Physicochemical Property and Pharmacological Activity of Exopolysaccharide From Endophytic Bacterium Bacillus Sp. B3 Isolated From Polygonatum Sibiricum


 Polygonatum sibiricum is a widely used traditional Chinese medicine with various functions. In this study, Bacillus sp. B3 as its endophytic bacterium, which produced the maximum amount of exopolysaccharide (EPS), was isolated. The EPS-B3 produced by Bacillus sp. B3 was purified and identified. As suggested by the analysis of monosaccharide composition, EPS-B3 with an average molecular weight of 1.863×104 Da, was composed of three main monosaccharides, including galactose, glucose and mannose with a molecular ratio of 15: 3: 1. The monosaccharide component of EPS-B3 was different from P. sibiricum polysaccharide (PSP), but they exhibited similar antioxidant and antitumor activity. It is worth to point out that, EPS-B3 exhibited higher Ferric reducing power than PSP. As suggested by the results, the EPS produced by endophytic bacterium from the medicinal plants might have the potential to serve as an alternative of the medicinal herbs.


Introduction
Traditional Chinese medicine (TCM), which was used by a total of 80% the world's population [1] , plays a signi cant role in people's life and health. As the indispensable and valuable asset of natural and cultural heritages, they are commonly used to treat human and animal diseases. Many components with remarkable bioactivities from TCM have been isolated and characterized, such as saponins, cyrtonema, lectins, polyphenolics, and polysaccharides [2,3] . An increasing number of evidences indicated that polysaccharides derived from TCM have potent bioactivities, such as immune enhancing property, antitumor activity and antiviral activity [4][5][6] . Thus, it is urgent to nd out the alternatives of the medicinal plants.
Polygonatumsibiricum is a common traditional Chinese medicine with various functions widely used in some Asian countries [3] to help facilitate secretion of uid and quench thirst, treat dryness and cough, and so on [7,8] . The polysaccharide is one of the main bioactive components from P. sibiricum exhibiting remarkable pharmacological applications and biological activities [2,9,10] , such as antioxidant [3] , anti-tumor [11,12] , anti-atherosclerosis activity and immunity enhancement effect [10,13] .
Microbial exopolysaccharides (EPS) with high molecular weight and complex chemical structures are generated and secreted out of the cells by a broad range of microbial species [14] . The EPS have lots of advantages, such as safety, unique biological properties, and the relatively easy preparation [15,16] . The endophytic bacteria from medicinal plants are a group of microorganisms, which have symbiotic relationship with host medicinal plants. And they normally produce similar or identical metabolite to the host plants, which can be used as an important source of active substances.
Therefore, the EPS of endophytic bacteria from medicinal plants with quite signi cant biological functions could be regarded as the substitute or supplement of polysaccharides from medicinal plants. Due to their diverse functions, an increasing number of researchers are focusing on screening for valuable EPS producing bacteria.
In this study, we aim to screen and identify endophytic strains isolated from P. sibiricum producing the highest amount of EPS. Afterwards, the chemical characteristics, monosaccharide compositions, antioxidant and antitumor activity of EPS produced by Bacillus sp. B3 (EPS-B3) were investigated. Compared with PSP, the EPS-B3 exhibited similar antioxidant and antitumor activity, which provides evidences for the clinical application of EPS-B3 as an alternative of PSP.
BALA/c mice were purchased from Shandong University laboratory animal center. Bacterial Physiological and Biochemical Identi cation Reaction Tube were obtained from Binhe Co. Ltd. (Hangzhou, China).

Isolation and identi cation of the endophytic bacteria from Polygonatum sibiricum
After the surface of Polygonatum sibiricum root was disinfected, the internal tissues were placed on the separation plate evenly and cultivated in an incubator (37°C, several days). The single colonies were picked according to the colony morphology, ampli ed and conserved by a conventional method. Speci cally, the monoclonal colony was selected in the LB solid medium containing NaCl 10 g/L, yeast 5 g/L, tryptone 10 g/L and agar 15%, respectively. To isolate the targeted strains, plates were incubated in incubator (37°C, 48-72 h). The isolated single colonies were cultivated in LB liquid medium on a shaker (37°C and 200 rpm, 72 h).
The bacterium named as B3 exhibited the highest production of EPS was identi ed based on morphologic, physiological and biochemical features. After that, the 16S rDNA fragment was synthesized by PCR with two universal primers 9F (5'-GGTTACCTTGTTACGACTT-3') and 1492R (5'-AGAGTTTGATCCTGGCTCAG-3') [17] . Furthermore, it was sequenced (Beijing Genomics Institute, Beijing, China) and compared with the 16S rDNA sequences of BLAST database online. Its evolutionary with other bacteria was analyzed by MEGA (version 6.1) software through the 16S rDNA sequence analysis [18] .

Preparation of the EPS-B3
Endophytic bacterium Bacillus sp. B3 was cultured on a shaker (200 rpm, 37°C and 72 h). And fermentation broth was centrifuged (8000 g, 10 min) to remove cells. The supernatants were precipitated in ethanol (1:4, v/v) at 4°C for 12 h and then centrifuged (8000 g, 10 min). Precipitates were gathered and freeze-dried to obtain crude polysaccharides. In addition, the proteinase was added to remove the extra proteins in the supernatant using the savage method [19] . After decoloration by hydrogen peroxide, desalting through dialysis, the sample was further puri ed with the DEAE column and Sephacryl-300 HR chromatography as previous research [20] . Beyond that, polysaccharide content of the puri ed sample was analyzed as previously reported [21] and the fractions were combined and lyophilized after desalting.

Molecular Weight Determination of EPS-B3
High-performance gel permeation chromatography (HPSEC) and multi-angle laser light scattering (MALLS) spectrometer (Wyatt, DAWN HELEOS-II) were used to analyzed the homogeneity and average molecular weight (Mw) of EPS-B3 were analyzed as previously reported [20,22] . Before 250 μL samples (5 mg/mL) was added into MALLS system at ow velocity of 0.5 mL/min, the EPS sample was completely dissolved in 0.2 mol/L NaNO 2 solution which contained 0.02% NaN 3 .

Analysis the Monosaccharide composition of EPS-B3
Before monosaccharide ingredient of the sample was analyzed by ion chromatography (IC) (Dionex, Sunnyvale USA), 10 mg EPS-B3 was hydrolyzed with 3 M of tri uoroacetic acid (TFA) in a total volume of 1 mL at 110°C for 6 h. After the excess acid was discarded, 1 mL acid hydrolyzed sample (10 mg/L) was analyzed by IC using carbopac PA10 column (4 mm × 250 mm, 10 μm) (Dionex) with 10 mM NaOH at a 0.3 mL/min ow rate. Finally, the monosaccharide mixtures (containing L-fucose, L-rhamnose, D-galactose, D-glucose, D-mannose and D-fructose) were used as standard sugars and the content of each component was analyzed according to the relative area of the peak.

FT-IR determination
Fourier transform infrared (FT-IR) spectral of the puri ed sample was analyzed by using FT-IR spectrum (Bruker ALPHA-T). The EPS-B3 was mixed with KBr powder (sample: KBr approximately 1:100) and compressed into pellet for FT-IR determination between wave numbers 450 cm −1 and 4000 cm -1 .

Antioxidant property analysis of EPS-B3 and PSP
2.7.1 DPPH free-radical scavenging activity DPPH free-radical scavenging ability of PSP and EPS-B3 were investigated as previous research [23] with slight modi cations [20] . Brie y, 250 μL ethanol solution including 0.02% DPPH was added to the different concentrations (0, 2, 4, 6, 8, and 10 mg/mL) of sample solutions (1 mL). After the mixed solution was kept away from light for 6 min, absorbance of each group at 517 nm was measured. All groups were performed in triplicate and the scavenging activities of all groups were calculated by the formula followed: Where A s was the absorbance of experimental group; A c was the absorbance of the group without samples; A b was the absorbance of the group without DPPH solution.

Superoxide radical scavenging ability
The superoxide radical scavenging activities of samples were estimated by monitoring the inhibition of pyrogallol autoxidation as previous method [24] . 1 mL sample at different concentrations (0, 2, 4, 6, 8 and 10 mg/mL) was added into the test tube which contains 4.5 mL PBS (0.05 M, pH 8.2) preheated to 25°C. Afterward, pyrogallol solution (0.4 mL, 45 mmol/L) was added into the mixture and test tube was incubated (25°C, 15 min) in water bath. Finally, 3mL HCl solution (8 M) was added to interrupt the reaction and the absorbance of solutions at 320nm was analyzed. The superoxide radical scavenging activity could be calculated by the formula followed: Where A 0 was the absorbance of group without the sample, while A s was the absorbance of the experimental group.

Hydroxyl radical scavenging ability
The ·OH scavenging effect was analyzed via the Fenton's reaction [20,25] . Firstly, the mixture of 1, 10-phenanthroline (1. (0.01% v/v) was added into solutions to start the interaction. The absorbance of all groups at 510 nm was recorded. The ·OH scavenging effect was analyzed as follows: Where A s was the absorbance of experimental group; A 0 was the absorbance of group containing H 2 O 2 without samples; A b was the absorbance of group without H 2 O 2 .

Ferric reducing activity
The Ferric reducing abilities of EPS-B3 and PSP were assessed by previous modi ed method [20] . Brie y, samples (1 mL) at different concentrations (0, 2, 4, 6, 8, and 10 mg/mL) were added into 5 mL PBS (0.2 M, pH 6.6) which contained K 3 Fe(CN) 6 (5%, w/v) and incubated (50°C, 20 min) in water bath. Next, 2.5 mL TFA (10%, w/v) was added to interrupt the reaction and the mixture was centrifuged at 1200 g for 10 min. Then, 3.0 mL H 2 O containing FeCl 3 (0.0167%, w/v) were added into the supernatant (2.5 mL) and incubated (50°C, 10 min). The absorbance of the mixture was measured at 700 nm. The equal volume of H 2 O instead of polysaccharide solution was used as control. The absorbance of reaction solution at 700 nm was positively correlated with the Ferric reducing power.
2.8 Determination of antitumor activity 2.8.1 Determination of antitumor activity in vitro EPS-B3 and PSP at various concentrations (2.5-80 μg/mL) were incubated with S180 tumor cells, respectively. The blank group was treated with only the CCK8 solution without either polysaccharides or culture medium. The control group was treated with the CCK8 solution in culture medium without polysaccharides. After incubated in an incubator (37°C, 4 h), the absorbance of mixture at 450 nm was measured. The inhibition rate for S180 tumor cells was calculated as follows: The inhibition rate (%) = (A c -A e ) / (A c -A 0 ) ×100 Where A e signi ed the absorbance of experimental group; A c signi ed the absorption of blank group; A 0 signi ed the absorbance of control group.

Determination of antitumor activity in vivo
The tumor-bearing mouse model was established by injecting 0.2 mL PBS buffer containing 4 × 10 5 S180 cells into the right armpit of the mice. 7 days later model was constructed. The mice were divided randomly into four groups: a control group (PBS), a group of PSP and two groups for EPS-B3 at dosages of 100, 300 mg/kg, respectively. The mice were treated with above solution through single intragastrical administration every day in 24 days.
To monitor growth of tumors, the length and width of the tumor in armpit of S180 tumor-bearing mice were accurately measured every 4 days. The volumes and inhibition rate (TIR) of tumor were analyzed and plotted as formula follows: Tumor volume = (Tumor length × tumor width × tumor height)/2

TIR (%) = (M−NM) ×100
Where M signi ed the percentage of tumor weight to mouse weight in control group; and N signi ed the percentage of tumor weight to mouse weight in the PSP or EPS-B3 treatment groups 2.9 Determination anti-polysaccharide antibody titers Antibody titers of anti-PSP and anti-EPS-B3 in serum of S180 tumor-bearing mice were detected by ELISA. 2.5 μg/mL (100 μL/well) of PSP and EPS-B3 in 0.05 M Na 2 CO 3 (pH 9.6) buffer was added into ELISA plates and incubated at 4°C overnight. 250 μL/well wash buffer (PBS including 0.1% Tween 20) was used to wash plates for four times and 200 μL/well PBS (containing 1% BSA) was used to block plates at 37°C for 2 h. 100 μL serum dilutions from 1:100 to 1:102400 in PBS (containing 0.05% Tween 20, pH 7.4) were added into the plates and incubated at 37°C for 2 h. 250 μL/well wash buffer was used to wash plates for four times again. 100 μL/well PBS containing Horseradish peroxidaseconjugated polyclonal goat (HRP-Goat) anti-mouse IgG (1:3000, 0.05% Tween 20) were added into the plates and incubated at 37°C for 1 h. After the plates were washed as mentioned above, 100μL/well 3, 3, 5, 5-tetramethyl benzidine (TMB, Thermo sher) was added and incubated at 25°C for 15 min in the dark. 100 μL H 3 PO 4 (1 mol/L) was added to interrupt the reaction, and absorbance of the mixture at 450 nm was measured. The absorbance values of test samples were matched into the linear regression curve of a calibrated control (reference serum) to analyze the titers.

Morphological Analysis of Endophytic strains from Polygonatum
Six rhizobium endophytic strains isolated from the medicinal plant P. sibiricum were incubated at 37°C for 48h (Table S1). The colony with maximum EPS production (~1.7 g/L, about 67.1% of the total sugar content) strain B3 was round, small, moist, adhesive drawing and opaque (Table S1). And it was a gram-negative bacterium, which was positive for D-glucose, citrate salt, malonate, and etc, but negative for arabinose, H 2 S, catalase, urease, and so on (Table S2). Strain B3 was identi ed as Bacillus sp. by combining with the bacterial morphologic, physiological and biochemical features and the phylogenetic analysis results (Fig.1).Bacillus sp. is a kind of microbial species producing EPS with excellently physicochemical and biological properties [6,20,26] .

Characterization of the Molecular weight (Mw) and the monosaccharide composition of EPS-B3
The exopolysaccharide of Bacillus sp. B3 (EPS-B3) was puri ed with the DEAE cellulose column and Sephacryl-300 column as displayed in Fig.S1. The main fraction with 88.9% of the total sugar content was obtained. The homogeneity (Mw/Mn) and average molecular weights of EPS-B3 were assessed by SEC-MALLS. The peak shape in Fig.S2 has the characteristics of single, narrow, and symmetrical, indicating that EPS-B3 was a homogenized exopolysaccharide. As shown in Table 1, the EPS-B3 was a sugar including 103 monosaccharides with the Mw value 1.863×10 4 Da. And the distribution coe cient (Mw/Mn) of EPS-B3 was 1.24, suggesting that EPS-B3 has a small coe cient dispersion.
The components of EPS-B3 were analyzed and compared with standard monosaccharides by Ion chromatogram. As results shown in Fig.2, the EPS-B3 was composed of galactose, glucose, mannose with a molecular ratio of 15:3:1, respectively, which suggested that monosaccharide composition of EPS-B3 was relatively simple. It was interesting that the monosaccharide composition of EPS-B3 was totally different from that of PSP, which consists of ve different monosaccharides [3,27] .

The feature peak analysis of EPS-B3
FT-IR spectroscopy was used to measure feature peaks and glucosidic bonds of EPS-B3 (Fig.3). The sample had a broad peak at about 3400 cm −1 , indicating that EPS-B3 showed the polysaccharide characteristic which was identi ed as the O-H stretching vibration in the sugar ring [28,29] and the weak peak around 2930 cm −1 was attributed to the stretch vibration of C-H bond [28] . The spectrum at around 1650 cm −1 was the absorption of C=O bond and attributed to the bound water [30] . Some peaks between 1400 and 1000 cm −1 also stand for the characteristics of polysaccharides, such as C-O-C glycosidic bonds and C-O-H link bonds of the pyranose unit are symbol of polysaccharide [20,31] . Feature peak at 966.48 cm −1 , 814.33 cm −1 and 537.42 cm −1 were attributed to the pyranose units. Regarding the spectrum, signi cant differences were observed between PSP (Fig.S3) and EPS-B3, which indicated that EPS-B3 was a different kind of polysaccharide from PSP.

The comparative analysis of antioxidant activities between EPS-B3 and PSP
Antioxidants could effectively delay the development of degenerative disease and ageing by reducing the oxidation free radicals [32] . Herein, DPPH radical scavenging activity, O 2 -· scavenging assay, ·OH quenching assay and Ferric reducing power are used to analyze the antioxidant activities of samples. The comparison of the antioxidant activity of the puri ed EPS-B3 and the PSP was shown in Fig.4.
In the DPPH and superoxide radical (O 2 -·) assay, EPS-B3 could scavenge DPPH· or O 2 − · in a concentration-dependent manner from 2 to 10 mg/mL. The DPPH and O 2 − · scavenging ability of EPS-B3 were 39.0 ± 3.21% and 40.1 ± 4.55% at the concentration of 10.0 mg/mL, which were similar with PSP (39.5 ± 3.45% and 39.3 ± 5.21% respectively) (Fig.4A, B). In Fig.4C, the EPS-B3 could also scavenge ·OH and its scavenging ability increases with increasing EPS-B3 concentration, but its increase rate was slower than that of PSP. The maximum scavenging rate of EPS-B3 was 46.9 ± 4.41% at 8 mg/mL, lower than PSP (66.1 ± 6.12% at 8 mg/mL). Besides, it is interesting that EPS-B3 exhibited higher reducing power than PSP against the Ferric reagents (Fig.4D). The maximum reducing power of EPS-B3 (0.56) was observed at the concentration of 8 mg/mL, which is higher than that of PSP (0.35) at 10 mg/mL. The reducing power was a signi cant indicator of the antioxidant property [33] .
Overall, the antioxidant activity of EPS-B3 is relatively poor in its natural neutral form [3] . The chemically modi ed or degraded polysaccharides exhibited stronger biological activities [3,27] , so further work could be performed to improve the antioxidant activity of EPS-B3.

Analysis of the antitumor activities
It was believed that PSP showed inhibition for the proliferation of cancer cell as a potential candidate for anti-tumor therapy [11,12] . In this study, S180 tumor cells were exposed to different concentrations (2.5, 5, 10, 20, 40, and 80 μg/mL) of EPS-B3 or PSP to investigate the tumor inhibition effect in vitro. As shown in Fig. 5A, EPS-B3 can inhibit the growth of S180 tumor cells and the highest inhibition rate was 32.5% (40 μg/mL of EPS-B3), which was a little lower than PSP (48.5% inhibition at 40 μg/mL).
To measure the tumor inhibition effect in vivo, tumor bearing mouse model was established using S180 cells and treated with EPS-B3 or PSP. As shown in Fig.5B, the tumor weights in PSP group and EPS-B3 group were all signi cantly declined from the 8th day, compared with that in PBS group. In the same concentration (100 mg/kg), the inhibitory rate of PSP was higher than EPS-B3, while high concentration (300 mg/kg) of EPS-B3 exhibited stronger inhibition. This indicates that EPS-B3 possesses notable anti-tumor effects against tumors in mice. The antibody titers of anti-PSP and anti-EPS-B3 polysaccharides in the serum of treated mice were detected by ELISA. The results showed that PSP and EPS-B3 can signi cantly stimulate the mice to produce corresponding anti-polysaccharide antibodies (Fig.5C). The PSP was found to stimulate autophagy and inhibit the proliferation of tumor [12] mediated by TLR4-MAPK/NF-κB signaling pathways [11] , which was affected by many extracellular polysaccharide [16] . However the molecular mechanism of EPS-B3 for tumor inhibition needs further exploration.

Conclusion
In this study, the exopolysaccharide produced by the endophytic bacterium Bacillus sp. B3 isolated from medicinal plant P.

Con icts of interest
The authors declare that they have no known competing nancial interests or personal relationships that could have appeared to in uence the work reported in this paper.
Availability of data and material The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.
Code availability Not applicable.

Ethics approval
This work was approved by Shandong University.