OBJECTIVETo investigate the ability of invasion and migration of breast cancer MDA-MB-231 cells under serum starvation and hypoxia, and the effect of antiangiogenic drugs, rh-endostatin and bevacizumab, on the ability of invasion and migration of breast cancer cells under serum starvation and/or hypoxia, in order to explore the potential risk of antiangiogenic therapy in clinics.

METHODSThe cells were randomized into 4 groups, i.e., group A: 10% fetal bovine serum (FBS) group; group B: hypoxia + 10% FBS group; group C: serum starvation group; group D: hypoxia + serum starvation group; each group was further divided into three subgroups as blank control, treated with rh-endostatin and bevacizumab, respectively. Cell counting kit-8 (CCK-8) was used to assess the inhibition rate of cell growth induced by endostatin and bevacizumab, in order to determine the proper working concentration and time of the two drugs. Transwell assay was conducted to detect the cell invasion and migration in vitro. The expressions of c-Met and MMP-9 were detected by Western blot. The cells treated with rh-endostatin or bevacizumab under serum starvation were tested by hybridization using Exiqon miBase 18.0 microarray. The miRNAs which exibited significant differences (P < 0.05) in miRNA hybridization were verified by real-time PCR assay.

RESULTSCCK-8 assay showed that the inhibition rates of MDA-MB-231 cells cultured with 800 mg/L rh-endostatin for 48 h and 24 h were (32.2 ± 2.5)% and (27.0 ± 1.3)%, respectively, showing a significant difference (P = 0.023). The inhibition rates of MDA-MB-231 cells cultured with 80 mg/L bevacizumab for 48 h and 24 h were (30.5 ± 1.4) % and (26.1 ± 2.4) %, respectively, showing also a significant difference (P = 0.015). The Transwell assay showed that in the starvation blank group, the number of invaded and penetrated cells were 28.8 ± 2.2 and 31.4 ± 1.5, respectively, significantly different from that in the rh-endostatin and bevacizumab groups (P < 0.05). The relative expressions of c-Met and MMP-9 were 0.213 ± 0.017 and 0.542 ± 0.048, respectively, with a significant difference from those of the groups treated with each drug (P < 0.05 for both). The numbers of penetrated cells in the Transwell assay treated with rh-endostatin in hypoxia were 17.5 ± 2.1 and 16.5 ± 2.8, respectively, and the numbers of penetrated cells in the Transwell assay treated with bevacizumab were 16.3 ± 3.5 and 17.5 ± 2.4, respectively, showing no significant difference among them (P > 0.05 for both). The ability of migration and invasion of MDA-MB-231 cells and the expression of c-Met and MMP-9 were not impacted by hypoxia (P > 0.05). Real-time PCR assay showed that only the levels of miR-2355 and miR375 were significantly and stably decreased in the cells which had increased ability of invasion and migration. The relative expression levels of miR375 and miR-2355 in the serum starvation blank group were 0.550 ± 0.036 and 0.852 ± 0.121, respectively, significantly lower than that in the groups treated with rh-endostatin or bevacizumab (P<0.05). In the serum starvation group, the expression levels of miR375 and miR-2355 of cells treated with rh-endostatin were 0.295 ± 0.012 and 0.253 ± 0.011, and the expression levels of cells treated with bevacizumab were 0.234 ± 0.020 and 0.309 ± 0.022, respectively, (P > 0.05 for all). Compared with the serum starvation blank group, the expression levels of miR2355 and miR375 were significantly decreased when cells were treated with rh-endostatin/bevacizumab under serum starvation, but no significant difference was found between the two drugs (P > 0.05). However, hypoxia did not affect the expressions of miR2355 and miR375 (P > 0.05).

CONCLUSIONSThe results of this study suggest that serum starvation can increase the ability of invasion and migration of breast cancer cells. Furthermore, both rh-endostatin and bevacizumab may enhance their invasion and penetration ability under serum starvation condition.