Cross-talk between Human Spinal Cord μ-opioid Receptor 1Y Isoform and Gastrin-releasing Peptide Receptor Mediates Opioid-induced Scratching Behavior

Abstract

Background: Although spinal opioids are safe and effective, pruritus is common and distressing. The authors previously demonstrated in mouse spinal cord that interactions between μ-opioid receptor isoform 1D and gastrin releasing peptide receptor mediate morphine-induced scratch. The C-terminal of 1D inhibits morphine-induced scratch without affecting analgesia. The authors hypothesize that human spinal cord also contains itch-specific μ-opioid receptor isoforms which interact with gastrin releasing peptide receptor.

Methods: Reverse transcription polymerase chain reaction was performed on human spinal cord complimentary DNA from two human cadavers. Calcium responses to morphine (1 μM) were examined using calcium imaging microscopy on human cells (HEK293) coexpressing gastrin releasing peptide receptor and different human μ-opioid receptor isoforms. The authors assessed morphine-induced scratching behavior and thermal analgesia in mice following intrathecal injection of morphine (0.3 nmol) and a transactivator of transcription peptide designed from C-terminal sequences of 1Y isoform (0, 0.1, and 0.4 nmol).

Results: The authors demonstrated 1Y expression in the spinal cord dorsal horn. Morphine administration evoked a calcium response (mean ± SD) (57 ± 13 nM) in cells coexpressing both gastrin releasing peptide receptor and the 1Y isomer. This was blocked by 10 μM naltrexone (0.7 ± 0.4 nM; P < 0.0001), 1 μM gastrin-releasing peptide receptor antagonist (3 ± 2 nM; P < 0.0001), or 200 μM 1Y-peptide (2 + 2 nM; P < 0.0001). In mice, 0.4 nmol 1Y-peptide significantly attenuated morphine-induced scratching behaviors (scratching bouts, vehicle vs. 1Y-peptide) (92 ± 31 vs. 38 ± 29; P = 0.011; n = 6 to 7 mice per group), without affecting morphine antinociception in warm water tail immersion test (% of maximum possible effect) (70 ± 21 vs. 67 ± 22; P = 0.80; n = 6 mice per group).

Conclusions: Human μ-opioid receptor 1Y isomer is a C-terminal splicing variant of Oprm1 gene identified in human spinal cord. Cross-talk between 1Y and gastrin releasing peptide receptor is required for mediating opioid-induced pruritus. Disrupting the cross talk may have implications for therapeutic uncoupling of desired analgesic effects from side effects of opioids.

Fig. 1.

Detection of hMOR1Y from human spinal cord. (A) Gel image of RT-PCR using C-terminal splicing variants specific primers showed that bands corresponding to hMOR1Y were detected in the dorsal horn of the spinal cord by primer pair 4 (687 bp) and primer pair 10 (361 bp), respectively. RT-PCR bands corresponding to hMOR1, hMOR1A, hMOR1B5, and hMOR1O were also detected by primer pair 1 (325 bp), primer pair 2 (306 bp), primer pair 6 (459 bp), primer pair 7 (389 bp), and primer pair 8 (298 bp), respectively. (B) In the ventral horn, hMOR1 and hMOR1A transcripts were detectable by RT-PCR. A faint band of hMOR1Y was also detected by primer pair 10. (C) hGrpr transcript was detected in the dorsal horn (lane 1), but not in the ventral horn (lane 2) by RT-PCR. (D) Alignment of amino acid sequences of hMOR1Y and mMOR1D shows that MOR is highly conserved with 93.8% identity and 97.2% similar in 388 aa amino acids overlap (1 to 388:1 to 386). (E) Alignment of amino acid sequences of human GRPR (M1-V384) and mouse GRPR (M1-V384) shows that GRPR is highly conserved with 89.9% identity. Rectangles in parts D and E highlight differences between two sequences. DL, DNA ladder; GRPR, human gastrin-releasing peptide receptor human; hMOR, human μ-opioid receptor; MOR, μ-opioid receptor; mMOR, mouse μ-opioid receptor; RT-PCR, reverse transcription polymerase chain reaction

Fig. 2.

Morphine cross activates hGRPR/Ca²⁺ signaling pathway when hGRPR is co-expressed with hMOR1Y. (A) Both morphine (1 μM) and GRP (1 nM) induced calcium responses in hGRPR-expressing HEK 293 cells transfected with and hMOR1Y. (B) Morphine (1 μM) did not induce calcium response in hGRPR-expressing HEK293 cells transfected with hMOR1. (C) Morphine (1 μM) did not induce calcium response in HEK 293 cells singly expressing hMOR1Y (purple) or hGRPR (grey). Data are presented as mean ± SD. Each experiment was done three times and at least 30 cells were analyzed each time. GRP, gastrin-releasing peptide; hGRPR, human gastrin-releasing peptide receptor; hMOR, human μ-opioid receptor.

Fig. 3.

Morphine induced hGRPR calcium responses through hMOR1Y. (A) Naltrexone (10 μM, red) blocked morphine-induced calcium responses in HEK 293 cells co-expressing hGRPR and hMOR1Y. The GRPR antagonist (1 μM) (blue trace) completely blocked morphine and GRP-induced Ca²⁺ increase. (B) Quantified data of (A) show that naltrexone (red) significantly inhibited the peak concentrations of intracellular calcium ([Ca²⁺]_i) induced by morphine. ***P < 0.001 versus vehicle, F (2, 87) = 604.3 for morphine, F (2, 87) = 717.6 for GRP, one-way ANOVA followed by Dunnett post hoc test. (C and D) Incubation of Tat-1Y with hMOR1Y/hGRPR coexpressing cells significantly blocked morphine-induced calcium spikes. High concentration of Tat-1Y (200 μM, red), but not low concentration of Tat-1Y (50 μM, green), significantly inhibited GRP-induced calcium responses. Data are presented as mean ± SD. *P < 0.05, ***P < 0.001 versus vehicle, F (2, 87) = 426.9 for morphine, F (2, 87) = 30.8, one-way ANOVA followed by Dunnett post hoc test. Each experiment was done three times and at least 30 cells were analyzed each time. GRP, gastrin-releasing peptide; hGRPR, human gastrin-releasing peptide receptor; hMOR, human μ-opioid receptor; ns, not significant; Tat, transactivator of transcription.

Fig. 4.

C-terminal end of hMOR1Y is responsible for the interactions between hGRPR and hMOR1Y. (A) Preinjection of Tat-1Y (0.4 nmol, i.t.) for 1 h significantly reduced scratching behaviors induced by i.t. injection of morphine (0.3 nmol). n = 6 to 7 mice per group. *P < 0.05 versus Tat-1Y 0 nmol, F (2, 16) = 5.04, one-way ANOVA followed by Dunnett’s post hoc test. (B) Morphine-induced analgesia was not affected by Tat-1Y (0.4 nmol, i.t.) as tested by warm water tail-immersion assay. n = 6 mice per group. P = 0.796, t = 0.27, degrees of freedom = 10, two-tailed unpaired t test. (C) Tat-1Y dose-dependently inhibited scratching behaviors induced by i.t. injection of GRP (0.05 nmol). n = 7 mice per group. *P < 0.05 versus Tat-1Y 0 nmol, F (2, 18) = 4.78, one-way ANOVA followed by Dunnett post hoc test. (D and E) Incubation of Tat-1Y (200 μM, red) for 2 h significantly inhibited morphine and GRP-induced calcium responses in mMOR1D/mGRPR coexpressing cells. *** P < 0.001 versus vehicle, two-tailed unpaired t test, t = 18.19, degrees of freedom = 58 for morphine, t = 23.68, degrees of freedom = 58 for GRP. (F) A diagram shows that morphine activates hMOR1Y, which cross-activates hGRPR-mediated the PLC-Ca²⁺ signaling pathway for itch that can be blocked by either naltrexone or TAT-hMOR1Y. Data are presented as mean ± SD. GRP, gastrin-releasing peptide; hGRPR, human gastrin-releasing peptide receptor; hMOR, human μ-opioid receptor; i.t., intrathecal; mMOR, mouse μ-opioid receptor; Tat, transactivator of transcription.