Perspectives in Pharmacology Molecular Mechanisms for Heterologous Sensitization of Adenylate Cyclase

The nine membrane-bound isoforms of the enzyme adenylate cyclase (EC 4.6.1.1) are highly regulated by neurotransmitters and drugs acting through G protein-coupled receptors to modulate intracellular cAMP levels. In general, acute activation of G s-coupled receptors stimulates cAMP accumulation, whereas acute activation of G i/o-coupled receptors typically inhibits cAMP accumulation. It is also well established that persistent activation of G-protein coupled receptors will alter subsequent drug-modulated cAMP accumulation. These alterations are thought to represent cellular adaptive responses following prolonged receptor activation. One phenomenon commonly observed, heterologous sensitization of adenylate cyclase, is characterized by an enhanced responsiveness to drugstimulated cAMP accumulation following persistent activation of G i/o-coupled receptors. Heterologous sensitization of adenylate cyclase was originally proposed to explain tolerance and withdrawal following chronic opiate administration and may be a mechanism by which cells adapt to prolonged activation of inhibitory receptors. Such an adaptive mechanism has been suggested to play a role in the processes of addiction to and withdrawal from many drugs of abuse and in psychiatric disorders including schizophrenia and depression. Although the precise mechanisms remain unknown, research over the last decade has led to advances toward understanding the molecular events associated with heterologous sensitization of recombinant and endogenous adenylate cyclases in cellular models. These events include the pertussis toxin-sensitive events that are associated with the development of heterologous sensitization and the more recently identified G s-dependent events that are involved in the expression of heterologous sensitization. Historical Perspective Acute activation of G i/o-coupled receptors inhibits cAMP accumulation, whereas prolonged activation enhances drugstimulated cAMP accumulation. This enhanced responsiveness was first observed following persistent activation of the -opioid receptor in the laboratory of Dr. Marshall Nirenberg (National Institute of Mental Health, Bethesda, MD), who proposed that the increased responsiveness was a mechanism of opiate tolerance and dependence (Sharma et al., 1975). This phenomenon has since been described using many different names, including cAMP overshoot, supersensitivity, superactivation, supersensitization, and heterologous sensitization of adenylate cyclase (EC 4.6.1.1). The term heterologous sensitization will be used throughout this article to describe observations where persistent activation of a G i/o-coupled receptor induces an enhanced response to drug-stimulated cAMP accumulation (Fig. 1). Those initial observations from the Nirenberg laboratory (Sharma et al., 1975) prompted a number of subsequent investigations aimed at determining the receptor and tissue specificity of heterologous sensitization and identifying the molecular mechanisms responsible for this phenomenon. These studies revealed that persistent activation of several G i/o-coupled receptors (including opioid, 2-adrenergic, adenosine, somatostatin, and muscarinic receptors) induces heterologous sensitization in both neuronal and non-neuronal cellular models (see review by Thomas and Hoffman, 1987). Based on the results from these studies, Thomas and Hoffman (1987) proposed the following model: chronic agonist stimulation of a G i-coupled receptor induces heterologous sensitization through a pertussis toxin-sensitive G protein, invoking an unknown mechanism that may ultimately alter G s, G i, or adenylate cyclase to contribute to the enhanced cAMP response. Furthermore, this mechanism does not involve desensitization or tolerance of the G i-coupled receptor, and the This work was supported by the National Institute of Mental Health, MH60397, the National Alliance for Research on Schizophrenia and Depression, and Purdue University. 1 I have used adenylate cyclase as the common name of the enzyme EC 4.6.1.1 based on the recommendation of the International Union of Biochemistry and Molecular Biology (IUBMB) (http://www.chem.qmw.ac.uk/iubmb/enzyme/EC4/6/1/1.html). Other sources that list adenylate cyclase as the preferred common name include Dorland’s Medical Dictionary, the Sourcebook of Enzymes by White and White, Stedman’s Medical Dictionary (27th edition), and the Oxford Dictionary of Biochemistry and Molecular Biology. ABBREVIATIONS: AC, adenylate cyclase; PKA, cAMP dependent protein kinase A. 0022-3565/02/3021-1–7$7.00 THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS Vol. 302, No. 1 Copyright © 2002 by The American Society for Pharmacology and Experimental Therapeutics 35105/995628 JPET 302:1–7, 2002 Printed in U.S.A. 1 at A PE T Jornals on M ay 8, 2017 jpet.asjournals.org D ow nladed from precise role of cAMP inhibition is unknown (Thomas and Hoffman, 1987). Although the general characteristics of heterologous sensitization proposed by Thomas and Hoffman (1987) are consistent with more recent studies of G i/o-coupled receptors, a number of advances have provided additional information regarding the molecular mechanisms of heterologous sensitization using a variety of cellular models. The cloning and biochemical characterization of the nine membrane-bound adenylate cyclase isoforms has revealed that different isoforms have distinct regulatory properties (Taussig and Zimmerman, 1998). Several research groups have sought to exploit the differences among isoforms to identify mechanisms of heterologous sensitization, and the results of their studies suggest that sensitization is isoform-dependent so that the characteristics of heterologous sensitization in a given cell depend on the complement of adenylate cyclase isoforms present in the cell (Thomas and Hoffman, 1996; Watts and Neve, 1996; Avidor-Reiss et al., 1997; Cumbay and Watts, 2001). In the present article, the mechanisms of heterologous sensitization in neuronal and non-neuronal cultured cells will be discussed as they relate to the early (development) and the late (expression) events associated with heterologous sensitization (Fig. 1). The development of heterologous sensitization will refer to events that are more closely associated with the persistent stimulation of the G i/o-coupled receptor (e.g., activation of G i/o proteins). The expression of heterologous sensitization will refer to events that influence cAMP accumulation in response to G s-coupled receptor agonists, forskolin (a direct activator of adenylate cyclase), or selective activators of adenylate cyclase isoforms (Cumbay and Watts, 2001). Studies over the last 15 years have focused on the mechanisms involved in the development of heterologous sensitization, and more recent studies have identified potential mechanisms specifically involved in the expression of heterologous sensitization. Although the present article will often focus on heterologous sensitization following persistent stimulation of the D2 dopamine receptor (Fig. 2), it is likely that the mechanisms associated with one G i/o-coupled receptor are similar to those associated with other G i/o-coupled receptors, including -opioid, CB1 cannabinoid, 2 adrenergic, M2 and M4 muscarinic, A3 adenosine, and 5-hydroxytryptamine1A serotonin receptors (Avidor-Reiss et al., 1995; Hensler et al., 1996; Thomas and Hoffman, 1996; Palmer et al., 1997; Nevo et al., 1998; Rhee et al., 2000). G Protein Subunit Specificity for Heterologous Sensitization Pertussis toxin treatment prevents heterologous sensitization of both endogenous and recombinant adenylate cyclases in several cellular models (Watts and Neve, 1996; Palmer et al., 1997; Watts et al., 1998, 1999; Rhee et al., 2000; Rubenzik et al., 2001). Because pertussis toxin prevents receptor coupling to G i1, G i2, G i3, and G o in a nondiscriminating fashion, one important question is which pertussis toxinsensitive G protein is responsible for heterologous sensitization. This question has been investigated for D2L dopamine receptors and several of the opioid receptors (Watts et al., 1998; Tso and Wong, 2000; Tso and Wong, 2001). The D2L dopamine receptor study used viral-mediated gene delivery of individual genetically engineered pertussis toxin-resistant G proteins (G x*) to determine the G protein specificity for heterologous sensitization in NS20Y neuroblastoma cells (Watts et al., 1998). These experiments examined the ability of individual recombinant subunits G x* to rescue heterologous sensitization in pertussis toxin-treated cells. Selective activation of G o* by D2L receptors was found to be responsible for heterologous sensitization of forskolin-stimulated cAMP accumulation in NS20Y-D2L cells. In contrast, expression of mutant G i1*, G i2*, and G i3* subunits did not rescue heterologous sensitization in pertussis toxintreated cells. Similar studies in pertussis toxin-treated human embryonic kidney-D2L cells revealed that heterologous Fig. 1. Hypothetical model for the development and expression of G i/ocoupled receptor-induced heterologous sensitization. Acute activation of G i/o-coupled receptors inhibits cAMP accumulation, whereas prolonged activation enhances drug-stimulated cAMP accumulation following agonist removal. This enhanced responsiveness or heterologous sensitization of adenylate cyclase has also been referred to as cAMP overshoot, supersensitivity, supersensitization, and superactivation of adenylate cyclase. Fig. 2. A model of heterologous sensitization in cells expressing D2 dopamine receptors. Persistent dopamine (DA) stimulation of D2 dopamine receptors promotes the dissociation of G and subunits in a pertussis toxin-sensitive matter, which in turn, induces sensitization through a G s-dependent mechanism. The signaling events that follow the activation of the G i/o subunits and the release of the subunits (represented by a ? in the black box) are thought to produce an enhanced interaction between G s and adenylate cyclase (represented by green outline around G s). These drug-induced changes enhance cAMP accumulation that can be observed following stimulation of adenylate cyclase with forskolin or using isoproterenol (Iso) to stimulate the 1 adrenergic receptor (G s pathway). 2 Watts at A PE T Jornals on M ay 8, 2017 jpet.asjournals.org D ow nladed from sensitization is rescued by the expression of either G o* or G i1* (B. L. Wiens, V. J. Watts, and K. A. Neve, unpublished results). The abundant expression of G o throughout the central nervous system and recent studies demonstrating that D2 dopamine receptors couple efficiently to G o in native brain tissue suggest an important role for G o in heterologous sensitization (Jiang et al., 2001). Heterologous sensitization may also involve the simultaneous activation of multiple G i/o proteins. For example, the magnitude of selective G o*-induced heterologous sensitization seems to be reduced when compared with sensitization in cells where the entire endogenous G i/o pool was available (Watts et al., 1998). In fact, a role for multiple G subunits has been proposed for opioid-induced heterologous sensitization because the expression of G i1* only partially rescues -opioid receptor-induced sensitization, whereas expression of G i3* or a pertussis toxin-insensitive G i2/z* chimera has no effect (Tso and Wong, 2000, 2001). Furthermore, the expression of G i1*, G i3*, or a pertussis toxin-insensitive G i2/z* fails to rescue or -opioid receptor-induced sensitization (Tso and Wong, 2000, 2001). Although the importance of G o in opioid receptor-induced heterologous sensitization has not been examined, these results would be consistent with the contribution of multiple G subunits to sensitization. In addition, it is also possible that the specificity among G i/o subtypes for heterologous sensitization is governed by receptor/G protein coupling. In other words, all pertussis toxin-sensitive G proteins may have the ability to induce heterologous sensitization if activated sufficiently by a receptor. G Protein Subunit Expression and Heterologous Sensitization It has been proposed that persistent activation of G i/o induces heterologous sensitization through changes in the abundance of G s and G i/o. In some studies, chronic activation of G i/o-coupled receptors results in reduced levels of G i/o or increased levels of G s (Hadcock and Malbon, 1993; Watts et al., 1999), either of which would be predicted to enhance subsequent adenylate cyclase activity. In contrast, other studies have demonstrated that alterations in the abundance of G i/o or G s are not required for heterologous sensitization of adenylate cyclase (Palmer et al., 1997; Watts et al., 1999; Bayewitch et al., 2000). The ability of agonist treatment to induce changes in G i/o or G s levels seems to be dependent on the cellular model or cell line under investigation. Furthermore, the magnitude of and time course for changes in G i/o or G s do not seem to correlate directly with measures of heterologous sensitization. For example, heterologous sensitization can be observed within 15 min of agonist treatment (Thomas and Hoffman, 1996; Watts and Neve, 1996; Jones et al., 1987), and robust sensitization is commonly observed following 2 to 4 h of drug treatment (AvidorReiss et al., 1995; Watts and Neve, 1996; Palmer et al., 1997; Nevo et al., 1998; Cumbay and Watts, 2001; Watts et al., 2001). In contrast, agonist-induced changes in G levels, which probably involve changes in gene expression, typically require long-term receptor activation (Hadcock and Malbon, 1993; Watts et al., 1999). Together, these observations suggest that, although agonist-induced changes in G i/o or G s levels are not required for the development of heterologous sensitization, they may influence the magnitude or expression of heterologous sensitization. It is also possible that heterologous sensitization could involve a change in the localization of G proteins, as opposed to a change in overall protein levels. For example, persistent agonist treatment of -opioid receptors decreases the detergent solubility (and presumably lipid microdomain localization) of G i and the 1 subunit (Bayewitch et al., 2000). These changes occur rapidly ( 4 h) and correlate with agonistinduced heterologous sensitization of adenylate cyclase. Similar agonist-induced changes in the detergent solubility of G i and 1 were also observed in cells expressing -opioid and M4 muscarinic receptors. Additional studies are necessary to determine the precise role that these changes play in both the development and expression of heterologous sensi-