Fig 1.
Positive correlations were observed between the abundance of LMP1 and GLUT1 in malignant cells and the number of CD33+ MDSCs in NPC tissues.
(A) Representative IHC staining for LMP1 and GLUT1 in NPC specimens. (B-D) Statistical analyses of the correlations between LMP1 expression and CD33+ MDSC number (B), GLUT1 expression and CD33+ MDSC number (C), and LMP1 and GLUT1 expression level (D), Pearson’s chi-square test was used to assess statistical significance. (E-F) Kaplan-Meier estimates of the progression-free interval were determined based on the median levels of LMP1 (E) and GLUT1 (F) in the patients with NPC. P < 0.05 was considered statistically significant.
Fig 2.
LMP1 increased glycolysis and the expression of glycolysis-related genes in NPC cells.
(A) Exogenous EBV oncoprotein LMP1 was stably overexpressed in the NPC cell lines CNE2 and TW03 following lenti-LMP1 transfection, and the lenti-vector was included as control. WB showed that the expression of LMP1 and GLUT1 was increased in CNE2-LMP1 and TW03-LMP1 cells compared with that in CNE2-vector and TW03-vector cells, and GAPDH was included as a control. One representative experiment of five independent experiments is shown. (B-D) ECAR assay results. CNE2-vector, CNE2-LMP1, TW03-vector and TW03-LMP1 cells were cultured in base DMEM media with no glucose or glutamine. ECAR was assessed after the addition of 10 mM glucose and in response to the metabolic inhibitor oligomycin and the glucose inhibitor 2-DG. The time course and calculations for (B) glycolytic capacity and (C) statistical analysis of glycolytic capacity and glycolytic reserve are shown. (D) The relative mRNA levels of glycolysis-related genes, including GLUT1, HK2, GPI, PFKB1, 2, 3 and 4, ALDOA, PGK1, PKM2 and LDHA, in CNE2-vector, CNE2-LMP1, TW03-vector and TW03-LMP1 cells were measured via real-time qRT-PCR. Statistical results are representative of at least three experiments performed in triplicate. All values are shown as the means ± SEM; * indicates P < 0.05. ** indicates P < 0.01.
Fig 3.
LMP1 promoted NPC-induced MDSC differentiation and the expression of MDSC-related molecules in NPC cells.
CD33+ cells were isolated from healthy PBMCs using human CD33 MicroBeads and were co-cultured with NPC or NPC-LMP1 cells in a Transwell system for 48 h. The percentage of CD33+CD11b+HLA-DR- MDSCs was measured by FACS staining. (A) Expression of the NLRP3, ASC, caspase-1, IL-1β, COX-2, Arg-1 and iNOS mRNAs was determined via real-time qRT-PCR, and the results were normalized against GAPDH expression. (B) NPC cells, including CNE2-vector, CNE2-LMP1, TW03-vector, and TW03-LMP1 cells, were cultured overnight following stimulation with LPS (0.1 μg/mL) and ATP (5 mM, 30 min) or no stimulation. Then, the culture supernatants were collected, and the concentrations of IL-1β, IL-6 and GM-CSF were measured by ELISA. (C) Representative density plots are shown as the CD33+CD11b+cells in the HLA-DR- gate induced by NPC or NPC-LMP1 cells in different combinations. Representative data from 1 of 5 experiments are presented. (D) Statistical analysis of the percentage of MDSCs mediated by NPC cells in different combinations. CFSE-labeled PBMCs were co-cultured with M-MDSCs, CNE2-vector-MDSCs, CNE2-LMP1-MDSCs, TW03-vector-MDSCs or TW03-LMP1-MDSCs at a ratio of 1:1 in OKT3-coated 96-well plates. After 3 days, the cells were collected, stained with anti-human mAbs against CD4 and CD8 and quantified by flow cytometry. Representative FACS density plots (E) and the statistical data (F) showed that the proliferation of PBMCs and CD4+ and CD8+ T cells was markedly suppressed by CNE2-LMP1-MDSCs or TW03-LMP1-MDSCs compared with the effects of CNE2-vector-MDSCs or TW03-vector-MDSCs. Data are presented as the means ± SEM of representative experiments performed in triplicate. *P < 0.05, **P < 0.01 compared with the control treatment.
Fig 4.
GLUT1-dependent glycolysis was required for NPC-LMP1-mediated MDSC differentiation.
(A) WB showed the level of GLUT1 in CNE2-LMP1-siNC, CNE2-LMP1-siGLUT1, TW03-LMP1-siNC and TW03-LMP1-siGLUT1 cells, and GAPD was included as the control. (B and C) Representative ECAR assays for CNE2-LMP1-siNC, CNE2-LMP1-siGLUT1, TW03-LMP1-siControl and TW03-LMP1-siGLUT1 cells. The time course and calculations for (B) glycolytic capacity and (C) statistical analysis of glycolytic capacity and glycolytic reserve are shown. (D) Expression of the NLRP3, ASC, caspase-1, IL-1β, COX-2, Arg-1 and iNOS mRNAs was determined via qRT-PCR in CNE2-LMP1-siControl, CNE2-LMP1-siGLUT1, TW03-LMP1-siControl and TW03-LMP1-siGLUT1 cells. (E) Representative immunoblotting of NPC-vector, NPC-LMP1, NPC-LMP1-siNC and NPC-LMP1-siGLUT1 cells stained with the indicated antibodies (n = 5). (F) Secretion of the cytokines IL-1β, IL-6 and GM-CSF from NPC-LMP1 cells following treatment with siGLUT1, siControl (siNC) or 2-DG was determined via ELISA. (G) Statistical analysis of the percentage of CD33+CD11b+HLA-DR-MDSCs mediated by CNE2-LMP1 or TW03-LMP1 cells following treatment with siGULT1, siNC or 2-DG. Data are presented as the means ± SEM of representative experiments performed in triplicate. *P < 0.05, **P < 0.01 compared with the control treatment.
Fig 5.
NLRP3 expression was involved in the accumulation of NPC-LMP1-mediated MDSCs.
(A) Representative immunoblots of CNE2-LMP1 and TW03-LMP1 cells treated with siNC or siNLRP3 and stained with the indicated antibodies (n = 3). (B) Secretion of the cytokines IL-1β, IL-6 and GM-CSF from NPC-LMP1 cells following treatment with si-NLRP3, siControl (siNC) or VX-765 was determined via ELISA. (C) Blockade of NLRP3 decreased the differentiation of HLA-DR-CD11b+CD33+ MDSCs mediated by NPC-LMP1. Representative data from 5 independent experiments are shown. (D) Statistical analysis of the percentage of CD33+CD11b+HLA-DR-MDSCs mediated by CNE2-LMP1 or TW03-LMP1 cells following treatment with si-NLRP3, siNC or VX-765. Representative data from 3 independent experiments are shown.
Fig 6.
LMP1 co-expression with GLUT1 in NPC-LMP1 cells stabilized GLUT1 proteins by disrupting GLUT1 K48 ubiquitination and p62-dependent lysosomal degradation.
(A) LMP1 physically interacts with endogenous GLUT1. NPC-LMP1 and NPC-vector cells were treated with MG-132 (1 μM) for 4 h to stabilize GLUT1, followed by immunoprecipitation (IP) with an anti-GLUT1 antibody. Immunoblotting was performed using anti-LMP1 and anti-GLUT1 antibodies. Anti-IgG was used as a negative control for IP. Representative data from 5 independent experiments are shown. (B) Immunofluorescence detection of EBV-LMP1 and GLUT1 in stable NPC-LMP1 cell lines. NPC-LMP1 and NPC-vector cells were seeded onto a chambered cover glass overnight. LMP1 and GLUT1 were detected with LMP1 and GLUT1 antibodies, nuclei were stained with DAPI, and confocal images were analyzed using ImageJ software. (C) NPC-LMP1 and NPC-vector cell lines were treated with CHX for 18 h. Proteins were harvest at 0, 3, 6, 12 and 18 h, and the expression of GLUT1 was measured via immunoblotting. Representative data from 5 independent experiments are shown, and GAPDH was included as a control. (D) NPC-LMP1 and NPC-vector cell lines were treated with the lysosome inhibitor BafA1 for 12 h, proteins were harvested at 0 and 12 h, and GLUT1 expression was measured by immunoblotting. Representative data from 5 independent experiments are shown, and GAPDH was included as a control. (E) Immunoblotting showed that GLUT1 was expressed at higher levels in 293T cells than in p62KO-293T cells, whereas GLUT1 was expressed at higher levels in 293T-LMP1 cells than in 293T cells but was not found at higher levels in p62KO-293T-LMP1 cells compared with p62KO-293T cells. Representative data from 1 of 5 experiments are shown. (F-G). NPC-LMP1 and NPC-vector cell lines cultured in 6-well plates were transfected with hemagglutinin (HA)-tagged Ub (4 μg/well) (F) HA-tagged Ub-K48, HA-tagged Ub-K48R, HA-tagged Ub-K63 or HA-tagged Ub-K63R (G) and then treated with 20 mM MG132 for 6 h prior to harvest. Cell lysates were immunoprecipitated with anti-HA antibodies and then subjected to WB with an anti-GLUT1 antibody to measure the levels of ubiquitinated GLUT1 proteins (upper panels). WCLs were blotted to evaluate the levels of GLUT1 proteins (lower panels). β-actin expression was used as a protein loading control. The experiment shown is a representative of three independent experiments.
Fig 7.
A schematic diagram illustrating the molecular signaling pathways regulating LMP1-mediated MDSC differentiation in NPC.
LMP1 up-regulates GLUT1 expression by activating p65 and binding to GLUT1 to delay GLUT1 p62-dependent autolysosomal degradation. The increased GLUT1 expression results in more extensive glycolysis. The increase in glycolysis triggers the activation of p65 and the NLRP3 inflammasome signaling pathway, leading to the production of IL-1β, IL6 and GM-CSF from NPC cells, which promotes tumor-associated MDSC expansion in NPC microenvironments.