The paradigm-shifting idea and its practice: from traditional abortion Chinese medicine Murraya paniculata to safe and effective cancer metastatic chemopreventives

Abstract

Recent large epidemiological studies demonstrated benefit of oral contraceptives in reducing cancer risk, and our analysis also showed molecular and cellular similarities between embryo implantation and CTCs adhesion-invasion to endothelium. We here hypothesize that abortion traditional Chinese medicine (TCM) may serve well for pre-metastatic chemoprevention. To test the hypothesis, we selected the safe and well-known abortifacient TCM Murraya paniculata and identified a most-promising extracted fraction G (containing flavonoids and coumarins) from its many raw ethanol/dichloromethane extracts by using the bioactivity-guided fast screen assay. G showed free radical scavenging effect, and specifically inhibited both embryo implantation to human endometrial bed and cancer HT29 cells to human endothelium in a concentration-dependent manner (1-30 μg/mL) without significant cytotoxicity demonstrated by its high adhesion inhibition ratio. The inhibition may result from its down-regulation on expression of integrin β1 and α6, and CD44 on HT29 cells, as well as E-selectin on endothelial cells. Furthermore, G inhibited invasion and migration of HT29 cells. Pretreatment followed by one-month oral administration of G to the immunocompetent mice inoculated with mouse melanoma cells produced significant inhibition on lung metastasis without marked side effects. Collectively, this paradigm-shifting study provides, for the first time, a new strategy to discover safe and effective pre-metastatic chemopreventives from abortion TCM.

Keywords: Murraya paniculata; cancer metastatic chemoprevention; cell adhesion molecules; circulating tumor cells; traditional abortion Chinese medicine.

Figure 1. Schematic illustration of bioactivity-guided fast screen for cancer metastatic chemopreventive materials from raw extracts of M. paniculata.

(A) Photographs of typical M. paniculata. (B) Procedures of extracting bioactive materials from the plant twigs: the twigs were collected and smashed to pieces, and immersed and refluxed with 75% ethanol overnight. The concentrated residual was extracted with dichloromethane, which was concentrated and subjected to gradient silica chromatographic separation. (C) Bioactivity-guided fast screen for active raw materials: the concentrated eluates (fractions A-H) were diluted with cell medium, and tested for their inhibition against cancer cell viability (IC₅₀) and mobility (EC₅₀). The fast screen index was calculated by dividing IC₅₀ by EC₅₀ of each eluted fraction. The fraction with the highest index was selected, i.e., G, as the best candidate fraction for further composition analysis and in vitro and in vivo tests (D).

Figure 2. Phytochemical analysis of main components in fraction G following chromatographic separation

(A) Fluorescent characteristics of M. paniculata. dichloromethane-extracted fractions A-H; Note, the G fraction showed the highest fluorescent intensity. (B) Phytochemical color tests for possible constituents existed in fraction G; (C) DPPH free radical scavenging effect of G in comparison with ascorbic acid; (D) HPLC analysis showed two main components (a and b), in fraction G, and semi-preparative HPLC resulted b with satisfying purity. (E) Further structural analysis indicated the chemical structure of component b to be HMFRR.

Figure 3. Inhibition of fraction G extracted from M. paniculataon

biomimetic embryo implantation to endometrium. (A) Laser confocal microscopy imaging showed the significant inhibition of fraction G (60 μg/mL) on spheroid outgrowth of human placental choriocarcinoma cells JEG-3 spheroids co-cultured with confluent monolayered human endometrial cells RL95-2 for 24 h. (B) Concentration–dependent inhibition of fraction G on embryo implantation (spheroid outgrowth of JEG-3 cells to RL95-2 cells). *P < 0.05, and **P < 0.01, compared with the control.

Figure 4. Low cytotoxicity of fraction G and its concentration-dependent (0, 1, 10, 30 μg/mL) inhibition on adhesion of cancer cell HT29 to human HUVECs and Fn-coated matrix

There was no significant effect of G on HT29 cell cycle distribution (A) and apoptosis (B). However, G produced dose-dependent inhibition on Rhodamine 123-labeled HT29 cells adhered to the HUVEC monolayer stimulated by IL-1β (1 ng/mL) (C). (D) Quantitative inhibition of G on adhesion between HT29 cells and HUVEC monolayer. (E) Quantitative inhibition of G on adhesion of HT29 cells to Fn-coated matrix. (F) AIR indicated that G specifically inhibited hetero-adhesion between HT29 cells and HUVECs or Fn-coated matrix; the larger the AIR is, the more likely the drug works as an adhesion inhibitor. **P < 0.01, compared to the control.

Figure 5. Effects of G on mobility and invasion capacity of HT29 cells

(A) Representative images of the transwell invasion of HT29 cells treated with G (0, 1, 10 and 30 μg/mL) for 24 h. (B) Quantitative analysis of the effect of G on cell invasion. (C) Representative images of the scratch mobility of HT29 cells treated with G (0, 1, 10, 30 μg/mL) for 24 h. (D) Quantitative analysis of the effect of G on cell mobility. *P < 0.05; and **P < 0.01, compared with the control.

Figure 6. Concentration-dependent effect of G on expression of integrin β1 (CD29), integrin α6 (CD49f), and CD44 on HT29 cells (A–C, E), as well as expression of E-selectin (CD62E) on HUVECs (D, F) **P < 0.01, compared with the control

Figure 7. Effect of oral G (100 mg/kg/day for 30 days) on development of lung metastasis induced by mouse B16-F10 melanoma cells inoculated (iv) into the immunocompetent mice

Images of lung metastasis (A) and typical lung histology (B) in the control (up) and experimental (down) groups (n = 10/group). Effect of G on mouse lung weight (C) and tumor nodule number (D). **P < 0.01, compared with the control.