Development of less toxic analogues of fascaplysin as novel therapeutic agents for the treatment of cancer

Cyclin-dependent kinase 4 (Cdk4) represents a crucially important target for the treatment of cancer because most human cancers are characterised by over-expression of its activating partner cyclin Dl (product of an oncogene), loss of the natural Cdk4 specific inhibitor p 16 INK4A or mutation in Cdk4' s catalytic subunit. Even the target of Cdk4, pRb, is found to be functionally inactive in many human tumours. All of these aberrations can cause deregulated cell growth, resulting in tumour formation. Therefore, a small synthetic molecule or a natural product that specifically and potently inhibits the kinase activity of Cdk4 in vitro and prevents cell growth and tumour volume in vivo could be of immense therapeutic value for the treatment of cancer. Fascaplysin is a marine natural product which is a potent Cdk4-specific inhibitor. However, fascaplysin also intercalates DNA resulting in S phase block in the cell division cycle and unusual toxicity at the cellular level. The purpose of these studies was to develop a less toxic analogue of fascaplysin that abolishes fascaplysin's DNA binding capacity but maintains its potency to inhibit Cdk4 so as to obtain a novel therapeutic agent for the treatment of cancer. The newly synthesized compounds (based on the structure of fascaplysin) were screened in five different Cdk enzyme assays: Cdk4-cyclin Dl, Cdk2-cyclin A, Cdk2-cyclin E, Cdkl-cyclin B1 and Cdk9-cyclin Tl. The results from these assays led to the identification of compounds that specifically inhibit Cdk4 enzyme activity in vitro (i.e. compounds that are, at least, 10-fold more potent in inhibiting Cdk4 than Cdk2, Cdkl or Cdk9). Interestingly, none of fascaplysin's analogues showed any interaction or intercalation with double-stranded DNA, proving that fascaplysin's ability to inhibit Cdk4 specifically can be separated from its deleterious DNA intercalating characteristic. These structural analogues of fascaplysin were tested for their ability to inhibit cancer cell growth in vitro. We have succeeded in identifying novel molecules (from the screening of four different series of compounds) which inhibit the growth of cancer cells in vitro at low micro-molar concentrations. The most active compounds CAI 99, CA224, AJW089 and DE002 inhibited growth of cancer cells at an average IC50 concentration of 7, 4.2, 2 and 0.74 µM respectively. The effects on the long term survival (colony forming ability) of cancer cells in vitro followed by the treatment with the compounds were also evaluated. Colony forming ability usually reflects an anticancer compound's efficacy in viva. It was found that these compounds significantly reduce the colony forming ability of cancer cells in vitro. Compounds showing potent anti-proliferative effects were studied further for their effects on the cell division cycle. To our surprise, some of the analogues manifested not only the expected Cdk4-specific inhibition by blocking at the G0/G1 phase of the cell cycle but also exhibited a G2/M phase block in a Cdk-independent manner. The unexpected G2/M block posed many questions regarding the possible dual mechanism of action of these compounds. Interestingly, we found that the fascaplysin analogues that show profound block at G2/M phase also inhibit tubulin polymerization both in an in vitro cell-free assay as well as in a cell-based assay. Further experiments were performed to evaluate the effect of these compounds on the expression levels of the tumour suppressor proteins pRb and p53 by Western blotting. It was found that CA199 and CA224 enhanced the expression levels of p53 but did not alter the expression levels of the pRb protein. CA 199 and CA224 were also found to induce the levels of cell cycle inhibitory proteins p27KJPJ and p21CIPJ/WAFJ _ Western analyses to check the phosphorylation status ofpRb protein at Cdk4 specific sites (Ser780, Ser795 and Ser8071811) indicated that the block of cancer cell growth at G0/G1 was mediated by Cdk4. Although most of the synthesised molecules are Cdk4-specific, yet they do not manifest their cellular activities only via Cdk4. Intriguingly, we confirm that some selected compounds manifest their cellular action by targeting multiple sites in the cancer cell division cycle. We also show that two of these selected compounds are highly efficacious in inhibiting tumour growth in human xenograft-SCID mice models. In animal studies, CA224 and DE002 significantly inhibited the in viva tumour growth ofHCTl 16 (human colon cancer) and NCI-H460 (human lung cancer).