Halofuginone Induces Post-Transcriptional Down-Regulation of Cyclin D1, Cell Cycle Arrest and Apoptosis In Mantle Cell Lymphoma Cells through Activation of Integrated Stress Response Pathways

Abstract 773

Mantle cell lymphoma (MCL) remains an incurable disease and has the worst outcome among B-cell lymphomas. Patients generally have a good response to first line treatment but most relapse and tend to have shorter responses or resistant disease. Thus, novel treatment strategies capable of providing and sustaining durable responses are clearly needed. The translocation t(11;14), a hallmark of MCL, leads to cyclin D1 overexpression and is invariably accompanied by different secondary genetic lesions that collaborate for lymphomagenesis. In a previous study, we found that several genes related to the AKT, WNT and TGFβ signaling pathways were aberrantly expressed in MCL. The role of the AKT and WNT pathways in MCL pathogenesis has been well established by other groups, but little is known about the role of the TGFβ pathway. To address this issue, we tested whether halofuginone, a small molecule with recognized anti-TGFβ and antifibrotic activity, would have cytotoxic effect against a panel of MCL cell lines. We found that halofuginone at nanomolar levels had significant cytotoxic activity against MCL cell lines as measured by the MTT assay. The IC₅₀'s for Mino and HBL-2 cell lines were 30 and 61 ng/mL at 48h, respectively, with IC₅₀'s for Jeko-1, JVM-2 and Granta-519 falling in between. Halofuginone induced apoptosis in Mino and HBL-2 cells in a time- and concentration-dependent fashion, as evidenced by annexin V/7-AAD staining by flow cytometry and electron microscopy studies. However, halofuginone failed to inhibit SMAD2 phosphorylation induced by recombinant TGFβ1 in Mino and HBL-2 cells, as shown by Western blot analysis, and co-treatment experiments with TGFβ1 failed to show antagonism, suggesting that the effect of halofuginone in MCL is not mediated by TGFβ inhibition. Cell cycle analysis of Mino and HBL-2 cells exposed to halofuginone revealed time- and concentration-dependent accumulation in G1 (83% of Mino cells at G1 upon exposure to 50 ng/mL for 24h vs. 48% in untreated Mino cells), and immunocytochemical analysis showed that this effect was accompanied by striking down-regulation of cyclin D1 protein levels starting as early as 3h after exposure to halofuginone, a finding that was reproduced in primary MCL cells. Real-time RT-PCR experiments, however, revealed up-regulation of cyclin D1 mRNA levels by halofuginone over time, suggesting a post-transcriptional mechanism for the observed down-regulation of cyclin D1 protein levels. Western blot analysis of Mino and HBL-2 cells exposed to halofuginone for 24h showed a concentration-dependent phosphorylation of GCN2, PERK and EIF2α, and up-regulation of ATF4. These findings point to an activation of integrated stress response pathways (amino acid starvation response and endoplasmic reticulum stress response) that causes a general shutdown in protein synthesis and explain, at least partially, the down-regulation in cyclin D1 levels. To further characterize the proteins targeted by halofuginone in MCL we employed a proteomic profiling approach in which differentially expressed proteins were revealed by label-free liquid chromatography tandem mass spectrometry (LC-MSE) analysis on a nanoAcquity system coupled to a Synapt MS Q-Tof mass spectrometer. A comprehensive catalogue representing 147 proteins was generated from this analysis and we found that several members of the heat shock protein family are up-regulated in Mino cells exposed to 100 ng/mL of halofuginone for 14h, the relevance of which is currently under investigation. Together, our data demonstrate that halofuginone at nanomolar levels has significant antiproliferative and cytotoxic effects in MCL cells that are induced by the activation of integrated stress response pathways. More importantly, our study provides a rationale for exploring the clinical activity of this oral agent in patients with MCL.