DZNep-mediated apoptosis in B-cell lymphoma is independent of the lymphoma type, EZH2 mutation status and MYC, BCL2 or BCL6 translocations

Abstract

Enhancer of zeste homolog 2 (EZH2) tri-methylates histone 3 at position lysine 27 (H3K27me3). Overexpression and gain-of-function mutations in EZH2 are regarded as oncogenic drivers in lymphoma and other malignancies due to the silencing of tumor suppressors and differentiation genes. EZH2 inhibition is sought to represent a good strategy for tumor therapy. In this study, we treated Burkitt lymphoma (BL) and diffuse large B-cell lymphoma (DLBCL) cell lines with 3-deazaneplanocin-A (DZNep), an indirect EZH2 inhibitor which possesses anticancer properties both in-vitro and in-vivo. We aimed to address the impact of the lymphoma type, EZH2 mutation status, as well as MYC, BCL2 and BCL6 translocations on the sensitivity of the lymphoma cell lines to DZNep-mediated apoptosis. We show that DZNep inhibits proliferation and induces apoptosis of these cell lines independent of the type of lymphoma, the EZH2 mutation status and the MYC, BCL2 and BCL6 rearrangement status. Furthermore, DZNep induced a much stronger apoptosis in majority of these cell lines at a lower concentration, and within a shorter period when compared with EPZ-6438, a direct EZH2 inhibitor currently in phase II clinical trials. Apoptosis induction by DZNep was both concentration-dependent and time-dependent, and was associated with the inhibition of EZH2 and subsequent downregulation of H3K27me3 in DZNep-sensitive cell lines. Although EZH2, MYC, BCL2 and BCL6 are important prognostic biomarkers for lymphomas, our study shows that they poorly influence the sensitivity of lymphoma cell lines to DZNep-mediated apoptosis.

Fig 1. Effect of mutated EZH2 on H3K27me3 protein expression in B-cell lymphoma.

(A) Western blot performed using total protein lysates from EZH2-mutated (SU-DHL-10, WSU-DLCL-2, KARPAS-422, Carnaval) and wild-type EZH2 (HT, U-2932 R2, U-2932 R1, DG-75, CA-46, BLUE-1, BL-2 and BL-41) cell lines showing an elevated H3K27me3 in EZH2-mutated cell lines as compared to wild-type EZH2 cell lines. (B) Densitometric quantification of EZH2 and H3K27me3 normalized for GAPDH.

Fig 2. Apoptosis mediated by DZNep is concentration-dependent and time-dependent.

(A) Flow cytometry and Western blot analysis showing apoptosis of 3 susceptible cell lines treated with increasing concentrations of DZNep for 72 hours. Flow cytometry data were obtained from 3 biological replicates and are shown as mean plus standard deviation (SD). (B) Lymphoma cell lines were treated with 5 μM DZNep and apoptosis was measured at 24 hours, 48 hours and 72 hours. The values shown are those normalized with that of the respective controls, and represent the mean plus standard deviation from triplicate measurements.

Fig 3. Impact of DZNep treatment on proliferation and apoptosis.

(A) Cell lines were seeded at a density of 2 x 10⁵ vital cells / ml. They were either untreated or treated with 5 μM DZNep on day 0. The number of vital cells was determined after 24 hours, 48 hours and 72 hours. The number of vital cells was recorded after exclusion of the apoptotic / dead cell population (annexin V/PI positive cells) by flow cytometry. (B) Cell lines were either untreated or treated with 5 μM DZNep for a duration of 72 hours. The percentage of apoptotic cells was measured using flow cytometry. (C) IC₅₀ determination and grouping of the cell lines into DZNep-sensitive, an intermediate and DZNep-resistant. For Fig 3A, 3B and 3C, data is shown as mean plus SD, n = 3 biological replicates.

Fig 4. Grouping of wild-type and EZH2 mutated lymphoma cell lines based on their response to apoptosis caused by DZNep.

Fig 5. DZNep induces stronger apoptosis at a lower concentration and within a shorter period than EPZ-6438.

(A) One DZNep-resistant wild-type EZH2 cell line (HT), two DZNep-sensitive EZH2-mutated cell lines (Carnaval and KARPAS-422) and one DZNep-sensitive wild-type EZH2 cell line (BL-41) were treated with 5 μM DZNep and 5 μM EPZ-6438 for 3 days. Apoptosis was measured afterwards by flow cytometry. Data was analyzed using GraphPad Prism 5 software (GraphPad Software, California, USA) and statistical significance was determined using the Mann-Whitney U test (n = 3 biological replicates). (B) The same cell lines as were treated with 5 μM and 10 μM EPZ-6438 for up to 13 days. Apoptosis measurement, splitting the cells, and medium change were done on day 3, day 6, day 10 and day 13. Fig 5A and 5B are shown as mean plus SD from three biological replicates.

Fig 6. Apoptosis caused by DZNep occurs via EZH2 inhibition and downregulation of H3K27me3.

(A) Downregulation of EZH2 and H3K27me3 in three DZNep-sensitive lymphoma cell lines treated with increasing concentration of DZNep. Western blot was performed using total protein lysates from the cell lines. (B) Western blot showing the treatment of two EZH2-mutated (WSU-DLCL-2 and Carnaval) and two wild-type EZH2 (BL-2 and BLUE-1) cell lines with DZNep.