Brd4 is essential for IL-1β-induced inflammation in human airway epithelial cells

Abstract

Background: Chronic inflammation and oxidative stress are key features of chronic obstructive pulmonary disease (COPD). Oxidative stress enhances COPD inflammation under the control of the pro-inflammatory redox-sensitive transcription factor nuclear factor-kappaB (NF-κB). Histone acetylation plays a critical role in chronic inflammation and bromodomain and extra terminal (BET) proteins act as "readers" of acetylated histones. Therefore, we examined the role of BET proteins in particular Brd2 and Brd4 and their inhibitors (JQ1 and PFI-1) in oxidative stress- enhanced inflammation in human bronchial epithelial cells.

Methods: Human primary epithelial (NHBE) cells and BEAS-2B cell lines were stimulated with IL-1β (inflammatory stimulus) in the presence or absence of H2O2 (oxidative stress) and the effect of pre-treatment with bromodomain inhibitors (JQ1 and PFI-1) was investigated. Pro-inflammatory mediators (CXCL8 and IL-6) were measured by ELISA and transcripts by RT-PCR. H3 and H4 acetylation and recruitment of p65 and Brd4 to the native IL-8 and IL-6 promoters was investigated using chromatin immunoprecipitation (ChIP). The impact of Brd2 and Brd4 siRNA knockdown on inflammatory mediators was also investigated.

Result: H2O2 enhanced IL1β-induced IL-6 and CXCL8 expression in NHBE and BEAS-2B cells whereas H2O2 alone did not have any affect. H3 acetylation at the IL-6 and IL-8 promoters was associated with recruitment of p65 and Brd4 proteins. Although p65 acetylation was increased this was not directly targeted by Brd4. The BET inhibitors JQ1 and PFI-1 significantly reduced IL-6 and CXCL8 expression whereas no effect was seen with the inactive enantiomer JQ1(-). Brd4, but not Brd2, knockdown markedly reduced IL-6 and CXCL8 release. JQ1 also inhibited p65 and Brd4 recruitment to the IL-6 and IL-8 promoters.

Conclusion: Oxidative stress enhanced IL1β-induced IL-6 and CXCL8 expression was significantly reduced by Brd4 inhibition. Brd4 plays an important role in the regulation of inflammatory genes and provides a potential novel anti-inflammatory target.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Figure 1. H2O2 induces intracellular ROS and enhances the inflammatory response.

BEAS-2B cells were pre-incubated with DCFH-DA for 30 minutes in loading media followed by wash with KRH buffer. Cells were then treated with different concentrations of H2O2 in KRH buffer for 2 hours and intracellular ROS was measured (A). Cells were exposed to a range of concentrations of H2O2 for 2 hours and cell viability was assessed using MTT assay (B). Results are presented as mean ± SEM. N = 4. *p<0.05; **p<0.01; ***p<0.0001; when compared to basal level (control). BEAS-2B cells were treated with H2O2 for 2 hours in the absence or presence of IL-β stimulation (overnight) or left untreated as a control. IL-6 (C) and CXCL8 (D) protein levels in cell culture supernatants were quantified by ELISA. IL-6 (E) and IL-8 (F) transcript levels were quantified by comparative real-time PCR and were normalised by measuring GNB2L1 transcript levels. Results are expressed as mean ± SEM of at least 4 independent experiments. * P<0.05; ** P<0.01; ***P<0.001 when compared to controls.

Figure 2. NF-κB p65 activates and translocates into the nucleus and IKK2 inhibitor AS602868 diminishes H2O2-enhanced IL-1β induction of cytokines.

Cells were harvested and nuclear proteins were extracted from BEAS-2B cells after treatment with H2O2 (100 µM) in the presence (+) or absence (-) of IL-1β (1ng/ml) for 2 hours. The quality and purity of the subcellular fractionation was determined by immunoblotting using anti-β-actin and TBP (A). Using Western blot analysis, NF-κB p65 nuclear protein was quantified (B). Densitometric analysis of each band is plotted above. NF-κB p65 DNA binding activity was measured using TransAM kit (C). Results are expressed as means ± SEM as ratio of NF-κB p65/TATA-binding protein (TBP) or relative to untreated cells. n = 4 independent experiments *p<0.05; **p<0.01; ***P<0.001 compared with unstimulated cells. Cells were pre-treated with AS602868 (5 µM, IKK2 inhibitor) followed by the treatment with H2O2 with or without IL-1β (1 ng/ml) for 16 hours. IL-6 (D) and CXCL8 (E) proteins were assayed by ELISA. The release of IL-6 and CXCL8 was completely inhibited in cells pre-treated with AS602868. Levels of IL-6 (F) and IL-8 (G) mRNA were quantified by real-time RT-QPCR and were normalised with respective to GNB2LI mRNA levels. The RT-QPCR findings are consistent with the ELISA data (H). AS602868 (5 µM) did not affect cell viability using MTT assay. Results are expressed as mean ± SEM. n = 4. ** P<0.01; ***P<0.001 when compared to H2O2+IL-1β alone.

Figure 3. NF-κB p65 acetylation and association with Brd4 protein.

(A) BEAS-2B cells were stimulated with H2O2 in the presence (+) or absence (-) of IL-1β (1 ng/ml) for 2 hours, nuclear (A) and cytoplasmic (B) extracts were fractioned by Western blot and membranes were probed with anti-acetylated NF-κB p65 antibody. The blots show that acetylated-310 (Ac310) NF-κB p65 is predominantly found in the nucleus when compared with the cytoplasm. (C) Brd4 protein was immunoprecipitated from whole cell extracts following treatments and separated by SDS-PAGE and subsequently analysed by Western blotting using an anti-NF-κB p65 antibody. Each blot is representative of 3 independent experiments and densitometric analysis of each band is plotted as bar graph above it. TBP: TATA-binding protein; *p<0.05; **p<0.01 compared with control (unstimulated).

Figure 4. H3 acetylation, p65 and Brd4 binding to IL-6 and IL-8 κB promoter sites.

Chromatin immunoprecipitation (ChIP) assays show that IL-1β induces p65 DNA binding to both IL-6 (A) and IL-8 (B) promoters. H2O2 by itself does not affect p65 DNA binding activity; however, when co-treated with IL-1β, the affinity is enhanced by 7-fold at IL-6 promoter site and 20-fold at IL-8 promoters. Brd4 is also recruited to the same κB promoter regions in the IL-6 (C) and IL-8 (D) promoters as p65. Histone 3 is acetylated at the IL-6 (E) and IL-8 (F) κB promoter sites following treatments. IgG is non-specific antibody used as a negative control. Furthermore, H3 acetylation was confirmed using confocal microscopy following IL-1β stimulation whereas H2O2 had no effect on AcH3 alone or in combination with IL-1β (G). Results are representative of at least 4 independent experiments.*p<0.05, **p<0.01 compared with control (unstimulated).

Figure 5. Concentration dependent reduction of IL-6 and CXCL8 by BET inhibitors.

Cells were pre-treated with JQ1 and JQ1 (-) enantiomers (A, B) or PFI-1 (C, D) for 4 hours followed by IL-1β (1 ng/ml) for 16 hrs. IL-6 (A, C) and CXCL8 (B, D) proteins were assayed by ELISA. The effect of JQ1 and JQ1 (-) (E) and PFI-1 (F) on cell viability was assessed by MTT assay. n = 3 independent experiments. Points represent mean ± SEM **p<0.01;***P<0.001 compared with IL-1β stimulation. #p<0.05 JQ1(-) versus JQ1. £££p<0.001when PFI-1 compared with IL-1β stimulation. $p<0.05; $$p<0.01when compared to control. (G) Cells were heat treated at 90°C or left untreated, mixed together and stained with LIVE/DEAD Fixable Aqua stain then analysed by flow cytometry. Cells were checked with forward scatter detector (FSC) and side scatter detector (SSC) and analysed by density graph to check cell size and granularity. Fragmented cells were excluded from the study. Histogram shows separation of live cells (left) and dead cells (right). These parameters were used to assess cell viability following treatment with JQ1 (0.5 µM) and PFI-1 (1 µM) for 16 hours. DMSO/Control (<1%) alone, PFI-1, JQ1(-) or JQ1 resulted in only 5% decrease of overall cell viability. The data is representative of 3 independent experiments.

Figure 6. The BET inhibitors (JQ1 and PFI-1) reduce inflammatory mediator production.

Cells were pre-treated with either JQ1 or JQ1(-) both at 500 nM for 4 hours followed by stimulation with H2O2 in the presence (+) or absence (-) of IL-1β (1 ng/ml) or both for 16 hours or left unstimulated. IL-6 (A) and CXCL8 (B) proteins were assayed by ELISA. IL-6 (C) and IL-8 (D) transcripts were quantified by RT-PCR. n = 4 independent experiments. Bar graph represents mean ± SEM *p<0.05, **p<0.01, when compared JQ1(-) with JQ1 treated cells. Under similar experimental conditions the effect of PFI-1 (1 µM) on IL-6 (E) and CXCL8 (F) proteins were assayed by ELISA. IL-6 (G) and IL-8 (H) mRNA levels were quantified by RT-QPCR. n = 4 independent experiments. Bar graph represent mean ± SEM *p<0.05, **p<0.01, when compared cells treated with or without PFI-1.

Figure 7. Knockdown of Brd4 reduces inflammation but not Brd2.

Cells were transfected with either pooled Brd4 or Brd2 siRNAs (20 nM) or non-specific siRNAs (20 nM). 72 hours post-transfection Brd4 mRNA (A) was quantified by RT-QPCR and nuclear extracts were analysed (B) by immunoblotting using anti-Brd4 and -TATA-binding protein (TBP). IL-6 (C) and CXCL8 (D) proteins were measured by ELISA in post-transfected cells following stimulation with H2O2 and IL-1β together. n = 4 independent experiments. 72 hours post-transfection, Brd2 mRNA (E) was quantified by RT-QPCR and nuclear extracts were analysed (F) by immunoblotting using anti-Brd2 and TBP. Brd2 siRNA knockdown had no effect on either IL-6 (G) or CXCL8 (H) expression. n = 4 independent experiments. Results represent mean ± SEM *p<0.05, **p<0.01, ***p<0.001, non-specific siRNA compared to Brd2 or Brd4 siRNA.

Figure 8. The effect of JQ1 on Brd4 and p65 binding to IL-6 and IL-8 promoters.

Chromatin immunoprecipitation (ChIP) assay shows that IL-1β (1 ng/ml) and H2O2 (100 µM) induces Brd4 (A) and p65 (B) DNA binding to the IL-6 promoter by 5-fold which is abolished in cells pre-treated with JQ1 (500 nM). Similarly, Brd4 (C) and p65 (D) DNA binding at the IL-8 promoter is increased following H2O2 and IL-1β stimulation in JQ1(-) (500 nM) pre-treated cells by 10- and 4-fold. This binding is diminished in cells pre-treated with the active JQ1 (500 nM). Results are representative of at least 3 independent experiments.*p<0.05, **p<0.01, ***p<0.001 when compared with unstimulated cells.

Figure 9. JQ1 reduces oxidative stress-enhanced IL-6 and CXCL8 expression in NHBE cells.

NHBE cells pre-treated with JQ1 but not JQ1(-) (both at 5×10−9−10−6 M) reduced IL-1β induced release of IL-6 (A) and CXCL8 (B) in a concentration-dependent manner. Points represent mean ± SEM *p<0.05; **p<0.01;***P<0.001 when compared with IL-1β stimulation. #p<0.05 JQ1(-) versus JQ1. NHBE cells were treated with a range of concentrations of H2O2 and intracellular ROS (C) and cell viability (D) were measured using DCFH-DA and MTT assay, respectively. Results are presented as mean ± SEM. N = 4. *p<0.05; **p<0.01; ***p<0.0001; when compared to untreated cells (control). Cells were pre-treated with either JQ1 or JQ1(-) both at 5×10-7 M for 4 hours followed by stimulation with IL-1β (1 ng/ml) in the presence (+) or absence (-) of H2O2 (100 µM) or both for 16 hours or left unstimulated. IL-6 (E) and CXCL8 (F) proteins were assayed by ELISA. IL-6 (G) and IL-8 (H) transcripts were quantified by RT-PCR. n = 5 independent experiments. Bar graph represents mean ± SEM *p<0.05; **p<0.01; ***p<0.001 when compared with controls. #p<0.05; ##p<0.01; ###p<0.001 when comparing JQ1- with JQ1(-)-treated cells.