Mechanisms of Sensitivity and Resistance of Primary Effusion Lymphoma to Dimethyl Fumarate (DMF)

Abstract

PEL is a rare B cell lymphoma associated with KSHV that mainly arises in immune-deficient individuals. The search for new drugs to treat this cancer is still ongoing given its aggressiveness and the poor response to chemotherapies. In this study, we found that DMF, a drug known for its anti-inflammatory properties which is registered for the treatment of psoriasis and relapsing-remitting MS, could be a promising therapeutic strategy against PEL. Indeed, although some mechanisms of resistance were induced, DMF activated NRF2, reduced ROS and inhibited the phosphorylation of STAT3 and the release of the pro-inflammatory and immune suppressive cytokines IL-6 and IL-10, which are known to sustain PEL survival. Interestingly, we observed that DMF displayed a stronger cytotoxic effect against fresh PEL cells in comparison to PEL cell lines, due to the activation of ERK1/2 and autophagy in the latter cells. This finding further encourages the possibility of using DMF for the treatment of PEL.

Keywords: ERK1/2; NRF2; PEL; ROS; STAT3; autophagy; inflammatory cytokines; mTOR/p-4EBP1.

Figure 1

DMF cytotoxicity in PEL cell lines is mediated by NRF2 activation and reduction of intracellular ROS and IL-6 and IL-10 release. (A,B) Percentage of cell viability of BC3 and BCBL1 cells, treated or not (CT) with 20 and 50 μM DMF for 24 h, or with 20 μM for 24 and 48 h, as evaluated by Trypan blue exclusion assay. (C) NQO1 and (D) p62/SQSTM (p62) expression levels in BC3 and BCBL1 cells, treated or not (CT) with 20 and 50 µM DMF for 24 h, assessed by western blotting. β-actin was used as loading control, and one representative experiment out of three is shown. The histograms represent the mean plus SD of the densitometric analysis of the NQO1/β-actin and p62/β-actin ratios of three different experiments. (E) p62 mRNA level of BC3 cells, treated or not (CT) with 50 µM DMF for 24 h, was quantitated by qRT-PCR. (F) FACS analysis using DC-FDA staining to measure the intracellular ROS in BC3 and BCBL1 cells, treated or not (CT) with 50 µM DMF for 24 h. (G) Western blotting showing MFN2 or (H) STAT3 phosphorylation (pSTAT3) in BC3 and BCBL1 cells, treated or not (CT) with 20 and 50 µM DMF for 24 h. β-actin was used as loading control, and one representative experiment out of three is shown. The histograms represent the mean plus SD of the densitometric analysis of the pSTAT3/STAT3 and STAT3/β-actin ratios of three different experiments. (I) IL-6 and IL-10 release by BC3 and BCBL1 cells, treated or not (CT) with 50 µM DMF for 24 h. (J) Percentage of cell viability of BC3 and BCBL1 cells, co-treated with 20 µM DMF and 5 nM Brusatol (BRU) for 24 h, evaluated by Trypan blue exclusion assay. In the figure, mean ± standard deviation (SD) of three independent experiments is shown. * p < 0.05.

Figure 2

mTOR pathway activation reduces DMF cytotoxicity in PEL cell lines. Western blot analysis was used to detect 4EBP1 phosphorylation (p-4EBP1) in BC3 and BCBL1 cells, (A) treated or not (CT) with 20 and 50 µM DMF for 24 h and (B) pre-treated with 100 nM mTOR inhibitor NVP-BEZ235 (BEZ) and then exposed to 20 µM DMF. β-actin was used as loading control, and one representative experiment out of three is shown. The histograms represent the mean plus SD of the densitometric analysis of the p-4EBP1/4EBP1and 4EBP1/β-actin ratios of three different experiments. (C) Percentage of cell viability of BC3 and BCBL1 cells, treated or not (CT) with 20 µM DMF and 100 nM of BEZ for 24 h, was evaluated by Trypan blue exclusion assay. (D) PARP cleavage (cl PARP) was assessed by western blotting in BC3 and BCBL1 cells, treated or not (CT) with 20 and 50 µM DMF for 24 h. β-actin was used as loading control, and one representative experiment out of three is shown. The histograms represent the mean plus SD of the densitometric analysis of the cl PARP/β-actin ratio of three different experiments. In the figure, mean ± standard deviation (SD) of three independent experiments is shown. * p < 0.05.

Figure 3

Lack of ERK1/2 activation underlies the higher cytotoxicity of DMF observed in PEL fresh tumors compared to BC3 PEL cell line. (A) IFA analysis to relieve LANA1 KSHV latent antigen (red) in PEL fresh tumors. DAPI (blue) was used to stain nuclei. (B) Percentage of cell viability of BC3 cells and PEL fresh tumors, treated or not (CT) with 20 and 50 µM DMF for 24 h, evaluated by Trypan blue exclusion assay. (C) PEL fresh tumors, treated or not (CT) with 20 and 50 µM DMF for 24 h, were analyzed by western blotting to assess p-4EBP1 level. β-actin was used as loading control, and one representative experiment out of three is shown. The histograms represent the mean plus SD of the densitometric analysis of the p-4EBP1/4EBP1and 4EBP1/β-actin ratios of three different experiments. (D) Western blotting of pERK1/2 in BC3 cells and PEL fresh tumors, treated or not (CT) with 20 and 50 µM DMF for 24 h. β-actin was used as loading control, and one representative experiment out of three is shown. The histograms represent the mean plus SD of the densitometric analysis of the pERK1/2/ERK1/2 and ERK1/2/β-actin ratios of three different experiments. (E) Percentage of cell viability of BC3 cells, treated with 50 µM DMF or not (CT) in the presence of 20 µM PD98059 EK1/2 inhibitor for 24 h. (F) PARP cleavage (cl PARP) evaluated by western blotting of BC3 cells, treated or not (CT) with 50 µM DMF for 24 h. β-actin was used as loading control, and one representative experiment out of three is shown. The histograms represent the mean plus SD of the densitometric analysis of the cl PARP/β-actin ratio of three different experiments. In the figure, mean ± standard deviation (SD) of three independent experiments is shown. * p < 0.05.

Figure 3

Lack of ERK1/2 activation underlies the higher cytotoxicity of DMF observed in PEL fresh tumors compared to BC3 PEL cell line. (A) IFA analysis to relieve LANA1 KSHV latent antigen (red) in PEL fresh tumors. DAPI (blue) was used to stain nuclei. (B) Percentage of cell viability of BC3 cells and PEL fresh tumors, treated or not (CT) with 20 and 50 µM DMF for 24 h, evaluated by Trypan blue exclusion assay. (C) PEL fresh tumors, treated or not (CT) with 20 and 50 µM DMF for 24 h, were analyzed by western blotting to assess p-4EBP1 level. β-actin was used as loading control, and one representative experiment out of three is shown. The histograms represent the mean plus SD of the densitometric analysis of the p-4EBP1/4EBP1and 4EBP1/β-actin ratios of three different experiments. (D) Western blotting of pERK1/2 in BC3 cells and PEL fresh tumors, treated or not (CT) with 20 and 50 µM DMF for 24 h. β-actin was used as loading control, and one representative experiment out of three is shown. The histograms represent the mean plus SD of the densitometric analysis of the pERK1/2/ERK1/2 and ERK1/2/β-actin ratios of three different experiments. (E) Percentage of cell viability of BC3 cells, treated with 50 µM DMF or not (CT) in the presence of 20 µM PD98059 EK1/2 inhibitor for 24 h. (F) PARP cleavage (cl PARP) evaluated by western blotting of BC3 cells, treated or not (CT) with 50 µM DMF for 24 h. β-actin was used as loading control, and one representative experiment out of three is shown. The histograms represent the mean plus SD of the densitometric analysis of the cl PARP/β-actin ratio of three different experiments. In the figure, mean ± standard deviation (SD) of three independent experiments is shown. * p < 0.05.

Figure 4

ERK1/2 activation triggers a pro-survival autophagy, reducing the DMF cytotoxicity in the BC3 cell line. (A) Western blot analysis to detect LC3II levels in BC3 cells and PEL fresh tumors, treated or not (CT) with 50 µM DMF for 24 h, in the presence of 20 µM chloroquine (CQ). β-actin was used as loading control, and one representative experiment out of three is shown. The histograms represent the mean plus SD of the densitometric analysis of the LC3II/β-actin ratio of three different experiments. (B) Percentage of cell viability of BC3 cells and PEL fresh tumors, treated or not (CT) with 50 µM DMF for 24 h, in the presence or absence of 20 µM CQ. (C) Western blot analysis to detect LC3II levels in BC3 cells, treated or not (CT) with 50 mM DMF for 24 h, in the presence or absence of 20 µM chloroquine (CQ) and PD98059 EK1/2 (PD) inhibitor. β-actin was used as loading control, and one representative experiment out of three is shown. The histograms represent the mean plus SD of the densitometric analysis of the LC3II/β-actin ratio of three different experiments. In the figure, mean ± standard deviation (SD) of three independent experiments is shown. * p < 0.05.