A Chinese Herbal Formula Suppresses Colorectal Cancer Migration and Vasculogenic Mimicry Through ROS/HIF-1 α/MMP2 Pathway in Hypoxic Microenvironment

Abstract

Various malignant tumors, including colorectal cancer, have the ability to form functional blood vessels for tumor growth and metastasis. Vasculogenic mimicry (VM) refers to the ability of highly invasive tumor cells to link each other to form vessels, which is associated with poor cancer prognosis. However, the antitumor VM agents are still lacking in the clinic. Astragalus Atractylodes mixture (AAM), a traditional Chinese medicine, has shown to inhibit VM formation; however the exact mechanism is not completely clarified. In this study, we found that HCT-116 and LoVo could form a VM network. Additionally, hypoxia increases the intracellular reactive oxygen species (ROS) level and accelerates migration, VM formation in colorectal cancer cells, while N-Acetylcysteine (NAC) could reverse these phenomena. Notably, further mechanical exploration confirmed that the matrix metalloprotease 2 (MMP2) induction is ROS dependent under hypoxic condition. On the basis, we found that AAM could effectively inhibit hypoxia-induced ROS generation, migration, VM formation as well as HIF-1α and MMP2 expression. In vivo, AAM significantly inhibits metastasis of colorectal cancer in murine lung-metastasis model. Taken together, these results verified that AAM effectively inhibits migration and VM formation by suppressing ROS/HIF-1α/MMP2 pathway in colorectal cancer under hypoxic condition, suggesting AAM could serve as a therapeutic agent to inhibit VM formation in human colorectal cancer.

Keywords: Astragalus Atractylodes mixture; colorectal cancer; hypoxia; migration; vasculogenic mimicry.

Figure 1

HPLC analysis of AAM extract. Base Peak chromatogram from HPLC Q-TOF/MS analysis of the ethanol extract of AAM (positive ions).

Figure 2

NAC inhibits colorectal cancer cell migration under hypoxia condition. Wound-healing assay for evaluating the migration ability of HCT-116 cells under normoxia (A) and hypoxia condition (B); cells were photographed at 0 and 24 h under 50× magnification. (C) The relative scratch area was measured by Image J (normoxia group, **P < 0.01; hypoxia group, ^#P < 0.05, ^##P < 0.01, ^###P < 0.001).

Figure 3

Hypoxia increases the ROS level in colorectal cancer cells and NAC could reverse this phenomenon. The (A) HCT-116 and (B) LoVo cells were cultured in hypoxic condition (1% O₂) and treated with NAC for 24 h, then treated with DCFH-DA (10 μM) for 30 min; the DCF fluorescence intensity was measured by flow cytometry.

Figure 4

Hypoxia accelerates migration and requires ROS generation. Transwell assay was used to evaluate the effect of hypoxia and ROS on migration ability. LoVo (A) and HCT-116 (B) cells were cultured in hypoxic condition (1% O₂) and treated with NAC for 24 h; cells on the lower surface of the Transwell membrane were stained with 0.1% crystal violet; the number of migrated cells was photographed under 50× magnification; Data are expressed as mean ± SD, n=3 (C, D). (*P < 0.05, **P < 0.01, ***P < 0.001).

Figure 5

Hypoxia accelerates vasculogenic mimicry and requires ROS generation. Vasculogenic mimicry of LoVo (A) and HCT-116 (B) cells was assayed after NAC treatment for 36 h at normoxia and hypoxia and photographed under 50× (LoVo) and 50× (HCT-116) magnification; data are represented as mean ± SD from at least three experiments. (*P < 0.05, **P < 0.01, ***P < 0.001).

Figure 6

Up-regulation of MMP2 under hypoxia requires ROS generation. (A, B) The expressions of EphA2, VE-cad, MMP2, MMP9 using western blotting analysis after the indicated treatment. Results showed that NAC markedly blocked hypoxia-stimulated MMP2 expression compared with the untreated group; (C, D) The expressions of MMP2 using RT-PCR after the indicated treatment (*P < 0.05).

Figure 7

AAM reduces ROS levels and inhibits proliferation in colorectal cancer cells. HCT-116 (A) and LoVo (B) cell proliferation was assayed at 24 h after treatment with AAM at different concentrations; cells were treated with 2.5 mg/ml AAM, and the ROS production of HCT-116 (C) and LoVo (D) cells was evaluated using flow cytometry (*P < 0.05, ***P < 0.001).

Figure 8

AAM inhibits migration of HCT-116 and LoVo cells. (A, B) Wound-healing assays of HCT-116 and LoVo cells. The cells were treated with medium alone, AAM (2.5 mg/ml), hypoxia medium, hypoxia medium with AAM (2.5 mg/ml) for 24 h. The representative images were shown in magnification of 100×. (C, D) Transwell assay was used to evaluate the effect of AAM on HCT-116 and LoVo cell migration. Migrated cells were stained and captured by microscope (50×); cell numbers are compared between groups. (*P < 0.05, **P < 0.01, ***P < 0.001).

Figure 9

AAM inhibits VM formation of HCT-116 and LoVo cells. VM formation assay. HCT-116 (A) or LoVo (B) cells were seed into 96-well plate which had been precoated with matrigel and incubated with AAM (2.5, 5, and 10 mg/ml) under normoxia and hypoxia. The representative images were shown in magnification of 50× (HCT-116) and 50× (LoVo).

Figure 10

AAM inhibits hypoxia-induced VM formation involving HIF-1α and MMP2. The HCT-116 (A) and LoVo (B) cells were treated with different concentrations of AAM under hypoxia and normoxia; the protein expression of HIF-1α and MMP2 was detected by western blotting. (C, D) The mRNA expression of HIF-1α and MMP2 was measured by RT-PCR (*P < 0.05, **P < 0.01, ***P < 0.001).

Figure 11

AAM inhibits metastasis of colorectal cancer in vivo. (A) Representative pictures of lungs of NAC, AAM treated and control groups; the number of metastatic tumor nodules on the surface of the lungs was counted (Right side). (B) HE staining was used to determine the metastatic lesions. (*P < 0.05).