Effects of inactivation-resistant agonists on the signalling, desensitization and down-regulation of bradykinin B(2) receptors

Abstract

Background and purpose: A peptide bradykinin (BK) B(2) receptor agonist partially resistant to degradation, B-9972, down-regulates this receptor subtype. We have used another recently described non-peptide agonist, compound 47a, as a tool to study further the effects of metabolically more stable and thus persistent, agonists of the BK B(2) receptor on signalling, desensitization and down-regulation of this receptor.

Experimental approach and key results: Compound 47a was a partial agonist at the B(2) receptor in the human umbilical vein, where it shared with B-9972 a very slow relaxation on washout, and in HEK 293 cell lines expressing tagged forms [myc, green fluorescent protein (GFP)] of the rabbit B(2) receptor. Compound 47a desensitized the umbilical vein to BK. In the cellular systems, the inactivation-resistant agonists induced [Ca(2+)](i) transients as brief as those of BK but affected other functions with a longer duration than BK [12 h; receptor endocytosis, endosomal beta-arrestin(1/2) translocation, protein kinase C-dependent extracellular signal-regulated kinases (ERK)1/2 phosphorylation and c-Fos expression]. The B(2) receptor-GFP was degraded in cells exposed to B-9972 or compound 47a for 12 h. The non-peptide B(2) receptor antagonist LF 16-0687 prevented all effects of compound 47a, which were also absent in cells lacking recombinant B(2) receptors.

Conclusion and implications: Inactivation-resistant agonists revealed a long-lasting assembly of the agonist-B(2) receptor-beta-arrestin complexes in endosomal structures and induce 'biased signalling' (in terms of activation of ERK and c-Fos) as a function of time. Further, B-9972 and compound 47a, unlike BK, efficiently down-regulated BK B(2) receptors.

Figure 1

Structures of compound 47a, an agonist of the bradykinin B₂ receptor, and of LF 16-0687, a competitive antagonist of this receptor. Circled areas show some structural similarities between 47a and LF 16-0687.

Figure 2

Antagonist effect of LF 16-0687 in a bioassay for the bradykinin (BK) B₂ receptor, the human umbilical vein stimulated with compound 47a. Values are the means ± SEM. (n= 6). Right: Schild plot analysis. See Results for details.

Figure 4

(A) Time for half-relaxation from the first maximal response for each bradykinin (BK) B₂ receptor stimulant after washout of the maximal concentration (same tissues as in Figure 3 for all three agonists). Kruskall–Wallis test indicated that the three groups were heterogeneous (P < 0.001). Further comparison with the values for BK using Dunn's multiple comparison test indicated statistically significant differences between BK and B-9972 (P < 0.05) and BK and compound 47a (P < 0.001). (B) Samples of cumulative concentration–effect curves for BK and compound 47a. Solid symbols represent drug application (nM cumulative concentrations indicated) and open symbols, the first of a series of tissue washouts.

Figure 3

Sequential stimulation of the human umbilical vein preparation to demonstrate agent-specific desensitization. Full concentration–effects curves were constructed at times 3 and 6 h relative to the beginning of the in vitro incubation in the order indicated above each of the four panels. Values are the per cent of maximal stimulant-induced contraction recorded at 3 h.

Figure 5

Displacement of [³H]BK (3 nM) bound to HEK 293 cells stably expressing BK B₂ receptor–GFP by three unlabelled ligands. Values are the means ± SEM. (duplicate determinations, n= 3–4). Average specific binding for this concentration of radioligand (100%) was 128 ± 27 fmol per well and the non-specific binding averaged 3.3 ± 0.5% of total binding.

Figure 6

Epifluorescence microscopy studies of live HEK 293 cells stably expressing bradykinin (BK) B₂ receptor–GFP and stimulated for 30 min, 3 h or 12 h with B₂ receptor ligands at the indicated concentrations. In the third row, ligands were applied for 30 min and then washed out for 2.5 h. Control cells generally exhibit sharply defined plasma membrane-associated green fluorescence. Blue fluorescence: nuclear counterstain with Hoechst 33258. Original magnification 1000×.

Figure 7

Calcium mobilization in HEK 293 cells stably expressing bradykinin (BK) B₂ receptor–GFP. The time course of effects elicited by acute application of B₂ receptor ligands (at 10 s, arrows) is shown. Values are means ± SEM. The number of replicates is indicated between parentheses.

Figure 8

Differential signalling responses induced by agonists with different resistance to inactivation in HEK 293 cells stably expressing bradykinin (BK) B₂ receptor–GFP (A, B, D, E, F) or HEK 293a cells transiently expressing myc–BK B₂ receptor (C). (A) Immunoblots for phospho-ERK1/2 and total ERK1/2 in response to BK or compound 47a (10 min treatments). B₂ receptor agonist-induced ERK1/2 phosphorylation in HEK 293 cells is receptor-dependent, as shown by the lack of effect of BK (10 nM; 10 min) or compound 47a (1 µM 10 min) on untransfected cells and by the antagonist effect of LF 16-0687 (1 µM; 10 min) on the activation of ERK1/2 induced by BK or compound 47a in cells stably expressing B₂ receptor–GFP. Total ERK1/2 is shown to compare loading in tracks. Representative results of two replicates. (B) Effect of duration of stimulation with agonists on activation of ERK1/2. Cells incubated for 10 min or 3 h were maintained in low serum medium, but not those stimulated for 12 h (*regular medium). (C) Effect of myc–B₂ receptor transient expression, agonists and time on ERK1/2 phosphorylation. Presentation as in (A, B). (D) Effect of the protein kinase C inhibitor GF109203x on kinin-induced ERK1/2 phosphorylation. (E) Induction of c-Fos expression in HEK 293 cells stably expressing B₂ receptor–GFP or in untransfected cells treated as indicated with one of the three agonists for various time periods; representative result of two experiments. (F) Effect of inhibitory drugs on c-Fos induction by compound 47a (1 µM, 12 h). Drug concentrations: LF 16-0687, 1 µM; GF109203x, 5 µM; PD98089, 25 µM.

Figure 10

Translocation of β-arrestin₁–cherryFP induced by 12 h treatment with bradykinin (BK) B₂ receptor agonists (concentrations as in Figure 9), and colocalization with B₂ receptor–GFP in HEK 293 cells. Original magnification 1000×.

Figure 9

Epifluorescence microscopy studies of HEK 293a cells transiently expressing β-arrestin₂–GFP and, optionally, myc–B₂ receptor and stimulated for 10 min to 12 h with B₂ receptor ligands at the indicated concentrations. Control cells exhibit diffuse cytosolic fluorescence. Original magnification 1000×.

Figure 11

Effect of agonists on B₂ receptor–GFP expressed in HEK 293 cells: immunoblot of total cell extracts based on anti-GFP antibodies. The cells were submitted to the indicated treatments for 12 h in the usual culture medium with heat-inactivated foetal bovine serum before extraction. Upper graph: Average densitometry values of immunoblots for four to six experiments with ligand treatments as shown. Lower figure: representative experiments also showing the prevention of stimulant-induced B₂ receptor–GFP degradation by LF 16-0687 co-treatment. anova indicated that the groups of values were different (P < 0.001). Comparison with control GFP/B₂ receptor–GFP density ratio (Dunnett's test): *P < 0.05; **P < 0.01.

Figure 12

Schematic representation of the mechanisms of bradykinin B₂ receptor-mediated responses to agonists in HEK 293 cells. The dotted arrow represents a pathway that has been proposed in other systems but is not supported in the present one. Thick arrows represent pathways that are favoured by agonists resistant to inactivation in long-term experiments.