Triptolide abrogates growth of colon cancer and induces cell cycle arrest by inhibiting transcriptional activation of E2F

Abstract

Despite significant progress in diagnostics and therapeutics, over 50 thousand patients die from colorectal cancer annually. Hence, there is urgent need for new lines of treatment. Triptolide, a natural compound isolated from the Chinese herb Tripterygium wilfordii, is effective against multiple cancers. We have synthesized a water soluble analog of triptolide, named Minnelide, which is currently in phase I trial against pancreatic cancer. The aims of the current study were to evaluate whether triptolide/Minnelide is effective against colorectal cancer and to elucidate the mechanism by which triptolide induces cell death in colorectal cancer. Efficacy of Minnelide was evaluated in subcutaneous xenograft and liver metastasis model of colorectal cancer. For mechanistic studies, colon cancer cell lines HCT116 and HT29 were treated with triptolide and the effect on viability, caspase activation, annexin positivity, lactate dehydrogenase release, and cell cycle progression was evaluated. Effect of triptolide on E2F transcriptional activity, mRNA levels of E2F-dependent genes, E2F1- retinoblastoma protein (Rb) binding, and proteins levels of regulator of G1-S transition was also measured. DNA binding of E2F1 was evaluated by chromatin immunoprecipitation assay. Triptolide decreased colon cancer cell viability in a dose- and time-dependent fashion. Minnelide markedly inhibited the growth of colon cancer in the xenograft and liver metastasis model of colon cancer and more than doubles the median survival of animals with liver metastases from colon cancer. Mechanistically, we demonstrate that at low concentrations triptolide induces apoptotic cell death but at higher concentrations it induces cell cycle arrest. Our data suggest that triptolide is able to induce G1 cell cycle arrest by inhibiting transcriptional activation of E2F1. Our data also show that triptolide downregulates E2F activity by potentially modulating events downstream of DNA binding. Therefore, we conclude that Triptolide and Minnelide are effective against colon cancer in multiple pre-clinical models.

Figure 1. Triptolide and its derivative Minnelide is effective in killing colon cancer cells both in vivo and in vitro

Triptolide decreases the viability of colon cancer cell lines (A) HCT116 and (B) HT29 in a dose and time dependent manner. (C) In a subcutaneous xenograft model of colon cancer cell line HCT116 treatment with Minnelide leads to decreased growth of tumor when compared to vehicle (saline) treated animals alone. In a liver metastasis model of colon cancer, Minnelide treatment leads to significantly decreased (D) weight and (E) number of liver metastases. (F) Representative picture of a liver metastases from HCT 116 colon cancer cell line. In the left panel (10X) tumor liver interface is shown. In the right panel, high magnification view of the liver metastases is shown. (G) In a separate liver metastasis model of colon cancer focused on survival, Minnelide treatment significantly improved survival of animals with liver metastases when compared with vehicle treatment alone (p <0.001). (H) Triptolide does not affect the viability of human colonic epithelium cells (HCEC) significantly.

Figure 2. Triptolide treatment leads to activation of apoptosis in colon cancer cells

Triptolide treatment leads to caspase-3 activation in both (A) HCT116 and (B) HT29 colon cancer cell lines. However this activation is not dose dependent and increases up to 25nM in HCT116 and 50nM in HT29 and then decrease at higher concentrations. Similarly triptolide treatment leads to increased annexin-V staining in both (C) HCT116 and (D) HT29 colon cancer cell lines. As observed with caspase-3 activation, annexin-V staining also is not dose dependent and increased up to 25nM in HCT116 and 50nM in HT29 and then decreased at higher concentrations. Representative flow plot of annexin assay are provided in supplementary figure. Triptolide treatment does not lead to significant increase in necrosis, as measured by LDH release in (E) HCT116 colon cancer cells. In (F) HT29 colon cancer cell line, triptolide treatment led to a small but significant increase in LDH release at 12.5-50nM dosage at 12hr time point but not at 100-200nM dose or with any dose at 24h.

Figure 3. Triptolide treatment leads to G1-S arrest

Triptolide treatment leads to significantly increased proportion of cells in G1 fraction at a concentrations of 100-200nM when compared to untreated controls in both (A) HCT116 and (B) HT29 cells. (C) A representative run of cell cycle analysis for both cell lines is shown.

Figure 4. Triptolide inhibits transcriptional activity of E2F

(A) Triptolide treatment leads to decreased E2F transcriptional activity as measured by luciferase promoter binding assay in HCT116 cells. Triptolide treatment also leads to decrease in the mRNA levels of genes known to be transcriptionally regulated by E2F1, including (B) ORC-1, (C) CDC-2 and (D) cyclin A-1, in both HCT116 and HT29 cells.

Figure 5. Triptolide does not modulate E2F1-Rb binding

Triptolide treatment does not decrease the level of cyclin D or cyclin E, CDK 4 or CDK 6 or p15 or p21, proteins which regulate E2F1-Rb binding, in either (A) HCT116 or (B) HT29 colon cancer cells. The full length blots demonstrating the protein of interest and actin are shown in supplementary figure 3 and 4. Triptolide does not modulate binding of E2F1 and Rb as measured by co-immunoprecipitation in either (C) HCT116 or (D) HT29 cancer cells. E2F1 or Rb was immunoprecipitated from colon cancer cells lysates treated with triptolide and the immunoprecipitates were separated on SDS page and probed for both Rb and E2F1. Triptolide treatment for 12h increases binding of E2F1 to promoter region of E2F1 responsive genes (CDC-2) in both (E) HCT116 and (F) HT29 colon cancer cells.