RIP1 is a central signaling protein in regulation of TNF-α/TRAIL mediated apoptosis and necroptosis during Newcastle disease virus infection

Abstract

Newcastle disease virus (NDV) is an oncolytic virus which selectively replicates in tumor cells and exerts anti-tumor cytotoxic activity by promoting cell death. In this study, we focus on characterization of the underlying mechanisms of NDV-induced cell death in HeLa cells. We find that NDV Herts/33 strain triggers both extrinsic and intrinsic apoptosis at late infection times. The activation of NF-кB pathway and subsequent up-regulation of TNF-α/TRAIL initiates extrinsic apoptosis, leading to activation of caspase 8 and cleavage of Bid into tBid. tBid transmits the extrinsic apoptotic signals to mitochondria and mediates intrinsic apoptosis, which is hallmarked by cleavage of caspase 9. Moreover, RIP1 is cleaved into RIP1-N and RIP1-C at D324 by caspase 8, and this cleavage promotes apoptosis. Surprisingly, over expression of RIP1 reduces apoptosis and depletion of RIP1 promotes apoptosis, suggesting full length RIP1 is anti-apoptotic. Moreover, necroptosis hallmark protein MLKL is activated by phosphorylation at 12-24 h.p.i., and RIP1 regulates the level of phosphor-MLKL. Immunostaining shows that RIP1 aggregates to stress granules (SGs) at 8-24 h.p.i., and phosphor-MLKL is also recruited to SGs, instead of migrating to plasma membrane to exert its necrotic function. Immunoprecipitation study demonstrates that RIP1 bind to phosphor-MLKL, and depletion of RIP1 reduces the aggregation of MLKL to SGs, suggesting that RIP1 recruits MLKL to SGs. Altogether, NDV infection initiates extrinsic apoptosis via activation of NF-кB and secretion of TNF-α/TRAIL. Activation of caspase 8 by TNF-α/TRAIL and subsequent cleavage of Bid and RIP1 transmit the death signals to mitochondria. Meanwhile, virus subverts the host defensive necroptosis via recruiting phosphor-MLKL by RIP1 to SGs. Thus, RIP1 is a central signaling protein in regulation of apoptosis and necroptosis during NDV infection.

Keywords: NDV; RIP1; apoptosis; necroptosis.

Figure 1. NDV infection induces both intrinsic and extrinsic apoptosis in HeLa cells

(A) Cytopathic effects of NDV infection. HeLa cells were infected with 1 MOI of NDV. Images were taken under microscope at 4, 8, 12, 16, 20, 24 h.p.i.. (B) Growth curve of NDV in HeLa cells. The culture medium in Figure 1A was collected at 0, 4, 8, 12, 16, 20, and 24 h.p.i., and subjected to TCID₅₀ assay using DF-1 cells. Virus growth curve was plotted based on three biological replicates, and error bars represent standard deviation of the mean (n=3). (C) Activation of intrinsic and extrinsic apoptosis by NDV infection. HeLa cells were mock-infected or NDV-infected, and harvested at 0, 4, 8, 12, 16, 20, and 24 h.p.i.. The NDV-induced apoptosis was analyzed with Western blot using anti-caspase 8, anti-caspase 9, anti-caspase 3, anti-PARP. NP was detected to monitor virus replication and β-actin was detected as loading control. The intensities of bands were determined by densitometry, normalized to β-actin, and shown as fold change (virus:mock).

Figure 2. TNF-α and TRAIL are up-regulated via NF-кB pathway and promotes caspase 8 activation in NDV-infected HeLa cells

(A) Transcriptional up-regulation of TNF-α and TRAIL during NDV infection. HeLa cells were mock-infected or NDV-infected for 12, 16, or 20 h. Total RNA was extracted and the expression levels of TNF-α and TRAIL were analyzed with semi-quantitative real time RT-PCR. The values represent means of three independent triplicate experiments ± S.D. (B) Secretion of TNF-α and TRAIL. The culture medium was collected at 12, 16, 20 h.p.i., and subjected to ELISA with anti-TNF-α or anti-TRAIL. The values represent means of three independent triplicate experiments ± S.D. (C) The nuclear translocation of p65 during NDV infection and blockage of p65 nuclear translocation by IKKβ inhibitor IKK16. HeLa cells were mock–infected or NDV-infected, followed by the treatment of DMSO or 5 μM IKK16. The translocation of p65 (red) and the expression of viral protein NP (green) were analyzed with immunostaining at 16 h.p.i. using anti-p65 and anti-NP. Nuclei were stained with DAPI (blue). Images were acquired by a META 510 confocal laser-scanning microscope (Zeiss). Merged images illustrate DAPI/NP/p65 fluorescence. (D) The inhibition of TNF-α and TRAIL transcription by IKK16. HeLa cells were mock-infected or NDV-infected, followed by the treatment of DMSO or 5 μM IKK16. Total RNA was extracted 16 and 20 h.p.i., and subjected to semi-quantitative real time RT-PCR using primers targeting to TNF-α and TRAIL mRNA, respectively. The values represent means of three independent triplicate experiments ± S.D. (E) Inhibition of apoptosis by IKK16. Cell samples in Figure 1D were lysed and analyzed by Western blot using anti-caspase 8 and anti-PARP. The intensities of bands were determined by densitometry, normalized to β-actin, and shown as fold change (IKK16+/−).

Figure 3. Bid is cleaved by caspase 8 and promotes apoptosis during NDV infection

(A) The cleavage of Bid during NDV infection. Cells samples in Figure 1A were analyzed with Western blot using indicated antibodies. The intensities of Bid and tBid bands were determined by densitometry, normalized to β-actin, and shown as fold change (virus:mock). (B) The subcellular distribution of Bid during NDV infection. HeLa cells were mock-infected or NDV-infected. Cells were harvested at 16 h.p.i. and the subcellular distribution of Bid (green) and mitochondria (red) were analyzed with immunostaining. Nuclei were stained with DAPI (blue). Images were acquired with a META 510 confocal laser-scanning microscope (Zeiss). Merged images illustrate DAPI/Bid/mitochondrial tracker fluorescence. (C) Bid cleavage and PARP cleavage are dependent on caspase 8 activity. HeLa cells were mock-infected or infected with NDV, followed by the treatment of DMSO or 40 uM caspase 8 inhibitor Z-IEVD-FMK. Cells were harvested at 20, and 24 h.p.i., and the cleavage of caspase 8, Bid, and PARP were analyzed with Western blot. The intensities of bands were determined by densitometry, normalized to β-actin, and shown as fold change (FMK+/−).

Figure 4. The cleavage of RIP1 at D324 by caspase 8 promotes apoptosis during NDV infection

(A) RIP1 is cleaved during NDV infection. Cells samples in Figure 1A were analyzed with Western blot to check the cleavage of RIP1, using anti-RIP1-N or anti-RIP1-C. The intensities of bands were determined by densitometry, normalized to β-actin, and shown as fold change (virus : mock). (B) The cleavage of RIP1 is dependent on caspase 8 activity. HeLa cells were mock-infected or infected with NDV, followed by the treatment of DMSO or 40 μM Z-IEVD-FMK. Cells were harvested at 16, 20, and 24 h.p.i. and the cleavage of RIP1 was analyzed with Western blot using anti-RIP1-C. The intensities of RIP1 and RIP1-C bands were determined by densitometry, normalized to β-actin, and shown as fold change (FMK+/−). (C) RIP1 is cleaved at D324 during NDV infection. HeLa cells were transfected with Flag14, Flag14-RIP1, and Flag14-D324K for 20 h, followed by NDV infection. Cells were harvested at 20 h.p.i. and subjected to Western blot analysis to examine the expression and cleavage of RIP1. The intensities of RIP1-C bands were determined by densitometry, normalized to β-actin, and shown as fold change (D324K/RIP1).

Figure 5. RIP1 clusteres to SGs during NDV infection

(A) Subcellular distribution of RIP1. HeLa cells were mock-infected or infected with NDV for 16 h. The subcellular distribution of RIP1 and NP were analyzed with immunostaining using anti-RIP1-N or anti-RIP1-C (green), and anti-NP (red). Nuclei were stained with DAPI (blue). Images were acquired by a META 510 confocal laser-scanning microscope (Zeiss). Merged images illustrate RIP1/NP/DAPI fluorescence. (B) RIP1 is not co-localized well with RIP3 either during NDV infection or by the treatment of necroptosis stimulators. HeLa cells were mock-infected, infected with NDV for 16 h, or treated with necroptosis stimulators TNF-α, Z-VAD-FMK, and AT406 (T+Z+A) for 10 h. The subcellular distribution of RIP1 and RIP3 were analyzed with immunostaining using anti-RIP1-N or anti-RIP1-C (green), and anti-RIP3 (red). Nuclei were stained with DAPI (blue). Merged images illustrate RIP1/RIP3/DAPI fluorescence. (C) RIP1 is co-localized with SG marker G3BP1 during NDV infection. HeLa cells were mock-infected or NDV-infected for 16 h. The subcellular distribution of RIP1 and G3BP1 were analyzed with immunostaining using anti-RIP1-N or anti-RIP1-C (green), and anti-G3BP1 (red). Nuclei were stained with DAPI (blue). Merged images illustrate RIP1/G3BP1/DAPI fluorescence.

Figure 5. RIP1 clusteres to SGs during NDV infection

(A) Subcellular distribution of RIP1. HeLa cells were mock-infected or infected with NDV for 16 h. The subcellular distribution of RIP1 and NP were analyzed with immunostaining using anti-RIP1-N or anti-RIP1-C (green), and anti-NP (red). Nuclei were stained with DAPI (blue). Images were acquired by a META 510 confocal laser-scanning microscope (Zeiss). Merged images illustrate RIP1/NP/DAPI fluorescence. (B) RIP1 is not co-localized well with RIP3 either during NDV infection or by the treatment of necroptosis stimulators. HeLa cells were mock-infected, infected with NDV for 16 h, or treated with necroptosis stimulators TNF-α, Z-VAD-FMK, and AT406 (T+Z+A) for 10 h. The subcellular distribution of RIP1 and RIP3 were analyzed with immunostaining using anti-RIP1-N or anti-RIP1-C (green), and anti-RIP3 (red). Nuclei were stained with DAPI (blue). Merged images illustrate RIP1/RIP3/DAPI fluorescence. (C) RIP1 is co-localized with SG marker G3BP1 during NDV infection. HeLa cells were mock-infected or NDV-infected for 16 h. The subcellular distribution of RIP1 and G3BP1 were analyzed with immunostaining using anti-RIP1-N or anti-RIP1-C (green), and anti-G3BP1 (red). Nuclei were stained with DAPI (blue). Merged images illustrate RIP1/G3BP1/DAPI fluorescence.

Figure 6. RIP1 regulates apoptosis and necroptosis during NDV infection

(A) Phosphorylation of MLKL at Ser358. HeLa cells were mock-infected or NDV-infected. Cells were harvested at 4, 8, 12, 16, 20, and 24 h.p.i., and the phosphorylation level of MLKL at Ser358 was examined with Western blot using anti-phosphor-MLKL (Ser358). The intensities of phosphor-MLKL bands were determined by densitometry, normalized to MLKL, and shown as fold change (NDV : mock). (B) Inhibition of RIP1 kinase activity reduces necroptosis and promotes apoptosis. HeLa cells were infected with NDV, followed by treatment of DMSO or 200 μM RIP1 inhibitor Necrostain. Cells were harvested at 20 h.p.i., and the phosphorylation level of MLKL and cleavage of PARP were examined with Western blot. The intensities of bands were determined by densitometry. Phosphor-MLKL was normalized to MLKL, PARP-C was normalized to β-actin, and were shown as fold change (Nec +/−). (C) RIP1 kinase activity is anti-apoptotic. HeLa cells were transfected with Flag14, RIP1, and K45A for 20 h, followed by NDV infection. Cells were harvested at 20 h.p.i., and analyzed with Western blot. The intensities of bands were determined by densitometry. Phosphor-MLKL was normalized to MLKL, PARP-C was normalized to β-actin, and were shown as fold change (K45A/RIP1). (D) Depletion of RIP1 promotes apoptosis. HeLa cells were transfected with siRNA targeting to RIP1 or control siRNA (sic) for 36 h, followed by NDV infection. Cells were harvested at 20 h.p.i., and analyzed with Western blot. The intensities of bands were determined by densitometry. Phosphor-MLKL was normalized to MLKL, RIP1 and PARP-C were normalized to β-actin, and were shown as fold change (siRIP1/sic). (E) Inhibition of caspase 8 activity and RIP1 cleavage promotes necroptosis and reduces apoptosis. HeLa cells were mock-infected or NDV-infected, followed by treatment of DMSO or 40 μM caspase 8 inhibitor Z-IEVD-FMK. Cells were harvested at 20 h.p.i., and the phosphorylation level of MLKL and cleavage of PARP was examined by Western blot. The intensities of bands were determined by densitometry. Phosphor-MLKL was normalized to MLKL, PARP-C was normalized to β-actin, and were shown as fold change (FMK +/−). (F) Blockage of RIP1 cleavage by mutating D324 to K promotes necroptosis and reduces apoptosis during NDV infection. HeLa cells were transfected with Flag14, RIP1, and D324K for 20 h, followed by NDV infection. Cells were harvested at 20 h.p.i., and analyzed with Western blot. The intensities of bands were determined by densitometry. Phosphor-MLKL was normalized to MLKL, PARP-C was normalized to β-actin in Figure 4C, and were shown as fold change (D324K/RIP1).

Figure 6. RIP1 regulates apoptosis and necroptosis during NDV infection

(A) Phosphorylation of MLKL at Ser358. HeLa cells were mock-infected or NDV-infected. Cells were harvested at 4, 8, 12, 16, 20, and 24 h.p.i., and the phosphorylation level of MLKL at Ser358 was examined with Western blot using anti-phosphor-MLKL (Ser358). The intensities of phosphor-MLKL bands were determined by densitometry, normalized to MLKL, and shown as fold change (NDV : mock). (B) Inhibition of RIP1 kinase activity reduces necroptosis and promotes apoptosis. HeLa cells were infected with NDV, followed by treatment of DMSO or 200 μM RIP1 inhibitor Necrostain. Cells were harvested at 20 h.p.i., and the phosphorylation level of MLKL and cleavage of PARP were examined with Western blot. The intensities of bands were determined by densitometry. Phosphor-MLKL was normalized to MLKL, PARP-C was normalized to β-actin, and were shown as fold change (Nec +/−). (C) RIP1 kinase activity is anti-apoptotic. HeLa cells were transfected with Flag14, RIP1, and K45A for 20 h, followed by NDV infection. Cells were harvested at 20 h.p.i., and analyzed with Western blot. The intensities of bands were determined by densitometry. Phosphor-MLKL was normalized to MLKL, PARP-C was normalized to β-actin, and were shown as fold change (K45A/RIP1). (D) Depletion of RIP1 promotes apoptosis. HeLa cells were transfected with siRNA targeting to RIP1 or control siRNA (sic) for 36 h, followed by NDV infection. Cells were harvested at 20 h.p.i., and analyzed with Western blot. The intensities of bands were determined by densitometry. Phosphor-MLKL was normalized to MLKL, RIP1 and PARP-C were normalized to β-actin, and were shown as fold change (siRIP1/sic). (E) Inhibition of caspase 8 activity and RIP1 cleavage promotes necroptosis and reduces apoptosis. HeLa cells were mock-infected or NDV-infected, followed by treatment of DMSO or 40 μM caspase 8 inhibitor Z-IEVD-FMK. Cells were harvested at 20 h.p.i., and the phosphorylation level of MLKL and cleavage of PARP was examined by Western blot. The intensities of bands were determined by densitometry. Phosphor-MLKL was normalized to MLKL, PARP-C was normalized to β-actin, and were shown as fold change (FMK +/−). (F) Blockage of RIP1 cleavage by mutating D324 to K promotes necroptosis and reduces apoptosis during NDV infection. HeLa cells were transfected with Flag14, RIP1, and D324K for 20 h, followed by NDV infection. Cells were harvested at 20 h.p.i., and analyzed with Western blot. The intensities of bands were determined by densitometry. Phosphor-MLKL was normalized to MLKL, PARP-C was normalized to β-actin in Figure 4C, and were shown as fold change (D324K/RIP1).

Figure 7. MLKL clusteres to SGs during NDV infection, instead of plasma membrane translocation

(A) MLKL aggregates to punctate structure in the cytoplasm during NDV infection. HeLa cells were mock-infected or NDV-infected for 16 h. The subcellular distribution of phosphor-MLKL or total MLKL (green), and NP (red) were examined with immunostaining. Nuclei were stained with DAPI (blue). Merged images illustrate p-MLKL/NP/DAPI or MLKL/NP/DAPI fluorescence. (B) MLKL moves to plasma membrane by treatment of necroptosis stimulators (T+Z+A). HeLa cells were treated with necroptosis stimulators TNF-α, Z-VAD-FMK, and AT406 (T+Z+A) for 10 h. The subcellular distribution of phosphor-MLKL or MLKL (green), and RIP3 (red) were examined with immunostaining. Nuclei were stained with DAPI (blue). Merged images illustrate MLKL/RIP3/DAPI fluorescence. (C) MLKL is co-localized with SGs hallmark G3BP1 during NDV infection. HeLa cells were mock-infected or NDV-infected for 16 h. The subcellular distribution of phosphor-MLKL or MLKL (green), and G3BP1 (red) were examined with immunostaining. Nuclei were stained with DAPI (blue). Merged images illustrate MLKL/G3BP1/DAPI fluorescence.

Figure 7. MLKL clusteres to SGs during NDV infection, instead of plasma membrane translocation

(A) MLKL aggregates to punctate structure in the cytoplasm during NDV infection. HeLa cells were mock-infected or NDV-infected for 16 h. The subcellular distribution of phosphor-MLKL or total MLKL (green), and NP (red) were examined with immunostaining. Nuclei were stained with DAPI (blue). Merged images illustrate p-MLKL/NP/DAPI or MLKL/NP/DAPI fluorescence. (B) MLKL moves to plasma membrane by treatment of necroptosis stimulators (T+Z+A). HeLa cells were treated with necroptosis stimulators TNF-α, Z-VAD-FMK, and AT406 (T+Z+A) for 10 h. The subcellular distribution of phosphor-MLKL or MLKL (green), and RIP3 (red) were examined with immunostaining. Nuclei were stained with DAPI (blue). Merged images illustrate MLKL/RIP3/DAPI fluorescence. (C) MLKL is co-localized with SGs hallmark G3BP1 during NDV infection. HeLa cells were mock-infected or NDV-infected for 16 h. The subcellular distribution of phosphor-MLKL or MLKL (green), and G3BP1 (red) were examined with immunostaining. Nuclei were stained with DAPI (blue). Merged images illustrate MLKL/G3BP1/DAPI fluorescence.

Figure 8. RIP1 binds to MLKL and recruits MLKL to SGs

(A) Depletion of RIP1 prevents MLKL aggregate to SGs. HeLa cells were transfected with non-target siRNA or RIP1 siRNA for 36 h, followed with NDV infection for 16 h. Mock-infected cells, DMSO, T+A+Z stimulated cells were included as control. The subcellular distribution of RIP1 (green) and MLKL (red) was examined with immunostaining with anti-RIP1-N, or anti-RIP1-C, and anti-MLKL. Nuclei were stained with DAPI (blue). Merged images illustrate RIP1-N/MLKL/DAPI or RIP1-C/MLKL/DAPI fluorescence. (B) RIP1 binds to MLKL upon T+Z+A stimulation or NDV infection. HeLa cells were mock-infected, NDV-infected for 16 h, or stimulated with T+Z+A for 10 h, and subjected to immunoprecipitation using anti-RIP1-N. The pull down proteins and whole cell lysates were analyzed with Western blot.

Figure 8. RIP1 binds to MLKL and recruits MLKL to SGs

(A) Depletion of RIP1 prevents MLKL aggregate to SGs. HeLa cells were transfected with non-target siRNA or RIP1 siRNA for 36 h, followed with NDV infection for 16 h. Mock-infected cells, DMSO, T+A+Z stimulated cells were included as control. The subcellular distribution of RIP1 (green) and MLKL (red) was examined with immunostaining with anti-RIP1-N, or anti-RIP1-C, and anti-MLKL. Nuclei were stained with DAPI (blue). Merged images illustrate RIP1-N/MLKL/DAPI or RIP1-C/MLKL/DAPI fluorescence. (B) RIP1 binds to MLKL upon T+Z+A stimulation or NDV infection. HeLa cells were mock-infected, NDV-infected for 16 h, or stimulated with T+Z+A for 10 h, and subjected to immunoprecipitation using anti-RIP1-N. The pull down proteins and whole cell lysates were analyzed with Western blot.

Figure 9. RIP1 is a central signaling protein in regulation of TNF-α/TRAIL mediated extrinsic apoptosis and necroptosis during NDV infection

NDV infection activates NF-κB, induces the expression and secretion of TNF-α and TRAIL. TNF-α and TRAIL binds to TNF receptor or TRAIL receptor on cell surface, and the pro-apoptotic complex is formed. Caspase 8 is activated and cleaves Bid and RIP1, promotes apoptosis. Meanwhile, full length RIP1 promotes the phosphorylation of MLKL. RIP1 binds to phosphor-MLKL and recruits it to SGs, prevents it move to plasma membrane, thereby inhibiting necroptosis.