Search

Previously, we showed that mice treated with cyclophosphamide (CTX) 4 days before intravenous injection of breast cancer cells had more cancer cells in the lung at 3 h after cancer injection than control counterparts without CTX. At 4 days after its injection, CTX is already excreted from the mice, allowing this pre-treatment design to reveal how CTX may modify the lung environment to indirectly affect cancer cells. In this study, we tested the hypothesis that the increase in cancer cell abundance at 3 h by CTX is due to an increase in the adhesiveness of vascular wall for cancer cells. Our data from protein array analysis and inhibition approach combined with in vitro and in vivo assays support the following two-prong mechanism. (1) CTX increases vascular permeability, resulting in the exposure of the basement membrane (BM). (2) CTX increases the level of matrix metalloproteinase-2 (MMP-2) in mouse serum, which remodels the BM and is functionally important for CTX to increase cancer abundance at this early stage. The combined effect of these two processes is the increased accessibility of critical protein domains in the BM, resulting in higher vascular adhesiveness for cancer cells to adhere. The critical protein domains in the vascular microenvironment are RGD and YISGR domains, whose known binding partners on cancer cells are integrin dimers and laminin receptor, respectively.

Previously, we showed that chemotherapy paradoxically exacerbated cancer cell colonization at the secondary site in a manner dependent on Atf3, a stress-inducible gene, in the non-cancer host cells. Here, we present evidence that this phenotype is established at an early stage of colonization within days of cancer cell arrival. Using mouse breast cancer models, we showed that, in the wild-type (WT) lung, cyclophosphamide (CTX) increased the ability of the lung to retain cancer cells in the vascular bed. Although CTX did not change the WT lung to affect cancer cell extravasation or proliferation, it changed the lung macrophage to be pro-cancer, protecting cancer cells from death. This, combined with the initial increase in cell retention, resulted in higher lung colonization in CTX-treated than control-treated mice. In the Atf3 knockout (KO) lung, CTX also increased the ability of lung to retain cancer cells. However, the CTX-treated KO macrophage was highly cytotoxic to cancer cells, resulting in no increase in lung colonization—despite the initial increase in cell retention. In summary, the status of Atf3 dictates the dichotomous activity of macrophage: pro-cancer for CTX-treated WT macrophage but anti-cancer for the KO counterpart. This dichotomy provides a mechanistic explanation for CTX to exacerbate lung colonization in the WT but not Atf3 KO lung.