A novel alternatively spliced isoform of the mu-opioid receptor: functional antagonism

Abstract

Background: Opioids are the most widely used analgesics for the treatment of clinical pain. They produce their therapeutic effects by binding to mu-opioid receptors (MORs), which are 7 transmembrane domain (7TM) G-protein-coupled receptors (GPCRs), and inhibiting cellular activity. However, the analgesic efficacy of opioids is compromised by side-effects such as analgesic tolerance, dependence and opioid-induced hyperalgesia (OIH). In contrast to opioid analgesia these side effects are associated with cellular excitation. Several hypotheses have been advanced to explain these phenomena, yet the molecular mechanisms underlying tolerance and OIH remain poorly understood.

Results: We recently discovered a new human alternatively spliced isoform of MOR (MOR1K) that is missing the N-terminal extracellular and first transmembrane domains, resulting in a 6TM GPCR variant. To characterize the pattern of cellular transduction pathways activated by this human MOR1K isoform, we conducted a series of pharmacological and molecular experiments. Results show that stimulation of MOR1K with morphine leads to excitatory cellular effects. In contrast to stimulation of MOR1, stimulation of MOR1K leads to increased Ca2+ levels as well as increased nitric oxide (NO) release. Immunoprecipitation experiments further reveal that unlike MOR1, which couples to the inhibitory Galphai/o complex, MOR1K couples to the stimulatory Galphas complex.

Conclusion: The major MOR1 and the alternative MOR1K isoforms mediate opposite cellular effects in response to morphine, with MOR1K driving excitatory processes. These findings warrant further investigations that examine animal and human MORK1 expression and function following chronic exposure to opioids, which may identify MOR1K as a novel target for the development of new clinically effective classes of opioids that have high analgesic efficacy with diminished ability to produce tolerance, OIH, and other unwanted side-effects.

Figure 1

MOR1K expression and binding pattern. (A) The schematic diagram illustrates the exonic composition and relative positions of PCR primers designed to amplify the major MOR1 isoform and the newly identified alternative MOR1K isoform. The relative positions of translation initiation start and stop codons are designated by ATG and TGA, respectively. The predicted protein structure of MOR1 and MOR1K isoforms is schematically presented. Translation of the MOR1K variant results in a 6TM receptor, truncated at the N-terminus. (B) Real-time PCR was performed on total RNA samples from the human brain regions known to express MOR1. Primers specific for exons 1 and 2 were used to measure MOR1 and primers specific for exons 13 and exon 2 were used to measure MOR1K[32]. GAP3DH was used as a control for cDNA loading and PCR efficiency. (C) Confocal images of C-terminally MYC-tagged MOR1 or FLAG-tagged MOR1K overexpressed in HEK293 cells and stained with either Anti-MYC-Tag Antibody (Alexa Fluor 647 Conjugate) or Anti-DYKDDDDK Tag Antibody (Alexa Fluor 555 conjugate). Cells transfected with MOR1 showed membrane expression of receptor, while cell transfected with MOR1K express receptor only intracellular. (D) Confocal images of C-terminally FLAG-tagged MOR1K overexpressed in Be2C cells and stained with either Anti-FLAG M2 Antibody Alexa Fluor Conjugate or fluorescent-labeled naloxone (FNAL). Cells transfected with MOR1K showed intracellular retention of FNAL that co-localized with antibody-labeled receptor. (E) The binding of naloxone to MOR1K was assessed using flow cytometry to measure FNAL retention. Be2C cells transfected with either MOR1 or MOR1K isoforms showed increased retention of FNAL at concentrations of 0.1 and 1 μM. FNAL retention was abolished in the presence of 10 μM unlabelled naloxone (Nal). In panel E, data are presented as mean + SEM. *P < 0.05 different from controls).

Figure 2

MOR1K activation stimulates cAMP and Ca²⁺. COS1 (A,B) or BE2C (C) cells were transiently transfected with MOR1K, MOR1, or empty vector control expressing constructs. (A) Forskolin (FSK, 10 μM) was used to increase cAMP levels prior to morphine treatment. Following morphine treatment, cells expressing MOR1 exhibited reduced cAMP levels, while those expressing MOR1K did not. In fact, cells expressing MOR1K exhibited a trend towards an increase in cAMP levels. (B) Following morphine treatment, COS1 cells expressing MOR1K exhibited substantial increases in Ca²⁺ levels, while those expressing MOR1 did not. (C) Stimulation of MOR1K with morphine produced a robust dose-dependent increase in Ca²⁺ levels in Be2C cells. This increase was significantly different from the moderate increase in morphine-evoked Ca²⁺ levels observed in Be2C cells transfected with MOR1 or empty vector, which were likely due to the high endogenous expression of MOR1K in Be2C cells. For all panels, both MOR1 and MOR1K, morphine-dependent effects were antagonized in the presence of naloxone (0.1 μM). Data are presented as mean ± S.E.M from at least 6 experiments. $P < 0.05 different from control and *P < 0.05 different from control and MOR1.

Figure 3

MOR1K activation stimulates NO production. Be2C cells were transiently transfected with MOR1K or MOR1 expressing constructs. (A) MOR1K produced robust concentration-dependent increases in the release of NO. (B) The time course associated with morphine-induced (1 μM) NO release was markedly different in response to stimulating MOR1K and MOR1. NO production was blocked by pretreatment with naloxone (0.1 μM) and was significantly different from the respective controls. Data are presented as mean + SEM. *P < 0.05 different from controls.

Figure 4

MOR1K couples to Gα_s. COS1 cells were transiently transfected with MYC-tagged MOR1 or FLAG-tagged MOR1K constructs. (A) Iimmunoprecipitation was performed with anti-FLAG (raised in rabbit) and anti-MYC (raised in rabbit) antibodies. Immunoblotting was conducted with specific anti-Gα antibodies as described. The MOR1K-FLAG construct was found to couple to Gα_s, while the MOR1-MYC construct coupled only to Gα_i/o. The exposure time for MOR1K-FLAG experiments were ~2 times longer to see band intensities comparable to MOR1-MYC. (B) Consistent with Gα_s binding to the MOR1K isoform, morphine-mediated increases in Ca²⁺ levels were not prevented by pre-treatment of MOR1K expressing cells with PTX (100 ng/ml for 15min).