Comparison of the effects of specific and nonspecific inhibition of nitric oxide synthase on morphine analgesia,tolerance and dependence in mice

The effects of L-Canavanine, a selective inducible nitric oxide synthase (NOS) inhibitor and N(G)-nitro-L-arginine methyl ester (L-NAME), a nonselective NOS inhibitor, on pain threshold and morphine induced analgesia, tolerance and dependence in mice were investigated and compared. Morphine was administered by subcutaneous implantation of a pellet containing 40 mg free base and withdrawal was precipitated by intraperitoneal (i.p.) injection of naloxone (2 mg/kg). L-Canavanine (200 mg/kg, i.p.) did not affect the pain threshold, morphine-induced analgesia and the induction and expression phases of morphine tolerance and dependence. L-NAME (20 mg/kg, i.p.) significantly (p < 0.05) enhanced the pain threshold, potentiated morphine-induced analgesia and attenuated the expression phase of morphine dependence which has been characterized by withdrawal signs and body weight loss, but did not modify the induction phase of morphine tolerance and dependence. It is concluded that constitutive NOS isoforms which were inhibited by L-NAME may be involved specifically in the mechanisms of morphine induced analgesia, tolerance and dependence.     

Potentiation of morphine analgesia by BQ123, an endothelin antagonist

Several neurotransmitter mechanisms have been proposed to play a role in the actions of morphine. The present study is the first to provide evidence that central endothelin (ET) mechanisms are involved in the modulation of pharmacological actions of morphine. The effect of intracerebroventricular (i.c.v.) administration of endothelin-A (ET(A)) antagonist, BQ123, on morphine-induced analgesia, hyperthermia, and catalepsy was determined in the rat. Morphine produced a significant increase in tail-flick latency as compared to control group. Pretreatment with BQ123 significantly potentiated the effect and duration of morphine (2 and 8 mg/kg, s.c.)-induced analgesia as compared to vehicle-pretreated control rats. The hyperthermic effect of morphine was not only significantly greater in BQ123-pretreated rats but also lasted for more than 6 h. ET antagonist, BQ123, did not affect the pharmacological effect of morphine on cataleptic behavior. These studies demonstrate that BQ123, a specific ET(A) receptor antagonist, significantly potentiated morphine-induced analgesia and hyperthermia in rats without affecting morphine-induced cataleptic behavior. [(3)H]-Naloxone binding was carried out to determine the possibility of BQ123 acting on opiate receptors. It was found that morphine could displace [(3)H]-naloxone but BQ123 did not affect [(3)H]-naloxone binding even at 1,000 nM concentration. Therefore, it can be concluded that BQ123 does not act on opioid receptors. This is the first report suggesting that an ET(A) antagonist, BQ123, significantly potentiates the analgesic effect of morphine, possibly through a nonopioid mechanism.     

Tolerance to morphine-induced analgesia in mice: Magnetic fields function as environmental specific cues and reduce tolerance development

Mice receiving daily injection of morphine (10 mg/kg) developed tolerance to morphine-induced analgesia, such that after 5–7 days of treatment their thermal response (paw licking) latencies in the hot plate test were indistinguishable from those of control animals. Exposure to a rotating magnetic field for thirty minutes before the daily morphine administrations significantly reduced the development of tolerance. These magnetic exposure also significantly increased over 7–10 days the basal nociceptive thresholds and paw licking response latencies of saline treated mice. Control and sham exposed mice that were fully tolerant to the analgesic effects of morphine failed to show any tolerance to morphine-induced analgesia when exposed to the magnetic stimuli prior to injection. Likewise, the partial tolerance to morphine shown by mice exposed to the rotating magnetic field pre-injection environmental cues was eliminated when control or sham pre-injection cues lacking the magnetic stimuli were provided. In all cases tolerance to morphine-induced analgesia was evident in the subsequent re-test with the original cues. These results indicate that magnetic field exposure can reduce the development of tolerance to the analgesic effects of morphine. They also show that magnetic stimuli function as significant environmental cues for the development of tolerance to morphine-induced analgesia. This suggests that magnetic stimuli affect both the associative (classical conditioning) and non-associative (physiological, pharmacological) mechanisms involved in the development of opiate tolerance.     

Effects of prolyl-leucyl-glycinamide and cyclo(leucyl-glycine) on morphine-induced antinociception and brain μ, δ and κ opiate receptors

The effects of prolyl-leucyl-glycinamide and cyclo (leucyl-glycine) on morphine-induced antinociception in mice and on $\text{in vitro}$ binding of ³H-ligands for opiate receptor subtypes (μ, δ and κ) the mouse brain homogenate were determined. Subcutaneous administration of either of the above peptides (1, 2, and 4 mg/kg) 10 min prior to the injection of morphine did not affect morphine-induced antinociception as evidenced by the identical ED₅₀ values of morphine in vehicle and peptide treated groups. The binding of ³H-dihydromorphine and ³H-naloxone ( μ receptors), ³HDAla²DLeu⁵-enkephalin (δ receptors), and ³H-ethylketocyclazocine (κ receptors) to opiate receptors in the mouse brain homogenate was also unaffected by both the peptides over a large concentration range. It is concluded that these peptides do not interact with brain opiate receptors.     

Circadian-Rhythm-dependent Effects of L-NG-Nitroarginine Methyl Ester (L-Name) on Morphine-Induced Analgesia

Circadian changes in the interactions between L-N^G-nitroarginine methyl ester (L-NAME), a nitric oxide synthase (NOS) inhibitor, and morphine-induced antinociception were investigated by the mouse hot-plate test. Born the basal pain sensitivity and morphine-induced analgesia undergo significant 24h variations. L-NAME (40 mg/kg, ip) alone did not show any antinociceptive activity, but potentiated morphine-induced analgesia when combined with morphine at all injection times. In terms of percentage absolute potentiation (%AP), L-NAME dramatically augmented the analgesic effect of morphine in the late dark period at 19 hours after lights on (HALO). It is concluded that nitric oxide (NO) is involved in the modulation of the analgesic effect of morphine; thus, the L-NAME and morphine combination might be beneficial in alleviating pain.     

The effect of an endogenous compound 1-methyl-1,2,3,4,-tetrahydroisoquinoline on morphine-induced analgesia, dependence and neurochemical changes in dopamine metabolism in rat brain structures.

1,2,3,4-Tetrahydroisoquinolines, among them the most interesting neuroprotective substance, an inhibitor of MAO, 1-methyl-1,2,3,4-tetrahydroisoquinoline (1MeTIQ), are endogenous compounds present in the central nervous system of mammals and humans. In this study, we investigated the effect of 1MeTIQ on morphine-induced analgesia, tolerance and abstinence syndrome as well as its effect on morphine-induced changes in dopamine metabolism in rat brain structures (nucleus accumbens, striatum, substantia nigra) using HPLC methodology. The experiments were carried out on male Wistar rats. Morphine analgesia was measured in the "hot-plate" test. To induce tolerance, morphine was given chronically (20 mg/kg i.p.) alone or following 1MeTIQ (50 mg/kg i.p.) injection. The development of dependence was assessed in the naloxone (2 mg/kg i.p.) precipitation test, after 10 days of morphine administration. The behavioral studies have shown that an endogenous compound, 1MeTIQ produced strong potentiation of morphine analgesia, prevented the development of morphine tolerance and inhibited expression of morphine abstinence syndrome in morphine-dependent rats. In neurochemical studies, we have demonstrated that 1MeTIQ antagonized morphine-induced changes in dopamine metabolism observed in rat brain structures. The main finding of this study was demonstration for the first time of an anti-abuse effect of an endogenous compound, 1MeTIQ, and its efficiency in counteracting morphine-induced addiction in the way useful from clinical point of view. The obtained results suggested a possibility of clinical application of 1MeTIQ in morphine addiction.     

Restoration of morphine-induced alterations in rat submandibular gland function by N-methyl-D-aspartate agonist

The effects of morphine, 1-aminocyclobutane-cis-1,3-dicarboxylic (ACBD; NMDA agonist) and 3-((R)2-carboxypiperazin-4-yl)-propyl-l-phosphoric acid (CPP; NMDA antagonist) and their concurrent therapy on rat submandibular secretory function were studied. Pure submandibular saliva was collected intraorally by micro polyethylene cannula from anaesthetized rats using pilocarpine as secretagogue. Intraperitoneal injection of morphine (6 mg/kg) induced significant inhibition of salivary flow rate, total protein, calcium, and TGF-beta1 concentrations. Administration of ACBD (10 mg/kg) and CPP (10 mg/kg) alone did not influence secretion of submandibular glands. In combination therapy, coadministration of CPP with morphine did not influence morphine-induced changes in salivary function while ABCD could restore all morphine-induced changes. In combination treatment, ACBD prevented morphine-induced reduction of flow rate, total protein, calcium, and TGF-beta1 and reached control levels. It is concluded that morphine-induced alterations in submandibular gland function are mediated through NMDA receptors.     

Glucagon is an antagonist of morphine bradycardia and antinociception

Glucagon and its receptors have been identified within the mammalian brain, and their anatomical distribution correlates well with the distribution of opioid peptides and their receptors. To evaluate possible physiological interactions between these two peptidergic systems, we examined the effects of glucagon on two opioid responses - bradycardia and antinociception. Glucagon administered either intravenously (iv) (100-1000 micrograms/kg) or intracerebroventricularly (icv) (5 micrograms) significantly attenuated morphine-induced (200 micrograms/kg, iv) bradycardia without producing any alterations in cardiovascular parameters when given alone. Furthermore, glucagon did not antagonize the bradycardia produced by phenyldiguanide (10 micrograms/kg, iv), a non-opioid substance. Peripheral (1 mg/kg, iv) and central (5 micrograms, icv) glucagon pretreatment antagonized morphine-induced (7.5 mg/kg, intraperitoneal) antinociception by 67% and 86%, respectively, at 30 minutes (as determined by the hot plate test). Glucagon treatment alone at these doses did not alter baseline response latencies. In both cases, central injections of glucagon were more effective than iv injections in antagonizing morphine's effects. These findings demonstrate a central action for glucagon and provide the first evidence that this neuropeptide may function as an endogenous antagonist of opioid actions.     

The influence of NMDA receptor agonist and antagonist on morphine state-dependent memory of passive avoidance in mice

The measurement of step-down latency in passive avoidance has been used to study memory in laboratory animals. The pre-training injection of 5 mg/kg morphine impaired memory, which was restored when 24 h later the same dose of the drug was administered. To explore the possible involvement of NMDA modulators on morphine-induced memory impairment, we have investigated the effects of intracerebroventricular (i.c.v.) administration of NMDA and the competitive NMDA antagonist, DL-AP5, on morphine-induced memory impairment or recall, on the test day. Morphine (5 mg/kg, s.c.) was administered 30 min before training to induce impairment of memory and 24 h later, 30 min before test to improve it. Pre-test administration of NMDA (0.00001, 0.0001 and 0.001 microg/mouse, i.c.v.) did not alter the retention latency compared to the saline-treated animals. But restored the memory impairment induced by pre-training morphine (5 mg/kg, s.c.). Pre-test administration of DL-AP5 (1, 3.2 and 10 microg/mouse, i.c.v.) by itself decreased the retention latencies. The same doses of DL-AP5 increased pre-training morphine-induced memory impairment. Co-administration of NMDA (0.0001 and 0.001 microg/mouse, i.c.v.) and morphine (5 mg/kg, s.c.) on the test day increased morphine memory improvement. Conversely, DL-AP5 (1, 3.2 and 10 microg/mouse, i.c.v.) inhibited morphine-induced memory recall. It is concluded that NMDA receptors may be involved, at least in part, in morphine state-dependent learning in mice.     

Is there an opiate receptor in the snail Megalobulimus sanctipauli? Action of morphine and naloxone

The effects of morphine sulfate (10, 15 and 20 mg/kg) or saline control (5 mg/kg) on the latency of the anterior body-lifting response to heat (avoidance response) were determined in four groups of snails Megalobulimus sanctipauli (n = 6) individually placed on a metal plate mounted on the surface of a water bath at 52 ± 1°C. The effects of pre-treatment with naloxone hydrochloride (5 mg/kg) or saline (2.5 ml/kg) control on the responses to morphine (15 mg/kg) were determined in two different groups of animals (n = 6). Administration of morphine resulted in an increase in the avoidance behavior latency with maximum effects occuring at 15 mg/kg, 10–15 min after injection. The effects of morphine disappeared within 90–120 min. Saline treatment had no detectable effects on the latency of the response to an aversive stimulus. Naloxone significantly blocked (P < 0.05, Student paired t-test) the increase in avoidance behavior latency. The present results indicate that: 1. morphine has an antinociceptive effect on the response of Megalobulimus sanctipauli to an aversive thermal stimulus; and 2. the morphine-induced “analgesia” may be caused by the stimulation of μ opiate receptors.     

Role of adrenals in morphine-induced hyperthermia in restrained rats

Role of adrenals in morphine-induced hyperthermia was studied in normal, neurotransmitter antagonist-pretreated, chemical-sympathectomized, adrenalectomized or adrenal-demedullated rats. In restrained female rats, 5 mg/kg morphine produced hyperthermia whereas 20 mg/kg and 40 mg/kg produced hypothermia. Pretreatment with either phenoxybenzamine, propranolol, pentolinium or scopolamine inhibited the hyperthermia. After adrenalectomy, neither 5 mg/kg nor chronic administration of 20 mg/kg morphine produced previously demonstrated hyperthermia. After adrenal-demedullation, a dose of 5 mg/kg morphine also did not produce hyperthermia. In contrast to female rats, restrained male rats showed no significant effect on body temperature after 5 mg/kg morphine, requiring 20 mg/kg and 40 mg/kg morphine to produce hyperthermia. In adrenalectomized male rats, 20 mg/kg morphine did not produce the usual hyperthermia. The results suggest that male rats are more resistant to the hyperthermic effects of morphine than female rats and that in the rat, the adrenals, likely the medulla, play an important role in morphine-induced hyperthermia.     

Effects of systemically administered monoamine reuptake blocking agents on patterns of buspirone-induced analgesia in rats

We have recently shown the serotonin 5-HT1A receptor agonist buspirone to produce analgesia in several pain tests in rats. As a 5-HT1A agonist, buspirone may change serotonergic (5-HT) tone to alter the balance of central monoaminergic (MA) systems that function in analgesia. MA-reuptake blocking drugs have been shown to produce analgesia, both when given alone and in combination with a variety of other agents, presumably via their action upon MA neurochemistry. The present study was undertaken to examine the effect of systemic administration of the 5-HT-reuptake blocker amitriptyline (AMI: 10 mg/kg), NE-reuptake blocker desipramine (DMI: 10 mg/kg) or DA-reuptake blocker GBR-12909 (7.5 mg/kg) on patterns of analgesia produced by buspirone (1-5 mg/kg) in thermal and mechanical pain tests in rats. Neither reuptake blocking agents or buspirone, when administered alone or in combination, produced overt changes in motor activity at the doses tested. AMI alone was not analgesic in either thermal or mechanical pain tests. In both assays, AMI reduced the analgesic action of buspirone, with greater effects seen in the thermal test. When administered alone, DMI produced significant analgesia against thermal and mechanical pain. DMI significantly attenuated the analgesic action of all doses of buspirone in both pain tests. Alone, GBR-12909 did not affect nociception in thermal or mechanical tests. GBR-12909 decreased buspirone-induced analgesia at all doses in the thermal test, while having no effect on buspirone-induced analgesia against mechanical pain. These results demonstrate that facilitation of 5-HT, NE and DA function with reuptake blocking drugs did not enhance the analgesic action of buspirone. These data indicate against the adjuvant use of reuptake blocking compounds and buspirone as analgesics. Furthermore, such findings may suggest other possible non-MA substrates of buspirone-induced analgesia.     

Attenuation of morphine dependence and withdrawal in rats by venlafaxine, a serotonin and noradrenaline reuptake inhibitor.

The effects of venlafaxine, a novel serotonin and adrenaline reuptake inhibitor, on the morphine withdrawal and activation of morphine conditioned place preference (CPP), were investigated in rats. Our results showed that the most morphine withdrawal signs, including jumping, writhing, shakes, exploring, lacrimation, piloerection, irritability, and diarrhea, were attenuated by pretreatment with 10 or 20 mg/kg venlafaxine. To investigate the effects of venlafaxine on relapse to opiate dependence, the morphine CPP was used and a dopamine D2 antagonist sulpiride was selected as a control drug. The morphine CPP disappeared following a 28-day drug-free period and appeared again after given a single injection of 1 mg/kg morphine. Acute treatment with sulpiride (25 or 50 mg/kg, i.p.) 30 min prior to 1 mg/kg morphine injection significantly blocked the reacquisition of CPP, while venlafaxine (10 or 20 mg/kg, i.p.) did not show significant effect. However, chronic treatment with venlafaxine (5 or 10 mg/kg, i.p. twice, daily, for seven consecutive days) significantly attenuated the reacquisition of morphine CPP, whereas chronic treatment with sulpiride (10 or 20 mg/kg, i.p.) have no significant effect. Our results demonstrated for the first time that venlafaxine strongly attenuates morphine withdrawal and morphine-induced reaquisition of     

Analgesic properties of a systemically-administered synthetic dipeptide of 5-hydroxytryptophan

Synthetic peptides of 5-hydroxytryptophan (5-HTP), including N-acetyl-5-HTP-5-HTP amide (5-HTP-ACETYL-DP), specifically inhibit the binding of serotonin to serotonin binding protein. 5-HTP-ACETYL-DP also produces a long-lasting, opiate-sensitive analgesia following central, but not systemic administration. The present study evaluated an apolar derivative of 5-HTP dipeptide, N-hexanoyl-5-HTP-5-HTP amide (5-HTP-HEX-DP), for its analgesic properties in rats following systemic administration. 5-HTP-HEX-DP (5–50 mg/kg) significantly increased jump thresholds in a dose-dependent manner with peak analgesia occurring at 2.5 hr after injection, and lasting up to 5 hr. In the tail-flick assay, 5-HTP-HEX-DP (20 mg/kg) produced a significant antinociceptive effect at 1 hr post-injection using both high and low intensity levels of radiant heat. While 5-HTP-HEX-DP and morphine each elicited analgesia following acute administration, chronic (14 days) incremental dosing with 5-HTP-HEX-DP or morphine resulted in persistent analgesia in 5-HTP-HEX-DP-treated animals, and a loss of analgesia in morphine-treated rats. Thus, significant tolerance to morphine, but not 5-HTP-HEX-DP analgesia developed using this protocol. Hence, 5-HTP-HEX-DP is a systemically-active analgesic which fails to develop tolerance when administered daily over 14 days.     

Naltrexone modulates growth in infant rats

Naltrexone, a potent opiate antagonist, had both stimulatory and inhibitory effects on somatic growth in preweaning rats depending on dose. Daily injections of 50 mg/kg naltrexone, which blocked morphine-induced analgesia for 24 hr/day, resulted in increased body and organ weights, and acceleration in the appearance of physical characteristics and maturation of spontaneous motor activity. Naltrexone in a dosage of 1 mg/kg, which blocked morphine-induced analgesia for 4 hr/day, had the opposite effects. These results show that naltrexone can modulate growth, and suggest a role for the endorphins and opiate receptors in developmental events.     

Modification of Morphine-induced Hyperlocomotion and Antinociception in Mice by Clorgyline,a Monoamine Oxidase-A Inhibitor

We evaluated the effects of pretreatment with clorgyline, an irreversible monoamine oxidase (MAO)-A inhibitor, on morphine-induced hyperlocomotion and antinociception. A single administration of morphine (30 mg/kg, i.p.) to male ICR mice induced a hyperlocomotion. ANOVA analysis revealed the statistical significance of the morphine effect on horizontal locomotion and of the clorgyline pretreatment × morphine interaction effect, but not of the effect of clorgyline pretreatment. The initial (5 min after challenge) phase of morphine actions vs. saline challenge appeared as if morphine had a strong inhibitory effect on locomotor activity in combination with different doses of clorgyline. The mice administered with morphine in combination of clorgyline (1 and 10 mg/kg) did not show any stereotypic behaviors. Clorgyline at a dose of 0.1 mg/kg but not other doses tested significantly potentiated morphine-induced antinociception evaluated by tail flick but not hot plate test. During the measurements of locomotor activity and antinociception, clorgyline at doses of 1 and 10 mg/kg significantly inhibited monoamine metabolism through MAO. These results suggest that clorgyline showed an inhibitory effect on morphine-induced hyperlocomotion, but not antinociception, through MAO inhibition. There is not a possibility that clorgyline pretreatment enhanced morphine action on motor activity, resulting in the abnormal behavior from hyperlocomotion to stereotypic movements.     

The bombesin/gastrin releasing peptide receptor antagonist RC-3095 blocks apomorphine but not MK-801-induced stereotypy in mice

Bombesin (BN)-like peptides might be involved in the pathogenesis of neuropsychiatric disorders such as schizophrenia. Stereotyped behaviors induced by the dopamine receptor agonist apomorphine or the N-methyl-D-aspartate glutamate receptor antagonist dizocilpine (MK-801) in rodents have been proposed as animal models of schizophrenic psychosis. In the present study we evaluated the effects of the BN/gastrin-releasing peptide receptor (GRP) antagonist (D-Tpi6, Leu13 psi[CH2NH]-Leu14) bombesin (6-14) (RC-3095) on apomorphine and MK-801-induced stereotyped behavior in mice. An intraperitoneal (i.p.) injection of RC-3095 (1.0, 10.0 or 100.0 mg/kg) blocked apomorphine-induced stereotypy. The inhibitory effect of RC-3095 on apomorhine-induced stereotypy was similar to that induced by haloperidol (0.5 mg/kg). RC-3095 did not affect stereotyped behavior induced by MK-801 (0.5 mg/kg). The results provide the first evidence that BN/GRP receptor antagonism blocks stereotyped behavior induced by a dopamine agonist. Together with previous evidence, the present study indicates that the BN/GRP receptor can be considered a drug target in the investigation of potential new agents for treating neuropsychiatric disorders.     

Soman and sarin induce a long-lasting naloxone-reversible analgesia in mice

Soman (50 micrograms/kg) and sarin (120 micrograms/kg), potent organophosphate anticholinesterase agents, produced an analgesic response in the mouse hotplate latency test. Naloxone antagonized but did not completely reverse the soman- and sarin-induced analgesia, whereas atropine antagonized completely the soman-and sarin-induced analgesia. Soman poisoning did not potentiate morphine-induced analgesia. It was simply an additive response. In survivors of soman (287 micrograms/kg) poisoning, the analgesia was more pronounced and was still apparent 96 hr after administration. This analgesia was completely antagonized by naloxone. Similar results were found in survivors of sarin (510 micrograms/kg) poisoning. The organophosphate-induced analgesia was not due to physical incapacitation as evidenced by performance on the accelerating rotorod. It is suggested that the organophosphate-induced analgesia is due to a combination of an increased concentration of acetylcholine due to inhibition of acetylcholinesterase combined with a reduced destruction of endogenous opioid-like substances due to organophosphate inhibition of proteases.     

Ethanol-induced analgesia

The effect of ethanol (ET) on nociceptive sensitivity was evaluated using a new tail deflection response (TDR) method. The IP injection of ET (0.5-1.5 g/kg) produced rapid dose-dependent analgesia. Near maximal effect (97% decrease in TDR) was produced with the 1.5 g/kg dose of ET ten minutes after injection. At ninety minutes post-injection there was still significant analgesia. Depression of ET-induced nociceptive sensitivity was partially reversed by a 1 mg/kg dose of naloxone. On the other hand, morphine (0.5 or 5.0 mg/kg IP) did not modify ET-induced analgesia, while 3.0 minutes of cold water swim (known to produce non-opioid mediated analgesia) potentiated ET-induced analgesic effect. The 0.5 g/kg dose of ET by itself did not depress motor activity in an open field test, but prevented partially the depression in motor activity produced by cold water swim (CWS). Thus the potentiation by ET of the depression of the TDR produced by CWS cannot be ascribed to the depressant effects of ET on motor activity.     

Cimetidine does not change the effect of Tyr-MIF-1 (Tyr-Pro-Leu-Gly-NH2) on the opiate form of footshock-induced analgesia.

It has been demonstrated that cimetidine blocks the effect of naloxone on footshock-induced analgesia. To study the effect of cimetidine on the antiopiate properties of an endogenous peptide Tyr-MIF-1, the opiate form of intermittent footshock-induced analgesia was elicited in the rat. The nociceptive responses were determined using the hot-plate test (52.5 degrees C). Intraperitoneal pretreatment with cimetidine (100 mg/kg) or chlorpheniramine maleate (20 mg/kg) did not affect the footshock-induced analgesia, and did not change the antagonizing effect of Tyr-MIF-1 (0.2 mg/kg) on this model of antinociception. It is concluded that cimetidine and chlorpheniramine maleate do not change the antagonizing effect of Tyr-MIF-1 on the opiate form of intermittent footshock-induced analgesia.