Targeting Ras-Signalling Pathway to Strike Hsf1 and Induce Apoptosis in Chronic Lymphocytic Leukemia

Abstract INTRODUCTION The Heat Shock Protein of 70kDa (HSP70) and its transcription factor, the Heat Shock Factor 1 (HSF1), are two cytoprotective molecules that we found overexpressed and correlated to poor prognosis in Chronic Lymphocytic Leukemia (CLL) B cells. We focused on RAS-signalling pathways, which involve proteins known to regulate HSF1 activity, such as RAS/RAF/MEK/ERK and RAS/PI3K/AKT. Taking advantage from a previous RPPA analysis (Frezzato et al., 2016), we already developed a model according to which patients with high HSP70 protein levels are characterized by the activation of RAS/PI3K/AKT pathway, while patients with low levels of HSP70 have the RAS/RAF/MEK/ERK pathway activated. The dissection of these aberrant pathways in CLL would help in providing new information on the pathobiology of CLL B cells, in order to define innovative prognostic factors and/or new therapeutic targets. METHODS Freshly isolated leukemic B cells from 20 therapy-free CLL patients were cultured in RPMI 1640 supplemented with antibiotics and 2% FBS and treated with: 10, 20 and 30µM Pterostilbene, a natural analogue of Resveratrol, which upregulates ERK and downmodulates AKT with the final effect of inhibiting HSF1 activity. Apoptosis was evaluated after 24 hours by Annexin V/Propidium iodide flow cytometry test and by the presence of cleaved PARP in Western blotting (WB). HSP70 and HSF1 expression levels were evaluated by WB analysis in leukemic B cells after in vitro and in vivo inhibition with anti-PI3K. RESULTS We previously found that molecules acting as Resveratrol on RAS signaling pathways (inhibiting HSF1 by upregulating ERK and downmodulating AKT), induce apoptosis of CLL B cells in a dose-dependent manner. Particularly, we already observed apoptosis after treatment with Triacetyl Resveratrol and Honokiol. We recently extended our preliminary data on Pterostilbene to a large cohort of patients, obtaining the following results starting from 10µM to 20µM and 30µM: 63.50 ± 14.47%, 48.50 ± 21.59% and 24.88 ± 22.03% of living cells vs untreated cells, 74.63 ± 11.77% respectively, with p value p<0.05%, p<0.05, and p<0.01; paired Student's t Test. To determine the robustness of our model, we analysed the effect of Idelalisib, a PI3K inhibitor, on HSP70 and HSF1 protein levels. As expected, referring to our model, the inhibition of RAS/PI3K/AKT pathway, led to the down-modulation of HSF1 and HSP70 protein levels. We further investigated the effect of PI3K inhibition by collecting sample of 2 CLL patients before, during and after anti-PI3K regimen. We found out that HSP70 and HSF1 protein levels, analysed by WB, decreased in both patients and this decrease positively correlated with the response to therapy. CONCLUSIONS Our preliminary data suggest a pivotal role for HSP70 and HSF1 in CLL B cell protection against apoptosis. The dissection of RAS pathways in CLL and, in particular, the studies on their effect on the regulation of HSF1/HSP70 axis, are essential to understand the mechanisms that sustain the leukemic clone survival. The alternative and innovative approach we are developing to downmodulate HSP70 by targeting upstream-signalling molecules, will reduce the unpredictable off-target effects caused by the direct inhibition of the ubiquitous HSP70 protein and may contribute to find innovative targets for new therapeutic strategies for CLL. Disclosures Visentin: janssen: Consultancy, Honoraria. Trentin:Abbvie: Honoraria; Janssen: Research Funding; Gilead: Research Funding; Roche: Membership on an entity's Board of Directors or advisory committees.