Immunomodulatory and antitumor activities of the exopolysaccharide produced by potential probiotic Lactobacillus plantarum YW11 in a HT-29 tumor-burdened nude mouse model

its Abstract In this study, the exopolysaccharides (EPSs)-producing L. plantarum YW11 was initially evaluated for its tolerance to simulated artificial gastric juice and bile salts, and cell surface hydrophobicity. Subsequent study using a HT-29 tumor-burdened nude mouse model showed that the EPS had potent immunomodulatory and antitumor activities. Administration of the EPS had no influence on body weight of the mice, and it could significantly prevent HT-29 tumor cells induced acute liver and kidney damages. The purified EPS at a dose of 5 mg/mL was able to enhance effectively the killing activities of splenic natural killer cells and cytotoxic T lymphocyte, and to stimulate the secretion of serum cytokines IL-2 and TNF-α. As evidenced by a cell culture assay, the EPS exhibited obvious inhibition on growth of colon cancer HT-29 cells treated with 600 μg/mL of the EPS for 72 h with an inhibition rate of 58.02%. Administration of the EPS obviously inhibited growth of HT-29 xenograft tumor to an extent near that of the positive control (5-Fu, 25 mg/kg bw). Therefore, the results of this study indicated the important roles of the EPS in the potential probiotic functions of the strain.


Introduction
Lactic acid bacteria (LAB) are widely used in a variety of foods or medicines due to their potential benefits for human health (Shao et al., 2014;Zendeboodi et al., 2020). These microorganisms are able to produce EPSs that are either covalently associated with the cell surface in the form of capsule or secreted into the growth medium in the form of slime. EPSs produced by LAB have aroused increasing attention in recent years, not only for their unique physicochemical properties, but also for biological activities including antioxidant, antibacterial, immunomodulatory and antitumor activities (Guo et al., 2013;Zhang et al., 2013;Li et al., 2014a;Kumar et al., 2021).
As an important Lactobacillus species, L. plantarum has been frequently used in food fermentation (Crowley et al., 2012;Mirlohi et al., 2014). Some L. plantarum strains were shown to produce EPSs, of which a few were reported with bioactivities (Wang et al., 2014b;Chakraborty et al., 2020). The EPS produced by L. plantarum YW32 was found to possess scavenging abilities toward hydroxyl and superoxide radicals, inhibitory effect on biofilm formation by several pathogenic bacteria, and inhibitory activity against colon cancer HT-29 cells (Wang et al., 2015a). The EPS from L. plantarum R315 exhibited antibacterial ability against pathogens, and scavenging activity against DPPH, hydroxyl, and superoxide radicals . Both the EPS and the sulfated EPS from L. plantarum ZDY2013 had radical scavenging activities, and the antioxidant activities increased after sulfonation (Zhang et al., 2016). The EPS from L. plantarum C88 was shown to be effective in scavenging reactive oxygen species in vitro (Zhang et al., 2013). However, the roles of the EPSs produced by L. plantarum strains in their probiotic functions were not well studied.
In this study, L. plantarum YW11 strain isolated from Tibet kefir grains was initially evaluated for its potential probiotic properties by in vitro gastrointestinal tolerance and bacterial surface hydrophobicity tests. The EPS produced by L. plantarum YW11, which was previously characterized with monosaccharide composition of glucose and galactose in a molar ratio of 2.71:1 and a molecular mass of 1.1 × 10 5 Da (Wang et al., 2015a), was further studied using a HT-29 tumor-burdened nude mouse model and the human colon cancer cell line HT-29 to evaluate its immunomodulatory and antitumor activities in order to understand the roles of the EPSs produced by L. plantarum strains in their probiotic functions.

Animals and cell lines
Human colon cancer cell line HT-29 was obtained from Chinese Academy of Medical Sciences Cancer Hospital (Beijing, China). Cells were grown in DMEM medium (high glucose) supplemented with 10% fetal bovine serum, 100 IU/mL penicillin, 100 μg/mL streptomycin, and incubated at 37 °C in a humidified atmosphere with 5% CO 2 . BALB/c nude mice (male, 6 weeks old) weighing 18-22 g were purchased from Beijing Vital River Laboratories Co., Ltd. (Beijing, China).

Bacterial cultures
L. plantarum YW11 isolated and identified from Tibet kefir grains, and L. rhamnosus GG (LGG) isolated from commercial yoghurt were stored in 30% (v/v) glycerol at -80 °C in the Dairy Laboratory of Beijing Technology and Business University, China. The strains were inoculated into MRS (Man Rogasa Sharpe) (Difco, Sparks, MD, USA) broth and grown at 37 °C for 12 h to obtain a viable count of log 8-9 CFU/mL for use in this study.

Determination of bacterial tolerance in simulated artificial gastric juice
The bacterial culture of L. plantarum K25 was taken, and centrifuged at 6000 × g 4 °C for 10 min. The cell pelletes were collected, washed twice with sterile PBS buffer (pH 6.5) and added with 5 mL PBS buffer (pH 6.5) to make bacterial suspension. 1 mL of bacterial suspension was added to a test tube containing 9 mL of simulated gastric juice (0.2 g/100 mL NaCI, 0.35 g/100 mL pepsin, filtered and sterilized) at pH 3.0, and placed at 37 °C for anaerobic incubation for 3 h. Samples were taken at 0 h and 3 h after inoculation, and then diluted and spread on MRS agar plates. After 48 h culture at 37 °C, viable bacterial counting was performed (Vinderola & Reinheimer, 2003). The survival rate of the strain is calculated as follows Equation 1 ( ) N 1 is the number of viable bacteria after gastric juice treatment for 3 h; N 0 is the number of viable bacteria after gastric juice treatment for 0 h

Determination of cell surface hydrophobicity
The L. plantarum YW11 culture was centrifuged at 6000 × g 4 °C for 10 min. The bacterial pelletes were collected and washed twice with sterile PBS buffer (pH 6.5). The bacterial concentration was adjusted with PBS to reach an absorbance of 1.0 at 560 nm. 3 mL of the adjusted bacterial suspension was added with 0.6 mL of xylene, chloroform and ethyl acetate, respectively. The mixture was oscillated by eddy. After static stratification, the lower aqueous phase was taken, and the absorbance was measured at 560 nm with blank set to zero by PBS. The hydrophobic rate was calculated as follows Equation 3: A 0 is the absorbance before mixing the bacterial suspention and adsorbent; A is the absorbance of the bacterial suspension after mixing with the adsorbent.

Isolation and purification of EPS
The EPS produced by L. plantarum YW11 was isolated and purified from the growth culture of L. plantarum YW11 according to our previously reported method (Wang et al., 2015b). The purified form of the EPS was used for evaluation of bioactivities in this study.

Establishment of HT-29 tumor-burdened nude mouse model
HT-29 cells were cultured in DMEM medium at 37 °C for 24 h.The concentration of the cells was adjusted to 1.0×10 7 cells/ mL in PBS. Then 0.2 mL of HT-29 cells suspension was inoculated subcutaneously into the right axilla of each mice at the beginning of the experiment. After inoculation for 24 h, the tumor bearing mice were divided into five groups (ten mice per group): model control group (0.9% saline), positive control group (5-Fu, 25 mg/kg bw), low-dose EPS group (10 mg/kg bw), mid-dose EPS group (25 mg/kg bw), high-dose EPS group (50 mg/kg bw). Another ten mice without tumor inoculation and drug administration were used as normal control group (0.9% saline). Drugs were orally administered to mice after grouping. The designated amounts of drugs were applied in 0.2-mL volume once a day for 14 days. In the course of treatment, tumor volumes and body weights were monitored every other day. The tumor volume was calculated using the following Equation 4: L and S (mm) were the largest and smallest diameters of the tumor, respectively.

Determination of tumor inhibition rate and organ indices
24 h after last drug administration, all the mice were weighed and killed by dislocation of cervical vertebra and the tumors, spleens, livers, kidneys and hearts were collected and weighed. The inhibition rate of tumor growth was calculated according to the following equation: inhibition rate (%) = [(tumor weight of the mice in model control group−tumor weight of the mice in EPS treatment group)/tumor weight of the mice in model control group] × 100. Relative spleen/liver/kidney/heart weights were expressed as the ratio of the spleen/liver/kidney/heart to body weight (mg/g).

Hematological and serum biochemical parameters assays
24 h after last drug administration, the blood of each mouse was collected with the anticoagulant in polystyrene tubes. The hematological parameters such as hemoglobin (Hb) content, red blood cell count (RBC), white blood cell count (WBC), total leukocytes count and lymphocyte count were determined by using analytical hematology system Symex XE-2100 (Symex Cooperation, Japan). After that, the serum of each mouse was obtained by centrifugation (3000 × g, 10 min, 4 °C) and stored at -80 °C for further analysis. The clinical parameters, including AST, ALT, BUN, Cr, SOD and MDA levels in serum were all determined spectrophotometrically using commercially available reagent kits (Nanjing Jiancheng Bioengineering Institute, China) following the manufacturer's instructions.

Assay of NK cell activity
The activity of NK cell was measured using the method described by Wang with slight modifications (Wang et al., 2014a). Briefly, colon cancer cell HT-29 (target cells) and splenocytes (effector cells) were mixed in 96-well U-bottom microtiter plate in a 50:1 ratio and incubated at 37 °C in 5% CO 2 atmosphere. After 24 h incubation, 50 μL of MTT solution (2 mg/mL) was added to each well and the plate was incubated for another 4 h and subjected to MTT assay. NK cell activity was calculated according to the following Equation 5: where A T is the absorbance value of target cells control, A S is the absorbance value of test samples, A E is the absorbance value of effector cells control.

Assay of CTL activity
The CTL activity was analyzed using MTT method as described above. Colon cancer cell HT-29 (target cells) and splenocytes (effector cells) were mixed in a 50:1 ratio. CTL activity was calculated according to the following equation: CTL activity where A T is the absorbance value of target cells control, A S is the absorbance value of test samples, A E is the absorbance value of effector cells control.

Assay of serum cytokine levels
The level of interleukin-2 (IL-2) and tumor necrosis factoralpha (TNF-α) in the serum were measured using commercially available kits from R&D systems Inc. (Minneapolis, USA), with a solidphase enzyme-linked immunoabsorbent assay (ELISA) as described by the manufacturers.

Assay of HT-29 cell proliferation as affected by EPS
The in vitro antitumor activity of the EPS was determined by a MTT assay using human colon cancer HT-29 cells. Briefly, HT-29 cells were incubated in 96-well microtiter plates (2.0 × 10 4 cells per well) under 5% CO 2 at 37 °C for 24 h. Then the tumor cells were treated with various concentrations of the EPS (50, 100, 200, 400 and 600 μg/mL) and 50 μg/mL of 5-FU (positive control) for 24 and 72 h, respectively. At the end of each treatment, MTT reagent (5.0 mg/mL) was added to each well (10 μL per well). After 4 h, the liquid was removed, and 100 μL of DMSO was added to each well. After dissolving the formazan crystal that formed, the absorbance was measured by a Synergy TM 2 microplate reader (BioTek Instruments, Inc., Burlington, VT) at 570 nm. The inhibition ratio was expressed as follows Equation 6: where A control and A blank were the absorbances of the system without the addition of the EPS and cells, respectively.

Statistical analysis
The results were presented as the mean ± standard deviation (SD) and were analyzed by one-way analysis of variance (ANOVA) using SPSS version 16.0. The values of P< 0.05 were used to identify statistically significant differences.

In vitro gastrointestinal tolerance of L. plantarum YW11
The prerequisite for probiotics to exert their health beneficial function in human body is to ensure their survival in the acidic environment of gastric juice and the bile salt environment of intestinal tract. The pH value of human gastric juice is generally between 1.3 and 1.8 on a fasting stomach, which rises to about 3.0 or higher after eating, and the food stays in the stomach for about 3 h, bile salt content in normal human small intestine fluctuates between 0.03% and 0.3% (Kosmerl et al., 2021). In this study, the tolerance of L. plantarum YW11 was evaluated by simulating the low pH environment and high bile salt in human gastrointestinal tract by in vitro experiments. LGG, the wellknown probiotic strain with clear probiotic characteristics, was used as the control strain. As shown in Table 1, L. plantarum YW11 showed good tolerance in simulated artificial gastric juice environment. The survival rate reached 82.28%, but lower than that (93.80%) of LGG after 3 h of culture. Assay of the tolerance to bile salt at 0.3, 0.5 and 1.0 g/100 mL for 3 h showed that the survival rate of L. plantarum YW11 decreased with the increase of bile salt concentration. When bile salt was at 0.6 g/mL, the survival rate of L. plantarum YW11 was 95.35%, higher than that of LGG (91.16%). However, increasing bile salt to 1.0% decreased significantly the survival rate of L. plantarum YW11 to about 70%.

Determination of cell surface hydrophobicity of L. plantarum YW11
Probiotics are required to not only tolerate the environment of gastrointestinal fluid, but also adhere to intestinal epithelial cells to ensure their colonization to exert probiotic functions. The adhesion ability of a strain is positively correlated with the bacterial hydrophobicity, and high hydrophobicity is conducive to the colonization of the strain in the intestinal tract (Zeng et al., 2020). In this study, the hydrophobic capacity of L. plantarum YW11 in three organic solvents, including xylene, chloroform and ethyl acetate was determined. Xylene can be used as a nonpolar solvent to test the surface hydrophobicity of the strain. As shown in Table 2, L. plantarum YW11 showed high hydrophobicity at a rate >50%, indicating that L. plantarum YW11 had a hydrophobic cell surface.
The electron acceptor and electron donor on the bacterial surface of a strain can be determined by detecting the affinity of the strain to chloroform (unipolar, acidic organic solvent, electron acceptor) and ethyl acetate (unipolar, alkaline organic solvent, electron donor). As shown in Table 2, L. plantarum YW11 had a high hydrophobicity to chloroform at a hydrophobicity rate of 62.25%, indicating that L. plantarum YW11 was a strong electron donor. The hydrophobicity of L. plantarum YW11 to ethyl acetate was lower (hydrophobicity <30%), indicating that the strain was a weak electron acceptor.

Effect of the EPS from L. plantarum YW11 on body weights, organ indices, serum clinical and hematological parameters
During the whole process of the in vivo assay experiments of this study, the activities of all mice, their body weights and physiological parameters were checked. As shown in Table 3, the gain of body weight of the mice in all doses of EPS treatment groups had no significant difference from those in the model control group, indicating that the EPS from L. plantarum YW11 had no inhibitory effects on the growth of weight. However, treatment with 5-FU significantly decreased the gain of body weight of the mice. Compared with the model control group, the spleen indices of mice in all doses of EPS treatment groups were slightly elevated in different degrees (P>0.05). Conversely, the spleen indices of the mice in positive control group were reduced,  Means in the same row followed by different superscripts are significantly different (P < 0.05); AB Means in the same column followed by different superscripts are significantly different (P < 0.05).
indicating that 5-FU caused a decline in immune function to a certain extent. In addition, there was no significant difference in liver, kidney and heart indices between the EPS treatment groups and the model control group. Bárbara F. Cordeiro reported that during colitis induction, mice that consumed the probiotic cheese exhibited reduced in the severity of colitis, with attenuated weight loss, lower disease activity index, limited shortening of the colon length, and reduced histopathological score (Cordeiro et al., 2021).
Serum clinical parameters including liver-related ALT and AST activities, and kidney-associated BUN and Cr levels, were measured to verify whether the EPS from L. plantarum YW11 would cause damage to the normal organs, especially liver and kidney. As shown in Figure 1A-1D, an obvious increase in ALT, AST, BUN and Cr was observed in HT-29 tumorbearing mice as compared with those of the normal control group. However, EPS treatments at different doses decreased significantly (P < 0.01) serum AST, ALT, BUN and Cr, and the activities of AST and BUN were reduced to the levels near those in the normal control group, indicating no damage to the body by EPS treatment. However, for the 5-FU treatment group, the serum ALT and AST activities increased, suggesting an adverse effect on liver function by 5-FU. Hematological parameters in HT-29 tumor-burdened nude mice on day 14 were shown in Table 4. All EPS-treated mice had significantly (P < 0.01) increased WBC and LYM counts as compared with those of the model control mice, confirming the immunomodulatory effect of the EPS. However, treatment with 5-FU caused declined immunity with significantly (P < 0.05) decreased WBC and LYM counts.
Inoculation of tumor cells in mice was found to cause hepatocellular and kidney cellular necrosis, characterized with formation of lipid peroxidation products and a decrease Figure 1. Effect of the EPS from L. plantarum YW11 on serum AST, ALT, BUN and Cr levels in HT-29 tumor-burdened nude mice. Data are expressed as mean ± SD (n = 10 mice per groups), *P < 0.05, **P < 0.01 compared with model control group; ## P < 0.01 compared with positive control group. Data were expressed as mean ± SD (n = 10 mice per groups). a P < 0.05, significantly different from model control group; aa P < 0.01, highly significantly different from model control group; bb P < 0.01, highly significantly different from positive control group; c P < 0.05, significantly different from normal control group; cc P < 0.01, highly significantly different from normal control group.
in antioxidative enzymes (Wang et al., 2014b). Serum ALT and AST enzymatic activities were considered to be sensitive indicators of hepatocellular injury, which caused leakage of the enzymes from the hepatocytes. Serum BUN and Cr levels were good biomarkers of renal injury, and their high content often indicated malfunctions of kidney (Fu et al., 2012). In this study, treatment with the EPS from L. plantarum YW11 had no influence on body weight, and no toxicity to liver and kidney. The significantly (P<0.01) decreased ALT and AST activities, and the levels of BUN and Cr in serum of HT-29 tumor-burdened nude mice indicated that treatment with the EPS from L. plantarum YW11 could significantly prevent HT-29 tumor cells induced acute liver and kidney damages.

Immunomodulatory activity of the EPS from L. plantarum YW11
The immunomodulatory activity of the EPS from L. plantarum YW11 was assayed by measuring the cytotoxic activities of NK cells and CTL of mice. As shown in Figure 2A, the EPS (25 and 50 mg/kg) significantly increased the splenic NK and CTL cytotoxic activities of HT-29 tumor-burdened nude mice when compared with those of the model control mice (P < 0.05 or P < 0.01). However, the NK cell and CTL activities of the model control mice were higher than those in 5-FU-treated mice.
EPS treatment of HT-29 tumor-burdened nude mice stimulated production of serum cytokines to different degrees when compared with the model control group, and significant (P < 0.01) increase of IL-10 and IL-2 levels was observed with EPS at 50 mg/kg ( Figure 2B). The TNF-α level also increased significantly (P < 0.05 or P < 0.01) in all doses of EPS-treated mice. Conversely, there was no significant difference in the serum levels of IL-10, IL-2 and TNF-a between the 5-FU treatment group and the model control group.
The immuno-regulating activity of EPS was thought to be influenced by its molecular weight, monosaccharide composition, electric charges, chain conformation, and water-solubility (Zjawiony, 2004). Hidalgo-Cantabrana hypothesized that EPSs having negative charges and/or low molecular weight were able to act as strong/mild stimulators of immune cells, whereas those neutral EPSs with a high molecular weight exhibited a weak stimulation and even suppressive profile (Hidalgo-Cantabrana et al., 2012). This was in agreement with the study by López, that the acidic and lower-molecular-weight EPSs produced by Bifidobacterium strains demonstrated stronger immune response stimulation than those of the neutral EPSs with high molecular weight (Lopez et al., 2012). The acidic nature and relatively low molecular weight (1.1 × 10 5 Da) of the EPS from L. plantarum YW11 as reported earlier (Wang et al., 2015a), might be relevant to its strong immunomodulatory activity as shown in this study. However, the mechanism of how the EPS from L. plantarum YW11 affected the intracellular immune system of the HT-29 tumor-burdened nude mice needs to be further studied.

Antitumor activity of the EPS from L. plantarum YW11
The in vitro inhibitory activity of the EPS from L. plantarum YW11 on the growth of HT-29 cells was determined by a MTT assay of the viability of the cells treated with increasing concentrations of the EPS for 24 and 72 h. As shown in Figure 3A, C, the EPS exerted inhibitory activities on HT-29 cells in time-and concentration-dependent manners. Treatment of the cells for 24 h with the EPS (50-600 μg/mL) resulted in inhibition rates of 30.65-55.11%, which were higher than those of the chemotherapy drug 5-FU (28.64 ± 2.39%). When treated with 600 μg/mL of the EPS for 72 h, the inhibition rate reached a maximum of 58.02 ± 3.93%, which was also higher than that of the treatment with 5-FU (50.30 ± 3.14%) for 72 h.
The in vivo antitumor activity of the EPS from L. plantarum YW11 was determined by monitoring the tumor growth of the male BALB/c nude mice bearing HT-29 colon cancer tumors during treatment with different doses of the EPS and daily injection of 5-FU for 14 days. As shown in Figure 3B, the tumor volume of model control group increased rapidly, while the 5-FU group and all EPS-treated groups had significant tumor inhibition effects. On day 14 after the last treatment, the Figure 2. Effect of the EPS from L. plantarum YW11 on splenic NK cells and CTL cytotoxic activities (A), and on serum cytokines IL-2 and TNF-α levels in HT-29 tumor-burdened nude mice (B). Data are expressed as mean ± SD (n = 10 mice per groups), *P < 0.05, **P < 0.01 compared with model control group; # P < 0.05, ## P < 0.01 compared with positive control group. average tumor volumes of 5-FU, low-dose (10 mg/kg), mid-dose (25 mg/kg) and high-dose EPS (50 mg/kg) groups were 52.20 ± 3.21%, 82.85 ± 8.24%, 73.74 ± 6.38% and 63.42 ± 5.43% of that of model control group, respectively. Therefore, the EPS treatment obviously inhibited the growth of HT-29 xenograft tumor in a concentration dependent manner and the maximal inhibition rate (36.58 ± 3.34%) achieved was approaching that of the treatment with 5-FU. Previously, EPSs from LAB were also reported to exert protective effects against cancer (Choi et al., 2006). The antitumor activity of polysaccharides was dependent on their chemical composition, molecular weight, monosaccharide composition, chain conformation and even charge characteristics . The immuno-regulating activity of the EPS from L. plantarum YW11 as shown in this study played a role in its antitumor activity, but the underlying mechanism needs to be further studied.