A Atractylodes lancea polysaccharide inhibits metastasis of human osteosarcoma U‐2 OS cells by blocking sialyl Lewis X (sLex)/E‐selectin binding

Abstract In this study, a new water and alkaline‐soluble polysaccharide (ALP), with an average molecular weight of 6.63 × 104 Da, was successfully purified from the rhizomes of Atractylodes lancea. GC analysis demonstrated that ALP was a kind of glucan. The effect of the ALP on the interaction between E‐selectin and sialyl Lewis X (sLex) was examined in human osteosarcoma U‐2 OS cells. It was obvious that the expression of sLex antigen on the surface of U‐2 OS cells was visible under fluorescence microscopy. The addition of ALP (0.5, 1 and 2 mg/mL) resulted in a marked inhibition on the adhesion, migration and invasion of U‐2 OS cells to human umbilical vein endothelial cells (HUVECs), which was achieved by the decreased sLex expression on U‐2 OS cells. Additionally, the induction of apoptosis can be observed in U‐2 OS cells following ALP treatment using TUNEL staining and Annexin V‐FITC/PI double‐staining analysis on flow cytometry. In conclusion, these results indicated that ALP exerted anti‐metastatic activity towards osteosarcoma cells via inhibition of sLex/E‐selectin binding, which suggested that ALP could be a potent agent for human osteosarcoma intervention.

tumour metastasis are identified in many types of carcinoma including colon cancer, 6 lung carcinoma 7 and melanoma. 8 Accordingly, any E-selectin antagonist blockading the adhesion of sLe x and E-selectin can decrease tumour cell motility and metastasis. A number of evidences have shown the synthesized anti-adhesion peptides are limited because of their short half-life and high dosage required. 9 To eliminate or ameliorate this drawback, a new series of derivative of synthesized peptides was designed and synthesized peptides with more repeat sequence exhibit a stronger anti-metastasis effect than non-repeat peptides. 10,11 This clue throw light on us to search if there is some natural polymers or its chemical derivative owning the same properties as repeated peptides, but with long half-life and long-lasting stability.
One of the most successful natural polymers is polysaccharide and it is easily accessible from natural source. The rhizomes of Atractylodes lancea (Thunb.) DC. are a very popular traditional Chinese medicine and have been used extensively to treat several disease including digestive disorders, influenza night blindness and rheumatic diseases. 12,13 Previous pharmacological studies have indicated that essential oils and aqueous extract from A lancea had promising apoptosis-inducing activity towards human cancer cells. 14,15 As far as we are aware, polysaccharides are principal components dissolved in aqueous extract. Increasing evidence has demonstrated that A lancea polysaccharides possessed intestinal immune system modulating activity 16,17 ; however, until now, the study regarding antitumour and anti-metastatic activities of A lancea polysaccharide are not fully investigated, let alone for OS. In an effort for discovering novel drug which were responsible for the antitumour activities of A lancea, we intend to purify one water-soluble polysaccharide from this plant and examine its antitumour and anti-metastatic effects on human osteosarcoma cell line U-2 OS. Furthermore, the effect of this polysaccharide on the adhesion of sLe x on U-2 OS cells to E-selectin on activated human umbilical vein endothelial cells (HUVECs) was also evaluated in U-2 OS cells.

| Materials and chemicals
Agrimonia pilosa was purchased from the local Drug store in were purchased from Gibco/Invitrogen (Grand Island, NY). All other chemical reagents used in this experiment were of analytical grade.

| Extraction and purification of polysaccharide from rhizomes of A lancea
The dried rhizomes of A lancea (1000 g) were first ground into powder in a disintegrator and then refluxed with 95% ethanol to get rid of lipophilic constituents, such as pigments, monosaccharides, oligosaccharides and other small molecule impurities. Subsequently, the refluxed residues were filtrated and washed until without phenolsulphuric acid reaction and then were immersed in 10 L of 2% NaOH aqueous solution overnight at room temperature for three times.
After the extract solution was combined and filtered through line cloth, the resulting suspension was added with hydrochloric acid The CALP (5 g) was dissolved in deionized water (10 mL) and filtered through 0.22 μm membrane. The supernatant was loaded to a DEAE Sepharose Fast Flow column (2.0 × 50 cm) and eluted with distilled water followed by 0.25, 0.5 and 1 M NaCl aqueous solutions respectively at a flow rate of 2.0 mL/min. Each fraction (10 mL/ tube) was collected by monitoring the carbohydrates content by phenol-sulphuric acid method at 490 nm. 18 Only one fraction (CALP) was obtained from the distilled water eluate. Then, this water-eluted fraction was further purified by size-exclusion chromatography on a Sephacryl S-300 gel column (1.6 × 100 cm) eluting with 0.15 M NaCl at 1 mL/min, yielding a purified polysaccharide (ALP, 116.5 mg).

| Molecular weight and chemical compositions analysis of polysaccharide from A lancea
The total carbohydrate content of polysaccharide was determined using the phenol-sulphuric method using glucose as the standard. 18 The protein content of polysaccharide was measured by Bradford assay using BSA as the standard. 19 The uronic acid content of the polysaccharide was analysed by the m-hydroxydiphenyl method with d-glucuronic acid as standard. 20 The homogeneity and molecular weight of the polysaccharide were determined by high performance gel permeation chromatography (HPGPC) on a Waters 1525 HPLC system (Waters) equipped with a Waters 2414 refractive index detector (RID) and a TSK-GEL G4000PWxlcolumn (7.8 × 300 mm × 5 μm). The operation conditions were set as follows: injection volume: 10 μL; mobile phase: 0.1 mol/L NaNO 3 ; flow rate: 0.9 mL/min; column temperature: 35°C; detecting temperature: 45°C. The elution time (T) of T-series Dextran standards (2000, 500, 70, 10, 5 kD) was plotted against their respective molecular weights to generate a regression equation under the conditions described above.
The monosaccharide composition of the polysaccharide was determined by GC method as recorded by Qin et al, 16 with some minor revision. Briefly, 5 mg of dried samples was hydrolysed with 2 mL of 2 M TFA at 120°C for 2 hours to release monosaccharide.
After the removal of the excess acid, the hydrolysates were then acetylated with 1 mL of pyridine and 1 mL of acetic anhydride in a water bath at 90°C for 20 minutes. The affording alditol acetate was further analysed on an Agilent7890A system gas chromatography fitted with a DB-5 MS capillary column capillary column (30 m × 0.25 mm × 0.25 μm) and a flame ionization detector (FID).
The reference of standard monosaccharide (Glc, Gal, Rib, Rha, Fuc, Man, Ara and Xyl) was derivatized under the same conditions and monosaccharide composition of sample was identified by matching with the retention times of the monosaccharide standards with inositol as the internal standard. The column temperature was initially programmed at 100-220°C at a rate of 5°C/min, increased to 240°C at a rate of 2°C/min, and then held at 240°C for 2 minutes, followed by gradient increment to 280°C at 10°C/min. The N 2 was used as the carrier gas at a constant flow rate of 20 mL/min.

| Detection of sLex by fluorescence microscopy
The presence of sLe x on the surface of cancer cells was determined using monoclonal antibodies specific for its epitope under fluorescence micrcscope examination according to the previous method. 7 At first, the 6-well glass slide was added with cancer cells (100 μL,

| Cell adhesion experiments under static conditions
Adhesion assay was performed as described by Yue et al, 21 with slight modification. Briefly, HUVECS (100 μL, 2 × 10 5 cells/well) in exponential phase of growth was incubated in 96-well plates with 5% CO 2 at 37°C. When the cell monolayer was formed, 5 ng/ mL TNF-α was added to stimulate the E-selectin production. After

| Cell invasion and migration assays
The cell migration and invasion assay was performed as described For migration assay, all experimental operations were conducted in the same condition except that no HUVECs coating on polycarbonate filters.

| Detection of sLex by flow cytometry analysis
To qualify the expression change of sLe x on the surface of U-2 OS cells by ALP, flow cytometry analysis was performed as previously described. 23 Briefly, U-2 OS cells were allowed to grow until logarithmic phase and adhering to bottom of 6-well culture plates and

| Annexin V-FITC/PI double-staining analysis
To qualify the apoptosis rate, apoptotic cells stained with Annexin V-FITC/PI were subjected to flow cytometry analysis according the manufacture's protocols. 24 Briefly, following different ALP treatment, cancer cells were collected, rinsed twice with PBS and

| Statistical analysis
The data are presented as means ± SD The differences between the groups were analysed with analysis of variance (ANOVA). P-values of <0.05 were accepted to be statistically significant.

| Isolation, purification and characteristic of different polysaccharide from A lancea
The crude polysaccharides CALAP were extracted from the defat-

| Expression of sLe x antigen on the surface of U-2 OS cells
As shown in Figure 3, only sLe x was obviously seen on U-2 OS cells as compared with the control sample without the addition of the antibody, while no sLe a was expressed. This observing indicated that sLe x antigen was present on the surface of U-2 OS cells.

| ALP from A lancea inhibits adhesion of U-2 OS cells to HUVECs
As shown in Figure 4, the maximum adhesion effect was observed in cell upon stimulating HUVECs with TNF-a (5 ng/mL) and this

| ALP from A lancea suppresses cell invasion and migration of U-2 OS cells in vitro
As seen from Figure 5A, a significant loss of cells number invading to the lower surface of HUVECs membrane, as evidenced by the attenuation of crystal violet intensity, was observed in cells treated by ALP, which was present in a dose-dependent manner. The same result occurred in the migration assay for ALP ( Figure 5B). Both results suggested that ALP was able to effectively inhibit the invasion and migration of U-2 OS cells in vitro.

| ALP from A lancea inhibits sLe x expression of U-2 OS cells
The result from flow cytometric analysis ( Figure 6A,B) showed that untreated U-2 OS cells had more sLe x expression, whereas pretreatment with ALP (0.5, 1 and 2 mg/mL) resulted in a decreased number of the cells expressing sLe x in a dose-dependent manner, especially at the concentrations of 1 and 2 mg/mL when compared with control cells (P < 0.05).The present study verified that the cell adhesion inhibitory effect of ALP on U-2 OS cells to HUVECs was achieved by mitigating the sLe x expression on U-2 OS cells.

F I G U R E 3
The expression pattern of sLe x /sLe a antigens on U-2 OS cells F I G U R E 4 The effect of ALP from Atractylodes lancea on the TNF-α-induced adhesion of U-2 OS to HUVECs. All data are represented as the means ± SD (n = 3). *P < 0.05, **P < 0.01 and ***P < 0.001 vs control

| ALP from A lancea induces apoptosis of U-2 OS cells in vitro
TUNEL staining assay ( Figure 7A) indicated that the addition of ALP led to a significant increase of apoptotic cells than that of the control (P < 0.01), as showed by the increased number of purplish redstained TUNEL-positive nuclei. As illustrated in Figure 7B, the result of Annexin V-FITC/PI double staining on flow cytometry further confirmed that ALP was found to induce apoptosis in U-2 OS cells in a concentration-dependent manner.

| D ISCUSS I ON
Cancer metastasis, as one of the major causes of cancer-associated mortality, involves a multiple of cell movements, that is cancer cell

| CON CLUS ION
The present experimental results revealed that the adhesion, migration and invasion of U-2 OS cells to HUVECs were significantly inhibited by ALP. It is possibly inferred that ALP is prone to bind E-selectin instead of sLe x , thus hindering the combination of E-selectin and sLe x to promote tumour metastasis. In addition, U-2 OS cells underwent the apoptosis following ALP treatment.

F I G U R E 7
The effect of ALP from Atractylodes lancea on the apoptosis of U-2 OS cells. A, Intracellular apoptotic nuclei was detected by TUNEL assay in U-2 OS cells. B, Cell apoptosis was detected by Annexin V-FITC/PI dual-labelling technique by flow cytometry in U-2 OS cells