Receptor-selective changes in µ-, δ- and κ-opioid receptors after chronic naltrexone treatment in mice

Animals and treatment

Male C57/Bl6 mice, 2 months old, were obtained from Charles River (L'Arbresle, France) and were housed in groups of four mice in extended Macrolon^© type I cages with water and food pellets available ad libitum. Environmental conditions were controlled (22 °C and 50% humidity; lights on at 07.00 h and off at 19.00 h). After transport the mice were allowed to acclimatize for at least 1 week before the experiment. The experimental procedures were approved by the Ethical Committee for Animal Experiments of Utrecht University.

Pellets containing 15 mg NTX or corresponding placebo pellets were implanted subcutaneously in the nape of the neck under halothane anaesthesia (5%/95% O₂) (day 1). NTX- and placebo-treated mice were housed together (two of both per cage). On day 8 the pellet was removed (5% halothane/95% O₂). Twenty-four hours after pellet removal the mice were killed by cervical dislocation. Brains were rapidly dissected, frozen in crushed dry ice and stored at −80 °C until use.

Autoradiographic procedures

The autoradiographic procedures used have been described previously (Kitchen et al., 1997). Briefly, 20-µm coronal sections were cut 300 µm apart using a cryostat (Leica, Rijswijk, the Netherlands) and thaw-mounted on gelatin-subbed slides. The tissue was left to dry at −20 °C in air-tight boxes containing CaSO₄ until further processing.

Tissue was preincubated for 30 min in 50 mm Tris–HCl pH 7.4 containing 0.9% NaCl. Thereafter, the sections were incubated in 50 mm Tris–HCl pH 7.4 containing either 5 nm[³H]DAMGO (µ-opioid receptors), 8 nm[³H]deltorphin-1 (δ-opioid receptors) or 2.5 nm[³H]CI-977 (κ-opioid receptors). Ligand concentrations approximated 3–4 × K_D. Nonspecific binding was determined on adjacent sections in the presence of 1 µm naloxone (µ- and κ-opioid receptors) or 10 µm naloxone (δ-opioid receptors). After incubation the slides were rinsed three times in ice-cold 50 mm Tris–HCl pH 7.4 and subsequently dried in cold air.

Slides were apposed to [³H]Hyperfilm (Amersham, UK) for 3, 4 or 6 weeks for µ-, δ- and κ-opioid receptors, respectively. Slides of both NTX and placebo treatment groups were apposed to the same film (two mice/group/film). Films were developed using Kodak D19 developer and fixed using Kodak Rapid Fixer (Sigma, UK).

Image analysis

Quantitative analysis of receptor binding was carried out by video-based computerized densitometry using an MCID image analyser (Imaging Research, Canada). Specific binding was determined by subtraction of the nonspecific binding from the total binding, using the overlay function. Specific binding is expressed in fmol/mg tissue, as derived from calibrated [³H] microscale standards (Amersham, UK) laid down with each film. For each region, quantified measures were taken from both hemispheres. Measures of binding therefore represent a duplicate determination for each brain region. Cortical areas, olfactory tubercle and hippocampus were analysed by sampling five–eight times with a box tool. All other regions were analysed by freehand drawing of anatomical areas. Structures were identified according to the mouse brain atlas of Paxinos & Franklin (2001). The n-value refers to the number of mice per treatment group.

Statistical analysis

For statistical analysis of the data SPSS 10.1 was used. The data for the three opioid receptor subtypes were analysed separately. Because binding was determined for both hemispheres, paired-sampled Student's t-tests were performed to determine hemisphere effects. Because there were no significant differences between hemispheres, the data for left and right measurements were pooled. The mean values per area per animal were used in further analyses. An overall analysis (two-way anova) was performed with treatment and area as factors. In addition, separate analyses were carried out for cortical regions, limbic structures and regions involved in movement or in pain/sensory functions. Data are represented as mean ± SEM; significance was accepted at P < 0.05.

µ-Opioid receptor autoradiography: specific [³H]DAMGO binding

Quantitative analysis of µ-opioid receptors revealed up-regulation of µ-opioid receptors after chronic NTX treatment (Table 1). The mean percentage change of [³H]DAMGO binding across all regions was 80%; changes in µ-opioid receptor levels per region ranged from 22 to 308%. There was a significant overall effect of chronic NTX treatment upon µ-opioid receptor densities (P < 0.001). Further, two-way anova revealed a significant treatment by area interaction (P < 0.001), indicative of interregional differences in up-regulation. Separate analyses for distinct neuroanatomical and functional systems revealed significant increases in µ-opioid receptor expression in cortical regions (102.1%, P < 0.001), limbic structures (64.5%, P < 0.001), regions involved in motor activity (58.6%, P < 0.01) and in pain/sensory-related structures (62.5%, P < 0.001). The data suggest an overall up-regulation of µ-opioid receptors after chronic NTX treatment, which was highest in cortical regions and was less pronounced in a small number of regions, e.g. septum and substantia nigra compacta.

δ-Opioid receptor autoradiography: specific [³H]deltorphin-1 binding

δ-Opioid receptor binding was increased with a mean percentage change of 39% across all regions; the percentage change per area varied from 1 to 109% (Table 2). There was a main effect of chronic NTX exposure upon [³H]deltorphin-1 binding (P < 0.001) and a significant treatment by area interaction (P < 0.001), indicative of regional differences in δ-opioid receptor up-regulation. Further analyses for separate functional systems revealed significant increases of δ-opioid receptors in cortical regions (34.4%, P < 0.001), limbic regions (43.2%, P < 0.001) and regions involved in motor activity (38.8%, P < 0.001) or in pain/sensory functions (39.6%, P < 0.001). The data for δ-opioid receptor binding suggest an overall up-regulation of δ-opioid receptors after chronic NTX pretreatment. It should be noted that the up-regulation of δ-opioid receptors was less pronounced in some regions, e.g. amygdala, medial geniculate nucleus, ventral tegmental area and hypothalamus.

κ-Opioid receptor autoradiography: specific [³H]CI-977 binding

Across all regions the mean percentage change in κ-opioid receptor binding was 11% (Table 3). Statistical analysis of the [³H]CI-977 binding revealed a significant effect of chronic NTX treatment (P < 0.01), but no significant treatment by area interaction. Further analysis of the data revealed significant increases in κ-opioid receptor expression in cortical regions (24.7%, P < 0.001). There was no significant change in κ-opioid receptor binding in limbic regions (2.5%) and regions involved in motor activity (6.5%) or in pain/sensory functions (−4.1%). Thus, chronic NTX-induced up-regulation of κ-opioid receptors was restricted to cortical regions.

µ-Opioid receptor autoradiography

Without exception, up-regulation of µ-opioid receptors was observed for all regions. The data from this study show interregional variation in up-regulation, but provide no evidence for up-regulation of µ-opioid receptors in specific functional neuroanatomical systems. For example, µ-opioid receptor densities were enhanced in limbic structures but also in basal ganglia and structures involved in pain and sensory motor functions, such as the periaqueductal grey, red nuclei and colliculi. Although different authors suggested there were regional differences, they disagree on exactly which areas are sensitive to chronic NTX-induced µ-opioid receptor up-regulation. In fact, comparison of the different previous studies on this matter, which all used rats, reveals that the data to support regional differences are rather weak. Zukin et al. (1982) suggested high increases in opioid receptors in the limbic system, whereas others found relatively low increases in µ-opioid receptors in limbic structures compared with other areas (Tempel et al., 1984; Morris et al., 1988). Further, Diaz et al. (2002) described highest increases in [³H]DAMGO binding in caudate putamen and nucleus accumbens, while Unterwald et al. (1998) found highest increments in µ-opioid receptors in central grey and hypothalamus. The data from the present mapping study suggest that chronic NTX exposure induces a general up-regulation of µ-opioid receptors in mice, although the percentage increase may be highest in some cortical regions.

δ-Opioid receptor autoradiography

There was an up-regulation of δ-opioid receptors after treatment with NTX, although the mean percentage change was lower as compared with the µ-opioid receptor data. This finding corresponds well to data of homogenate binding assays for chronic NTX treatment in mice (Yoburn et al., 1986b, 1988, 1989, 1995; Duttaroy et al., 1999) and an autoradiographic study by Morris et al. (1988). Although there was variation between regions, up-regulation of δ-opioid receptor binding was not restricted to a specific functional neuroanatomical system, which contrasts with earlier work by Morris et al. (1988). In summary, chronic NTX treatment results in a general increase in δ-opioid receptor binding across the brain.

κ-Opioid receptor autoradiography

The nature of the differences between placebo- and NTX-treated mice for κ-opioid receptor binding did not mirror those for µ- and δ-opioid receptors. Cortical regions showed up-regulation, whereas in noncortical regions there was no change in binding compared with placebo-treated controls. It can be concluded that 15 mg NTX pellets implanted subcutaneously for 1 week causes an up-regulation of κ-opioid receptors in merely cortical structures in mice. Morris et al. (1988) described effects of chronic naloxone treatment in rats upon κ-opioid receptor binding, but this was found for the high dose (3.0 mg/kg/h) and not the low dose (0.5 mg/kg/h). Apparently dependent on the dose, naloxone, and probably also NTX, can induce up-regulation of κ-opioid receptors.

In this study a single time-point after chronic NTX treatment is addressed: 24 h after NTX was removed; based upon studies which showed functional opioid supersensitivity in mice after chronic NTX treatment. At this time-point most of the NTX is eliminated (t_½ = 4.6 h (Yoburn et al., 1986a); 23 h postremoval approximately 97% eliminated). Tempel et al. (1982) have studied the time course of [³H]etorphine binding after chronic NTX treatment in rats. [³H]etorphine binding was enhanced by chronic NTX after 8 days of NTX treatment, which is the time-point of pellet removal in the present study. During NTX withdrawal, Tempel et al. (1982) showed that receptor levels declined and reached baseline levels by day 6 after removal of the pellets.

Quantitative autoradiography cannot discriminate between changes in B_max and K_D values. However, whole brain homogenate binding assays consistently showed increases in B_max without any changes in K_D for µ-, δ- and κ-opioid receptors (Yoburn et al., 1989, 1995; Giordano et al., 1990; Cote et al., 1993; Unterwald et al., 1995; Castelli et al., 1997; Duttaroy et al., 1999). It is therefore likely that the results described here in fact represent increases in B_max as opposed to changes in affinity.

The mechanism of opioid receptor up-regulation induced by chronic NTX is not known, although different theories have been addressed. Opioid receptor up-regulation is probably not due to increased transcription of the opioid receptor gene or related to altered opioid receptor mRNA stability (Jenab et al., 1995; Unterwald et al., 1995; Castelli et al., 1997; Duttaroy et al., 1999). Danks et al. (1988) and Rothman et al. (1989) observed µ- and δ-opioid receptor up-regulation induced both by chronic morphine and NTX, which they suggested might be explained by agonist- and antagonist-induced release of ‘antiopiates’, i.e. endogenous peptides such as cholecystokinin-8, α-MSH, dynorphin, β-endorphin and Met-enkephalin, which participate in opioid receptor up-regulation. Data concerning G-protein changes are contradictory; one study reported increased sensitivity of the newly synthesized or unmasked opioid receptors to guanyl nucleotides (Tempel et al., 1985), although in another study no changes in G-protein mRNA levels after chronic exposure to NTX were observed (Rubino et al., 1994). Further, mechanisms such as changes in receptor protein stability, changes in receptor turnover or degradation and unmasking of ‘silent’ receptors have been proposed (Castelli et al., 1997), and Unterwald et al. (1995) suggested that NTX might inhibit normal down-regulation of opioid receptors, presumably by preventing endogenous opioids to bind. Finally, discrepancies between quantitative µ-opioid receptor immunoreactivity and quantitative µ-opioid receptor autoradiography led to suggestions that the percentage of active receptors may be increased without a change in the total number of receptors (Unterwald et al., 1998).

The data of the present study clearly show chronic NTX-induced up-regulation of µ- and δ-opioid receptors, while κ-opioid receptor expression was enhanced in cortical but not in noncortical regions. The pattern of up-regulation induced by chronic NTX treatment (µ > δ > κ) as described here does not parallel the ratio in affinity of NTX for µ-, δ- and κ-opioid receptors [µ > κ > δ (Kieffer, 1995; Gutstein & Akil, 2001)]. This suggests that the affinity of NTX to a receptor is not predictive of the extent of NTX-induced supersensitivity of that receptor. It is further interesting to note that µ- and κ-opioid receptors are thought to have different, even opposing, functions (for review, see Pan, 1998; Narita et al., 2001). For example, µ-opioid agonists induce antinociception and tolerance, µ-opioid agonists are rewarding (van Ree et al., 1999), µ-agonists impair memory, cause euphoria, increase dopamine release from the nucleus accumbens (Di Chiara & Imperato, 1988) and are proconvulsant. In contrast, κ-opioid receptor agonists have anti µ-opioid actions, i.e. they block morphine analgesia, reduce morphine tolerance and improve µ-opioid agonist-induced memory impairment. Moreover, κ-opioid agonists induce µ-opioid opposing effects (but see van Ree et al., 1999): they induce dysphoria, are aversive, improve memory processes, act as anticonvulsants and decrease dopamine release from the nucleus accumbens. The present data suggest that µ- and κ-opioid receptors are also differentially regulated by chronic NTX.

In conclusion, the present full quantitative mapping study of µ-, δ- and κ-opioid receptors revealed clear effects of chronic NTX exposure upon opioid receptors in mice. Chronic NTX induced overall increases in µ- and, although to a lesser extent, in δ-opioid receptor binding. Changes in κ-opioid receptors were restricted to cortical regions. The findings described here suggest opioid receptor subtype-selective regulation mechanisms.