Abstract
Opioids have decreased analgesic potency (but not efficacy) in aged rodents compared with adults; however, the neural mechanisms underlying this attenuated response are not yet known. The present study investigated the impact of advanced age and biological sex on opioid signaling in the ventrolateral periaqueductal gray (vlPAG) in the presence of chronic inflammatory pain. Assays measuring µ-opioid receptor (MOR) radioligand binding, GTPγS binding, receptor phosphorylation, cAMP inhibition, and regulator of G-protein signaling (RGS) protein expression were performed on vlPAG tissue from adult (2–3 months) and aged (16–18 months) male and female rats. Persistent inflammatory pain was induced by intraplantar injection of complete Freund's adjuvant (CFA). Adult males exhibited the highest MOR binding potential (BP) and highest G-protein activation (activation efficiency ratio) in comparison to aged males and females (adult and aged). No impact of advanced age or sex on MOR phosphorylation state was observed. DAMGO-induced cAMP inhibition was highest in the vlPAG of adult males compared with aged males and females (adult and aged). vlPAG levels of RGS4 and RGS9-2, critical for terminating G-protein signaling, were assessed using RNAscope. Adult rats (both males and females) exhibited lower levels of vlPAG RGS4 and RGS9-2 mRNA expression compared with aged males and females. The observed age-related reductions in vlPAG MOR BP, G-protein activation efficiency, and cAMP inhibition, along with the observed age-related increases in RGS4 and RGS9-2 vlPAG expression, provide potential mechanisms whereby the potency of opioids is decreased in the aged population.
SIGNIFICANCE STATEMENT Opioids have decreased analgesic potency (but not efficacy) in aged rodents compared with adults; however, the neural mechanisms underlying this attenuated response are not yet known. In the present study, we observed age-related reductions in ventrolateral periaqueductal gray (vlPAG) µ-opioid receptor (MOR) binding potential (BP), G-protein activation efficiency, and cAMP inhibition, along with the observed age-related increases in regulator of G-protein signaling (RGS)4 and RGS9-2 vlPAG expression, providing potential mechanisms whereby the potency of opioids is decreased in the aged population. These coordinated decreases in opioid receptor signaling may explain the previously reported reduced potency of opioids to produce pain relief in females and aged rats.
Introduction
Clinical studies examining pain management in the elderly are challenging; comorbid conditions such as diabetes and high blood pressure and participant use of concomitant medications affect patient outcomes, contributing to difficulties in the interpretation of results (Naples et al., 2016; Prostran et al., 2016). Additionally, there exist age-related and sex-related individual differences in the likelihood of reporting pain in a clinical setting which may lead to misrepresentations of analgesic efficacy (Reddy et al., 2012; Dampier et al., 2013). Particularly, aged populations are known to underreport pain because of fears of institutionalization and concerns with addiction and overdose (Ferrell et al., 1990; Hofland, 1992).
Using a preclinical model of persistent inflammatory pain [intraplantar complete Freund's adjuvant (CFA)], we have previously reported a significant impact of age and sex on morphine potency. Specifically, we showed that aged rats (18 months) exhibit decreased morphine potency compared with adults (2 months), with aged males requiring greater than 2× the concentration of morphine than their adult counterparts to produce equivalent analgesia (Fullerton et al., 2021). Similar results have been reported in prior preclinical studies, suggesting that aged rodents require higher doses of opioids to produce antinociception (Kavaliers et al., 1983; Kramer and Bodnar, 1986; Webster et al., 1976). The mechanisms underlying the attenuation of opioid potency in the aged population are currently unknown.
Morphine-induced analgesia is mediated primarily via binding to µ-opioid receptors (MORs), seven-transmembrane domain G-protein-coupled receptors (GPCRs) located predominantly on neuronal cell membranes (Martin, 1963; Wolozin and Pasternak, 1981; Goodman and Pasternak, 1985; Serohijos et al., 2011). Following agonist binding, coupled G-proteins undergo a conformational change in which the guanosine diphosphate (GDP) bound to the inactive α subunit is replaced by guanosine triphosphate (GTP), activating the subunit and promoting opioid signaling through interaction with downstream effectors (Senese et al., 2020). MORs are coupled to a family of G-proteins called Gi/o that signal via inhibition of adenylyl cyclase and subsequently decrease cAMP (Koehl et al., 2018; Bouchet et al., 2021). This G-protein signaling is downregulated or terminated by regulator of G-protein signaling (RGS) proteins, which act as GTPase activating proteins (GAPS), hydrolyzing the active GTP back into GDP and terminating downstream signaling (Gerber et al., 2016). RGS proteins, particularly RGS4 and RGS9-2, have been previously implicated in opioid-mediated G-protein signaling, and have been shown to attenuate analgesia (Garnier et al., 2003; Psifogeorgou et al., 2007; Avrampou et al., 2019; Senese et al., 2020). A primary target of MOR downstream signaling is the effector adenylate cyclase (AC). AC enzymatically converts adenosine triphosphate (ATP) to the excitatory second messenger cAMP. MOR binding inhibits AC, thereby reducing cAMP expression, facilitating neuronal hyperpolarization, and promoting analgesia (Santhappan et al., 2015). Like other GPCRs, MOR signaling is subject to desensitization via the cellular mechanism of receptor phosphorylation, which limits agonist binding and recruits arrestin proteins and thus attenuates opioid signaling (L. Zhang et al., 1996; Yu et al., 1997; Groer et al., 2011).
The present studies test the hypothesis that the observed age-induced reduction in morphine potency is mediated by changes in MOR signaling within the midbrain periaqueductal gray (PAG), a CNS region critical for the opioid modulation of pain. (Basbaum et al., 1976; Behbehani and Fields, 1979; Morgan et al., 1992, 2006). The ventrolateral PAG (vlPAG) contains a large population of MOR+ neurons, and direct administration of MOR agonists into the PAG produces potent analgesia (Satoh et al., 1983; Jensen and Yaksh, 1986; Bodnar et al., 1988; Loyd et al., 2008) Intra-PAG administration of MOR antagonists or lesions of PAG MOR significantly attenuate the analgesic effect of systemic morphine, suggesting a critical role for PAG MOR in mediating morphine action (Ma and Han, 1991; Y. Zhang et al., 1998; Loyd et al., 2008).
We have previously shown that aged rats exhibit reduced vlPAG MOR protein expression and reduced MOR agonist binding in the vlPAG compared with adult rats (Fullerton et al., 2021), suggesting that diminished levels of functioning MOR in the vlPAG contribute to the attenuated opioid potency seen in the aged. The present studies build on these previous findings to assess age and sex differences in MOR availability, ligand affinity, phosphorylation, G-protein activation, cAMP inhibition, and expression of RGS proteins.
Materials and Methods
Experimental subjects
Adult (2–3 months) and aged (16–18 months) male and regularly cycling female Sprague Dawley rats were used in these experiments (Charles River Laboratories). Rats were co-housed in same-sex pairs on a 12/12 h light/dark cycle (lights on at 8 A.M.). Access to food and water was ad libitum throughout the experiment, except during testing. All studies were approved by the Institutional Animal Care and Use Committee at Georgia State University and performed in compliance with Ethical Issues of the International Association for the Study of Pain and National Institutes of Health. All efforts were made to reduce the number of rats used in these experiments and to minimize pain and suffering. All assays were performed on separate cohorts of rats (n = 8/sex/age; N = 32), with the exception of the radioligand binding and GTPγS assays, which were run simultaneously on PAG tissue from a single cohort.
Vaginal cytology
Beginning 10 d before testing, vaginal lavages were performed daily on adult and aged female rats to confirm that all rats were cycling regularly and to keep daily records of the stage of estrous. Proestrus was identified as a predominance of nucleated epithelial cells, and estrus was identified as a predominance of cornified epithelial cells. Diestrus 1 was differentiated from diestrus 2 by the presence of leukocytes. Rats that appeared between phases were noted as being in the more advanced stage (Loyd et al., 2007).
CFA-induced chronic pain treatment
Seventy-two hours before experimentation, persistent inflammatory pain was induced by injection of CFA (mycobacterium tuberculosis; Sigma; 200 µl), suspended in an oil/saline (1:1) emulsion, into the plantar surface of the right hind paw. Edema was present within 24 h of injection, indicated by a >100% change in paw diameter, determined using calibrated calipers applied midpoint across the plantar surface compared with handled paw.
Membrane preparation for radioligand binding and GTPγS assays
vlPAG membrane protein lysates were prepared from adult and aged, male and female, handled and CFA-treated rats to be used for radioligand binding and GTPγS assays. Seventy-two hours post-CFA injection or handling, rats were restrained using DecapiCones and decapitated. Brains were removed rapidly, flash-frozen in 2-methyl butane on dry ice, and stored at −80°C. vlPAG tissue from caudal PAG (bregma −7.5 through −8.5) was dissected from each brain using a straight edge razor at −20°C. On the day of the assay, PAG sections were placed in ice-cold assay buffer (50 mm Tris-HCl, pH 7.4). Tissue was homogenized with a glass dounce and centrifuged at 20,000 × g at 4°C for 30 min. The supernatant was discarded and the pellet resuspended in assay buffer (Zollner et al., 2003; Shaqura et al., 2016a,b; Li et al., 2018). Membrane protein concentration was calculated using the Bradford Assay, and lysates of vlPAG membrane protein from adult and aged, male and female, naive and CFA-treated rats (n = 4; N = 32) were used immediately for radioligand binding and GTPγS assays.
Saturation radioligand binding assay
Saturation binding experiments were performed on vlPAG membranes using [3H]DAMGO (specific activity 50 Ci/mmol, American Radiolabeled Chemicals). Briefly, 100 µg of membrane protein was incubated with various concentrations of [3H]DAMGO (0.5–5 nm), in a total volume of 1 ml of binding buffer (50 mm Tris-HCl, pH 7.4). Nonspecific binding was defined as radioactivity remaining bound in the presence of 10 µm unlabeled naloxone. At the end of the incubation period [1 h at room temperature (RT)] bound and free ligands were separated by rapid filtration over Whatman brand Grade GF/C glass filters (Sigma-Aldrich) using a sampling vacuum manifold (MilliporeSigma). Filters were washed four times with 5 ml of cold dH2O. Bound radioactivity was determined by liquid scintillation spectrophotometry after overnight extraction of the filters in 3 ml of scintillation fluid. All experiments were performed in triplicate. Bmax and Kd values were determined by nonlinear regression analysis of concentration-effect curve in GraphPad Prism 9.1.
Agonist-stimulated [35S]GTPγS binding
DAMGO-stimulated [35S]GTPγS binding to PAG membrane protein was assessed by incubating membrane protein (100ug) in the presence or absence of [35S]GTPγS (0.1 nm; specific activity 1250 Ci/mmol, American Radiolabeled Chemicals) and various concentrations of DAMGO (2–30,000 nm) in assay buffer (20 mm Tris, 10 mm MgCl2, 100 mm NaCl, and 0.2 mm EGTA, pH 7.4) for 30 min at 30°C. Stimulated [35S]GTPγS binding was compared with unstimulated binding at each measurement point and presented as percent basal binding. At the end of the incubation period, bound and free ligands were separated by rapid filtration over Whatman brand Grade GF/B glass filters (Sigma-Aldrich) using a sampling vacuum manifold (MilliporeSigma). Filters were washed four times with 5 ml of cold buffer (50 mm Tris-HCl, pH 7.4). Bound radioactivity was determined by liquid scintillation spectrophotometry after overnight extraction of the filters in 3 ml of scintillation fluid. All experiments were performed in duplicate. Efficacy (Emax) is defined as the maximum percent stimulation by an agonist; potency (EC50) is defined as the concentration of DAMGO required for half the maximal response. Emax and EC50 values were determined by nonlinear regression analysis of concentration-effect curves using GraphPad Prism 9.1.
Phosphorylated MOR analysis
To assess the impact of advanced age and sex on MOR phosphorylation state, levels of MOR phosphorylation were analyzed by Western blotting. The lysis buffers used and general methods are the same as reported by Lei et al. (2017). Briefly, rat PAG samples were homogenized in ∼500 µl of lysis buffer using a glass dounce and rotated overnight at 4°C. The samples were then spun down at 13,000 × g for 10 min at 4°C. The resulting lysates were quantified by DC Protein Assay kit (Bio-Rad, #500-0111) according to the manufacturer's instructions. A total of 50–75 µg of soluble protein per sample (same for all samples in a set) was run on 10% Bis-Tris Bolt PAGE gels (Fisher Scientific, #NW00100BOX) and wet-transferred to nitrocellulose membrane (Protran 0.2 µm NC, #45-004-001 from Fisher Scientific) at 30V at 4°C for 2–3 h. Membranes were blocked using 5% nonfat dry milk in TBS, then blotted for primary antibody target in 5% BSA in TBST overnight rocking at 4°C. Primary antibodies used were: pMOR (1:1000, Bioss #bs-3724R), tMOR (1 µg/ml, R&D #MAB6866), and GAPDH (1:1000, Invitrogen #MA5-15738). All three targets were always probed for on the same blot, using low pH stripping buffer between each set. Secondary antibodies were used at 1:5000 each in 5% nonfat dry milk in TBST, rocking for ∼1 h at RT. Secondary antibodies used were: Goat-α-Mouse-IRDye800CW (Fisher Scientific #NC9401841) and Goat-α-Mouse-IRDye680LT (Fisher Scientific #NC0046410). Target signal was acquired using an Azure Sapphire imager using the near-infrared channels (658 and 784 nm). Band density was quantified, and background subtracted using the onboard AzureSpot analysis software. Samples were run with at least two representatives of each experimental group on the same gel. Adult and aged, male and female, naïve, and CFA-treated rats (n = 4; N = 32) were used for these experiments. pMOR signal was normalized to the same sample tMOR or GAPDH as indicated, and all samples were normalized to the Adult Male average on each gel before combining data from different gels.
DAMGO-induced cAMP inhibition
DAMGO-induced cAMP inhibition was analyzed using a cAMP assay. vlPAG membrane protein lysates were prepared from adult and aged, male and female, handled and CFA-treated rats. Seventy-two hours post-CFA injection or handling, rats were restrained using DecapiCones and decapitated. Brains were removed rapidly, flash-frozen in 2-methyl butane on dry ice, and stored at −80°C. vlPAG tissue from caudal PAG (bregma −7.5 through −8.5) was dissected from each brain using a straight edge razor at −20°C. Briefly, vlPAG tissues were homogenized in ice-cold lysis buffer containing 0.25 m sucrose, 50 mm Tris–HCl, pH 7.5, 5 mm EGTA, 5 mm EDTA, 1 mm phenylmethylsulfonyl fluoride, 0.1 mm dithiothreitol, and 10 µg/ml leupeptin. The homogenized tissues were then centrifuged at 1000 × g for 5 min (4°C), and the supernatant was centrifuged at 35,000 × g for 10 min (Viganò et al., 2003). vlPAG membrane protein samples (100ug) from adult and aged, male and female, naive and CFA-treated rats (n = 4; N = 32) were incubated in 1 µm forskolin (FSK) and in the presence or absence of 10 µm DAMGO to stimulate AC activity. cAMP levels in vlPAG tissue were determined using the LANCE Ultra cAMP Detection kit (PerkinElmer), a time-resolved fluorescence resonance energy transfer (TR-FRET) cAMP immunoassay. Data are plotted as % change in TR-FRET signal comparing FSK baseline measurements to measurements following DAMGO-stimulated cAMP inhibition.
Single-molecule fluorescence in situ hybridization (smFISH)
smFISH (RNAscope) assays were used to determine mRNA expression of OPRM1, RGS9-2, and RGS4 in the vlPAG. Rats were restrained using DecapiCones and decapitated. Brains were removed rapidly, flash frozen in 2-methyl butane on dry ice, and stored RNase free at −80°C. Frozen tissue was sectioned in a 1:6 series of 20-µm coronal sections at –20°C with a Leica CM3050S cryostat. Sections were immediately mounted onto Superfrost slides (20°C) and stored at −80°C until the time of the assay. vlPAG sections from adult and aged, male and female, naive and morphine treated rats were used (n = 4–7; N = 45). For morphine dosing paradigm, see (Fullerton et al., 2021). Tissue was processed for smFISH according to the RNAscope Multiplex kit protocol (Advanced Cell Diagnostics) using probes for OPRM1, RGS4, and RGS9-2. To facilitate cellular mRNA quantification within the vlPAG, sections were counterstained with DAPI. mRNA puncta were visualized as fluorescent signals. Fluorescent images were captured on Zeiss LSM 700 Confocal Microscope at 40×, and mRNA expression (target puncta/DAPI) was calculated using Imaris software. To determine RGS4 and RGS9-2 expression levels in MOR+ neurons, quantification was restricted to puncta located within 10 µm of DAPI that co-expressed OPRM1 mRNA. mRNA expression values were determined for the left and right ventrolateral subdivisions of each PAG image from two representative levels of the mid-caudal PAG (bregma −7.74 and −8.00). As there was no significant effect of rostrocaudal level in the analyses, data were collapsed and presented as vlPAG and were averaged for each rat. RGS4 and RGS9-2 mRNA values are expressed as the mean ± SEM. All images were collected and analyzed by an experimenter blinded to the experimental condition.
Statistical analysis and data presentation
All values are reported as mean ± SEM. Data were assessed for normality and homogeneity of variance using Shapiro–Wilk and Bartlett's tests. Significant main effects of sex, age, and treatment were assessed using ANOVA; p < 0.05 was considered statistically significant. Tukey's post hoc tests were conducted to determine significant mean differences between groups that were a priori specified. Data are expressed either as fmol/protein or nm DAMGO for radioligand binding assay or Emax or EC50 for GTPγS assay.
Results
Advanced age and sex impact vlPAG MOR binding properties
To assess the impact of advanced age, biological sex, and pain on vlPAG MOR signaling, we first used radioligand saturation binding assays to determine MOR binding parameters. The saturation binding curves generated with [3H]DAMGO are shown in Figure 1A. CFA treatment did not significantly impact Kd (F(1,22) = 3.429, p = 0.078) or Bmax (F(1,24) = 0.0023, p = 0.963) so CFA and handled groups were combined.
Saturation binding curve of bound [3H]DAMGO (A). No significant effect of age or sex was observed in MOR affinity (Kd; B). MOR availability (Bmax) was significantly lower in aged males compared with adult males (C). Adult females, aged males, and aged females exhibited reduced MOR BP (Bmax/Kd) compared with adult males, indicating attenuated agonist interaction at the receptor level (D). CFA-treated rats are represented as circles, handled rats are represented as triangles. ns, not significant. *Significant difference between adult males vs aged males, or adult males vs adult females, aged males, and aged females; p < 0.05 calculated by Tukey's post hoc test. Graphs indicate mean ± SEM. Values indicate % change from adult male.
Kd values, indicative of receptor affinity, were determined using the concentration-effect curves generated from each sample. Analysis of Kd values indicated no significant impact of age (F(1,26) = 0.770, p = 0.388) or sex (F(1,26) = 1.241, p = 0.275), or a significant interaction (F(1,26) = 1.278, p = 0.269; Fig. 1B).
Analyses of Bmax values, indicative of receptor availability, showed no significant impact of age (F(1,28) = 2.661, p = 0.114) or sex (F(1,28) = 0.002, p = 0.967), but a significant interaction (F(1,28) = 4.995, p = 0.034). Post hoc analysis showed a significant difference in vlPAG MOR availability between adult males and aged males, as evidenced by reduced Bmax values in the aged males compared with their adult counterparts (p = 0.025; Fig. 1C). Bmax values for aged males were markedly reduced compared with adult males, while adult females and aged females exhibited smaller reductions in Bmax compared with their adult male counterparts (%Δ −23.6 and −17.1, respectively; Fig. 1C).
Analyses of binding potential (BP) values, a measure that takes into consideration the density of available receptors and the affinity of the receptor for its agonist, showed a significant impact of sex, with males exhibiting greater BP than females (F(1,28) = 4.631, p = 0.040) and a significant interaction between age and sex (F(1,28) = 9.110, p = 0.005). No significant impact of age (F(1,28) = 3.312, p = 0.079) was observed. Post hoc analyses showed a significant difference in vlPAG MOR BP between adult males and adult females (p = 0.006) and adult males and aged males (p = 0.010), indicating that adult males exhibit greater vlPAG MOR BP than their aged male counterparts and their adult female counterparts (Fig. 1D). Adult females, aged males, and aged females all exhibited marked reductions in MOR BP compared with their adult male counterparts (%Δ −52, −48.7, and −40, respectively; Fig. 1D).
Advanced age and sex impact opioid-induced G-protein activation in the vlPAG
We next conducted GTPγS binding assays to determine whether advanced age or biological sex impacted MOR mediated G-protein activation. Our initial studies revealed a significant main effect of chronic pain (F(1, 24) = 29.04, p < 0.0001; Fig. 2E). To improve the translatability of these results to the target population of aged patients suffering from chronic pain, all other data displayed from GTPγS experiments are exclusively from CFA-treated rats. Concentration curves generated from [35S]GTPγS assays are shown in Figure 2A. Analyses of Emax values, a measure of G-protein availability combined with ligand efficacy, indicated no significant impact of age (F(1,24) = 0.035, p = 0.8532) or sex (F(1,25) = 2.852, p = 0.1042), or a significant interaction (F(1,24) = 2.023, p = 0.1678; Fig. 2B).
[35S]GTPγS binding curve of agonist-stimulated GTPγS binding (A). No significant effect of age or sex was observed in Emax (B). Aged males, adult females, and aged females exhibited increased EC50 values compared with adult males, indicating that adult males have the greatest potency (C). Aged males, adult females, and aged females exhibited reduced G-protein activation efficiency (Emax/EC50) compared with adult males, indicating attenuated G-protein signaling (D). Initial studies using both handled and CFA-treated rats indicated a significant main effect of chronic pain on EC50 values (E). CFA-treated rats are represented as circles, handled rats are represented as triangles. ns, not significant. *Significant difference between males vs females, or adult males vs adult females, aged males, and aged females, or handled vs CFA; p < 0.05 calculated by Tukey's post hoc test. Graphs indicate mean ± SEM. Values indicate % change from adult male.
Analyses of EC50 values, a measure of the concentration of DAMGO required for half-maximal G-protein binding, indicated a significant impact of sex, with males exhibiting lower effective concentration values than females, reflecting a higher potency of activation (F(1,24) = 5.177, p = 0.0321). There was no significant impact of age (F(1,24) = 2.339, p = 0.1393), and no significant interaction (F(1,24) = 2.473, p = 0.1289; Fig. 2C). Although not statistically significant, aged males, adult females, and aged females all exhibited marked increases in G-protein EC50 compared with their adult male counterparts, as evidenced by their percent change values (%Δ 60.5, 114.8, and 98.6, respectively), reflecting a lower potency of activation (Fig. 2C). Together, these results suggest that opioid-induced G-protein signaling is impaired in the PAG of the aged rat and the female rat.
Next, we calculated the coefficient ratio of analysis of Emax/EC50, which is indicative of overall G-protein activation efficiency. A significant impact of age was noted, with adults exhibiting higher vlPAG G-protein activation efficiency than aged rats (F(1,24) = 6.491, p = 0.0177). There was also a significant impact of sex, with males exhibiting higher vlPAG G-protein activation efficiency compared with females (F(1,24) = 10.39, p = 0.0036). A significant interaction was also observed (F(1,24) = 13.93, p = 0.0010). Post hoc analyses showed a significant difference in vlPAG G-protein activation efficiency between adult males and aged males (p = 0.0009), with adult males exhibiting greater vlPAG G-protein activation efficiency than aged males. A significant difference in vlPAG G-protein activation efficiency between adult males and adult females (p = 0.0003) was also observed, with adult males exhibiting greater vlPAG G-protein activation efficiency than adult females (Fig. 2D). Aged males, adult females, and aged females all exhibited marked reductions in G-protein activation efficiency compared with their adult male counterparts (%Δ −40.0, −52.1, and −41.9, respectively; Fig. 2D).
Advanced age and sex do not impact phosphorylated MOR in the vlPAG
The results above showed significant reductions in MOR BP and G-protein activation efficacy. Therefore, we next examined whether there was an effect of age and sex on MOR phosphorylation, as increased levels of pMOR would likely contribute to these reductions. Western blottings were used to determine the impact of advanced age on MOR phosphorylation at serine-375, a known site of phosphorylation-mediated desensitization (Schulz et al., 2004). There was no significant impact of treatment on our initial analyses (F(1,16) = 0.0033, p = 0.9547), so CFA and handled groups were combined. Although increased pMOR/tMOR was observed in aged males compared with their adult counterparts (%Δ 34.0), no significant main effect of age (F(1,41) = 2.168, p = 0.1486), sex (F(1,41) = 1.807, p = 0.1863), or age × sex interaction (F(1,41) = 0.4615, p = 0.5008) was observed (Fig. 3).
No significant impact of advanced age or biological sex on the ratio of phosphorylated MOR/total MOR (A). Representative phosphoMOR blots are shown below the graph (B). CFA-treated rats are represented as circles, handled rats are represented as triangles. ns, not significant. Graph indicates mean ± SEM. Values indicate % change from adult male.
Advanced age and sex impact opioid-induced cAMP inhibition in the vlPAG
We next determined whether advanced age, biological sex, or persistent pain impacted opioid-induced cAMP inhibition in the vlPAG. FSK-induced cAMP release was used as a baseline measurement for each group, while FSK + DAMGO was used to assess the degree to which cAMP was inhibited by DAMGO. Percent change from baseline was used to compare across treatment groups (Fig. 4A). There was no significant impact of persistent pain on percent change from baseline (F(1,26) = 0.7619, p = 0.3907), so CFA and handled groups were combined. Our analyses revealed a significant main effect of age (F(1,30) = 9.314, p = 0.0047), a significant main effect of sex (F(1,30) = 10.31, p = 0.0031), and a significant interaction (F(1,30) = 14.24, p = 0.0007). Post hoc analyses showed a significant difference in cAMP inhibition between adult males and adult females (p = 0.0002), adult males and aged males (p = 0.002), and adult males and aged females (p = 0.004), indicating that DAMGO elicits greater cAMP inhibition in adult males compared with aged males and females (both adult and aged; Fig. 4B).
TR-FRET immunoassay showing % change from FSK-stimulated baseline (1 μm FSK) to DAMGO-induced cAMP inhibition (10 μm DAMGO). cAMP levels are inversely proportional to TR-FRET signal. Adult males exhibited the highest level of agonist-dependent % change in cAMP (A). Individual values in B indicate each rat's % change from FSK baseline to DAMGO-induced cAMP inhibition. 2 × 2 ANOVA was used to compare across treatment groups. Aged males and females exhibited reduced agonist-dependent cAMP inhibition compared with adult males, indicating attenuated downstream opioid signaling (B). CFA-treated rats are represented as circles, handled rats are represented as triangles. *Significant difference between adult males vs adult females, aged males, and aged females; p < 0.05 calculated by Tukey's post hoc test. Graphs indicate mean ± SEM. Values indicate % change from FSK baseline.
Advanced age and sex impact RGS4 and RGS9-2 expression in the vlPAG
RGS proteins act as GAP accelerators to negatively modulate G-protein signaling. RGS protein family members RGS4 and RGS9-2 are expressed in the vlPAG and have both been shown to regulate opioid signaling by reversing G-protein activation. Therefore, we next used smFISH to determine whether RGS4 and RGS9-2 expression in the vlPAG was altered by advanced age and biological sex. In these studies, following CFA administration, a cohort of rats was administered morphine to examine the relationship between morphine EC50 and RGS levels. No significant impact of morphine treatment on RGS4 (F(1,22) = 0.8521, p = 0.3664), RGS9-2 (F(1,22) = 0.0258, p = 0.8739), or OPRM1 (F(1,22) = 0.3146, p = 0.5805) was observed in our initial analyses, so naive and morphine treated groups were combined. We first assessed total vlPAG expression of RGS4 and RGS9-2. These analyses revealed a significant main effect of age on both RGS4 (F(1,37) = 18.15, p = 0.0001) and RGS9-2 (F(1,37) = 17.09, p = 0.0002), with aged rats exhibiting increased levels of RGS4 and RGS9-2 mRNA compared with their adult counterparts (Fig. 5A,B). No significant main effect of sex on RGS4 (F(1,37) = 0.0247, p = 0.8763) or RGS9-2 (F(1,37) = 0.4855, p = 0.4903), or significant interactions between age and sex for RGS4 (F(1,37) = 0.4994, p = 0.4842) or RGS9-2 (F(1,41) = 0.3246, p = 0.5723) were observed (Fig. 5A,B).
mRNA expression of RGS4 (A) and RGS9-2 (B) in the vlPAG was significantly greater in aged rats compared with adults. mRNA expression of RGS4 (C) and RGS9-2 (D) on MOR+ cells in the vlPAG was significantly greater in aged rats compared with adults. No significant impact of advanced age or biological sex on mRNA expression of OPRM1 (E) or mRNA expression of OPRM1 on MOR+ cells in the vlPAG (F). Photomicrograph of RGS expression compared an adult male and an aged male PAG section (G). CFA-treated rats are represented as circles, CFA + morphine treated rats are represented as diamonds. ns, not significant. *Significant difference between adults and aged rats; p < 0.05 calculated by 2 × 2 ANOVA. Graphs indicate mean ± SEM.
Following our assessment of total RGS4 and RGS9-2, we next restricted our analyses to RGS4 and RGS9-2 mRNA expressed on MOR+ neurons. We first assessed overall OPRM1 mRNA expression in the vlPAG and found no significant main effect of age (F(1,36) = 2.793, p = 0.1034), or sex (F(1,36) = 0.076, p = 0.7844), or significant interactions between age and sex (F(1,36) = 0.7626, p = 0.7892). Similarly, no significant main effect of age (F(1,41) = 3.556, p = 0.0664), or sex (F(1,41) = 1.170, p = 0.2858), or significant interactions between age and sex (F(1,41) = 0.071, p = 0.7908) was observed for OPRM1 mRNA expressed specifically on MOR+ neurons. We next determined whether RGS4 and RGS9-2 expression was increased preferentially in OPRM1+ neurons. Similar to what was noted above, a significant main effect of age on expression of RGS4 (F(1,41) = 26.47, p < 0.0001) and RGS9-2 (F(1,41) = 21.69, p < 0.0001) was observed, with aged rats exhibiting increased levels of RGS4 and RGS9-2 mRNA in MOR+ neurons compared with their adult counterparts (Fig. 5C,D). No main effect of sex for RGS4 (F(1,41) = 0.7881, p = 0.3799) or RGS9-2 (F(1,41) = 0.0006, p = 0.9799), or significant interactions between age and sex for RGS4 (F(1,41) = 0.6583, p = 0.4219) or RGS9-2 (F(1,41) = 0.0075, p = 0.9314) were observed (Fig. 5C,D).
Discussion
The present studies are the first to show that advanced age results in a significant attenuation in MOR signaling within the vlPAG of male and female rats. Specifically, aged males and females (regardless of age) showed decreased MOR BP, decreased G-protein activation, and decreased agonist-stimulated cAMP inhibition in comparison to adult males (Fig. 6). These changes, along with the observed increase in RGS4 and RGS9-2 expression, provide a mechanism whereby morphine potency is significantly reduced in aged rats (Fullerton et al., 2021).
Model and table of MOR signaling impairments in the vlPAG of the aged and female rat. Aged and female rats exhibit decreased MOR BP, decreased G-protein activation, and decreased cAMP inhibition, while aged rats alone demonstrate increased RGS4 and RGS9-2 expression. No impact of age or sex was found in phosphorylation of MOR at serine-375.
Impact of advanced age and sex on MOR BP
We previously reported that morphine potency is reduced in aged and female rats, likely because of a reduction in DAMGO binding in the vlPAG (Fullerton et al., 2021). Here, we report significant reductions in vlPAG MOR BP in aged males and females of either age in comparison to adult male rats, suggesting the observed decrease in opioid potency is driven, in part, by reduced expression/availability of MOR in the vlPAG. Notably, no impact of advanced age or sex on the MOR's affinity for its ligand (Kd values) was noted. However, aged males had reduced MOR availability as evidenced by lower Bmax values compared with their adult male counterparts, a finding consistent with reports of reduced DAMGO binding in the vlPAG of aged males (Fullerton et al., 2021) and, together, suggest an age-induced downregulation of PAG MOR in males matching a consistently lower level of MOR expression in females of any age.
No impact of persistent inflammatory pain on vlPAG MOR BP was noted, consistent with previous studies. For example, patients with chronic fibromyalgia pain exhibit no changes in MOR binding in the PAG, despite significant reductions in the nucleus accumbens and amygdala (Harris et al., 2007). Similarly, no change in PAG MOR binding was noted following chronic pain induced by proximal nerve injury (Maarrawi et al., 2007). In rodents, persistent inflammatory pain increased MOR binding in the DRG at 24 and 96 h post-CFA (Mousa et al., 2001; Zollner et al., 2003), with no change noted for the hypothalamus or spinal cord (Shaqura et al., 2004). Similarly, no change in PAG MOR availability or expression was observed following sciatic nerve injury, although reductions were noted for the insula, caudate putamen, and motor cortex (Thompson et al., 2018). These results, together, suggest that chronic pain alters MOR signaling in a CNS region-specific manner.
Impact of advanced age and sex on G-protein activation
Agonist binding at MOR activates Gαi/o -proteins and downstream signaling cascades, critical for opioid-mediated hyperpolarization (Laugwitz et al., 1993; Connor and Christie, 1999; Koehl et al., 2018; Mondal et al., 2020). Presently, adult males exhibited greater G-protein activation efficiency compared with aged males, suggesting that the decreases in opioid potency observed in aged males are driven, in part, by attenuated G-protein-MOR coupling. A significant impact of biological sex on G-protein potency (EC50) was also observed. The mechanism(s) by which age and sex attenuate potency is not known; these studies used membrane preparations that would be devoid of soluble signaling regulators. Similarly, the GTPγS molecule itself is not hydrolyzable and is thus not subject to regulation by GAPs (e.g., RGS proteins). This suggests that the observed decrease in G-protein activation potency may be because of decreased opioid receptor expression/availability observed in our binding studies.
Interestingly, the GTPγS assay was the only analysis where an impact of persistent inflammatory pain was observed and may contribute to previous findings that the analgesic effects of intrathecal DAMGO are potentiated following inflammation in male rats (Hurley and Hammond, 2000). While reductions in Gα subunit expression have been reported within the rostral ventromedial medulla and dorsal horn following intraplantar CFA in adult male rats (Wattiez et al., 2017), the mechanism whereby this would impact G-protein activation, and not the other pharmacodynamics of MOR, is unknown. Also unknown are the mechanisms by which advanced age and sex alter G-protein-MOR signaling. Advanced age results in a global downregulation of Gαi/o, most notably in the prefrontal cortex (Young et al., 1991; Alemany et al., 2007; de Oliveira et al., 2019), that is not associated with overall cell loss. Indeed, the Gαi/o subunit, in particular, appears susceptible to aging, with estimates of reduced expression as high as 65% in the frontal cortex, hippocampus, substantia nigra, and striatum (de Oliveira et al., 2019). Although age-induced changes in Gαi/o expression were not assessed specifically in PAG, a widespread reduction in Gαi/o would likely impact PAG, thereby limiting MOR-G-protein coupling and reducing both G-protein activation efficiency and opioid potency. Alternatively, an uncoupling between MOR and Gαi/o and/or a switch in Gα subunit from Gαi/o to Gαs cannot be ruled out (Gintzler and Chakrabarti, 2000, 2004, 2006). Indeed, a shift from Gαi/o to Gαs would similarly increase AC activity, resulting in reduced hyperpolarization and decreased morphine potency (Lamberts et al., 2011).
Impact of advanced age and sex on MOR phosphorylation
The results of the radioligand binding and GTPγS assays suggested that aged and female rats exhibit decreased receptor expression and activation potency. As MOR can be desensitized by phosphorylation in its basal state, thereby limiting agonist activation (L. Zhang et al., 1996; Yu et al., 1997; Groer et al., 2011), we tested this possibility. Although MOR phosphorylation at serine-375 was higher in aged males compared with adults, this result was not significant, suggesting that our observed reductions in activation potency are driven by an alternative mechanism. Furthermore, this finding suggests that age and sex may not impact receptor desensitization and internalization, which are generally thought to be MOR phosphorylation-dependent.
MOR desensitization is canonically mediated by GPCR kinase (GRK)-dependent phosphorylation of the receptor (J. Zhang et al., 1998; Schulz et al., 2004; Dang et al., 2009). Although our results suggest no impact of advanced age or sex on GRK-mediated MOR phosphorylation, MOR signaling is also desensitized via phosphorylation through the extracellular signal-regulated kinase 1 and 2 (ERK1/2) pathway (Dang et al., 2009). ERK1/2 phosphorylation stimulates the activity of Gα-interacting protein (GAIP), an RGS protein that acts as a GTPase activator to reduce opioid signaling at the level of G-protein activation (Ogier-Denis et al., 2000). Indeed, pharmacological inhibition of ERK1/2 phosphorylation in a rat model leads to improved morphine analgesia (Popiolek-Barczyk et al., 2014; Melkes et al., 2020). Thus, the age-induced and sex-induced changes in morphine potency may be driven in part by age-induced hyperphosphorylation of ERK1/2 and result in the downregulation of G-protein signaling. This represents an alternative explanation for our phosphorylation results, by which the receptor could still be functionally desensitized/downregulated without changing canonical phosphorylation.
Impact of advanced age and sex on cAMP inhibition
Agonist binding at MOR elicits a conformational change in the receptor to allow G-protein α and βγ subunits to interact with downstream effectors. Notably, the Gα subunit binds to AC and inhibits the conversion of ATP to cAMP, limiting the activation of cAMP-dependent protein kinase (PKA) and ultimately inducing higher levels of hyperpolarization (Christie, 2008; Seseña et al., 2014; Santhappan et al., 2015). Here, aged males and adult and aged females exhibited significantly lower levels of DAMGO-induced cAMP inhibition compared with adult males, suggesting an attenuated activity of the α subunit at the level of AC or a weakened relationship between AC and cAMP. This effect may result from decreased MOR expression and G-protein activation potency observed above.
Impact of advanced age and sex on RGS protein levels
The inactivation of G-protein signaling is modulated by the activity of RGS proteins via enhancement of GTPase activity of the α subunit. By promoting the hydrolyzis of the α-bound GTP during the active state, RGS proteins hasten the return of the α subunit to the GDP-bound inactive state (Roman and Traynor, 2011). RGS proteins play a critical role in negatively modulating opioid signaling, as morphine analgesia is increased in male mice lacking RGS9-2 (Garzón et al., 2001; Zachariou et al., 2003) and overexpression of RGS4 attenuates MOR signaling in reconstituted MORs in vitro (Ippolito et al., 2002). We observed increased expression of RGS4 and RGS9-2 in the vlPAG of aged rats compared with adults, suggesting greater GTPase activity and reduced G-protein signaling. Similar results were observed when the analysis was limited to MOR+ cells, indicating opioid-induced G-protein signaling is subjected to greater negative regulation in the vlPAG of the aged rat. These results are consistent with Kim et al. (2005), who reported higher RGS9-2 protein levels in the PAG of one-year-old male rats compared with three-week-old rats (Kim et al., 2005). In humans, advanced age is associated with increased RGS4 expression in the prefrontal cortex (Rivero et al., 2010), suggesting that our observed increase in RGS4 not be specific to the vlPAG.
No significant impact of advanced age on OPRM1 mRNA expression in the vlPAG was noted. This result is interesting given our finding of reduced MOR binding in aged rats and protein expression in aged rats (Fullerton et al., 2021), and suggests that the observed age-induced reduction of vlPAG MOR protein is a function of impaired translation. Ori et al., 2015 reported reduced expression of several markers necessary to initiate translation, namely EIF3A, EIF4G3, EIF4A1, and MDN1 in aged mice (Ori et al., 2015), and a recent study using both mice and fish found that aged brains exhibit decreased proteasome activity and decreased ribosome assembly (Sacramento et al., 2020). These results suggest an age-induced dysregulation of protein synthesis in the brain that is conserved across species.
In conclusion, the present studies are the first to show that sex and advanced age lead to attenuated vlPAG opioid signaling compared with adult male rats. Taken together with our previous findings, these results suggest that age-induced and sex-induced reductions in vlPAG MOR expression and binding, combined with attenuated downstream MOR signaling, contribute to the diminished opioid potency reported in aged and female rats (Fullerton et al., 2021). The results of our analyses demonstrate that aged and female rats exhibit reductions in MOR expression/availability, G-protein activation, and cAMP inhibition, and increased G-protein regulation by RGS proteins, each of which provides potential therapeutic targets for improved pain management in the elderly.
Footnotes
This work was supported by National Institutes of Health Grants DA041529 (to A.Z.M.) and UG3DA047717 JMS and P50 MH100023 (to L.J.Y.) and by Georgia State University Molecular Basis of Disease Fellowship (E.F.F.) and Provost's Dissertation Fellowship (E.F.F.). Confocal images were acquired in the Georgia State University Imaging Core Facility.
J.M.S. has an equity stake in Botanical Results, LLC and is co-founder and equity holder in Teleport Pharmaceuticals, LLC. No company products or interests were tested in this study. All other authors declare no competing financial interests.
- Correspondence should be addressed to Anne Z. Murphy at amurphy{at}gsu.edu
