**2. Results**

### *2.1. Fibroblast Growth Factor 2 Bound to Propylene Glycol Alginate Sodium Sulfate with High A*ffi*nity*

PSS was investigated by surface plasmon resonance (SPR) analysis for its a ffinity with FGF2 and vascular endothelial growth factor 165 (VEGF165). As shown in Figure 1, PSS bound directly to FGF2, and the equilibrium dissociation constant (KD) was 2.73 × 10−<sup>8</sup> M, which is comparable to heparin (2.76 × 10−<sup>8</sup> M). In contrast, weak a ffinity was detected between PSS and VEGF165 (Supplementary Table S1), while the KD of heparin was 8.09 × 10−<sup>7</sup> M. As FGF2 is a crucial growth factor to regulate angiogenesis and the function of tumor cells, these data indicate that FGF2 was probably a potential target for PSS to improve tumor environment.

**Figure 1.** Analysis of the affinity between propylene glycol alginate sodium sulfate (PSS) and fibroblast growth factor 2 (FGF2). (**A**,**B**) were binding response curves of PSS and heparin with FGF2, respectively. PSS or Heparin (1–5 mM) was immobilized on Graft-to-PCL sensor chips. The mobile phase was FGF2 solution (dissolved in phosphate buffer solution (PBS)), and the concentrations were 125, 250, and 500 nM. The data obtained were analyzed and fitted by a PLEXERA SPR Data Analysis Module (DAM) to obtain the equilibrium dissociation constant (KD). Images are representative of three independent experiments with similar results.

### *2.2. Propylene Glycol Alginate Sodium Sulfate Inhibited the Fibroblast Growth Factor 2-Induced Invasion of B16-F10 Cells*

FGF2 has been reported to play an important role in tumor metastasis; thus, we next wanted to evaluate the role of PSS in inhibiting tumor cell invasion and metastasis. PSS significantly inhibited the serum-induced invasion of B16-F10 in a dose-dependent manner. The Matrigel barrier approach was used to evaluate tumor cell metastasis. As shown in Figure 2A,B, after treating the cells for 16 h, PSS inhibited invasion by 33.8%, 45.7%, and 61.8% at concentrations of 25, 50, and 100 μg/mL, respectively, in a dose-dependent manner. We further detected the anti-invasive effect of PSS on FGF2-mediated invasion. As shown in Figure 2C,D, 10% FBS and 200 ng/mL FGF2 induced a comparable number of B16-F10 cells to penetrate the growth factor-reduced Matrigel, and PSS significantly inhibited FGF2-induced invasion by 59.1% at a concentration of 50 μg/mL after treating the cells for 8 h.

### *2.3. Propylene Glycol Alginate Sodium Sulfate Had No E*ff*ect on Cell Viability*

To ensure that the inhibitory effect of PSS on invasion was not due to the direct killing of tumor cells by PSS, we detected the proliferation ability of tumor cells after treatment with PSS for 48 h. No inhibition on proliferation was observed, even when the concentration was more than 1000 μg/mL (Figure 3), suggesting that the inhibitory effect of PSS on invasion was not due to a reduction in the viability of cancer cells, but was likely due to the inhibition of tumor cell migration or the suppression of matrix-degrading enzymes.

**Figure 2.** The effect of PSS on the invasion of B16-F10 cells. B16-F10 cells (1.5 × 10<sup>4</sup> cells/well) were seeded onto a membrane coated with Matrigel (**A**,**B**) or growth factor-reduced Matrigel (**C**,**D**), and were treated with various concentrations of PSS (25, 50, 100 μg/mL) for 16 h or 8 h (FGF2-mediated invasion of B16-F10). Cells that penetrated through to the lower surface of the membrane were stained with crystal violet and photographed under a light microscope at 40× magnification. Then, crystal violet was dissolved in 10% acetic acid, and the absorbance of the resulting solution was measured at 600 nm using a microplate reader (SpetraMAX i3, Molecular Devices, Sunnyvale, CA, USA). The controls included a negative control, complete medium only; control (with 10% FBS) complete medium with 10% FBS; control (with FGF2) complete medium with 200 ng/mL FGF2; and FGF2 + PSS 50 μg/mL complete medium with 200 ng/mL FGF2 and 50 μg/mL PSS. The results are from three independent experiments. \*\* *p* < 0.01, significant difference between PSS-treated groups and the untreated control by Student's *t*-test.

**Figure 3.** The effect of PSS on cell viability. B16-F10 cells (0.5 × 10<sup>4</sup> cells/well) were seeded in 96-well plates and incubated for 24 h to allow adherence. Various concentrations of PSS were added to the plates, and then the cells were further incubated for another 48 h. The resazurin assay (1 mg/mL) was used for detection. The absorbance of each well was measured at 405 nm by a microplate reader (SpetraMAX i3, Molecular Devices, Sunnyvale, CA, USA). The data obtained are from three independent experiments.

### *2.4. Propylene Glycol Alginate Sodium Sulfate Had No E*ff*ect on the Migration of B16-F10 Cells*

Based on the results above, we further detected the effect of PSS on the migration of B16-F10 cells. As shown in Figure 4A,B, no obvious difference in the number of the migratory cells was found between the untreated control and the treatment with PSS at 400 μg/mL. Meanwhile, the results of the scratch wound migration assays also revealed no significant difference across the wounded region after treatment with PSS at 400 μg/mL for 24 h (Figure 4C). PSS also had no effect on the migration of human umbilical vein endothelial (HUVEC) cells (Supplementary Figure S1).

**Figure 4.** The effect of PSS on the migration of B16-F10 cells. B16-F10 cells (1 × 10<sup>4</sup> cells/well) were seeded into the upper chamber of Transwell plates, treated with various concentrations of PSS, and allowed to migrate for 16 h. Cells that penetrated through to the lower surface of the membrane were stained with crystal violet and photographed under a light microscope at 40× magnification (**A**). Then, crystal violet was dissolved in 10% acetic acid, and the absorbance of the resulting solution was measured at 600 nm using a microplate reader (SpetraMAX i3, Molecular Devices, Sunnyvale, CA, USA) (**B**). For the wound healing assay (**C**), 70% confluency B16-F10 cells in 12-well culture plates were scratched and then treated with FBS-free medium with PSS (100, 400 μg/mL) or 400 μg/mL heparin. Three randomly selected views along the wound line in each well were photographed under an inverted microscope at 0 h and 24 h after incubation. The percentage of void area with respect to time 0 was determined using ImageJ software (ImageJ 1.8.0, Rawak Software Inc., Stuttgart, Germany). The results are from three independent experiments. \*\* *p* < 0.01, significant difference between the PSS-treated groups and the untreated control by Student's *t*-test.

### *2.5. Propylene Glycol Alginate Sodium Sulfate Down-Regulated the Expression of Activated Matrix Metalloproteinase 2 and Matrix Metalloproteinase 9*

The process of invasion involves the degradation of the ECM and the subsequent migration of tumor cells. Based on the data above, we confirmed that PSS inhibited the invasion of B16-F10 cells in a dose-dependent manner, but had no effect on tumor cell migration. Therefore, we examined whether PSS affected the expression or activity of crucial ECM-degrading enzymes. MMP-2 and MMP-9 are capable of degrading type IV collagen, which is the most abundant component of the basement membrane. Degradation of the basement membrane is an essential step for the metastatic progression of most cancers. Matrigel is an analog of the basement membrane. PSS inhibits the degradation of Matrigel by B16-F10. Therefore, it is necessary to detect whether PSS inhibits the expression and activity of MMP-2 and MMP-9. We first detected the effect of PSS on the expression of MMP-2 and MMP-9. After treating B16-F10 cells for 24 h, PSS was found to decrease the expression of activated MMP-2 in a dose-dependent manner. PSS inhibited the level of MMP-2 by 1%, 15%, 35%, and 67% compared to the untreated control at concentrations of 12.5, 25, 50, and 100 μg/mL (Figure 5A,B), respectively. PSS also exerted an inhibitory effect on the expression of MMP-9. As shown in Figure 5C,D, PSS at a concentration of 100 μg/mL suppressed the expression of MMP-9 by 32%.

**Figure 5.** PSS down-regulated the level of activated matrix metalloproteinase 2 (MMP-2) and matrix metalloproteinase 9 (MMP-9) proteins in B16-F10 cells. Cells (2 × 10<sup>6</sup> cells/dish) were treated with PSS (12.5, 25, 50, 100 μg/mL) for 24 h. Then the cells were harvested, total protein was determined, and SDS-PAGE was performed, as described in the Materials and Methods section. The levels of activated MMP-2 (**A**,**B**) and MMP-9 (**C**,**D**) were estimated by Western blotting, also as described in the Materials and Methods section. The numbers underneath the blots represent the band intensities (normalized to the loading controls, means of three independent experiments) measured by ImageJ software (ImageJ 1.8.0, Rawak Software Inc., Stuttgart, Germany). The standard deviations were all within ±15% of the means (data not shown); β-actin was used as an equal loading control. The experiments were repeated three times.

### *2.6. Propylene Glycol Alginate Sodium Sulfate Decreased the Activity of Matrix Metalloproteinase 2 and Matrix Metalloproteinase 9*

We further detected the activity of MMP-2 and MMP-9 in B16-F10 cells after treatment with PSS for 24 h. As shown in Figure 6B,C, PSS inhibited the activity of MMP-2 in a dose-dependent manner. At 25 μg/mL, PSS slightly inhibited MMP-2, while at 50 and 100 μg/mL, it suppressed the activity by 21.8% and 28.7%, respectively. PSS also suppressed the activity of MMP-9 in a similar manner. The inhibition of PSS was 2.7%, 17.2%, and 25.1% at concentrations of 25, 50, and 100 μg/mL, respectively (Figure 6E,F).

**Figure 6.** The effect of PSS on the activity of MMP-2 and MMP-9 in B16-F10 cells. The activity of MMP-2 (**A**–**C**) and MMP-9 (**D**–**F**) was tested using the GENMED Kit (Genmed Scientifics Inc., Wilmington, DE, USA). Cells (1 × 10<sup>5</sup> cells/well) were seeded in 12-well culture plates with a glass slide to reach 50% confluence. After being washed with PBS, fresh serum-free culture medium was added to the plate in the presence or absence of serial treatments of PSS (25, 50, 100 μg/mL) for 24 h. The GENMED kit was used to detect the fluorescence intensity. The green dots represent the catalytic sites for MMP-2 or MMP-9, and the fluorescence intensity represents the amount of the degraded gelatin. Fluorescence was visualized in 10 randomly selected fields of view for each cell slide under a fluorescence microscope (Colibri 7, ZEISS, Jena, Germany), and the fluorescence intensity was quantified by ZEN 2.3 lite software (ZEISS, Jena, Germany). The data shown above are from three independent experiments. \* *p* < 0.05, \*\* *p* < 0.01, significant difference between the PSS-treated groups and the untreated control by Student's *t*-test.

### *2.7. Propylene Glycol Alginate Sodium Sulfate Down-Regulated the Protein Expression of Vimentin*

Carcinoma cells make use of the epithelial–mesenchymal transition (EMT) as they become invasive. Indeed, the transition typically features the loss of cell–cell adherence proteins like cadherin, followed by the loss of apico-basal polarity, and finally, gaining the ability to migrate and invade. To investigate the potential role of PSS in the EMT, two crucial EMT markers in B16-F10 cells were examined by western blot. The expression of Vimentin was decreased by 41% and 47% in the presence of PSS at concentrations of 50 and 100 μg/mL, respectively (Figure 7), while a slight increase was detected in the level of E-cadherin. Additionally, other proteins involved in the regulation of invasion and migration did not show any change in expression after treatment with PSS.

**Figure 7.** PSS down-regulated the protein expression of Vimentin in B16-F10 cells. Cells (2 × 10<sup>6</sup> cells/dish) were treated with PSS (25, 50, 100 μg/mL) for 24 h. The cells were harvested, total protein was determined, and SDS-PAGE was performed as described in the Materials and Methods section. The levels of ERK 1/2, p-ERK 1/2, AKT, p-AKT, p38, p-p38, NF-κB, p-NF-κB, Vimentin, and E-cadherin were estimated by Western blotting, as described in the Materials and Methods section. (**A**,**B**) The protein expression level of Vimentin and E-cadherin in B16-F10 cells. (**C**) The levels of other proteins mentioned above in B16-F10 cells. The numbers underneath the blots represent the band intensities (normalized to the loading controls, means of three independent experiments) measured by ImageJ software (ImageJ 1.8.0, Rawak Software Inc., Stuttgart, Germany). The standard deviations were all within ±15% of the means (data not shown); β-actin was used as an equal loading control. The experiments were repeated three times.

### *2.8. Propylene Glycol Alginate Sodium Sulfate Inhibited Angiogenesis*

We next assessed the capacity of PSS to inhibit angiogenic activity using rat aortic rings and chick chorioallantoic membrane models. As shown in Figure 8A,B, PSS inhibited the outgrowth of new microvessels in a dose-dependent manner. The molecular weight of sulfated polysaccharides including PSS played crucial roles to determine their bioactivities. To illuminate the interaction between molecular weight and anti-angiogenesis potency of PSS, we further detected the effects of PSS fractions with various molecular weights (Mw) (H1: Mw 21.91 kDa; H3: Mw 13.68 kDa; H5: Mw 6.56 kDa; H7: Mw 3.10 kDa; H8: Mw 2.26 kDa) [17] on new blood vessel formation, using the same model. The data showed that the inhibitory effect of the fractions closely correlated with the molecular weight. The H1 fraction (Mw = 21.91 kDa) exhibited the most potent effect among all five fractions, and the inhibitory effect was reduced as the molecular weight decreased (Figure 8C,D). Meanwhile, in the chick chorioallantoic membrane model, PSS suppressed vessel formation at doses of 100 and 200 μg/egg (Figure 8E,F).

*Mar. Drugs* **2019**, *17*, 257

**Figure 8.** *Cont*.

**Figure 8.** PSS suppresses angiogenesis. The inhibitory effect of PSS on microvessel outgrowth arising from rat aortic rings. Aortic rings were embedded in Matrigel in 96-well plates and then cultured with medium containing various concentrations of PSS (50, 100, 150 μg/mL) for seven days (**A**,**B**). The effect of fractionated PSS on angiogenesis in the rat aortic ring model. The method used was the same as above (**C**,**D**). The effect of PSS on angiogenesis in the chick chorioallantoic membrane assay. Gelatin sponges (5 mm × 5 mm) saturated with various concentrations of PSS solution or normal saline were inserted into seven-day-old fertilized eggs (**E**,**F**). After incubation for another 48 h, the zones of neovascularization under and around the gelatin sponge were photographed under an anatomic microscope (Colibri 7, ZEISS, Jena, Germany) and the black arrows in the photos pointed at the site of gelatin sponge. The total vessel number was quantified by ImageJ software (ImageJ 1.8.0, Rawak Software Inc., Stuttgart, Germany). \* *p* < 0.05, \*\* *p* < 0.01, significant difference between the PSS or PSS fraction-treated groups and the untreated control by the Student's *t*-test.

### *2.9. Propylene Glycol Alginate Sodium Sulfate Inhibited Fibroblast Growth Factor 2-Mediated Angiogenesis*

To confirm the possible effect of PSS on FGF2-mediated angiogenesis, we cultured rat aortic rings in a serum-free medium, with 200 ng/mL FGF2 in the plate coating with growth factor-reduced Matrigel. As shown in Figure 9, little sprouting was observed in the negative control ring, while substantially more sprouting was observed in the FGF2-treated ring. PSS at 100 μg/mL decreased the level of sprouting to the amount in the negative control, and 200 μg/mL completely suppressed sprouting.

**Figure 9.** PSS suppresses FGF2-induced microvessel outgrowth in a rat aortic ring model. (**A**) Aortic rings were embedded in growth factor-reduced Matrigel in 96-well plates, and were then cultured with medium containing 200 ng/mL FGF2 and various concentrations of PSS (100, 200 μg/mL) for seven days. Images are representative of three independent experiments with similar results. (**B**) The total vessel number was quantified by ImageJ software (ImageJ 1.8.0, Rawak Software Inc., Stuttgart, Germany). \*\* *p* < 0.01, significant difference between the PSS-treated groups and the FGF2 control by Student's *t*-test.
