Perspectives in Pharmacology The Ambiguities of Opioid Tolerance Mechanisms: Barriers to Pain Therapeutics or New Pain Therapeutic Possibilities

Identification of adaptations to chronic morphine that are causally associated with opioid tolerance formation has long been intensely pursued by the opioid research community. There is an impressive array of components of signaling pathways that are influenced by chronic opioid administration. This underscores the importance to tolerance mechanisms of the complex interplay of cellular adaptations that are downstream from the opioid receptor. A major impetus for this research remains the need to develop opioid agonists that are potent and efficacious activators of analgesic mechanisms without triggering opioid tolerance-producing adaptations. Implicit in most models of opioid tolerance is that their underlying mechanisms are invariant and independent of the system in which they have been observed. Reports that prior acute morphine treatment and pain could influence tolerance mechanisms were not understood on mechanistic levels and, consequently, were not incorporated into commonly used models of opioid tolerance. The recent demonstration that adenylyl cyclase/cAMP-related cellular adaptations to chronic morphine depend on cell state demonstrates that ongoing cell physiology is a critical determinant of tolerance mechanisms. The plasticity and pliability of cellular adaptations that mediate tolerance formation indicate that mechanisms underlying opioid analgesic tolerance could be a moving target. Although this might represent a daunting barrier to developing antitolerance pharmacotherapies, appreciation of this complexity could lead to the development of new pharmacotherapeutic approaches. Of the armamentarium of pharmacological agents available to manage postsurgical and neuropathic pain, morphine and its congeners remain among the most widely employed. Nevertheless, the propensity of narcotics to induce analgesic tolerance (operationally defined as a reduction in responsiveness to an agent after repeated exposure) profoundly limits their therapeutic usefulness. It is not surprising that an enormous research effort has been expended over the years toward elucidating the mechanistic underpinnings of opioid tolerance. A major impetus for this research remains the need to develop opioid agonists that are potent and efficacious activators of analgesic mechanisms without triggering opioid tolerance-producing adaptations. Central to this effort is the identification of tolerance substrates, i.e., adaptations causally associated with tolerance formation, on all organizational and functional levels. In this pursuit, conceptual formulations of tolerance become critical, because they determine the contour map guiding the journey as well as the resting stops that are targeted along the way. It is certainly humbling to recognize that the search for nontolerance-forming potent narcotic analgesics, alone or in combination with adjunctive pharmacotherapy, has been ongoing for at least the past 50 years without notable success. Moreover, this failure has occurred in the face of huge advances in our molecular and cellular knowledge of opioid receptors and the cell signaling pathways that are activated by them. This could indicate that opioid analgesic and tolerance mechanisms are so inextricably intertwined that they cannot be differentially targeted. Alternatively, our conceptual models of tolerance might not be sufficiently inclusive to provide the perspectives needed to develop opioid-based medications with which to treat pain in the absence of tolerance. This perspective will advance the concept that models of tolerance need to embrace the influence of ongoing physioArticle, publication date, and citation information can be found at http://jpet.aspetjournals.org. doi:10.1124/jpet.107.135533. ABBREVIATIONS: AC, adenylyl cyclase; GPCR, G protein-coupled receptor; MOR, -opioid receptor; MOR-CHO, Chinese hamster ovary cells stably expressing -opioid receptor; AC1-MOR-CHO, MOR-CHO overexpressing AC1; AC2-MOR-CHO, MOR-CHO overexpressing AC2; PKC, protein kinase C; CNS, central nervous system. 0022-3565/08/3253-709–713$20.00 THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS Vol. 325, No. 3 Copyright © 2008 by The American Society for Pharmacology and Experimental Therapeutics 135533/3343692 JPET 325:709–713, 2008 Printed in U.S.A. 709 at A PE T Jornals on A uust 9, 2017 jpet.asjournals.org D ow nladed from logical state on opioid tolerance mechanisms that are utilized. Because this article is not intended to be a review, aspects of opioid tolerance have been selected that advance the idea that opioid tolerance mechanisms are pliable and context-dependent. Cellular Tolerance versus Adaptations Involving Neuronal Networks Adaptations to chronic morphine that occur on the level of individual neurons all occur within neuronal networks, which can amplify (or diminish) the functional consequences of those adaptations. A good exemplar of this is the ability of chronic opioids to increase activity of glutamatergic neurons and consequently augment N-methyl-D-aspartate receptor activity (Mao, 1999). This illustrates the important consideration that chronic morphine can induce tolerance adaptations in neurons that do not bear opioid receptors, which might amplify or compensate for the consequences of cellular adaptations that occur within individual neurons. This greatly complicates the generation of tolerance models of translational utility. This notwithstanding, delineation of cellular adaptations to chronic morphine enables the identification of putative pharmacologic cellular targets for the amelioration of tolerance development. It also can facilitate the development of organizing concepts and principals that would apply to chronic morphine-induced adaptations on cellular as well as network organizational levels. Translational Utility of Tolerance Models It is important to keep in mind that opioid tolerance is not a unitary entity and that variable mechanisms might underlie the development of tolerance to each of the multiple effects of the same agonist in the same or different experimental systems. It is also essential to remember that there are multiple forms of tolerance, each of which could be mediated via a different subset of adaptations. These can often be differentiated by their specificity and temporal characteristics. For example, opioid receptor homologous desensitization resulting from G protein uncoupling has a very rapid (minutes) onset (Law and Loh, 1999), whereas adaptations involving up-regulation of the adenylyl cyclase (AC) cascade requires hours for full manifestation (Nestler et al., 1994; Nestler and Aghajanian, 1997). Yet another proposed adaptation to chronic morphine that is heterologous and that has a much more delayed onset involves down-regulation of the sodium pump and a reduction in its electrogenic contribution to membrane potential (Fleming, 1999). The existence of i) multiple organizational levels on which opioid tolerance can occur and ii) multiplicity of functions influenced by opioids require that mechanistic models attempting to define tolerance be appropriately constrained and qualified in relationship to the system under study. Impairment of Opioid Receptor Functionality Most models of opioid tolerance frequently employed revolve around the conceptual rubric that it is the direct result of the actual loss of specific opioid receptor-mediated signaling, i.e., opioid receptor desensitization. This desensitization is frequently envisioned to involve a reduction in spare opioid receptors (Chavkin and Goldstein, 1984), increased opioid receptor internalization (Bohn et al., 2000), decreased opioid receptor density (Chakrabarti et al., 1995), and altered content of G proteins (Ammer and Schulz, 1995). A pivotal aspect of such models is the enhanced phosphorylation of the opioid receptor via G protein receptor kinase that accompanies its activation and is a prelude to it forming a complex with -arrestin (Pei et al., 1995; Kovoor et al., 1997; Appleyard et al., 1999). This results in its targeting to clathrin-coated pits, G protein uncoupling, and its subsequent internalization and intracellular trafficking to subcellular compartments, e.g., lysozomes where receptor degradation can occur. These events directly parallel those that have been extensively described for the 2-adrenergic receptor, and they are shared by most, if not all, G protein-coupled receptors (GPCRs). The relevance of these events and models that revolve around them to in vivo pharmacological opioid tolerance is certainly suggested by the coincidence of the temporal characteristics of opioid receptor phosphorylation and G protein uncoupling with the onset of the acute loss of opioid receptor functionality, i.e., receptor desensitization or “acute tolerance” (Zhang et al., 1996; Appleyard et al., 1997). It is important that the above mechanistic formulations of opioid tolerance are invariably thought of as being invariant responses to chronic morphine, independent of ongoing physiological state. Post-Opioid Receptor Adaptations to Chronic Morphine Numerous post-opioid receptor cellular adaptations to chronic morphine have been identified that challenge the centrality of the uncoupling theory of opioid tolerance by suggesting alternative mechanisms. Furthermore, these underscore the complex interplay of tolerance mechanisms with cell physiology and the need for more complex working models that reflect it. Many tolerance-related adaptations pertain to AC/cAMP (Duman et al., 1988; Nestler and Tallman, 1988; Guitart and Nestler, 1989; Lane-Ladd et al., 1997; Kim et al., 2006) and protein kinase C (PKC) (Mao et al., 1995; Mayer et al., 1995; Narita et al., 1995; Wang et al., 1996; Wei and Roerig, 1998; Zeitz et al., 2001) signaling pathways. More recently, chronic morphine was shown to up-regulate specific AC isoforms (Chakrabarti et al., 1998a; Rivera and Gintzler, 1998), increase phosphorylation of AC (Chakrabarti et al., 1998b) and the G subunit of G proteins (Chakrabarti et al., 2005b), and increase association of the -opioid receptor (MOR) with Gs (Chakrabarti et al., 2005a; Chakrabarti and Gintzler, 2007). It is noteworthy that all of these adaptations not only occur concomitantly but also have convergent signaling consequences; in the aggregate, they shift acute MOR-coupled signaling from AC inhibitory to stimulatory (Gintzler and Chakrabarti, 2006). These cellular adaptations to long-term morphine underscores that at least a subset of tolerance mechanisms does not cause the loss of opioid receptor functionality but rather the alteration of the consequences of opioid receptor activation. This is very revealing because it indicates that the protective function served by opioid tolerance formation, i.e., the reinstatement of initial steady-state conditions, does not result solely from unidirectional adaptations, e.g., restricted opioid receptor functionality but from the active assertion of compensatory opioid receptor-coupled cell signaling strategies. 710 Gintzler and Chakrabarti at A PE T Jornals on A uust 9, 2017 jpet.asjournals.org D ow nladed from Opioid Receptor Pleiotropy and Duality of Signaling Identification of the inter-related cellular adaptations to long-term morphine treatment highlighted above is a poignant reminder that formation of tolerance utilizes the flexibility that is inherent in receptor G protein signaling. Opioid receptor pleiotropy (tolerance-associated enhanced coupling to Gs) (Chakrabarti et al., 2005a) and the duality of G protein signaling via the G and G subunits (tolerance-associated enhanced AC G stimulatory AC signaling) (Chakrabarti et al., 1998a) are both recruited in response to persistent opioid receptor activation. GPCR pleiotropy and duality of G protein subunit signaling are pillars of cell signaling plasticity and underlie much of the richness and diversity of signaling that is characteristic of GPCRs. Thus, it should not be surprising that they also underlie many of the cellular adaptations that enable cell survival in the face of prolonged opioid exposure. Chronic morphine-induced enhanced opioid receptor pleiotropy and duality of signaling enable opioid tolerance mechanisms to be pliable, as reflected by the shift in MOR-coupled signaling from Gi /Go inhibitory to Gs /G stimulatory AC signaling (Gintzler and Chakrabarti, 2006). This multidimensionality of cellular adaptations to long-term morphine treatment demands the development of mechanistic models of opioid tolerance that include a much broader spectrum of adaptational mechanisms than has thus far been the case if they are to be medicinally relevant. Influence of Prior Treatment on Spinal Opioid Tolerance and Addiction Although not concluded at the time of publication, some early behavioral studies do indicate that tolerance adaptations could depend on physiological state. For example, there is provocative data predating the biochemical demonstration of the opioid receptor, that the antinociceptive effect of morphine can be reduced by a single dose of systemic morphine administered months earlier (Cochin and Kornetsky, 1964). More recently (Lim et al., 2005), it was shown that the rate of onset and the magnitude of antinociceptive tolerance increase with serial intrathecal morphine injections. The authors suggest that repeated cycles of morphine exposure produce sustained changes in the spinal cord that modulate the development of opioid tolerance to subsequent morphine exposure. The demonstration that prior history of morphineinduced plasticity can influence the magnitude of subsequently observed tolerance adaptations can be construed to indicate that at least some of the adaptations to chronic morphine are not set in stone and in fact resonate with evolving physiological state. However, in the absence of any formal direct demonstration of this concept, these phenomena remained an enigma. Behavioral studies conducted months after opioid withdrawal also support the notion that ongoing physiological state can be a major determinant of addiction predisposition. For example, morphine-dependent rats that had been successfully detoxified and showed no significant signs of morphine dependence consumed significantly larger volumes of morphine solution than opiate naive controls and had recurrence of morphine tolerance and dependence (Dai et al., 1984). Subsequently, it was suggested (Bartoletti et al., 1987) that modification of the neuronal mechanism subserving the excitatory component of the action of opiates by chronic morphine treatment that had occurred months earlier could represent a neurobiological basis for recidivism in addicts. Although the mechanisms responsible for the pliability of opioid responsiveness have remained unidentified, such observations further support the notion of the state dependence of the processes of tolerance and dependence. Dependence of Cellular Opioid Tolerance Mechanisms on Cell State A review of the opioid tolerance literature reveals that, although cellular biochemical parameters of morphine administration are often considered in studies of tolerance, the influence of ongoing physiology and cell state are not. The complexity and multidimensionality of cellular mechanisms underlying opioid tolerance are underscored by the recent report that a subset of inter-related cell signaling adaptations to chronic morphine exposure does not represent a fixed set of adaptations; rather, these inter-related cell signaling adaptations are cell state-dependent (Shy et al., 2008). This notion was directly put to the test by comparing AC/ cAMP-related adaptations to long-term morphine treatment among Chinese hamster ovary cells (CHO) stably expressing MOR (MOR-CHO) and MOR-CHO overexpressing either AC2 (AC2-MOR-CHO) or AC1 (AC1-MOR-CHO). These cells manifest qualitatively opposite consequences of acute MOR activation as a result of differences in the relative abundance of specific AC isoforms (Federman et al., 1992; Tsu et al., 1995; Yoshimura et al., 1996) that are differentially regulated by G (Tang and Gilman, 1991). The qualitative difference in the consequences of acute MOR activation (AC inhibition versus stimulation) has a profound effect on the manifestation of multiple, complementary AC-related adaptations to chronic morphine, many of which are diametrically opposite (Shy et al., 2008). It is striking that none of the AC/cAMP-related adaptations to chronic morphine observed in MOR-CHO and AC1-MORCHO (increased AC and G phosphorylation, membrane protein kinase C translocation and MOR Gs association (Chakrabarti et al., 1998b, 2005a,b; Chakrabarti and Gintzler, 2003) is observed in AC2-MOR-CHO. Instead, overexpression of AC2 negates the increment in G phosphorylation and PKC translocation and reverses the increment in AC phosphorylation and MOR Gs association to a decrement (Shy et al., 2008). These experiments formally tested the inter-relatedness of tolerance adaptations and cell state. Results directly demonstrate that adaptational strategies in the AC/cAMP signaling pathway elicited by chronic morphine are not hard-wired but instead are conditional on cell state. In this particular case, the default acute responsiveness of cells to MOR activation is a determinant of the mechanisms harnessed by cells to cope with the persistent activation of MOR. In cells in which acute MOR activation inhibits AC activity (MOR-CHO, AC1-MORCHO; Fig. 1, left), chronic morphine elicits adaptations that augment a stimulatory arm of MOR-G protein coupling. In contrast, in cells manifesting acute stimulatory AC responsiveness to MOR (AC2-MOR-CHO; Fig. 1, right), the same treatment with morphine elicits adaptations that enhance AC inhibitory responsiveness. It is noteworthy that the subPlasticity of Tolerant Mechanisms 711 at A PE T Jornals on A uust 9, 2017 jpet.asjournals.org D ow nladed from