Refugee Receptors Switch Sides

See allHide authors and affiliations

Science  26 Mar 2010:
Vol. 327, Issue 5973, pp. 1586-1587
DOI: 10.1126/science.1188538

Human heart muscle cells (cardiomyocytes) express two different types of β adrenergic receptor: β1AR and β2AR. Both respond to the same hormones and couple to the same biochemical signaling effectors, yet they elicit different cellular responses. On page 1653 of this issue, Nikolaev et al. (1) show that β1ARs are widely distributed at the plasma membrane (crest) of a normal cardiomyocyte, whereas β2ARs localize to transverse-tubular (T-tubule) regions at the cell surface, thus segregating their signaling. Remarkably, failing cardiomyocytes exhibit membrane remodeling that disrupts compartmentalization of βARs. This may be a primary mechanism for the pathological effects of βAR signaling in heart failure and disease.

β2AR is the prototype of receptors that span the membrane seven times and are associated with a heterotrimeric guanine nucleotide–binding protein (G protein). Before 1990, the conventional view was that norepinephrine and epinephrine (adrenaline) secreted by neuronal and adrenal tissues activate βARs that are coupled to the stimulatory subtype of G protein (Gs). This then triggers the enzyme adenylate cyclase to produce cyclic adenosine monophosphate (cAMP), a signaling molecule that was thought to diffuse throughout the cytoplasm to activate intracellular targets (2). However, this view changed after the description of macromolecular “signalosomes,” complexes consisting of β2ARs or other receptors and proteins that transduce and modulate the cAMP signal (3, 4). Such compartmentalization of components has emerged as a ubiquitous biological theme for orchestrating cell signal transduction to confer precision and specificity of response.

Receptor redistribution.

Cell membrane remodeling in heart failure produces T-tubule loss, displacing β2ARs to cell crests where they exhibit β1AR-like nonlocalized cAMP signaling. A key question is whether β2ARs that translocate to cell crests exert the same cardiotoxic effects as β1ARs. AC, adenylate cyclase; PDE, phosphodiesterase.


Upon stimulation, β1ARs increase the force and frequency of cardiac muscle contraction by inducing cAMP-dependent phosphorylation of factors that regulate the intracellular concentration of calcium. This consequently affects calcium-dependent proteins that control the contraction of myofilaments (see the figure). However, in heart failure, β1AR induces cell remodeling and programmed cardiomyocyte death, which contribute to the tissue deterioration seen in heart failure. Thus, treatment with β1AR antagonists protects against morbidity and mortality in heart failure, even though cardiac pump function is depressed (5). The pathophysiological role of cardiac β2ARs is less clear. Although they also couple to Gs and adenylate cyclase, cardiac β2AR-generated cAMP signaling is not transmitted to calcium-regulatory and myofilament proteins that control contraction. Instead, β2AR activity opposes β1AR signaling by switching sequentially from Gs to an inhibitory G protein (Gi) that blocks adenylate cyclase (6). This switching results in the production of different cAMP pools in the cardiomyocyte (7, 8).

Selective generation of intracellular pools of cAMP implies subcellular compartmentalization of signals. Indeed, fluorescence resonance energy transfer (FRET) techniques for real-time live-cell imaging of intracellular signals have localized βAR-stimulated cAMP to discrete subcellular microdomains of the cardiomyocyte T-tubule system (4). T-tubules are plasma membrane invaginations containing many proteins that couple plasma membrane depolarization (excitation) to calcium-mediated myofilament shortening (contraction). β2ARs in normal cardiac myocytes are associated with T-tubules, thus generating spatially restricted cAMP production. This signaling is locally inactivated by phosphodiesterases located in T-tubules, which limits the diffusion of cAMP. By contrast, β1ARs produce cytosolic cAMP that diffuses the distance of a dozen or more myofilament sarcomeres to enhance contractility (8). Nikolaev et al. combined FRET with live-cell scanning ion conductance microscopy to colocalize β2ARs and their cAMP signaling events. By comparing spatiotemporal patterns of signaling in cardiomyocytes from normal and failing hearts of rats and from mice lacking either β1AR or β2AR, they observed redistribution of β2ARs from T-tubules to cell crests, the normal province of β1ARs. Consequently, β2ARs in failing cardiac myocytes acquired a pattern of broad cAMP signaling more characteristic of β1ARs. Because β1AR signaling is linked to cardiotoxicity in heart failure, the transition of β2ARs from a discrete to a generalized pattern of cAMP signaling may accelerate heart failure.

Altered excess βAR trafficking is characteristic of heart failure (9). Prolonged adrenergic stimulation induces receptor down-regulation by internalization and degradation. Even if internalized receptors are not degraded, internal sequestration uncouples them from signaling effectors (desensitization). But βAR down-regulation in human heart failure is specific to β1ARs, whereas β2ARs are spared. The net result is a change in the relative proportion of β1ARs versus β2ARs from ∼4:1 in normal hearts to ∼3:2 in heart failure (10). Furthermore, β1ARs that remain in failing hearts are largely uncoupled from Gs–adenylate cyclase–cAMP signaling because of desensitization. Receptor internalization also appears to be necessary for switching β2AR from Gs to Gi signaling (6, 11). Because the net effect of β1AR down-regulation and desensitization combined with β2AR switching to Gi signaling is to increase cardiac β2AR signaling at the expense of β1AR signaling, these events have widely been assumed to benefit the failing heart (9).

A pathological role for T-tubule loss in heart failure has been attributed to compromised excitation-contraction coupling (12) and to the redistribution of voltage-dependent calcium channels (13). By linking β2AR redistribution in heart failure with plasma membrane remodeling and the loss of T-tubules, Nikolaev et al. add another layer of complexity to β2AR regulation in heart failure. Future studies will determine whether diffusible cAMP signals generated by displaced β2ARs couple to contraction and cell death pathways in the same manner as those generated by β1ARs.

Compartmentalization of β2ARs and restriction of their cAMP signals has not been observed in cells that normally do not have T-tubules or analogous intracellular membrane structures. T-tubules may therefore not only define β2AR signaling in normal versus failing cardiomyocytes, but also in cardiomyocytes versus other cell types. This establishes an imperative to study receptor signaling in cells directly relevant to the disease of interest and emphasizes necessary caution in extrapolating findings from one cell type to another. Likewise, there are important species differences between humans and rodents with respect to T-tubule density, β1AR/β2AR expression ratio, and in the capacity for β2ARs to undergo Gs-to-Gi signal switching. For these reasons, it is premature to consider changing the clinical use of compounds that block βARs in human heart failure. However, the finding of Nikolaev et al. that β2ARs in heart failure can adopt pathological characteristics of β1ARs when displaced from T-tubules to cell crests does support the need to revisit prior clinical studies that concluded that β1AR-specific blockade is desirable in human heart failure.


View Abstract

Stay Connected to Science

Navigate This Article