**3. Discussion**

In this study, we observed that PSS has a major impact on invasion and angiogenesis in murine melanoma B16-F10 cells. FGF2 is a proangiogenic factor involved in tumor angiogenesis, invasion and migration. The present findings show that FGF2 bound to PSS with high affinity and inhibited FGF2-mediated angiogenesis in a rat aortic ring model. Moreover, PSS could suppress a FGF2-mediated invasion. The results of a further mechanism study indicate that PSS down-regulated the expression of activated MMP-2 and MMP-9, and also suppressed their activity. In addition, PSS was found to decrease the protein levels of Vimentin, which is known to participate in EMT. Notably, PSS did not elicit any changes in cancer cell viability, even though the concentration was more than 1000 μg/mL.

Previous data indicated that the KD value for heparin binding to FGF2 ranges from 1 to 71 nM [19–22]. These values were affected by a variety of factors, including the method used to determine them, the ionic strength of the buffer, the size and source of the heparin, and the source of the growth factor. In our system, we obtained similar KD values for FGF2 binding to PSS or heparin, indicating that PSS was comparable to heparin in terms of binding to FGF2. We further analyzed the electrostatic potential surface of FGF2. As shown in Supplementary Figure S2, FGF2 displayed a mass area of positive charge in the surface; hence, it could easily bind negative charged compounds. PSS and heparin possess similar sulfate group contents (32.39% and 34%, respectively) and similar KD values to FGF2; therefore, we presume that the sulfate of the two polysaccharides probably accounted for their affinity to FGF2.

FGF2 is known to interact with *N*-sulfoglucosamine (GlcNS) and 2-*O*-sulfated iduronate residues (IdoUA (2S)) in heparin and HS [23,24], but the additional presence of 6-*O*-sulfation is required for biological activity [25,26]. PSS is a heparin-like drug, which is composed of repeating units of mannuronic acid (M) and guluronic acid (G), with 2-*O* and 3-*O* sulfate groups in the sugar rings. Groups that are 2-*O*-sulfated play a crucial role in mediating the binding of heparin with FGF2. Because PSS and heparin exhibited a comparable affinity to FGF2 and, we presume that 2-*O*-sulfated PSS might also be crucial for promoting the interaction of PSS with FGF2. To confirm this presumption, further research should be performed to elucidate the structure–activity relationship.

FGF2 and VEGF165 are the most important growth factors, and can be blocked by heparin to reduce angiogenesis. We also detected the a ffinity of PSS with VEGF165; however, PSS exhibited weaker a ffinity to VEGF165 (KD = 1.78 × 10−<sup>4</sup> M) than heparin (KD = 8.09 × 10−<sup>7</sup> M). Zhao et al. [27] reported that the specific structural features of heparin, such as the content of sulfate, sugar ring stereochemistry, and conformation, determined the a ffinity of heparin-derived oligosaccharides to VEGF165. Moreover, the positive charge on the surface of VEGF165 was distributed in a dispersed state (Supplementary Figure S2). Based on the above information, we presume that the conformation of PSS might not fit the stereochemical structure of VEGF165.

It was well-documented that the molecular weight of sulfated polysaccharides including PSS played crucial roles in determining their bioactivities. Our previous study showed that the average molecular weight of PSS was about 17 kDa and the distribution range of molecular weight was about 2~20 kDa [17]. Unlike the small molecular compounds, which generally bind to domains with catalytic activity of targeting proteins, sulfated polysaccharides possessed large amount of negative charge and generally interacted with proteins rich in positive potential on the surface of proteins. Theoretically, the longer sugar chains (the higher molecular weight) of the negative charged polysaccharide was endowed with the stronger binding a ffinity to target proteins and further exerted obvious bioactivities. Here, the higher molecular weight of PSS fractions exerted stronger inhibitory e ffect on angiogenesis which was consistent with the trend in previous publications [17,28]. Similarly, we speculated that it might be the same trends of PSS fractions in other experiments of the manuscript.

For the first time, we evaluated the e ffect of PSS on the highly metastatic B16-F10 melanoma cells and the related tumor environment. PSS itself has no inhibitory e ffect on the growth of B16-F10 cells—however, it suppressed FGF2-mediated angiogenesis and invasion of B16-F10 cells, and also decreased the level of Vimentin, which might help enhance the sensitivity of tumor cells to chemotherapy. Moreover, to fully elucidate the e ffects of PSS on the tumor microenvironment, further research should be conducted to investigate whether PSS exerts inhibitory e ffects on other cells involved in the tumor microenvironment, such as endothelial cells, fibroblasts, and immune cells. Meanwhile, further research should be done to combine PSS with chemotherapeutic drugs to check whether a synergistical effect happens.
