3.1. Determining the Optimal Combination of Enzymes
Enzyme preparations suitable for
Ulva polysaccharide extraction were screened according to the characteristics of
U. lactuca cell walls (
Table 4).
U. lactuca has a bilayer cell wall structure consisting mainly of cellulose, hemicellulose and pectin [
33]. Therefore, CEL, hemicellulase (AXC and GMA) and PEC were included in the screening set. In addition, the
U. lactuca cell wall contains a small amount of protein. Kevin et al. [
34] showed that proteases could effectively improve the extraction yield of
Ulva polysaccharides, so ALC, PAP and NP were included. Polysaccharide extraction was carried out under the optimal conditions for each enzyme, and the screening experiment results are shown in
Figure 1. The extraction yield of
Ulva polysaccharides was 6.43% without any enzyme preparation. In comparison, all the enzyme preparations used in the experiment increased the extraction rate of polysaccharides. For example, CEL, AXC, GMA, PEC, ALC, NP and PAP treatments could increase the polysaccharide content in extracts to 14.46%, 12.02%, 8.18%, 8.85%, 10.51%, 7.91% and 9.19%, respectively (
Figure 1). This indicates that enzyme treatment may be an effective way to break through the cell wall, an important barrier for polysaccharide extraction, thus facilitating the rapid release of polysaccharides from the cells. Specifically, CEL, AXC, PEC and ALC had a higher polysaccharide extraction rate than the same kind enzyme preparations (
Figure 1), so they could be used as the basic components of the complex enzyme preparations.
Further, the microstructure of
U. lactuca was observed by SEM to determine the beneficial effects of enzymes on cell wall destruction. As shown in
Figure 2, the intact surface of the untreated sample tissue was smooth (
Figure 2A), while the ultrasonic-assisted hot water treatment caused tissue damage and wrinkles (
Figure 2B), resulting in the dissolution of some polysaccharides. As expected, the addition of CEL significantly disrupted the cell walls and membranes, resulting in a lamellar structure and a large number of gaps and cavities (
Figure 2C), which contributed to the release of polysaccharides. The results were consistent with the screening experiment, indicating that the enzyme treatment caused the disruption or loosening of the cell wall, which facilitated the infiltration of solvent molecules into the cells and the enhanced polysaccharide extraction.
Considering the complex composition of
U. lactuca’s cell wall, the degrading effect of a single enzyme was limited, as the complex action of multiple enzymes is conducive to the destruction of the cell wall [
35]. Therefore, CEL, AXC, PEC and ALC were used as the basic components of complex enzyme preparations and coupled. The experiments were performed according to the simplex lattice design, with each point corresponding to a specific mixture composition, and the obtained results are shown in
Table 1. The Scheffé canonical polynomial model (special cubic) was used for regression analysis. The variation in
Ypol values corresponding to different ratios of the enzyme was fitted using Design-Expert to obtain the reduced model for CEL (
x1)-AXC (
x2)-PEC (
x3)-ALC (
x4):
According to the analysis of variance (
Table 5), Prob (P) >
F < 0.001, indicating that the modified equation had a high fitting accuracy. The correlation coefficient R
2 = 0.9842 and adjusted R
2 = 0.9500, indicating a good fit of the model. Furthermore, no significant deviation from the basic assumptions of ANOVA was found, and the
p-value of the lack of fit was 0.9690, indicating that the model had high stability (
Table 5). In addition, CV = 7.36% < 10%, indicating high confidence in the experiment. In summary, the fitted regression equation was consistent with the test principle and had good adaptability, which can be used for subsequent optimization designs.
The coefficient of the positive term in the fitting equation represents the positive correlation between the factor and the response value, and the larger the coefficient, the stronger the correlation [
36]. According to Equation (7), the single-enzyme preparation had a beneficial effect on the dissolution of polysaccharides, and the influence of the four enzyme preparations on the polysaccharide extraction was in the following order: CEL > AXC > ALC > PEC, which was consistent with the results of the screening experiments (
Figure 1). Three-dimensional response surfaces and two-dimensional contours could show the interaction between the different factors, for example, surface convexity reflects synergistic effects and surface concavity vice versa. As shown in
Figure 3 and
Figure S1, ternary mixed-enzyme preparations have both synergistic and antagonistic effects at the same time. For example, when CEL-PEC-ALC were used together, they showed antagonism; the mixed-enzyme preparations of CEL-AXC-PEC, CEL-AXC-ALC and AXC-PEC-ALC showed synergistic effects and CEL-AXC-ALC had the greatest synergistic effect. For this purpose, numerical calculations were performed by maximizing the corresponding variables in the equation, and the following the numerical calculation results were obtained:
x1 = 0.353,
x2 = 0.345,
x3 = 0.302 and
Ylop = 28.34%. Validation experiments were performed with the above mixed-enzyme preparations and the resulting polysaccharide content of 26.68% differed from the predicted value by <2%.
3.2. Optimization of Ultrasound-Assisted Enzymatic Extraction
The BBD was used to optimize the UAEE process for the extraction of
Ulva polysaccharides. With the extraction rate of
Ulva polysaccharides as the response value, the relationship between the predicted response value and various factors can be expressed by a polynomial equation, as follows:
The statistical analysis results of this model are shown in
Table 6, with
p < 0.0001 and the
F-value (24.17) indicating that the fitted polynomial could well characterize the relationship between the parameters [
37]. The
p-value of the lack of fit was 0.6440, confirming the validity of the experimental model and that unknown factors had little effect. Furthermore, predictions were made for the response values of the regression Equation (8): the correlation coefficient R
2 = 0.9603, which means that the model has a 96.03% agreement with the actual test fit. In addition, the coefficient variability of the model, CV = 3.34%, indicates a high degree of experimental reproducibility. In general, the fitted model was an ideal model which was sufficient to cover the experimental design area and could be used in subsequent experiments.
The response surface diagram more intuitively reflected the influence of two factors (other variables fixed at 0 level) on the extraction yield of
Ulva polysaccharides. The greater the slope of the 3D surface, the stronger the effect of the independent variables on the extraction rate [
38]. Moreover, the ellipticity of the contour lines shows whether the factor has a significant effect on the response value [
39]. As shown in
Figure 4 and
Figure S2, the effects of enzyme concentration and ultrasonic time on polysaccharide yield were more significant compared to enzymatic time and enzymatic temperature. Furthermore, the interaction of
X1X2,
X1X3 and
X2X3 was significant and the interaction effect of
X1X4,
X2X4 and
X3X4 was not significant. The analysis results of the RSM were in good agreement with the results of the analysis of variance of the regression model (
Table 6), which proved that the test results were highly representative. According to the BBD results, the optimal technological parameters of
Ulva polysaccharide extraction were obtained: an enzyme concentration of 1.49%, enzymatic time of 1.08 h, ultrasonic time of 89.42 min and enzymatic temperature of 58.82 °C. According to the actual situation, the modified parameters were 1.5% enzyme concentration, 1.1 h enzymatic time, 90 min ultrasonic time and 60 °C enzymatic temperature for experimental verification. The polysaccharide yield was 30.14%, and the relative error of the predicted value was 0.41%. This shows that the predicted values fit well with the actual values, and the model has good practical reference significance.
Compared with the reported extraction methods of
Ulva polysaccharides, the optimized UAEE process can not only significantly improve the extraction rate of polysaccharides but also save energy consumption and time. For example, Xu et al. [
40] obtained a 21.96% extraction yield of
Ulva polysaccharides by using a 90 °C hot water treatment for 4 h and obtained a 20.22% extraction yield by using cellulase to promote polysaccharide dissolution. Lü et al. [
41] obtained 27.75% of
Ulva polysaccharides by protease-assisted extraction. Although the extraction yield of acid and alkaline extraction was higher, for example, 33.30% polysaccharides could be obtained by using an alkaline solution at 90 °C for 2 h [
42] and up to 38.35% polysaccharides could be produced by acid extraction at 80 °C for 24 h [
43], acid and alkaline extraction could break the glycosidic bond and change the polysaccharide configuration, and special reactions during extraction may produce by-products. On the other hand, in order to prevent environmental pollution, the liquid after acid and alkaline extraction should be pH neutralized, and the post-processing is more complicated. Therefore, acid and alkaline extraction of polysaccharides is not considered a promising method.
3.3. Ulva Polysaccharide Extract Effectively Scavenges Free Radicals
In molecular biology, high levels of free radicals have been closely linked to the onset of degenerative processes. They could enhance oxidative stress, leading to inadequate cell function, aging and even disease [
44]. Therefore, the antioxidant capacity of polysaccharides is an important index to evaluate their biological activity. DPPH scavenging ability detection is a simple, rapid and reliable method for the study of antioxidant properties of natural products. As shown in
Figure 5A, in the range of 0~6.0 mg/mL, the scavenging activity of DPPH free radicals was significantly enhanced with the increase in the concentration of
Ulva polysaccharides. A total of 8.0 mg/mL
Ulva polysaccharides can effectively remove 69.80% of DPPH free radicals. Subsequently, it was calculated that the sample concentration required to scavenge half of the free radicals (SC
50) was 5.46 mg/mL. Compared with other methods, the
Ulva polysaccharides extracted by the UAEE method had better DPPH scavenging activity. For example, the DPPH scavenging SC
50 values of
Ulva polysaccharides obtained by enzyme-assisted and ultrasonic-enzyme-assisted extraction were 6.52 and 9.90 mg/mL, respectively [
45].
The hydroxyl radical is the most harmful free radical for an organism, and it is capable of having a free radical chain reaction with almost any biological macromolecule in living cells [
46]. As shown in
Figure 5B,
Ulva polysaccharide extracts of different concentrations had scavenging effects on the hydroxyl radical. However, when the extract concentration was higher than 4 mg/mL, the scavenging effect was not significantly improved. Compared with
Ulva polysaccharides obtained by pressurized water-assisted extraction, the polysaccharide extracted with UAEE showed a better hydroxyl radical scavenging effect. For example, the hydroxyl radical scavenging rates of 2 mg/mL UAEE-extracted and pressurized water-assisted-extracted
Ulva polysaccharides were 49.12% and 45% [
3], respectively. In addition, previous studies have shown that the hydroxyl radical scavenging activity is related to the molecular weight of the compound [
43]. The high hydroxyl radical scavenging activity of
Ulva polysaccharides suggests that they have a lower molecular weight, which affects the solubility and viscosity of the polysaccharide, thus improving the antioxidant activity.
The ABTS free radical scavenging method is widely used to determine the total antioxidant capacity of biological samples [
47]. The ABTS scavenging activity of
Ulva polysaccharides is shown in
Figure 5C. As expected, increasing the concentration of
Ulva polysaccharides resulted in an increase in ABTS scavenging. When the concentration of
Ulva polysaccharides increased from 0 mg/mL to 4 mg/mL, the scavenging rate of ABTS radical increased from 0% to 67.85%. In the range of 6.0~10.0 mg/mL, the ABTS free radical scavenging rate of
Ulva polysaccharides remained about 73%. Moreover, the highest ABTS scavenging rate of
Ulva polysaccharides extracted by UAEE was 73.81%, which was higher than those extracted by the hot water (68.06%), alkali (61.01%) and acid (71.87%) methods [
43].
Superoxide free radicals play an important role in the oxidative and reductive metabolism of cells, can participate in many physiological activities such as cell proliferation and apoptosis, and are closely related to body aging and disease [
48]. Therefore, the scavenging activity of superoxide free radicals is very important to antioxidant work. As shown in
Figure 5D, the scavenging effect of 0~8 mg/mL
Ulva polysaccharides on superoxide free radicals was concentration-dependent. Among them, 8 mg/mL
Ulva polysaccharides could remove 64.26% of superoxide free radicals. Subsequently, with the increase in polysaccharide concentration, the free radical scavenging rate did not increase significantly, and the inhibitory concentration (IC
50) was 5.32 mg/mL. For polysaccharides with special conformation, the hydrogen in the oxygen–hydrogen bond is easily released, thus stabilizing the superoxide free radicals [
49]. The mechanism of polysaccharide removal of superoxide free radicals may be related to the dissociation energy of the oxygen–hydrogen bond.
3.4. Ulva Polysaccharides Protect SH-SY5Y Cell Damage Induced by H2O2
Based on the study of cell-free systems, we further investigated the protective effect of
Ulva polysaccharide pretreatment on oxidative stress in intact cell models. H
2O
2 destroyed the protein structure through oxidative reaction, triggered mitochondrial dysfunction and led to apoptosis, and was a common compound used to establish cell models of oxidative damage [
50]. An MTT assay was used to investigate the mitigating effect of
Ulva polysaccharides on oxidative damage in cells, and the cell viability of the blank control group was defined as 100%. As shown in
Figure 6A, as expected, 250 μM H
2O
2 reduced cell viability to 70.31%, indicating severe cell damage induced by H
2O
2.
Ulva polysaccharide pretreatment was effective in alleviating the oxidative damage caused by H
2O
2 in a dose-dependent manner, specifically, 25, 50, 100 and 200 μg/mL
Ulva polysaccharide treatment increased the cell activity to 72.05%, 77.20%, 82.47% and 89.11%, respectively.
ROS can induce oxidative stress and lead to apoptosis by regulating active transcription factors [
51]. The antioxidant activity of
Ulva polysaccharides was evaluated by detecting ROS levels. H
2O
2-stimulated SH-SY5Y showed significantly higher ROS levels than normal cultured cells, indicating oxidative stress. However, the H
2O
2-induced elevation of ROS levels gradually decreased to normal levels with the increase in
Ulva polysaccharides. In particular, 200 μg/mL of
Ulva polysaccharides could reduce the ROS to 112.26% (
Figure 6B). Consistent with the results of Zhang et al. [
52], pretreatment with antioxidant substances could alleviate cytotoxicity and inhibit ROS production to play a cytoprotective role in oxidatively stressed cells.