Mechanism regulating proasthmatic effects of prolonged homologous beta2-adrenergic receptor desensitization in airway smooth muscle

Abstract

Use of long-acting beta(2)-adrenergic receptor (beta2AR) agonists to treat asthma incurs an increased risk of asthma morbidity with impaired bronchodilation and heightened bronchoconstriction, reflecting the adverse effects of prolonged homologous beta2AR desensitization on airway smooth muscle (ASM) function. Since phosphodiesterase 4 (PDE4) regulates ASM relaxation and contractility, we examined whether the changes in ASM function induced by prolonged homologous beta2AR desensitization are attributed to altered expression and action of PDE4. Cultured human ASM cells and isolated rabbit ASM tissues exposed for 24 h to the long-acting beta2AR agonist salmeterol exhibited impaired acute beta2AR-mediated cAMP accumulation and relaxation, respectively, together with ASM constrictor hyperresponsiveness. These proasthmatic-like changes in ASM function were associated with upregulated PDE4 activity due to enhanced expression of the PDE4D5 isoform and were prevented by pretreating the ASM preparations with the PDE4 inhibitor rolipram or with inhibitors of either PKA or ERK1/2 signaling. Extended studies using gene silencing and pharmacological approaches demonstrated that: 1) the mechanism underlying upregulated PDE4D5 expression following prolonged beta2AR agonist exposure involves PKA-dependent activation of G(i) protein signaling via its betagamma-subunits, which elicits downstream activation of ERK1/2 and its induction of PDE4D5 transcription; and 2) the induction of PDE4 activity and consequent changes in ASM responsiveness are prevented by pretreating the beta2AR agonist-exposed ASM preparations with inhibitors of G(i)-betagamma signaling. Collectively, these findings identify that the proasthmatic changes in ASM function resulting from prolonged homologous beta2AR desensitization are attributed to upregulated PDE4 expression induced by G(i)-betagamma-mediated cross-talk between the PKA and ERK1/2 signaling pathways.

Fig. 1.

Inhibition of phosphodiesterase 4 (PDE4) prevents the induction of impaired β₂-adrenergic receptor (β2AR)-mediated cAMP accumulation accompanying homologous β2AR desensitization in cultured human airway smooth muscle (ASM) cells. In contrast to vehicle-exposed (control) ASM cells, acute isoproterenol (ISO; 1.0 μM)-induced cAMP accumulation is markedly attenuated (P < 0.01) in cells initially exposed for 24 h to salmeterol (10 μM). By comparison, in ASM cells pretreated with rolipram (10 μM), the cAMP responses to isoproterenol are unaltered by preexposing the cells to salmeterol. Data represent means ± SE values from 3 paired experiments.

Fig. 2.

Regulation of cAMP PDE activity in cultured human ASM cells following prolonged exposure to β2AR agonists. Relative to vehicle-treated (control) ASM cells, levels of PDE activity are significantly increased in cells exposed for 24 h to 10 μM of either salmeterol or isoproterenol. The stimulated PDE activity is ablated in β2AR agonist-exposed ASM cells by pretreatment either with the PDE4 inhibitor rolipram (A), the transcription or protein synthesis inhibitors actinomycin D (Act D) or cycloheximide (CHX), respectively (B), the PKA inhibitor H89, or the ERK1/2 inhibitor U0126 (C). Data represent means ± SE values based on 3–4 determinations made under each treatment condition. **P < 0.01.

Fig. 3.

Inhibition of PDE4 and either the PKA and ERK signaling pathways prevents the induction of altered ASM tissue constrictor and relaxation responsiveness accompanying prolonged homologous β2AR desensitization. Relative to vehicle-treated controls, rabbit ASM tissues exposed for 24 h to salmeterol (10 μM) exhibit significantly increased constrictor responses to ACh (A) and impaired relaxation responses to isoproterenol (B). Pretreatment with either the PDE4-selective inhibitor rolipram (10 μM), the PKA inhibitor H89 (10 μM), or the ERK1/2 inhibitor U0126 (5 μM) prevents the salmeterol-induced changes in ASM constrictor and relaxation responsiveness. Data represent means ± SE values from 6–8 paired experiments.

Fig. 4.

Regulation of PDE4D5 mRNA expression in β2AR agonist-exposed ASM cells. ASM cells treated with salmeterol (10 μM) (A) or isoproterenol (10 μM) (B) exhibit temporal increases in PDE4D5 mRNA expression, with peak induction of mRNA transcripts detected at 6 h. C: β2AR agonist-induced upregulated expression of PDE4D5 transcripts is ablated in ASM cells pretreated with inhibitors of PKA (H89), MEK1/2 (U1026), Src tyrosine kinase (SU-6656), or by ADP ribosylation of G_i protein with pertussis toxin (PTX), whereas pretreatment with the JNK inhibitor (SP-600125) or the p38 MAPK inhibitor (SB-202190) has no effect.

Fig. 5.

β2AR agonist-exposed ASM cells exhibit PKA-dependent activation of the CRE-regulating transcription cofactors CREB and ATF1 and upregulation of PDE4D5 protein. A: Western blot depicting that β2AR stimulation elicits transiently upregulated expression of phosphorylated CREB and ATF1 proteins in ASM cells, with peak phosphorylation detected at 30 min. B: β2AR agonist-induced phosphorylation of CREB/ATF1 is prevented in ASM cells transfected with siRNA duplexes directed against the PKAα and PKAγ catalytic subunits, whereas transfection with a scrambled (control) siRNA duplex (scRNA) has no effect. C: salmeterol-induced upregulation of PDE4D5 protein expression is also prevented in ASM cells transfected with siRNAs directed against the PKAα and PKAγ catalytic subunits, whereas transfection with scrambled siRNA has no effect.

Fig. 6.

β2AR stimulation elicits PKA-dependent activation of ERK1/2 in ASM cells. A: Western blot depicting that isoproterenol and salmeterol acutely evoke enhanced ERK1/2 phosphorylation. B: isoproterenol-induced phosphorylation of ERK1/2 is prevented in ASM cells that are transfected with the PKA siRNA duplexes, whereas transfection with a scrambled (control) siRNA duplex (scRNA) has no effect. C: corresponding densitometric analysis of the changes in ERK1/2 phosphorylation demonstrates that, relative to control cells (lane 1), β2AR-stimulated cells exhibit a mean 2.3-fold increase in ERK1/2 phosphorylation (lane 2). The latter β2AR agonist-induced phosphorylation of ERK1/2 is prevented in cells pretreated with the PKA siRNA duplexes (lane 6), whereas pretreatment with the scRNA duplex has no effect (lane 4). Data represent means ± SE values based on 4 measurements obtained under each treatment condition. *P < 0.05.

Fig. 7.

β2AR agonist-induced phosphorylation of ERK1/2 in ASM cells is mediated by G protein βγ-subunit-mediated activation of the Ras signaling cascade. A: Western blot showing that, in contrast to ASM cells transfected with adeno-LacZ (i.e., negative control), β2AR agonist-induced ERK1/2 phosphorylation is prevented in ASM cells wherein Gβγ signaling is inhibited by transfection with adeno-βARK-ct. B: Western blot depicting that pretreatment with the c-Raf1 inhibitor GW5074 prevents isoproterenol-induced phosphorylation of ERK1/2 in ASM cells, whereas inhibition of Rap1 with GGTI-298 has no effect.

Fig. 8.

G_i-βγ signaling mediates upregulated PDE activity in β2AR-stimulated ASM cells. PDE4 activity is significantly increased in ASM cells exposed for 24 h to 10 μM isoproterenol or salmeterol. Stimulation of cAMP PDE activity is inhibited in β2AR-exposed ASM cells by ADP ribosylation of G_i protein with PTX or by blockage of Gβγ signaling with the membrane-permeable anti-βγ MPS-coupled peptide, whereas pretreatment with the membrane-permeable anti-G_iα3 peptide has no effect. Data represent means ± SE values based on 4 measurements obtained under each treatment condition. **P < 0.01.

Fig. 9.

Changes in ASM constrictor and relaxation responsiveness accompanying salmeterol-induced homologous β2AR desensitization are mediated by G_i-βγ signaling. Relative to vehicle-treated controls, rabbit ASM tissues exposed for 24 h to salmeterol exhibit significantly increased constrictor responses to ACh (A) and impaired relaxation responses to isoproterenol (B). The salmeterol-induced changes in ASM constrictor and relaxation responsiveness are prevented in tissues that are pretreated with the membrane-permeable anti-βγ MPS-coupled peptide, whereas pretreatment with the membrane-permeable anti-G_iα3 peptide has no effect. Data represent means ± SE values from 4 paired experiments.

Fig. 10.

Schematic representation of the proposed mechanism of induction of proasthmatic changes in ASM responsiveness following prolonged homologous β2AR desensitization. During early β2AR signaling initiated by ASM exposure to a β2AR agonist (e.g., salmeterol or isoproterenol), activation of PKA leads to acute uncoupling of the phosphorylated β2AR from G_s protein (10, 23) and activation of G_i-βγ signaling, a phenomenon that may be reflective of PKA-induced switching of β2AR coupling from G_s to G_i (8, 23). Activation of the G_i-βγ-subunit then initiates Src-mediated stimulation of the Ras/Raf1/MEK1/2 signaling pathway, resulting in downstream ERK1/2 activation which, in turn, leads to phosphorylation of the transcription factors CREB and ATF1 and their induction of CRE-driven PDE4D5 gene transcription. The latter induction of PDE4 expression and action results in homologous β2AR desensitization (i.e., late β2AR hyporesponsiveness) that is evidenced by attenuated cAMP accumulation in response to subsequent β2AR agonist exposure, which is associated with impaired ASM relaxation and heightened ASM contractility (16). The inhibitors used herein to identify the above signaling pathways are indicated within the boxes shown. It is important to note that this schematic representation of the proposed mechanism underlying prolonged homologous β2AR desensitization in ASM does not depict the reported acute interactions between PKA and PDE4 activity and between ERK and PDE4 activity and its acute feedback modulation by PKA activation that have been implicated in other cell types (15). AC, adenylyl cyclase.