Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Feb 1.
Published in final edited form as: Pharmacol Biochem Behav. 2022 Jan 22;213:173339. doi: 10.1016/j.pbb.2022.173339

Age-related changes in CB1 receptor expression and function and the behavioral effects of cannabinoid receptor ligands

Brett C Ginsburg 1, Julie G Hensler 2,3
PMCID: PMC8973309  NIHMSID: NIHMS1775849  PMID: 35077729

Abstract

Cannabinoid use has increased among aging individuals. However, little information on age-related differences in the behavioral effects of these agents is available. To explore potential differences in the behavioral effects of cannabinoids, we determined effects of Δ9-tetrahydrocannabinol (THC, 1–10 mg/kg) or rimonabant (0.3–3.2 mg/kg) on operant fixed-ratio responding (FR10) for food in young adult (6 months) and aged (29 months) rats. THC dose-dependently decreased responding for food. Rimonabant alone had little or no effect on responding up to 1.0 mg/kg, but disrupted responding following a 3.2 mg/kg dose. Rimonabant (1.0 mg/kg) partially antagonized response disruption by THC. These effects were similar in young adult and aged rats. However, aging has been reported to change the neurobiology of cannabinoid CB1 receptors. To confirm our rats exhibited such differences, we assessed CB1 receptor binding sites and function in six subcortical (caudate, nucleus accumbens CA1, and CA2/CA3), and three cortical regions (medial prefrontal, temporal, entorhinal) in young adult (6 months) or aged (26 months) male Lewis rats using quantitative autoradiography. CB1 receptor binding sites were reduced in cortical, but not subcortical brain regions of aged rats. CB1 receptor function, at the level of receptor-G protein interaction, was not different in any region studied. Results indicate that down-regulation of CB1 receptor binding sites observed in cortical regions of aged rats was not accompanied by a commensurate decrease in CB1 receptor-stimulated [35S]GTPγS binding, suggesting a compensatory increase in receptor function in cortical areas. Together, our results provide additional evidence of age-related changes in central CB1 receptor populations. However, the functional compensation for decreased CB1 receptor binding may mitigate changes in behavioral effects of cannabinoids. With the rising use of cannabinoid-based therapeutics among aging populations, further evaluation of age-related changes in the cannabinoid system and the impact of these changes on effects of this class of drugs is warranted.

1. Introduction

The use of cannabinoids as therapeutic and recreational drugs have increased among aging populations recently, yet, only limited data are available on any age-related differences in cannabinoid effects (Minerbi et al., 2019; Weinstein and Sznitman, 2020). Age-related changes in the endogenous cannabinoid system could influence the effects of therapies that target the cannabinoid system in aging populations (Minerbi et al., 2019). Though aging generally appears to result in decreased availability of CB1 receptor binding sites, age-related changes in CB1 receptor expression appear to depend on the brain region observed. For example, in aged rats, CB1 receptor binding sites are reported to be decreased in the lateral (but not medial) caudate-putamen, globus pallidus, substantia nigra, and the dentate gyrus of the hippocampus, when compared to young adults (Berrendero et al., 1998; Liu et al., 2003; Mailleux and Vanderhaeghen, 1992; Romero et al., 1998). Reduced CB1 receptor binding sites in aged rats compared with young adults have also been reported in the molecular layer of the cerebellum, the hypothalamus, and deep layer VI of the cortex (Berrendero et al., 1998), and reduced receptor protein levels were observed in postrhinal cortex (Liu et al., 2003). Thus, CB1 receptors may be diminished in cortical and subcortical regions of aged subjects.

Aging has also been reported to decrease the function of CB1 receptors in subcortical regions. The maximal stimulation of GTPγS binding by the synthetic cannabinoid HU-210 is reduced in limbic forebrain membrane preparations (but not cerebellum, amygdala, or hippocampus) from aged versus young mice (Wang et al., 2003). In rats, age-related decreases in CB1 receptor function were reported using stimulation of GTPγS binding by WIN 55212-2 in the lateral caudate-putamen, substantia nigra, and globus pallidus (Romero et al., 1998). Together these studies reveal a consistent pattern of age-related decline in CB1 receptor function in subcortical regions. Changes in cortical CB1 receptor function have not been reported.

While the functional consequences of these age-related changes in the cannabinoid system remain unclear, there is some evidence that suggests diminished behavioral effects of cannabinoids with age. For example, recreational use of cannabinoid agonists including THC is less common among older individuals, though this may be due to cultural or social influences rather than a change in the reinforcing effectiveness of the drugs themselves (Center for Behavioral Health Statistics and Quality, 2017).

Better understanding of age-related changes in CB1 receptor expression and function and the subsequent changes in behavioral effects of cannabinoid agonists may impact the therapeutic and recreational use of cannabinoids in aging populations. There is increasing interest in the therapeutic use of cannabinoids among the aged for a variety of indications including memory and cognitive function as well as pain and inflammation (Bilkei-Gorzo et al., 2017; Minerbi et al., 2019).

To further explore whether cannabinoid effects may differ in young versus aged individuals, we evaluated the behavioral effects of the CB1 agonist Δ9-tetrahydrocannabinol (THC) and the CB1 antagonist/inverse agonist rimonabant, and their interaction on disruption of operant fixed-ratio food-maintained responding using a cumulative dosing paradigm. This procedure allowed drug effects to be compared in young adult and aged rats using a food-maintained, operant behavior that is under exquisite experimental control. We also examined changes in CB1 receptor binding sites and CB1 receptor function, at the level of receptor-G protein interaction, in the brains of adult rats at two different ages (approximately 6 and 26 months). The binding of the agonist radioligand [3H]CP55,940 to CB1 receptors sites and CP55,940 - stimulated [35S]GTPγS binding to G proteins was measured in various cortical and subcortical regions using quantitative autoradiography.

2.0. Methods

2.1. Subjects:

Male Lewis rats (Harlan, Inc, Frederick, MD) were 6 weeks old upon arrival. For the next two weeks, rats were allowed to habituate to vivarium routines. During the second week, food rations were restricted (~12g/d for all rats) to maintain consistent weights of about 330g (± 5g), to facilitate food-maintained behavioral studies, and to provide consistent feeding conditions across all experimental subjects. Food restriction also extends the quality and quantity of rat lifespan, facilitating the retention of aged rats for this study (Yu, 1994). Further, all subjects were trained under the behavioral procedure to equate experiences across the duration of the study across groups. However, rats used for the quantitative autoradiography studies never received injections of any drug or vehicle. A timeline of study events is shown in Figure 1.

Figure 1:

Figure 1:

Open in a new tab

Timeline of study events. Events on the timeline are indicated as post-natal day. Note all rats were exposed to the behavioral procedure to equate handling, feeding, and behavioral experience across rats. However rats in the quantitative autoradiography studies were never exposed to any drug prior to sacrifice.

2.1. Behavior:

Sixteen rats were used for these studies. In the first study, effects of THC or rimonabant alone were examined. Behavioral testing occurred between the ages of PND 141 and 207 (4.7 – 6.9 months) in eight animals, which constituted the young adult group. For the other eight rats, testing occurred between the ages of PND 141–207 (28.1 – 30.3 months), which constituted the aged group. In the second study, interactions between THC and rimonabant were examined in separate groups of rats.

2.1.3. Quantitative autoradiography:

Sixteen rats were used for these studies. Eight of the rats used were sacrificed at PND 175 (5.8 months). These animals constituted the young adult group. The other eight rats were sacrificed at PND 784 (26.1 months). These rats constituted the aged group.

2.2. Behavioral Studies:

2.2.1. Training:

Rats were initially trained in commercially available equipment (Med-Associates, Georgia, VT) to respond on a lever located to the left of a receptacle into which a magazine dispensed 45-mg food pellets (Rodent grain-based diet, Bio-Serv, Frenchtown, NJ, USA). Illumination of a light above the lever signaled the availability of food. Initially, a single response on the lever resulted in the delivery of a pellet, illuminated a houselight above the chamber and turned off the lever light for 10-s. Responses while the light was not illuminated had no programmed consequence. Once rats reliably responded for food during a 2-hour session, response requirements were increased to 10 over a few sessions (FR10), and session length was reduced to 1-hour. After rats consistently responded under the FR10 schedule, the session was altered such that rats responded during five 5-min components each preceded by a 15-min timeout in which no lights were illuminated and responses had no programmed consequence.

2.2.2. Cumulative dosing:

Once rats responded reliably during each of the five components in a daily session, subjects involved in behavioral studies began receiving saline injections immediately before each session began on Tuesday, Thursday, and Friday of each week. Rats were placed in the conditioning chamber for 15 min, and then lights were illuminated and rats were allowed to respond for food for 5 min. Subsequently, each subject was removed from the operant chamber and injected once again, then replaced in the chamber. A 15 min timeout and subsequent 5 min period of responding for food followed, then rats received the next injection across five such response components. Once behavior was stable across sessions in which saline injections were administered, rats received injections of vehicle on Tuesdays and Thursdays of each week, and saline on Friday. Once responding following 5 doses of vehicle was stable, injections of vehicle or saline were administered before each component on Tuesday and Friday, respectively, as controls. On Thursday, rats received cumulative doses of rimonabant (0.32, 0.56, 1.00, 1.78, and 3.20 mg/kg, sequentially). Subsequently, rats received vehicle on Tuesday and saline on Friday, and cumulative doses of THC (1.00, 1.78, 3.20, 5.60, and 10.0 mg/kg) on Thursday.

2.2.3. Interaction Studies:

Rats were then treated with rimonabant (0.32, 1.0 mg/kg) or vehicle 15 min before the first dose of THC was administered. Responding was assessed 15-min later, as described above, and additional cumulative doses of THC (1.00, 1.78. 3.2, 5.6, and 10.0 mg/kg) were administered 15-min before each 5-min response period as described above. Responses during each 5-min response period were compared across rimonabant pretreatment condition. At least two weeks separated any drug treatments to allow for recovery of behavioral stability and to minimize the development of tolerance.

2.2.4. Behavioral Analysis:

Effects of rimonabant and THC were expressed as a percentage of control responses per 5-min component. Control values were from the corresponding component following vehicle administration on Tuesday of the same week for each rat. A mixed-effects analysis of variance (ANOVA) was performed for effects of each drug with dose as a within-subject factor and age as a between subject factor. When no significant effect of age was present, post-hoc analysis by dose over age was not performed. Determination of ED50 values was complicated by the lack of substantial disruption of responding by rimonabant in several subjects across the dose range tested. Instead, the maximal effect observed in each subject across the dose ranges of rimonabant and THC were compared between young and aged rats to provide a measure of drug potency using a Student’s t-test. Analysis was performed using R version 4.0 with linear mixed effect regression from the NLME package and ANOVA from the CAR package.

2.3. Quantitative Autoradiography:

Rat brains were rapidly removed, frozen on dry ice and stored at −80 °C until use. Coronal sections (20 μm) were cut in a cryostat microtome at −16 °C at the level of the medial prefrontal cortex (from Bregma 3.2 to 2.2 mm), caudate putamen (1.6 to 1.0 mm), dorsal hippocampus (−3.14 to −3.6 mm) and entorhinal cortex (−6.72 to −7.04 mm) according to the Paxinos atlas of the rat brain (Paxinos and Watson, 2007). Sections were thaw mounted onto gelatin-coated glass slides, desiccated at 4 °C for 18 hours under vacuum and stored at −80 °C until use.

2.3.1 Autoradiography of binding of the agonist radioligand [3H]CP55,940 to CB1 receptors was performed as we have done previously (Seillier et al., 2010). Slide-mounted sections were pre-incubated for 15 minutes in Tris–HCl buffer (50 mM, pH 7.4) supplemented with 3 mM MgCl2, 0.2 mM EGTA, 100 mM NaCl, and then incubated for 2 hours in the same Tris-HCl buffer containing 0.5% BSA and 4 nM [3H]CP55,940. Non-specific binding was determined by incubating adjacent sections with the CB1 antagonist AM251 (1 mM). Slides were washed for 1 hour and then 3 hours in ice-cold Tris–HCl (50 mM) containing 0.3% BSA, and then rinsed by briefly dipping in cold deionized water. Sections were dried and exposed to Kodak Biomax MR film (Amersham) for 3 weeks.

2.3.2 Autoradiography of CP55,940-stimulated [35S]GTPγS binding in brain sections was performed as we have done previously (Seillier et al., 2010). Slide-mounted sections were equilibrated in Tris–HCl buffer (50 mM, pH 7.4), supplemented with 3 mM MgCl2, 0.2 mM EGTA, 100 mM NaCl, 0.2 mM dithiothreitol and 0.5% bovine serum albumin (BSA) for 10 minutes at 25 °C. Sections were pre-incubated in Tris–HCl buffer containing GDP (2 mM) and DPCPX (1 μM), an adenosine A1 receptor antagonist, for 15 minutes at 25 °C, and then incubated in the same buffer plus 40 pM [35S]GTPγS either in the absence or presence of the CB1 receptor agonist CP55,940 (1 μM), for 2 hours at 25 °C. This concentration of CP55,940 produces maximal stimulation (Emax) of [35S]GTPγS binding (Seillier et al., 2010). In addition to characterizing the concentration-response relationship, we have also previously assessed the specificity of CP55,940-stimulated [35S]GTPγS binding. The CB1 receptor antagonist AM251 (100 nM) markedly reduces CP55,940- stimulated [35S]GTPγS binding, whereas the CB2 receptor antagonist JTE907 (100 nM) has no effect, indicating that this effect of CP55,940 is CB1 receptor-mediated (Seillier et al., 2010). Basal [35S]GTPγS binding was defined in the absence of CP55,940. Non-specific [35S]GTPγS binding was defined in the absence of CP55,940 and in the presence of 10 μM GTPγS. The incubation was stopped by two washes (2 minutes each) in ice-cold 50 mM Tris–HCl, followed by a rinse in ice-cold deionized water (30 seconds). Sections were dried on a slide warmer and exposed to Kodak Biomax MR film for 48 hours.

2.3.3 Digitized autoradiograms were analyzed using the NIH Image software (ImageJ 1.42q). [3H]CP55,940 binding was quantified using simultaneously exposed [3H]-labeled standards (ART-123; American Radiochemicals, USA), which had been calibrated according to the method of Geary and colleagues (Geary et al., 1985; Geary and Wooten, 1983). The amount of ligand bound was determined by converting optical density measurements to femtomoles per milligram of protein. Specific binding was calculated by subtracting nonspecific binding from total binding on adjacent sections. Autoradiograms of CP55,940-stimulated [35S]GTPγS binding were quantified using simultaneously exposed [14C]-labeled standards (ARC-146; American Radiochemicals, USA). Standard curves were fitted to pixel data obtained from [14C] standards and tissue equivalent values (nCi/g) provided by American Radiochemicals, and were used to transform regional densitometric values into relative radioactivity measures. Non-specific binding of [35S]GTPγS was subtracted from basal values and from the binding of [35S]GTPγS in the presence of CP55,940.

2.4. Drugs:

Vehicle was prepared by dissolving 0.5 ml Tween-80 (Sigma-Aldrich, St. Louis, MO) and 0.5 ml propylene glycol (Sigma-Aldrich, St. Louis, MO) in 9 ml saline, resulting in a 5% tween, 5% propylene glycol solution. Rimonabant (NIDA, Rockville MD) was dissolved in tween, which was then emulsified with propylene glycol and diluted with saline to provide a 10 mg/ml stock solution in the same vehicle described above. From this stock solution, five working solutions were made by diluting with vehicle. These solutions were 0.32, 0.26, 0.44, 0.80, and 1.2 mg/ml, and were injected in ascending order before each component of the test session at a volume of 1 ml/kg. THC (NIDA, Rockville, MD) was shipped dissolved in 100% ethanol, which was evaporated with dry nitrogen. The resulting THC residue was dissolved in Tween-80 and emulsified with propylene glycol, then diluted to a stock solution of 25 mg/ml in the vehicle described above. From this stock solution, five working solutions were made by diluting with vehicle. These solutions were 1.0, 0.8, 1.4, 2.4, and 4.4 mg/ml), and were also injected in ascending order before each component of the test session at a volume of 1 ml/kg.

3. Results

3.1. Behavior

3.1.1 Behavioral stability across weekdays. As shown in Figure 2, ANOVA revealed a significant interaction between day of the week and session component on responses (F[16,336]=3.3, p<0.001). Significant main effects of day of the week (F[4,336] = 5.5, p<0.001) and session component (F[4,336]=17.4, p<0.001) were also present. No significant main effect of age, nor any interactions between age and day of the week or component were present.

Figure 2:

Figure 2:

Open in a new tab

Behavioral stability for each session component across weekdays. Symbols represent response rates for each component for each weekday, open are young adults and filled are aged rats (n=8 per group). Data are only from days (18 ± 3 assessments for all subjects on each day) on which rats received saline, vehicle, or no treatment at all.

3.1.2 Rimonabant dose-dependently decreased responding for food. Generally, the effect of rimonabant to disrupt responding for food was greater at any given dose in the aged rats (Figure 3). However, this did not reach statistical significance. An analysis of variance revealed a main effect of rimonabant dose (F[4, 56]=4.5, p<0.005) but not of age (p>0.1) on responding for food. There was no significant interaction between dose and age (p>0.2).

Figure 3:

Figure 3:

Open in a new tab

Effects of Cumulative doses of rimonabant on response rate during operant responding maintained by food in young adult (left) and aged (right) rats. Points represent mean response rate ± SEM of n=8 rats in each group following each dose. White ribbon represent the 95% confidence interval for responding observed after vehicle treatment.

3.1.3 THC (Figure 4, open circles) also dose-dependently decreased responding for food (F[4,56]=8.4, p<0.001). Post-hoc comparison revealed THC significantly reduced responding compared to vehicle control at all dose levels of THC (p<0.05), even after Benjimini-Hochberg correction for multiple comparisons. ANOVA also revealed a significant effect of age (F[1,14]=9.4, p<0.01). However, post-hoc comparisons showed only effects in the first component (after 1 mg/kg THC) differed between young and aged rats, and this difference did not withstand Benjamini-Hochberg correction for multiple comparison. The interaction between dose and age (F[4, 336] = 2.3, p<0.06) also fell short of significance.

Figure 4:

Figure 4:

Open in a new tab

Effects on response rate maintained by food of THC alone (open symbols) or in combination with rimonabant (0.32 mg/kg – upward pointing triangles or 1.0 mg/.kg - downward pointing triangles) administered 15-min before the first THC dose. Effects in young adult (left) or aged (right) rats are shown (n=8 per group). Points represent the mean response rate ± SEM during each response period following the THC dose indicated. White ribbon represents the 95% confidence interval.

3.1.4 Interactions between THC and Rimonabant. Rimonabant pretreatment (1.0 mg/kg, but not 0.32 mg/kg) dose-dependently partially antagonized effects of THC on responding (Figure 4, filled triangles). This was confirmed with ANOVA, which revealed an interaction between THC dose and rimonabant pretreatment (F[16, 336]=2.2, p<0.005) as well as main effects of rimonabant pretreatment (F[4,336]=3.6, p<0.05) and THC dose (F[4,336]=8.4, p<0.001). Post-hoc comparisons revealed that 0.32 mg/kg rimonabant did not significantly affect THC effects. However, 1.0 mg/kg rimonabant significantly attenuated the effects of 3.2, 5.6, and 10 mg/kg THC (p<0.05, after Benjimini-Hochberg correction. Subsequent analysis revealed that each of these results also significantly differed from responding following vehicle alone, suggesting a partial reversal of the THC effect by rimonabant. Because no age-effect was present, these analysis were collapsed across both age groups.

3.1.5 Effects on responding the day after THC or rimonabant exposure. While unexpected, an effect was present on responding the day after treatment with THC and/or rimonabant (Figure 5). As described above (Section 2.2.2), only repeated administration of saline occurred on this testing day. Responding on Friday (the day after rats received vehicle alone) was similar to responding on Thursday (the day vehicle was administered) (compare white polygons in Figure 3 and 4 with those in Figure 5). Age did not influence this effect, nor did age interact with the treatment or response component (p>0.1). A main effect of the treatment on the previous day was present (F[4,336]=9.5, p<0.0001). Subsequent analysis revealed rimonabant alone (Figure 5, top panel), THC alone (Figure 5 bottom panel; open circles), or 1.0 mg/kg rimonabant in combination with THC (Figure 5 bottom panel; filled triangles) resulted in significant differences in responding across all five response periods on the Friday following these exposures, when compared with Thursday treatment with vehicle alone. These results were significant (p<0.05) even after Benjimini-Hochberg correction for multiple comparisons.

Figure 5:

Figure 5:

Open in a new tab

Responding for food in young versus aged rats the day after rimonabant (top panel) or THC alone or combined with rimonabant (bottom panels) administration. Details are as described for Fig. 3, except that rats were administered saline 15-min prior to each 5-min component the day after cumulative exposure the previous day. The white polygons represent responding on Friday after Thursday vehicle treatment. Responding following rimonabant alone, THC alone, and THC combined with rimonabant were all significantly reduced compared with responding following vehicle (p<0.05, see text). Responding following rimonabant alone and THC combined with rimonabant were not significantly different from each other, but were significantly different from responding following THC alone (p<0.05).

Further, responding the day after rats were exposed to rimonabant alone or rimonabant and THC combined resulted in significant differences in responding when compared to responding the day after exposure to THC alone, as indicated in Figure 5. In contrast, responding the day after exposure to rimonabant alone did not differ from responding the day after exposure to rimonabant and THC combined. All comparisons were significant (p<0.05) after correction for multiple comparisons.

3.2. Autoradiography:

3.2.1 The effect of age on CB1 receptor binding sites in cortical and subcortical brain regions was examined by measuring the binding of the agonist radioligand [3H]CP55,490. [3H]CP55,490 binding was significantly reduced (p<0.05 after Bonferroni correction for multiple comparisons) in medial prefrontal cortex, temporal cortex and entorhinal cortex of aged rats (Figure 6A). Interestingly, the binding of [3H]CP55,490 in anterior cingulate cortex was similar in aged and young adult animals (Figure 6A). In contrast to what was observed for many cortical areas, [3H]CP55,490 binding was not altered in subregions of dorsal hippocampus, caudate putamen or nucleus accumbens (Figure 6B). Our data indicate that there was a region-specific effect of age on CB1 receptor sites, specifically a reduction in cortical regions, but not in subcortical areas.

Figure 6:

Figure 6:

Open in a new tab

Region-specific decreases CB1 receptor binding in the brains of aged versus young adult rats. Coronal brain sections were incubated with the CB1 receptor agonist [3H]CP55,940 (4 nM). Non-specific binding was defined in the presence of AM251 (1 mM). Specific binding of [3H]CP55,940 is expressed as fmol/mg protein. (A) CB1 receptor binding was reduced in medial prefrontal cortex (mPFCx), temporal cortex (TmpCx) and entorhinal cortex (EntCx), but not in anterior cingulate cortex (aCgCx). (B) CB1 receptor binding in CA1, CA2/3, dentate gyrus (DG) regions of dorsal hippocampus, caudate (CPu) and nucleus accumbens (NAc) was statistically similar in young adult and aged rats. Data are presented as mean±SEM, n = 8 per group. * p < 0.05, per a Student’s t-test corrected for multiple comparisons using the Bonferroni method.

3.2.2 We also assessed the effect of age on CB1 receptor function, specifically CB1 receptor-stimulated binding of [35S]GTPγS to G proteins. The binding of [35S]GTPγS stimulated by the CB1 receptor agonist CP55,490 (1 μM) was not statistically different between young adult and aged rats in any area of brain examined (Figure 7). There was an increase in CP55,490-stimulated [35S]GTPγS binding in the CA2/3 region of the hippocampus in aged rats (p<0.05), however this increase failed to meet statistical significance after Bonferroni correction for multiple comparisons. Our data indicate that the down-regulation of CB1 receptor binding sites observed in cortical regions was not accompanied by a commensurate decrease in CB1 receptor-stimulated [35S]GTPγS binding. This could be interpreted as an increase in receptor function at the level of receptor-G protein interaction in cortical areas to compensate for the decrease in binding sites observed in those regions.

Figure 7:

Figure 7:

Open in a new tab

CB1 receptor function was not altered in (A) cortical or (B) subcortical brain regions of aged versus young adult rats. Coronal brain sections were incubated with [35S]GTPγS (40 pM). Nonspecific binding was defined in the presence of 10 mM GTPγS. [35S]GTPγS binding was stimulated by CP55,940 (1 mM). Specific binding of [35S]GTPγS is expressed as % above basal. medial prefrontal cortex (mPFCx); temporal cortex (TmpCx); entorhinal cortex (EntCx); anterior cingulate cortex (aCgCx); CA1, CA2/3, dentate gyrus (DG) regions of dorsal hippocampus; caudate (CPu); nucleus accumbens (NAc). Data are presented as mean ± SEM, n = 8 per group.

4. Discussion

Taken together, these results suggest that aging may produce an increase in CB1 receptor function, even while CB1 receptor number declines with age. However, these age-related changes did not alter the behavioral effects of THC, rimonabant, or their combination. Together, our results provide further evidence that age-related changes in central CB1 receptor populations occur, yet these changes do not appear to result in grossly different behavioral effects of drug action at this receptor site.

In the present study CB1 receptor binding sites were reduced in cortical regions of aged rats, i.e. medial prefrontal, temporal, and entorhinal cortices. Others have also shown decreases in CB1 receptor binding or receptor protein expression in cortical regions of rat brain, particularly deep layer VI and postrhinal cortex (Berrendero et al., 1998; Romero et al., 1998). In contrast, Liu et al. (2003) showed increased expression of CB1 receptor protein in entorhinal and temporal cortex. The reasons for this inconsistency is not clear, however it is important to note that Liu used Western blotting in whole tissue preparations, so changes in CB1 receptor expression do not necessarily reflect changes in cell-surface binding site availability.

In humans, postmortem studies reveal age-related decreases in CB1 receptor binding sites in cortical regions (Mato and Pazos, 2004; Westlake et al., 1994). More recently, a study utilizing PET imaging in vivo revealed a positive relationship between CB1 receptor binding in cortical areas and age in women, though no such relationship was present for men (Van Laere et al., 2008). The reasons for this discrepancy are unclear, however high inter-subject variability in CB1 receptor binding potential and possible differences in the origin and histories of the sample populations may contribute to the differences reported.

In the present study, no changes in CB1 receptor binding were observed in subcortical regions. Previous studies have found age-related declines in CB1 receptor binding, especially in the striatum and basal ganglia. In an early study of CB1 receptor binding sites, a 40% decrease in binding in the caudate was reported in rats from age 3 months to 24 months (Mailleux and Vanderhaeghen, 1992). However, few details of the housing conditions for the rats were provided. In a later study in rats of similar ages, CB1 receptor binding in the lateral (but not medial) caudate-putamen, the substantia nigra, the peduncularpontine nucleus, and the dentate gyrus of the hippocampus were found to be decreased (Romero et al., 1998). This discrepancy between our results and this prior report may be due to differences in food access, as those rats had ad libitum access to food, while rats in the present study were provided daily food rations to maintain a stable weight and behavioral stability. Additionally, rats in the present study were housed individually, while housing conditions were not reported by Romero et al., (Romero et al., 1998). Based on these results in both cortical and subcortical regions, housing and feeding conditions may influence CB1 receptor binding sites, though further investigation is required to demonstrate this.

In the present study, the down-regulation of CB1 receptor binding sites in cortical regions was not accompanied by a commensurate decrease in CB1 receptor-stimulated [35S]GTPγS binding. Instead, our data suggest an increase in receptor function in these same cortical areas, specifically at the level of receptor-G protein interaction in compensation for the decrease in binding sites. In subcortical regions, we observed no change in CB1 receptor binding sites or in CB1 receptor-stimulated [35S]GTPγS binding. Our data are in marked contrast to other studies examining age-related changes in CB1 receptor function in rodent brain. Romero et al. (1998) reported decreased GTPγS binding in the basal ganglia (i.e., lateral caudate-putamen, substantia nigra, and globus pallidus) of aged rats. Aging has also been reported to decrease the function of CB1 receptors in the limbic forebrain (nucleus accumbens and anterior cingulate cortex) of mice (Wang et al., 2003). These results in rodents are extended to postmortem human brain by (Mato and Pazos, 2004). The authors found a mean reduction in CB1 binding sites and coupling efficiency of approximately 10% per decade in layers IV, V, and VI of frontal cortex (Mato and Pazos, 2004). This effect is not due to reduced G-protein expression because G-protein levels do not decline with age (Mato and Pazos, 2004; Wang et al., 2003). Nor is this effect likely due to age-dependent changes in extracellular endocannabinoid levels as no changes in extracellular levels of endocannabinoids are apparent in aging in CD1 or C57BL/6J mice (Maccarrone et al., 2001; Maccarrone et al., 2001; Wang et al., 2003). Thus, data from several species suggests that although endocannabinoid levels and G-protein expression may remain relatively stable, CB1 receptor function changes with age. The reasons for these discrepancies between our study and previous work are unclear. The influence of strain or environmental conditions (e.g., feeding and housing) during the lifespan could account for different outcomes across studies (e.g. Judge et al., 2009). Additional research is necessary to investigate these possibilities.

There was no difference in the potency and efficacy of THC to disrupt operant responding for food between aged rats and young adult rats. Thus, the observed decrease in CB1 receptor binding but not function in aged rats does not result in a change in the behavioral effects of THC. This is perhaps not surprising as there is a large CB1 receptor reserve, and THC appears to act as a full agonist in behavioral assays (McMahon et al., 2008; Wiley et al., 2007). While chronic treatment (which presumably results in reductions in both receptor expression and function) can reveal efficacy differences among CB1 agonists in behavioral assays, these effects are subtle and are not conferred to the rate-disruptive effects of the drugs (Hruba et al., 2012; McMahon, 2011; Paronis et al., 2012). Thus, it is not surprising that THC effects were similar in young adult and aged rats (in which CB1 receptor number was reduced but function remained unchanged).

Rimonabant alone had only modest effects on responding across the dose range tested, with similar effects in both young and aged rats. We selected this range based on evidence that the highest dose of rimonabant tested occupies over 65% of CB1 receptors, which is sufficient to affect feeding-related behavior, but mitigates the likelihood non-selective effects at higher doses of rimonabant (Need et al., 2006). There was some evidence that higher doses of rimonabant began to produce some disruption. This could be due to low efficacy at CB1 receptors (inverse or partial agonism, see: Dore et al., 2013; De Vry and Jentzsch, 2004) or to activity at other, non-CB1 receptor sites of action. Consistent with its activity as a CB1 receptor antagonist, rimonabant partially reversed the disruptive effects of THC. This effect was similar in young adult and aged rats. The failure of rimonabant to fully reverse THC effects could be due to rate-disruptive effects of rimonabant itself.

The disruption of responding on the day after rimonabant and THC was unexpected. This did not occur on the Tuesday or Friday after repeated vehicle administration, suggesting it is specific to drug exposure. These effects may reflect something similar to a ‘hangover’. Indeed, in humans, driving impairment is most pronounced hours after THC exposure, long after acute subjective effects have passed (Ginsburg, 2019). This result could reflect lingering cognitive impact. There is some evidence in humans that young individuals are more susceptible to acute cognitive effects and withdrawal effects of cannabis (Sexton et al., 2018), though we did not observe age-related differences on our behavioral measure. Preliminary studies on this topic reveal a complex relationship between cannabis use and cognition among aged individuals (Benitez et al., 2020).

Our behavioral results indicate that despite age-related changes in the cannabinoid system, behavioral effects of THC, rimonabant, and their interactions are similar in young adult and aged rats. This may be due to an increase in CB1 receptor function accompanying a decrease in CB1 receptor expression in aged rats. Such changes in central cannabinoid effects may be relevant to therapeutic as well as to recreational uses of these drugs. The increasing therapeutic use of such agents in aging populations warrants further evaluation of the impact of age-related changes in the cannabinoid system and the impact of these changes on the efficacy of drugs acting on this system.

Highlights:

  • Cannabinoid CB1 receptor agonist THC and antagonist rimonabant dose-dependently decreased responding for food similarly in young adult and aged rats. Rimonabant antagonized THC effects in responding similarly in aged versus young rats.

  • THC disrupted responding 24-hours after exposure in both age groups, and this effect was partially reversed by rimonabant.

  • Cannabinoid CB1 receptor binding sites were reduced in cortical, but not subcortical brain regions of aged rats, compared to young rats.

  • CB1 receptor function, at the level of receptor-G protein interaction, was not different between ages in any region studied.

  • Together, our results suggest that CB1 receptor function may compensate for age-related changes in binding sites, such that behavioral effects of cannabinoids are similar in young and aged rats. Additional studies to determine if more subtle behavioral differences emerge with age are warranted.

Acknowledgements:

The authors wish to thank NIDA for generously providing THC and rimonabant for this study. The authors gratefully acknowledge the technical assistance of Teresa Burke, Olivia Dominguez, Gerardo Martinez, and Marisela Valdez. The authors also thank Dr. Tushar Advani for assistance with the [35S]GTPγS binding. This study was funded by PHS grant 021195.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Declarations of Interest

None. The views and information presented are those of the authors and do not represent the official position of the U.S. Army Medical Center of Excellence, the U.S. Army Training and Doctrine Command, or the Department of the Army, Department of Defense, or U.S. Government.

References

  1. Benitez A, Lauzon S, Nietert PJ, McRae-Clark A, Sherman BJ, 2020. Self-reported cognition and marijuana use in older adults: Results from the national epidemiologic survey on alcohol and related conditions-III. Addict Behav 108, 106437. 10.1016/j.addbeh.2020.106437 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Berrendero F, Romero J, García-Gil L, Suarez I, De la Cruz P, Ramos JA, Fernández-Ruiz JJ, 1998. Changes in cannabinoid receptor binding and mRNA levels in several brain regions of aged rats. Biochim. Biophys. Acta 1407, 205–214. [DOI] [PubMed] [Google Scholar]
  3. Bilkei-Gorzo A, Albayram O, Draffehn A, Michel K, Piyanova A, Oppenheimer H, Dvir-Ginzberg M, Rácz I, Ulas T, Imbeault S, Bab I, Schultze JL, Zimmer A, 2017. A chronic low dose of Δ9-tetrahydrocannabinol (THC) restores cognitive function in old mice. Nat. Med 23, 782–787. 10.1038/nm.4311 [DOI] [PubMed] [Google Scholar]
  4. Center for Behavioral Health Statistics and Quality, 2017. 2016 National Survey on Drug Use and Health: Detailed Tables.
  5. De Vry J, Jentzsch KR, 2004. Partial agonist-like profile of the cannabinoid receptor antagonist SR141716A in a food-reinforced operant paradigm. Behav Pharmacol 15, 13–20. [DOI] [PubMed] [Google Scholar]
  6. Dore R, Valenza M, Wang X, Rice KC, Sabino V, Cottone P, 2013. The inverse agonist of CB1 receptor SR141716 blocks compulsive eating of palatable food. Addict Biol. 10.1111/adb.12056 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Geary WA 2nd, Toga AW, Wooten GF, 1985. Quantitative film autoradiography for tritium: methodological considerations. Brain Res 337, 99–108. [DOI] [PubMed] [Google Scholar]
  8. Geary WA 2nd, Wooten GF, 1983. Quantitative film autoradiography of opiate agonist and antagonist binding in rat brain. J. Pharmacol. Exp. Ther 225, 234–240. [PubMed] [Google Scholar]
  9. Ginsburg BC, 2019. Strengths and limitations of two cannabis-impaired driving detection methods: a review of the literature. The American Journal of Drug and Alcohol Abuse 45, 610–622. 10.1080/00952990.2019.1655568 [DOI] [PubMed] [Google Scholar]
  10. Hruba L, Ginsburg BC, McMahon LR, 2012. Apparent inverse relationship between cannabinoid agonist efficacy and tolerance/cross-tolerance produced by Δ9- tetrahydrocannabinol treatment in rhesus monkeys. J. Pharmacol. Exp. Ther 342, 843–849. 10.1124/jpet.112.196444 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Judge MK, Zhang Y, Scarpace PJ, 2009. Responses to the cannabinoid receptor-1 antagonist, AM251, are more robust with age and with high-fat feeding. J. Endocrinol 203, 281–290. 10.1677/JOE-09-0210 [DOI] [PubMed] [Google Scholar]
  12. Liu P, Bilkey DK, Darlington CL, Smith PF, 2003. Cannabinoid CB1 receptor protein expression in the rat hippocampus and entorhinal, perirhinal, postrhinal and temporal cortices: regional variations and age-related changes. Brain Res 979, 235–239. [DOI] [PubMed] [Google Scholar]
  13. Maccarrone M, Attinà M, Bari M, Cartoni A, Ledent C, Finazzi-Agrò A, 2001. Anandamide degradation and N-acylethanolamines level in wild-type and CB1 cannabinoid receptor knockout mice of different ages. J. Neurochem 78, 339–348. [DOI] [PubMed] [Google Scholar]
  14. Maccarrone M, Attinà M, Bari M, Cartoni A, Ledent C, Finazzi-Agrò A, 2001. Anandamide degradation and N-acylethanolamines level in wild-type and CB1 cannabinoid receptor knockout mice of different ages. J. Neurochem 78, 339–348. 10.1046/j.1471-4159.2001.00413.x [DOI] [PubMed] [Google Scholar]
  15. Mailleux P, Vanderhaeghen JJ, 1992. Age-related loss of cannabinoid receptor binding sites and mRNA in the rat striatum. Neurosci. Lett 147, 179–181. [DOI] [PubMed] [Google Scholar]
  16. Mato S, Pazos A, 2004. Influence of age, postmortem delay and freezing storage period on cannabinoid receptor density and functionality in human brain. Neuropharmacology 46, 716–726. 10.1016/j.neuropharm.2003.11.004 [DOI] [PubMed] [Google Scholar]
  17. McMahon LR, 2011. Chronic Δ9-tetrahydrocannabinol treatment in rhesus monkeys: differential tolerance and cross-tolerance among cannabinoids. Br. J. Pharmacol 162, 1060–1073. 10.1111/j.1476-5381.2010.01116.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. McMahon LR, Ginsburg BC, Lamb RJ, 2008. Cannabinoid agonists differentially substitute for the discriminative stimulus effects of Delta(9)-tetrahydrocannabinol in C57BL/6J mice. Psychopharmacology (Berl.) 198, 487–495. 10.1007/s00213-007-0900-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Minerbi A, Häuser W, Fitzcharles M-A, 2019. Medical Cannabis for Older Patients. Drugs Aging 36, 39–51. 10.1007/s40266-018-0616-5 [DOI] [PubMed] [Google Scholar]
  20. Need AB, Davis RJ, Alexander-Chacko JT, Eastwood B, Chernet E, Phebus LA, Sindelar DK, Nomikos GG, 2006. The relationship of in vivo central CB1 receptor occupancy to changes in cortical monoamine release and feeding elicited by CB1 receptor antagonists in rats. Psychopharmacology 184, 26–35. 10.1007/s00213-005-0234-x [DOI] [PubMed] [Google Scholar]
  21. Paronis CA, Nikas SP, Shukla VG, Makriyannis A, 2012. Δ9-Tetrahydrocannabinol acts as a partial agonist/antagonist in mice. Behav Pharmacol 23, 802–805. 10.1097/FBP.0b013e32835a7c4d [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Paxinos G, Watson C, 2007. The Rat Brain in Stereotaxic Coordinates, Sixth Edition: Hard Cover Edition, 6th ed. Academic Press. [Google Scholar]
  23. Romero J, Berrendero F, Garcia-Gil L, de la Cruz P, Ramos JA, Fernández-Ruiz JJ, 1998. Loss of cannabinoid receptor binding and messenger RNA levels and cannabinoid agonist-stimulated [35S]guanylyl-5’O-(thio)-triphosphate binding in the basal ganglia of aged rats. Neuroscience 84, 1075–1083. [DOI] [PubMed] [Google Scholar]
  24. Seillier A, Advani T, Cassano T, Hensler JG, Giuffrida A, 2010. Inhibition of fatty-acid amide hydrolase and CB1 receptor antagonism differentially affect behavioural responses in normal and PCP-treated rats. Int. J. Neuropsychopharmacol 13, 373–386. 10.1017/S146114570999023X [DOI] [PubMed] [Google Scholar]
  25. Sexton M, Cuttler C, Mischley LK, 2018. A Survey of Cannabis Acute Effects and Withdrawal Symptoms: Differential Responses Across User Types and Age. The Journal of Alternative and Complementary Medicine 25, 326–335. 10.1089/acm.2018.0319 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Van Laere K, Goffin K, Casteels C, Dupont P, Mortelmans L, de Hoon J, Bormans G, 2008. Gender-dependent increases with healthy aging of the human cerebral cannabinoid-type 1 receptor binding using [(18)F]MK-9470 PET. Neuroimage 39, 1533–1541. 10.1016/j.neuroimage.2007.10.053 [DOI] [PubMed] [Google Scholar]
  27. Wang L, Liu J, Harvey-White J, Zimmer A, Kunos G, 2003. Endocannabinoid signaling via cannabinoid receptor 1 is involved in ethanol preference and its age-dependent decline in mice. Proc Natl Acad Sci U S A 100, 1393–1398. 10.1073/pnas.0336351100 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Westlake TM, Howlett AC, Bonner TI, Matsuda LA, Herkenham M, 1994. Cannabinoid receptor binding and messenger RNA expression in human brain: an in vitro receptor autoradiography and in situ hybridization histochemistry study of normal aged and Alzheimer’s brains. Neuroscience 63, 637–652. [DOI] [PubMed] [Google Scholar]
  29. Weinstein G, Sznitman SR, 2020. The implications of late-life cannabis use on brain health: A mapping review and implications for future research. Ageing Research Reviews 59, 101041. 10.1016/j.arr.2020.101041 [DOI] [PubMed] [Google Scholar]
  30. Wiley JL, O’connell MM, Tokarz ME, Wright MJ, 2007. Pharmacological effects of acute and repeated administration of Delta(9)-tetrahydrocannabinol in adolescent and adult rats. J. Pharmacol. Exp. Ther 320, 1097–1105. 10.1124/jpet.106.108126 [DOI] [PubMed] [Google Scholar]
  31. Yu BP, 1994. How diet influences the aging process of the rat. Proc. Soc. Exp. Biol. Med 205, 97–105. [DOI] [PubMed] [Google Scholar]

RESOURCES