Bufalin suppresses tumour microenvironment-mediated angiogenesis by inhibiting the STAT3 signaling pathway

Anti-angiogenesis therapy has increasingly become an important strategy for the treatment of colorectal cancer. Recent studies have shown that tumor microenvironment (TME) promotes tumour angiogenesis. Bufalin is an active compound whose anti-tumor ecacy has been proven by previous studies. However, there are very few studies on the anti-angiogenic effects of bufalin. umbilical vein endothelial cells (HUVEC) tube formation, migration and adhesion test were used to assess angiogenesis in vitro. Western blot and quantitative PCR were used to detect relevant protein levels and the expressions of mRNAs. Subcutaneous xenograft tumor model and hepatic metastasis model in mice were established to investigate the inuence of bufalin on angiogenesis-mediated by TME in vivo.


Results
We found that the angiogenesis mediated by tumor microenvironment cells was signi cantly inhibited in the present of bufalin. The results demonstrated that the pro-angiogenic gene in HUVEC such as VEGF, PDGFA, E-selectin and P-selectin were downregulated by bufalin, and the downregulation was regulated by inhibiting the STAT3 pathway. Overexpression STAT3 could reverse the inhibitory effect of bufalin on angiogenesis. What is more, few reduction of angiogenesis when bufalin directly acted on tumor microenvironment cells.

Conclusion
Our ndings demonstrate that bufalin suppresses tumour microenvironment-mediated angiogenesis by inhibiting the STAT3 signaling pathway of vascular endothelial cells, which reveals that bufalin may be used as a new anti-angiogenic adjuvant therapy medicine in the treatment of colorectal cancer.

Background
Colorectal cancer (CRC) is the third most common cancer worldwide, with a mortality rate of 9% of all cancer-related deaths (1), and the incidence is increasing. It is now widely accepted that angiogenesis plays a key role in tumour development, progression and metastasis. Excessive abnormal angiogenesis is a hallmark of solid tumours (2). The concept of anti-angiogenic therapy arose from the seminal observations of Judah Folkman and colleagues. It has been reported that anti-angiogenic therapy can effectively improve the survival rate of CRC patients, which manifests that inhibiting tumour angiogenesis is a latent method for treating CRC (3)(4)(5)(6).
Many studies have demonstrated that the tumour microenvironment, including cancer-associated broblasts (CAFs) and tumour-associated macrophages (TAMs), promotes tumour angiogenesis (7)(8)(9). The tumour microenvironment, which is also termed the tumour mesenchyme or tumour stroma, includes CAFs, TAMs, blood vessels and extracellular matrix and substantially in uences the initiation, growth and dissemination of colorectal cancer (2). Signal transducer and activator of transcription 3 (STAT3) belongs to a family of transcription factors that regulate the expression of genes involved in the pathogenesis of many human malignancies (10,11). It was also reported that the TME activates STAT3 signalling in human umbilical vein endothelial cells (12), and the tumour microenvironment may affect angiogenesis through the STAT3 signalling pathway. In summary, the STAT3 pathway in blood vessels may become a target for the treatment of angiogenesis.
Bufalin (BU), the major bioactive component isolated from toad venom, has been con rmed to be a potent antitumour drug through its effect on tumour cell apoptosis, metastasis and proliferation (13,14).
In addition, bufalin also inhibits angiogenesis, and it was reported that the anti-angiogenic effect of sorafenib was signally increased in combination with bufalin by targeting AKT/VEGF in HUVECs(15). Wu et al. showed that bufalin enhanced cytocidal effects by targeting the STAT3 pathway (16).
Previously, we performed numerous studies on bufalin in the treatment of colorectal cancer (13,17,18).
Here, we found that bufalin could reverse the pro-angiogenic effects mediated by TME. In the present study, we used in vitro angiogenesis-related experiments and in vivo subcutaneous tumour models and liver metastasis models to demonstrate that bufalin can inhibit TME-mediated STAT3 activation in endothelial cells to reduce angiogenesis and identi ed a new mechanism of bufalin in the treatment of CRC.

Materials And Methods
Cell culture Cell were cultured in a humidi ed incubator with an atmosphere of 5% CO 2 at 37°C under normal oxygen conditions. Human umbilical vein endothelial cells (HUVECs, #8000, Sciencell, USA) were grown in endothelial cell medium (ECM, # 1001, Sciencell, USA) and only early passages (< p6) were used. CT26 cells were obtained from the Cell Bank of the Chinese Academy of Science s and were cultured in RPMI-1640 containing 10% FBS and 1% penicillin/streptomycin. The STAT3 overexpression plasmid were purchased from GeneChem (Shanghai, CN).
Conditioned medium (CM) preparation CT26 cells and CAF were con rmed by morphological observation (Fig. S1a) and WB (Fig. S1b). The tumor cell supernatant polarizes the M0 macrophages, which was con rmed by morphological observation and ow cytometry (Fig. S1c-d). After tumour-associated macrophages, cancer-associated broblasts, CT26 cells have grown to 80% of the bottom area of the ask, replace with FBS-free medium, at 48h post-treatment, cell suspensions were collected as conditioned medium. The CM was collected after high speed centrifugation and then pass 0.22 μm microporous mebrance, stored at -20˚C. In the same way, the cells were treated with 10 nM of bufalin for 24 hours, and then treated as above to obtain the relevant conditioned medium after drug action.
Tube formation assay.
HUVEC cells were incubated with conditioned medium for 24 hours before tube formation. Matrigel (BD, #356234, USA) was thawed overnight at 4˚C a day in advance, 50 μl of Matrigel was spread in a 96-well plate and polymerized at 37˚C for 30 minutes. Next, 3 × 10 4 HUVECs in 50 μl ECM were seeded to each well and incubated at 37°C in 5% CO 2 for 4 hours, photographed under the microscope.
Cell migration assay HUVEC cells were incubated with conditioned medium for 24 hours before migration assay. 3×10 4 HUVECs in 300 μl of serum-free ECM mentioned earlier in the upper chamber and were seeded into a transwell insert (8.0 μm pore size, #353097, FALCON, USA), and allowed to migrate towards 700 μl of complete ECM. The non-migrated cells were removed from the upper part of the transwell with a cotton swab after 6 hours of incubation, and the insert was xed with 4% paraformaldehyde for 10 minutes at room temperature. Transwell inserts were stained in 500 μl of 0.03% crystal violet solution for 30 minutes at 37℃.Then the insert was immersed and washed three times with PBS solution for 5 minutes each time, dried and photographed under the microscope.

Adhesion assay
After HUVEC cells were overgrown in a 24-well plate, they were treated with conditioned medium for 24 hours. Then 2×10 4 HCT116-GFP cells in 300 μl of serum-free ECM were inoculated on the treated HUVEC cells, and then incubated in a 37℃ incubator for 2 hours, washed with PBS three times, and then photographed under a uorescence microscope.

Quantitative PCR
Total RNA was extracted from HUVEC using TRIzol (Invitrogen). The concentration of total RNA was quanti ed by measuring the absorbance at 260 nm. For SYBR Green-based quantitative PCR ampli cation, the reaction were carried out in a 20 μl reaction volume (Applied Biosystems). Using 2 −ΔΔCt method to determine the relative expression level of each cell line in each group.

Western Blot (WB)
The proteins separated by SDS-PAGE and subjected to immunoblotting to analyze antibodies (Abcam, USA). Blocked and incubated overnight at 4°C with primary antibodies. The membrane was further probed with horseradish peroxidase-conjugated anti-rabbit/mouse IgG antibody (Cell Signaling Technology, 1:5,000). Using ImageJ software to quantify protein bands.

In Vivo Xenograft Model
To determine the in vivo antiangiogenic activity of bufalin treatment, CT26-LUC cells (2× 10 6 ) were injected into the anks or spleen of male Balb/c mice (6 weeks old). One weeks after injection, bufalin (1 mg/kg) was administered by intraperitoneal (i.p.) injection once every other day for 21 days ( ank) or 14 days (spleen). Subcutaneous tumour size was measured every three days and live imaging once a week after the treatment. The estimated tumor volumes (Vs) were calculated by the formula V=W 2 ×L×0.5, where W represents the largest tumor diameter in centimeters and L represents the next largest tumor diameter. Tumor-bearing mice were sacri ced after 21 or 14 days of treatment, and tumour tissues, spleen and liver were harvested, weighed, and then immediately xed in formalin for follow-up experiment.
All experiments conformed to the ethical principles of animal experimentation stipulated by institutional animal care and use committee of Putuo Hospital, Shanghai University of Traditional Chinese Medicine, China.

Histopathological assay
The tissues were harvested after animals were sacri ced. The procedure of histopathological assay was accomplished by conventional hematoxylin-eosin (H&E) staining in accordance with standard techniques.
Immuno uorescence 2×10 4 HUVECs were seeded and cultured overnight on microscope coverslips (Thermo Fisher Scienti c, Waltham, MA, USA). After treated with CM with or without bufalin for 24 hours as described previously, the cells were washed with PBS twice and xed in methanol for 15 minutes, then permeabilized with 0.2% Triton X-100 (Beyotime, Shanghai, China) /PBS for 5 minutes and blocked by 3% BSA for 1 hours at room temperature. The coverslips were incubated with primary antibodies at 4°C overnight, and then incubated with secondary antibodies for 2 hours at 37°C (protected from light). Nuclear localization was assessed with 4′,6-diamidino-2-phenylindole (DAPI, Beyotime, Shanghai, China).
Tissue sections were permeabilized with cold methanol for 5 minutes and incubated with 5% BSA in PBS for 1h. Primary antibodies were applied in blocking buffer and incubated overnight at 4°C. Dyeconjugated secondary antibodies were added in blocking buffer and incubated for 2 hours. Nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI, Beyotime, Shanghai, China). Images were taken using Zeiss LSM880 confocal microscopy at the same voltage level and analyzed using ZEN Software.

Immunohistochemistry (IHC)
The tissue was xed in 10% formalin, embedded in para n, and then sectioned (5 mm thick). The IHC of CD31 was performed as follows. The slides were dewaxed and incubated with 3% H 2 O 2 aqueous solution for 10 minutes to quench the endogenous peroxidase activity. The heat-induced antigen recovery method was used to detect the antigen. Incubate the tissue with 5% BSA for 30 minutes at room temperature, and then incubate with the primary antibody in PBS at 4°C overnight. Using the appropriate secondary antibody to apply the indirect avidin-biotin-peroxidase method at room temperature for 30 minutes. EnVision (K4007, Dako) signal enhancement system was used to develop bound antibodies. The sections were stained with Harris hematoxylin, dehydrated and xed. For quanti cation, 30 random images (400×) were captured with a microscope (Leica, Wetzlar, Germany) under each microscope.
Elisa VEGF in HUVEC cells culture supernatants was evaluated by Human VEGF Quantikine ELISA Kit (R&D, Minnesota, USA) according to the manufacturer's protocol. VEGF in serum was evaluated by Mouse VEGF Quantikine ELISA Kit (R&D, Minnesota, USA) according to the manufacturer's protocol.

Statistical Analysis
Each experimental value is expressed as the mean ± SD. Statistical analysis was performed using t-test to evaluate the signi cance of the difference between the different groups, and the signi cance was accepted at *p<0.05, **p<0.01 and ***p<0.001. All data points represent the average of three repeated measurements. Spearman rank statistical test and Mann-Whitney test were used for statistical analysis of tissue samples to assess the signi cance of differences between groups.

Bufalin suppresses angiogenesis induced by tumour microenvironment cells
Angiogenesis programming in tumour tissue is a multi-dimensional process, which is jointly regulated by cancer cells and various tumour-associated stromal cells and their biologically active products (7). CAFs, TAMs, and tumour cells are important components of the TME and contribute to tumour growth, metastasis, and neovascularization. Bufalin, as an active compound (Fig. 1a), has been reported to inhibit migration and invasion of hepatoma cells and lung cancer cell lines by down-regulating VEGF, which plays an important primary in tumour angiogenesis (19).
To clarify the effect of bufalin on angiogenesis caused by TME cells, we collected the supernatant of CT26 cells, CAFs and TAM as TME-conditioned media (CMs). Furthermore, we selected a low concentration IC 15 (10 nM) and short treatment time (24 hours) by measuring the IC 50 (30 nM) value of bufalin for HUVECs for 24 hours (Fig. 1b), to examine whether bufalin inhibits angiogenesis under these conditions to explore the mechanism by which bufalin affects the occurrence and development of colorectal cancer. The cell proliferation experiment showed that TME have no signi cant effect on HUVECs proliferation after treated by TME-CMs (CT26-CM, CAF-CM and TAM-CM) for 24 hours (Fig. 1c). The effect of bufalin on TME-mediated angiogenesis were determined by HUVEC tube formation, migration and adhesion experiments, and the results showed that bufalin inhibit angiogenesis induced by TME (Fig. 1d-f).

Bufalin suppresses TME-mediated angiogenesis by inhibiting angiogenic factors
To determine the mechanism of bufalin inhibit TME-mediated angiogenesis, we observed the expression of the angiogenesis-related and migratory factors VEGF and PDGFA and the vascular adhesion genes Eselectin and P-selectin in HUVECs treated with or without bufalin in combination with TME-CMs by quantitative PCR. The results showed that bufalin could signi cantly down-regulate those gene expression in HUVECs increased by TME (Fig. 2a-c).
Previous studies have shown that the STAT3 signaling pathway play a critical role in angiogenesis by inducing angiogenic factor, like VEGFs, PDGFs and so on (20,21). In our study, the WB experiment showed bufalin could directly decease STAT3 phosphorylation in HUVECs activated by TME (Fig. 2d-f), and the results of immuno uorescence experiment also con rmed these ndings (Fig. 2g). In addition, we found bufalin could inhibit VEGF expression by WB (Fig. 2d-f) and ELISA (Fig. 2h-j). Taken together, these results suggest that bufalin could inhibit STAT3 phosphorylation to decreased angiogenic factor expression induced by TME in vascular endothelial cells.
Bufalin suppresses TME-mediated angiogenesis by the STAT3 signaling pathway To investigate whether the STAT3 signaling pathway is a key factor by which bufalin inhibits TMEmediated angiogenesis, we constructed a STAT3-overexpression (STAT3-OE) plasmid. STAT3 expression was con rmed by WB (Fig. S2a) and qPCR (Fig. S2b). We found that the STAT3-OE plasmid prevented bufalin-mediated inhibition of HUVEC tube formation (Fig. 3a), migration (Fig. 3b) and adhesion (Fig. 3c) induced by TME. These results suggest that bufalin inhibits TME-induced angiogenesis by the STAT3 signaling pathway, and STAT3 plays an important role in this process.
Bufalin suppresses TME-mediated angiogenesis by directly affecting vascular endothelial cells but not altering tumour microenvironment cells Next, we determined whether bufalin also indirectly affects endothelial cells by changing TME cells. We rst treated TME-cells (CT26, CAF & TAM) with 10 nM bufalin for 24 hours, replaced the medium with serum-free medium and collected the cell supernatant after an additional 24 hours for use as (TME+BU)-CM, which was then used to treat HUVECs for 24 hours. Then, HUVEC tube formation (Fig. 4a), migration (Fig. 4b) and adhesion (Fig. 4c) were analysed. The results showed that bufalin did not affect vascular endothelial cells by affecting TME cells, and as the same as well, the WB results also proved that there are no signi cant differences in the increase in phosphorylated STAT3 levels in HUVECs induced by (TME+BU)-CMs and TME-CMs (Fig. 4d-f). Taken together, these results suggest that bufalin suppresses TME-mediated angiogenesis by directly affecting vascular endothelial cells but not altering TME-cells.

Bufalin inhibits CRC cells growth via anti-angiogenesis in vivo
To determine the effect of bufalin in vivo, a CRC cells xenograft model was established by using CT26 cells expressing luciferase (CT26-LUC). The schedule and scheme are shown in (Fig. 5a). All mice started treatment one week after xenotransplantation. The vehicle group was treated with normal saline, and the treatment group was administered bufalin (1 mg/kg). The volumes of the subcutaneous tumours (Fig.   5b) and in vivo imaging (Fig. 5c, Fig. S3b) were recorded during the treatment cycle (three weeks). Twenty-eight days after inoculation, the mice were sacri ced, subcutaneous tumours in the two groups of mice were weighted and photographed after the treatment was over (Fig. 5d). The results showed that bufalin signi cantly inhibited the growth of tumours compared with the vehicle without affecting animal body weights (Fig. S3a), which proved that bufalin had no serious toxic effects on the body. To further con rm the anti-angiogenesis effect obtained in vitro, we conducted IHC to evaluate the expression of CD31 and Ki67 in tumours, the results of IHC showed the reduction of proliferation and subcutaneous tumor blood vessels in the bufalin group (Fig. 5e). Next, we measured the serum VEGF levels in the two groups of mice by ELISA, and the results showed that the serum VEGF levels in the bufalin group were signi cantly lower than those in the vehicle group (Fig. 5f). Moreover, we also examined the activation of vascular p-STAT3 in solid tumours. The immuno uorescence results showed that the number of blood vessels in subcutaneous tumours and the proportion of activated p-STAT3-positive blood vessels in the bufalin group were signi cantly lower than those in the vehicle group (Fig. 5g). These data indicated that bufalin inhibits angiogenesis by targeting the activation of p-STAT3 in tumour blood vessels, thereby inhibiting tumour growth in a CRC cells xenograft model.

Bufalin inhibits CRC cells metastasis via anti-angiogenesis in vivo
Cancer metastasis remains a major challenge to the successful management of malignant diseases. The liver is the main site of metastatic disease and a major cause of death from colorectal cancer (22). To investigate the effect on bufalin inhibiting CRC cells metastasis via anti-angiogenesis in vivo, we established a CRC cells liver metastasis model (Fig. 6a), and tumour metastasis was observed by an in vivo imaging system from day 7 to day 21 (Fig. 6b, Fig. S4b). The in vivo imaging results showed that after one week of bufalin treatment, bufalin began to inhibit liver metastasis compared to that of the vehicle group, and until the mice were sacri ced on day 21, bufalin signi cantly inhibited liver metastasis by more than 50% compared to that of the vehicle without affecting animal body weights (Fig. S4a). Metastatic foci of considerable sizes were visible in the livers of mice treated with vehicle. Haematoxylineosin-stained liver sections were examined under a microscope, and as expected, the formation of metastases in the liver was reduced by approximately 80% by bufalin treatment (Fig. 6c-d). The results of IHC showed the reduction of blood vessels in spleen and liver metastases in the bufalin group (Fig. 6e). Similarly, mice treated with bufalin had a signi cantly lower serum VEGF levels than mice treated with the vehicle (Fig. 6f). Consistent with the subcutaneous tumours, the immuno uorescence results showed that the number of blood vessels in liver tumours and p-STAT3-positive blood vessels signi cantly decreased after bufalin treatment (Fig. 6g). Interestingly, p-STAT3 in endothelial cells was only activated in the tumour site, while it was rarely activated in normal liver tissues. This result further suggests that bufalin inhibits liver metastasis by targeting STAT3 in tumour blood vessels not only in primary tumours but also in metastatic tumours.

Discussion
Accumulating evidence has substantiated that angiogenesis plays a critical role in tumour progression and that inhibiting angiogenesis is a promising strategy for tumour treatment. Angiogenic programming in neoplastic tissue is a multidimensional process regulated by tumour cells in conjunction with various tumour-associated stromal cells, as well as the TME (9,(23)(24)(25). We found that bufalin could reverse angiogenesis mediated by the TME.
Anti-angiogenesis therapy is an important strategy for the treatment of CRC. Previous studies have reported that bufalin can synergistically enhance the anti-angiogenic effect of sorafenib via AKT/VEGF signalling (15). We found that bufalin could inhibit the tube formation, adhesion and migration of HUVECs mediated by CAFs, TAMs and tumour cells by inhibiting the activation of HUVEC STAT3 and thereby decreasing the expression of VEGF, PDGFA, E-selectin, and P-selectin. Similarly, we established subcutaneous tumour models and liver metastasis models in vivo and found that bufalin inhibited the growth and metastasis of CRC by signi cantly reducing the number of blood vessels and STAT3 phosphorylation in vascular endothelial cells. In addition, the serum concentration of VEGF after bufalin treatment was also signi cantly reduced. This nding indicated that bufalin targets the STAT3 signalling pathway to reduce TME-mediated angiogenesis.
STAT3 is a transcription factor that regulates various kinds of cellular events, including differentiation, apoptosis and proliferation. Previous studies have shown that STAT3 activation promotes tumour angiogenesis by increasing VEGF expression (26,27). Intercellular communication between the TME and vascular endothelial cells is promoted by STAT3 (28). Notably, STAT3 is one of the important targets of bufalin, and bufalin can inhibit STAT3 activity. Moreover, we found that STAT3 overexpression could reverse the inhibitory effect of bufalin on angiogenesis. Collectively, we propose a model in which bufalin reduces the expression of angiogenesis genes by inhibiting the phosphorylation of STAT3 on endothelial cells, thereby antagonizing the pro-angiogenic effect of the tumor microenvironment (Fig. 7) More interestingly, we found that bufalin could suppress TME-mediated angiogenesis. Furthermore, we treated CAFs, TMAs and tumour cells with bufalin and then used the conditioned cell supernatant to treat HUVECs and found no promotion of tube formation, migration or adhesion. Notably, compared to some previous antitumour studies of bufalin (29,30), we chose a lower bufalin concentration and a shorter treatment time for both in vitro and in vivo experiments, which ruled out the direct effect of bufalin on TME cells. These results suggest a new mechanism by which low concentrations of bufalin affect TMEmediated angiogenesis by directly acting on the STAT3 signalling pathway on vascular endothelial cells but not on TME cells, and this effect is characterized by low toxicity and high e ciency.
In summary, our results show that the TME promotes tumour angiogenesis by activating STAT3 in vascular endothelial cells and that bufalin can precisely inhibit angiogenesis by targeting STAT3.
Through our research, we have enriched the understanding of the antitumour effect of bufalin, which can indirectly inhibit TME-mediated angiogenesis. In the future, bufalin may be developed as a new type of anti-angiogenic auxiliary drug.

Conclusions
In summary, bufalin suppresses tumour microenvironment-mediated angiogenesis by inhibiting the STAT3 signaling pathway. Tumor microenvironment promotes tumour angiogenesis by activating STAT3 in vascular endothelial cells, and that bufalin can precisely inhibit angiogenesis by targeting STAT3 which reveals that bufalin may be used as a new anti-angiogenic adjuvant therapy medicine in the treatment of colorectal cancer.

Declarations
This project was sponsored by the Science and technology innovation project of Putuo District Health system (ptkwws201905), the Budget project of Shanghai University of Traditional Chinese Medicine (2019LK039) and the Natural Science Foundation of Shanghai (20ZR1450500,19ZR1413800). This project was also sponsored by the National Nature Science Foundation of China (81873137).

Availability of data and materials
The datasets during and/or analysed during the current study available from the corresponding author on reasonable request.
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.  b Cell viability of HUVEC after treated with BU for 24h. c Cell proliferation of HUVEC after treated with different TME-CMs for 24h. The effect of bufalin (BU) on the tube formation (d), migration (e) and adhesion (f) of HUVECs in response to different TME-CMs. *P<0.05, **P<0.01, ***P<0.001. Each bar represents the mean ± SD of three independent experiments. BU, bufalin in HUVECs treated with different TME-CMs in the presence or absence of BU, membranes were stripped and re-probed with total β-actin as a control. g-i The concentration of VEGFA in HUVEC supernatant after treatment with different TME-CMs in the presence or absence of BU. j Immuno uorescence analysis showing p-STAT3+ HUVECs after treatment with different TME-CMs in the presence or absence of BU. *P<0.05, **P<0.01, ***P<0.001. Each bar represents the mean ± SD of three independent experiments. BU, bufalin  Bufalin suppresses TME-mediated angiogenesis by the STAT3 signaling pathway. Tube formation (a), migration (b) and adhesion (c) of HUVECs after treatment with TME-CMs and BU with or without the STAT3-OE plasmid. *P<0.05, **P<0.01, ***P<0.001. Each bar represents the mean ± SD of three independent experiments. BU, bufalin Figure 4 Bufalin suppresses TME-mediated angiogenesis by directly affecting vascular endothelial cells but not altering tumour microenvironment cells. Tube formation (a), migration (b), and adhesion (c) of HUVECs treated with TME-CMs and TME+BU-CMs for 24 hours. d-f WB showing the protein expression of p-STAT3 and STAT3 in HUVECs treated with different TME-CMs and TME+BU-CMs, membranes were stripped and re-probed with total β-actin as a control. N.S. indicates no signi cant difference between the two groups, P>0.05. BU, bufalin Figure 5 Bufalin inhibits CRC cells growth via anti-angiogenesis in vivo. a Scheme and schedule of imaging and treatments. b Tumour volumes from day 0 to day 28. c Tumour growth was visualized by an in vivo imaging system from day 7 to day 28. d Tumour weights and images. e IHC analysis of CD31 and Ki67 in tumours. f VEGF expression level in serum. g Immuno uorescence analysis of CD31 and p-STAT3 in tumours. **P<0.01, ***P<0.001. Each bar represents the mean ± SD of three independent experiments. IHC, immunohistochemistry Figure 6 Bufalin inhibits CRC cells metastasis via anti-angiogenesis in vivo. a Scheme and schedule of imaging and treatments. b Tumour metastasis was visualized by an in vivo imaging system from day 7 to day 21.
c Representative images of liver and H&E-stained liver tissue. d Tumour number and area in liver. e IHC analysis of CD31 in spleen and liver tissue. f VEGF expression level in serum. g Immuno uorescence analysis of CD31 and p-STAT3 in liver. **P<0.01, ***P<0.001. Each bar represents the mean ± SD of three independent experiments. IHC, immunohistochemistry Figure 7 Page 21/21 The mechanism of BU inhibiting tumor microenvironment-mediated angiogenesis.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.