Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

A role for the endogenous opioid system in the central regulation of many types of ingestive behaviors has been characterized by using selective opioid agonists and antagonists (see reviews: Bodnar, 1996; Gosnell and Levine, 1996; Morley et al., 1983). Agonists for all three major classes of opioid receptors (mu, kappa and delta) typically stimulate spontaneous food intake, whereas general opioid antagonists decrease spontaneous intake and body weight. Further, specific opioid receptor subtype antagonists against mu, kappa and delta receptors differentially reduce food intake as a function of the ingestive situation.

The initial cloning of DOR-1 (Evans et al., 1992; Kieffer et al., 1992) quickly led to the identification of other structurally related G protein-mediated receptors (MOR-1, KOR-1, KOR-3, ORL-1; see review: Uhl et al., 1994) and has opened new areas of research, including investigation of the relationship of the cloned receptors to opioid actions in vivo. The cloning of the specific opioid receptors made possible the identification of short (15-25 bases) AS ODN sequences that are complementary to specific regions of mRNA that can down-regulate receptor proteins. The efficacy and specificity of AS ODNs directed against opioid receptor clones have been confirmed functionally and biochemically (see review: Pasternak and Standifer, 1995), particularly in analgesic assays. AS ODNs directed against the 5-untranslated regions of DOR-1, MOR-1 or KOR-1 selectively and respectively reduced delta-mediated (Bilsky et al., 1994; Lai et al., 1994; Standifer et al., 1994; Tseng et al., 1994), mu-mediated (Chen et al., 1995; Rossi et al., 1994, 1995a, 1995b) and kappa₁-mediated (Adams et al., 1994; Chien et al., 1994) forms of analgesia.

Although MOR-1 encodes a mu opioid receptor, its relationship to the pharmacologically defined mu receptor subtypes has been unclear. Using the AS ODN technique to map individual exons within MOR-1, we were able to determine which individual exons modulate mu-mediated analgesia (Rossi et al., 1995a, 1995b, 1996, 1997). AS ODN probes directed against exons 1 and 4 of MOR-1 blocked morphine and mu agonist-mediated analgesia, whereas probes targeted against exon 2 or 3 of MOR-1 were ineffective. In contrast, AS ODN probes directed against exon 2 or 3 of MOR-1 blocked analgesia induced by the morphine metabolite M6G, whereas AS ODNs directed against exon 1 or 4 of MOR-1 were ineffective. Analgesic responses induced by heroin, fentanyl and etonitazine are reduced by AS ODNs directed against either exon 1 or 2 of MOR-1 (Rossi et al., 1996). Thus these studies raised the possibility that various mu receptor subtypes could result from alternative splice variants of MOR-1 (Pasternak and Standifer, 1995).

The AS ODN strategy has recently been applied to opioid modulation of ingestive behavior (Leventhal et al., 1996). Body weight and food intake were significantly reduced by AS ODNs directed against each of the four exons of MOR-1. In contrast, a MS ODN control was ineffective. The sensitivity of spontaneous intake and body weight to all four exons suggests that the receptor responsible for this action is encoded by MOR-1. mu-selective opioid agonists such as morphine and DAMGO stimulate food intake after systemic and central administration (Bakshi and Kelley, 1993; Gosnell et al., 1986a, 1986b; Sanger and McCarthy, 1980), and this effect is blocked by mu-selective opioid antagonists (Levine et al., 1991). The first goal of the present study was to determine whether AS ODNs directed against MOR-1 would block hyperphagia elicited by the mu agonist DAMGO. If so, the second goal was to determine whether the profile of MOR-1 AS ODN effects mirrored those effects observed for mu agonist analgesia (only exons 1 and 4 AS ODNs effective) or those effects observed for spontaneous food intake and body weight (all four exons effective). Thus the present study examined whether i.c.v. administration of AS ODNs directed against exon 1, 2, 3 or 4 of MOR-1 or administration of a MS ODN altered hyperphagia elicited by the selective mu agonist DAMGO. Although AS ODNs produce quite dramatic behavioral and physiological effects, they are accompanied by rather modest (30%-40%) reductions in receptor protein levels (see review: Pasternak and Standifer, 1995). Therefore, it is possible that the presence or absence of AS ODN effects on agonist-induced hyperphagia depends on the dose of the agonist employed as well as on the efficacy of the AS ODN. Thus our third goal was to determine whether effective AS ODNs would block DAMGO-induced hyperphagia across a range of effective DAMGO doses. Finally, the present study confirmed that DAMGO-induced hyperphagia was a mu-mediated effect by blocking this response with the mu antagonist FNA (Levine et al., 1991).

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Subjects. Male albino Sprague-Dawley rats (90-120 days of age, Charles River Laboratories, Kingston, NY) were housed individually in wire mesh cages and maintained on a 12-h light: 12-h dark cycle with water and rat chow available ad libitum. Each rat was pretreated with chlorpromazine (3 mg/kg i.p.) and anesthetized with Ketamine HCl (120 mg/kg i.m.). A stainless steel guide cannula (22-gauge, Plastics One, Roanoke, VA) was implanted stereotaxically (Kopf Instruments, Tujunga, CA) into the left lateral ventricle using the following coordinates: incisor bar (+5 mm), 0.5 mm anterior to the bregma suture, 1.3 mm lateral to the sagittal suture and 3.6 mm from the top of the skull. Cannulas were secured to the skull by three anchor screws with dental acrylic. All animals were allowed at least 2 weeks to recover from stereotaxic surgery before behavioral testing began. Rats weighed between 275 and 300 g before surgery and weighed 400 to 550 g after completion of testing. After completion of behavioral testing, all animals were killed with an overdose of anesthetic, and cannula placements were verified visually.

AS ODNs and opioid agonists and antagonists. All phosphodiester oligodeoxynucleotides (Midland Certified Reagent Company, Midland, TX) were dissolved (5 µg/µl) in 0.9% normal saline and purified in our (G.W.P., G.C.R.) laboratory. AS ODNs (19-22 bases long) were directed against four regions of MOR-1 (table 1). An AS ODN control consisted of a MS ODN in which four bases from the AS1 sequence were switched without altering the remaining sequence. All sequences are specific to MOR-1 and are not present in other opioid receptor cDNAs. Three infusions were administered i.c.v. at 48-h intervals over 15 s through a stainless steel internal cannula (28-gauge, Plastics One, Roanoke, VA). DAMGO (0.5-5 µg, Peninsula Laboratories, Belmont, CA) and FNA (0.2-20 µg, Research Biochemicals Intl., Natick, MA) were dissolved in 0.9% normal saline and administered i.c.v. in 5-µl volumes over 30 s through an internal cannula. FNA was administered 24 h before agonist administration to allow for full development of irreversible mu antagonist effects (Portoghese et al., 1980).

Protocols. All rats were tested over 4 to 10 days at 3 to 9 h into the light cycle to ensure the stability of base-line spontaneous food intake. Preweighed pellets were placed directly on the floor of the wire mesh cages to optimize accessibility, because this factor can interfere with DAMGO-induced feeding (Badiani et al., 1995). Cumulative intakes were assessed 2 and 4 h before and after each condition, adjusting for spillage that was collected on paper placed under the cage. After intake stabilization, all rats received a VEH condition (PRE-VEH, 5 µl of 0.9% normal saline i.c.v.). Because DAMGO is known to produce sedative and hypoactive effects, we treated each animal twice with DAMGO (1 µg i.c.v.) without measuring intake. In the assessment of DAMGO-induced hyperphagia, rats received one of three (0.5, 1 or 5 µg) doses (PRE-DAMGO), and intake was assessed after 2 and 4 h. Intake elicited by each DAMGO dose was matched across the subgroups of rats receiving AS ODNs and the MS ODN control. During the test phase of the experiment, rats received one of four AS ODN sequences (AS1, AS2, AS3 or AS4: 10 µg, 2 µl, i.c.v.) or a mismatch sequence (MS1) (table 1) on days 1, 3 and 5 as previously described (Leventhal et al., 1996). This AS ODN dose was most efficacious in analgesic dose-response assays (Rossi et al., 1997). This time course of treatment both down-regulates the synthesis of new receptors and permits the turnover of existing receptors (see review: Pasternak and Standifer, 1995). Twenty-four hours after the last AS ODN or MS ODN treatment (day 6), rats were retested with their respective DAMGO dose (0.5-5 µg), and food intake was assessed after 2 and 4 h.

Statistics. Separate split-plot analyses of variance were performed on cumulative food intake data at 2 and 4 h for different doses of DAMGO (0.5, 1 and 5 µg) as a function of either AS ODN treatment (AS1, AS2, AS3, AS4 or MS1) or FNA treatment. Significant differences in intake measures were determined for each subgroup relative to both corresponding VEH control values and corresponding DAMGO doses before AS ODN or antagonist treatment (Tukey comparisons, P < .05).

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

DAMGO-induced hyperphagia. Significant dose-dependent differences in DAMGO-induced hyperphagia were observed after 2 [F(3,86) = 217.89, P < .0001] and 4 [F = 291.43, P < .0001] h. DAMGO significantly increased food intake relative to VEH after 4 h (fig. 1). These effects were dose-dependent in that the effect of the 5-µg dose was significantly greater than those of the 0.5- and 1-µg doses, and the effect of the 1-µg dose was significantly greater than that of the 0.5-µg dose (fig. 1). The patterns of effects after 2 h in this and the other conditions were identical.

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Alterations (mean ± S.E.M.) in food intake after i.c.v. administration of 0 (n = 87), 0.5 (n = 24), 1.0 (n = 39) or 5.0 (n = 24) µg of DAMGO. Significant differences are denoted relative to vehicle (), 0.5 µg (*) or 1.0 µg (#) DAMGO doses (Tukey comparisons, P < .01).

MOR-1 AS ODNs and DAMGO (0.5 µg)-induced hyperphagia. Significant differences in food intake were observed among conditions after 2 [F(5,23) = 67.92, P < .0001] and 4[F = 69.87, p < .0001] h. The 0.5-µg dose of DAMGO significantly increased intake after 2 and 4 h, and this effect was significantly reduced by pretreatment with either AS1 (76%) or AS4 (70%) (fig. 2A). In contrast, neither AS2 (17% increase) nor AS3 (10% increase) significantly altered DAMGO hyperphagia (fig. 2A).

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Alterations (mean ± S.E.M.) in DAMGO-induced hyperphagia at doses of 0.5 (panel A, n = 24) or 1.0 (panel B, n = 38) µg after i.c.v. administration of AS ODNs directed against exons 1 (AS1: panel A, n = 7; B, n = 8), 2 (AS2: A, n = 6; B; n = 8), 3 (AS3: A, n = 5; B, n = 8) or 4 (AS4: A, n = 6; B, n = 8) of MOR-1 or a missense control (MS1: B, n = 6). Panel C depicts alterations in food intake as a function of the DAMGO dose (0.5-5.0 µg i.c.v.) after the administration of effective (AS1 and AS4) AS ODNs of MOR-1. Significant differences are denoted relative to either VEH-VEH () or VEH-DAMGO (*) values (Tukey comparisons, P < .01).

MOR-1 AS ODNs and DAMGO (1.0 µg)-induced hyperphagia. Significant differences in food intake were observed among conditions after 2 [F(6,37) = 66.42, P < .0001] and 4 [F = 73.38, P < .0001] h. The 1.0-µg dose of DAMGO significantly increased intake after 2 and 4 h, and this effect was significantly reduced by pretreatment with either AS1 (100%) or AS4 (53%) (fig. 2B). In contrast, neither AS2 (10% increase) nor AS3 (7% increase) significantly altered DAMGO hyperphagia (fig. 2B). Further, MS1 failed to alter DAMGO-induced hyperphagia significantly at this dose (fig. 2B).

MOR-1 AS ODNs and DAMGO (5.0 µg)-induced hyperphagia. Significant differences in food intake were observed among conditions after 2 [F(4,17) = 15.93, P < .0001] and 4 [F = 21.74, P < .0001] h. The 5.0-µg dose of DAMGO significantly increased intake after 2 and 4 h. In contrast to the differential effectiveness of AS ODNs on DAMGO-induced hyperphagia at lower doses, neither AS1 nor AS4 significantly altered hyperphagia induced by a 5-µg dose of DAMGO (fig. 2C). Again, MS1 also failed to affect DAMGO-induced hyperphagia at this dose (data not shown).

FNA and DAMGO-induced hyperphagia. Significant differences were observed for DAMGO-induced hyperphagia among conditions for doses of 0.5 [F(2,23) = 91.24, P < .0001], 1.0 [F(4,37) = 112.28, P < .0001] and 5.0 [F(2,17) = 31.53, P < .0001] µg. A fixed 20-µg dose of FNA significantly and dose-dependently reduced DAMGO-induced hyperphagia at doses of 0.5 (88%), 1.0 (100%) and 5.0 (59%) µg (fig. 3A). Further, hyperphagia induced by a fixed 1.0-µg dose of DAMGO was significantly reduced by a dose range (0.2-20 µg) of FNA (80%-100%; fig. 3B).

View larger version (18K): [in this window] [in a new window]   Fig. 3.   A) Alterations (mean ± S.E.M.) in DAMGO-induced hyperphagia after pretreatment with a fixed (20-µg) dose of the mu opioid antagonist FNA as a function of the DAMGO dose. B) Alterations (mean ± S.E.M.) in DAMGO (1.0 µg)-induced hyperphagia after pretreatment with different doses (0.2-20 µg) of FNA. Significant differences induced by FNA (n = 5 or 6/dose) are denoted relative to either VEH-VEH () or VEH-DAMGO (*) values (Tukey comparisons, P < .01).

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The present findings confirmed the major goals of the study. First, specific AS ODNs directed against MOR-1 blocked hyperphagia elicited by the mu agonist DAMGO. In contrast, a MS ODN control failed to alter the magnitude of DAMGO-induced hyperphagia. Second, the observed effects of AS ODNs directed against MOR-1 were specific to the targeted regions of the clone in that AS ODNs directed against exon 1 or 4 significantly reduced DAMGO-induced hyperphagia. In contrast, AS ODNs directed against exon 2 or 3 of MOR-1 failed to exert significant effects. This pattern of AS ODN effects for mu agonist-induced hyperphagia mirrored the pattern of AS ODN effectiveness observed for mu agonist-induced analgesia (Rossi et al., 1995a, 1995b, 1996, 1997). The present pattern of results for DAMGO-induced hyperphagia stands in marked contrast to reductions in spontaneous food intake and body weight after the administration of AS ODNs directed against each of the four exons of MOR-1 (Leventhal et al., 1996). Third, there are dose-dependent limitations to the AS ODN approach, such that effective AS ODNs reduced hyperphagia at DAMGO doses of 0.5 and 1 µg but not at a higher, 5-µg dose. Hence the presence or absence of AS ODN effects on agonist-induced hyperphagia appears to depend on the dose of the agonist employed as well as on the efficacy of the AS ODN. This pattern can be explained by the modest (40%) reductions in receptor protein levels resulting from AS ODN administration (see review: Pasternak and Standifer, 1995). Finally, the ability of the mu-selective antagonist FNA to reduce DAMGO-induced hyperphagia confirmed previous findings (Levine et al., 1991). However, the magnitude of mu antagonist effects on dose-response relationships for DAMGO-induced hyperphagia paralleled AS ODN effects such that the potency of a fixed (20 µg) FNA dose to reduce DAMGO-induced hyperphagia declined as a function of increased DAMGO doses.

Our previous study (Leventhal et al., 1996) failed to observe any significant AS ODN effects on hyperphagia elicited by 2-deoxy-D-glucose or hyperdipsia elicited by angiotensin II, a surprising finding given the effectiveness of mu antagonists to block both responses (Arjune et al., 1990; Koch and Bodnar, 1994; Ruegg et al., 1994). This apparent discrepancy might be explained by the present findings of differential AS ODN effectiveness as a function of the DAMGO dose. The 2-deoxy-D-glucose (500 mg/kg i.p.) and angiotensin II (20 ng i.c.v.) doses utilized in the previous (Leventhal et al., 1996) study were at the high end of their respective dose-response curves for hyperphagia and hyperdipsia. It will be important to determine whether lower effective doses of the two physiological compounds are sensitive to AS ODNs against MOR-1.

The activity of AS ODNs that target various regions of mRNA that encode opioid receptors suggests that alternative splicing of transcripts can be explored by mapping exons with AS ODN probes (see review: Pasternak and Standifer, 1995). This approach was especially fruitful in distinguishing the analgesic responses of morphine and its metabolite M6G (Pasternak et al., 1987; Paul et al., 1989). Supraspinal and spinal mediation of morphine analgesia have been defined pharmacologically by the respective actions of mu₁ and mu₂ receptors (see review: Pasternak, 1993). M6G labels the traditional mu receptors in binding assays with an affinity slightly less than that of morphine. Yet when it is administered centrally in mice, its analgesic potency is 100-fold greater. The efficacy differences between the two drugs raised the possibility that M6G might act through a subtype of the mu receptor. Supraspinal morphine analgesia was significantly and selectively reduced by AS ODNs directed against only exon 1 or 4 of the MOR-1 clone, whereas supraspinal M6G analgesia and heroin analgesia were potently and selectively reduced by AS ODNs directed against only exon 2 or 3 (Rossi et al., 1995a, 1995b, 1996, 1997). Further, spinal morphine analgesia is blocked only by AS ODNs directed against exon 4. Thus the selectivity profiles of the AS ODN probes are not consistent with a single receptor, which implies that these subtypes might represent isoforms of MOR-1 generated by alternative splicing (Rossi et al., 1995a).

The reductions in DAMGO-induced hyperphagia by AS ODNs directed against only exon 1 or 4 parallel effects observed for mu analgesic actions, which suggests that a common splice variant of the MOR-1 clone may be mediating both analgesic and hyperphagic actions. It should be noted that inactivity of a particular AS ODN might result from a variety of technical factors, including unanticipated mRNA structures. This concern is alleviated by the fact that the inactive probes in the DAMGO-induced hyperphagia paradigm are active in reducing both M6G analgesia (Rossi et al., 1995a, 1995b, 1997) and spontaneous food intake and body weight (Leventhal et al., 1996). Further, although the same mRNA receptor substrate mediates both responses, the substrate is probably located in different supraspinal loci. Thus mu-mediated analgesic responses are most potently elicited from the periaqueductal gray, the rostroventromedial medulla and the locus ceruleus (Bodnar et al., 1988, 1991; Fang et al., 1986; Smith et al., 1988). In contrast, mu-mediated hyperphagic responses are most potently elicited from the hypothalamic paraventricular nucleus (Koch et al., 1995), the nucleus accumbens (Bakshi and Kelley, 1993; Bodnar et al., 1995; Cador et al., 1986; Majeed et al., 1986) and the ventral tegmental area (Mucha and Iversen, 1986; Noel and Wise, 1993, 1995). The differential actions of multiple MOR-1 splice variants may explain some of the different mu-mediated ingestive effects obtained using selective opioid antagonists. Thus mu antagonism with FNA significantly reduces food intake under spontaneous, deprivation, glucoprivic and palatable conditions (see review: Bodnar, 1996). In contrast, mu₁ antagonism with naloxonazine significantly reduces food intake only under spontaneous and deprivation conditions.

Two possible profiles of AS ODN effects on DAMGO-induced hyperphagia were hypothesized: the observed differential pattern described in the previous sections and consistent with mu agonist-induced analgesia, and equal effectiveness of AS ODNs directed against each exon of the MOR-1 clone as observed in spontaneous intake and weight studies (Leventhal et al., 1996). That the former, but not the latter, hypothesis was confirmed strongly suggests that the mu mediation of mu agonist-induced hyperphagia and the mu mediation of spontaneous intake and weight are different. Spontaneous food intake is an outcome of multiple factors acting on organisms, and the experimental alteration of any of them may increase feeding. For instance, mu opioid agonists increase feeding by altering the palatability of certain constituents of food, including either the macronutrient itself, such as fat (Marks-Kaufman, 1982; Marks-Kaufman and Kanarek, 1980), or the preference for a preferred macronutrient (Doyle et al., 1993; Gosnell et al., 1990). It is conceivable that DAMGO elicits feeding by acting on such specific mechanisms and that AS ODNs directed against exon 1 or 4 block this effect. In contrast, body weight and spontaneous intake may be influenced by additional factors mediated by all four exons of MOR-1. Such a distinction could not be made in traditional opioid antagonist studies. The mu antagonist FNA both significantly reduced DAMGO-induced hyperphagia (Levine et al., 1991) and significantly reduced spontaneous intake and weight under both acute (Arjune et al., 1990; Ukai and Holtzman, 1988) and chronic (Cole et al., 1995) microinjection conditions. Therefore, the AS ODN approach appears to be a more precise tool in dissecting differences in the functional roles of specific receptors. The selective actions of AS ODNs directed against different exons of MOR-1 in reducing DAMGO-induced hyperphagia support the hypothesis that multiple splice variants of MOR-1 exist and raise the possibility of further opioid receptor subclassifications.