Danshensu (PubChem CID: 439435) was purchased from Helin Biological Engineering Co., Ltd. (Xi’an, China). Oxaliplatin (OXA) were purchased from Aladdin Biological Technology Co., Ltd. (Shanghai, China).

SW620 human colon cancer cell line was purchased from Jiangyin Cambridge Biotechnology Co., Ltd. (Nanjing, China), and maintained in Dulbecco’s modified eagle medium (DMEM) (Invitrogen, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS) (Gibco, Grand Island, NY, USA).

Cell viability was determined by the MTT (3- [4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide) assays according to the manufacturer’s protocol (n = 6). In short, SW620 cells (1 $\times$ 10${}^4$ cells/well) were seeded in a 96-well plate. Different concentrations of OXA were added to treat the SW620 cells for 24 h. The absorbance value was measured by a microplate reader (BioTek Instruments, Inc., Winooski, USA).

Platelet-rich plasma for research use only was provided by Jiangsu Province Blood Center (Nanjing, China). Platelets were prepared as described previously [15]. The platelets (2 $\times$ 10${}^8$/mL) were then exposed to colon cancer cells (1 $\times$ 10${}^6$/mL) or thrombin (0.5 U/mL) for 15 min following treatment with Danshensu or solvent (basal medium) for 30 min at 37 °C. The supernatant was filtered using a sterile 0.22-micron filter. Before incubation with the SW620 cancer cells, 10% FBS was added into the supernatants. The supernatants from the co-cultured platelets and colon cancer cells with direct contact (SCP) and the supernatant of platelets activated by thrombin (STP) were collected according to the protocol as described previously and stored at –80 °C freezer (Supplementary Fig. 1). This study was approved by the Institutional Ethics Committee of Nanjing University of Chinese Medicine and conducted in accordance with the principles of the Helsinki Declaration. All patients have provided written informed consent before sample collection and subsequent analysis.

SW620 cells were incubated with platelets, STP or SCP collected by centrifugation and filtration. The cells were maintained in the supernatants and observed every 1 h by IncuCyte ZOOM™ Live Content Imaging System (Essen BioScience Inc. (Essen) IncuCyte™, Michigan City, Indiana, USA).

The levels of VEGF, IL-6, IL-1$\beta$ and TNF-$\alpha$ in the supernatants were measured by ELISA kits (Peprotech, Rocky Hill, NJ, USA) according to the manufacturer’s instructions (n = 3).

ATP levels in the supernatants were assessed using an ATP assay kit (Beyotime Biotechnology, China) (n = 3). In brief, the platelets (2 $\times$ 10${}^8$/mL) were exposed to colon cancer cells (1 $\times$ 10${}^6$/mL) or thrombin (0.5 U/mL) for 15 min after treatment with Danshensu or solvent (basal medium) for 30 min at 37 °C. The supernatants were filtered using a sterile 0.22-micron filter after centrifugation. ATP detection working solution was added to a black 96-well culture plate and was incubated for 3 minutes at room temperature. Then, the supernatants were added to the wells, and the luminescence was measured immediately. The concentration of ATP in the sample was calculated based on the standard curve.

24-well Transwell migration chambers with 8 $\mu$m pore size were purchased from Corning incorporated (Kennebunk, ME, USA). Briefly, the treated cells were re-suspended in 200 $\mu$L serum-free culture medium (1 $\times$ 10${}^6$/mL). Then, 800 $\mu$L of complete medium containing 10% FBS was supplemented to the lower chamber. After 24 h, the non-migrated and non-invasion cells were removed by a wet cotton swab. The migration and invasion of the SW620 cells were assessed using 0.1% crystal violet. The SW620 cells were counted in five randomly selected fields, and images were obtained under a microscope (200$\times$) (Leica Microsystems, Watzlar, Germany). The number of cells was calculated through the membrane, and then the migration scores of SW620 cells were examined (n = 3).

The SW620 cells were cultured in DMEM with 10% FBS in the 6-well plates, and 500 SW620 cells were seeded into each well of the plates (n = 3). The plates were incubated at 37 °C for 14 days, and the colony formation of the cells was assessed using 0.1% crystal violet, after which the formed colonies were photographed (Leica Microsystems, Watzlar, Germany).

Total protein was extracted as previously reported [16]. Briefly, the total protein was extracted using RIPA buffer containing 1 mM PMSF and 1:100 dilution of protease inhibitor cocktail (BestBio, Shanghai, China). The total protein was transferred to a polyvinylidene difluoride (PVDF) membrane (Millipore, USA) that was then incubated with mouse anti-human E-cadherin, N-cadherin, vimentin, TGF-$\beta$, Smad2, phospho-Smad2, Smad3, phospho-Smad3, multidrug resistance 1 (MDR1) and GADPH polyclonal antibodies (Cell Signaling Technology, 1:1000), followed by incubated with rabbit or mouse secondary monoclonal antibody (Cell Signaling Technology, 1:1000) (n = 3). Finally, the protein bands were visualized with Bio-Rad Gel Doc XR+ gel imaging system (Bio-Rad, USA).

All tissue specimens were embedded in paraffin according to standard histological procedures, sectioned (4 $\mu$m), and stained with hematoxylin and eosin (H&E) or MDR1 antibody (Cell Signaling Technology, 1:500) as described previously [16]. Then, the sections were examined and quantitatively analyzed using PerkinElmer MANTRA microscope (Mantra, PerkinElmer, USA). The mean DAB staining intensity was quantified by Mantra quantitative pathology workstation (Mantra, PerkinElmer, USA) (n = 4).

The whole blood sample was placed at room temperature for 2 h, centrifuged at 2500 R/min for 10 min under the condition of 4 °C, and the upper serum was separated. The full-automatic biochemical analyzer (Hitachi 7020) was used. After debugging, according to the glutamic oxaloacetic transaminase determination kit and alanine aminotransferase assay kit (Wako Pure Chemical Industries, Ltd, R1: el157; R2: el158; R1: ej715; R2: eh029) (n = 4).

All procedures were approved by the Institutional Animal Care and Use Committee of Nanjing University of Chinese Medicine, and performed in accordance with the use of laboratory animal guidelines of Nanjing University of Chinese Medicine (Nanjing, China, Approval No. ACU170905). The SW620 cells were pre-incubated in the SCP or the supernatants collected from co-cultured tumor cells and platelets that treated with 25 $\mu$M Danshensu for 24 h, after which the residues were removed by centrifugation. Six-week-old male BALB/c nu/nu mice were supplied by Model Animal Research Center of Nanjing University (Nanjing, China) and housed under specific pathogen-free (SPF) conditions. Subsequently, 2 $\times$ 10${}^6$ SW620 cells suspended in a mixture of 100 $\mu$L PBS and 100 $\mu$L Matrigel (BD Bioscience, USA) were subcutaneously injected into each nude mouse. Control animals were subcutaneously injected with untreated SW620 cells. One week later. The mice were intraperitoneally administered with 3 mg/kg OXA on a weekly basis for 4 weeks. The length and width of tumors were measured using a digital caliper every 4 days for 28 days following tumor cell injection. The mice were sacrificed on day 28 after subcutaneous injection of SW620 cells, and the tumor and liver tissues were fixed in formalin and prepared for histopathological examination.

The data were represented as mean $\pm$ standard deviation (SD) and statistically analyzed using one-way ANOVA or Student’s T-test. All results were analyzed by SPSS 22.0 software (SPSS Inc., Chicago, USA). p 0.05 was considered statistically significant.

It has been well accepted that platelets can induce EMT-like transition and promote metastasis in vitro [17]. We observed the morphological changes of SW620 cells by IncuCyte Zoom, a real-time dynamic imaging system. SW620 cells underwent EMT-like morphological changes following the treatment of SCP, STP or platelets for 24 h (Fig. 1A). SW620 cells incubated with SCP had pronounced EMT-like phenotypic changes compared with those incubated with STP. Further, the expression of E-cadherin was down-regulated while the levels of N-cadherin and Vimentin in the SW620 cells were significantly up-regulated following the treatment of SCP for 24 h compare with control (Fig. 1B,C). These data indicated that SW620 exerted significant effects on activating secretion ability of platelets and platelet secretion stimulated the formation of an EMT-like invasive phenotype.

Fig. 1.

SW620 cancer cells treated with SCP induced an EMT-Like invasive phenotype. (A) Morphological changes of SW620 cancer cells incubated with platelets, STP or SCP for 48 h. Scale bar indicates 200 $\mu$m. (B,C) Western blot bands of E-cadherin, N-cadherin and Vimentin in the SW620 cells treated with platelet secretions for 24 h (n = 3). The relative expression of E-cadherin, N-cadherin and Vimentin was normalized to GAPDH. Data were represented as mean $\pm$ SD. * p 0.05, ** p 0.01 versus control.

Danshensu was reported to suppress platelet adhesion and aggregation by inhibiting cyclooxygenase-2 (COX-2) and thromboxane B2 [10] (Fig. 2A). We first explored the effects of Danshensu on the proliferation and membrane permeability of SW620 cancer cells by MTT and LDH release assays (Supplementary Fig. 2 and Fig. 2B). Strikingly, 25 $\mu$M Danshensu attenuated the ATP secretion from platelets induced by SW620 colon cancer cells, whereas 25 $\mu$M Danshensu did not have obvious impacts on the cytotoxicity of SW620 cells (Fig. 2B,C, Supplementary Fig. 2). Subsequently, we observed the morphological changes of SW620 cells after incubation with SCP and Danshensu using IncuCyte Zoom, and detected the expression levels of EMT-related proteins by Western blot assay (Fig. 2D). After SW620 cells were stimulated with SCP, the invasive ability of the cells, as well as the expression levels of N-cadherin and Vimentin were all increased (Fig. 3A,B). However, the levels of E-cadherin were up-regulated while the expression of N-cadherin and vimentin were significantly down-regulated in Danshensu treated group compared with those in the SCP group (Fig. 3C,D). Taken together, the EMT phenotype was reversed in the presence of Danshensu. We next explored the effect of Danshensu on cell migration using Transwell assay in the SW620 cells. The results showed that the platelets treated with Danshensu was able to attenuate the migration of SW620 cells induced by SCP (p 0.001, Fig. 3A,B).

Fig. 2.

Danshensu interfered with the interactions between platelets and colon cancer cells. (A) The structure of Danshensu. (B) LDH release assay showing the effects of Danshensu on the cytotoxicity of SW620 cells (n = 3). (C) ATP levels in the supernatants were assessed using an ATP assay kit (n = 3). (D) Microscopic images of the SW620 cells in the absence or presence of SCP. Data were represented as mean $\pm$ SD from three independent experiments. ### p 0.001 versus control, * p 0.05 versus SCP group.

Fig. 3.

Danshensu attenuated the migration of SW620 cells mediated by platelets. (A,B) The effect of Danshensu on migration of SW620 cells using Transwell. The results were represented as mean $\pm$ SD from three independent experiments (n = 3). (C,D) Western blotting analysis of E-cadherin, N-cadherin and Vimentin in the SW620 cells treated with SCP or Danshensu for 24 h (n = 3). Data were represented as mean $\pm$ SD from three independent experiments. # p 0.05, ### p 0.001 versus control group. * p 0.05, *** p 0.001 versus SCP group.

Chemoresistance is a common problem in cancer treatment. Fischer et al. [18] reported that EMT phenotype increased tumor cell chemoresistance, eventually leading to tumor development and difficulty in cancer treatment. OXA is used for colorectal cancer treatment, but the development of chemoresistance is inevitable in clinical practice [19]. Drug efflux protein MDR1 is a main marker for the drug chemoresistance in the colorectal cancer cells [20]. Therefore, we examined the role of SCP in the chemoresistance of SW620 cells. Compared with the control group, SCP group significantly increased the clonogenic survival of SW620 cancer cells. However, prophylactic administration of Danshensu onto platelets significantly decreased the clonogenic survival (p 0.001, Fig. 4A,B). MTT assay showed that the resistance of SW620 cells to OXA was increased following the stimulation of SCP (Fig. 4C). Nevertheless, intervening with platelets with Danshensu alleviated SCP-induced chemoresistance and MDR1 protein expression (Fig. 4C–E).

Fig. 4.

Danshensu diminished SCP-induced clonogenic survival and chemoresistance. (A,B) The effect of Danshensu on SW620 cell survival using colony formation assay (n = 3). (C) MTT assay showing the chemoresistances of SW620 cells to OXA following the stimulation of SCP (n = 6). (D,E) Western blotting analysis of MDR1 expression level in the SW620 cells, GAPDH was used as a loading control (n = 3). Data were represented as mean $\pm$ SD from three independent experiments. #p 0.05, ## p 0.01, ### p 0.001 versus control group. * p 0.05, ** p 0.01, *** p 0.001 versus SCP group.

It has been known that platelets contain numerous growth factors and cytokines, and the secretions of co-cultured tumor cells and platelets play a crucial role in the process of EMT. Notably, we previously reported that a variety of secreted factors from platelets were involved in the malignant progression of tumor cells [15]. In the present study, we examined the levels of various cytokines in the supernatants by ELISA assay. Our results demonstrated that the levels of IL-6, TNF-$\alpha$, IL-1$\beta$ and VEGF were increased following the co-culture of tumor cells and platelets. We uncovered that Danshensu administration inhibited the production of TNF-$\alpha$, IL-6, and IL-1$\beta$ in the co-cultured tumor cells and platelets, which was associated with the inhibition of the TGF-$\beta$/Smad signaling pathway (Fig. 5). Additionally, inactivation of the TGF-$\beta$/Smad signaling pathway suppressed EMT and chemoresistance. To reveal the mechanisms underlying Danshensu limited EMT and chemoresistance, we further investigated whether the TGF-$\beta$/Smad signaling pathway was regulated by SCP. Compared with those in the control group, the levels of TGF-$\beta$, p-Smad2 and p-Smad3 in the SW620 cells were significantly increased in the SCP group (Fig. 5E,F). More interestingly, prophylactic administration of Danshensu onto platelets decreased the levels of TGF-$\beta$, p-Smad2, and p-Smad3 in the SW620 cells (Fig. 5E,F).

Fig. 5.

Danshensu modulated the cytokine secretion from platelets induced by SW620 cancer cells. (A–D) The levels of VEGF (A), IL-6 (B), IL-1$\beta$ (C) and TNF-$\alpha$ (D) in the secretions of platelets interacting with tumor cells by ELISA (n = 3). (E,F) Western blotting analysis of TGF-$\beta$, p-Smad2, Smad2, p-Smad3 and Smad3 expression levels in the SW620 cancer cells. GAPDH was used as a loading control (n = 3). Data were represented as mean $\pm$ SD from three independent experiments. #p 0.05, ### p 0.001 versus control group. * p 0.05, ** p 0.01, *** p 0.001 versus SCP group.

To demonstrate the role of Danshensu in the progression of SW620 tumors in vivo, SCP-stimulated SW620 cancer cells were subcutaneously injected into mice, and the growth of tumors after intraperitoneal administration of oxaliplatin (OXA) was monitored accordingly (Fig. 6A). It was observed that SCP-stimulated SW620 cancer cells were insensitive to OXA and the volume of SCP-stimulated tumors were significantly increased compared to that of the control group (Fig. 6B). However, prophylactic administration of Danshensu boosted the sensitivity of SW620 cancer cells to OXA, and the tumor volume was strikingly decreased following the intervention of Danshensu compared with SCP alone (p 0.01, Fig. 6B). Moreover, all mice in the SCP group displayed remarkable liver metastasis and the percentage of liver metastasis was significantly up-regulated compared with that of control group (p 0.001, Fig. 6C,D). Surprisingly, prophylactic administration of Danshensu onto platelets prevented the liver metastasis of SCP-stimulated SW620 cancer cells (p 0.001, Fig. 6C,D). Additionally, MDR1 levels in the primary tumors were detected by immunohistochemistry. It was revealed that the expression of MDR1 was profoundly impaired following the treatment of Danshensu, in comparison to SCP alone (p 0.001, Fig. 6E,F). In addition, liver metastasis can cause the increase of ALT and AST (Fig. 6G,H). Therefore, SW620 cells incubated with SCP can induced chemoresistance and facilitated metastasis in vivo, which could be reversed in the presence of Danshensu.

Fig. 6.

Danshensu retarded the progression of SCP-stimulated SW620 tumors in vivo. (A) Schematic diagram of animal experimental procedure. (B) Growth curve of SW620 tumors following different treatments (n = 4). (C) H&E staining of the livers. (D) Quantification of liver metastasis (n = 4). (E) Immunohistochemical staining of MDR1 expression in the primary tumors (n = 4). Scale bar indicates 50 $\mu$m. (F) Quantification of MDR1 expression in tumors (n = 4). Scale bar indicates 50 $\mu$m. (G,H) Detection of ALT and AST levels in serum (n = 4). Data were represented as mean $\pm$ SD from four independent experiments. ### p 0.001 versus control group. ** p 0.01, *** p 0.001 versus SCP group.