Cantharidin (Spanish Fly) is a natural toxin and an inhibitor of protein phosphatases 1 (PP1) and 2A (PP2A), which have key roles in cell cycle progression. We have synthesised two series of demethylated cantharidin analogues, one displaying an open-ring lactone configuration in solution (Novo-1 to Novo-5) similar to cantharidin, the other showing a closed-ring lactone configuration (Novo-6 to Novo-10). In the present study, these ten agents were screened for in vitro PP1 and PP2A inhibition and cellular cytotoxicity in nine cancer cell lines of haematopoietic (L1210, HL60), ovarian (A2780, ADDP), osteo (143B), and colon (HCT116, HT29, WiDr, SW480) origin and one normal colon cell line (CCD-018). The open-ring series (IC50, PP1=2.0-4.8 muM, PP2A=0.2-0.5 muM) maintained the PP2A selectivity of cantharidin (IC50, PP1=1.8 muM, PP2A=0.2 muM), although some were less potent. The closed-ring series (IC50, PP1 = 12.5->1000 muM, PP2A=5->1000 muM) were considerably less potent inhibitors, confirming the need of ring opening for inhibition. The cytotoxicity (IC50, 72 h, MTT assay) of cantharidin ranged from 6-15 muM, while the new analogues ranged from 14 to >1000 muM. Cytotoxicity of the agents did not consistently parallel the in vitro potency of protein phosphatase inhibition. A number of analogues showed colon cancer selectivity, particularly Novo-6, where the cytotoxicity ranged from 14-88 muM in the colon cancer cells and 275-680 muM in all other cell lines including normal colon cells. The reason for this selectivity was not apparent and may involve additional intracellular targets. Cell cycle analysis showed cantharidin to enhance cell cycle progression as evident from an increased S-phase population and enhanced DNA synthesis, culminating in G(2)/M arrest and apoptosis. With Novo-1 and Novo-6, the cell cycle changes paralleled the cytotoxicity responses, with the predominant effect of G(2)/M cell cycle arrest followed by cell death. In conclusion, we have synthesised new anticancer agents that show selective cytotoxicity in colon cancer cells while remaining inactive in normal colon cells, and which mediate their effects via the G(2)/M phase of the cell cycle.