The chaperone Hsp90 is required for the correct folding and maturation of certain "client proteins" within all cells. Hsp90-mediated folding is particularly important in cancer cells, because upregulated or mutant oncogenic proteins are often Hsp90 clients. Hsp90 inhibitors thus represent a route to anticancer agents that have the potential to be active against several different types of cancer. Currently, various Hsp90 inhibitors that bind to Hsp90 at its ATP-binding site are in preclinical and clinical trials. Some of the most promising Hsp90 ATP-binding site inhibitors are the well characterized geldanamycin derivative 17-AAG and the recently described compounds PU-H71 and NVP-AUY922. An undesirable characteristic of these compounds is the transcriptional upregulation of Hsp70 that has prosurvival effects. Here we characterize the activity of a new type of chaperone inhibitor, 1,6-dimethyl-3-propylpyrimido[5,4-e][1,2,4]triazine-5,7-dione (named C9 for simplicity). Using purified protein components in vitro, C9 prevents Hsp90 from interacting with the cochaperone HOP and is thus expected to impair the Hsp90-dependent folding pathway in vivo. We show that this compound is effective in killing various breast cancer cell lines including the highly metastatic MDA-MB-231. An important property of this compound is that it does not induce the transcriptional upregulation of Hsp70. Moreover, when cells are treated with a combination of C9 and either 17-AAG or NVP-AUY922, the overexpression of Hsp70 is counteracted considerably and C9's lethal-IC50 decreases compared to its value when added alone.