Abstract
Chronic opioid exposure induces tolerance to the pain-relieving effects of opioids but sensitization to some other effects. While the occurrence of these adaptations is well understood, the underlying cellular mechanisms are less clear. This study aimed to determine how chronic treatment with morphine, a prototypical opioid agonist, induced adaptations to subsequent morphine signaling in different subcellular contexts. Opioids acutely inhibit glutamatergic transmission from medial thalamic (MThal) inputs to the dorsomedial striatum (DMS) via activity at μ-opioid receptors (MORs). MORs are present in somatic and presynaptic compartments of MThal neurons terminating in the DMS. We investigated the effects of chronic morphine treatment on subsequent morphine signaling at MThal–DMS synapses and MThal cell bodies in male and female mice. Surprisingly, chronic morphine treatment increased subsequent morphine inhibition of MThal–DMS synaptic transmission (morphine facilitation) in male, but not female, mice. At MThal cell bodies, chronic morphine treatment decreased subsequent morphine activation of potassium conductance (morphine tolerance) in both male and female mice. In knock-in mice expressing phosphorylation-deficient MORs, chronic morphine treatment resulted in tolerance to, rather than facilitation of, subsequent morphine signaling at MThal–DMS terminals, suggesting phosphorylation deficiency unmasks adaptations that counter the facilitation observed at presynaptic terminals in wild-type mice. The results of this study suggest that the effects of chronic morphine exposure are not ubiquitous; rather adaptations in MOR function may be determined by multiple factors such as subcellular receptor distribution, influence of local circuitry, and sex.
Significance Statement
Repeated opioid use causes tolerance to their pain-relieving effects but can exacerbate some undesirable effects limiting their clinical utility. A detailed understanding of the physiological adaptations that contribute to the development of tolerance is critical to develop mitigation strategies. This study found that within medial thalamic projection neurons, chronic morphine treatment induced adaptations that were not ubiquitous. Instead, prior morphine exposure increased morphine effects at thalamic terminals in the dorsomedial striatum only in male mice but decreased morphine effects at medial thalamic cell bodies in both sexes. In mice lacking phosphorylation sites on MOR, chronic morphine treatment decreased, rather than increased, morphine effects at thalamic terminals in the dorsomedial striatum, implicating receptor phosphorylation in driving adaptations observed in wild-type mice.
Introduction
Repeated exposure to opioids such as morphine results in tolerance to their pain-relieving properties, whereby increasing doses of drug are required to achieve the same effect (McQuay, 1999). Conversely, behavioral sensitization develops to other properties, notably conditioned reward and locomotor stimulation, whereby repeated exposure enhances drug response (Lett, 1989; Gaiardi et al., 1991; Lamb et al., 1991; Stewart and Badiani, 1993). While behavioral tolerance and sensitization are well described, the underlying cellular adaptations are not. Defining these mechanisms is challenging given that different opioid responses to which tolerance or sensitization develop are primarily mediated through the same receptor, the μ-opioid receptor (MOR; Matthes et al., 1996).
The MORs are distributed throughout neurons, regulating neuronal excitability in somatodendritic (somatic) regions and inhibiting neurotransmitter release in axonal (presynaptic) compartments. Previous work has established important differences in how presynaptic and somatic MOR signaling adapts to chronic opioid exposure. In the somatic compartment, chronic morphine generally results in reduced opioid efficacy, or tolerance, although the degree of tolerance can vary between brain regions (Christie et al., 1987; Bagley et al., 2005; Levitt and Williams, 2018). It is becoming recognized that many aspects of cellular morphine tolerance in the somatic compartment are mediated by MOR phosphorylation, and loss of phosphorylation sites within the C-terminal tail of MOR attenuates somatic cellular tolerance. Within the presynaptic compartment, multiple adaptations to chronic opioid exposure have been observed, tolerance in some instances (Fyfe et al., 2010; Atwood et al., 2014; Matsui et al., 2014) and enhanced opioid efficacy, or facilitation, in others (Chieng and Williams, 1998; Ingram et al., 1998; Hack et al., 2003; Pennock et al., 2012). The MOR phosphorylation also regulates presynaptic cellular tolerance in cultured striatal neurons (Jullié et al., 2020, 2022). However, the role of phosphorylation in mediating presynaptic facilitation and tolerance in intact brain circuits is not established. Complicating the matter, studies investigating these differences have been done across species and brain regions making it difficult to generalize how a particular cell type or synapse will adapt to repeated opioid exposure. Differences in how male and female human patients and rodents respond to opioids are well documented. In rodents, MOR-selective opioids are generally more potent in males but in humans are more potent in females (Craft, 2003). Numerous studies in rats have also demonstrated greater or more rapid morphine tolerance in males (Kest et al., 2000). Given these phenotypic differences, there may also be important sex differences in how opioids alter cellular signaling. However, the influence of sex on opioid-induced cellular adaptations is poorly understood.
Neurons in the medial thalamus (MThal), centered around the mediodorsal nucleus, provide an ideal system to investigate presynaptic and somatic adaptations to opioids. MThal neurons express relatively high levels of MOR, and single thalamic neurons send broad axonal projections to both the striatum and many cortical areas. MThal projections provide a major source of glutamatergic innervation to the dorsomedial striatum (DMS; Hunnicutt et al., 2014, 2016). Signaling within the MThal and DMS is involved in numerous opioid-sensitive processes, including motivated learning, movement, and perception of pain affect (Graybiel et al., 1994; Peyron et al., 2000; Price, 2000; Johansen et al., 2001; Balleine et al., 2007; Navratilova et al., 2015; Vogt, 2015; McDevitt et al., 2021). Glutamatergic thalamic innervation of the DMS, specifically, has been implicated in learning processes, including reinforcement-based learning (Johnson et al., 2020; Kato et al., 2021). Because of their behavioral relevance and high levels of MOR expression, MThal–DMS projections serve as a relevant system to directly compare chronic morphine effects on subsequent morphine signaling within somatic and presynaptic compartments of the same neuronal population. The objective of the present study was to determine how chronic morphine exposure modulated subsequent morphine signaling within MThal axon terminals synapsing in the DMS and MThal neuronal cell bodies and whether the observed effects were sex-specific. We further investigated the role of MOR phosphorylation in mediating these adaptations.
Materials and Methods
Drugs
Reagent | Source | Identifiers |
---|---|---|
Naloxone | Hello Bio | HB2451 |
Picrotoxin | Hello Bio | HB0506 |
Dizocilpine (MK-801) | Hello Bio | HB0004 |
Baclofen | Hello Bio | HB0953 |
Bestatin | Sigma Aldrich | B8385 |
Thiorphan | Sigma Aldrich | T6031 |
[met5]enkephalin | Sigma Aldrich | M6638 |
ML-297 | Tocris Bioscience | 5380 |
Morphine sulfate | Sigma Aldrich | 1448005 |
Morphine sulfate | Spectrum Chemical | M1167 |
rAAV2-hsyn-hChR2(H134R)-EYFP-WPRE-PA | Gift of Karl Deisseroth produced by UNC Vector Core | Lot # AV4384H |
Cholera toxin subunit B Alexa Fluor 488 conjugate | Thermo Fisher | C22841 |
Animals
All procedures were conducted in accordance with the National Institutes of Health guidelines and with approval from the Institutional Animal Care and Use Committee at the University of Michigan. Mice were maintained on a 12 h light/dark cycle and given ad libitum access to food and water. C57Bl/6J mice were obtained from Jackson Laboratory, and MOR 10 S/T-A mice were created by Dr. Stefan Schulz (Kliewer et al., 2019). Mice were 4–8 weeks old at the time of viral injection and 6–10 weeks old at the time of brain slice preparation. Mice of both sexes were used.
Chronic opioid treatment
Morphine-treated mice were implanted with an osmotic minipump (Alzet Model 2001) continuously releasing morphine (80 mg/kg/d) for 7 d prior to brain slice preparation. Drug concentrations were calculated based on the mean pump rate and mouse mass at the time of surgery to achieve the desired dose. Mice were anesthetized with isoflurane (4% induction, 2% maintenance), and an incision was made along the lower back. Pumps were inserted subcutaneously, and the incision was closed with surgical glue and wound clips. Pumps remained implanted until mice were killed for brain slice preparation. Brain slices were incubated in the absence of morphine for a minimum of 1 h prior to experimentation to ensure no residual drug was present in the slices during the baseline recordings, representing a state of acute morphine withdrawal.
Stereotaxic injections
For evoked synaptic responses, mice were injected bilaterally with an adeno-associated virus type 2 encoding channelrhodopsin-2 [ChR2; AAV2-hsyn-ChR2(H134R)-EYFP] targeting MThal. Mice were anesthetized with isoflurane (4% induction, 2% maintenance) and placed on a stereotaxic frame (Kopf Instruments). An incision was made along the scalp and holes drilled through the skull above the injection sites. A glass pipette filled with virus was inserted into the brain and lowered to the appropriate depth. In total, 60–70 nl of virus was injected bilaterally into the medial thalamus (A/P, −1.2 mm; M/L, ± 0.6 mm; D/V, 3.6 mm). Virus was delivered using a microinjector (Nanoject II, Drummond Scientific). For somatic recordings, mice were injected bilaterally with cholera toxin subunit B conjugated to Alexa 488 (Ctx-488; Thermo Fisher) into DMS for retrograde labeling of DMS-projecting medial thalamic neurons. Injections were performed identically to viral injections, with the exceptions that 130–140 nl was injected and the following stereotaxic coordinates were used for DMS: A/P + 0.8, M/L ± 1.2, and D/V 3.6 mm.
Brain slice electrophysiology
Brain slices were prepared 2–3 weeks following injection of ChR2 or 1–2 weeks following injection of Ctx-488. Mice were deeply anesthetized with isoflurane and decapitated. Brains were removed and mounted for slicing with a vibratome (Model 7000 smz, Campden Instruments). During slicing brains were maintained at room temperature in carbogenated Krebs’ solution containing the following (in mM): 136 NaCl, 2.5 KCl, 1.2 MgCl2–6H2O, 2.4 CaCl2–2 H2O, 1.2 NaH2PO4, 21.4 NaHCO3, 11.1 dextrose supplemented with 5 µM MK-801. Coronal sections (250–300 µM) containing the DMS or MThal were made and incubated in carbogenated Krebs’ solution supplemented with 10 µM MK-801 at 32°C for 30 min. Slices were then maintained at room temperature in carbogenated Krebs’ solution until used for recording. Only one cell was recorded from each slice to ensure baseline recordings were not contaminated from prior drug application. A maximum of three cells per animal were included in each dataset to avoid oversampling.
For DMS recordings, borosilicate glass patch pipettes (Sutter Instrument) were pulled to a resistance of 2.0–4.0 MΩ and filled with a potassium gluconate-based internal solution (in mM: 110 potassium gluconate, 10 KCl, 15 NaCl, 1.5 MgCl2, 10 HEPES, 1 EGTA, 2 Na ATP, 0.4 Na GTP, 7.8 Na2 phosphocreatine). Slices were placed in the recording chamber and continuously perfused with carbogenated Krebs’ solution supplemented with 100 µM picrotoxin at 32–34°C. Striatal medium spiny neurons (MSNs) were identified based on cell morphology, resting membrane potential, and firing frequency (Kreitzer, 2009), and MSN subtype (D1 vs D2 expressing) was not distinguished. Whole-cell recordings were made in MSNs in voltage-clamp mode at −70 mV holding potential. All drug solutions for DMS recordings were prepared in carbogenated Krebs’ solution supplemented with 100 µM picrotoxin.
For MThal recordings, patch pipettes were pulled to a resistance of 2.0–4.0 MΩ and filled with a potassium methanesulfonate-based internal solution containing the following (in mM): 115 potassium methane sulfonate, 10 KCl, 15 NaCl, 1.5 MgCl2, 10 HEPES-K, 10 BAPTA 4 K, 2 Na ATP, 0.4 Na GTP, 7.8 Na2 phosphocreatine. Slices were placed in the recording chamber and continuously perfused with carbogenated Krebs’ solution at 32–34°C. Whole-cell recordings were made in DMS-projecting thalamic neurons, identified based on cell morphology and the presence of Ctx-488 in the soma. Recordings of G-protein-gated inwardly rectifying K+ channel (GIRK) currents were made in voltage-clamp mode, and cells were maintained at a holding potential of −60 mV. During recording, all solutions were supplemented with 10 µM ML-297 to enhance the size of the GIRK-mediated outward currents for quantification purposes.
Whole-cell recordings were made with a MultiClamp 700B amplifier (Molecular Devices) digitized at 20 kHz (National Instruments BNC-2090A). Synaptic recordings were acquired using Matlab WaveSurfer (MathWorks). Currents were evoked every 30 s by illuminating the field of view through the microscope objective (Olympus BX51WI) using a transistor-transistor logic (TTL)-controlled LED driver and a 470 nm LED (Thorlabs). The LED stimulation duration was 1 ms and power output measured after the microscope objective ranged from 0.5 to 3 mW, adjusted to obtain consistent current amplitudes across cells. Somatic responses were recorded using LabChart (ADInstruments) to passively record and measure drug-induced changes in holding current. For both presynaptic and somatic responses, series resistance was monitored throughout the recordings, and only recordings in which the series resistance remained <15 MΩ and did not change more than 18% were included.
Data analysis
For synaptic responses, raw data were analyzed using Matlab or AxoGraph. Peak current amplitude was calculated for each sweep after baseline subtraction, with baseline defined as the average holding current during the first 10 ms of each sweep, prior to optical stimulation. For each condition (baseline, drug, washout/reversal), baseline subtracted sweeps were averaged together, and peak current amplitude of the averaged trace was calculated. For the baseline condition, the first 2–4 sweeps were omitted from the average to allow the currents to stabilize. For the drug and washout/reversal conditions, the first 4–8 sweeps were omitted from the average to allow for equilibration of drug or washout of drug within the tissue. Average drug and washout/reversal current amplitudes were normalized to the average baseline current peak amplitude and plotted as percentage of baseline to analyze sensitivity of MThal terminals to opioid-mediated presynaptic inhibition. For somatic responses, raw data were analyzed using AxoGraph. Average holding current was calculated for each condition, and morphine-induced GIRK current was normalized to baclofen-induced GIRK conductance. Across all conditions, perfusion of 10 µM ML-297 alone induced an average current of 12.59 ± 1.96 pA (mean ± SEM) with no differences across sex or treatment. To account for a small changes in holding current or GIRK current induced by ML-297 alone, a two-region sloping baseline subtraction was performed using AxoGraph, in which a line interpolated between the ML-297 baseline and naloxone reversal was subtracted from the trace. Cells in which morphine responses were not distinguishable from the ML-297 baseline were considered nonresponders and excluded from analysis. Statistical analysis was performed using GraphPad Prism (GraphPad Software). Statistical comparisons were made using a t test or one-way or two-way ANOVA with Tukey’s (one-way ANOVA) or Šidák’s (two-way ANOVA) post hoc analysis. Complete datasets for male and female mice were obtained and analyzed to examine sex as a factor for all experiments involving morphine. For all experiments, statistical significance was defined as p < 0.05. For all comparisons, n (number of cells) and N (number of animals) are both reported.
Results
MOR agonists inhibit optically evoked MThal–DMS glutamate release
Agonist-induced activation of MOR decreases the amplitude of optically evoked excitatory postsynaptic currents (oEPSCs) in DMS MSNs via presynaptic inhibition (Atwood et al., 2014; Birdsong et al., 2019; Adhikary et al., 2022). We first demonstrated opioid-mediated inhibition of oEPSCs in MThal–DMS terminals was induced by opioid agonists morphine and [Met5]-enkephalin (ME) by performing whole-cell recordings in voltage-clamp mode in DMS MSNs following viral expression of ChR2 in MThal neurons (Fig. 1A,B) in male and female mice. After recording a stable baseline of oEPSCs, agonist was perfused onto the slices, followed by the opioid receptor antagonist naloxone or Krebs’ solution to reverse inhibition (Fig. 1C–E). Naloxone was perfused to reverse inhibition by morphine, and ME was reversed with Krebs’ solution due to ME readily washing out of the slice. Consistent with opioid-mediated presynaptic inhibition of glutamate release from MThal terminals, perfusion of morphine (3 µM) caused a significant decrease in the mean amplitude of the MThal–DMS oEPSC relative to baseline, and this effect was largely reversed upon perfusion of naloxone (Fig. 1C,E,F; p < 0.0001, main effect of condition, repeated measures one-way ANOVA; baseline vs morphine, p < 0.0001; baseline vs naloxone, p = 0.0091; morphine vs naloxone, p = 0.0052, Tukey’s multiple-comparisons test). Like morphine, perfusion of ME (3 µM) also caused a significant decrease in oEPSC mean amplitude in a reversible manner (Fig. 1D,E,G; p = 0.0018, main effect of condition, repeated measures one-way ANOVA; baseline vs ME, p = 0.0082; baseline vs washout, p = 0.0615; ME vs washout, p = 0.0091, Tukey’s multiple-comparisons test).
μ-Opioid receptor agonists inhibited glutamatergic MThal–DMS oESPCs. A, Schematic showing viral-mediated expression of ChR2 in the MThal and axonal projections onto striatal MSNs. B, Examples of acute brain slices showing EYFP fluorescence in the injection site (MThal, left) and axon terminals in the DMS (right). C, Representative traces showing oEPSCs in an MSN evoked by 470 nm LED light pulses during baseline (gray), perfusion of morphine (3 µM, orange), and perfusion of naloxone (1 µM, black). D, Representative traces showing oEPSCs in an MSN evoked by 470 nm LED light pulses during baseline (gray), perfusion of ME (3 µM, pink), and washout (black). E, Time course of normalized oEPSC amplitude during baseline, perfusion of morphine (3 µM) or ME (3 µM), followed by perfusion of naloxone (1 µM), or washout [morphine, N = 12 (6 M, 6 F); n = 14 (7 M, 7 F); ME, N = 6 (2 M, 4 F); n = 6 (2 M, 4 F)]. F, Raw amplitudes of oEPSCs in individual striatal MSNs during baseline, perfusion of morphine (3 µM), and perfusion of naloxone [1 µM; main effect of condition, F(1.687, 21.93) = 28.01; p < 0.0001; N = 12 (6 M, 6 F); n = 14 (7 M, 7 F); filled circles represent males; open circles represent females; repeated measures 1-way ANOVA; baseline vs morphine, p < 0.0001; baseline vs naloxone, p = 0.0091; morphine vs naloxone, p = 0.0052; Tukey’s multiple-comparisons test]. G, Raw amplitudes of oEPSCs in individual striatal MSNs during baseline, perfusion of ME (3 µM), and washout [main effect of condition, F(1.259, 6.297) = 24.17; p = 0.0018; N = 6 (2 M, 4 F); n = 6 (2 M, 4 F); repeated measures 1-way ANOVA; baseline vs ME, p = 0.0082; baseline vs washout, p = 0.0615; ME vs washout, p = 0.0091; Tukey’s multiple-comparisons test]. H, Summary data comparing oEPSC inhibition following perfusion of 3 µM (circles) and 10 µM (squares) morphine (orange) and ME (pink) in male and female mice [3 µM morphine, 72.26 ± 2.44% of baseline; N = 16 (7 M, 9 F); n = 21 (10 M, 11 F); 10 µM morphine, 72.32 ± 4.24% of baseline; N = 7 (5 M, 2 F); n = 8 (5 M, 3 F); 3 µM ME, 31.61 ± 5.38% of baseline; N = 11 (7 M, 4 F); n = 11 (7 M, 4 F); 10 µM ME, 25.16 ± 4.20% of baseline; N = 6 (5 M, 1 F); n = 8 (6 M, 2 F); filled data points represent males; open data points represent females; main effect of concentration, F(1, 44) = 0.6015; p = 0.4421; main effect of agonist, F(1, 44) = 113.8; p < 0.0001; concentration x agonist interaction, F(1, 44) = 0.6253; p = 0.4333; ordinary 2-way ANOVA; 3 µM morphine vs 10 µM morphine, p > 0.9999; 3 µM ME vs 10 µM ME, p = 0.8815; 3 µM morphine vs 3 µM ME, p < 0.0001; 10 µM morphine vs 10 µM ME, p < 0.0001; Šidák’s multiple-comparisons test]. I, Summary data comparing oEPSC inhibition following perfusion of morphine (3 µM) in male versus female mice (males, 74.70 ± 4.06% of baseline; N = 7; n = 10; females, 70.04 ± 2.85% of baseline; N = 9; n = 11; t(19) = 0.9533; p = 0.3524; unpaired t test). Lines and error bars represent mean ± SEM. ****p < 0.0001.
The inhibition of oEPSCs by 3 µM morphine was significantly less than inhibition induced by the same concentration of ME (Fig. 1H; 3 µM morphine, 72.26 ± 2.44% of baseline; N = 15; n = 21; 3 µM ME, 31.61 ± 5.38% of baseline; N = 10; n = 11; 3 µM morphine vs 3 µM ME, p < 0.0001; ordinary two-way ANOVA with Šidák’s multiple-comparisons test). Likewise, our previous work has shown that perfusion of the MOR full agonist DAMGO inhibits MThal–DMS oEPSCs to 38.2 ± 6.1% of baseline (Birdsong et al., 2019). Inhibition of oEPSCs by 10 µM morphine was similar to inhibition by 3 µM morphine, indicating that 3 µM morphine is a saturating concentration (Fig. 1H; 3 µM morphine, 72.26 ± 2.44% of baseline; 10 µM morphine, 72.32 ± 4.24% of baseline; 3 µM morphine vs 10 µM morphine, p > 0.9999; ordinary two-way ANOVA with Šidák’s multiple-comparisons test). Inhibition of oEPSCs by 3 µM ME and 10 µM ME were also similar, indicating that inhibition by ME is saturated at 3 µM (Fig. 1H; 3 µM ME, 31.61 ± 5.38% of baseline; 10 µM ME, 25.16 ± 4.20% of baseline; 3 µM ME vs 10 µM ME, p = 0.8815; ordinary two-way ANOVA with Šidák’s multiple-comparisons test).
These results indicate that, at this synapse, morphine acts as a partial agonist for inhibition of MThal–DMS EPSCs. Morphine was selected for future experiments because, as a partial agonist, observable changes in the sensitivity of MORs are less likely to be occluded by receptor reserve than with a full agonist. Under these conditions, there were no statistically significant differences in oEPSC inhibition by morphine between slices from untreated male and female mice (Fig. 1I; males, 74.70 ± 4.06% of baseline; N = 7; n = 10; females, 70.04 ± 2.85% of baseline; N = 9; n = 11; t(19) = 0.9533; p = 0.3524; unpaired t test).
Chronic morphine treatment increased morphine sensitivity at MThal–DMS terminals in male, but not female, mice
We next investigated whether exposing mice to chronic morphine altered the sensitivity of MThal–DMS oEPSCs to inhibition by a subsequent morphine challenge in a sex-dependent manner. Chronic morphine treatment was achieved through implantation of an osmotic minipump continuously releasing morphine (80 mg/kg/d) for 7 d prior to recording (Fig. 2I). To ensure no morphine from the minipump was present in the slices during the baseline recordings, slices were incubated in the absence of morphine for a minimum of 1 h before performing electrophysiology recordings. After recording a stable baseline, morphine (3 µM) was perfused onto the slices, followed by naloxone (1 µM; Fig. 2A–F). Surprisingly, morphine caused greater inhibition of oEPSCs in morphine withdrawn slices from morphine-treated mice compared with slices from drug-naive mice in males. However, in slices from female mice, no differences were observed between morphine inhibition of oEPSCs in drug-naive and chronically morphine treated mice (Fig. 2G; naive male, 74.70 ± 4.06% of baseline; chronic morphine male, 54.76 ± 4.11% of baseline; naive female, 70.04 ± 2.85% of baseline; chronic morphine female, 70.44 ± 2.82% of baseline; treatment × sex interaction, p = 0.0052; male naive vs chronic morphine, p = 0.0004; female naive vs chronic morphine, p = 0.9957; Šidák’s multiple-comparisons test). To ensure the effects observed in males were a result of morphine exposure rather than an effect associated with minipump implantation, recordings were performed in slices taken from a cohort of male mice implanted with saline-containing osmotic minipumps (Fig. 2H). Morphine inhibition of oEPSCs in slices from saline-treated mice did not differ from naive mice but was significantly less than morphine inhibition in slices from chronic morphine-treated mice, suggesting that facilitation of morphine inhibition was an effect of morphine exposure rather than surgical manipulation (saline treated, 69.10 ± 4.11% of baseline; F(2, 29) = 6.6910; p = 0.0035; ordinary one-way ANOVA; naive vs saline treated, p = 0.6038; saline treated vs chronic morphine treated, p = 0.0413; Tukey’s multiple-comparisons test). These results suggest that chronic morphine exposure resulted in facilitation of, rather than tolerance to, morphine inhibition of glutamatergic MThal–DMS oEPSCs and that this adaptation occurred in a sex-specific manner.
Chronic morphine treatment increased morphine sensitivity at MThal–DMS terminals in male, but not female, mice. A, B, Representative traces showing oEPSCs in MSNs during baseline (gray), following perfusion of morphine (3 µM, orange), and following perfusion of naloxone (1 µM, black) in drug-naive male (A) and chronic morphine-treated male (B) mice. C, Time course of normalized oEPSC amplitude during baseline, perfusion of morphine (3 µM), and perfusion of naloxone (1 µM) in drug-naive (black) and chronically treated (purple) male mice (naive, N = 5; n = 7; chronic morphine, N = 6; n = 9). D, E, Representative traces showing oEPSCs in MSNs during baseline (gray), following perfusion of morphine (3 µM, orange), and following perfusion of naloxone (1 µM, black) in drug-naive female (D) and chronically treated female (E) mice. F, Time course of normalized oEPSC amplitude during baseline, perfusion of morphine (3 µM), and perfusion of naloxone (1 µM) for drug-naive (black) and chronically treated (purple) female mice (naive, N = 6; n = 7; chronic morphine, N = 6; n = 10). G, Summary of normalized oEPSC inhibition following perfusion of morphine in drug-naive and chronically treated male and female mice (naive male, 74.70 ± 4.064% of baseline; chronic morphine male, 54.76 ± 4.11% of baseline; naive female, 70.04 ± 2.85% of baseline; chronic morphine female, 70.44 ± 2.82% of baseline; main effect of treatment, F(1, 40) = 8.048; p = 0.0071; main effect of sex, F(1, 40) = 2.524; p = 0.1200; treatment × sex interaction, F(1, 40) = 8.723; p = 0.0052; N = 7–9; n = 10–12 for each group; ordinary 2-way ANOVA; male naive vs chronic morphine, p = 0.0004; female naive vs chronic morphine, p = 0.9957; Šidák’s multiple-comparisons test). H, Summary of normalized oEPSC inhibition following perfusion of morphine in naive, saline-treated, and morphine-treated male mice (untreated, 74.70 ± 4.064% of baseline; morphine treated, 54.76 ± 4.11% of baseline; saline treated, 69.10 ± 4.11% of baseline; F(2, 29) = 6.6910; p = 0.0035; ordinary one-way ANOVA; naive vs saline treated, p = 0.6038; saline treated vs chronic morphine, p = 0.0413; Tukey’s multiple-comparisons test). I, Schematic of chronic morphine treatment. Morphine (80 mg/kg/d) was continuously administered via osmotic minipump for 7 d prior to brain slice preparation and recording. Lines and error bars represent mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001.
Chronic morphine treatment attenuated morphine-activated GIRK current amplitude at MThal cell bodies in male and female mice
We next investigated whether morphine treatment affected subsequent morphine signaling at MThal cell bodies in male and female mice. Within the somatodendritic compartment, activation of MOR can activate GIRK channels. When measuring somatic GIRK conductance, chronic opioid exposure has been shown to induce varying degrees of tolerance (or decreased response amplitude) to morphine in a cell type-specific manner (Christie et al., 1987; Bagley et al., 2005; Levitt and Williams, 2018). Using exogenously expressed MOR, we have previously shown that MOR agonists can activate GIRK in MThal cell bodies and that this signaling desensitizes over time in an MOR phosphorylation-dependent manner (Birdsong et al., 2015). However, to our knowledge, chronic opioid effects at MThal cell bodies specifically have not yet been investigated.
To address this, the retrograde tracer Ctx-488 was injected into the DMS to fluorescently label DMS-projecting MThal neurons (Fig. 3A–C). Whole-cell voltage-clamp recordings were made from identified Ctx-488-positive MThal neurons in acute brain slices prepared 1–2 weeks later. GIRK currents were activated by bath perfusion of the partial MOR agonist morphine (3 µM) and the GABAB receptor agonist baclofen (3 µM; Fig. 3D–G). To compensate for varying degrees of GIRK expression between cells, the morphine response was normalized to the baclofen response within each cell. Six cells across sexes and treatment conditions (three male naive, one male chronic morphine, and two female chronic morphine) did not respond to morphine perfusion and were excluded from analysis. Comparing the effects of chronic morphine treatment and sex, we observed a main effect of treatment, where baclofen-normalized morphine responses were significantly smaller in slices from chronically treated mice (Fig. 3H; male naive, Imorphine = 52.61 ± 4.22% of Ibaclofen; male chronic morphine, Imorphine = 36.30 ± 7.10% of Ibaclofen; female naive, Imorphine = 45.85 ± 7.87% of Ibaclofen; female chronic morphine, Imorphine = 33.93 ± 4.37% of Ibaclofen; main effect of treatment, p = 0.0266; main effect of sex, p = 0.4595; treatment × sex interaction, p = 0.7217; ordinary two-way ANOVA). Post hoc analysis did not show a statistically significant effect of chronic morphine treatment in males or females (male naive vs chronic morphine, p = 0.1296; female naive vs chronic morphine, p = 0.3209; Šidák’s multiple-comparisons test). These results indicate that chronic morphine treatment induced small but significant tolerance to a subsequent morphine challenge at MThal cell bodies in both male and female mice.
Chronic morphine treatment attenuated morphine-activated GIRK current amplitude at MThal cell bodies in male and female mice. A, Schematic showing retrograde labeling of MThal projection neurons following injection of Ctx-488 into the DMS. B, Examples of acute brain slices showing fluorescence in the injection site (DMS, left) and retrograde labeling site (MThal, right). C, Example of an acute brain slice at 5× magnification showing fluorescence in the cell bodies of individual DMS-projecting MThal neurons. D, E, Representative traces showing GIRK conductance at medial thalamic cell bodies following perfusion of morphine (3 µM), perfusion of naloxone (1 µM), and perfusion of baclofen (3 µM) in drug-naive male (D) and chronically treated male (E) mice. F, G, Representative traces showing GIRK current at medial thalamic cell bodies following perfusion of morphine (3 µM), perfusion of naloxone (1 µM), and perfusion of baclofen (3 µM) in drug-naive female (F) and chronically treated female (G) mice. H, Summary of morphine-induced (3 µM) GIRK currents normalized to baclofen-induced GIRK currents in drug-naive and chronically treated male and female mice (naive male, Imorphine = 52.61 ± 4.22% of Ibaclofen; chronic morphine male, Imorphine = 36.30 ± 7.10% of Ibaclofen; naive female, Imorphine = 45.85 ± 7.87% of Ibaclofen; chronic morphine female, Imorphine = 33.93 ± 4.37% of Ibaclofen; main effect of treatment, F(1, 36) = 5.342; p = 0.0266; main effect of sex, F(1, 36) = 0.4595; p = 0.4595; treatment × sex interaction, F(1, 36) = 0.1289; p = 0.7217; N = 6–8; n = 10 per group; ordinary 2-way ANOVA; male naive vs chronic morphine, p = 0.1296; female naive vs chronic morphine, p = 0.3209; Šidák’s multiple-comparisons test). I, Summary of raw morphine-induced (3 µM) GIRK currents in drug-naive and chronically treated male and female mice (naive male, Imorphine = 90.56 ± 8.66 pA; chronic morphine male, Imorphine = 65.17 ± 12.14 pA; naive female, Imorphine = 81.75 ± 11.81 pA; chronic morphine female, Imorphine = 66.50 ± 13.75 pA; main effect of treatment, F(1, 36) = 2.998; p = 0.0920; main effect of sex, F(1, 36) = 0.1014; p = 0.7520; treatment × sex interaction, F(1, 36) = 0.1870; p = 0.6680; N = 6–8; n = 10 per group; ordinary 2-way ANOVA). Lines and error bars represent mean ± SEM. *p < 0.05.
Raw GIRK current amplitudes induced by perfusion of morphine were also examined. No significant effect of chronic morphine treatment was observed in raw GIRK currents induced by morphine in slices from male or female mice (Fig. 3I; I, male naive, Imorphine = 90.56 ± 8.67 pA; male chronic morphine, Imorphine = 65.17 ± 12.14 pA; female naive, Imorphine = 81.75 ± 11.81 pA; female chronic morphine, Imorphine = 66.50 ± 13.75 pA; main effect of treatment, p = 0.0920; ordinary two-way ANOVA). However, the high degree of variability in raw current amplitude makes these findings difficult to interpret. Together, these results indicate that chronic morphine treatment induced small but significant tolerance to subsequent morphine signaling in the somatic compartment in MThal projection neurons in both male and female mice.
Mice lacking phosphorylation sites in the MOR C terminus are more sensitive to morphine at MThal–DMS terminals and develop tolerance following chronic morphine treatment
Receptor phosphorylation is a key regulator of MOR signaling. However, the role of phosphorylation in regulating MOR signaling in the presynaptic compartment is not well established. Using a knock-in mouse line in which mice express MORs with 10 serine (S) and threonine (T) to alanine (A) mutations in the MOR C-terminal tail (10 S/T-A; Fig. 4L,K; Kliewer et al., 2019), we first determined whether loss of phosphorylation sites altered basal sensitivity to morphine at MThal–DMS terminals. Phosphorylation-deficient MORs have been shown to display reduced receptor internalization and desensitization; however, binding affinity, activation kinetics, and signaling through the G protein pathway are similar to WT receptors. MOR 10 S/T-A mice display enhanced opioid analgesia and reduced tolerance, further suggesting phosphorylation plays an important role in regulating opioid effects following acute and chronic exposure (Kliewer et al., 2019). To our knowledge, the effects of the 10 S/T-A mutations have not been characterized in presynaptic terminals. We first aimed to determine whether opioid-mediated inhibition of synaptic transmission was altered under baseline conditions in MOR 10 S/T-A male and female mice relative to WT mice. Inhibition of glutamate release from MThal–DMS terminals by perfusion of ME or morphine in slices from untreated 10 S/T-A mice was quantified and compared with WT mice (Fig. 4A–D,G).
Mice lacking phosphorylation sites in the MOR C terminus are more sensitive to morphine at MThal–DMS terminals and develop tolerance following chronic morphine treatment. A, Representative traces showing oEPSCs in DMS MSNs during baseline (gray), following perfusion of ME (3 µM, pink), and following washout (black) in drug-naive MOR 10 S/T-A mice. B, Summary data comparing oEPSC inhibition following perfusion of ME (3 µM) in male and female WT and 10 S/T-A mice [WT, 31.61 ± 5.38% of baseline; N = 11 (7 M, 4 F); n = 11 (7 M, 4 F); 10 S/T-A, 30.83 ± 6.03% of baseline; N = 5 (3 M, 2 F), n = 9 (6 M, 3 F); t(18) = 0.09722; p = 0.9236; filled circles represent males; open circles represent females]. C, Summary data comparing oEPSC inhibition following perfusion of morphine (3 µM) in male and female WT and 10 S/T-A mice (WT male, 74.70 ± 4.064% of baseline; 10 S/T-A male, 57.79 ± 4.49% of baseline; WT female, 70.04 ± 2.85% of baseline; 10 S/T-A female, 59.74 ± 6.35% of baseline; main effect of strain, F(1, 38) = 8.957; p = 0.0048; main effect of sex, F(1, 38) = 0.08938; p = 0.7666; strain × sex interaction, F(1, 38) = 0.5285; p = 0.4717; N = 6–9; n = 10–11 for each group; ordinary 2-way ANOVA; male WT vs 10 S/T-A, p = 0.0244; female WT vs 10 S/T-A, p = 0.2210; Šidák’s multiple-comparisons test). D, E, Representative traces showing oEPSCs in DMS MSNs during baseline (gray), following perfusion of morphine (3 µM, orange), and following perfusion of naloxone (1 µM, black) in drug-naive male (D) and chronically treated male (E) 10 S/T-A mice. F, Time course of normalized oEPSC amplitude in DMS MSNs during baseline, perfusion of morphine (3 µM), and perfusion of naloxone (1 µM) for drug-naive male (gray) and chronically treated male (cyan) mice (naive, N = 9; n = 11; chronic morphine, N = 5; n = 8). G, H, Representative traces showing oEPSCs in DMS MSNs during baseline (gray), following perfusion of morphine (3 µM, orange), and following perfusion of naloxone (1 µM, black) in drug-naive female (G) and chronically treated female (H) 10 S/T-A mice. I, Time course of normalized oEPSC amplitude in DMS MSNs during baseline, perfusion of morphine (3 µM), and perfusion of naloxone (1 µM) for drug-naive female (gray) and chronically treated female (cyan) mice (naive, N = 5; n = 7; chronic morphine, N = 5; n = 7). J, Summary of normalized oEPSC inhibition following perfusion of morphine in drug-naive and chronically treated male and female 10 S/T-A mice (naive male, 57.79 ± 4.49% of baseline; chronic morphine male, 70.03 ± 3.70% of baseline; naive female, 59.74 ± 6.35% of baseline; chronic morphine female, 69.79 ± 2.46% of baseline; main effect of treatment, F(1, 39) = 6.485; p = 0.0149; main effect of sex, F(1, 39) = 0.03786; p = 0.8467; treatment × sex interaction, F(1, 38) = 0.06250; p = 0.8039; N = 5–9; n = 10–11 for each group; ordinary 2-way ANOVA; male naive vs chronic morphine, p = 0.1018; female naive vs chronic morphine, p = 0.197; Šidák’s multiple-comparisons test). K, Summary of normalized oEPSC inhibition following perfusion of morphine in drug-naive and chronically treated WT and 10 S/T-A male mice (main effect of strain, F(1, 40) = 0.04962; p = 0.8249; main effect of treatment, F(1, 40) = 0.8821; p = 0.3533; strain × treatment interaction, F(1, 40) = 15.88; p = 0.0003; ordinary 2-way ANOVA; WT vs 10 S/T-A naive, p = 0.0117; WT vs 10 S/T-A chronic morphine, p = 0.0190, Šidák’s multiple-comparisons test). L, Schematic of MOR C-terminal phosphorylation site mutations in 10 S/T-A mice. Lines and error bars represent mean ± SEM. *p < 0.05.
Perfusion of ME (3 µM) reduced oEPSC amplitude to 30.82 ± 6.03% of baseline in slices from 10 S/T-A mice, similar to the 31.61 ± 5.38% of baseline observed in WT mice (Fig. 4B). This result suggests 10 S/T-A and WT mice have similar sensitivity to a full agonist at a saturating concentration in MThal–DMS terminals and is not surprising because receptor reserve has been previously demonstrated at presynaptic MORs (Fyfe et al., 2010; Jullié et al., 2022). Thus, we would not expect an increase in the number of functional membrane receptors to translate to an increase in maximum inhibition. In contrast to the effect of ME, morphine (3 µM) inhibited oEPSC amplitude significantly more in 10 S/T-A mice than WT mice using an ordinary two-way ANOVA (Fig. 4C). We observed a main effect of strain, with morphine perfusion causing greater oEPSC inhibition in 10 S/TA mice compared with WT, but no main effect of sex or strain × sex interaction (WT male, 74.70 ± 4.06% of baseline; 10 S/T-A male, 57.79 ± 4.49% of baseline; WT female, 70.04 ± 2.85% of baseline; 10 S/T-A female, 59.74 ± 6.35% of baseline; main effect of strain, p = 0.0048; main effect of sex, p = 0.766; strain × sex interaction, p = 0.4717; ordinary two-way ANOVA). These results indicate that morphine is more efficacious at inhibiting oEPSCs in 10 S/T-A mice than WT mice, consistent with enhanced analgesic potency observed at the behavioral level (Kliewer et al., 2019). Post hoc analysis revealed a significant difference between WT and 10 S/T-A mice in male, but not female mice, suggesting the overall effect is primarily driven by males (WT vs 10 S/T-A male, p = 0.0244; WT vs 10 S/T-A female, p = 0.2210; Šidák’s multiple-comparisons test).
We next determined whether the morphine facilitation observed at MThal–DMS synapses in WT mice was dependent on MOR phosphorylation by comparing morphine-mediated inhibition of MThal–DMS oEPSCs in drug-naive and morphine-treated 10 S/T-A mice. Although facilitation was observed only in WT males, mice of both sexes were used to determine whether any adaptations occur in females that may be unmasked in the phosphorylation-deficient MOR mice. While chronic morphine treatment enhanced inhibition of oEPSC amplitude by subsequent morphine in WT mice, we observed decreased morphine inhibition of oEPSC amplitude in slices from chronically morphine-treated 10 S/T-A mice, indicated by a main effect of treatment (Fig. 4D–J; naive male, 57.79 ± 4.49% of baseline; chronic morphine male, 70.03 ± 3.70% of baseline; naive female, 59.74 ± 6.35% of baseline; chronic morphine female, 69.79 ± 2.46% of baseline; main effect of treatment, p = 0.0149; ordinary two-way ANOVA). We did not observe a main effect of sex or a treatment × sex interaction. Post hoc analysis did not reveal statistical significance in either sex separately (main effect of sex, p = 0.8467; treatment × sex interaction, p = 0.8039; ordinary two-way ANOVA; male naive vs chronic morphine, p = 0.1018; female naive vs chronic morphine, p = 0.2197; Šidák’s multiple-comparisons test). These findings indicate a small but significant tolerance effect, but the experiments were insufficiently powered to determine whether this tolerance was driven preferentially in one sex. When comparing WT and 10 S/T-A mice across treatment conditions in males only, we observed a treatment × strain interaction with naive WT males being less sensitive to morphine than naive 10 S/T-A males and chronically treated WT males being more sensitive to morphine than chronically treated 10 S/T-A males (Fig. 4K; WT naive, 74.70 ± 4.06% of baseline; 10 S/T-A naive, 57.79 ± 4.49% of baseline; WT chronic morphine, 54.76 ± 4.11% of baseline; 10 S/T-A chronic morphine, 70.03 ± 3.70% of baseline; treatment × sex interaction, p = 0.0003; ordinary two-way ANOVA; WT vs 10 S/T-A naive, p = 0.0117; WT vs 10 S/T-A chronic morphine, p = 0.0190; Šidák’s multiple-comparisons test). Together, the results of these experiments suggest that MOR C-terminal phosphorylation-deficient mice were initially more sensitive to morphine inhibition at MThal–DMS terminals but developed tolerance, rather than facilitation, to subsequent morphine signaling following chronic morphine exposure. These changes resulted in a reversal in sensitivity to morphine inhibition following chronic morphine treatment (WT were more sensitive than 10 S/T-A) and suggest that loss of MOR phosphorylation sites did not simply mimic or occlude the effect of morphine treatment seen in WT male mice but may have unmasked counteradaptations leading to the development of tolerance.
Discussion
The present study provided a direct comparison of how chronic morphine treatment differentially alters morphine signaling at presynaptic and somatic subcellular compartments within the same neuronal population in a sex-specific manner. Seven days of continuous morphine exposure facilitated morphine responses at MThal–DMS terminals in male, but not female mice, but induced tolerance at MThal cell bodies in both sexes. In MOR phosphorylation-deficient mice, chronic morphine treatment induced tolerance, rather than facilitation, at MThal–DMS presynaptic terminals, indicating that receptor phosphorylation may regulate processes that drive facilitation. One caveat of our study is that because brain slices were maintained in the absence of morphine, slices from chronically treated mice were in an acutely withdrawn state during baseline recordings. Chronic morphine induces both desensitization and tolerance in locus coeruleus neurons. Desensitization recovers in <1 h, but tolerance to morphine persists beyond 6 h (Levitt and Williams, 2012). Thus, conditions in our study represent an acutely withdrawn “tolerant,” but not desensitized, state. In this withdrawn state, opioid signaling in presynaptic periaqueductal gray terminals has been shown to utilize additional adenylyl cyclase-dependent (AC) signaling pathways (Ingram et al., 1998). In this context, the loss of facilitation in phosphorylation-deficient MOR mice may suggest that receptor phosphorylation could regulate the ability of MOR to either induce adaptive responses or signal through this alternate withdrawn signaling pathway. However, the effects of morphine and the role of receptor phosphorylation in regulation of morphine effects may be different in opioid-maintained conditions.
Facilitation of morphine signaling at MThal–DMS presynaptic terminals
Previous studies have observed facilitation of opioid signaling at GABAergic terminals in various brain regions (Chieng and Williams, 1998; Ingram et al., 1998; Hack et al., 2003; Pennock et al., 2012), while others have observed tolerance (Fyfe et al., 2010; Matsui et al., 2014). At excitatory synapses in the striatum, a single exposure to oxycodone blocked the subsequent induction of long-term depression by MOR agonists, possible evidence of opioid tolerance (Atwood et al., 2014). Studies which have described presynaptic facilitation have primarily attributed the effect to a compensatory upregulation of AC that drives hyperexcitability of the terminals (Sharma et al., 1975). We hypothesized that because AC upregulation is proposed to be blunted by MOR internalization (Finn and Whistler, 2001), phosphorylation-deficient/low-internalizing 10 S/T-A mice might display increased morphine facilitation. Instead, chronic morphine treatment eliminated facilitation and induced tolerance in 10 S/T-A mice, suggesting that either AC upregulation may not mediate facilitation observed in our study, that receptor phosphorylation and receptor trafficking/ localization may be important for AC upregulation (Zhao et al., 2006), or that alternative mechanisms may be involved. Other possible mechanisms of presynaptic facilitation include increases in functional receptor number, increases in receptor–effector coupling efficiency or circuit-level changes in the strength of innervation of MOR-expressing thalamic inputs to DMS. It is also possible that, rather than blocking the process of facilitation, phosphorylation deficiency could mimic and, therefore, occlude morphine-induced facilitation. Alternatively, phosphorylation deficiency may unmask other forms of morphine-induced tolerance that act in opposition to the morphine-induced facilitation. These competing mechanisms can be investigated in future studies.
From the data we cannot conclude that morphine facilitation at MThal–DMS terminals is driven by presynaptic, rather than postsynaptic, adaptations given that opioids have been shown to induce synaptic plasticity (Gerdeman et al., 2003). We have recently shown that morphine acting at postsynaptic sites can negatively modulate tonic adenosine signaling at glutamatergic presynaptic terminals in the DMS, suggesting that presynaptic effects on glutamate release can be influenced by postsynaptic adaptations (Adhikary et al., 2022). Precise differences in local circuitry such as these could provide insight as to why facilitation occurs at thalamic presynaptic terminals in DMS, while presynaptic tolerance has been shown within other circuits.
Sex differences in the development of morphine facilitation
Sex differences in the development of analgesic tolerance are well known, with numerous studies showing greater tolerance in males than females (Bodnar and Kest, 2010), but the underlying physiological mechanisms are not yet clear. Female and castrated male rats developed tolerance more slowly than testosterone-pretreated females or intact males, suggesting that testosterone may influence the development of tolerance (South et al., 2001). Morphine tolerance has also been shown to develop in male and proestrus female rats, but not ovariectomized females or females in other estrous phases (Shekunova and Bespalov, 2004, 2006). Repeated morphine administration has been associated with a decrease in the number of periacqueductal gray to rostroventral medulla-projecting (PAG-RVM) output neurons activated by morphine in male but not female rats, providing a neural correlate with sex differences in opioid tolerance (Loyd et al., 2008).
Many studies investigating chronic opioid effects on somatic or presynaptic effects of opioids either did not report sex differences or conducted experiments only in males. To our knowledge, this study is the first to describe sex-specific facilitation of opioid effects. However, from our data we cannot determine a mechanism driving the observed sex differences or whether facilitation at MThal–DMS terminals is not present in females at all, or if these effects are masked by additional counteradaptations that are not present in males.
Tolerance to morphine signaling at medial thalamic cell bodies
Cellular tolerance at somatic MORs induced by chronic morphine treatment has been observed in many brain regions including the locus coeruleus, Kölliker–Fuse neurons and PAG (Christie et al., 1987; Bagley et al., 2005; Levitt and Williams, 2018). In agreement with these studies, we observed cellular tolerance at MThal cell bodies following continuous morphine exposure in both males and females. To our knowledge, decreased coupling of opioids to GIRK conductance in thalamic regions following chronic opioid exposure has not been previously demonstrated. Recently, GIRK currents induced by DAMGO in paraventricular thalamic (PVT) neurons were not found to be different between saline- and morphine-treated mice; however, tolerance developed to the DAMGO-mediated inhibition of PVT neuron firing (Hou et al., 2023). This lack of tolerance may be due to DAMGO being a full agonist, making detection of tolerance difficult, or the high variability of GIRK currents seen in both their study and ours. While the tolerance in our study was statistically significant, the magnitude of the effect was small. This may be due, in part, to morphine-mediated GIRK current that was small under physiological conditions, prompting the use of ML-297 to enhance the amplitude of the currents for quantification. This indicates that, at the soma, MOR expression may be low or GIRK coupling weak, making a decrease in receptor–effector coupling difficult to capture. Furthermore, we observed a high degree of variability in the amplitude of morphine-activated GIRK currents, and several cells were excluded from analysis due to a lack of response to morphine. These results suggest that there may be heterogeneity in MOR expression patterns across MThal (and possibly PVT) neurons.
The role of MOR C-terminal phosphorylation in driving presynaptic opioid adaptations
Phosphorylation of serine and threonine residues in the C-terminal tail of MOR promotes arrestin recruitment to MORs, uncoupling of associated G proteins, and receptor internalization. Deletion of phosphorylation sites in the C-terminal tail of MOR has been shown to reduce both acute desensitization and tolerance in the somatic compartment (Williams et al., 2013; Arttamangkul et al., 2018; Arttamangkul et al., 2019). Given the small effect of chronic morphine treatment on morphine signaling at MThal cell bodies and the large variability of morphine-mediated responses, we did not investigate chronic morphine effects at MThal cell bodies in 10 S/T-A mice. However, based on the findings of previous studies, we would not expect tolerance to occur in phosphorylation-deficient mice.
To our knowledge, the role of MOR phosphorylation in mediating chronic opioid effects at presynaptic receptors has not previously been examined in brain slices. However, based on current knowledge of MOR regulation by phosphorylation, it is not clear why loss of MOR phosphorylation produced tolerance, whereas facilitation was observed in WT mice at MThal–DMS presynaptic terminals. In 10 S/T-A mice, behavioral tolerance to morphine is reduced, but not eliminated, while tolerance to fentanyl is eliminated (Kliewer et al., 2019). Receptor phosphorylation, desensitization, and tolerance are agonist-dependent, and unlike other agonists, desensitization by morphine is modest and dependent on phosphorylation by protein kinase C (Dang and Christie, 2012). A recent study demonstrated that agonist-induced MOR phosphorylation can vary by brain region (Fritzwanker et al., 2023); methadone elicited very little phosphorylation at residues S375, T376, and T379 in the striatum but robust phosphorylation in the medial habenula and spinal cord. It is possible that the facilitation observed in our study is driven by a unique phosphorylation pattern that is dependent on agonist and brain region, and loss of phosphorylation in 10 S/T-A mice unmasks underlying tolerance. Several kinases, including GRK2/3, protein kinase C, and c-Jun N-terminal kinase, are involved in phosphorylation of the MOR C-terminal tail (Williams et al., 2001). It is not known which of the 10 mutated phosphorylation sites or kinases mediate the chronic opioid effects we observed. MOR phosphorylation patterns within the axonal compartment and variability between sexes are unknown making a determination of which phosphorylation sites are affected by morphine treatment difficult to predict.
This study highlights that the effects of chronic opioid treatment on MOR signaling are not ubiquitous but may instead be specific to agonist, brain region, and subcellular location of the receptor. A better understanding of the precise mechanisms by which these adaptations occur is critical to the development of safer, more effective therapeutics for chronic and severe pain for both females and males.
Footnotes
This work was supported by National Institute on Drug Abuse Grant Numbers R01DA042779 (to W.T.B.) and T32DA007281 (to E.R.J.) and Benedict and Diana Lucchesi Fellowship (to E.R.J.). We thank Dr. Erwin Arias-Hervert for technical assistance and comments on this manuscript, Dr. Erica Levitt for comments on this manuscript, and Dr. Alberto Perez-Medina for technical assistance.
The authors declare no competing financial interests.
- Correspondence should be addressed to William T. Birdsong at wtbird{at}umich.edu.