Abstract
Among new treatment options for mantle cell lymphoma (MCL), the targeted drug lenalidomide appears as one of the most efficient molecules. Lenalidomide has multiple modes of action targeting the tumor cell and its environment including the immune system. It is widely reported that cancer patients are deficient in vitamin D3 (1,25-dihydroxyvitamin D3, VD3) and recent studies have shown in non Hodgkin lymphomas (NHL) that VD3 levels have a prognostic value on survival (Drake, J Clin Oncol. 2010;28:4191). While the relations between VD3 and cancer incidence remain unresolved, it has been shown that VD3 displays anti-tumoral properties via its anti-proliferative, pro-differentiation, anti-inflammatory and anti-angiogenic properties. We assessed the efficacy of VD3 to potentiate cell death induced by lenalidomide in MCL cell lines and patients’ samples and explored the mechanisms of cell death in this context. Experiments were conducted on a panel of 6 MCL cell lines (JEKO-1, MINO, GRANTA-519, UPN-1, REC-1 and Z138) and 8 primary peripheral blood samples.
After 6 days of treatment, MCL cells were weakly sensitive to low doses of lenalidomide (1µM and 10µM for cell lines and samples, respectively). Addition of physiological doses of VD3 (100nM) significantly and synergistically increased cell death in 67% of cell lines (Z138, JEKO-1, MINO, REC-1) and in 63% of primary samples (p<0.05). However resistance to lenalidomide alone was not reversed by VD3 since both GRANTA-519 and UPN-1 remained unsensitive. Apoptosis, characterized by Annexin V staining, appearance of a subG1 peak and caspase 9 activation, was dependent on Bax expression, since transient extinction of BAX by siRNA in JEKO-1 cells inhibited cell death (mean of inhibition 30%±5%, p=0.03). The combination of lenalidomide and VD3 dramatically increased expression of the BH3-only Bik (Bcl2-Interacting Killer) protein in sensitive (Z138, JEKO-1, MINO, REC-1) but not resistant (GRANTA-519, UPN-1) cell lines, without affecting the expression of other molecules of the Bcl2 family. By immunoprecipitation assays, we showed that induced-Bik was not bound to the anti-apoptotic molecules Bcl2, BclxL or Mcl1 in treated cells but was free to activate such pro-apoptotic molecules as Bax. Moreover, siBIK RNA significantly decreased the proportion of Annexin V+ cells observed after treatment with lenalidomide and VD3, respectively by 36%±9% (p=0.04) and 28%±4% (p=0.04) in JEKO-1 and MINO cells. This confirmed the involvement of Bik in the cell death induced by this synergistic combination. Q-RT-PCR assays disclosed that Bik accumulation was related to an increase in BIK mRNA expression. BIK expression is controlled by the transcription factor TEF and is regulated by epigenetic modifications, its expression being silenced by methylation in many cancer cells. We showed that Bik accumulation induced by lenalidomide and VD3 was not related to an increase in TEF expression. To determine whether Bik expression could be induced or increased upon demethylation in MCL, we treated cell lines for 3 days with 1µM 5-azadecytidine (5-aza). Indeed, higher expression of Bik was observed after this treatment in the four cell lines sensitive to lenalidomide. Of note, cell death induced by 5-aza correlated linearly to that induced by lenalidomide and VD3 (p<0.001, r=0.95, n=6), suggesting that BIK demethylation could be a key point in the response to this combination. To directly assess the level of BIK methylation in MCL cell lines, we then performed a DNA methylation specific PCR assay on bisulfite-treated DNA, which targets the CpG rich region located within intron 1, as previously described by Hatzimichael et al (Leuk Lymphoma. 2012;53:1709). Indeed, we showed that lenalidomide and VD3 increased the proportion of unmethylated over methylated BIKDNA CpG islands in sensitive (2 to 5-fold increase) but not in resistant cell lines.
These data show that the association of lenalidomide and VD3, by increasing BIK expression through DNA demethylation, is an efficient combination to induce the apoptosis of MCL cells. They also underline the interest of measuring the level of VD3 in MCL patients especially those receiving lenalidomide, since supplementation in deprived patients might improve the effect of therapy.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.