OPRM1, the only identified gene encoding the mu-opioid receptor (MOR), (Chen et al. 1993) generates a remarkable diversity of MOR subtypes (Pasternak 2001). Most are C-terminal variants that are generated through alternative splicing between exon 3 and multiple downstream exons (Pasternak and Pan 2013). The discovery of these and other splice variants (e.g., exon 11 variants with extensive 5' splicing, exon 1 variants containing a single transmembrane domain) provide a molecular biology framework for the concept of multiple mu-opioid receptors inferred from much earlier pharmacological studies (Wolozin and Pasternak 1981).
Some functional differences have been identified among MOR splice variants. Many opioids show marked differences in efficacy and potency at C-terminal MOR variants (Bolan et al. 2004; Pasternak et al. 2004; Pan et al. 2005a,b). Morphine-induced internalization also varies among the C-terminal variants, for example, morphine given intracerebroventricularly internalizes mMOR-1C1 but not mMOR-1 (Abbadie and Pasternak 2001).
Differences among MOR splice variants also vary with respect to agonist selectivity. Knockout of exon 1 abolishes morphine analgesia, but not that of either morphine-6β-glucuronide or heroin (Schuller et al. 1999). Conversely, knockout of exon 11 greatly reduces analgesic responsiveness to morphine-6β-glucuronide and heroin, but not to morphine or methadone (Pan et al. 2009). MOR transcripts also vary in their expression patterns and levels of expression across brain regions (Abbadie et al. 2000b,c), implying region-specific processing, and differing physiological functions. MOR splice variants have also been implicated to explain incomplete cross-tolerance among agonists and variability of agonist potency/effectiveness in patients.
Strikingly, MOR splice variants have not been studied with respect to the well-documented sexual dimorphism in MOR antinociception in laboratory animals (Cicero et al. 1996, 1997; Boyer et al. 1998; Krzanowska and Bodnar 1999; Cook et al. 2000; Barrett et al. 2002; Peckham and Traynor 2006) and, in humans, (Sarton et al. 2000). This laboratory has recently reported two striking manifestations of sexually dimorphic properties of MOR in spinal cord: (i) in opioid naïve spinal tissue the propensity of MOR to heterodimerize with the kappa opioid receptor (KOR) is substantially greater in females than males, the magnitude of which is dependent on stage of estrus cycle, that is, spinal levels of the MOR/KOR heterodimer during proestrus are significantly greater than during diestrus (Chakrabarti et al. 2010; Liu et al. 2011) and (ii), chronic systemic morphine treatment augments MOR Gs coupling in the spinal cord of males but not females (Chakrabarti et al. 2012).
The structure of MOR could be a determinant of both the facility with which spinal MOR heterodimerizes with KOR as well as the G proteins with which MOR preferentially couples. Accordingly, we investigated the influence of stage of estrus cycle and chronic systemic morphine (in both males and females) on spinal cord levels of mRNA encoding MOR splice variants. We selected MOR variants representing each of the two main classes of splicing that have been identified in rodents and humans. For example, MOR-1C1 and MOR-1B2 represent variants resulting from extensive 3' splicing at the end of the C-terminus but containing the seven transmembrane spanning domains and the N-terminus of MOR-1. MOR-1G1 represents variants resulting from extensive 5' splicing containing six transmembrane spanning domains and an altered N-terminus.
Results reveal a striking sexual dimorphism in the ability of chronic systemic morphine to selectively influence the generation of spinal MOR variants. Implications for sex-based opioid tolerance mechanisms are discussed.
- Top of page
- Materials and methods
- Acknowledgments and conflict of interest disclosure
This is the first report to suggest that chronic systemic treatment with morphine, sufficient to produce profound analgesic tolerance and dependence at the time of sacrifice (Villar and Bhargava 1992), not only alters the content of spinal cord mRNAs that encode specific MOR splice variants but also does so in a sex-dependent fashion. Sexually dimorphic alterations in the relative abundance of selective MOR variants, which could have unique signaling attributes, provides a molecular context for understanding sex-based accommodations to chronic morphine that have been reported previously (Chakrabarti et al. 2012).
In spinal cord of males, but not females, chronic morphine was found to augment mRNA encoding rMOR-1C1 and rMOR-1B2. Poor correlation between steady-state isoform-specific mRNA and the protein it encodes has been reported (Mons and Cooper 1994). Nevertheless, it seems reasonable to conjecture that a selective increase in the mRNA encoding rMOR-1B2 and rMOR-1C1 likely suggests a parallel change in the content and activity of rMOR-1B2 and rMOR-1C1 receptor protein. This inference is supported by saturation binding analyses, which revealed that a small but significant increase in MOR receptor density accompanies the increment in rMOR-1B2/rMOR-1C1 mRNA following chronic systemic morphine.
Analgesic tolerance formation was not directly assessed in this study. However, the regimen of chronic systemic morphine exposure is well established to produce profound analgesic tolerance and dependence at the time of animal sacrifice (Villar and Bhargava 1992). Importantly, sexual dimorphism in the magnitude of analgesic tolerance remains controversial in both rats (Craft et al. 1999; Holtman et al. 2004) and mice (Kest et al. 2000; Hosseini et al. 2010). Thus, it is highly unlikely that differential magnitudes of analgesic tolerance in males versus females per se could explain the observed sex-dependent effects of chronic morphine on spinal rMOR-1B2/rMOR-1C1 mRNA. Moreover, as the magnitude of differences in MOR agonist efficacy and potency among splice variants is relatively small in comparison to the magnitude of tolerance that can develop, it also seems unlikely that presumptive tolerance-associated up-regulation of rMOR-1C1 and rMOR-1B2 per se would significantly contribute to tolerance by altering the agonist response profile to mu agonists (Pasternak and Pan 2013).
C-terminal splice variants of MOR have been reported to differ in their endocytotic response to chronic morphine (Koch et al. 1998, 2001). However, given the well-documented disconnects between the ability of opioid agonists to induce MOR endocytosis and/or G protein uncoupling versus analgesic tolerance (Sternini et al. 1996; Madia et al. 2012), the differential intracellular trafficking of C-terminal splice variants, per se, is not likely to be a primary factor modulating morphine analgesic tolerance in vivo.
There are, however, unique structural characteristics of rMOR-1C1 and rMOR-1B2, which reside in the composition of the tip of the C-terminus that could influence their functionality. The C-terminal twelve amino acids (aa) of rMOR-1, LENLEAETAPLP, are replaced in rMOR-1C1 by 65 aa, PALAVSVAQIFTGYPSPTHGEKPCKSYRDRPRPCGRTWSLKSRAESNVEHFHCGAALI-YNNVNFI [encoded for by exon 7 (bp 44966586-4496614), exon 8 (bp 45027017-450027076), exon 9a (bp 45034092-45034145) and exon 9b (bp 45032373-45032473) of the OPRM gene located on chromosome 1]. In rMOR-1B2, the C-terminal twelve aa of rMOR-1 are replaced by seven aa, EPQSVET, (Zimprich et al. 1995; Pan 2005; Pan et al. 2005b), encoded by exon 5 [exon 5b (bp 44845374- 44847120) and 5a (bp 44847121-44847400) of the OPRM1 gene]. Importantly, the unique C-terminus of rMOR-1B2/rMOR-1C1 contains numerous consensus sequences for phosphorylation by a variety of kinases that are not present in rMOR-1. (e.g., protein kinase C, protein kinase A, casein kinase II, ribosomal factor S6 kinase, cyclin-dependent kinase 5, and cGMP-dependent protein kinase, etc).
Phosphorylation patterns of G protein coupled receptors not only influence the nature of the particular G protein to which receptors couple (Lefkowitz et al. 2002; Zamah et al. 2002), but also influence biased agonism, agonist-specific stabilization of receptor conformations that preferentially activate specific downstream signaling pathways (Kahsai et al. 2011). Moreover, the induction by chronic morphine of newly synthesized MOR variants and the intracellular trafficking of newly synthesized MORs could themselves constitute signals that herald novel adaptations. These considerations suggest that the male-specific up-regulation of spinal rMOR-1B2/rMOR-1C1 in response to chronic morphine could substantially influence the aggregate signaling consequences of mu-opioids. In this regard, it is relevant to note that a derivative of the endogenous MOR ligand, endomorphin 2, has been shown to elicit analgesic tolerance and related biochemical sequelae (Muranyi et al. 2013). Thus, male-specific adaptations to chronic morphine described herein could be relevant to physiological adaptations to stimuli that result in sustained utilization/activation of the endogenous MOR analgesic system. This is particularly so as MOR-1C1 has been shown to be present in laminae I/II of rat and human spinal cord (Abbadie et al. 2000a,b), areas rich in endomorphin 2 (Schreff et al. 1998; Martin-Schild et al. 1999).
Recently, we reported on the heterodimerization of MOR with KOR in spinal cord that was not only sex-dependent but also dependent on the stage of the estrus cycle (Chakrabarti et al. 2010; Liu et al. 2011), the bases for which remain unknown. We had speculated that sex and estrus cycle dependence of MOR/KOR heterodimerization resulted from sex- and estrus cycle-dependent expression of spinal MOR variants. However, this is unlikely in light of the current finding that spinal expression levels of mRNA encoding each of the six splice variants studied did not vary between diestrus and proestrus females.
Surprisingly, spinal levels of mRNA encoding rMOR-1B2 and rMOR-1G1 were significantly higher in the spinal cord of diestrus female vs. male rats. The physiological/pharmacological significance of this difference remains to be defined. However, it should be noted that differences in analgesic responsiveness among diestrus, proestrus, and male rats have been reported (Boyer et al. 1998). Furthermore, as diestrus is the longest single stage of the estrus cycle, studies comparing males and (non-cycled) females most likely involve females in diestrus. Therefore, differences in spinal rMOR-1B2/rMOR-1G1 could undergird, at least in part, male-female differences in opioid antinociception that have been reported, definitive proof of which will require the development of MOR variant-selective agonists and antagonists.
In summary, the sexually dimorphic ability of chronic systemic morphine to up-regulate spinal rMOR-1C1 and rMOR-1B2 mRNA could suggest that spinal MOR variants are differentially recruited in males versus females to mediate sex-dependent adaptations to chronic morphine, for example, (Chakrabarti et al. 2012). Knowledge of unique signaling properties of rMOR-1C1 and rMOR-1B2, which reside in the composition of the tip of the C-terminus, could point the way to defining the molecular determinants of sex-dependent tolerance and withdrawal mechanisms and treatments thereof.