Ultrasound-assisted extraction, optimization, characteristics and antioxidant activity of Piper nigrum L. polysaccharides

Abstract

The process conditions, content of ingredients and in vitro antioxidant activity of Piper nigrum L. polysaccharides (PNP) by ultrasound-assisted extraction (UAE) and PNP by hot water extraction (HWE) were compared. The findings demonstrated that the UAE produced greater polysaccharides content (74.41 %) with the yield of PNP (2.9 %) than HWE. The ideal conditions were 324 W of ultrasonic power, 36 mL/g of liquid to material ratio, 70 min of ultrasonic time, and 78 °C of temperature. Structural analysis showed UAE-PNP was the α-type polysaccharides with a pyran ring structure, which were mainly neutral polysaccharides. In addition, UAE-PNP had great antioxidant activity, especially in its ability to scavenge ABTS free radicals. According to the experimental results, the UAE method was an efficient way to extract PNP. This experiment showed for the first time the structure and antioxidant activity of HWE-PNP and UAE-PNP, which provided some theoretical proof for the application of PNP in food additives and biopharmaceuticals.

1. Introduction

Originally found in India, Piper nigrum L. (PNL) is a climbing vine that is related into the Pepperaceae family. It is currently widely planted in many tropical regions of the world and has a great economic value, which is frequently referred to as the “black gold” and “the king of tropical spices” [1]. In Traditional Chinese Medicine (TCM), in addition to being used for flavoring, it is believed to have anti-inflammatory and pain-relieving properties [2,3]. In our daily lives, it is a common spice that is processed into powder. Based on recent studies, PNL contains alkaloids [4,5], monoterpenes [6], sesquiterpenes [6], lignans, and other biologically active substances [7], of which its alkaloids as amides specific to the genus PNL have been extensively studied [8,9]. On the other hand, its polysaccharides have not yet received much attention and its structure and its activity are not understood. It is crucial to isolate polysaccharides from PNL and to study their structure and bioactivity in order to enhance the analysis of PNL components and to investigate their bioactivity.

Polysaccharides can be extracted from natural plants in a variety of ways [10]. The most popular hot water extraction (HWE), which has the benefits of being cheap as well as simple to operate, has been replaced by more effective techniques like enzyme-assisted extraction (EAE), microwave-assisted extraction (MAE) and ultrasound-assisted extraction (UAE) because of its low extraction rate [11,12]. Many studies have chosen UAE because of its high rate of extraction, which is related to its mechanical effect, which significantly facilitates the mass transfer between unmixed phases by hyperstirring [13,14]. By altering the chemical or spatial structure of plant polysaccharides, UAE can also break down or modify polysaccharides with various physicochemical characteristics and biological activities [15]. There are disadvantages, nevertheless, such as the fact that excessive ultrasonic intensity damages the polysaccharides' structure, lowers its production, and most certainly lowers its biological activity [10]. Therefore, optimization of the UAE technique is essential in order to maintain its biological activity and obtain more Piper nigrum L. polysaccharides (PNP).

Math and statistics are used in response surface methodology (RSM) experiments. It has been effectively used to identify optimal processes and multivariate interactions [16]. In this research, Box-Behnken design (BBD) has been used for the optimisation of the production process of UAE of PNP. A BBD with three factors and three levels was used to investigate the effects of varying temperatures, times and the ratio of liquid to material on the production of PNP. This study will also use HWE to extract PNP and compare the polysaccharides content, protein content, and in vitro antioxidant activity of HWE-PNP and UAE-PNP. To provide some indications on the development and application of PNP.

2. Materials and methods

2.1. Material pre-treatment and laboratory equipment

PNL was produced from Wanning City, Hainan Province, China, which was washed with H₂O, dried at 60°C, crushed, screened through 60 mesh, sealed, and retained [11]. Then mix the powder with petroleum ether in a ratio of 1:5 (w/v), reflux and extract at 50 ℃ for 2 h to remove lipids, pigments, etc., and dry for later use.

Experimental instruments: SCIENTZ-10 N Vacuum freeze-dryer (Ningbo Xinzhi Biotech Co., Ltd.), IR408 Fourier Transform Infrared (FT-IR) spectrometer (Shimadzu, Japan), ANAVCEII Nuclear Magnetic Resonance (NMR) instrument (Brugg, Switzerland), L550 benchtop low-speed large-capacity centrifuge (Changsha Xiangyi Instrument), UV-2550 Ultraviolet–visible spectro-photometer (Shimadzu Instruments).

2.2. HWE of PNP

10 G of PNL powder was taken into a round bottom flask and the PNP was extracted with hot water, when the extraction was complete, cooled, centrifuged and filtered to obtain the polysaccharides extract, distilled under reduced pressure to a volume of 15 mL, taken to add 4 times the amount of anhydrous ethanol and placed in a refrigerator set at 4°C for about 12 h [17]. The solution was centrifuged at 4200 rpm for 5 min to isolate and get the extract. About 10 mL of H₂O was put in to re-solubilize the solution, which was then frozen as a solid and placed in a freeze-dryer for 48 h to get HWE-PNP, and the yield was calculated [18]. The yield of HWE-PNP was calculated using the formula

$$\mathit{Yield}=\frac{\mathit{Weight}ofPNP}{\mathit{Weight}ofskimmedPipernigrumLPowder}\ast 100\%$$ (2.1)

2.3. UAE of PNP

10 G of PNL powder and a certain amount of H₂O were mixed in a beaker, and then the beaker was subjected to ultrasonic waves at a frequency of 40 kHz for extraction, after completion it was cooled to room temperature, centrifuged and filtered to get the polysaccharides extract[18], after which the steps were referred to the method of 2.2 to obtain UAE-PNP, and the yield of UAE-PNP was calculated using Eq. (2–1)

2.4. HWE optimization of PNP

2.4.1. One-factor experiment of HW

This experiment will examine the effect of liquid to material ratio, extraction time and temperature on the yield of HWE-PNP. Weigh 10 g of PNL powder into a beaker and refer to the experimental method in 2.2 to extract HWE-PNP and calculate its yield according to Eq. (2–1). Keeping temperature and extraction time constant to find the best values of the single-factor liquid/material ratio experiment, with conditions of 15, 20, 25, 30 and 35 mL/g. Keeping temperature and extraction time constant to find the best values of one-factor experiment of the extraction time, with conditions of 60, 90, 120, 150, and 180 min. Keeping extraction time and liquid to material ratio constant, to find the best value of the one-factor experiment of temperature, the conditions were 50, 60, 70, 80 and 90 ℃ [19]. The details are shown in Table 1.

Table 1.

One-factor experiment conditions of HWE-PNP.

Single factor
Liquid to material ratio (mL/g)	15	20	25	30	35
Extraction time (min)	60	90	120	150	180
Temperature (℃)	50	60	70	80	90

2.4.2. RSM experiments of HWE-PNP

According to the Box-Behnken principle, three factors, namely, the desired liquid to material ratio, extraction time and temperature, have been organized as varying conditions that carried out the design of the experiment described in Table 2. The combined impact of these factors on yield was mainly examined [20].

Table 2.

BBD test factors and levels of HWE-PNP.

Levels	Factors
	Liquid to material ratio (mL/g)	Extraction time (min)	Temperature (℃)
−1	25	90	70
0	30	120	80
1	35	150	90

2.5. UAE optimization of PNP

2.5.1. One-factor experiment of UAE

The yield of PNP will be investigated in this experiment in relation to temperature, ultrasonic power, ultrasonic time, and liquid/material ratio. Weigh 10 g of PNL powder into a beaker and refer to the experimental method in 2.3 to extract UAE-PNP and calculate its yield. To investigate the optimum value of liquid to material ratio in one-factor experiments with constant temperature, ultrasonic power and ultrasonic time, the conditions were 20, 30, 40, 50 and 60 mL/g. Maintain the temperature, liquid/material ratio, and ultrasound power constant, and explore the optimal ultrasound time under experimental conditions of 30, 45, 60, 75, and 90 min. The optimal values of one-factor experiments were investigated by maintaining the ultrasonic time, the liquid/material ratio and the ultrasonic power constant, and the conditions were 40, 50, 60, 70 and 80 ℃. The optimal values of the one-factor experiment of ultrasonic power were investigated by maintaining liquid/material ratio, temperature and ultrasonic time constant, and the conditions were 216, 252, 288, 324 and 360 W [19]. The one-factor experiments design are shown in Table 3.

Table 3.

One-factor experiment conditions for UAE-PNP.

Single factor
Liquid to material ratio (mL/g)	20	30	40	50	60
Ultrasound time (min)	30	45	60	75	90
Temperature (℃)	40	50	60	70	80
Ultrasonic power (W)	216	252	288	324	360

2.5.2. RSM experiments of UAE-PNP

According to the Box-Behnken principle, three factors, namely, the desired liquid/material ratio, ultrasound time and temperature, have been organized as varying conditions that carried out the experimental design shown in Table 4. The combined impact of these three factors on yield was mainly examined [11].

Table 4.

BBD test factors and levels of UAE-PNP.

Levels	Factors
	Liquid to material ratio (mL/g)	Ultrasound time (min)	Temperature (℃)
−1	30	60	60
0	40	75	70
1	50	90	80

2.6. Deproteinization of PNP

HWE-PNP and UAE-PNP are a mixture which has proteins present. It is therefore necessary to deproteinize [21]. Among the methods for deproteinization of polysaccharides, the Sevag method is more common, in which PNP solution: trichloromethane: n-butanol = 15:4:1 by volume is added to the separatory funnel and mixed thoroughly for about 15 min, and the protein layer and organic layer at the bottom are removed to obtain the supernatant [22]. The supernatant was then subjected to alcohol precipitation and re-solubilization with reference to the experimental method in 2.2, and freeze-dried to obtain PNP after protein removal.

2.7. Identification of PNP chemical components

The PNP composition obtained by the optimized process using the methods in 2.4 and 2.5 includes not only polysaccharides, but also other substances such as proteins, and even if it is further purified by the method in 2.6, it will still have other substances present, and therefore it needs to be tested for its composition. The concentration of the standard solution to be measured is set to 0, 0.02, 0.04, 0.06, 0.08 and 0.10 mg/mL. The following equation “y” represents the absorbance at different wavelengths and “x” represents the concentration of the component.

2.7.1. Identification of polysaccharides content

This experiment is based on the phenol sulfate method for calculating the polysaccharides content in PNP. It is shown in Fig. 1(a), which is the linear regression equation of the glucose standard curve:

$$y=13.749x+0.779(R^2=0.9993)$$ (2.2)

Fig. 1.

Standard curves for various components.

2.7.2. Identification of protein content

This experiment was based on the determination of protein content in PNP by the method of Caulmers Brilliant Blue staining. As shown in Fig. 1(b), which shows the linear regression equation for the standard curve of bovine serum protein:

$$y=5.676x+0.675(R^2=0.9986)$$ (2.3)

2.7.3. Identification of uronic acid content

This experiment was based on the carbazole sulfate method for the determination of content of uronic acid in PNP as shown in Fig. 1(c), which shows the linear regression equation of the standard curve for uronic acid:

$$y=11.531x+0.144(R^2=0.9983)$$ (2.4)

2.7.4. Identification of total phenol content

This experiment was based on the determination of total phenol content in PNP by the forintol method, as shown in Fig. 1(d), which is a linear regression equation for the standard curve of gallic acid:

$$y=9.714x+0.022(R^2=0.9997)$$ (2.5)

2.8. Structural analysis

2.8.1. FT-IR spectral analysis

About 4 mg of HWE-PNP and UAE-PNP were taken and analyses were carried out by FT-IR using KBr compression method for both polysaccharides [23]. The scan range was set to 4000 ∼ 500 cm⁻¹.

2.8.2. NMR analysis

About 50 mg UAE-PNP was taken and dissolved in 0.45 mL of heavy water, centrifuged and the supernatant was taken into NMR tube. It was analyzed by 300 MHz NMR carbon spectroscopy C13CPD [23].

2.9. Analysis of antioxidant activity of PNP in vitro

The absorbance values of HWE-PNP and UAE-PNP reaction solutions at specific wavelengths were determined by UV spectrophotometry to determine and test the magnitude of their antioxidant capacity from four aspects [24]. Vitamin C (Vc) solution was used as a positive control in all of the following antioxidant experiments.

2.9.1. Determination of DPPH radical scavenging capacity of PNP

The DPPH radical scavenging ability was determined with reference to the experiment of Teng et al. [5]. According to the actual situation, with partial modifications, 2 mL of different concentrations of HWE-PNP and UAE-PNP solutions were added to the same volume of DPPH-ethanol solution (0.5 mmol/L). The reaction was carried out in a dark environment for about 30 min, and the absorbance values were measured at 517 nm. DPPH radical scavenging ability by different concentrations of HWE-PNP and UAE-PNP was calculated as follows:

$$E\left(\%\right)=\frac{A_{\alpha }-A_{\beta }}{A_{\alpha }}\ast 100\%$$ (2.6)

A_α is the absorbance of the blank control group, A_β is the absorbance of the sample group.

2.9.2. Determination of hydroxyl radical scavenging capacity of PNP

The method was modified with reference to Chen et al. [25]. 1 mL of different concentrations of HWE-PNP and UAE-PNP solutions were added with the same volume of FeSO₄ (10 mmol/L), salicylic acid–ethanol solution (10 mmol/L) and H₂O₂ (10 mmol/L) mixed and shaken, and then a reaction occurred out for 25 min at 40 °C in a water bath, and the absorbance was measured at 510 nm. The hydroxyl radical scavenging capacities of different concentrations of HWE-PNP and UAE-PNP were calculated as follows:

$$E\left(\%\right)=\frac{A_{\alpha }-A_{\beta }}{A_{\alpha }}\ast 100\%$$ (2.7)

A_α is the absorbance of the blank control group, A_β is the absorbance of the sample group.

2.9.3. Characterization of ABTS radical scavenging ability of PNP

Referring to the method of Liu et al. [26], for improvement, ABTS diammonium salt solution (7.4 mmol/L) and K₂S₂O₈ solution (2.6 mmol/L) were configured, and the same volume of the above solutions was mixed and added to a brown bottle for 12 h of storage in a dark environment, and it was adjusted to an absorbance at 0.7 at 734 nm using phosphate buffer (pH = 7.2 ∼ 7.4) to get the ABTS radical working solution. 1 m L of different concentrations of HWE-PNP and UAE-PNP, and 3 mL of ABTS radical working solution were mixed and shaken well, and reacted for 10 min in a dark environment, and the absorbance values were measured at 734 nm. ABTS radical scavenging ability of different concentrations of HWE-PNP and UAE-PNP were calculated as follows:

$$E\left(\%\right)=\frac{A_{\alpha }-A_{\beta }}{A_{\alpha }}\ast 100\%$$ (2.8)

A_α is the absorbance of the blank control group, A_β is the absorbance of the sample group.

2.9.4. Characterization of the reducing power of PNP

The method was improved by referring to Song et al. [27]. 1 mL of phosphate buffer (pH = 6.6) and K₃Fe(CN)₆ solution were configured using hydrochloric acid and added to 1 mL of HWE-PNP and UAE-PNP solutions at different concentrations, oscillated uniformly and then water-bathed at 50 °C for 30 min. 1 mL of Cl₃CCOOH (10 %) was then added to terminate the reaction. Centrifuged 2 mL the supernatant, 2 mL H₂O and 0.4 mL of FeCl₃ solution (1 %), mix thoroughly, and detect at 700 nm.

3. Results and Discussion

3.1. One-factor optimization

3.1.1. Single factor optimization of HWE-PNP

Temperature, time, and liquid to material ratio, which are three typical single variables of HWE, were chosen for single factor optimization in order to compare the yield of HWE-PNP by altering one of the factors while holding the other two constant. As seen in Fig. 2, the yield of HWE-PNP was taken as the vertical axis, while the single factors set at various gradients were used as the horizontal axis.

Fig. 2.

Effect of different single factors on the yield of HWE-PNP.

Based on Fig. 2(a), it can be noticed that yield increases faster when the liquid/material ratio does not reach 30 mL/g. This is because as the amount of H₂O increases, more H₂O molecules adsorb the polysaccharides, a hydrophilic substance containing a large number of hydroxyl and other polar groups, thus increasing its solubility. However, when liquid/material ratio exceeds 30 mL/g, the curve begins to fall gradually. It was shown that increasing the solvent at this point had little effect on the yield and that doing so would prolong the time required for further processing [12]. Therefore, 30 mL/g was chosen as the optimum liquid/material ratio.

As observed in Fig. 2(b), when the extraction time was raised to more than 120 min, the yield began to decrease. It was hypothesized that prolonged extraction time might promote more polysaccharides solubilization, but too long extraction time simultaneously triggered the hydrolysis of polysaccharides to oligosaccharides, which reduced the yield of HWE-PNP [28]. Therefore, the ideal extraction time was determined to be 120 min.

As can be seen in Fig. 2(c), the yield increases until temperature reaches 80°C. This may be due to the fact that the elevated temperature disrupts the cell wall structure of PNL, allowing more polysaccharides to be released into the solvent. Elevated temperature also accelerated the molecular movement of polysaccharides and increased their solubility. When the temperature was increased to 90°C, the yield decreased due to the hydrolysis and denaturation of the polysaccharides caused by the high temperature [20]. It was judged that 80°C was the optimal extraction temperature.

3.1.2. Single factor optimization of UAE-PNP

The optimal conditions for three single factors i.e., temperature, time, liquid to material ratio were explored at a fixed ultrasonic power of 324 W. One of the factors was varied and other 3 factors were kept constant to compare the yield of HWE-PNP. The single factors set at different gradients are shown as the horizontal axis and the yield of UAE-PNP is shown as the vertical axis. As can be seen in Fig. 3.

Fig. 3.

Effect of different single factors on the yield of UAE-PNP.

As can be seen in Fig. 3(a), the yield started to decrease when the liquid/material ratio rose above 40 mL/g. It is generally believed that too much solvent will weaken the intermolecular interactions and reduce the penetration ability of the ultrasonic system, which will reduce the solubilization effect of polysaccharides and ultimately reduce the yield [29].

Fig. 3(b) indicates that yield increased significantly when ultrasound time was less than 75 min. This may be due to the fact that the longer cavitation time disrupts the cell wall of the feedstock, thus reducing the binding of polysaccharides to the cell wall and other structures. Beyond 75 min, there was little change or even a decrease in yield. It is assumed that the polysaccharides solubility has reached its maximum at this point, and that the excessively long ultrasound time has no effect on increasing the yield, and may even lead to the hydrolysis of the polysaccharides, thus reducing the yield [30].

With reference to Fig. 3(c), the yield of UAE-PNP significantly rose from 40°C to 70°C, whereas the yield began to decline at 80°C. It's possible that the solvent's surface tension decreases at a particular temperature, lowering the intensity needed for ultrasonic cavitation and aiding in the solubilization of polysaccharides. However, if the temperature keeps rising, eventually the hydrolysis and inactivation of polysaccharides will cause the temperature to rise above the point at which the mass transfer rate increases [31].

Fig. 3(d) indicates that the enhancement of ultrasonic power has an influence on the increase of UAE-PNP yield in a certain range. Higher power increases cavitation and system pressure, thereby increasing the likelihood that the polysaccharides will break the cell wall and release solubilization. However, too much ultrasonic power leads to polysaccharides hydrolysis, which reduces the yield [31].

3.2. RSM optimization and results of HWE-PNP

According to the results of HWE one-factor, 3-factor, 3-level experiment design was conducted out based on BBD. The design selected liquid to material ratio (A), extraction time (B) and temperature (C) as the three factors as variables, and R1 as yield of HWE-PNP. Seventeen sets of exploratory experiments were conducted. The specific experimental design and results are presented in Table 5.

Table 5.

RSM experiment design and results of HWE-PNP.

Serial number	A: Liquid to material ratio (mL/g)	B: Extraction time (min)	C: Temperature (℃)	Actual yield (%)	Predicted yield (%)
1	30	120	80	2.62	2.62
2	35	120	90	2.4	2.4
3	35	90	80	2.2	2.19
4	35	120	70	2.17	2.17
5	30	90	90	2.53	2.53
6	30	120	80	2.61	2.62
7	30	150	90	2.58	2.58
8	30	120	80	2.63	2.62
9	25	150	80	2.29	2.30
10	30	120	80	2.63	2.62
11	35	150	80	2.25	2.25
12	25	120	70	2.2	2.2
13	30	90	70	2.26	2.26
14	30	150	70	2.34	2.34
15	25	120	90	2.47	2.47
16	25	90	80	2.23	2.23
17	30	120	80	2.62	2.62

3.2.1. Fitting of the HWE-PNP regression equation and analysis of variance

The equation for quadratic regression was determined by fitting the data in Table 5 using multiple recursion and establishing a mathematical model:

$$R1=-14.26650+0.610150\ast x+0.038433\ast y+0.123225\ast z-0.000017\ast x\ast y-0.000200\ast x\ast z-0.000025\ast y\ast z-0.009940\ast x^2-0.000146\ast y^2-0.000635\ast z^2$$ (3.1)

(In the equation “x” stands for Liquid to material ratio, “y” stands for Extraction time and “z” stands for Temperature). As can be seen from Table 6, according to the larger F value, which represents the larger influence of the factor to the yield of HWE-PNP, the order of the influence of A: Liquid to material ratio, B: Extraction time and C: Temperature on the yield of UAE-PNP was C > B > A [32]. This indicates that the effect of temperature is the most significant. The regression model had a p-value < 0.0001 and an F-value of 883.65, showing that the model was significant. It can also be determined that the model constructs of B,C,A²,B²,C² are highly significant [30]. Table 7 shows a correlation coefficient of R² = 0.9991, Adjusted R² = 0.9980, and a difference of 0.0011, indicating that the experimental model is well-fitted and can predict the optimal extraction conditions and yield of HWE-PNP.

Table 6.

Analysis of variance of the regression model for the HWE-PNP.

Sourse	Sum of Square	df	Mean Square	F-value	p-value	Significant differences
Model	0.5169	9	0.0574	883.65	< 0.0001	**
A: Liquid to material ratio	0.0036	1	0.0036	55.58	0.0001	*
B: Extraction time	0.0072	1	0.0072	110.77	< 0.0001	*
C: Temperature	0.1275	1	0.1275	1961.73	< 0.0001	**
AB	0	1	0	0.3846	0.5548
AC	0.0004	1	0.0004	6.15	0.0422	*
BC	0.0002	1	0.0002	3.46	0.1051
A2	0.26	1	0.26	4000.15	< 0.0001	**
B2	0.0723	1	0.0723	1111.64	< 0.0001	**
C2	0.017	1	0.017	261.2	< 0.0001	**
Residual	0.0005	7	0.0001
Lack of Fit	0.0002	3	0.0001	0.8333	0.5413	−-
Pure Error	0.0003	4	0.0001
Cor Total	0.5174	16

Annotation: no significant (−-); highly significant (**); significant (*).

Table 7.

ANOVA regression model of HWE-PNP.

Std. Dev.	0.0081	R2	0.9991
Mean	2.41	Adjusted R2	0.9980
C.V. %	0.334	Predicted R2	0.9937
PRESS	0.0032	Adeq Precision	72.6938

3.2.2. Interaction of HWE-PNP factors

The 3D response surface plots and contour plots formed by two-by-two interactions of A: Liquid to material ratio, B: Extraction time and C: Temperature are seen in Fig. 4 to determine the interaction and significance effects. The interaction between B: Extraction time and C: Temperature is significant, as indicated by the steepness of the response surface in Fig. 4(B-1), the roughly elliptical form of the contour lines in Fig. 4(B-2), and the p-value < 0.05 in Table 6 [32]. The three-dimensional surfaces shown in Fig. 4(A-1) and Fig. 4 (C-1) are relatively flat, while the contour shapes in Fig. 4(A-2) and Fig. 4(C-2) differ from the elliptical shapes, indicating that the interaction between B: Extraction time and A: Liquid to material ratio and C: Temperature is not significant, and there are some differences. All the response surfaces in Fig. 4(A ∼ C-1) have openings facing downwards, proving that the model has optimal extraction conditions and optimal yields [32].

Fig. 4.

3D and contoured maps of the interaction of different variables in the HWE-PNP model.

3.2.3. Determination and validation of optimal conditions for HWE-PNP

The results of 3.2.1 and 3.2.2 showed that the results of the designed model were reliable. The model predicted optimal conditions of liquid/material ratio of 31.289 mL/g, extraction time of 121.666 min and temperature of 83.09 °C, at which time the predicted optimal yield was 2.633 %. From the operational simplicity of the experiment, the conditions were modified to a liquid/material ratio of 31 mL/g, an extraction time of 122 min and a temperature of 83°C. The results of these experiments are summarized below. Three parallel experiments were carried out according to these conditions and the final average yield obtained was 2.61 %. The relative error to the predicted optimum yield was 0.87 %. The reliability of the experimental design model was again demonstrated.

3.3. RSM optimization and results of UAE-PNP

Based on the experimental results of UAE one-factor, a 3-factor, 3-level experimental design was conducted according to Box-Behnken central combination. The design selected liquid to material ratio (A), ultrasound time (B) and extraction temperature (C) as the three factors as variables and R2 as yield of UAE-PNP. Seventeen sets of exploratory experiments were conducted. The experimental design is shown in Table 8.

Table 8.

RSM experiment design and results of UAE-PNP.

Serial number	A: Liquid to material ratio (mL/g)	B: Ultrasound time (min)	C: Temperature (℃)	Actual yield (%)	Predicted yield (%)
1	40	90	60	2.58	2.57
2	40	75	70	2.9	2.92
3	50	60	70	2.61	2.60
4	50	75	60	2.41	2.41
5	30	75	80	2.88	2.88
6	40	60	80	2.82	2.83
7	40	75	70	2.93	2.92
8	50	90	70	2.71	2.72
9	40	75	70	2.93	2.92
10	50	75	80	2.79	2.79
11	40	60	60	2.35	2.36
12	40	75	70	2.93	2.92
13	30	90	70	2.78	2.79
14	30	60	70	2.73	2.72
15	30	75	60	2.52	2.52
16	40	75	70	2.92	2.92
17	40	90	80	2.83	2.82

3.3.1. Fitting of the UAE-PNP regression equation and analysis of variance

The equation for quadratic regression was determined by fitting the data in Table 8 using multiple recursion and establishing a mathematical model:

$$R2=-12.55425+0.069175\ast x+0.098750\ast y+0.277900\ast z+0.000083\ast x\ast y+0.000050\ast x\ast z-0.000367\ast y\ast z-0.001047\ast x^2-0.000488\ast y^2-0.001672\ast z^2$$ (3.2)

(where “x” represents the liquid to material ratio, “y” represents the ultrasound time, and “z” represents the temperature). From the Table 9, according to the larger F-value, which represents the larger impact on the yield of UAE-PNP, the sequence of the influence of A: Liquid to material ratio, B: Ultrasound time and C: Temperature on the yield of UAE-PNP is C ＞ B = A [33]. The model is not significant for pure error and has an excellent significance effect, as indicated by its F-value of 246.21 and p-value ＜ 0.0001. It shows that the experimental results predicted by the model have a greater chance of matching the actual results. Finally, the significant effects on extraction rate included model constructs of A,B,C,A²,B²,C² [33], where the coefficients of the quadratic terms of this equation are all negative, indicating that the response surface curve opens downward and there is a response maximum. According to the Table 10, it can be observed that R² = 0.9969 and Adjusted R² = 0.9928, which is a difference of 0.0041, indicating that the results of the experimental model are well fitted and the data are reliable, which can be used as the optimal extraction conditions for analyzing the UAE-PNP and in order to predict its optimal extraction rate.

Table 9.

Analysis of variance of the regression model for the UAE-PNP.

Sourse	Sum of Square	df	Mean Square	F-value	p-value	Significant differences
Model	0.5556	9	0.0617	246.21	< 0.0001	**
A: material-liquid ratio	0.019	1	0.019	75.83	< 0.0001	**
B: Ultrasound time	0.019	1	0.019	75.83	< 0.0001	**
C: Temperature	0.2664	1	0.2664	1062.76	< 0.0001	**
AB	0.0006	1	0.0006	2.49	0.1584
AC	0.0001	1	0.0001	0.3989	0.5477
BC	0.0121	1	0.0121	48.26	0.0002	*
A2	0.0462	1	0.0462	184.27	< 0.0001	**
B2	0.0507	1	0.0507	202.29	< 0.0001	**
C2	0.1178	1	0.1178	469.77	< 0.0001	**
Residual	0.0018	7	0.0003
Lack of Fit	0.0011	3	0.0004	2.11	0.242	−
Pure Error	0.0007	4	0.0002
Cor Total	0.5573	16

Annotation: no significance (−); highly significant (**); significant (*)

Table 10.

ANOVA regression model for UAE-PNP.

Std. Dev.	0.0153	R2	0.9969
Mean	2.74	Adjusted R2	0.9928
C.V. %	0.5774	Predicted R2	0.9672
PRESS	0.018	Adeq Precision	46.3806

3.3.2. Interaction of UAE-PNP factors

Combined with Fig. 5 to analyze the interaction and influence between the two factors. The steeper the 3D curve, the more elliptical the shape of the contours and the denser the contours, the more significant the interaction between the two factors and the greater the influence on yield[32]. Among them, the steeper 3D curve in Fig. 5(C-1) and the nearly elliptical shape of contour lines in Fig. 5(C-2) indicate that the interaction of B: Ultrasound time and C: Temperature is significant. Additionally supporting this finding is the p-value < 0.05 from Table 9. Meanwhile, from (A-1 ∼ 2) and (B-1 ∼ 2) in Fig. 5, it can be noticed which the interactions of A: Liquid to material ratio and B: Ultrasound time, and C: Temperature are not significant, and the p-value > 0.05 in Table 9 is consistent with this result [32].

Fig. 5.

3D and contoured maps of the interaction of different variables in the UAE-PNP model.

3.3.3. Determination and validation of optimal conditions for UAE-PNP

Based on the results of the response surface modeling analysis, the predicted optimal process for UAE-PNP was the liquid/material ratio of 35.643 mL/g, the ultrasound time of 69.128 min and the temperature of 77.525 °C at the ultrasound power of 324 W, at which time the yield was 2.947 %. Based on the simplicity of operation, the predicted optimal conditions were appropriately adjusted, and three experiments were conducted in parallel at the ultrasonic power of 324 W, the liquid/material ratio of 36 mL/g, the ultrasonic time of 69 min, and the temperature of 78 °C, resulting in an average yield of 2.9 %. The relative error to the theoretical optimum yield was 1.59 %, which proved the reliability of the experimental model.

3.4. The effect of various extraction methods on PNP

3.4.1. HWE-PNP and UAE-PNP extraction conditions and yield comparison

The optimal reaction conditions and optimal yields obtained from response surface optimization experiments based on are shown in the Table 11. From the comparison of HWE-PNP and UAE-PNP in the Table 11, it is concluded that ultrasonic extraction of PNP requires a higher liquid/material ratio, whereas the extraction time is substantially reduced and the required temperature is lowered. It shows that UAE of PNP is more efficient than HWE of PNP, and the final optimization to get the best yield shows that UAE of PNP has a higher yield. Therefore, UAE of PNP is more recommended. The experimental results again proved that UAE has the advantage of high efficiency [34]. There are still many extraction methods for PNP that have not been studied in depth, and in the future, we can consider the joint use of two extraction methods to extract PNP.

Table 11.

Optimized extraction conditions and yields of HWE-PNP and UAE-PNP.

Product	Liquid to material ratio (mL/g)	Time (min)	Temperature (℃)	Actual optimal yield (%)
HWE-PNP	31	122	83	2.61
UAE-PNP	36	69	78	2.9

3.4.2. Comparison of compositional analysis of HWE-PNP and UAE-PNP

Configure 0.1 mg/mL solutions of HWE-PNP and UAE-PNP and calculate the absorbance of the reaction solutions at various wavelengths according to the experimental method in 2.7. Calculate the content of components in HWE-PNP and UAE-PNP extracted under optimal extraction conditions by referring to the formula in Fig. 1. As shown in Table 12, the polysaccharides content of the PNP obtained by both extraction methods was predominant, followed by proteins, with the least amount of phenolics. The comparison in the Table 12 indicates that the polysaccharides content of the UAE-PNP was higher than that of the HWE-PNP, with 74.41 % compared to 70.91 %. The protein content of UAE-PNP was slightly smaller than that of HWE-PNP protein content. It is hypothesized that under the influence of ultrasound, it can destroy the chemical bonding force between polysaccharides and protein and promote the separation of polysaccharides and protein, at the same time, ultrasound can disrupt the structure of the cell wall and decrease the molecular weight of polysaccharides [35]. This can enhance the solubilization of polysaccharides, to achieve the increase of polysaccharides content and decrease of protein content. Before the two polysaccharides were removed from proteins, the polysaccharides content of UAE-PNP was 70.77 %, and the polysaccharides content of HWE-PNP was 67.13 %, which indicated that Sevag method removal of proteins had a certain effect on its enhancement of polysaccharides content. However, this method also reduces the yield of HWE-PNP and UAE-PNP, and the residue of organic solvents in it may affect the subsequent detection and application. Meanwhile, when comparing the uronic acid content and total phenol content, there was little difference between the two extraction methods. In terms of in order to extract PNP with higher purity, UAE is recommended.

Table 12.

Comparison of component analysis of HWE-PNP and UAE-PNP.

Name of polysaccharides	Polysaccharides content (%)	Protein content (%)	Uronic acid content (%)	Phenol content (%)
HWE-PNP	70.91	7.40	3.99	0.31
UAE-PNP	74.41	5.81	4.16	0.21

3.5. Structural characterization

3.5.1. FT-IR analysis of HWE-PNP and UAE-PNP

Infrared spectroscopy is commonly used to detect the characteristic functional groups of polysaccharides, as well as their characteristic absorption peaks mostly appear at 4000 ∼ 500 cm⁻¹. The strong absorption peaks at 3287 cm⁻¹ and 3277 cm⁻¹ for HWE-PNP and UAE-PNP, respectively, which are generated by the vibrations of stretching of the non-free hydroxyl groups of polysaccharides chain, are most clearly seen in the Fig. 6. Both polysaccharides have the same absorption peak at 2930 cm⁻¹, which is generated by the stretches vibration of the C-H bonds of the the methyl and methylene groups in the polysaccharides [20]. No C=O absorption peaks were found in the 1700 ∼ 1750 cm⁻¹ range, suggesting that both polysaccharides may or may not contain small amounts of uronic acid. This conjecture is also supported by the results in Table 12. The absorption peaks at 1635 and 1633 cm⁻¹ are generated by bound water with the polysaccharides [19]. The angular vibration of the C-H bonds is what causes faint peaks on 1415 and 1410 cm⁻¹. Pyranose tends to have consecutive absorption peaks in the range of 1200 ∼ 1000 cm⁻¹. HWE-PNP shows a faint absorption peak on 1077 cm⁻¹ and a strong absorption peak on 1019 cm⁻¹, and UAE-PNP shows fragile absorption peaks on 1145 cm⁻¹, 1075 cm⁻¹ and a strong absorption peak on 1019 cm⁻¹. This indicates that both samples include the functional groups C-O-H and C-O-C. In contrast, the peaks at the typical interval of furanose from 846 to 758 cm⁻¹ were not particularly pronounced [20]. The weak absorption peaks at 934 and 932 cm⁻¹ indicate the possible presence of β-type glycosidic bonds in both polysaccharides, however, relatively, the absorption peaks on 857 and 852 cm⁻¹ indicate that they contain α-type glycosidic bonds, which are presumed to contain α-type glycosidic bonds considering that the intensity of the absorption peaks on 857 and 852 cm⁻¹ is greater than that of the absorption peaks on 934 and 932 cm⁻¹ [23]. This is demonstrated in the NMR analysis in 3.5.2. The functional groups of HWE-PNP and UAE-PNP are basically the same, and the differences are mainly concentrated in the range of 1200 to 1000 cm⁻¹, with HWE-PNP having one less peak characteristic of a pyranose. HWE-PNP and UAE-PNP are both pyranoses.

Fig. 6.

FT-IR of HWE-PNP and UAE-PNP.

3.5.2. NMR analysis of UAE-PNP

Considering that the nuclear magnetic resonance images of HWE-PNP and UAE-PNP are almost the same, in order to avoid repeated description, only ¹³C NMR spectrum of UAE-PNP is analyzed here. The Fig. 7 shows the ¹³C spectrum of UAE-PNP, and the signal peaks from C2 to C5 on its sugar ring are mainly concentrated in the region from 69.44 to 76.88 ppm. Three C1 peaks (99.64 ppm, 95.87 ppm, 91.99 ppm) with gradually weakening signals appeared in the signal region of the heterohead carbons, suggesting that UAE-PNP may contain three monosaccharide residues and that the amount of monosaccharide residues generating the resonance peaks at 95.87 ppm, 91.99 ppm is relatively low compared to the amount of monosaccharide residues generating the resonance peak at 99.64 ppm. Also based on the chemical shifts at 90 ∼ 102 ppm is an α-type glycosidic bond and at 102 ∼ 112 ppm is a β-type glycosidic bond [36]. Determining that UAE-PNP has an α-type glycosidic bond. Between 60 ppm and 110 ppm, the signals of all the signal peaks, except for the signal of the heterohead carbon atom, are less than 80 ppm, demonstrating that the UAE-PNP is made up only of the pyran ring without the involvement of the furan ring, which is compatible with the hypothesis of the FI-IR analysis. The signal peak appearing at 60 ppm, which is presumed to be generated by its C6. In addition, no signal peaks were found in the region from 160 to 180 ppm, representing that the NMR did not detect the carboxyl carbon signal of uronic acid, consistent with the results of low uronic acid content in Table 12.

Fig. 7.

¹³C NMR spectra of UAE-PNP.

3.6. In vitro antioxidant activity

As indicated by Fig. 8, the antioxidant activity of both HWE-PNP and UAE-PNP were somewhat dose-dependent. In a series of antioxidant studies, there is no discernible difference between antioxidant properties of HWE-PNP and UAE-PNP. Compared to HWE-PNP and UAE-PNP, Vc has a much better antioxidant capacity. It is expected that the effects of chemical modifications (carboxymethylation and phosphorylation, etc.) and isolation and purification on the antioxidant activity of HWE-PNP and UAE-PNP will be investigated in the future, and the application of PNP to areas such as food additives and biopharmaceuticals.

Fig. 8.

Comparison of in vitro antioxidant activities of HWE-PNP and UAE-PNP.

3.6.1. DPPH radical scavenging activity assay

DPPH with stable nitrogen centered radicals is often used to evaluate the in vitro antioxidant properties of polysaccharides [37]. When a single electron with the nitrogen atom of a DPPH radical comes into contact with an electron absorbing antioxidant, the DPPH ethanol solution will fade [38]. The scavenging ability of Vc, HWE-PNP, and UAE-PNP towards DPPH radicals at specific concentration gradients is shown in Fig. 8(a). The scavenging rate of UAE-PNP is slightly higher than that of HWE-PNP.

3.6.2. Hydroxyl radical scavenging activity assay

The majority of biological macromolecules can react with hydroxyl radical, an incredibly aggressive reactive oxygen species, to see how well they can withstand radicals [39]. The hydrogen provided by the antioxidant combines with hydroxyl radical in order to stop the radical reaction chain and achieve the antioxidant effect [40]. From Fig. 8(b), it was obtained that HWE-PNP and UAE-PNP have some ability to scavenge hydroxyl radicals, which is comparable.

3.6.3. ABTS radical scavenging activity assay

The ABTS radical is one of the radicals that is commonly employed to evaluate the antioxidant potential of natural polysaccharides. The antioxidant's hydroxyl and carboxyl groups can supply electrons to change ABTS radical into a stable form, which will have an antioxidant effect [5,41]. The scavenging rates of Vc, HWE-PNP, and UAE-PNP all rose as solubility increased. As shown in Fig. 8(c), the clearance rates of Vc, HWE-PNP, and UAE-PNP increase with increasing concentration. The clearance efficiency of HWE-PNP is slightly higher than that of UAE-PNP.

3.6.4. Reduction capacity assay

K₃Fe(CN)₆ was reduced to K₄Fe(CN)₆ (pH = 6.6) by the active polysaccharides, which then combined with Fe³⁺ to generate Fe[Fe(CN)₆]₃ [42]. Fig. 8(d) illustrates the positive correlation between the magnitude of the reducing ability and the absorbance of the complex solution. Fig. 8(d) indicates that the reduction capacity of HWE-PNP is marginally greater than that of UAE-PNP.

4. Conclusion

This study used RSM to optimize the extraction of PNP by HWE and UAE. The liquid to material ratio of 31.289 mL/g, the extraction duration of 121.666 min, and the temperature of 83.09 °C were the ideal parameters for the former, and the expected ideal yield was 2.633 %. For the latter, the ideal parameters were 324 W of ultrasonic power, 35.643 mL/g of the liquid to material ratio, 69.128 min of ultrasonic time, and 77.525 °C of temperature. Ideal yield is 2.947 %. Additionally, the actual yield matches the model forecast quite well. It demonstrates the correctness and dependability of RSM design model presented in this paper. UAE decreased the protein level of PNP while simultaneously increasing its polysaccharides content as compared to the other two. The results of both polysaccharides obtained through different extraction methods were found to be basically the same by modern analytical techniques such as FT-IR and NMR, which are composed of α-pyran rings. The experiments showed that HWE-PNP and UAE-PNP had certain antioxidant activity and the difference between them was not significant. The major objectives of this research are to enhance and add to PNL's compositional analysis and to offer theoretical underpinnings and technical assistance for PNP's continued advancement and use. Its good antioxidant activity may be useful in food and pharmaceutical fields. The high-level structure of PNP and its conformational connection, however, remain unclear and are now awaiting experimentalists' further exploration.

CRediT authorship contribution statement

Laiqing Deng: Investigation, Methodology, Writing – original draft. Gangliang Huang: Supervision, Writing – review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.