Graded hypoxia modulates the invasive potential of HT1080 fibrosarcoma and MDA MB231 carcinoma cells.

Spatial and temporal oxygen heterogeneity exists in most solid tumour microenvironments due to an inadequate vascular network supplying a dense population of tumour cells. An imbalance between oxygen supply and demand leads to hypoxia within a significant proportion of a tumour, which has been correlated to the likelihood of metastatic dissemination in both rodent tumour models and human patients. Experimentally, it has been demonstrated that near-anoxic in vitro exposure results in transiently increased metastatic potential in some tumour cell lines. The purpose of this study was to examine the effect of graded low oxygen conditions on the invasive phenotype of human tumour cells using an in vitro model of basement membrane invasion, in which we measured oxygen availability directly at the invasion surface of the transwell chamber. Our results show a relationship between culture vessel geometry and time to achieve hypoxia which may affect the interpretation of low oxygen experiments. We exposed the human tumour cell lines, HT1080 and MDA MB231, to graded normobaric oxygen (5% O(2)-0.2% O(2)) either during or prior to in vitro basement membrane invasion to simulate conditions of intravasation and extravasation. A secondary aim was to investigate the potential regulation of matrix metalloproteinase activity by oxygen availability. We identified significant reductions in invasive ability under low oxygen conditions for the HT1080 cell line and an increase in invasion at intermediate oxygen conditions for the MDA MB231 cell line. There were differences in the absolute activity of the individual matrix metalloproteinases, MMP-2, -9, -14, between the two cell lines, however there were no significant changes following exposure to hypoxic conditions. This study demonstrates cell line specific effects of graded oxygen levels on invasive potential and suggests that intermediate levels of low oxygen may increase metastatic dissemination.