Roger Barber

Partial agonists and G protein-coupled receptor desensitization

Weak or partial agonists induce less desensitization of G protein-coupled receptors (GPCRs) than do strong agonists. However, there have been few attempts to relate partial agonism quantitatively with the various parameters of agonist-induced desensitization, and to elucidate the mechanisms involved. Our understanding of how the treatment of cells and tissues with partial agonists affects their capacity to activate receptors is based on continued progress in defining partial agonism and the mechanisms of desensitization in which protein kinases, phosphatases, endocytosis and recycling play various roles. In this review, current research concerning partial-agonist-induced desensitization of GPCRs and the nature of partial agonism is summarized, and an attempt is made to put the existing knowledge into a working hypothesis concerning the mechanisms that account for the reduced desensitization in response to partial agonists.

Salmeterol-induced desensitization, internalization and phosphorylation of the human β2-adrenoceptor

Partial agonists of the β2‐adrenoceptor which activate adenylyl cyclase are widely used as bronchodilators for the relief of bronchoconstriction accompanying many disease conditions, including bronchial asthma. The bronchodilator salmeterol has both a prolonged duration of action in bronchial tissue and the ability to reassert this activity following the temporary blockade of human β2‐adrenoceptors with antagonist. We have compared the activation and desensitization of human β2‐adrenoceptor stimulation of adenylyl cyclase induced by salmeterol, adrenaline and salbutamol in a human lung epithelial line, BEAS‐2B, expressing β2‐adrenoceptor levels of 40–70 fmol mg−1, and in human embryonic kidney (HEK) 293 cell lines expressing 2–10 pmol mg−1. The efficacy observed for the stimulation of adenylyl cyclase by salmeterol was only ≈percnt;10% of that observed for adrenaline in BEAS‐2B cells expressing low levels of β2‐adrenoceptor, but similar to adrenaline in HEK 293 cells expressing very high levels of receptors. Salmeterol pretreatment of these cells induced a rapid and stable activation of adenylyl cyclase activity which resisted extensive washing and β2‐adrenoceptor antagonist blockade, consistent with binding to a receptor exosite and/or to partitioning into membrane lipid. The desensitization and internalization of β2‐adrenoceptors induced by the partial agonists salmeterol and salbutamol were considerably reduced relative to the action of adrenaline. Consistent with these observations, the initial rate of phosphorylation of the receptor induced by salmeterol and salbutamol was much reduced in comparison to adrenaline. Our data suggest that the reduction in the rapid phase of desensitization of β2‐adrenoceptors after treatment with salmeterol or salbutamol is caused by a decrease in the rate of β2‐adrenoceptor kinase (βARK) phosphorylation and internalization. In contrast, the rate of cyclic AMP‐dependent protein kinase (PKA)‐mediated phosphorylation by these partial agonists appears to be similar to adrenaline.

Examination of the effects of increasing Gs protein on β2-adrenergic receptor, Gs, and adenylyl cyclase interactions☆

We have examined the effect of increased Gs protein levels on the abilities of three different beta2-agonists to induce GTP shifts and stimulate adenylyl cyclase response in an effort to investigate the kinetic association between the beta2-adrenergic receptor Gs and adenylyl cyclase. Agonist competition binding analysis and adenylyl cyclase concentration-response assays revealed that increases in Gs protein resulted in proportional increases in the areas of the GTP shift and adenylyl cyclase activity. Changes in the magnitude of the GTP shift were evaluated with a novel and straightforward approach for analyzing the GTP shift data that allowed us to determine the proportion of high agonist affinity binding receptor population and the apparent dissociation constant between the agonist bound receptor and Gs, regardless of the Gs protein level or the type of beta2-agonist. Using this method, we concluded that increased Gs results in the accumulation of the receptor population displaying high affinity towards agonist (HRGs) by increasing the number of receptor-Gs complexes (to a receptor:Gs protein ratio of about 0.7 at maximal Gs expression) without affecting the affinity between hormone bound receptor and Gs. Using the Gs protein levels determined with our novel analysis, we ran simulations using the theoretical shuttle model equation that relates the EC50 to available Gs. Fitting the simulations to experimental data required a receptor to catalytic unit ratio of 0.45 and revealed at least two distinct stages for beta2-agonist-stimulated adenylyl cyclase activity, namely, the activation of Gs by the beta2-adrenergic receptor (a step whose rate is dependent on the type of agonist used to stimulate activity), and the activation of adenylyl cyclase by active Gs (a step whose rate is independent of the type of agonist).

Evidence for the shuttle model for GSα activation of adenylyl cyclase

Abstract Knowledge of the nature of the interaction between the stimulatory G protein (G s ) and the adenylyl cyclase catalytic unit (C) is essential for interpreting the effects of G s mutations and expression levels on cellular response to a wide variety of hormones, drugs, and neurotransmitters. It has been proposed that s-adrenergic receptor activation of adenylyl cyclase occurs either by a two-step “shuttle” mechanism where the receptor activates G s independently of cyclase followed by G s α activation of cyclase independent of the receptor; or the receptor activates a “precoupled” G s —C complex in a single step. Simulations of the two models revealed that the two forms of activation are distinguishable by the effect of G s levels on epinephrine-stimulated EC 50 values for cyclase activation; specifically, the shuttle model predicts an increased potency of epinephrine stimulation as levels of G s α increase. To address this problem, S49 cyc − cells were stably transfected with the gene for G sα long regulated by the MMTV LTR promoter, which allowed for an induction of G sα long expression levels over a 40-fold range by incubation of the cells for various times with 5 μM dexamethasone. Expression of G sα was strongly correlated to the appearance of GTP shifts in the competitive binding of epinephrine with [ 125 I]iodocyanopindolol to the β-adrenergic receptors and epinephrine-stimulated adenylyl cyclase activity. Most importantly, high expression of G s α resulted in lower EC 50 values for epinephrine and prostaglandin E 1 stimulation of adenylyl cyclase activity. The decrease in EC 50 did not occur as a result of a change in β 2 -adrenergic receptor, G i α, Gβα, or adenylyl cyclase levels. These novel findings demonstrate that a change in the level of a protein downstream of a plasma membrane receptor can influence hormone potency. We explain these results by using kinetic arguments to suggest that some fraction of hormone-activated adenylyl cyclase occurs via a shuttle mechanism, and not a purely precoupled mechanism.

Discrimination between intact cell desensitization and agonist affinity changes

The relationship between epinephrine-induced receptor-epinephrine affinity changes and intact cell adenylate cyclase activity was investigated using S49 lymphoma cells. It was demonstrated that 20 nM epinephrine caused desensitization (defined as a reduction in the rate of cAMP synthesis with time), but no significant change in receptor-epinephrine affinity as measured by competition with [125I]iodopindolol. On the other hand, treatment with 5 microM epinephrine caused no desensitization (as defined above), but did cause a very significant reduction in receptor-epinephrine affinity. It is suggested that there is a desensitization process which is distinct from the agonist affinity shift and which is expressed primarily as a change in the EC50 of the hormone-induced activation. The demonstration of extensive desensitization at low concentrations of hormone may have important physiological implications.

Effects of desensitization on the responses of WI-38 and S49 cells to hormones

Experiments with intact cells were used to quantitate the effects of hormone desensitization on the cAMP accumulation in cultured fibroblasts (WI-38) and lymphoma cells (S49). Desensitization of WI-38 cells to prostaglandin E1 and epinephrine and of S49 cells to epinephrine was associated with a shift in the ED50 in each case to higher hormone concentrations. This led to the situation where cells under continuous stimulation by low concentrations of hormone demonstrated a greater relative decline in cAMP synthesis with time than the same cells treated with high hormone concentrations. Therefore, any general attempt to quantitate desensitization in terms of cAMP accumulation in these cells must take the hormone concentration of the assay into account.

A hypothesis about cellular signaling with nitric oxide in the earliest life forms in evolution

We propose that nitric oxide participated as an extracellular and intracellular messenger in the early evolution of life. From a toxic and noxious substance it evolved into an important material for cellular communication and regulation with unique chemistry and properties. The presence of some nitric oxide complexes in extraterrestrial samples may support evidence for life forms in the past or present. Although nitric oxide probably participated in the evolution and maintenance of life, if pollution continues at an ever-increasing rate, it could also end life on the planet as we know it today.

Hormone and methylxanthine action on breast epithelial cells

Human mammary carcinoma cells and normal mouse breast epithelial cells desensitized as the result of treatment with beta-adrenergic agonists. Accumulation of cAMP in the same cells was affected only slightly by caffeine and there was no detectable desensitization or hypersensitization as a result of that treatment. However, as in many other cell types, caffeine was an effective inhibitor of adenosine action. These observations do not support the hypothesis made by Minton and his co-workers (1), that treatment of breast epithelial cells with agents that increase cAMP accumulation leads to hypersensitization rather than desensitization.