Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Opioid analgesics such as morphine are the standard therapeutic agents for the treatment of moderate-to-severe pain. Although these agents have demonstrated efficacy in treating many types of pain, their clinical use can be limited in certain situations due to side effects mediated primarily through the µ-opioid receptor (Ellison, 1993). These side effects include the development of tolerance and physical dependence, respiratory depression, constipation, and urinary retention. One approach to limit these side effects is to selectively target - and -opioid receptors. This approach, however, has seen only limited success due to inherent side effects mediated by -opioid receptors, pharmacokinetic factors with opioid peptides, or limited efficacy/toxicity with nonpeptidic -agonists (Porreca et al., 1997). A second approach is to develop compounds that possess mixed opioid activity at the different opioid receptors. This has led to a number of interesting compounds that are in clinical use today (O'Brien and Benfield, 1989; Preston and Jasinski, 1991; Strain et al., 1996a,b; Eder et al., 1998; Greenwald et al., 1999).

Studies in the early 1990s from the laboratories of Takemori and Portoghese suggested that blockade of the -opioid receptor attenuates the development of morphine tolerance and physical dependence. Initial studies assessed the effects of naltrindole and naltrindole 5'-isothiocyanate on development of morphine tolerance and physical dependence (Abdelhamid et al., 1991). Using -subtype-selective antagonists, follow-up studies provided evidence that this effect was mediated through the -2 receptor rather than -1 receptor (Miyamoto et al., 1993a,b, 1994). Subsequent studies by Suzuki and colleagues have provided additional evidence that -antagonists dose dependently suppress the behavioral and biochemical changes seen after morphine withdrawal (Suzuki et al., 1995, 1997).

Results from recent studies have provided additional support to the concept behind the development of µ-agonist/-antagonist ligands. Of particular significance are the observations of Pintar and coworkers in their experiments with -receptor knockout mice (Zhu et al., 1999). They reported that genetic deletion of the cloned -opioid receptor does not disrupt µ-receptor-mediated analgesic activities of drugs such as morphine. Interestingly, in contrast to wild-type animals in which analgesic response to a fixed dose of morphine is lost within 5 days due to development of tolerance, the -receptor knockout mice failed to develop tolerance to the analgesic effects of morphine. These results provide direct evidence regarding the central role that -receptors play in the development of tolerance to µ-agonists and are in conformity with the earlier observations that blockade of -receptors decreases the development of tolerance without diminishing the analgesic potency of µ-agonists.

Schiller and coworkers have been engaged in the development of peptides possessing mixed µ-agonist/-antagonist profiles of activity. In a recent publication they disclose the identification of H-Dmt-Tic[CH₂NH]Phe-Phe-NH₂ as one such peptide ligand possessing a balanced µ-agonist/-antagonist profile (Schiller et al., 1999). This compound given i.c.v. produced potent analgesic activity in the tail-flick test in rats. Furthermore, the compound produced less acute tolerance than morphine and produced no physical dependence on chronic administration at high doses.

In our investigations on nonpeptide opioid ligands derived from naltrexone, we identified a pyridomorphinan compound SoRI 9409, 5'-(4-chlorophenyl)-17-(cyclopropylmethyl)-6,7-didehydro-3,14-dihydroxy-4,5-epoxypyrido-[2',3':6,7]morphinan (Fig. 1), as a lead structure possessing µ-agonist/-antagonist activity (Ananthan et al., 1999). This compound displayed high affinity at the -receptor, high antagonist potency in the mouse vas deferens bioassay, and moderate agonist potency in the guinea pig ileum bioassay. Antinociceptive evaluations in mice showed that i.c.v. injections of SoRI 9409 produced a partial agonist effect in the 55°C tail-flick assay and a full agonist effect in the acetic acid writhing assay. No signs of overt toxicity were observed with this compound in the dose ranges tested. Moreover, repeated i.c.v. injections of an A₉₀ dose did not induce any significant development of antinociceptive tolerance in the acetic acid writhing assay.

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Chemical structures of morphine, naltrexone, naltrindole, and SoRI 9409.

Nonpeptide ligands in general do not suffer from the bioavailability problems that are associated with peptide ligands. The nonpeptide nature of SoRI 9409 coupled with the promising µ-agonist/-antagonist profile that this compound displayed led us to undertake a more complete in vivo characterization of SoRI 9409, including agonist and antagonist profiles after systemic administration to male ICR mice.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals. Male, ICR mice (20-30 g) (Harlan Industries, Cleveland, OH) were housed in groups of five in Plexiglas chambers with food and water available ad libitum before any procedures. Animals were maintained on a 12-h light/dark cycle in a temperature-controlled animal colony. Studies were carried out in accordance with the Guide for the Care and Use of Laboratory Animals as adopted and promulgated by the National Institutes of Health.

Chemicals. SoRI 9409 was synthesized as previously described (Ananthan et al., 1999). Morphine sulfate, DAMGO, U69,593, -FNA, naltrindole, and nor-BNI were obtained through the National Institute on Drug Abuse drug supply program. Naloxone HCl was purchased from RBI/Sigma (Natick, MA). DPDPE and [D-Ala²,Glu⁴]deltorphin were generous gifts from Dr. Victor Hruby at the University of Arizona, Tucson, AZ.

Drug Solutions and Injections. SoRI 9409 and U69,593 were initially dissolved in 100 µl of glacial acetic acid and 900 µl of distilled water. The solution was brought up to approximately pH 5.5 with 1.74 M NaOH and then the final volume adjusted with distilled water. Vehicle was prepared in a similar manner without the drug. Additional compounds used in these studies were dissolved in distilled water (i.c.v. injections) or physiological saline (i.p. injections). Intracerebroventricular injections were performed as previously described (Porreca et al., 1984). Briefly, mice were lightly anesthetized with ether and an incision was made in the scalp. An injection was made with a 25-µl Hamilton syringe at a point 2 mm caudal and 2 mm lateral from bregma. Compounds were injected at a depth of 3 mm in a volume of 5 µl.

Antinociceptive Testing. Antinociception was assessed using either the 55°C warm-water tail-flick test or the acetic acid writhing assay. For the tail-flick test, the latency to the first sign of a rapid tail-flick was taken as the behavioral endpoint (Jannsen et al., 1963). Each mouse was first tested for baseline latency by immersing its tail in the water and recording the time to response. Mice not responding within 5 s were excluded from further testing. Mice were then administered the test compound and tested for antinociception at 10, 20, 30, 45, 60, and 90 min postinjection. A maximum score was assigned (100%) to animals not responding within 15 s to avoid tissue damage. Antinociception was calculated by the following formula: % antinociception = 100 × (test latency  control latency)/(15  control latency). For the acetic acid writhing test, mice were injected i.p. with doses of morphine or SoRI 9409, followed 10 min later by an i.p. injection of 0.9% acetic acid. They were then placed in a clear Plexiglas observation jar and the number of abdominal writhes recorded for 15 min (Mogil et al., 1999). Percentage of antinociception was calculated using the formula % maximum possible effect = 100  ((number of writhes individual mouse/mean number of writhes control group) × 100).

Tolerance Studies. Mice received repeated twice (7 AM and 7 PM) or thrice daily (7 AM, 1 PM, and 7 PM) i.p. injections of vehicle or A₉₀ doses of morphine or SoRI 9409 for 3 days. On the morning of the 4th day, i.p. dose-response curves were generated for each agonist in the acetic acid writhing assay. The A₅₀ values for each compound in vehicle-treated and drug-treated mice were compared to generate tolerance shifts.

Agonist Effects of SoRI 9409. To further determine the in vivo opioid receptor profile of SoRI 9409, mice were pretreated with a µ- (-FNA, 19 nmol i.c.v., 24 h), - (naltrindole, 20 mg/kg i.p., 20 min), or - (nor-BNI, 1 nmol i.c.v., 24 h) selective antagonist. Control mice received a vehicle injection (5 µl of distilled water, i.c.v., 24 h). These times and doses have previously been shown to produce selective blockade of µ-, -, and -receptors, respectively (Portoghese et al., 1988; Jiang et al., 1991; Horan et al., 1992). Mice then received an A₉₀ dose of SoRI 9409 (30 nmol i.c.v.) followed 10 min later by an i.p. injection of 0.9% acetic acid.

Antagonist Actions of SoRI 9409. Mice were pretreated with vehicle or various doses of SoRI 9409 (i.p., 20 min) followed by injection of i.c.v. A₉₀ doses of selective µ- (DAMGO, 0.1 nmol), -1 (DPDPE, 30 nmol), -2 ([D-Ala²,Glu⁴]deltorphin (20 nmol), or - (U69,593, 60 nmol) agonists at 0 min. Antinociception was assessed 10 min after agonist injection, which corresponded to the time of agonist peak effect (Horan et al., 1992). In a second set of studies, i.c.v. antinociceptive dose-response curves were generated for [D-Ala²,Glu⁴]deltorphin and DAMGO in vehicle or SoRI 9409 (10 mg/kg i.p., 20 min) pretreated mice.

Physical Dependence. To assess development of morphine physical dependence, both an acute and chronic assay in mice was used (Yano and Takemori, 1977; Bilsky et al., 1996). Mice were pretreated with a single injection of morphine (100 mg/kg s.c., 4 h) or a morphine pellet (75 mg s.c., 72 h). Withdrawal was precipitated by an injection of the opioid antagonist naloxone (10 mg/kg i.p.), SoRI 9409 (10 mg/kg i.p.), or a combination of both. Mice were immediately placed in a clear Plexiglas cylinder and observed for 15 min. The number of vertical jumps was recorded during this time.

Statistical Analysis. For antinociceptive tests, dose-response lines were constructed at times of agonist peak effect and analyzed using linear regression with FlashCalc software. All A₅₀ values (95% confidence limits) shown are calculated from the linear portion of the dose-response curve. Antagonist ID₅₀ values were calculated in a similar manner. Single-dose data were analyzed using one-way analysis of variance (ANOVA) followed by the Scheffé test or Student's t test for between groups comparisons (significance set at p < 0.05). A minimum of 10 mice was used at each dose level.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

The antinociceptive dose- and time-response curves (55°C tail-flick) for morphine and SoRI 9409 are depicted in Figs. 2 (i.c.v. administration) and 3 (i.p. administration). As expected, morphine produced a full agonist effect via each route of administration. The calculated A₅₀ values (and 95% confidence intervals) were 1.7 nmol (0.8-3.7 nmol) and 8.0 mg/kg (6.3-10.0 mg/kg), respectively. In contrast, SoRI 9409 produced only partial agonist effects after i.c.v. administration (40.3% maximum possible effect at 100 nmol) and had no measurable antinociceptive effect after i.p. administration at doses up to 60 mg/kg. In the acetic acid writhing test, both compounds produced dose-related antinociception after i.c.v. and i.p. administration (Fig. 4).

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Dose- and time-response curves for i.c.v. morphine and SoRI 9409 in the 55°C tail-flick test.

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Dose- and time-response curves for i.p. morphine and SoRI 9409 in the 55°C tail-flick test.

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Comparison of the morphine and SoRI 9409 i.c.v. (A) and i.p. (B) dose-response curves in the acetic acid writhing test.

The full agonist activity of morphine and SoRI 9409 in the acetic acid writhing assay allowed us to assess the development of antinociceptive tolerance to each compound. Repeated i.p. administration of approximate A₉₀ doses of morphine (6 mg/kg, ×2 or ×3 daily for 3 days) shifted the morphine dose-response curve approximately 2.4- and 5.1-fold, respectively (Fig. 5; Table 1). In contrast, repeated i.p. injections of A₉₀ doses of SoRI 9409 (30 mg/kg, ×2 or ×3 daily for 3 days) did not significantly shift the SoRI 9409 dose-response curve (Fig. 5; Table 1).

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Dose-response curves for morphine (A) and SoRI 9409 (B) in mice treated repeatedly (×2 or ×3 daily, 3 days) with vehicle or A90 doses of the respective agonist.

To determine the opioid receptor(s) through which SoRI 9409 produces its antinociceptive actions, mice were pretreated with vehicle or a selective µ-, -, or -antagonist. Mice were then injected with an A₉₀ i.c.v. dose of SoRI 9409 and antinociception was assessed in the acetic acid writhing assay. An ANOVA of the data depicted in Fig. 6 yielded an F(4,65) = 11.2, p < 0.001. Post hoc analysis using a Scheffé test indicated that the 30-nmol dose of SoRI 9409 significantly decreased the number of writhes (p < 0.001). This effect was blocked by pretreatment with -FNA (p < 0.002) but not by naltrindole (p > 0.99) or nor-BNI (p > 0.13). There was also no difference between the vehicle control and nor-BNI group (p > 0.23).

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Nociceptive responses to 0.9% acetic acid (i.p.) in mice treated with vehicle or an A90 dose of i.c.v. SoRI 9409. Mice were pretreated with either vehicle, µ-, -, or -selective opioid antagonists at various times before SoRI 9409 injection (under Materials and Methods). *Indicates a significant (p < 0.05) difference in number of writhes between vehicle-injected and SoRI 9409-injected mice. Indicates a significant (p < 0.05) difference in number of writhes between SoRI 9409-injected mice and -FNA/SoRI 9409-treated mice.

The antagonist actions of SoRI 9409 were initially assessed by pretreating mice i.p. with doses of SoRI 9409. Mice were then injected with A₉₀ doses of selective µ-, -, or -agonists and antinociception was assessed in the 55°C tail-flick test. The antagonist dose-response curves are depicted in Fig. 7 with the corresponding ID₅₀ values and 95% confidence limits displayed in Table 2. SoRI 9409 potently antagonized the actions of the -2-selective agonist [D-Ala²,Glu⁴]deltorphin. The compound was much less potent at antagonizing the actions of the -1 agonist DPDPE or the µ-agonist DAMGO. In addition, doses of up to 60 mg/kg SoRI 9409 did not affect the antinociceptive actions of the -agonist U69,593 (data not shown). To further confirm the -selective antagonist actions of the compound, mice were pretreated with i.p. vehicle or a 10-mg/kg dose of SoRI 9409. Full i.c.v. antinociceptive dose-response curves for [D-Ala²,Glu⁴]deltorphin or DAMGO were then constructed. SoRI 9409 produced a significant 8.2-fold shift in the [D-Ala²,Glu⁴]deltorphin dose-response curve (Fig. 8A), while having no significant effect (1.5-fold shift) on the DAMGO dose-response curve (Fig. 8B).

View larger version (22K): [in this window] [in a new window]   Fig. 7.   Antagonist dose-response curves for i.p. SoRI 9409 against fixed i.c.v. A90 doses of selective µ- (DAMGO) or - (DPDPE and [D-Ala2,Glu4]deltorphin) agonists in the 55°C tail-flick test.

View larger version (14K): [in this window] [in a new window]   Fig. 8.   Agonist dose-response curves for i.c.v. [D-Ala2,Glu4]deltorphin (A) or DAMGO (B) in vehicle or SoRI 9409 (10 mg/kg i.p., 20 min) pretreated mice in the 55°C tail-flick test.

Because SoRI 9409 displayed a -antagonist activity profile similar to that of naltrindole, we assessed the compounds ability to elicit a withdrawal syndrome in morphine-dependent mice and to attenuate a naloxone-precipitated withdrawal (Fig. 9). These effects were assessed in both an acute (Fig. 9A) and chronic (Fig. 9B) model of morphine dependence. For the acute studies, an ANOVA yielded an F(2,56) = 12.1, p < 0.0001. Post hoc analysis (Student's t test) indicated that an i.p. injection of SoRI 9409 (10 mg/kg) precipitated significantly less vertical jumps than naloxone injection (10 mg/kg i.p.) (p < 0.001). Coadministration of SoRI 9409 and naloxone also produced significantly less jumps than naloxone alone (p < 0.05). Implantation of a 75-mg morphine pellet for 72 h produced significantly greater physical dependence than single injection of morphine (Student's t test, p < 0.001). Qualitatively similar results were, however, seen in both models. The ANOVA for the chronic model yielded an F(2,37) = 6.2, p < 0.01. Post hoc analysis (Student's t test) indicated that an i.p. injection of SoRI 9409 (10 mg/kg) precipitated significantly less vertical jumps than naloxone injection (10 mg/kg i.p.) (p < 0.006). Coadministration of SoRI 9409 and naloxone also produced significantly less jumps than naloxone alone (p < 0.05).

View larger version (12K): [in this window] [in a new window]   Fig. 9.   Elicitation of opioid withdrawal (vertical jumping) by naloxone (10 mg/kg i.p.), SoRI 9409 (10 mg/kg i.p.), or a combination of naloxone and SoRI 9409 in mice pretreated with morphine acutely (100 mg/kg s.c., 4 h) or chronically (morphine pellet, 75 mg s.c., 72 h). *Indicates a significant (p < 0.05) difference in number of jumps from the naloxone/morphine control.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Considerable evidence supports the hypothesis that blockade of -opioid receptors attenuates the development of morphine tolerance and physical dependence (Abdelhamid et al., 1991; Miyamoto et al., 1993a,b; Suzuki et al., 1995; Ananthan et al., 1999; Schiller et al., 1999; Zhu et al., 1999). Through modification of a naltrexone-based pyridomorphinan, we have synthesized SoRI 9409, a compound that possesses both µ-agonist and -antagonist properties in vitro and in vivo. An advantage of SoRI 9409 is its nonpeptidic structure, which presumably is relatively stable in blood serum and able to penetrate the blood-brain barrier after systemic routes of administration. The compound produced a full agonist effect in the acetic acid writhing assay after i.c.v. or i.p. administration. It is interesting that the potency ratios of SoRI 9409/morphine are much closer to one after i.p. versus i.c.v. administration. This may be due to the highly lipophilic nature of SoRI 9409 at physiological pH. The compound would be expected to quickly diffuse out of the central nervous system after i.c.v. injection, yet cross the blood-brain barrier more readily than morphine after systemic routes of administration. A spinal or peripheral mechanism of action may also help explain why the compound was significantly less potent than morphine after i.c.v. administration, but nearly equipotent after i.p. injection. A definitive explanation for these finding will require a more detailed study of the compound.

The agonist effects of SoRI 9409 were mediated primarily through the µ-opioid receptor, because the µ-antagonist -FNA almost completely blocked the antinociceptive actions of the compound. There may, however, be a weak -agonist component associated with SoRI 9409, because nor-BNI partially blocked the antinociceptive effects of the compound. This action is not totally unexpected because SoRI 9409 has moderate (~20 nM) affinity at the -opioid receptor and nor-BNI shifted the SoRI 9409 dose-response curve 1.6-fold to the right in the guinea pig ileum bioassay (Ananthan et al., 1999).

The potency and central nervous system bioavailability of the compound after i.p. administration allowed us to assess the development of antinociceptive tolerance after repeated i.p. injections of an A₉₀ dose of the drug. As predicted from our i.c.v. studies (Ananthan et al., 1999), repeated twice-daily injections of i.p. SoRI 9409 for 3 days produced significantly less antinociceptive tolerance than the administration of equianalgesic doses of morphine. One potential criticism of the study might be the limited morphine tolerance-shift seen with this regimen. To partially address this concern, additional experiments were run using thrice-daily injections of agonist for 3 days. A greater rightward shift in the morphine, but not SoRI 9409, dose-response curve was observed with this protocol. Other standard tolerance regimens (i.e., escalating-dose) were not feasible. A standard escalating dose procedure would have required doses of SoRI 9409 in excess of 200 mg/kg. The concentrations required to attain these doses in mice surpassed the solubility limits of the compound. By lowering the pH of the solution, or using dimethyl sulfoxide as a solvent, we were able to dissolve the compound at concentrations up to 30 mg/ml. These solvents were, however, incompatible with repeated systemic injections of the compound. In fact, repeated i.p. injections of the vehicle for the 30-mg/kg dose of SoRI 9409 (pH = ~5.5) produced body weight loss and malaise in animals injected over a 6-day period (data not shown).

Other potential explanations for the lack of tolerance seen with SoRI 9409 include pharmacokinetic and pharmacodynamic differences between morphine and SoRI 9409. The antinociceptive effects of both compounds after i.p. administration are similar (~60 min for A₉₀ doses). Furthermore, the antagonist actions of the compound last for at least 60 min (data not shown). The lower efficacy of SoRI 9409 could also confound interpretation of the results. Based on previous research, however, we would expect a greater tolerance-shift to repeated injections of SoRI 9409 than to morphine. Specifically, when equieffective antinociceptive doses of opioid compounds are injected, there is an inverse relationship between the amount of tolerance produced and the intrinsic efficacy of the particular compound (Stevens and Yaksh, 1989; Sosnowski and Yaksh, 1990; Duttaroy and Yoburn, 1995).

The complete lack of tolerance seen with SoRI 9409 in the current studies supports the hypothesis that -antagonists can prevent µ-agonist tolerance. Presumably, this action is mediated via the fairly potent and selective antagonist actions of SoRI 9409 at the pharmacologically characterized -2 opioid receptor (Fig. 7). The compound was significantly less potent in antagonizing -1-, µ-, or -opioid-mediated antinociception (Fig. 7). Additional studies confirmed that a 10-mg/kg dose of SoRI 9409 significantly shifted the [D-Ala²,Glu⁴]deltorphin antinociceptive dose-response curve to the right, while having no significant effect on the antinociceptive actions of DAMGO (Fig. 8). The blockade of DAMGO antinociception with the 60-mg/kg dose of SoRI 9409 may be due to greater receptor occupancy of the µ-opioid receptor with a partial agonist. Somewhat lower doses of SoRI 9409 (30 mg/kg) are needed to block the antinociceptive actions of the less efficacious µ-agonist morphine (data not shown).

The partial µ-agonist/-antagonist profile of SoRI 9409 may also explain why the compound precipitated a less severe withdrawal syndrome than naloxone in morphine-dependent mice and, furthermore, attenuated the expression of naloxone-precipitated withdrawal. In severely dependent mice (morphine pellet), the partial µ-agonist effects of SoRI 9409 may not be able to completely substitute for morphine, because these mice did display some signs of withdrawal (e.g., vertical jumping, diarrhea, and urination). These results are similar to those seen with other partial µ-opioid agonists such as buprenorphine.

In addition to limited antinociceptive tolerance and physical dependence liability, other clinical advantages may be seen with a compound possessing mixed µ-agonist/-antagonist actions. A recent study in rats indicated that naltrindole infusion reversed alfentanil-induced respiratory depression as indexed by changes in pCO₂ levels (Su et al., 1998). The -selective doses of naltrindole that were used (0.1-0.5 mg/kg i.v.) did not affect the antinociceptive actions of alfentanil and did not alter respiration by themselves. The -antagonists naltriben (NTB) and 7-benzylidenenaltrexone produced similar effects in this assay. These effects were also seen after i.c.v. infusion of the -antagonists naltrindole, NTB, 7-benzylidenenaltrexone, and H-Tyr-Tic-Phe-Phe(). Naltrindole and NTB have also been shown to reverse sufentanil-induced respiratory depression in mongrel dogs (Freye et al., 1992). Although the specific role of -receptors is unclear in the regulation of respiration, it appears that naltrindole-like drugs reverse µ-opioid-mediated respiratory depression. Further studies will be needed to assess whether SoRI 9409 produces less respiratory depression than morphine and the fentanyl analogs.

An additional side effect of µ-opioid analgesics is the inhibition of gastrointestinal motility. During reflex neuromuscular activity, opioid peptides are released in a coordinated manner at precise locations to modulate sphincteric and peristaltic activity. Opioid peptides acting at the -receptor may promote segmentation, nonpropulsive contractions, and delay transit through the gut. Foxx-Orenstein and colleagues hypothesized that the administration of -antagonists would stimulate colonic propulsion. Administration of naltrindole dose dependently enhanced colonic propulsion in isolated guinea pig colonic segments (Foxx-Orenstein et al., 1998). Interestingly, i.p. administration of SoRI 9409 (3-30 mg/kg) produced only moderate (30%) inhibition of gastrointestinal motility in rats (T. Vanderah, unpublished observations).

In summary, SoRI 9409 represents a lead structure for developing nonpeptidic compounds that possess a µ-agonist/-antagonist profile. The potential therapeutic advantages of a µ-agonist/-antagonist may be further explored using this compound. In this report, for example, we demonstrated the lack of antinociceptive tolerance to repeated injections of SoRI 9409. Although the lack of antinociceptive activity of the compound in a thermal nociceptive assay may limit the compounds clinical utility in certain pain states, other pain states may be adequately suppressed. Current studies are attempting to modify the SoRI 9409 structure to enhance the antinociceptive efficacy of the compound at µ-opioid receptors, preserve the potent -antagonist properties of the compound, and eliminate -agonist activity. Furthermore, the partial µ-agonist/-antagonist profile may be of interest in the treatment of heroin, cocaine, or methamphetamine dependence (Reid et al., 1993, 1995; Strain et al., 1996b; O'Connor et al., 1998). In this regard, SoRI 9409 was able to attenuate opioid withdrawal in mice made acutely dependent to morphine. Future studies will address other potential benefits of the compound.