INTRODUCTION

We recently found that the Heat Shock Protein of 70kDa (HSP70), an ATP-dependent chaperone that is induced by cellular stress and protects cells against various apoptotic stimuli, was particularly overexpressed in neoplastic B cells from Chronic Lymphocytic Leukemia (CLL) vs normal B lymphocytes. HSP70 responds to a wide variety of physiological and environmental stress signals, thus allowing cells to survive to lethal conditions. The primary responsible for the transcription of HSP70 is the heat shock factor 1 (HSF1), being the major regulator of HSP70 expression. In response to stress, HSF1 becomes phosphorylated, forms homotrimers, binds DNA and activates heat shock gene transcription.

Considering that the search for molecules involved in the apoptosis resistance and increased survival of B cells from CLL is still ongoing, with this as a background, we were aimed at studying and targeting HSP70 or players related to it (i.e. HSF1) in view of their clinical, prognostic and therapeutical relevance in CLL.

METHODS

HSP70/HSF1 axis was analysed in freshly isolated leukemic B cells from CLL patients. Expression levels of HSP70, HSF1 and HSF1-Ser326 were assessed by Western blotting analysis with specific antibodies and the obtained expression data have been correlated with clinical features of the patients. HSP70 subcellular localization has been determined by confocal microscopy and cell fractionation. HSP70 expression and localization was also assessed by immunohistochemistry in lymph nodes from CLL patients. Leukemic B cells from 15 CLL therapy-free patients were treated with different concentrations of: i) Zafirlukast, an oral leukotriene receptor antagonist used to prevent asthma symptoms and acting also as HSP70 inhibitor and ii) Fisetin, a dietary flavonoid acting as anti-inflammatory and anti-carcinogen, that inhibits HSF1 activity through the block of its binding to the HSP70 promoter. Apoptosis induction in CLL cells was evaluated by Annexin V/Propidium Iodide flow cytometry test and by the presence of cleaved PARP observed in Western blotting.

RESULTS

We found that HSP70 and HSF1 proteins were overexpressed in leukemic vs normal B cells and correlated to poor prognosis. In particular, IGHV unmutated or ZAP70 positive patients presented higher levels of HSP70 and HSF1 with respect to patients with a favorable prognosis. Moreover, the two proteins presented a positive correlation (p<0.0001, r=0.84; Pearson's correlation) thus hypothesizing a positive loop feedback for their expression. We found that, in CLL, HSF1 was constitutively phosphorylated at activatory Ser326, thus being positively regulated, in a large part of our patients. In addition, patients presenting a higher phosphorylation of HSF1 at Ser326 were mostly ZAP70 positive patients. We also observed an abnormal constitutive nuclear localization of HSP70 in leukemic cells. On the basis of these results and the pro-survival role played by HSP70 and HSF1, we analyzed the effects of their inhibition in leukemic cells of our patients by using two inhibitors of this axis, Zafirlukast and Fisetin. Both inhibitors have been proven to be effective in inducing a dose-dependent cell apoptosis in CLL B cells.

CONCLUSIONS

HSP70 overexpression is involved in a diminished response to treatment by promoting the adaptation of tumor cells to changes (i.e. toxic conditions) currently induced by chemotherapy thus revealing critical roles for HSP70 in cancer initiation and progression. It has been shown that HSP70 depletion results in an increased sensitivity to chemotherapy. For this reason, and considering its prognostic implications and functional role in cancers, including CLL, HSP70 represents an interesting target for antileukemic therapies.

In this context, our results suggest: i) an involvement of HSP70/HSF1 axis in the pathogenesis of CLL; ii) an input for further studies that consider the possible involvement of ZAP70 in HSP70/HSF1 axis in CLL; iii) the putative usage of Zafirlukast, which is a drug already available for clinical use and in the targeting of HSF1 in CLL.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution