Ultrasonic-Assisted Extraction of Codonopsis pilosula Glucofructan: Optimization, Structure, and Immunoregulatory Activity

In recent years, multiple edible polysaccharides from Codonopsis pilosula were mainly isolated with high average molecular weights and exhibited various bioactivities, but it was proven that low-molecular-weight polysaccharides could exert stronger activities due to the superior water solubility and permeability. In the present study, the water-soluble polysaccharide C. pilosula with low molecular weight was isolated under ultrasonic assistance at 30 °C, the extraction process was optimized via response surface method (RSM), and the structure and immunoregulatory activity were further investigated. The maximum yield (4.86%) for crude polysaccharides (cCPPs) was obtained under following parameters: ultrasonic power of 370 W, liquid/material ratio of 33 mL/g, ultrasonic time of 81 min. Subsequently, the cCPPs were further purified through dialysis and Sephadex G-25 column to acquire purified polysaccharide (CPPs). Structural analysis indicated that CPPs was a glucofructan (average molecular weight of 4.23 × 103 Da) with (2→1)-β-D-Fruf and (1→)-α-D-Glcp as the backbone branched by (2→6)-β-D-Fruf. Additionally, CPPs could enhance immunoregulatory function by stimulating NO production and cytokine (IL-6 and TNF-α) secretion of RAW264.7 macrophages dose-dependently, which presented no cytotoxic effects. These data suggest that CPPs have the potential to be used as a nutritional dietary compound and natural immunostimulant supplement in the food industry.


Introduction
Recently, herbaceous plants have become an increasing research focus because of the potential application prospects in food, medicine, and cosmetics fields. Codonopsis pilosula Nnannf (belonging to the bellflower family) is a well-known edible and functional herbaceous perennial mainly located in northern regions of China, and has been usually served as a ginseng alternative (more expensive) due to the similar biological features, including effectively strengthening health, invigorating the spleen, and nourishing the lungs [1,2]. The polysaccharides are identified as the major effective constituents of C. pilosula and exhibit multiple biological functions, including immunomodulatory [3], antivirus [4], and anticancer activities [5]. However, diverse extraction methods would contribute to different extraction yields of polysaccharides, as well as the structural properties and biological effects [6]. Therefore, more gentle extraction technology for polysaccharides needs to be explored, with the aim of maintaining the physicochemical and functional properties.
Classical extraction methods for polysaccharides usually require elevated temperatures and a long period of time, which might cause the degradation and denaturation of the structure, leading to weakened bioactivities [7,8]. Presently, various modern extractive techniques have been applied to commercially extract functional biopolymers with a high extraction yield and low depletion of resources under moderate conditions [9]. Ultrasound treatment could produce acoustic cavitation bubbles in water solution, and the bubbles' collapse would accelerate substance dissolution via disrupting cell walls in a eco-friendly green environment [10], contributing to the incremental extraction yields with curtate processing time and less energy consumption [11,12]. As reported, ultrasonic extraction had been extensively used in the preparation of multiple bioactive polysaccharides to improve the extraction yields [13]. Response surface methodology (RSM) could evaluate the impacts of processing parameters on response values by planning experiments and constructing models [14], and has been applied for optimizing the extraction processes of bioactive compounds for higher yields [15]. Therefore, in this study, crude C. pilosula polysaccharides (cCPPs) were prepared at a moderate temperature (30 • C) under ultrasonic extraction, and the extraction conditions were optimized by RSM, which would be beneficial to efficient utilization and industrial production.
The body immune responses (including innate and adaptive responses) are elicited by various environmental antigens and orchestrated by multiple immune cells [16], which are essential for immunodeficiency-related diseases [17]. Macrophages mainly exist in tissues or peripheral blood, and play a central role in host defense by removing pathogens and dead cells [18,19]. Therefore, the in vitro evaluation of macrophages activities after bioactive substance exposure is an important indicator to reflect the immunoregulatory capacity of compounds [20]. As reported, the molecular weights of polysaccharides would significantly affect the solubility and bioactivities, and low-molecular-weight polysaccharides could exhibit stronger activities due to the higher dissolubility and permeability compared with those of higher molecular weight [21]. As a consequence, purified C. pilosula polysaccharides (CPPs) with low molecular weight were prepared through dialysis and column chromatography, and the immunostimulatory effects on macrophages were evaluated.
In the present study, cCPPs were extracted from edible roots of C. pilosula under an ultrasonic-assistant process at 30 • C, and the extraction conditions were optimized to improves yields using RSM on the basis of single-factor experiments. Subsequently, the CPPs with low molecular weight were isolated and the structural properties were analyzed. Finally, the immunoregulatory action on macrophages of the CPPs were investigated. These results would promote the practical applications of CPPs as an edible compound in the food industry and help to improve the nutritional values of C. pilosula.

Preparation Process of cCPP
The root powder of C. pilosula was immersed and extracted with distilled water at 30 • C using a DXIENTZ-IID ultrasonic homogenizer twice. The collected leach liquor was merged, concentrated, and treated by 3 volumes of absolute ethyl alcohol at 4 • C overnight. The sediments were dissolved in distilled water and treated by the Sevag reagent to obtain cCPP. The yield of cCPP was calculated following the formula below: Yield of cCPP (%) = (cCPP weights/C. pilosula roots powder weights) × 100

Single-Factor Experiments
Four crucial factors, including the liquid-solid ratios (10-50 mL/g), ultrasonic power (200-520 W), ultrasonic time (20-100 min), and extraction times (1-5 times), were optimized in single-factor experiments for cCPP preparation. In each single-parameter experiment, one contributing element was researched and the other parameters were constant. The unchanged values were identified as having a liquid-material ratio of 20 mL/g, ultrasonic power of 360 W, ultrasonic time of 60 min, and 2 extraction times. Each experiment was put into practice three times.

Experimental Design and RSM Modeling
The extraction conditions for cCPP were further optimized by Box-Behnken design (BBD) of the RSM according to the above results. Three operating parameters, including liquid-material ratio (A, 25−35 mL/g), ultrasonic time (B, 70−90 min), and ultrasonic power (C, 320−400 W), were coded at three levels of −1, 0, and +1, respectively, which are shown in Table 1. The 17 experimental combinations were run with random sequence, and the predicted and actual yields are displayed in Table 2. As presented, the extraction yields of obtained cCPP in the actual experiments were close to those of the predicted values, suggesting a well-fitted model with high precision [21].

Purification of CPPs
The prepared cCPPs under the above optimal conditions were dialyzed with a molecular weight cut off (MWCO) of 1000 Da against tap and distilled water for three days and centrifuged (8000 rpm for 10 min) to remove sediment. The polysaccharide solution was further purified by sephadex-G25 column. Finally, the purified C. pilosula polysaccharides (CPPs) were obtained after collection and lyophilization.

Monosaccharide Constituents of CPPs
The monosaccharide compositions and the contents of the CPPs were detected using high-performance liquid chromatography (HPLC, Agilent-1200, USA) equipped with an amino column and refractive index detector. In short, 5 mg of dried CPPs were treated by 2 M trifluoroacetic acid at 100 • C for 1 h. The products were neutralized by NaOH solution, followed by HPLC detection. The determination parameters were set as below: The column and detector temperature was 35 • C, the injection volume was 10 µL, the flow rate was 1.0 mL/min, and the mobile phase was acetonitrile-water (7:3). In addition, the fructose content was also further confirmed according to the Seliwanoff colorimetric method.

Cell Viability Detection
An MTT assay was utilized to evaluate the viability of murine macrophages (RAW264.7 cells) after CPP treatments. Briefly, 20,000 RAW264.7 cells per well were planted into a 96-well plate, cultured at 37 • C for 24 h, and then treated with various concentrations of CPPs (6.25-400 µg/mL) for 20 h. Lipopolysaccharide (LPS) in a concentration of 1 µg/mL was regarded as positive control, and the isovolumetric medium as negative control. Subsequently, an MTT solution (5 mg/mL) of 10 µL/well was inoculated and cultured for another 4 h. The medium was abandoned, followed by a 150 µL DMSO addition into each well to dissolve the formed blue-purple formazans. Finally, the solution could be measured at 490 nm using a microplate spectrophotometer (Synergy HTX, Bio Tek, Winooski, VT, USA).

Phagocytosis Detection by Neutral Red Uptake
Neutral red reagent was applied to assess the phagocytosis abilities of RAW264.7 cells. RAW264.7 cells in a concentration of 20,000 cells/well were inoculated in a 96-well plate and cultured under an atmosphere (5% CO 2 , 37 • C) for 24 h. LPS in a concentration of 1 µg/mL was regarded as positive control, and the isovolumetric medium as negative control. Then CPPs (6.25-400 µg/mL) were added for 24 h incubation, and 100 µL of 0.075% neutral red solution was added to each well for another 1 h incubation. Subsequently, the superfluous neutral red was removed by washing three times with PBS, and 200 µL of cell lysis solution containing 1% acetic acid-anhydrous alcohol (1:1, v:v) was added to each well to release the neutral red from macrophages at room for 2 h. The absorbance at 540 nm was measured and the cell phagocytic uptake rates were calculated [26].

NO, TNF-α, and IL-6 Level Detection
The nitric oxide (NO) assay kit was employed to evaluate the NO levels in the medium supernatant. About 400,000 RAW264.7 cells were seeded into each well of a 24-well plate for 24 h cultivation and then treated with different concentrations (6.25-400 µg/mL) of CPPs for 24 h. LPS in a concentration of 1 µg/mL was regarded as positive control, and the isovolumetric medium as negative control. After incubation, the supernatant of each well was collected separately, and the NO level was determined according to the kit instruction. The absorbance (540 nm) was measured under the microplate spectrophotometer, and the NO levels were calculated by establishing a standard curve of different sodium nitrite concentrations. Similarly, the TNF-α and IL-6 levels produced by RAW264.7 cells were evaluated through enzyme-linked immunosorbent assay (ELISA) kits following the corresponding manufacturer's instructions.

Statistical Analysis
All values are displayed as the mean ± standard deviation (S.D.), and each experiment was repeated three times. The significance of between-group differences was measured by Student's t-test and one-way analysis of variance (ANOVA). A p-value of <0.05 was identified as significant. Figure 1 presents the preliminary optimized results of the extraction conditions via single-factor experiments. As displayed in Figure 1A, the cCPP yields significantly increased with an increasing liquid-to-material ratio from 10 to 30 mL/g, and then gradually stabilized, which is consistent with previous related research [27]. As reported, the improved liquid-to-material ratio could have been beneficial for polysaccharide dissolution, resulting in higher polysaccharide yields until reaching the saturated polysaccharide dissolution [28]. Therefore, a liquid-to-material ratio of 30 mL/g was chosen in subsequent optimizations. Figure 1B shows the influence of the extraction time (20-100 min) on cCPP yields. The extraction time also demonstrated a significant impact on polysaccharide yields, which gradually increased to maximum as the extraction time extended to 80 min, and then also slightly decreased [29]. This phenomenon may also have resulted from the degradation of the cCPPs, which was similar with ultrasonic power. Thus, an extraction time of 80 min was selected for subsequent experiments.

Single-Factor Experiment Analysis
Then the effects of ultrasonic power on cCPP yields were investigated, ranging from 200 to 520 W, and the results are presented in Figure 1C. As displayed, the cCPP yields rapidly increased as ultrasonic power improved from 200 to 360 W, and then gradually decreased when the ultrasonic power was higher than 360 W, which was attributed to the polysaccharide degradation under superior powers, which is in agreement with a related report [30]. Therefore, an ultrasonic power of 360 W was adopted in subsequent BBD experiments.
The effects of extraction time (1, 2, 3, 4, 5) on cCPP yields were investigated, and the results are presented in Figure 1D. The cCPP yields were remarkably improved as the extraction frequency increased from 1 to 2, and then gradually became a stable value, which is consistent with a previous paper [31]. As a result, the extraction frequency of 2 was selected for cCPP preparation. rapidly increased as ultrasonic power improved from 200 to 360 W, and then gradually decreased when the ultrasonic power was higher than 360 W, which was attributed to the polysaccharide degradation under superior powers, which is in agreement with a related report [30]. Therefore, an ultrasonic power of 360 W was adopted in subsequent BBD experiments. The effects of extraction time (1, 2, 3, 4, 5) on cCPP yields were investigated, and the results are presented in Figure 1D. The cCPP yields were remarkably improved as the extraction frequency increased from 1 to 2, and then gradually became a stable value, which is consistent with a previous paper [31]. As a result, the extraction frequency of 2 was selected for cCPP preparation.

ANOVA for cCPP Extraction
Design-Expert software (version 10.0) was used to evaluate the ANOVA for the response of the cCPP extraction, and the results of the ANOVA for the quadratic model are displayed in Table 3. The experimental variables were associated with the following formula: The variance analysis demonstrated that coefficients A, B, C, AC, BC, A 2 , B 2 , and C 2 were extremely significant (p < 0.01), whereas the interaction effects of AB showed no significant difference on cCPP yields, which indicates that the variables A, B, C, AC, BC, A 2 , B 2 , and C 2 could significantly influence the extraction yields of cCPPs. As shown, the p < 0.0001 of this mode and the p = 0.7258 for the lack-of-fit value indicate insignificant effects

ANOVA for cCPP Extraction
Design-Expert software (version 10.0) was used to evaluate the ANOVA for the response of the cCPP extraction, and the results of the ANOVA for the quadratic model are displayed in Table 3. The experimental variables were associated with the following formula: The variance analysis demonstrated that coefficients A, B, C, AC, BC, A 2 , B 2 , and C 2 were extremely significant (p < 0.01), whereas the interaction effects of AB showed no significant difference on cCPP yields, which indicates that the variables A, B, C, AC, BC, A 2 , B 2 , and C 2 could significantly influence the extraction yields of cCPPs. As shown, the p < 0.0001 of this mode and the p = 0.7258 for the lack-of-fit value indicate insignificant effects of deviation and suggest a fitted model [32]. Besides, the difference of less than 0.2 between the Pred R 2 and Adj R 2 values also indicates a good correlation between the predicted and actual cCPP extraction yields [33]. Figure 2 presents the interaction effects of the liquid-to-material ratio (A, 25-35 mL/g), ultrasonic time (B, 70-90 min), and ultrasonic power (C, 320-400 W) on cCPP yields. As shown, all response surfaces curves exhibited a maximum point in these selected experimental ranges, suggesting a reasonable range of factors. As reported, the elliptical/circular contour plots reflected significant/indistinctive interaction effects between these variables [34]. Therefore, the liquid-to-material ratio and ultrasonic power (AC), ultrasonic time, and ultrasonic power (BC) exhibited obvious interaction effects (p < 0.01) on the cCPP extraction yields, which is consistent with previous results. The optimal extraction conditions for maximum cCPP yields were determined by Design-Expert software: liquid-to-material ratio of 32.81 mL/g, ultrasonic time of 80.86 min, and ultrasonic power of 368.02 W, with a predicted yield of 4.81%. These conditions were slightly modified for more convenience and feasibility in practical production (liquid-tomaterial ratio of 33 mL/g, ultrasonic time of 81 min, and ultrasonic power of 370 W), and the obtained extraction yields of the cCPPs (three parallel trials) reached 4.86 ± 0.16%. The optimized extraction parameters not only helped remarkably improve the extraction yields of the cCPPs, but also effectively avoided the waste of energy and resources, which could also provide material foundation for further polysaccharides purification and bioactivity determination.

Chemical Constitutions and Primary Structure Analysis of CPPs
In this study, the cCPPs were purified through dialysis and a sephadex-G25 column to obtain CPPs with a purification yield of 31.79%. Chemical constitution analysis showed that the sugar and protein contents of the CPPs were 95.64 ± 4.29% and 1.85 ± 0.11%, respectively. In addition, the uronic acid and total phenolic contents were 0.89 ± 0.07% and 1.13 ± 0.12%, respectively, indicating that the CPPs were a kind of neutral polysaccharide.
The average molecular weight (Mw) of the CPPs was estimated, and the results are shown in Figure 3A. As presented, the retention time (R t ) of the CPPs in the HPGPC spectrum was 12.196 min, demonstrating that the average molecular weight of the CPPs was about 4.23 × 10 3 Da, combining the calibration curve of T-series dextran, taking R t as the x-coordinate and lg(Mw) as the y-coordinate. The monosaccharide compositions and the relative contents were determined by HPLC analysis. As presented in Figure 3B,C, the HPLC spectrum showed two main peaks, which were confirmed to be fructose and glucose, compared to the standards. Moreover, the corresponding contents of fructose and glucose in the CPPs were calculated as 93.14 ± 4.57% and 6.86 ± 0.36%, respectively, according to the calibration curves drawn by various concentrations of standards.
As reported, the structural properties of polysaccharides could significantly affect various properties, including bioactivities, and it has been proven that low-molecular-weight polysaccharides can exhibit superior bioactive effects due to the good water solubility Nutrients 2022, 14, 927 9 of 14 and permeability [35]. Up to now, the average molecular weights of most polysaccharide isolated from C. pilosula were usually higher than 5.00 × 10 3 Da [36], indicating a relatively weak bioactivity presentation. Therefore, in this study, CPPs with low molecular weight were prepared, and the specific structural characteristics and immunomodulatory activity in vitro were investigated.

NMR Spectra Analysis of CPPs
The signals of 1 H and COSY NMR spectra were characterized and are presented in Figure 4A,B. As shown, the chemical shifts in the CPPs occurred from 5.50 ppm to 3.45 ppm in 1 H NMR spectra, and there were no signals between 6 and 8 ppm, indicating that no phenol or ferulic acid existed in the CPPs. According to the 1 H and COSY NMR analysis, the anomeric protons signals at about 5.45 ppm were attributed to the H1 presence of α-Glc [37]. The H signals at about 4.27 ppm and 4.12 ppm were assigned to Fru-H3 and Fru-H4, respectively. The signals from 3.48 ppm to 3.92 ppm corresponded to Fru-H1, Fru-H5, Fru-H6, and Glc-H2−H6, which were definitely assigned through the cross-peaks of adjacent hydrogens in the COSY spectrum, and would be helpful for further carbon attribution [38]. Additionally, fructose is a ketose sugar, suggesting that the C2 of fructose is anomeric carbon without linked hydrogen. Thus, the cross-absorption signals involved in Fru-H2 are nonexistent in the COSY spectrum. In terms of the 13 C NMR spectrum in Figure 5A, the signals of the CPPs appeared from 106.50 ppm to 62.90 ppm. On the basis of the DEPT135 spectrum in Figure 5B, the carbon signals at 106.02 ppm/106.48 ppm were determined to be one quaternary carbon atom; the carbon signals at 79.77 ppm/76.92 ppm, 77.06 ppm/84.04 ppm, and 83.86 ppm/76.66 ppm were confirmed to be the tertiary carbons; and the carbon signals at 63.68 ppm/62.91 ppm and 64.92 ppm/63.26 ppm were the secondary carbons, which means the structures of the A and C units were furan fructose residues with a β glycosidic bond configuration [39]. Figure 5C displays the results of the HSQC (blue signals) and HMBC (red signals) NMR spectra of the CPPs. Combined with the COSY spectrum analysis, the chemical shifts in the carbons were attributed accurately in the HSQC spectrum [40], and β-Fru and α-Glc were marked as A/C and B, respectively, which is shown in Table 4. Based on the above assignments of the 1 H and 13 C signals, six cross-peaks, including A1,2 (A1 ,2), A6C2, C6,2, B1A2, and B1,5, were observed in the HMBC spectrum. It can be inferred that CPPs mainly contained (2→1)-β-D-Fruf and (1→)-α-D-Glcp as the backbone, which is analogous to the structure of inulin [41]. Additionally, the branch of (2→6)-β-D-Fruf was demonstrated according to the HSQC and HMBC spectra analysis. In view of the above results, the CPPs can be identified as a glucofructan, and the possible structural formula is shown in Figure 5D, which is quite different from the other inulin-type fructan isolated from C. pilosula in previous literatures [42,43]. These diversities were probably caused mainly by diverse extraction and purification methods. As reported, the branched degree presented crucial roles affecting various bioactivities of polysaccharides [44]. Compared with the structure of inulin, the CPPs presented the branch of (2→6)-β-D-Fruf, which might have been responsible for the activity development.   The effects of the CPPs on RAW264.7 cell viability were determined using an MTT assay, which was a critical indicator for further application, and the determination results are shown in Figure 6A. As presented, the CPPs exhibited no significant cytotoxicity on macrophages even at a concentration of 400 µg/mL. The viabilities of RAW264.7 cells after various concentrations of CPP (6.25, 25, 100, 200, 400 µg/mL) treatments were 98.80 ± 1.54%, 101.29 ± 1.26%, 104.31 ± 2.31%, 100.22 ± 2.51%, and 102.81 ± 1.96%, respectively, suggesting that CPPs would not directly induce any side effects in RAW264.7 cells and could be used as a novel effective immunostimulant. As reported, multiple polysaccharides could not directly exert cytotoxic effects on immune cells, and could be employed for immune cell activation [45], which is consistent with our results. Figure 6. Immunoregulatory effects of the CPPs on Raw264.7 macrophages ((A) cytotoxic evaluation, (B) phagocytic rates, (C) NO production, (D) TNF-α and IL-6 cytokines secretion). Significant differences are designated as * p < 0.05 and ** p < 0.01, compared with the control group.

Macrophage Phagocytosis Analysis
In the present research, the effects of the CPPs on RAW264.7 cell phagocytosis were determined, and the results are presented in Figure 6B. As shown, compared to the control group, the phagocytic rates of RAW264.7 cells after different concentrations of CPP (6.25-400 µg/mL) treatments for 24 h were significantly enhanced by 1.05, 1.30, 1.73, 1.81, and 1.97 times those of the control, indicating that the CPPs could activate RAW264.7 cells and promote the phagocytic capacities. As is known to us, the phagocytic ability is a crucial indicator to evaluate macrophage activities, which would further reflect the immunoregulatory capacity of macrophages after bioactive compound activation [46]. In this study, the edible CPPs exerted an immune-modulating function by enhancing the phagocytic ability of macrophages, which presented similar effects with a neutral polysaccharide from ginger [47].
3.6.3. NO, TNF-α, and IL-6 Content Analysis The effects of the CPPs on NO, IL-6, and TNF-α production of macrophages were determined and the results are shown in Figure 6C,D. As displayed, CPP treatments of 6.25 µg/mL to 400 µg/mL remarkably improved NO production and TNF-α and IL-6 secretion compared to the control group, dose dependently (p < 0.05 or p < 0.01). When the CPP concentration reached 400 µg/mL, the contents of NO, TNF-α, and IL-6 increased to 25.03 ± 1.70 µM, 207.77 ± 7.91 pg/mL, and 206.16 ± 9.04 pg/mL, respectively. As reported, NO, IL-6, and TNF-α are secreted by activated macrophages under bioactive compound stimulation, contributing to eliminating pathogenic microorganisms [48]. In general, the NO was first released in the proinflammatory response of the macrophages, and then the cytokines including TNF-α and IL-6 were synthesized to destroy abnormal cells or activate T-helper cells [49,50]. These results suggest that the CPPs could enhance macrophage phagocytosis by improving the NO, TNF-α, and IL-6 levels, demonstrating that the CPPs have the potential to be a novel edible compound from C. pilosula to improve the immunoregulatory activity of the body.

Conclusions
In summary, C. pilosula polysaccharides (cCPPs) were separated from C. pilosula through ultrasonic-assisted extraction, and the maximum yields of the cCPPs reached 4.86 ± 0.16% under the obtained optimal conditions, including a liquid-to-material ratio of 33 mL/g, ultrasonic time of 81 min, and ultrasonic power of 370 W, which was close to the predicted value (4.81%). Subsequently, the novel low-molecular-weight glucofructan (CPPs) was isolated through dialysis and sephadex G-25 column purification, which presented an average molecular weight of 4.23 × 10 3 Da with (2→1)-β-D-Fruf and (1→)-α-D-Glcp as the backbone and (2→6)-β-D-Fruf as branches. Meanwhile, the CPPs exhibited strong immunostimulation activity on RAW264.7 cells, which make them valuable for further application in the functional food industry as a novel edible compound.