Beta-arrestin 2 is required for B1 receptor-dependent post-translational activation of inducible nitric oxide synthase

Abstract

A major source of "high-output" NO in inflammation is inducible nitric oxide synthase (iNOS). iNOS is primarily transcriptionally regulated and is thought to function as an uncontrolled generator of high NO. We found that iNOS in cytokine-stimulated human lung microvascular endothelial cells (HLMVECs) is highly regulated post-translationally via activation of the B1 kinin G protein-coupled receptor (B1R). We report here that B1R-mediated iNOS activation was significantly inhibited by knockdown of beta-arrestin 2 with siRNA in cytokine-treated HLMVECs or HEK293 cells transfected with iNOS and B1R. In contrast, beta-arrestin 1 siRNA had no effect. The prolonged phase of B1R-dependent ERK activation was also inhibited by beta-arrestin 2 knockdown. Furthermore, robust ERK activation by the epidermal growth factor receptor (a beta-arrestin 2 independent pathway) had no effect on iNOS-derived NO production. beta-arrestin 2 and iNOS coimmunoprecipitated, and there was significant fluorescence resonance energy transfer between CFP-iNOS and beta-arrestin 2-YFP (but not beta-arrestin 1-YFP) that increased 3-fold after B1R stimulation. These data show that beta-arrestin 2 mediates B1R-dependent high-output NO by scaffolding iNOS and ERK to allow post-translational activation of iNOS. This could play a critical role in mediating endothelial function in inflammation.

Figure 1.

B1R-mediated NO production is dependent on β-arrestin 2. A) HLMVECs were transfected by electroporation (Amaxa Nucleofector) where indicated with vehicle, 160 pmol control siRNA, or 160 pmol β-arrestin 2 (βarr2) specific siRNA. At 50 h postelectroporation, cells were treated with 5 ng/ml IL-1β and 100 U/ml IFN-γ (22 h) to up-regulate B1R and iNOS expression. Cells were then stimulated with 100 nM DAKD, and NO was measured in real time with a porphyrinic microsensor. Cells were then lysed and subjected to Western blot analysis to detect iNOS and β-arrestin 2 expression. Data are expressed as means ± se of NO concentration achieved at 20 min. (n=3). *P < 0.05. B) HEK-B1R cells were transiently transfected with iNOS alone (control) or iNOS and control siRNA (NC) or increasing amounts of siRNA specific for β-arrestin 2. After transfection (48–72 h), cells were serum starved overnight. Cells were then treated with 100 nM DAKD, and NO was measured in real time with a porphyrinic microsensor. Data are expressed as means ± se of NO concentration achieved at 20 min (n=3). *P < 0.05, **P<0.01 vs. control; #P < 0.001 vs. control. C) HEK-B1R cells were transiently transfected with iNOS alone or with iNOS and β-arrestin 2 siRNA (75 pmol for 72 h). Cells were stimulated with 100 nM DAKD, and NO production was measured with porphyrinic microsensor for 65 min. Vertical bar on left axis represents electrode response to 700 nM NO standard.

Figure 2.

B1R and iNOS-dependent NO production is not affected by β-arrestin1 knockdown. A) HEK-B1R cells were transiently transfected with iNOS and siRNA specific to β-arrestin 1 (βarr1) or β-arrestin 2 (βarr2) as above (Fig. 1), and NO production stimulated with 100 nM DAKD was measured in real time with a porphyrinic microsensor. Data are expressed as means ± se of NO concentration achieved at 20 min. (n=3). *P < 0.001. B) HEK-B1R cells were transiently transfected with iNOS and then stimulated with increasing concentrations of B1R agonist DAKD, and the NO concentration achieved at 20 min was measured. C) HEK-B1R cells were transiently transfected with iNOS and 75 pmol of siRNA for β-arrestin 1 (βarr1), β-arrestin2 (βarr2), or nonspecific control (NC). Cells were then stimulated with 20 or 100 nM DAKD, and the NO concentration achieved at 20 min was measured. Samples were subjected to Western blot analysis and probed for iNOS, β-arrestin 1, β-arrestin 2, and β-actin. Data are expressed as means ± se (n=3). *P < 0.05.

Figure 3.

EGF activates ERK without stimulating iNOS-dependent NO and does not affect B1R-mediated NO production. A) HEK-B1R cells were stimulated with either DAKD (100 nM) or EGF (100 ng/ml) for various times. Where indicated, cells were preincubated 30 min with 100 nM of EGFR kinase inhibitor PD153035. Cell lysates were analyzed by Western blot analysis for phospho-ERK, total ERK, and GAPDH. Results are representative of ≥3 independent experiments B) HEK-B1R cells were transiently transfected with iNOS. Cells were stimulated with either 100 nM DAKD, 100 ng/ml EGF, or 100 ng/ml EGF for 10 min, then 100 nM DAKD (without or with preincubation for 30 min with PD153035), and NO concentration was measured. Cells were then lysed, and iNOS expression was detected by Western blot analysis. Data are expressed as means ± se (n=3), *P < 0.01 vs. 100 nM DAKD.

Figure 4.

β-Arrestin 2 interacts with iNOS. A) HEK-B1R cells were transfected with iNOS and β-arrestin 2-V5 where indicated. After transfection for 36–48 h, cells were treated with B1R agonist at the indicated time points; cells were collected and lysed, and aliquots were removed for lysate control (bottom panel). Remaining supernatant was incubated with polyclonal antibody to iNOS and protein-G beads. After washing, immunoprecipitates were boiled in Laemmli sample buffer (2×) and analyzed by immunoblotting (IB) with antibody to iNOS (immunoprecipitation control) or the V5 epitope to detect coimmunoprecipitated β-arrestin 2. For controls, samples were immunoprecipitated with normal IgG (IgG) or with protein-G beads without primary antibody (bead). Results are representative of ≥3 independent experiments. B) HEK-B1R stable cells were transiently transfected with CFP-iNOS and β-arrestin 2 (βarr2)-YFP or CFP-iNOS and β-arrestin 1 (βarr1)-YFP (negative control). After treatment with 100 nM DAKD for various times, cells were fixed and analyzed for FRET using the acceptor photobleaching method. Top: CFP and YFP were visualized at excitation/emission wavelengths of 458/485 nm and 514/545 nm, respectively, after treatment with DAKD for 5 min. Images show enhanced CFP emission after YFP photobleaching in 2 ROIs (outlined in white) from cells transfected with CFP-iNOS and βarr2-YFP (left) but not β-arr1-YFP (right). Bottom: plot of time course of change in CFP emission corresponding to a decrease in YFP emission caused by photobleaching at 514 nm. Scale bars = 10 μm. C) FRET efficiency for CFP-iNOS and βarr2-YFP vs. CFP-iNOS and βarr1-YFP (negative control) in the absence of agonist (0 min) or after 100 nM DAKD stimulation for various times (15–20 cells/time point). FRET efficiency was calculated as described in Materials and Methods. Data are expressed as means ± se (n=20). *P < 0.05, **P < 0.001 vs. corresponding control.

Figure 5.

Depletion of β-arrestin 2 with siRNA reduces the prolonged phase of B1R-mediated ERK activation. HEK-B1R cells were transfected with 75 pmol of siRNA specific for βarr2 (open squares) or control siRNA (solid circles) for 72–96 h. Cells were then stimulated with 100 nM DAKD for the indicated time and lysed, and ERK activation was assayed by Western blot analysis for phospho-ERK. Membranes were also probed for β-arrestin 2 and GAPDH. Phospho-ERK was quantitated by densitometry using ImageJ software after correcting for loading based on intensity of the GAPDH bands. β-Arrestin 2-mediated ERK activation was determined by subtracting values for phospho-ERK in β-arrestin 2 siRNA treated cells from those obtained in cells treated with control siRNA (open triangles). Data are expressed as means ± se (n=5). *P < 0.05, **P < 0.01 vs. corresponding control.