IKKε modulates RSV-induced NF-κB-dependent gene transcription

Abstract

Respiratory syncytial virus (RSV), a negative-strand RNA virus, is the most common cause of epidemic respiratory disease in infants and young children. RSV infection of airway epithelial cells induces the expression of immune/inflammatory genes through the activation of a subset of transcription factors, including Nuclear Factor-κB (NF-κB). In this study we have investigated the role of the non canonical IκB kinase (IKK)ε in modulating RSV-induced NF-κB activation. Our results show that inhibition of IKKε activation results in significant impairment of viral-induced NF-κB-dependent gene expression, through a reduction in NF-κB transcriptional activity, without changes in nuclear translocation or DNA-binding activity. Absence of IKKε results in a significant decrease of RSV-induced NF-κB phosphorylation on serine 536, a post-translational modification important for RSV-induced NF-κB-dependent gene expression, known to regulate NF-κB transcriptional activity without affecting nuclear translocation. This study identifies a novel mechanism by which IKKε regulates viral-induced cellular signaling.

Fig. 1. Expression of IKKε induces IL-8 gene transcription

(A) Nuclear extracts were prepared from tetracycline-treated 293 cell lines expressing the empty vector (EV) or IKKε wild type (WT), fractionated on a 10% SDS-PAGE, transferred to PVDF membranes and probed with anti-Flag antibody. (B) 293 stable cell lines expressing EV or IKKε WT were treated with tetracycline for 48 h and infected with RSV for the indicated time points. Cell culture supernatants were assayed for IL-8 production by ELISA. Data are expressed as mean ± standard deviation of triplicate samples. (C) 293 stable cell lines expressing EV or IKKε WT were treated with tetracycline for 48 h and transfected with a plasmid containing the human IL-8 promoter linked to the luciferase gene. Cells were harvested 12 h after transfection to measure luciferase activity. For each plate luciferase was normalized to the β-galactosidase reporter activity. Data are expressed as mean ± standard deviation of normalized luciferase activity. Bars represent the mean value of triplicate samples of one representative experiment.

Fig. 1. Expression of IKKε induces IL-8 gene transcription

(A) Nuclear extracts were prepared from tetracycline-treated 293 cell lines expressing the empty vector (EV) or IKKε wild type (WT), fractionated on a 10% SDS-PAGE, transferred to PVDF membranes and probed with anti-Flag antibody. (B) 293 stable cell lines expressing EV or IKKε WT were treated with tetracycline for 48 h and infected with RSV for the indicated time points. Cell culture supernatants were assayed for IL-8 production by ELISA. Data are expressed as mean ± standard deviation of triplicate samples. (C) 293 stable cell lines expressing EV or IKKε WT were treated with tetracycline for 48 h and transfected with a plasmid containing the human IL-8 promoter linked to the luciferase gene. Cells were harvested 12 h after transfection to measure luciferase activity. For each plate luciferase was normalized to the β-galactosidase reporter activity. Data are expressed as mean ± standard deviation of normalized luciferase activity. Bars represent the mean value of triplicate samples of one representative experiment.

Fig. 1. Expression of IKKε induces IL-8 gene transcription

(A) Nuclear extracts were prepared from tetracycline-treated 293 cell lines expressing the empty vector (EV) or IKKε wild type (WT), fractionated on a 10% SDS-PAGE, transferred to PVDF membranes and probed with anti-Flag antibody. (B) 293 stable cell lines expressing EV or IKKε WT were treated with tetracycline for 48 h and infected with RSV for the indicated time points. Cell culture supernatants were assayed for IL-8 production by ELISA. Data are expressed as mean ± standard deviation of triplicate samples. (C) 293 stable cell lines expressing EV or IKKε WT were treated with tetracycline for 48 h and transfected with a plasmid containing the human IL-8 promoter linked to the luciferase gene. Cells were harvested 12 h after transfection to measure luciferase activity. For each plate luciferase was normalized to the β-galactosidase reporter activity. Data are expressed as mean ± standard deviation of normalized luciferase activity. Bars represent the mean value of triplicate samples of one representative experiment.

Fig. 2. IKKε modulates RSV-induced NF-κB transcriptional activity

(A) 293 stable cell lines expressing the empty vector (EV) or a dominant negative (DN) mutant of IKKε were treated with tetracycline for 48 h and then infected with RSV. Culture supernatants, from uninfected and infected cells, were assayed 24 h later for IL-8 production by ELISA. Data are expressed as mean ± standard deviation of triplicate samples. *, P< 0.05 relative to RSV-infected EV-expressing cells. (B) 293 cells were transiently transfected with a plasmid containing the human IL-8 promoter linked to the luciferase gene and either a plasmid expressing IKKε DN or the empty vector (EV). Cells were infected with RSV and harvested at 12 h p.i. to measure luciferase activity. For each plate luciferase was normalized to the β-galactosidase reporter activity. Data are expressed as mean ± standard deviation of normalized luciferase activity. Bars represent the mean value of triplicate samples of one representative experiment. *, P< 0.05 relative to RSV-infected EV-expressing cells. (C) 293 cells were transiently transfected with a plasmid containing multimers of the IL-8 NF-κB sites and either the IKKε DN expressing plasmid or the empty vector (EV). Cells were infected with RSV and harvested at 12 h p.i. to measure luciferase activity, as described above. *, P< 0.05 relative to RSV-infected EV-expressing cells.

Fig. 2. IKKε modulates RSV-induced NF-κB transcriptional activity

(A) 293 stable cell lines expressing the empty vector (EV) or a dominant negative (DN) mutant of IKKε were treated with tetracycline for 48 h and then infected with RSV. Culture supernatants, from uninfected and infected cells, were assayed 24 h later for IL-8 production by ELISA. Data are expressed as mean ± standard deviation of triplicate samples. *, P< 0.05 relative to RSV-infected EV-expressing cells. (B) 293 cells were transiently transfected with a plasmid containing the human IL-8 promoter linked to the luciferase gene and either a plasmid expressing IKKε DN or the empty vector (EV). Cells were infected with RSV and harvested at 12 h p.i. to measure luciferase activity. For each plate luciferase was normalized to the β-galactosidase reporter activity. Data are expressed as mean ± standard deviation of normalized luciferase activity. Bars represent the mean value of triplicate samples of one representative experiment. *, P< 0.05 relative to RSV-infected EV-expressing cells. (C) 293 cells were transiently transfected with a plasmid containing multimers of the IL-8 NF-κB sites and either the IKKε DN expressing plasmid or the empty vector (EV). Cells were infected with RSV and harvested at 12 h p.i. to measure luciferase activity, as described above. *, P< 0.05 relative to RSV-infected EV-expressing cells.

Fig. 2. IKKε modulates RSV-induced NF-κB transcriptional activity

(A) 293 stable cell lines expressing the empty vector (EV) or a dominant negative (DN) mutant of IKKε were treated with tetracycline for 48 h and then infected with RSV. Culture supernatants, from uninfected and infected cells, were assayed 24 h later for IL-8 production by ELISA. Data are expressed as mean ± standard deviation of triplicate samples. *, P< 0.05 relative to RSV-infected EV-expressing cells. (B) 293 cells were transiently transfected with a plasmid containing the human IL-8 promoter linked to the luciferase gene and either a plasmid expressing IKKε DN or the empty vector (EV). Cells were infected with RSV and harvested at 12 h p.i. to measure luciferase activity. For each plate luciferase was normalized to the β-galactosidase reporter activity. Data are expressed as mean ± standard deviation of normalized luciferase activity. Bars represent the mean value of triplicate samples of one representative experiment. *, P< 0.05 relative to RSV-infected EV-expressing cells. (C) 293 cells were transiently transfected with a plasmid containing multimers of the IL-8 NF-κB sites and either the IKKε DN expressing plasmid or the empty vector (EV). Cells were infected with RSV and harvested at 12 h p.i. to measure luciferase activity, as described above. *, P< 0.05 relative to RSV-infected EV-expressing cells.

Fig. 3. Lack of IKKε affects RSV-induced NF-κB-dependent gene expression

A549 cells were transfected with 100 nM siRNA targeting IKKε (siIKKε) or a scrambled control (siScr). A549 cells were mock infected or infected with RSV and harvested to prepare total RNA, total cell lysates or to collect cell supernatants. IKKε mRNA levels were measured at 24 h p.i. by Q-RT-PCR (A, left panel), IKKε protein levels were measured at 24 h p.i. by western blot (A, right panel), while IL-8 production was measured at various time points p.i. by ELISA (B). MEFs derived from wild type (WT) or IKKε−/− mice were infected with RSV and harvested at different times p.i. to measure KC production by ELISA. Data for ELISAs are expressed as mean ± standard deviation of triplicate samples (C). MEFs derived from WT or IKKε−/− mice were transiently transfected with a plasmid containing multimers of IL-8 NF-κB sites and either infected with RSV for 15 h or stimulated with 10 ng/ml of TNF-α for 6 h. Cells were harvested to measure luciferase activity (D).

Fig. 3. Lack of IKKε affects RSV-induced NF-κB-dependent gene expression

A549 cells were transfected with 100 nM siRNA targeting IKKε (siIKKε) or a scrambled control (siScr). A549 cells were mock infected or infected with RSV and harvested to prepare total RNA, total cell lysates or to collect cell supernatants. IKKε mRNA levels were measured at 24 h p.i. by Q-RT-PCR (A, left panel), IKKε protein levels were measured at 24 h p.i. by western blot (A, right panel), while IL-8 production was measured at various time points p.i. by ELISA (B). MEFs derived from wild type (WT) or IKKε−/− mice were infected with RSV and harvested at different times p.i. to measure KC production by ELISA. Data for ELISAs are expressed as mean ± standard deviation of triplicate samples (C). MEFs derived from WT or IKKε−/− mice were transiently transfected with a plasmid containing multimers of IL-8 NF-κB sites and either infected with RSV for 15 h or stimulated with 10 ng/ml of TNF-α for 6 h. Cells were harvested to measure luciferase activity (D).

Fig. 3. Lack of IKKε affects RSV-induced NF-κB-dependent gene expression

A549 cells were transfected with 100 nM siRNA targeting IKKε (siIKKε) or a scrambled control (siScr). A549 cells were mock infected or infected with RSV and harvested to prepare total RNA, total cell lysates or to collect cell supernatants. IKKε mRNA levels were measured at 24 h p.i. by Q-RT-PCR (A, left panel), IKKε protein levels were measured at 24 h p.i. by western blot (A, right panel), while IL-8 production was measured at various time points p.i. by ELISA (B). MEFs derived from wild type (WT) or IKKε−/− mice were infected with RSV and harvested at different times p.i. to measure KC production by ELISA. Data for ELISAs are expressed as mean ± standard deviation of triplicate samples (C). MEFs derived from WT or IKKε−/− mice were transiently transfected with a plasmid containing multimers of IL-8 NF-κB sites and either infected with RSV for 15 h or stimulated with 10 ng/ml of TNF-α for 6 h. Cells were harvested to measure luciferase activity (D).

Fig. 3. Lack of IKKε affects RSV-induced NF-κB-dependent gene expression

A549 cells were transfected with 100 nM siRNA targeting IKKε (siIKKε) or a scrambled control (siScr). A549 cells were mock infected or infected with RSV and harvested to prepare total RNA, total cell lysates or to collect cell supernatants. IKKε mRNA levels were measured at 24 h p.i. by Q-RT-PCR (A, left panel), IKKε protein levels were measured at 24 h p.i. by western blot (A, right panel), while IL-8 production was measured at various time points p.i. by ELISA (B). MEFs derived from wild type (WT) or IKKε−/− mice were infected with RSV and harvested at different times p.i. to measure KC production by ELISA. Data for ELISAs are expressed as mean ± standard deviation of triplicate samples (C). MEFs derived from WT or IKKε−/− mice were transiently transfected with a plasmid containing multimers of IL-8 NF-κB sites and either infected with RSV for 15 h or stimulated with 10 ng/ml of TNF-α for 6 h. Cells were harvested to measure luciferase activity (D).

Fig. 4. IKKε regulates RSV-induced p65 phosphorylation at Ser536

A549 cells were transfected with 100 nM siRNA targeting IKKε (siIKKε) or a scrambled control (siScr). A549 cells were mock infected or infected with RSV and harvested to prepare nuclear extracts or total cell lysates. Nuclear translocation of p65 was assessed by Western blot. Membrane was stripped and reprobed with Lamin B to determine equal loading of the samples (A). Ser536 phosphorylation levels were detected by Western blot of total cell lysates. Membrane was stripped and reprobed for total p65 and β-actin to determine equal loading of the samples. Densitometric analysis of band intensity of Ser 536 p65 was performed using the histogram function of Adobe Photoshop. Results are shown after corrections to p65. Results are representative of two independent experiments (B). MEFs derived from wild type (WT) or IKKε−/− mice were infected with RSV and harvested to prepare nuclear extracts or total cell lysates. Nuclear translocation of p65 was assessed by Western blot. Membrane was stripped and reprobed with Lamin B to determine equal loading of the samples (C). Ser536 phosphorylation levels were detected by Western blot of total cell lysates. Membrane was stripped and reprobed for total p65 and β-actin. Densitometric analysis of band intensity of Ser536 p65 was performed using the histogram function of Adobe Photoshop. Results are shown after corrections to p65. Results are representative of two independent experiments (D).

Fig. 4. IKKε regulates RSV-induced p65 phosphorylation at Ser536

A549 cells were transfected with 100 nM siRNA targeting IKKε (siIKKε) or a scrambled control (siScr). A549 cells were mock infected or infected with RSV and harvested to prepare nuclear extracts or total cell lysates. Nuclear translocation of p65 was assessed by Western blot. Membrane was stripped and reprobed with Lamin B to determine equal loading of the samples (A). Ser536 phosphorylation levels were detected by Western blot of total cell lysates. Membrane was stripped and reprobed for total p65 and β-actin to determine equal loading of the samples. Densitometric analysis of band intensity of Ser 536 p65 was performed using the histogram function of Adobe Photoshop. Results are shown after corrections to p65. Results are representative of two independent experiments (B). MEFs derived from wild type (WT) or IKKε−/− mice were infected with RSV and harvested to prepare nuclear extracts or total cell lysates. Nuclear translocation of p65 was assessed by Western blot. Membrane was stripped and reprobed with Lamin B to determine equal loading of the samples (C). Ser536 phosphorylation levels were detected by Western blot of total cell lysates. Membrane was stripped and reprobed for total p65 and β-actin. Densitometric analysis of band intensity of Ser536 p65 was performed using the histogram function of Adobe Photoshop. Results are shown after corrections to p65. Results are representative of two independent experiments (D).

Fig. 4. IKKε regulates RSV-induced p65 phosphorylation at Ser536

A549 cells were transfected with 100 nM siRNA targeting IKKε (siIKKε) or a scrambled control (siScr). A549 cells were mock infected or infected with RSV and harvested to prepare nuclear extracts or total cell lysates. Nuclear translocation of p65 was assessed by Western blot. Membrane was stripped and reprobed with Lamin B to determine equal loading of the samples (A). Ser536 phosphorylation levels were detected by Western blot of total cell lysates. Membrane was stripped and reprobed for total p65 and β-actin to determine equal loading of the samples. Densitometric analysis of band intensity of Ser 536 p65 was performed using the histogram function of Adobe Photoshop. Results are shown after corrections to p65. Results are representative of two independent experiments (B). MEFs derived from wild type (WT) or IKKε−/− mice were infected with RSV and harvested to prepare nuclear extracts or total cell lysates. Nuclear translocation of p65 was assessed by Western blot. Membrane was stripped and reprobed with Lamin B to determine equal loading of the samples (C). Ser536 phosphorylation levels were detected by Western blot of total cell lysates. Membrane was stripped and reprobed for total p65 and β-actin. Densitometric analysis of band intensity of Ser536 p65 was performed using the histogram function of Adobe Photoshop. Results are shown after corrections to p65. Results are representative of two independent experiments (D).

Fig. 4. IKKε regulates RSV-induced p65 phosphorylation at Ser536

A549 cells were transfected with 100 nM siRNA targeting IKKε (siIKKε) or a scrambled control (siScr). A549 cells were mock infected or infected with RSV and harvested to prepare nuclear extracts or total cell lysates. Nuclear translocation of p65 was assessed by Western blot. Membrane was stripped and reprobed with Lamin B to determine equal loading of the samples (A). Ser536 phosphorylation levels were detected by Western blot of total cell lysates. Membrane was stripped and reprobed for total p65 and β-actin to determine equal loading of the samples. Densitometric analysis of band intensity of Ser 536 p65 was performed using the histogram function of Adobe Photoshop. Results are shown after corrections to p65. Results are representative of two independent experiments (B). MEFs derived from wild type (WT) or IKKε−/− mice were infected with RSV and harvested to prepare nuclear extracts or total cell lysates. Nuclear translocation of p65 was assessed by Western blot. Membrane was stripped and reprobed with Lamin B to determine equal loading of the samples (C). Ser536 phosphorylation levels were detected by Western blot of total cell lysates. Membrane was stripped and reprobed for total p65 and β-actin. Densitometric analysis of band intensity of Ser536 p65 was performed using the histogram function of Adobe Photoshop. Results are shown after corrections to p65. Results are representative of two independent experiments (D).

Fig. 5. Ser536 is important for RSV-induced NF-κB-dependent gene expression

A549 cells were transiently transfected with a plasmid containing the human IL-8 promoter linked to the luciferase gene and either a plasmid expressing WT p65 or the 536A mutant. Cells were infected with RSV and harvested at 12 h p.i. to measure luciferase activity. For each plate luciferase was normalized to the β-galactosidase reporter activity. Data are expressed as mean ± standard deviation of normalized luciferase activity*, P< 0.05 relative to RSV-infected cells (A). MEFs derived from p65−/− mice were transfected with a plasmid encoding p65 either wild type (WT) or mutated on serine 536 (Ser536Ala). Cells were either infected with RSV for 15 and 24 h or stimulated with TNF for 1 h and harvested to prepare total RNA. Gro-β gene expression was analyzed by Q-RT-PC (B).

Fig. 5. Ser536 is important for RSV-induced NF-κB-dependent gene expression

A549 cells were transiently transfected with a plasmid containing the human IL-8 promoter linked to the luciferase gene and either a plasmid expressing WT p65 or the 536A mutant. Cells were infected with RSV and harvested at 12 h p.i. to measure luciferase activity. For each plate luciferase was normalized to the β-galactosidase reporter activity. Data are expressed as mean ± standard deviation of normalized luciferase activity*, P< 0.05 relative to RSV-infected cells (A). MEFs derived from p65−/− mice were transfected with a plasmid encoding p65 either wild type (WT) or mutated on serine 536 (Ser536Ala). Cells were either infected with RSV for 15 and 24 h or stimulated with TNF for 1 h and harvested to prepare total RNA. Gro-β gene expression was analyzed by Q-RT-PC (B).