Modulation the crosstalk between tumor-associated macrophages and non-small cell lung cancer to inhibit tumor migration and invasion by ginsenoside Rh2

G-Rh2 was obtained from National Standard Material Center (Beijing, China). Dulbecco’s modified Eagle’s medium (DMEM), fetal bovine serum (FBS), and trypsin were bought from GIBCO/BRL (Grand Island, NY, USA). VEGF-ELISA kit was purchased from R&D Systems (Minneapolis, MN, USA). VEGF antibody was from Santa Cruz Biotechnology (Santa Cruz, CA, USA).MMP9 and MMP2antibodies were purchased from Abcam (Cambridge, UK). The flow cytometry antibodies CD206, CD16/32were purchased from Peprotech (New Jersey, NJ, USA). Lipopolysaccharide (LPS) was from Sigma-Aldrich (St. Louis, MO, USA).Interferon-γ (IFN-γ) and interleukin-4 (IL-4) were produced by BioLegend (San Diego, CA, USA).

The murine macrophage-like cell line RAW264.7, human lung adenocarcinoma cell lines A549 and H1299, and human THP-1 cells were purchased from Shanghai Institute of Biological Science (Catalogue Number TCM13, TCHu150, TCHu160 and SCSP-648, respectively. Shanghai, China).

These cells were cultured in DMEM media supplemented with 10% FBS,100 U/mL of penicillin, 100 μg/mL of streptomycin at 37 °C in a humidified atmosphere containing 5% CO2.RAW264.7 and THP-1cells were polarized into M1 and M2 macrophages with different stimulation. Combination LPS (100 ng/mL) and IFN-γ(20 ng/mL) were used to generate M1 subset macrophages. IL-4 (20 ng/mL) was used to differentiate cells into M2 subset macrophages.

Transwell plate from Corning (NY, USA) with a pore size 0.4 μM was used as a co-culture system. RAW264.7 (5 × 10⁵/mL) or THP-1 (1.5 × 10⁵/mL) were loaded on the upper chamber. Cells were treated with IL-4 (20 ng/mL) for 48 h to differentiate into M2 macrophages. Lung cancer cells A549 or H1299 (2.5 × 10⁵/mL) were loaded in the lower chamber for 24 h. Then, M2 macrophages and lung cancer cells were co-cultured under conditions without serum for 24 h to generate co-cultured lung cancer cells, using lung cancer cells without co-cultured as control. These cells were used for further experiments to be treated with G-Rh2.

After 48 h stimulation, differentiated cells were harvested and identified by flow cytometry with specific makers i.e. CD16/32 for M1 and CD206 for M2 macrophages. M2 macrophages were sorted out using flow cytometry with CD206 marker for further co-culture experiment.

In brief, A549, H1299 cells, and respective co-cultured cells were seeded in 96-well plates (3 × 10³ cells/well) at the logarithmic phase. After 24 h, cells were treated with different concentrations of G-Rh2 (5, 10, 20, 40, 60, 80, 100, 120 μM) for 72 h. Then, the proliferation of the cells was determined by CCK-8 assay according to manufacturer’s instruction.

The cells were seeded in a 12-well plate to form a monolayer one day before the assay. After making a uniform straight scratch with a pipette tip, cells were incubated for 24 h. Cell motility was assessed by measuring the speed of wound closure at intervals. Each experiment was performed in triplicate.

The concentration of VEGF in the supernatant was determined by ELISA Kit (R&D System). Samples from each group were collected in sterile tubes and centrifuged at 1500 rpm for 15 min to obtain supernatants. The supernatants were analyzed according to the manufacturer’s instructions. Results were presented as picograms of VEGF per milliliter.

Briefly, cells were washed twice with ice-cold phosphate buffer saline(PBS) after treatment with G-Rh2 for 24 h. Next, cells were harvested with ice-cold lysis buffer. Then, cell lysates were centrifuged at 12,000×g for 10 min at 4 °C and collected the supernatant. The total of 50 μg protein per sample was separated by electrophoresis on 8 to 10% SDS-PAGE gel. Then, protein was transferred onto a nitrocellulose membrane. The membrane was blocked with 5% non-fat dry milk for 1 h and incubated with MMP2, MMP9, and VEGF-C (1:1000) primary antibodies overnight at 4 °C. β-actin was used as a loading control.

Total RNA isolated from cells using an RNeasy Micro kit (Qiagen) was converted to first-strand cDNA using a high-capacity cDNA reverse transcription kit (Applied Biosystem). Quantitative real-time PCR assays were performed with SYBR Green PCR Master Mix (Applied Biosystems) and a 7900HT Fast Real-time PCR System (Applied Biosystems). All primers were synthesized in Huada Biotechnology Corporation (Shenzhen, China). The sequence of primers was shown in the Table 1. All data were normalized by β-actin.

It was performed as previously described [11]. Briefly, paraffin-embedded tumor samples were cut into 4 μm-thick sections and mounted on polylysine-coated slides. Samples were dewaxed in xylene and rehydrated using a graded series of ethanol solutions. After deparaffinization, endogenous peroxidase activity was blocked by incubation with 3% peroxide-methanol solution at room temperature (RT) for 10 min, and then antigen retrieval was performed at 100 °C in an autoclave for 7 min. After washing with PBS, sections were incubated with primary antibodies against theCD206 monoclonal antibody (clone 10D6, Zhongshan Goldenbridge Biotechnology Co., LTD., Beijing, China) and VEGF-C (Santa Cruz Biotechnology, Santa Cruz, CA, USA)overnight at 4 °C. Next, sections were incubated with aDAKO EnVision kit (DAKO, Glostrup, Denmark) following the manufacturer’s instructions. Finally, sections were faintly counter-stained with hematoxylin and mounted with glycerol gelatin.

Female 5-week-old C57BL/6 mice (n = 14) were purchased from Shanghai Silaike Experiment Animal Co., Ltd. (Shanghai, China). Animal experiments were conducted in animal room with Specific Pathogen Free (SPF) standards. All animal experiment protocols were approved by Institutional Animal Care and Use Committee. Each mouse was subcutaneously injected 5 × 10⁵ murine lewis lung carcinoma (LLC) cells on right should blade. Then, mice were randomly divided into two groups: vehicle control (n = 7) and G-Rh2 (n = 7) which was administered i.p. at 40 mg kg^− 1 daily for 21 days. Tumor size was measured daily. Then mice were sacrificed after CO₂ anesthesia. Tumor tissues were isolated and fixed in formalin immediately for further immunohistochemistry experiments.

M2 macrophages are considered as an important subtype of TAMs to affect tumor metastasis [19, 20]. In order to investigate how G-Rh2 affects the function of M2 macrophage, unstimulated RAW264.7 cells (M0) were classically treated with LPS (100 ng/mL) and INF-γ(20 ng/mL) for 48 h and differentiated into M1 subset (Fig. 1a) whereas cells stimulated withIL-4 (20 ng/mL) promoted M2macrophage polarization, exhibiting different cellular morphologies between two subsets of macrophages (Fig. 1a). These cells were further identified with specific markers throughflow cytometry analysis. CD206 is a crucial marker for M2 macrophages which was dramatically upregulated after induction by cytokines (Fig. 1b and c). In contrast, markers specific for M1 subtype CD16 and CD32 were remarkably decreased in M2 subtype (Fig. 1b and d). Further examination to detect other biomarkers demonstrated that tumor necrosis factor alpha (TNF-α) and inducible nitric oxide synthase (iNOS) were significantly upregulated in M1 macrophages, whereas arginase 1(ARG-1) was remarkably elevated in M2 subtype (Fig. 1e). To confirm the cell polarization, human THP-1 monocyte was treated with the same combination cytokines as above. M1 and M2 macrophages had different morphologies (Fig. 1f). And M2 subtype displayed higher levels of CD206 (Fig. 1g and h), whereas M1 macrophages had higher levels of CD16/32 than that in M2 subtype (Fig. 1g and i). The expression pattern of TNF-α, iNOS, and ARG-1 in THP-1 derived M1 and M2 macrophages was similar to that derived from RAW264.7 cells (Fig. 1j and e). All of these results suggest that combination of these inflammatory factors is an effective way to polarize M1 and M2 subtypes of macrophage.

To mimic the original tumor microenvironment as much as possible, two co-cultured NSCLC cell lines were developed by co-culturing A549 and H1299 with RAW264.7 derived M2 macrophages being alternatively induced, which was named as MA549 and MH1299 cells. A549 and H1299 cells, as well asMA549/MH1299 cells were treated with different concentrations of G-Rh2 for 72 h. As shown in the Fig. 2a and b, g-Rh2had a potential to suppress the growth of A549, H1299 and MA549/MH1299 cells in a dose-dependent manner. It indicated a trend that G-Rh2 at the high doses over 100 μM could inhibit more MA549/MH1299 cell growth than that of not co-cultured cells, but without statistical significance (Fig. 2a and b). In line with above results, A549 cells were co-cultured with THP-1 derived M2 macrophages. G-Rh2 inhibited more co-cultured cell growth than that of A549 cells, but without significant difference (Fig. 2c). To further study the inhibitory effects of G-Rh2on cell migration, we performed a scratch wound model in the presence of mitomycin C to inhibit cell proliferation. Compared with negative control cells (NC), Co-cultured A549cells migrated faster at two time-points of 24 and 48 h, indicating the mobility of NSCLC cells after co-culture was intensively increased (Fig. 2d). Aftertreatment with G-Rh2 (100 μM), the mobility of co-cultured A549 cells was effectively blocked after 24 h (Fig. 2d). Furthermore, cells almost lost mobility after48 hours exposure to G-Rh2 (Fig. 2d). This finding indicates that G-Rh2 is a potent compound to prevent the migration of co-cultured NSCLC cells.

Since the interaction between TAMs and cancer cells is the key factor to promote cancer metastasis [21], a question was raised concerning whether G-Rh2 affects the phenotype of M2 TAMs. M2 macrophages were identified and sorted out using flow cytometry by detection of the CD206 marker. Then, sorted M2 cells were treated with different doses of G-Rh2 for 24 h. These treated cells were harvested and analyzed through flow cytometry with characteristic markers. As shown in Fig. 3, G-Rh2 significantly reduced the expression of CD206expression in M2 macrophages derived from RAW264.7 in a dose-dependent manner (Fig. 3a and b). It was particularly interesting to detect that M1 markers CD16/32 expression was simultaneously increased in a dose-responsive way (Fig. 3c and d) after G-Rh2 treatment. To further confirm this function of G-Rh2,M2 macrophages differentiated from human THP-1 cells were treated with different concentrations of G-Rh2 for 24 h. The similar subtype switch was observed that M1 markers CD16/32 was increased whereas M2 phenotype CD206 was decreased (Fig. 3e-h). All of these results indicate that G-Rh2 has a potential to shift M2 phenotype to M1 thereby affecting the biological function of TAMs.

Compelling evidence indicates that VEGF and MMPs are important factors involving in the cancer metastases, which may be regulated by M2 macrophages in tumor microenvironment [22‐24]. The secretory levels of VEGF-C were measured by ELISA. Results showed that VEGF levels were significantly increased in co-cultured A549 cells after being co-cultured 12 and 24 h with M2 macrophages derived from RAW264.7, compared with that in the media of A549 cells (Fig. 4a). G-Rh2 reduced the basal levels of VEGF in A549 culture media and decreased more in co-culturing system (Fig. 4a). As for another NSCLC cell line H1299, there was a tendency to upregulate VEGF-C levels after co-cultured with M2 macrophages derived from RAW264.7, but without significant difference (Fig. 4b). Similarly as in A549 or co-cultured A549, G-Rh2 remarkably inhibited the secretion of VEGF in H1299 and co-cultured H1299, especially in co-culturing system (Fig. 4b). When A549 cells were co-cultured with M2 macrophages differentiated from THP-1 cells, secretion of VEGF-C was increased and G-Rh2 remarkably inhibited the up-regulation of VEGF-C (Fig. 4c). In agreement with the secretory levels of VEGF, VEGF-C mRNA expression levels were increased in Co-A549 and Co-H1299 cells, compared to their respective controls. G-Rh2 significantly reduced the mRNA levels of VEGF-C in A549 and H1299 cells and effectively blocked the induction of VEGF by co-culturing with M2 macrophages derived from RAW264.7 or THP-1 (Fig. 4d-f). Similar regulatory patterns were observed in the expression of MMP9 and MMP2 mRNA by G-Rh2 in two lung cancer cell lines with or without being co-cultured (Fig. 4d-f). These observations suggest that M2 macrophages promote the expression of VEGF and MMPs in lung cancer cells.

Since VEGF and MMPs promote cancer cell invasion and metastasis mainly through their respective proteins [24‐26], thereby protein expression levels were measured by Western blotting. In line with the regulation of mRNA expression, the protein expression levels of VEGF-C were decreased by G-Rh2 at high concentration of 100 μM in A549 cells, whereas G-Rh2 started to decrease VEGF-C at concentration of 60 μM in co-cultured A549 with M2 derived from RAW264.7 cells. Given the same concentration of G-Rh2 at 100 μM, it was more potent to reduce VEGF-C protein levels in Co-A549 cells over than that in A549 cells without being co-cultured with M2 macrophages (Fig. 5a). As for theMMP9, G-Rh2 weakly reduced the protein levels in A549 even at high concentration (Fig. 5b). In contrast, G-Rh2 clearly decreased MMP9 protein expression at 60μMin co-cultured A549 cells (Fig. 5b). Similarly, G-Rh2 significantly blocked the MMP2 protein levels at high concentration in A549 cells, while MMP2 protein levels were remarkable reduced at low concentration in co-cultured A549 cells (Fig. 5c). The quantification results were consistent with that from immunoblotting (Fig. 5c-f). In another co-culturing system that A549 with M2 differentiated from THP-1 cells, VEGF protein levels was weakly reduced by G-Rh2 at high concentration (100 μM) in A549 cells, but G-Rh2 remarkably decreased VEGF protein levels at low concentration of 60 μM (Fig. 5g). Interestingly, total VEGF levels in co-cultured A549 with M2 derived from THP-1 cells were lower than that of A549 cells (Fig. 5h). This was different from that in co-cultured A549 with M2 derived from RAW264.7 cells expressing higher levels of VEGF than A549 cells (Fig. 5d). Our results demonstrated that M2 macrophages modulate the biological behaviors of lung cancer cells and G-Rh2 displays a special preference to block expression of these aggressive factors related with cancer malignancy under co-cultured conditions.

To confirm the regulatory function of G-Rh2 in the communication between TAMs and lung cancer cells, murine lewis lung carcinoma cells (LLC) were injected subcutaneously in C57BL/6 mice. Then, mice were divided into two groups i.e. vehicle control and G-Rh2 administered group. Tumor size was measured daily. After 21 days, mice were sacrificed and tumor tissues were fixed for immunohistochemistry. Strong cytoplasmic staining of VEGF-C was predominantly observed in cancer cells and tumor stromal cells (Fig. 6a) and G-Rh2 significantly inhibited VEGF-C expression (Fig. 6b). As for the marker of M2 macrophages, CD206 was highly expressed on the cell membrane and cytoplasm in the infiltrative macrophages among the tumor cells (Fig. 6c). G-Rh2 remarkably decreased CD206 expression (Fig. 6d). Importantly, G-Rh2 also significantly reduced the tumor size compared with vehicle control group (Fig. 6e). These results clued that G-Rh2 can prevent macrophages from differentiation into the M2 subtype, which might disassociate the communication between TAMs and lung cancer cells.