3.1 Purification, molecular weight, monosaccharide composition, and FT-IR spectrum of MBP-2
As shown in Fig. 1A, the main DEAE-elution fraction comprised the majority of crude MBPs, marked as MBP-1. The MBP-1 was dialyzed using a 3 kDa dialysis bag, and the glucan gel-purified sample was labelled as MBP-2.
As analyzed by HPGPC (Fig. 1B), a single symmetric peak occurred at around 21.5 min, suggesting favorable homogeneity of the MBP-2. Further molecular weight fitting analysis demonstrated its number averaged molecular weights (Mn) as 15.3 kDa and weight-averaged molecular weights (Mw) as 21.7 kDa, and the Mw was smaller than the previous report[19, 20].
The ion chromatogram of monosaccharide composition of the MBP-2 was shown in Fig. 1C. Glucose was the main monosaccharide, indicating the potential glucan structure of the MBP-2. There was a distinct difference from the previous report in terms of the monosaccharide composition of MBPs[19, 20].
As displayed in FT-IR in Fig. 1D, a broad and intense absorption peak for stretching vibration of O-H appeared at 3431 cm− 1, while an absorption peak for stretching vibration of C-H as the characteristic peak of saccharides appeared at 2926 cm− 1[21]. The absorption peaks at 1634 cm− 1 was attributed to the stretching vibration of C-O[22]. The appearance of an absorption peak at 931 cm− 1 demonstrated asymmetric stretching vibration of the pyranose ring of d-glucose[23]. Besides, the stretch peak was at 849 cm− 1, demonstrating α-glycosidic linkage[24]. Thus, the MBP-2 might be composed of d-glucose and α-glycosidic linkages[25]. No distinct peak was noticed at 1730 cm− 1, which implied the absence of uronic acid[26], consistent with the monosaccharide composition analysis.
3.2 Methylation analysis
GC-MS was performed following methylation reactions, and the total ion chromatogram was shown in Fig. S1. Peak analysis demonstrated that, there were three main peaks (peak 1, 2, and 3) with the retention time of 8.87, 14.12, and 18.27 min, and the relative molar amount of 11.4%, 78.5%, and 6.75%, respectively (Table 2). The secondary MS data pointed out that the MBP-2 sample was mainly composed of a backbone linked by →4)-α-D-Glcp-(1→ and contains α-D-Glcp-(1→ and →4, 6)-α-D-Glcp-(1→ (Fig. S2).
Table 2
Methylation analysis for MBP-2
Time(min)
|
Major mass fragment
(m/z)
|
Deduced residues
|
Molar ratio
|
8.87
|
239.19, 205.11, 162.08, 145.08, 118.05, 102.06, 87.04
|
T-Glc(p)
|
1
|
14.12
|
233.09, 162.07, 118.06, 87.05
|
4-Glc(p)
|
6.88
|
18.27
|
261.09, 231.07, 201.06, 142.05, 118.05, 102.06, 59.04
|
4, 6-Glc(p)
|
0.59
|
3.3 NMR analysis
NMR was performed to further characterize the structure of saccharide subunits of MBP-2. The major hydrogen signals of MBP-2 varied between δ 3-5.5 ppm in 1H-NMR (Fig. 2A) while between δ 60–105 ppm in 13C-NMR (Fig. 2B). Usually, most of the α-anomeric protons vary between 5–6 ppm and the β-anomeric protons vary between 4–5 ppm[27]. Here, an anomeric proton signal appeared at δ 5.3 ppm in 1H-NMR, demonstrating an α-configuration. In the meantime, an anomeric carbon signal appeared at δ 99.6 ppm in 13C-NMR. Both signals were attributed to the H1 and C1 signals of →4)-α-D-Glcp-(1→[28]. According to the relative intensity, the residues were labelled as A, B, and C from high to low. The results of GC-MS showed that residue A had the highest intensity as 78.5%. Correspondingly, residue A had the strongest signals in 1H-NMR and 13C-NMR, and the signals at δ 3.9, 3.78, 3.77, 3.58 ppm, except δ5.34 ppm, all could be attributed to the hydrogen signals of residue A. Similarly, the signals at δ 99.6, 76.8, 73.3, 71.6, 71.2, and 60.5 ppm in 13C-NMR were carbon signals of residue A. Based on the cross-peak signals in 1H-1H COSY, the hydrogen signals of residue A, including H1, H2, H3, H4, and H5, were δ 5.34, 3.55, 3.91, 3.61, and 3.77 ppm, respectively (Fig. 2C). While based on the cross-peak signals in HSQC, the carbon signals C1, C2, C3, C4, and C5 were respectively δ 99.56, 71.58, 73.29, 76.81, and 71.24 ppm (Fig. 2F). The chemical shifts of H6a/6b (δ 3.79, and 3.54 ppm) and C6 (δ 60.51 ppm) were also obtained from HSQC. The downfield shift of C4 also indicated that the residue A was →4)-α-D-Glcp-(1→. The above results were also supported by previous reports[29, 30].
As for residue B, the signals at δ 4.91, 3.54, and 3.88 ppm were H1-H3 signals. The other hydrogen signals of residue B were directly displayed by HSQC due to a high overlap between signals in COSY. Relative to A-C4, B-C4 did not develop downfield shift given the absence of substituents at the C4 site, consistent with the previous literature[31]. The chemical shift of residue C signals and corresponding attributes were obtained using the same method and shown in Table 3.
Table 3
1H and 13C NMR chemical shifts of MBP-2 recorded in D2O
|
Glycosyl residues
|
H1/C1
|
H2/C2
|
H3/C3
|
H4/C4
|
H5/C5
|
H6a,b/C6
|
|
A
|
→4)-α-D-Glcp-(1→
|
5.34
|
3.55
|
3.91
|
3.61
|
3.77
|
3.79
|
3.54
|
|
|
99.56
|
71.58
|
73.29
|
76.81
|
71.24
|
60.51
|
|
B
|
α-D-Glcp-(1→
|
4.91
|
3.54
|
3.88
|
3.36
|
3.77
|
3.89
|
3.37
|
|
|
98.58
|
71.7
|
73.29
|
69.32
|
76.71
|
60.51
|
|
C
|
→4,6)-α-D-Glcp-(1→
|
5.34
|
3.59
|
3.89
|
3.90
|
3.78
|
3.98
|
|
|
|
99.94
|
71.66
|
73.33
|
76.72
|
71.27
|
70.22
|
|
HMBC spectrum can display the inter-residue connectivities with glycosidic bonds. As shown in Fig. 2E, there were cross-peaks between A-H4 (3.36 ppm) and A-C1 (99.56 ppm), and between A-H1 (5.34 ppm) and A-C4 (76.81 ppm), demonstrating a →4)-α-D-Glcp-(1→4)-α-D-Glcp-(1→ (ie. 4A1→4A1) structure. The cross-peaks between A-H1 (5.34 ppm) and C-C4 (76.72 ppm) and between C-H4 (3.90 ppm) and A-C1 (99.56 ppm) indicated a →4)-α-D-Glcp-(1→4, 6)-α-D-Glcp-(1→4)-α-D-Glcp-(1→ (ie. 4A1→4, 6C1) structure. The cross-peaks between A-H4 (3.61 ppm) and C-C1 (99.94 ppm) implied a →4, 6)-α-D –Glcp-(1→4)-α-D-Glcp-(1→ (ie. 4A1→4, 6C1) structure. The cross-signals between B-H1 (4.91 ppm) and C-C6 (70.22 ppm) suggested a α-D-Glcp-(1→6, 4)-α-D-Glcp-(1→ (ie. B1→6, 4C1) structure.
In the NOESY spectrum (Fig. 2D), the cross-peaks between A (H1) and A (H4), except for the existing signals in 1H-1H COSY, further demonstrated the presence of the 4A1→4A1 structure, while the cross-peaks between A (H1) and C (H4) indicated the 4A1→4, 6C1 structure. Additionally, the glucose at positions 1 and 5 formed a ring, resulting in cross-peaks between A (H1) and A (H5) and between B (H1) and B (H5). The results were in line with the results of HMBC. Combining the above findings, the structure of MBP-2 might be characterized as that in Fig. 3.
Currently, glucan polysaccharides have been only reported in the root of white mulberry, while they are starchy polysaccharides with low water solubility[32]. The present study, for the first time, identified glucan polysaccharides from mulberry branch, which are water-soluble and worthy of further development and utilization.
3.4 Immunoregulatory activity
3.4.1 Cell survival and NO content
To assess the safety of MBP-2, CCK-8 was performed to detect the viability of RAW 264.7 cells treated with MBP-2 (1-800 µg/mL) (Fig. 4A). The result demonstrated that MBP-2 at a low concentration (1–10 µg/mL) significantly promoted the proliferation in RAW 264.7 cells (P < 0.05). Besides, the cell proliferation decreased when the concentration of MBP-2 increased, and the decline was extremely significant upon 800 µg/mL (P < 0.01). MBP-2 at 400 µg/mL contributed to the most promotive effect on cell proliferation and thus selected as the highest concentration in further analysis.
NO is an important molecular messenger that can mediate a series of host-defense functions executed by activated macrophages. In addition, it is also a type of cytotoxic agent with immunoregulatory activity[33]. To explore the potential immunoregulatory activity of MBP-2, the NO produced by RAW 264.7 cells was examined. As shown in Fig. 4B, MBP-2 at 50–400 µg/mL remarkably activated the production of NO by RAW 264.7 cells, consistent with the positive control (LPS). This suggested that MBP-2 might have immunostimulatory effects.
3.4.2 Expression of relevant cellular genes
Inducible nitric oxide synthase (iNOS) can act through NO synthesis[34]. The immunoregulatory activity of native products is largely attributed to their capability of inducing multiple chemokines (e.g., MCP-1) and cytokines (e.g., TNF-α, IL-6) [35, 36]. This study analyzed the effect of MBP-2 on expression of relevant genes in RAW 264.7. It was found that MBP-2 at 50–400 µg/mL significantly stimulated the expression of iNOS in RAW 264.7 cells (Fig. 5A), consistent with the positive control (LPS) and NO production trend (Fig. 4B). The result indicated that MBP-2 could activate the cellular defense functions by up-regulating iNOS expression and inducing NO synthesis. Moreover, MBP-2 also led to distinct up-regulation of IL-1β, IL-6, and MCP-1 gene in RAW 264.7 cells (Fig. 5B-D), thereby exerting its immunoregulatory activity.
3.4.3 Cytokine level
Cytokine is a class of bioactive macromolecular proteins that are produced upon an external stimulus to immune cells and play an important role in regulating the body's immunity and inflammatory response[37, 38]. Pro-inflammatory cytokines TNF-α and IL-1β can act on macrophages to enhance immune responses and induce the expression of other immunoregulatory factors[39]. The current study found that MBP-2 significantly elevated the expression of TNF-α and IL-6 (Fig. 6A-B), thereby exerting its immunoregulatory activity.
The above findings demonstrated favorable immunoregulatory activity of MBP-2. It was reported that the immunoregulatory activity of MBPs might be attributed to their (1→4)-α-D- glucan backbone[40]. The immunoregulatory polysaccharides from traditional Chinese medicines have been applied in development of vaccines as adjuvants to promote the immune response of the body[41, 42]. Therefore, the MBP-2 may have the potential to be used as vaccine adjuvants.
The secretion of cytokines (IL-1β and TNF-α) can be mediated by multiple signals, among which Toll-like receptor (TLR) is the most significant[43]. Native polysaccharides can modulate immune cell functions in vitro by interactions with immunoreceptors, such as TLRs[44]. To investigate the mechanism underlying the immunoregulatory activity of MBP-2, this study selected C29 and TAK-242 for analysis. C29 is a TLR2 inhibitor that can block hTLR2/1 and hTLR2/6 signals[45]. TAK-242, also known as Resatorvid, is a selective inhibitor of TLR4 signal transduction that can down-regulate the expression of TLR4 downstream signaling molecule Myeloid differentiation factor 88 (MyD88) and TIR-domain-containing adaptor inducing interferon-β (TRIF) [46, 47]. As analyzed, TAK-242 dramatically suppressed the production of IL-6 (P < 0.01), and the IL-6 level was not significantly different with that of the Ctrl group (Fig. 6C). The result suggested that TAK-242 can inhibit the immunoregulatory activity of MBP-2. In addition, IL-6 production was also decreased by C29, but the effect was largely different with that of the Ctrl and TAK groups (P < 0.01). Similar results were obtained for TNF-α (Fig. 6D). However, there was a large difference between the Ctrl and TAK groups when the MBP-2 concentration was too high or the concentration of inhibitors was too low, which might be due to the large stimulating effect of 10 µg/mL MBP-2 on TNF-α production. In general, the trend for IL-6 and TNF-α production was similar, and the TAK-242 was more potent than C29 in inhibiting their production (P < 0.01). Collectively, MBP-2 may exert its immunoregulatory activity via TLR4, during which TLR2-related pathways are potential participants.
Research has revealed that TLR2 regulates downstream mitogen-activated protein kinase (MAPK) and nuclear factor kappa-B (NF-κB) via MYD88-dependent pathways, thereby playing its immune-stimulating effect. Except the MYD88-dependent pathways, TLR4 can also play its role through the TRIF pathway[48]. The current study found that the TLR4 inhibitor, TAK-242, had more significant suppressive effect on cytokine production in cells treated with MBP-2, as compared to the TLR2 inhibitor C29, indicating that MBP-2 acted mainly via the TRIF-dependent pathways while the MYD88 pathways might also make some contributions. Since the immunoregulatory activity of MBPs has been rarely reported, further studies are required.