1.5% concentration of α-amylase, a 1.2% concentration of complex protease, and a 35 min ultrasonic time. The experimental yield of SDF under these circumstances was 10.25%, which was approximately the expected yield. Compared with enzymatic extraction, ultrasonic treatment combined with complex enzymes can extensively extract DF and improve the extraction efficiency. The use of UAE technology to extract P. citrinopilestus SDF has been optimized from the perspective of yield, but the differences in structure and the physiological activity of SDF extracted by different processes could not be compared, and additional studies should be conducted in the future.