Epigenetic silencing of tumor suppressor genes during in vitro Epstein-Barr virus infection

Abstract

DNA-methylation at CpG islands is one of the prevalent epigenetic alterations regulating gene-expression patterns in mammalian cells. Hypo- or hypermethylation-mediated oncogene activation, or tumor suppressor gene (TSG) silencing mechanisms, widely contribute to the development of multiple human cancers. Furthermore, oncogenic viruses, including Epstein-Barr virus (EBV)-associated human cancers, were also shown to be influenced by epigenetic modifications on the viral and cellular genomes in the infected cells. We investigated EBV infection of resting B lymphocytes, which leads to continuously proliferating lymphoblastoid cell lines through examination of the expression pattern of a comprehensive panel of TSGs and the epigenetic modifications, particularly methylation of their regulatory sequences. EBV infection of primary B lymphocytes resulted in global transcriptional repression of TSGs through engagement of hypermethylation. Therefore, CpG methylation profiles of TSGs may be used as a prognostic marker as well as development of potential therapeutic strategies for controlling acute infection and EBV-associated B-cell lymphomas.

Keywords: B-cell lymphoma; EBV; lymphoblastoid cell lines; promoter methylation; tumor suppressor genes.

Fig. 1.

Methyl-array profiling of human TSGs in response to EBV infection in resting PBMCs. (A) Approximately 10 million PBMCs were infected with GFP-tagged BAC-EBV. Genomic DNA was isolated on days 0-, 2-, 7-, and 15-d postinfection. DNA was quantified and subjected to overnight digestion with combinations of methylation-sensitive and -dependent enzymes, as suggested in the manual provided by the manufacturer [SABioscience-Methyl-Profiler DNA Methylation PCR Array System (335211)]. Data analysis (algorithm provided by the manufacturer) also included the quality check to include or exclude the real-time PCR data points for individual genes. Relative change in the percent methylation is represented with respect to the day 0 control sample. Changes in the methylation status are displayed as a heat map normalized separately for individual gene promoter. (B) The results are depicted as four significantly different categories which represented the methylation patterns: (i) gain of methylation, (ii) loss of methylation, (iii) distinct methylation status, and (iv) fluctuating methylation status. The similar experiments were carried out for three independent donors and a representative figure is shown from donor 1. The red-marked gene promoters were chosen for subsequent real-time PCR experiments.

Fig. S1.

In vitro infection of PBMCs with BAC-GFP-EBV. Approximately 10 million PBMCs from healthy donors were mixed with virus supernatant in 1 mL of RPMI 1640 with 10% FBS for 4 h at 37 °C. Cells were centrifuged for 5 min at 500 × g, and the supernatant discarded. Pelleted cells were resuspended in 2 mL of complete RPMI medium 1640 in six-well plates. EBV-GFP expression was visualized by fluorescence microscopy (20× magnification) and used to quantify infection.

Fig. 2.

Gene-expression analyses of selected TSGs and viral transcripts in response to EBV infection of resting PBMCs. In a similar experimental set-up as described in Fig. 1, total RNA was isolated at indicated days (0-, 2-, 7-, and 15-d postinfection) from resting PBMCs infected with BAC-GFP-EBV, converted to cDNA, and subjected to real-time PCR for the indicated cellular (A–C) and viral transcripts (D) using SYBR green PCR master mix as per the manufacturer’s protocol. Similar experiments were carried out for three independent donors and a representative figure is shown from donor 2. Relative transcriptional changes are represented with respect to the day 0 uninfected sample, and displayed as a heat map normalized separately for individual gene expressions.

Fig. 3.

PCR array analysis for chromatin-modification enzymes and chromatin-remodeling factors in response to EBV infection in resting PBMCs. In a similar experimental set-up as described in Figs. 1 and 2, total RNA was isolated from ∼10 million PBMCs infected with BAC-GFP-EBV at different time points (days 0, 2, 7, and 15 postinfection), cDNA was made and subjected to PCR array analyses using (A) “Epigenetic Chromatin Modification Enzymes PCR Array” and (B) “Human Epigenetic Chromatin Remodeling Factors PCR Array” from SABiosciences, according to the manufacturer’s instructions for two independent donors. The relative changes in transcripts using the 2^−ΔΔCt method were represented as a heat map in comparison with the day 0 uninfected sample control for donor 1. The red marked transcripts—particularly HDACs, DNA methyl transferases, and methyl binding proteins—were chosen for further analyses.

Fig. 4.

Real-time PCR validation of selected DNMTs, MBDs, and HDACs in response to EBV infection. (A and B) Total RNA was isolated from either PBMCs infected with BAC-GFP-EBV at different postinfection time points (0, 2, 7, and 15 d) or EBV-induced in vitro transformed LCLs. cDNA was made and subsequently subjected to real-time PCR analyses for the selected genes. The difference of individual gene expressions compared with uninfected PBMC samples is represented as a heat map. Similar experiments were carried out for two independent donors and a representative figure is shown for donor 1. (C) The cartoon diagram represents the working model of global repression of TSGs through epigenetic modification.

Fig. S2.

Expression analyses of DNMTs, MBDs, and HDACs in EBV⁻ and EBV⁺ B cells. Total RNA was isolated from freshly cultured 10 million of both EBV⁻ BJAB and BL41 cells and EBV⁺ Jijoyee and BL41-B95.8. cDNA was made and subsequently subjected to real-time PCR analyses for the indicated genes. The difference of individual gene expressions in EBV⁺ cells compared with EBV⁻ cells is represented as a heat map. The results shown are representative of three independent experiments.

Fig. 5.

Real-time PCR analyses of the selected genes in the presence and absence of 5′-azacytidine treatment. (A) Total RNA was isolated from either DMSO or 50 μM Aza-treated 10 million PBMCs infected with BAC-GFP-EBV at 7-d postinfection, and three different LCL clones (GFP-LCL10, LCL1, and LCL2) for 12 h. Subsequently, cDNA was generated and subjected to real-time PCR analyses for the selected genes as described in Materials and Methods. The difference in individual gene expression (fold-change) compared with the DMSO control is represented as a heat map. (B) Approximately 0.3 × 10⁵ LCL clones were plated into each well of the six-well plates, exposed to either 1 µM Aza or DMSO vehicle control treatment for 3 d at 37 °C humidified CO₂ chamber. Cell viability was determined through the Trypan blue exclusion method and only viable cells were counted every 24 h using an automated cell counter. Experiments were performed in duplicate and repeated two times.