DNA methyltransferase inhibitor 5-azacytidine enhances neuroblastoma cell lysis by an oncolytic parainfluenza virus

Abstract

Studies with neuroblastoma have shown that the presence of aberrant DNA epigenetic modifications mediated by DNA methyltransferases correlates with poor prognosis, making these enzymes a target for therapeutics based on synthetic epigenetic modulators such as DNA methyltransferase inhibitors (DNMTi). Here, we have used a neuroblastoma cell line model to test the hypothesis that treatment with a DNMTi would enhance cell killing when used in combination with oncolytic Parainfluenza virus 5 (P/V virus), a cytoplasmic-replicating RNA virus. Pretreatment of SK-N-AS cells with the DNMTi 5-azacytidine substantially enhanced P/V virus-mediated cell death in a dose- and multiplicity of infection-dependent manner. Infection with the virus alone and the combination treatment with 5-azacytidine and P/V virus infection led to the activation of caspases-8, -9, and -3/7. Inhibition of caspases using a pan-caspase inhibitor minimally affected cell killing by P/V virus alone, but by contrast, largely reduced cell death mediated by 5-azacytidine treatment alone or in combination with P/V virus infection. 5-Azacytidine pretreatment dampened P/V virus gene expression and growth within the SK-N-AS cell population, which correlated with enhanced expression of important antiviral genes such as interferon-β and OAS2 . Taken together, our data support the role of combination treatment using 5-azacytidine and an oncolytic P/V virus for neuroblastoma therapy.

Fig. 1

Enhanced killing of neuroblastoma cells by combination treatment with DNMT inhibitor 5-azacytidine and P/V virus infection. SK-N-AS cells expressing nuclear red fluorescent protein (SK-N-AS NLR cells) were pretreated with either (b) DMSO as vehicular control, or with 1, 5, or 10 μM 5-azacytidine (c–e respectively) for ~19 h. Cells were mock-infected or infected with P/V virus at an MOI of 10 PFU/cell and cultured in the presence of indicated drug concentration. Red object count (ROC) was recorded in real-time via the IncuCyte system every 6 h by quantifying the number of cells with intact nuclear red fluorescent protein expression. ROC per well was calculated and normalized to ROC at time 0 (ROC^t0) when the infection was performed. Values were expressed as percentage of time 0 and are the mean of three replicates with error bars representing SD. *** indicates the point where P value <0.001 first begins when comparing virus-infected to mock-infected samples, and this statistical significance extends for later timepoints. The vertical line in each graph indicates the time taken to achieve a 50% reduction in ROC relative to time 0. (a) shows representative fluorescence images taken by the IncuCyte detector at 0 and 30 hpi showcasing cells expressing red nuclei (ROC) and P/V virus-infected cells (GFP). DNMT, DNA methyltransferase; GFP, green fluorescence protein; hpi, hours post-infection; MOI, multiplicity of infection; NLR, Nuc-Light-Red; P/V virus, Parainfluenza virus 5.

Fig. 2

Enhanced killing of neuroblastoma cells with combination pretreatment of 5-azacytidine and P/V virus infection is MOI dependent. SK-N-AS NLR cells were pretreated with (a) DMSO or (b) 5 μM 5-azacytidine for ~19 h. Cells were either mock-infected or infected with the P/V virus at increasing MOIs of 0.1, 1, and 10 PFU/cell and cultured in the presence of indicated drug concentration. ROC per well was normalized to ROC at time 0 (ROC^t0) when infection was initiated, as described in Fig. 1. Values are the mean of three replicates with error bars representing SD. ** and *** indicate the point where P values <0.002 or <0.001 respectively, first begin when comparing virus-infected to mock-infected samples, and this statistical significance extends for later timepoints. DMSO, dimethyl sulfoxide; MOI, multiplicity of infection. NLR, Nuc-Light-Red; P/V virus, Parainfluenza virus 5; ROC, red object count.

Fig. 3

Combination treatment of 5-azacytidine and P/V virus infection induces cell death and caspase activation in neuroblastoma cells. SK-N-AS cells were pretreated with either DMSO or 5 μM 5-azacytidine for ~19 h. Cells were mock-infected or infected with P/V virus at MOI 10 PFU/cell and were then cultured with either DMSO or 5 μM 5-azacytidine. Cytotoxicity was measured using CytoTox-Glo assay at (a) 30 hpi as described in Materials and Methods. Alternatively, samples were assayed for (b) caspase-3/7, (c) caspase-8, and (d) caspase-9 activity by Caspase-Glo Assays at 20 hpi. Samples labelled Z-VAD-FMK are from cells incubated with the pan-caspase inhibitor prior to performing the assay. ** and *** indicate P values of <0.002 and <0.001 respectively, when comparing mock-infected samples to corresponding samples within DMSO or 5-azacytidine-treated groups. DMSO, dimethyl sulfoxide; hpi, hours post-infection; MOI, multiplicity of infection; P/V virus, Parainfluenza virus 5.

Fig. 4

Pan-caspase inhibitor Z-VAD-FMK reduces cell death in neuroblastoma cells largely induced by 5-azacytidine but minimally by P/V virus infection. SK-N-AS NLR cells were pretreated with either (a, c) DMSO or (b, d) 5 μM 5-azacytidine for ~19 h. Cells were (a and b) mock-infected or (c and d) infected with P/V virus at MOI 10 PFU/cell and cultured with the indicated drug, in the presence or absence of 20 μg/ml Z-VAD-FMK. ROC per well was calculated and normalized to ROC at time 0 (ROC^t0) when the infection was performed (as described in Fig. 1). Values were expressed as percentage of time 0 and are the mean of three replicates with error bars representing SD. *, **, and *** indicate when P values of <0.033, <0.002, and 0.001, respectively, and first begin when comparing virus-infected to mock-infected samples, and this statistical significance extends until a new significance is achieved at later timepoints. DMSO, dimethyl sulfoxide; MOI, multiplicity of infection; NLR, Nuc-Light-Red; P/V virus, Parainfluenza virus 5; ROC, red object count.

Fig. 5

Pretreatment of neuroblastoma cells with 5-azacytidine reduces virus gene expression and progeny virus production. SK-N-AS cells were pretreated with DMSO or 5 μM 5-azacytidine for ~19 h. Cells were either mock-infected or infected with P/V virus at MOI of 0.1, 1, and 10 PFU/cell and cultured with the indicated drug concentration. At 24 hpi, (a) bright field and fluorescence images were captured and (b) the percentage of infected GFP+ cells was quantified by flow cytometry. Mean fluorescence intensity (MFI) for samples is indicated in parentheses. Media from infected cells was collected at 0, 16, and 24 hpi, and plaque assays were performed to quantify (c) infectious virus units. Viral titres were normalized to 10⁶ cells and expressed as PFU/ml in the logarithmic scale. Values are the mean of three samples, with error bars representing SD. ** and *** indicate P values of <0.002 and <0.001 when comparing to DMSO control and drug-treated samples. DMSO, dimethyl sulfoxide; GFP, green fluorescence protein; hpi, hours post-infection; MOI, multiplicity of infection; P/V virus, Parainfluenza virus 5.

Fig. 6

Combination treatment of neuroblastoma cells with 5-azacytidine and P/V virus infection increases expression of antiviral genes. SK-N-AS cells were pretreated with DMSO or 5 μM 5-azacytidine for ~19 h and then mock-infected or infected with P/V virus at MOI of 10 PFU/cell. Cells were then cultured with the indicated drug concentration. Total RNA was isolated at (a, c) 16 or (b, d) 24 hpi to determine levels of (a and b) IFN-β and (c and d) OAS2 using quantitative real-time PCR. Media was also collected at (e) 16 and (f) 24 hpi and biological levels of IFN-I were quantified using a functional detection assay, as described in the Materials and Methods. *, **, and *** indicate P values of <0.033, <0.002, and <0.001 when comparing DMSO-treated and mock-infected cells to other samples. ‘u’ in (b, f) denote undetectable levels of IFN-I. DMSO, dimethyl sulfoxide; hpi, hours post-infection; IFN-I, type I interferon; MOI, multiplicity of infection; P/V virus, Parainfluenza virus 5.

Fig. 7

Enhanced killing of SK-N-SH cells by combination treatment with 5-azacytidine and P/V virus infection. SK-N-SH NLR cells were pretreated with either (c) DMSO, or with (d and e respectively) 1 or 5 μM 5-azacytidine for ~19 h. Cells were mock-infected or infected with P/V virus at an MOI of 10 PFU/cell and cultured in the presence of indicated drug concentration. (a and b) show representative fluorescence images taken by the IncuCyte detector at 0 and 30 hpi respectively ROC per well was normalized to ROC at time 0 (ROC^t0) as described in Fig. 1. Values were expressed as percentage of time 0 and are the mean of three replicates with error bars representing SD. *, **, and *** indicate when P values of <0.033, <0.002, and 0.001 first begin when comparing virus-infected to mock-infected samples, and this statistical significance extends until a new significance is achieved at later timepoints. DMSO, dimethyl sulfoxide; hpi, hours post-infection; MOI, multiplicity of infection; NLR, Nuc-Light-Red; P/V virus, Parainfluenza virus 5; ROC, red object count.