Contribution of G-Protein α-Subunits to Analgesia, Hyperalgesia, and Hyperalgesic Priming Induced by Subanalgesic and Analgesic Doses of Fentanyl and Morphine

Abstract

While opioids produce both analgesia and side effects by action at μ-opioid receptors (MORs), at spinal and supraspinal sites, the potency of different opioids to produce these effects varies. While it has been suggested that these differences might be because of bias for signaling via β-arrestin versus G-protein α-subunits (Gα), recent studies suggest that G-protein-biased MOR agonists still produce clinically important side effects. Since bias also exists in the role of Gα subunits, we evaluated the role of Gα_i/o subunits in analgesia, hyperalgesia, and hyperalgesic priming produced by fentanyl and morphine, in male rats. We found that intrathecal treatment with oligodeoxynucleotides antisense (AS-ODN) for Gα_i2, Gα_i3, and Gα_o markedly attenuated hyperalgesia induced by subanalgesic dose (sub-AD) fentanyl, while AS-ODN for Gα_i1, as well as Gα_i2 and Gα_i3, but not Gα_o, prevented hyperalgesia induced by sub-AD morphine. AS-ODN for Gα_i1 and Gα_i2 unexpectedly enhanced analgesia induced by analgesic dose (AD) fentanyl, while Gα_i1 AS-ODN markedly reduced AD morphine analgesia. Hyperalgesic priming, assessed by prolongation of prostaglandin E₂-induced hyperalgesia, was not produced by systemic sub-AD and AD fentanyl in Gα_i3 and Gα_o AS-ODN-treated rats, respectively. In contrast, none of the Gα_i/o AS-ODNs tested affected priming induced by systemic sub-AD and AD morphine. We conclude that signaling by different Gα_i/o subunits is necessary for the analgesia and side effects of two of the most clinically used opioid analgesics. The design of opioid analgesics that demonstrate selectivity for individual Gα_i/o may produce a more limited range of side effects and enhanced analgesia.

SIGNIFICANCE STATEMENT Biased μ-opioid receptor (MOR) agonists that preferentially signal through G-protein α-subunits over β-arrestins have been developed as an approach to mitigate opioid side effects. However, we recently demonstrated that biased MOR agonists also produce hyperalgesia and priming. We show that oligodeoxynucleotide antisense to different Gα_i/o subunits play a role in hyperalgesia and analgesia induced by subanalgesic and analgesic dose (respectively), of fentanyl and morphine, as well as in priming. Our findings have the potential to advance our understanding of the mechanisms involved in adverse effects of opioid analgesics that could assist in the development of novel analgesics, preferentially targeting specific G-protein α-subunits.

Introduction

Approximately 30% of medicines in use today are G-protein-coupled receptor (GPCR) ligands, as are 20% of drugs in clinical trials (Hauser et al., 2017). GPCRs mediate their effects through the activation of guanine nucleotide binding proteins (G-proteins; Childers, 1988; Reisine and Bell, 1993; Uhl et al., 1994; Pasternak and Standifer, 1995; Standifer et al., 1996) and β-arrestins (Gainetdinov et al., 2004; Williams et al., 2013). G-proteins are composed of three distinct subunit families (α, β, and γ), which couple GPCRs to downstream second messengers and effectors. α-Subunits are pharmacologically relevant because of their intrinsic GTPase activity (Standifer and Pasternak, 1997). Different α-subunits (e.g., Gα_i, Gα_o, Gα_s, Gα_q, Gx/α_z) establish the differential modulation of cAMP by GPCRs (Standifer and Pasternak, 1997). Activated Gα proteins have a variety of effects, including activation of mitogen-activated protein kinase (MAPK) and phosphatidylinositol 3-kinase (Leurs et al., 2005). Activation of the enzyme phospholipase (PL) A2 may also occur, which induces the release of arachidonic acid, as well as the inhibition of the Na⁺/H⁺ exchanger in the plasma membrane, and the lowering of intracellular Ca²⁺ (Leurs et al., 2005). How these various downstream signaling pathways are differentially activated by GPCRs remains to be elucidated.

Oligodeoxynucleotides (ODNs) antisense (AS) for α-subunit mRNA have been used to implicate different α-subunits in various opioid receptor-mediated functions (Pasternak and Standifer, 1995; Rossi et al., 1995; Standifer et al., 1996; Hadjimarkou et al., 2002; Silva et al., 2002; Wainford and Kapusta, 2012). For example, while intracerebroventricular administration of AS-ODN directed against either Gα_i2 or Gα_o reduces morphine, but not morphine-6β-glucuronide (M6G) analgesia, AS-ODN directed against Gα_i1 or Gx/α_z reduces M6G, but not morphine analgesia (Rossi et al., 1995; Standifer et al., 1996). In contrast, fentanyl displays high potency for activation of Gα_i1 and Gα_o (Saidak et al., 2006).

While opioids produce both analgesia and side effects by action at the μ-opioid receptor (MOR), their relative potencies that produce these diverse effects vary between opioids (Araldi et al., 2018a, c, 2019; Ferrari et al., 2019; Khomula et al., 2021). It has been suggested that this difference between opioid analgesics might be because of biased signaling by GPCR via β-arrestin versus Gα_i (Raehal and Bohn, 2014). However, several recent studies of MOR agonists biased toward G-proteins indicate that they still produce substantial side effects, such as constipation, tolerance, and respiratory depression (Viscusi et al., 2016; Olson et al., 2017; Hill et al., 2018; Gillis et al., 2020). We recently found that two such biased MOR agonists (i.e., TRV130 and PZM21) still produce hyperalgesia [opioid-induced hyperalgesia (OIH)] and hyperalgesic priming (opioid-induced hyperalgesic priming), as well as analgesia (Araldi et al., 2018c). In this study, we tested the hypothesis that spinal opioid-induced analgesia, OIH, and hyperalgesic priming produced by two clinically important opioid analgesics, fentanyl and morphine, are mediated by different Gα_i/α_o proteins.

Materials and Methods

Animals.

Experiments were performed on 260–380 g adult male Sprague Dawley rats (Charles River Laboratories). Given the large number of experiments required to establish the role of multiple Gαs in the analgesic and side effects of different doses of multiple opioid analgesics, we elected to perform experiments in female rats in a subsequent study. Animals were housed three per cage, under a 12 h light/dark cycle, in a temperature- and humidity-controlled animal care facility at the University of California, San Francisco. Food and water were available ad libitum. Experimental protocols were approved by the Institutional Animal Care and Use Committee (IACUC) at the University of California, San Francisco, and adhered to the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Effort was made to minimize the number of animals used and their suffering.

Nociceptive threshold testing.

Mechanical nociceptive threshold was quantified using an analgesy-meter (Randall Selitto - paw-withdrawal device, Ugo-Basile), which applies a linearly increasing mechanical force to the dorsum of the hindpaw of a rat, as previously described (Taiwo et al., 1989; Taiwo and Levine, 1989; Araldi et al., 2015, 2017, 2018a,c, 2019). Rats were placed in cylindrical acrylic restrainers designed to provide ventilation, allow hindleg extension from lateral ports in the cylinder during the assessment of nociceptive threshold, and minimize restraint stress. To acclimatize rats to the testing procedure, they were placed in restrainers for 1 h before starting each training session, for 3 consecutive daily training sessions, and for 40 min before subsequent experimental manipulations. Nociceptive threshold was defined as the force in grams at which a rat withdrew its paw. Baseline paw-pressure nociceptive threshold was defined as the mean of the three readings taken before test agents were injected. To minimize experimenter bias, individuals conducting the behavioral experiments (D.A. and I.J.M.B.) were blinded to experimental interventions.

Drugs.

The following compounds were used in this study: fentanyl citrate salt (an MOR agonist), prostaglandin-E₂ (PGE₂; a direct-acting hyperalgesic agent that sensitizes nociceptors), and morphine sulfate salt pentahydrate (an agonist at μ, δ, and κ opioid receptors), all purchased from Sigma-Aldrich. The stock solution of PGE₂ (1 µg/μl) was prepared in 100% ethanol, and further dilutions were made with physiological saline (0.9% NaCl), yielding a final ethanol concentration of <2%. Fentanyl and morphine were dissolved in saline.

Intradermal administration of PGE₂ was performed on the dorsum of the hindpaw, at the site of nociceptive testing, using a 30 gauge hypodermic needle adapted to a 50 μl syringe (Hamilton) by a segment of PE-10 polyethylene tubing (Becton Dickinson). Our in vivo control experiments have previously shown that the final concentration of ethanol (2%), used to prepare the PGE₂ solution, alone had no effect on the mechanical threshold (Ferrari et al., 2016).

Systemic administration of fentanyl (sub-AD, 0.01 mg/kg, s.c.; AD, 0.03 mg/kg, s.c.) and morphine (sub-AD, 0.03 mg/kg, s.c.; AD, 3 mg/kg, s.c.) was performed at the nape of the neck (Araldi et al., 2018a, 2019). Rats received an injection of fentanyl or morphine, and mechanical nociceptive threshold was evaluated 1 h later (Araldi et al., 2018a, 2019). Fentanyl and morphine were diluted in saline and administered (100 μl/100 g body weight, s.c.).

Oligodeoxynucleotides antisense to G-protein α-subunit mRNAs.

To investigate the role of G-protein α-subunits in the analgesia, hyperalgesia, and priming, induced by systemic sub-AD and AD fentanyl and morphine, validated AS-ODNs for Gα_i1, Gα_i2, Gα_i3, and Gα_o mRNA (Rossi et al., 1995; Standifer et al., 1996; Hadjimarkou et al., 2002; Silva et al., 2002; Wainford and Kapusta, 2012) were used.

AS-ODN sequences, directed against unique regions of the rat mRNA for G-protein α-subunits, were as follows:

Gα_i1 AS-ODN: 5′-AGACCACTGCTTTGTA-3′ (Gnai1; GenBank accession no. NM_013145.1)

Gα_i2 AS-ODN: 5′-CTTGTCGATCATCTTAGA-3′ (Gnai2; GenBank accession no. NM_031035.3)

Gα_i3 AS-ODN: 5′-AAGTTGCGGTCGATCAT-3′ (Gnai3; GenBank accession no. NM_013106.1)

Gα_o AS-ODN: 5′-CGCCTTGCTCCGCTC-3′ (Gnao1; GenBank accession no. NM_017327.1)

As a control, the following sense (SE) ODN sequences were used:

Gα_i1 SE-ODN: 5′-TACAAAGCAGTGGTCT-3′

Gα_i2 SE-ODN: 5′-TCTAAGATGATCGACAAG-3′

Gα_i3 SE-ODN: 5′-ATGATCGACCGCAACTT-3′

Gα_o SE-ODN: 5′-GAGCGGAGCAAGGCG-3′

ODNs were synthesized by Thermo Fisher Scientific. A nucleotide BLAST (Basic Local Alignment Search Tool) search was performed to confirm that the mRNA sequences targeted by the AS-ODN were not homologous to any other sequences in the rat database and that SE-ODN control sequences were not homologous to any sequences in the rat database as well. Before use, lyophilized ODNs were reconstituted in nuclease-free 0.9% NaCl and then administered intrathecally at a dose of 6 µg/µl in a volume of 20 µl (120 µg/20 µl). AS-ODNs or SE-ODNs were injected daily for 4 consecutive days. On the fourth day, fentanyl (AD, 0.03 mg/kg; sub-AD, 0.01 mg/kg) or morphine (AD, 3 mg/kg; or sub-AD, 0.03 mg/kg) was then injected subcutaneously on the back of the neck. Mechanical nociceptive threshold was evaluated 1 h after systemic fentanyl and morphine; 5 d later, PGE₂ (100 ng) was injected intradermally and mechanical nociceptive threshold evaluated 30 min and 4 h later. As described previously (Alessandri-Haber et al., 2003), after rats were anesthetized with isoflurane (2.5% in O₂), ODN was injected using a 0.3 ml syringe (300 U/µl; Walgreens) with a 29 gauge hypodermic needle, inserted into the subarachnoid space between the L4 and L5 vertebrae. The intrathecal site of injection was confirmed by a sudden flick of the tail of the rat, a reflex that is evoked by accessing the subarachnoid space and bolus injection (Mestre et al., 1994). A total of 120 μg of ODN, in a volume of 20 μl, was then injected. Rats regained consciousness ∼2 min after anesthesia was stopped. Use of intrathecal AS-ODN to attenuate the expression of proteins, essential for their role in nociceptor sensitization, is well supported by previous studies (Song et al., 2009; Su et al., 2011; Bogen et al., 2012; Quanhong et al., 2012; Sun et al., 2013; Oliveira-Fusaro et al., 2017; Araldi et al., 2019; Ferrari et al., 2019; Pagliusi et al., 2020).

Data analysis.

All data are presented as the mean ± SEM of n independent observations. Statistical comparisons were made using GraphPad Prism 8.0 statistical software (GraphPad Software). A p-value < 0.05 was considered statistically significant. In the behavioral experiments, the dependent variable was change in the mechanical paw-withdrawal threshold, expressed as the percentage change from baseline. No significant difference in mechanical nociceptive thresholds was observed between G-protein α-subunit AS-ODN- and SE-ODN-treated groups of rats measured before the administration of fentanyl or morphine (as demonstrated in the figure legends) and immediately before the injection of PGE₂ (average mechanical nociceptive threshold before fentanyl- and morphine-induced priming, 141.5 ± 0.837 g; average mechanical nociceptive threshold before PGE₂ injection, 140.8 ± 0.839 g; n = 192 rats; paired Student's t test: t₍₁₉₁₎ = 1.87, p = 0.062. As specified in figure legends, Student's t test or two-way repeated-measures ANOVA followed by Bonferroni's post hoc test was performed to compare the magnitude of the analgesia and hyperalgesia induced by systemic fentanyl or morphine, or to compare the effect produced by the G-protein α-subunit AS-ODNs on the prolongation of hyperalgesia (evaluated 4 h after injection of PGE₂) compared with the SE-ODN control groups.

Results

OIH and priming produced by subanalgesic dose fentanyl

As most opioid analgesics produce hyperalgesia at sub-ADs, we investigated whether OIH and hyperalgesic priming produced by systemic sub-AD fentanyl (0.01 mg/kg, s.c.) are dependent on Gα_i/o (i.e., Gα_i1, Gα_i2, Gα_i3, and/or Gα_o). Rats were treated intrathecally with AS-ODN or SE-ODN to Gα_i1 (Fig. 1), Gα_i2 (Fig. 2), Gα_i3 (Fig. 3), or Gα_o (Fig. 4) mRNA, daily for 4 d. On the fourth day, ∼17 h after the third injection of ODN, sub-AD fentanyl (0.01 mg/kg) was administered systemically (subcutaneous, s.c.), and mechanical nociceptive threshold measured 1 h later. When compared with SE-ODN-treated rats the hyperalgesia induced by sub-AD fentanyl was markedly attenuated in rats pretreated with AS-ODN against Gα_i2 (Fig. 2A), Gα_i3 (Fig. 3A), or Gα_o (Fig. 4A). Treatment with Gα_i1 AS-ODN did not affect sub-AD fentanyl-induced hyperalgesia (Fig. 1A). At the end of the fourth day, rats received the last intrathecal administration of G-protein α-subunit AS-ODN or SE-ODN. Five days after systemic sub-AD fentanyl administration, PGE₂ (100 ng) was injected intradermally and the mechanical nociceptive threshold evaluated 30 min and 4 h later. In the rats treated with Gα_i3 AS-ODN, the prolongation PGE₂ hyperalgesia at the fourth hour was not present (Fig. 3C). However, prolonged PGE₂-induced hyperalgesia was still present in Gα_o, Gα_i1, and Gα_i2 AS-ODN- and SE-ODN-treated rats (Figs. 1C, 2C, and 4C). Our data support the following suggestions: (1) that some side effects of fentanyl (OIH and hyperalgesic priming) are Gα_i/o dependent; (2) that a side effect (OIH) may be dependent on multiple Gα_i/o subunits; and (3) that the Gα_i/o subunits mediating one side effect (Gα_i2, Gα_i3, and Gα_o mediating OIH) may differ from those mediating another side effect (only Gα_i3 mediating sub-AD fentanyl-induced priming).

Figure 1.

Role of G_i-protein α1 subunit (Gα_i1) in OIH and hyperalgesic priming produced by systemic sub-AD fentanyl and morphine. Rats received intrathecal (i.t.) injection of AS-ODN (120 μg/20 μl/d, i.t.) or SE-ODN (120 μg/20 μl/d, i.t.) to Gα_i1 mRNA for 3 consecutive days. A, B, On the fourth day, at which time mechanical nociceptive threshold was not different from the pre-ODN baseline (A: SE-ODN-treated group: t₍₅₎ = 0.5; p = 0.64; AS-ODN-treated group: t₍₅₎ = 0.82; p = 0.45; B: SE-ODN-treated group: t₍₅₎ = 2.17; p = 0.08; AS-ODN-treated group: t₍₅₎ = 0.62; p = 0.56 when the mechanical nociceptive threshold is compared before and ∼17 h after the third ODN injection; paired Student's t test), sub-AD fentanyl (A: 0.01 mg/kg, s.c.) or morphine (B: 0.03 mg/kg, s.c.) was injected and the mechanical nociceptive threshold was evaluated 1 h later. Gα_i1 AS-ODN did not affect sub-AD fentanyl-induced hyperalgesia (F_(1,10) = 1.94, p = 0.19, when hyperalgesia was compared between the Gα_i1 SE-ODN- and AS-ODN-treated groups 1 h after systemic sub-AD fentanyl; two-way repeated-measures ANOVA followed by Bonferroni post hoc test; A). However, in the Gα_i1 AS-ODN-treated group, OIH produced by sub-AD morphine was markedly attenuated (F_(1,10) = 189.4, ****p < 0.0001, when the hyperalgesia in the Gα_i1 SE-ODN- and the AS-ODN-treated groups is compared at 1 h after systemic sub-AD morphine; two-way repeated-measures ANOVA followed by Bonferroni post hoc test; B). At the end of the fourth day, rats again received intrathecal Gα_i1 AS- or SE-ODN. C, D, Five days after systemic sub-AD fentanyl and morphine, at which time mechanical nociceptive threshold was not different from preopioid baselines (C: SE-ODN-treated group: t₍₅₎ = 1.58; p = 0.18; AS-ODN-treated group that received sub-AD fentanyl: t₍₅₎ = 0.39; p = 0.71; D: SE-ODN-treated group: t₍₅₎ = 0.88; p = 0.42; AS-ODN-treated group that received sub-AD morphine: t₍₅₎ = 0.2; p = 0.85 when the mechanical nociceptive threshold is compared before and 5 d after systemic sub-AD opioid administration; paired Student's t test), PGE₂ (100 ng/5 μl, i.d.) was injected and the mechanical nociceptive threshold was evaluated 30 min and 4 h later. Treatment with Gα_i1 AS-ODN did not prevent PGE₂-induced prolonged hyperalgesia in either fentanyl-treated (C) or morphine-treated (D) groups (C: F_(1,10) = 2.12, p = 0.17; D: F_(1,10) = 1.02, p = 0.34, when the hyperalgesia in the Gα_i1 SE-ODN- and the AS-ODN-treated groups is compared at the fourth hour after intradermal PGE₂ administration; two-way repeated-measures ANOVA followed by Bonferroni post hoc test). These findings support the suggestion that Gα_i1 plays a role in OIH produced by systemic sub-AD morphine, but not fentanyl, and is not involved in hyperalgesic priming produced by sub-AD fentanyl or morphine (n = 6 paws/6 rats/group).

Figure 2.

Role of Gα_i2 in hyperalgesia and hyperalgesic priming produced by systemic sub-AD fentanyl and morphine. Rats received injections of AS-ODN (120 μg/20 μl/d, i.t.) or SE-ODN (120 μg/20 μl/d, i.t.) against Gα_i2 mRNA, for 3 consecutive days. A, B, Approximately 17 h after the third injection, at which time the mechanical nociceptive threshold was not different from pre-ODN baseline levels (A: SE-ODN-treated group: t₍₅₎ = 0.41, p = 0.70; AS-ODN-treated group: t₍₅₎ = 0.56; p = 0.57; B: SE-ODN-treated group: t₍₅₎ = 0.15, p = 0.89; AS-ODN-treated group: t₍₅₎ = 0.13, p = 0.90, when the mechanical nociceptive threshold is compared before and after the third ODN injection; paired Student's t test), sub-AD fentanyl (A: 0.01 mg/kg, s.c.) or morphine (B: 0.03 mg/kg, s.c.) was administered and mechanical nociceptive threshold was evaluated 1 h later. Treatment with Gα_i2 AS-ODN prevented hyperalgesia induced by both sub-AD fentanyl (A) and morphine (B), when it was compared with the SE-ODN-treated group (A: F_(1,10) = 75.5, ****p < 0.0001; B: F_(1,10) = 298.3, ****p < 0.0001, when the hyperalgesia in the Gα_i2 SE-ODN-treated and the AS-ODN-treated groups is compared at 1 h after systemic sub-AD fentanyl and morphine; two-way repeated-measures ANOVA followed by Bonferroni's post hoc test). Rats again received intrathecal Gα_i2 AS- or SE-ODN on the fourth day. C, D, Five days after sub-AD opioids, at which time the mechanical nociceptive threshold was not different from preopioid baseline (C: SE-ODN-treated group: t₍₅₎ = 1.0, p = 0.36; AS-ODN-treated group that received sub-AD fentanyl: t₍₅₎ = 0.54, p = 0.61; D: SE-ODN-treated group: t₍₅₎ = 0.73, p = 0.49; AS-ODN-treated group that received sub-AD morphine: t₍₅₎ = 0.59, p = 0.57, when the mechanical nociceptive threshold is compared before and 5 d after systemic sub-AD opioids; paired Student's t test), PGE₂ (100 ng/5 μl, i.d.) was injected and the mechanical nociceptive threshold was evaluated 30 min and 4 h later. In both the Gα_i2 AS-ODN-treated and SE-ODN-treated groups, the prolongation of PGE₂ hyperalgesia was present 5 d after systemic sub-AD fentanyl and morphine (C and D, respectively; C: F_(1,10) = 0.11, p = 0.74; D: F_(1,10) = 0.50, p = 0.49, when the hyperalgesia in the Gα_i2 AS-ODN-treated and the SE-ODN-treated groups is compared at the fourth hour after intradermal PGE₂; two-way repeated-measures ANOVA followed by Bonferroni's post hoc test). Our data indicate that OIH, but not priming, produced by systemic sub-AD fentanyl and morphine is Gα_i2 dependent (n = 6 paws/6 rats/group).

Figure 3.

Role of Gα_i3 in hyperalgesia and hyperalgesic priming produced by systemic sub-AD fentanyl and morphine. Rats received injection of AS-ODN (120 μg in 20 μl/d, i.t.) or SE-ODN (120 μg in 20 μl/d, i.t.) against Gα_i3 mRNA, daily for 3 consecutive days. A, B, On the fourth day, ∼17 h after the third intrathecal administration of ODNs, at which time mechanical nociceptive threshold was not significantly different from pre-ODN baselines (A: SE-ODN-treated group: t₍₅₎ = 0.67, p = 0.53; AS-ODN-treated group: t₍₅₎ = 0.18, p = 0.86; B: SE-ODN-treated group: t₍₅₎ = 0.39, p = 0.71; AS-ODN-treated group: t₍₅₎ = 0.15, p = 0.89, when the mechanical nociceptive threshold is compared before and after the third Gα_i3 ODN injection; paired Student's t test), sub-AD fentanyl (A: 0.01 mg/kg, s.c.) or morphine (B: 0.03 mg/kg, s.c.) was administered and the mechanical nociceptive threshold was evaluated 1 h later. In the Gα_i3 AS-ODN-treated group, systemic sub-AD of neither fentanyl (A) nor morphine (B) produced hyperalgesia, measured 1 h after its administration, as is observed in the Gα_i3 SE-ODN-treated group (A: F_(1,10) = 186.6, ****p < 0.0001; B: F_(1,10) = 193.9, ****p < 0.0001, when the hyperalgesia in the Gα_i3 SE-ODN-treated and the AS-ODN-treated groups was compared at 1 h after systemic sub-AD opioids; two-way repeated-measures ANOVA followed by Bonferroni post hoc test). At the end of the fourth day, rats again received intrathecal Gα_i3 AS-ODN or SE-ODN. C, D, Five days after systemic sub-AD fentanyl and morphine, at which time mechanical nociceptive threshold was not different from preopioid baselines (C: SE-ODN-treated group: t₍₅₎ = 1.98, p = 0.11; AS-ODN-treated group that received sub-AD fentanyl: t₍₅₎ = 1.07, p = 0.33; D: SE-ODN-treated group: t₍₅₎ = 0.22; p = 0.83; AS-ODN-treated group that received sub-AD morphine: t₍₅₎ = 2.23, p = 0.07, when the mechanical nociceptive threshold is compared before and 5 d after systemic sub-AD opioid; paired Student's t test), PGE₂ (100 ng/5 μl, i.d.) was injected and mechanical nociceptive threshold was evaluated 30 min and 4 h later. In the Gα_i3 AS-ODN-treated group, which received systemic sub-AD fentanyl, PGE₂-induced hyperalgesia was not present at the fourth hour (C: F_(1,10) = 42.9, ****p < 0.0001, when the hyperalgesia in the Gα_i3 AS-ODN-treated and the SE-ODN-treated groups is compared at the fourth hour after intradermal PGE₂; two-way repeated-measures ANOVA followed by Bonferroni's post hoc test). However, prolongation of PGE₂-induced hyperalgesia was not affected by the treatment with Gα_i3 AS-ODN in the systemic sub-AD morphine-treated group (D: F_(1,10) = 0.04, p = 0.85, when hyperalgesia was compared between the Gα_i3 SE-ODN-treated and AS-ODN-treated groups at the fourth hour after intradermal PGE₂; two-way repeated-measures ANOVA followed by Bonferroni's post hoc test). These findings indicate that Gα_i3 plays a role in hyperalgesia induced by both sub-AD fentanyl and morphine. However, priming induced by sub-AD fentanyl, but not sub-AD morphine, is dependent on Gα_i3 (n = 6 paws/6 rats/group).

Figure 4.

Role of Gα_o in hyperalgesia and hyperalgesic priming induced by systemic sub-AD fentanyl and morphine. Rats received injection of AS-ODN (120 μg in 20 μl/d, i.t.) or SE-ODN (120 μg in 20 μl/d, i.t.) against Gα_o mRNA daily for 3 consecutive days. On the fourth day, at which time the mechanical nociceptive threshold was not different from the pre-ODN baselines (A: SE-ODN-treated group: t₍₅₎ = 0.67; p = 0.53; AS-ODN-treated group: t₍₅₎ = 0.65; p = 0.54; B: SE-ODN-treated group: t₍₅₎ = 0.74; p = 0.49; AS-ODN-treated group: t₍₅₎ = 1.66; p = 0.15, when the mechanical nociceptive threshold is compared before and ∼17 h after the third ODN injection; paired Student's t test), sub-AD fentanyl (A, 0.01 mg/kg, s.c.) or morphine (B, 0.03 mg/kg, s.c.) was administered and the mechanical nociceptive threshold was evaluated 1 h later. In the group of rats treated with Gα_o AS-ODN, systemic sub-AD fentanyl-induced hyperalgesia was prevented (A, F_(1,10) = 53.3, ****p < 0.0001, when the hyperalgesia in the Gα_o SE-ODN-treated and the AS-ODN-treated groups was compared at 1 h after systemic sub-AD fentanyl; two-way repeated-measures ANOVA followed by Bonferroni's post hoc test). However, systemic sub-AD morphine-induced hyperalgesia was not affected by the treatment with Gα_o AS-ODN (B, F_(1,10) = 0.24, p = 0.63, when the hyperalgesia in the Gα_o SE-ODN-treated and the AS-ODN-treated groups was compared at 1 h after systemic sub-AD morphine; two-way repeated-measures ANOVA followed by Bonferroni's post hoc test). At the end of the fourth day, rats again received intrathecal Gα_o AS-ODN or SE-ODN. Five days after systemic sub-AD fentanyl and morphine, when the mechanical nociceptive threshold was not different from preopioid baselines (C: SE-ODN-treated group: t₍₅₎ = 0.75; p = 0.48; AS-ODN-treated group that received sub-AD fentanyl: t₍₅₎ = 2.15; p = 0.08; D: SE-ODN-treated group: t₍₅₎ = 0.68; p = 0.53; AS-ODN-treated group that received sub-AD morphine: t₍₅₎ = 1.90; p = 0.11, when the mechanical nociceptive threshold is compared before and 5 d after systemic sub-AD opioids; paired Student's t test), PGE₂ (100 ng/5 μl, i.d.) was administered and the mechanical nociceptive threshold was evaluated 30 min and 4 h later. Treatment with Gα_o AS-ODN did not prevent the prolongation of PGE₂-induced hyperalgesia in both fentanyl-treated (C) and morphine-treated (D) groups of rats (C: F_(1,10) = 2.15, p = 0.17; D: F_(1,10) = 0.18, p = 0.68, when the hyperalgesia in the Gα_o SE-ODN-treated and the AS-ODN-treated groups is compared at the fourth hour after intradermal PGE₂; two-way repeated-measures ANOVA followed by Bonferroni's post hoc test). These findings support the suggestion that the Gα_o subunit plays a role in OIH produced by systemic sub-AD fentanyl, but not morphine, and is not involved in hyperalgesic priming produced by sub-AD fentanyl and morphine. (n = 6 paws/6 rats/group).

OIH and priming produced by subanalgesic dose morphine

We next investigated whether OIH and hyperalgesic priming induced by sub-AD of another clinical opioid analgesic, morphine (0.03 mg/kg, s.c.; Araldi et al., 2019) are also Gα_i/o dependent and whether they are dependent on the same Gα_i/o subunits as these effects of sub-AD fentanyl. Rats received intrathecal AS-ODN or SE-ODN to Gα_i1 (Fig. 1), Gα_i2 (Fig. 2), Gα_i3 (Fig. 3), or Gα_o (Fig. 4) mRNA, daily for 4 d. On the fourth day, sub-AD morphine (0.03 mg/kg, s.c.) was administered and the mechanical nociceptive threshold was measured 1 h later. In rats treated with AS-ODN against Gα_i1 (Fig. 1B), Gα_i2 (Fig. 2B), and Gα_i3 (Fig. 3B) mRNA, sub-AD morphine did not induce hyperalgesia. However, in rats treated with Gα_o AS-ODN, systemic sub-AD morphine still induced hyperalgesia (Fig. 4B). At the end of the fourth day, rats again received AS-ODN or SE-ODN. Five days after systemic sub-AD morphine administration, PGE₂ (100 ng) was injected intradermally and the mechanical nociceptive threshold was evaluated 30 min and 4 h later. The prolongation of PGE₂-induced hyperalgesia was still present in all AS-ODN-treated rats (Figs. 1D, 2D, 3D, and 4D). These findings support the following suggestions: (1) that the Gα_i/o subunits mediating side effects (OIH and hyperalgesic priming) differ between the sub-AD of clinical opioid analgesics (fentanyl and morphine); and (2) that the same side effect (OIH) produced by two opioid analgesics may be dependent only partially on overlapping Gα_i/o subunits (while OIH produced by sub-AD morphine is dependent on Gα_i1, Gα_i2, and Gα_i3, OIH produced by sub-AD fentanyl is Gα_i2, Gα_i3, and Gα_o dependent). Whether opioid-induced hyperalgesic priming produced by sub-AD morphine is G-protein independent or dependent on one of the Gα subunits not evaluated in the present experiments remains to be established.

Analgesia and hyperalgesic priming produced by analgesic dose (AD) fentanyl

Since AD opioids produce hyperalgesic priming as well as analgesia, we next determined whether analgesia and priming induced by AD fentanyl (0.03 mg/kg, s.c.; Khomula et al., 2019, 2021) are Gα_i/o dependent and whether the same G-proteins mediate opioid-induced hyperalgesic priming produced by sub-AD and AD fentanyl. Groups of rats were treated with AS-ODN or SE-ODN for Gα_i1 (Fig. 5), Gα_i2 (Fig. 6), Gα_i3 (Fig. 7), or Gα_o (Fig. 8) mRNA, daily for 4 consecutive days. On the fourth day, ∼17 h after the third dose of ODN, AD fentanyl was administered (0.03 mg/kg, s.c.), and the mechanical nociceptive threshold was evaluated 1 h later. Unexpectedly, analgesia induced by AD fentanyl was increased in the Gα_i1 (Fig. 5A) and Gα_i2 (Fig. 6A) AS-ODN-treated groups of rats; while in the groups treated with either Gα_i3 (Fig. 7A) or Gα_o AS-ODN (Fig. 8A) fentanyl-induced analgesia was not affected. At the end of the fourth day, rats again received AS-ODN or SE-ODN. Five days after AD fentanyl, PGE₂ (100 ng, i.d.) was injected and mechanical nociceptive threshold evaluated 30 min and 4 h later. In rats treated with Gα_o (Fig. 8C), but not Gα_i1, Gα_i2, and Gα_i3 (Figs. 5C, 6C, and 7C) AS-ODN the prolongation of PGE₂-induced hyperalgesia was eliminated.

Figure 5.

Role of Gα_i1 in systemic AD fentanyl- and morphine-induced analgesia and priming. Rats received intrathecal injections of AS-ODN (120 μg in 20 μl/d, i.t.) or SE-ODN (120 μg in 20 μl/d, i.t.) against Gα_i1 mRNA, daily for 3 consecutive days. On the fourth day, AD fentanyl (A, 0.03 mg/kg, s.c.) or morphine (B, 3 mg/kg, s.c.) was administered and the mechanical nociceptive threshold was evaluated 1 h later. Treatment with Gα_i1 AS-ODN increased analgesia induced by systemic AD fentanyl (A), while it decreased AD morphine-induced analgesia (B), compared with their respective Gα_i1 SE-ODN-treated groups (A: t₍₁₀₎ = 3.24, ** p = 0.0088; B: t₍₁₀₎ = 4.81, *** p = 0.0007, when the analgesia in the Gα_i1 AS-ODN-treated and the SE-ODN-treated groups is compared at 1 h after systemic AD fentanyl and morphine; unpaired Student's t test). At the end of the fourth day, rats again received intrathecal Gα_i1 AS-ODN or SE-ODN. Five days after systemic AD fentanyl and morphine administration, at which time the mechanical nociceptive threshold was not different from preopioid baselines (C: SE-ODN-treated group: t₍₅₎ = 0.39, p = 0.71; AS-ODN-treated group that received systemic AD fentanyl: t₍₅₎ = 1.4, p = 0.22; D: SE-ODN-treated group: t₍₅₎ = 1.0, p = 0.36; AS-ODN-treated group that received systemic AD morphine: t₍₅₎ = 0.58, p = 0.59, when the mechanical nociceptive threshold is compared before and 5 d after systemic AD opioids; paired Student's t test), PGE₂ (100 ng/5 μl, i.d.) was administered and the mechanical nociceptive threshold was evaluated 30 min and 4 h later. Prolongation of PGE₂-induced hyperalgesia was not affected by treatment with Gα_i1 AS-ODN in both AD fentanyl-treated (C) and morphine-treated (D) groups of rats (C: F_(1,10) = 1.28, p = 0.28; D: F_(1,10) = 0.13, p = 0.72, when the hyperalgesia in the Gα_i1 SE-ODN-treated and the AS-ODN-treated groups is compared at the fourth hour after intradermal PGE₂ administration; two-way repeated-measures ANOVA followed by Bonferroni's post hoc test). These findings support the suggestion that Gα_i1 plays a role in analgesia, but not in hyperalgesic priming, produced by AD fentanyl and morphine (n = 6 paws/6 rats/group).

Figure 6.

Role of Gα_i2 in analgesia and hyperalgesic priming induced by systemic analgesic dose fentanyl and morphine. Rats received injections of AS-ODN (120 μg/20 μl/d, i.t.) or SE-ODN (120 μg/20 μl/d, i.t.) against Gα_i2 mRNA, daily for 3 consecutive days. Approximately 17 h after the third injection of ODNs, AD fentanyl (A: 0.03 mg/kg, s.c.) or morphine (B: 3 mg/kg, s.c.) was administered and the mechanical nociceptive threshold was evaluated 1 h later. Treatment with Gα_i2 AS-ODN increased analgesia induced by AD fentanyl (A); however, it did not affect analgesia induced by AD morphine (B), when compared with the SE-ODN-treated group (A: t₍₁₀₎ = 5.07, ***p = 0.0005; B: t₍₁₀₎ = 0.96, p = 0.36, when the analgesia in the Gα_i2 AS-ODN-treated and the SE-ODN-treated groups is compared at 1 h after systemic AD fentanyl and morphine, respectively; unpaired Student's t test). Rats again received Gα_i2 AS-ODN or SE-ODN later on the fourth day. Five days after AD opioids, at which time mechanical nociceptive threshold was not different from preopioid baselines (C: SE-ODN-treated group: t₍₅₎ = 2.5, p = 0.06; AS-ODN-treated group that received AD fentanyl: t₍₅₎ = 0.67; p = 0.53; D: SE-ODN-treated group: t₍₅₎ = 1.0, p = 0.36; AS-ODN-treated group that received AD morphine: t₍₅₎ = 0.12, p = 0.91, when the mechanical nociceptive threshold is compared before and 5 d after systemic AD opioids; paired Student's t test), PGE₂ (100 ng/5 μl, i.d.) was administered and the mechanical nociceptive threshold was evaluated 30 min and 4 h later. In both the Gα_i2 AS-treated and SE-ODN-treated groups, the prolongation of PGE₂-induced hyperalgesia was present 5 d after systemic AD fentanyl and morphine (C and D, respectively; C: F_(1,10) = 0.68, p = 0.43; D: F_(1,10) = 2.84, p = 0.12, when the hyperalgesia in the Gα_i2 AS-ODN-treated and the SE-ODN-treated groups is compared at the fourth hour after intradermal PGE₂ administration; two-way repeated-measures ANOVA followed by Bonferroni's post hoc test). Our data support the suggestion that the Gα_i2 subunit plays a role in analgesia produced by AD fentanyl, but not morphine; also, hyperalgesic priming produced by AD fentanyl and morphine is not Gα_i2 dependent (n = 6 paws/6 rats/group).

Figure 7.

Role of Gα_i3 in analgesia and hyperalgesic priming induced by systemic AD fentanyl and morphine. Rats received intrathecal injections of AS-ODN (120 μg in 20 μl/d, i.t.) or SE-ODN (120 μg in 20 μl/day, i.t.) against Gα_i3 mRNA, daily for 3 consecutive days. On the fourth day, ∼17 h after the third injection of ODNs, systemic AD fentanyl (A: 0.03 mg/kg, s.c.) or morphine (B: 3 mg/kg, s.c.) was injected and mechanical nociceptive threshold evaluated 1 h later. Treatment with Gα_i3 AS-ODN did not affect the analgesia produced by either AD fentanyl (A) or morphine (B), measured 1 h after their administration, as observed in the Gα_i3 SE-ODN-treated group (A: t₍₁₀₎ = 0.054; ***p = 0.96; B: t₍₁₀₎ = 0.51, p = 0.62, when the analgesia in the Gα_i3 AS-ODN-treated and the SE-ODN-treated groups is compared at 1 h after systemic AD fentanyl and morphine, respectively; unpaired Student's t test). At the end of the fourth day, rats again received intrathecal Gα_i3 AS-ODN or SE-ODN. Five days after systemic AD fentanyl and morphine, at which time the mechanical nociceptive threshold was not different from preopioid baseline levels (C: SE-ODN-treated group: t₍₅₎ = 1.38, p = 0.23; AS-ODN-treated group that received systemic AD fentanyl: t₍₅₎ = 1.06, p = 0.34; D: SE-ODN-treated group: t₍₅₎ = 2.0, p = 0.1; AS-ODN-treated group that received systemic AD morphine: t₍₅₎ = 0.72, p = 0.5, when the mechanical nociceptive threshold is compared before and 5 d after systemic AD opioids; paired Student's t test), PGE₂ (100 ng/5 μl, i.d.) was injected and the mechanical nociceptive threshold was evaluated 30 min and 4 h later. In both the Gα_i3 AS-ODN-treated and SE-ODN-treated groups, the prolongation of PGE₂-induced hyperalgesia was present 5 d after systemic AD fentanyl and morphine (C and D, respectively; C: F_(1,10) = 0.58, p = 0.46; D: F_(1,10) = 0.1, p = 0.75, when the hyperalgesia in the Gα_i3 AS-ODN-treated and the SE-ODN-treated groups is compared at the fourth hour after intradermal PGE₂; two-way repeated-measures ANOVA followed by Bonferroni's post hoc test). These findings support the suggestion that Gα_i3 did not play a role in analgesia and hyperalgesic priming induced by AD fentanyl or morphine (n = 6 paws/6 rats/group).

Figure 8.

Role of Gα_o in analgesia and hyperalgesic priming induced by systemic AD fentanyl and morphine. Rats received intrathecal injections of AS-ODN (120 μg in 20 μl/d, i.t.) or SE-ODN (120 μg in 20 μl/d, i.t.) against Gα_o mRNA, daily for 3 consecutive days. On the fourth day, AD fentanyl (A: 0.03 mg/kg, s.c.) or morphine (B: 3 mg/kg, s.c.) was administered and the mechanical nociceptive threshold was evaluated 1 h later. In the groups of rats treated with Gα_o AS-ODN, systemic AD fentanyl (A), and morphine (B) induced analgesia that was not different when compared with their respective Gα_o SE-ODN-treated groups (A: t₍₁₀₎ = 0.13, p = 0.90; B: t₍₁₀₎ = 0.005, p = 0.99, when the analgesia in the Gα_o AS-ODN-treated and the SE-ODN-treated groups is compared at 1 h after systemic AD fentanyl and morphine, respectively; unpaired Student's t test). At the end of the fourth day, rats again received Gα_o AS-ODN or SE-ODN. Five days after systemic AD fentanyl and morphine, when the mechanical nociceptive threshold was not different from the preopioids baselines (C: SE-ODN-treated group: t₍₅₎ = 2.03; p = 0.09; AS-ODN-treated group that received AD fentanyl: t₍₅₎ = 1.26; p = 0.26; D: SE-ODN-treated group: t₍₅₎ = 1.1, p = 0.32; AS-ODN-treated group that received AD morphine: t₍₅₎ = 1.52, p = 0.19, when the mechanical nociceptive threshold is compared before and 5 d after systemic AD opioids; paired Student's t test), PGE₂ (100 ng/5 μl, i.d.) was injected and the mechanical nociceptive threshold was evaluated 30 min and 4 h later. Treatment with Gα_o AS-ODN markedly attenuates PGE₂-induced prolonged hyperalgesia in rats that received AD fentanyl (C: F_(1,10) = 23.6, *** p = 0.0007, when the hyperalgesia in the Gα_o AS-ODN-treated and the SE-ODN-treated groups is compared at the fourth hour after intradermal PGE₂; two-way repeated-measures ANOVA followed by Bonferroni's post hoc test). However, the prolongation of PGE₂-induced hyperalgesia was not affected by treatment with Gα_o AS-ODN in the AD morphine-treated group (D: F_(1,10) = 2.89, p = 0.12, when hyperalgesia was compared between the Gα_o SE-ODN-treated and AS-ODN-treated groups at the fourth hour after intradermal PGE₂; two-way repeated-measures ANOVA followed by Bonferroni's post hoc test). These findings indicate that analgesia produced by systemic AD fentanyl and morphine is not Gα_o subunit dependent; however, the Gα_o subunit does play a role in hyperalgesic priming induced by AD fentanyl, but not morphine (n = 6 paws/6 rats/group).

To the best of our knowledge, our data demonstrating that Gα_i1 and Gα_i2 AS-ODN enhances a Gα_i/o–GPCR signaling provide the first evidence that Gα_i/o subunits can inhibit Gα_i/o GPCR signaling. Our results in which Gα_i1 and Gα_i2 subunits increased analgesia induced by AD fentanyl contrasts with a fentanyl saturation-binding analysis for MOR, using urea-washed MOR membranes, where fentanyl displayed high potency in activating both Gα_i1 and Gα_o (Saidak et al., 2006).

Furthermore, our data support the suggestion that different Gα_i/o subunits mediate priming induced by AD (Gα_o) and sub-AD (Gα_i3) fentanyl. While the basis for this difference is currently unknown, we have observed that sub-AD fentanyl-induced priming is attenuated by ODN antisense to Toll-like receptor 4 (TLR4) mRNA (unpublished data, Araldi D), while hyperalgesic priming induced by repeated AD fentanyl is attenuated by MOR antisense (Araldi et al., 2018a).

Analgesia and hyperalgesic priming produced by AD morphine

Finally, we investigated whether the analgesia and hyperalgesic priming induced by AD morphine (3 mg/kg, s.c.; Araldi et al., 2019) are also Gα_i/o dependent. Rats received AS-ODN or SE-ODN to Gα_i1 (Fig. 5), Gα_i2 (Fig. 6), Gα_i3 (Fig. 7), or Gα_o (Fig. 8) mRNA, intrathecally daily for 4 d. On the fourth day, AD morphine was administered systemically (3 mg/kg, s.c.), and the mechanical nociceptive threshold was measured 1 h later. AD morphine-induced analgesia was markedly reduced in the Gα_i1 AS-ODN-treated group, compared with its SE-ODN-treated group (Fig. 5B). On the other hand, intrathecal treatment with Gα_i2 (Fig. 6B), Gα_i3 (Fig. 7B), and Gα_o (Fig. 8B) AS-ODNs did not affect morphine analgesia. Rats again received AS-ODN or SE-ODN at the end of the fourth day. Five days after AD morphine, PGE₂ (100 ng, i.d.) was injected, and the mechanical nociceptive threshold was evaluated 30 min and 4 h later. PGE₂-induced prolonged hyperalgesia was present in all four groups of AS-ODN-treated rats (Figs. 5D, 6D, 7D, and 8D). Our findings support the suggestion that, in marked contrast to the role of Gα_i1 and Gα_i2, which increased analgesia induced by AD fentanyl, AD morphine analgesia is reduced by Gα_i1 AS-ODN. Thus, the role of Gα_i1 in morphine-induced analgesia is opposite to its role in analgesia induced by AD fentanyl. Also, unexpectedly, different Gα_i/o subunits mediate hyperalgesic priming induced by sub-AD and AD fentanyl, Gα_i3 versus Gα_o dependence, respectively; and, priming induced by both sub-AD and AD morphine are Gα_i1, Gα_i2, Gα_i3, and Gα_o independent.

Together, the experiments presented here support the following suggestions: (1) that OIH, opioid-induced hyperalgesia priming and analgesia induced by different doses of different opioid analgesics are mediated by different Gα_i/o proteins; (2) that different Gα_i/o subunits mediate OIH, opioid-induced hyperalgesic priming, and analgesia induced by fentanyl and morphine; (3) that OIH, opioid-induced hyperalgesic priming, and/or analgesia may be mediated by multiple Gα_i/o proteins, differing for different doses and opioid analgesics; and, (4) that Gα_i/o proteins may inhibit as well as mediate Gα_i/o–GPCR signaling.

Discussion

Although opioids remain the most effective treatment for many forms of moderate to severe pain, they produce several treatment-limiting side effects, including analgesic tolerance, addiction, OIH, induction of the transition from acute to chronic pain (modeled in the present experiments by opioid-induced hyperalgesic priming), respiratory depression, and constipation (Chu et al., 2008; Joseph et al., 2010; Lee et al., 2011; Trang et al., 2015; Roeckel et al., 2016; Araldi et al., 2015, 2017, 2018a, b, 2019; Khomula et al., 2019, 2021). Opioid analgesics, acting at MORs, produce their effects through the activation of G-proteins (Childers, 1988; Reisine and Bell, 1993; Uhl et al., 1994; Pasternak and Standifer, 1995; Standifer et al., 1996) and β-arrestins (Gainetdinov et al., 2004; Williams et al., 2013). While opioid agonists stimulate MOR phosphorylation, β-arrestin recruitment and receptor internalization, the signaling pathways involved can vary between MOR ligands. β-Arrestin-2-dependent signaling has been implicated in clinically important dose-limiting side effects of opioid analgesics (e.g., respiratory depression, constipation, and dependence), and biased MOR agonists that preferentially activate signaling via G-proteins over β-arrestins are currently being developed (DeWire et al., 2013; Manglik et al., 2016); and one biased MOR agonist, olinvyk (oliceridine), has recently been approved for use by the US Food and Drug Administration. While such biased opioids would be expected to extend the therapeutic window, increasing the safety of opioid analgesics (DeWire et al., 2013; Manglik et al., 2016), recent studies have raised the concern that they still produce substantial side effects (Kliewer et al., 2019; Ding et al., 2020; Kliewer et al., 2020), and we recently demonstrated that the biased MOR agonist PZM21 can induce OIH and hyperalgesic priming, mediated by a unique mechanism (Araldi et al., 2018c). Previous studies have suggested a potential role of Gα_i/o proteins in diverse effects of opioids (Pasternak and Standifer, 1995; Rossi et al., 1995; Standifer et al., 1996; Hadjimarkou et al., 2002; Silva et al., 2002; Wainford and Kapusta, 2012). In the present experiments, we tested the hypothesis that two side effects of clinically used opioid analgesics, OIH and hyperalgesic priming, induced by systemically administered fentanyl and morphine, acting at the level of the spinal cord and dorsal root ganglion (DRG) neurons, are Gα_i/o dependent.

While hyperalgesia induced by systemic sub-AD fentanyl was prevented by AS-ODN directed against three G-proteins, Gα_i2, Gα_i3, and Gα_o, antisense to only one of these G-proteins, Gα_i3, prevented hyperalgesic priming. In contrast, when sub-AD morphine was administered, a different set of Gα_i/o proteins, Gα_i1, Gα_i2 and Gα_i3, mediated hyperalgesia, while hyperalgesic priming was not affected by AS-ODN to any of the tested α-subunits. Thus, while OIH produced by sub-AD fentanyl and morphine share Gα_i2 and Gα_i3 subunit dependence, Gα_o plays a role in OIH produced only by sub-AD fentanyl, and Gα_i1 plays a role only in OIH produced by sub-AD morphine. In contrast, hyperalgesic priming produced by sub-AD fentanyl was prevented by Gα_i3 AS-ODN, while sub-AD morphine-induced priming was not affected by any of the G-protein subunit AS-ODNs. The basis of this differential contribution of G-proteins in OIH and hyperalgesic priming induced by different opioid analgesics remains to be elucidated.

We also investigated the role of Gα_i/o subunits in analgesia and hyperalgesic priming induced by ADs of fentanyl and morphine. While analgesia induced by AD morphine was attenuated by Gα_i1 AS-ODN, paradoxically knockdown of Gα_i1 and Gα_i2 enhanced fentanyl analgesia. To the best of our knowledge, this is the first demonstration that G-protein α-subunits can have a negative impact on opioid-induced analgesia, and that the same α_i/o subunit can inhibit analgesia produced by one opioid analgesic (morphine) while enhancing it to another (fentanyl).

While AD fentanyl-induced priming was prevented in Gα_o AS-ODN-treated rats, none of the Gα subunits tested contributed to AD morphine-induced priming. While we do not currently have an explanation for the difference in the requirement of Gα_i3 versus Gα_o for priming induced by sub-AD versus AD fentanyl, respectively, or for the lack of a role of any of the tested Gα subunits in priming induced by sub-AD or AD morphine, it has been suggested that some effects of sub-AD doses of opioid analgesics may be mediated by their action at a different receptor, TLR4, on nociceptors (Araldi et al., 2019). The role of Gα_i/o in TLR4 signaling is currently being examined.

Given the lack of effect of Gα_i/o AS-ODNs on hyperalgesic priming induced by both sub-AD and AD morphine, it may be of interest to evaluate whether other Gα proteins contribute, such as the closely related subunit Gα_z that is thought to mediate the desensitization of MORs at supraspinal sites (Garzon et al., 2005). Otherwise, with respect to the differences between the Gα subunits involved in hyperalgesia and analgesia induced by sub-AD and AD morphine, respectively, it has been suggested that the difference in the effect of sub-AD and AD morphine may be a result of a bimodal dose-dependent effect of opioid receptor agonists. Electrophysiology studies of the effects of opioids on mouse DRG neurons demonstrated that opioid receptors activate an excitatory signaling pathway when exposed to very low concentrations of opioid agonists, and an inhibitory pathway when exposed to higher concentrations (Shen and Crain, 1994). Alternatively, since none of the Gα_i/o AS-ODNs prevented sub-AD morphine-induced priming, it is possible that it could be because of an effect of sub-AD morphine on nociceptor TLR4 (Araldi et al., 2019).

Adenylyl cyclase (AC; Sharma et al., 1977), as well as Ca²⁺ channels (Hescheler et al., 1987), G-protein-coupled inwardly rectifying K⁺ channel (GIRK) activation (North et al., 1987; Henry et al., 1995), PLC stimulation (Spencer et al., 1997), and MAPK activation (Fukuda et al., 1996) are inhibited by MOR agonists. Low-dose morphine, administered intrathecally, can also couple to Gα_s to activate protein kinase C and L-type Ca²⁺ channels to induce hyperalgesia (Esmaeili-Mahani et al., 2008); however, at higher opioid concentrations, Gα_i/o-coupled pathways are activated (Shen and Crain, 1990). In a recent study, we found that the application of fentanyl (0.5 nm) to DRG neurons cultured from rats pretreated with AD fentanyl induced a MOR-dependent increase in [Ca²⁺]_i and significantly decreased action potential rheobase in weakly IB4⁺ and IB4^– small-diameter DRG neurons (Khomula et al., 2019). Since MOR can also couple to Gα_o, which links to attenuation of action potential duration (Groer et al., 2007; McPherson et al., 2010), these data corroborate our findings that hyperalgesic priming induced by systemic AD fentanyl is Gα_o dependent.

While widely accepted, the classical model of G-protein signaling, cyclical GDP–GTP exchange in response to ligand binding to seven transmembrane receptors followed by dissociation of the G-protein subunit and activation of intracellular signaling cascades, has been challenged. For example, studies have shown that Gα_i2 and Gα_i3 can display a complex interplay in GPCR signaling. Gα_i2 and Gα_i3 can be activated simultaneously by a single ligand, resulting in adenylyl cyclase-mediated effects that are dependent only on Gα_i2 (McClue et al., 1992), and Gα_i3 can have a regulatory role, its presence inhibiting Gα_i2-induced migration and GTPγS incorporation by hindering Gα_i2 binding to the receptor (Thompson et al., 2007). These studies provide evidence to support the existence of a regulatory role for Gα_i, independent of its effect on AC, and an interplay of Gα_i proteins in transmitting G-protein-coupled receptor signals (Thompson et al., 2007). In a study of the role of T cells in “graft-versus-host” (GVH) disease, the GVH response is abolished in mice adoptively transferred with Gα_i2^−/− T cells but exacerbated in mice with Gα_i3-deficient T cells (Jin et al., 2008). Many GPCRs are able to couple to multiple G-proteins, and different GPCR agonists can lead to signaling via different G-proteins, a phenomenon referred to as “agonist-directed trafficking” (Kenakin, 1995). For example, the dopamine D₂ receptor can initiate signals via Gα_i1, Gα_i2, Gα_i3, and Gα_o1 (Gazi et al., 2003); however, ligand pharmacology can be influenced greatly by the GPCR/G-protein expression ratio (Lane et al., 2007). The D₂ receptor ligand S-(—)−3-PPP shows partial agonism mediated by Gα_o1 and antagonism/inverse agonism mediated by Gα_i1, Gα_i2, and Gα_i3 in physiologically relevant end points (Arnt et al., 1983; Hjorth et al., 1983). Thus, an emerging signaling paradigm includes the capacity of one receptor to couple to and initiate signaling through multiple G-proteins and the capacity of one G-protein to activate multiple effectors (Woehler and Ponimaskin, 2009). Limitations of the use of AS-ODN to regulate protein levels in nociceptors include the following: (1) they do not completely eliminate their target protein; (2) their intrathecal administration affects cells in the spinal cord, so that additional approaches are required to specify the cells involved; and (3) their short half-life requires repeated administration.

In conclusion, the present data support the hypothesis that Gα_i/Gα_o signaling contributes to the side effects of clinically used opioid analgesics, providing a starting point for the design of analgesics with selectivity for individual Gα_i/α_o proteins, producing a more limited range of side effects. Our current findings are summarized in Table 1.

Table 1.

Gα_i/o subunits mediating analgesia, hyperalgesia, and priming induced by sub-ADs and ADs of systemic fentanyl and morphine

	Fentanyl		Morphine
	Sub-AD	AD	Sub-AD	AD
OIH	↓ Gαi2, Gαi3 and Gαo	NS	↓ Gαi1, Gαi2 and Gαi3	NS
Analgesia	NS	↑ Gαi1 and Gαi2	NS	↓ Gαi1
Priming	↓ Gαi3	↓ Gαo	None	None

NS, Not studied; ↑, enhances; ↓, attenuates.