Estrogen-receptor (ER) status is an important parameter in breast cancer management as ER-positive breast cancers have a better prognosis than ER-negative tumors. This difference comes essentially from the lower aggressiveness and invasiveness of ER-positive tumors. Here, we demonstrate, that interleukin-8 (IL-8) was clearly overexpressed in most ER-negative breast, ovary cell lines and breast tumor samples tested, whereas no significant IL-8 level could be detected in ER-positive breast or ovarian cell lines. We have also cloned human IL-8 from ERnegative MDA-MB-231 cells, and we show that IL-8 produced by breast cancer cells is identical to monocytederived IL-8. Interestingly, the invasion potential of ERnegative breast cancer cells is associated at least in part with expression of IL-8, but not with IL-8 receptor levels. Moreover, IL-8 increases the invasiveness of ER-positive breast cancer cells by two fold, thus confirming the invasion-promoting role of IL-8. On the other hand, exogenous expression of estrogen receptors in ERnegative cells led to a decrease of IL-8 levels. In summary, our data show that IL-8 expression is negatively linked to ER status of breast and ovarian cancer cells. We also support the idea that IL-8 expression is associated with a higher invasiveness potential of cancer cells in vitro, which suggests that IL-8 could be a novel marker of tumor aggressiveness.

Cathepsin-D is an independent marker of poor prognosis in human breast cancer. We previously showed that human wild-type cathepsin-D, as well as its mutated form devoid of proteolytic activity stably transfected in 3Y1-Ad12 cancer cells, stimulated tumor growth. To investigate the mechanisms by which human cathepsin-D and its catalytically-inactive counterpart promoted tumor growth in vivo, we quantified the expression of proliferating cell nuclear antigen, the number of blood vessels and of apoptotic cells in 3Y1-Ad12 tumor xenografts. We first verified that both human wild-type and mutated cathepsin-D were expressed at a high level in cathepsin-D xenografts, whereas no human cathepsin-D was detected in control xenografts. Our immunohistochemical studies then revealed that both wild-type cathepsin-D and catalytically-inactive cathepsin-D, increased proliferating cell nuclear antigen expression and tumor angiogenesis. Interestingly, wild-type cathepsin-D significantly inhibited tumor apoptosis, whereas catalytically-inactive cathepsin-D did not. We therefore propose that human cathepsin-D stimulates tumor growth by acting -directly or indirectly -as a mitogenic factor on both cancer and endothelial cells independently of its catalytic activity. Our overall results provide the first mechanistic evidences on the essential role of cathepsin-D at multiple tumor progression steps, affecting cell proliferation, angiogenesis and apoptosis.

No abstract