NMDA receptor antagonists and glycine site NMDA antagonists

Extracellular concentrations of excitatory amino acids increase substantially within cerebral tissue beds exposed to ischaemic conditions. This leads to excessive stimulation of N-methyl-D-aspartate (NMDA) receptors, a major cerebral excitatory neurotransmitter receptor that likely plays a critical role in the propagation of ischaemic injury in neurons. Pharmacological blockade of these receptors has proven to be an effective neuroprotective therapy by a number of animal models of central nervous system ischaemia. Clinical trials of these drugs were begun with high expectations for successful therapy of human stroke. These putative neuroprotective drugs included competitive or non-competitive inhibitors of the NMDA receptor itself, as well as inhibitors of a co-modulatory glycine site. Thus far, all clinical trials of NMDA antagonists have been unsuccessful in establishing benefit for human stroke.     

Peripherally acting NMDA receptor/glycineB site receptor antagonists inhibit morphine tolerance

The present study focused on the role of peripheral ionotropic N-methyl-D-aspartate (NMDA) receptors in the development of tolerance to morphine-induced antinociception. An initial experiment revealed that NMDA channel blocker memantine, and NMDA receptor/glycine(B) site antagonist MRZ 2/576 inhibited maximal electroshock-induced convulsions (MES) in female NMR mice with respective potency of 5.93 and 20.8 mg/kg, while other NMDA receptor/glycine(B) site antagonists MRZ 2/596 and MDL 105,519 were ineffective, supporting lack of CNS activity of the latter two agents. This observation was also supported by blood-brain barrier experiments in vitro. In male Swiss mice, morphine (10 mg/kg) given for 6 days twice a day (b.i.d.) produced tolerance to its antinociceptive effects in the tail-flick test. The NMDA receptor/glycine(B) site antagonists, MRZ 2/576 at 0.03, 0.1, 0.3 mg/kg and MRZ 2/596 at 0.1, 0.3, 3 and 10 mg/kg attenuated the development of morphine tolerance. Similarly, in male C57/Bl mice, morphine (10 mg/kg) given for 6 days b.i.d. produced tolerance to its antinociceptive effects in the tail-flick test. Like in Swiss mice, in C57/Bl mice morphine tolerance was attenuated by both MRZ 2/576 and MRZ 2/596. Another NMDA receptor/glycine(B) site receptor antagonist, MDL 105,519 (that very weakly penetrates to the central nervous system) also inhibited morphine tolerance at the dose of 1 but not 0.1 mg/kg. Moreover, both naloxone hydrochloride (5 and 50 mg/kg) and centrally inactive naloxone methiodide (50mg/kg) inhibited morphine tolerance suggesting the involvement of peripheral opioid receptors in this phenomenon. The present data suggest that blockade of NMDA receptor/glycine(B) sites in the periphery may attenuate tolerance to the antinociceptive effects of morphine.     

Sex-related differences in the enhancement of morphine antinociception by NMDA receptor antagonists in rats

The effect of the N-methyl-D-aspartate (NMDA) receptor antagonists dextromethorphan (DEX), ketamine (KET), and MK-801 on morphine (MOR)-induced antinociception has been investigated in male and female rats. DEX (7.5, 15, and 30 mg/kg), KET (0.75, 1.5, and 3 mg/kg), and MK-801 (0.075, 0.15, and 0.3 mg/kg) dose-dependently enhanced MOR-induced (3 mg/kg) analgesia in female rats. DEX and KET enhanced the peak effect, whereas MK-801 increased both magnitude and duration of analgesia. DEX also enhanced MOR-induced analgesia in male rats. However, the interaction was of less magnitude in male compared with female rats. The effects of KET and MK-801 on MOR-induced analgesia were negligible in male rats. A 3-mg/kg dose of MOR given alone produced greater analgesia in male than in female rats, but in the presence of NMDA antagonists, MOR elicited similar analgesic responses in both sexes.     

Effects of NMDA receptor antagonists on cocaine-conditioned motor activity in rats

NMDA receptor antagonists have been reported to affect learned behaviors conditioned with abused drugs, with the outcome dependent, in part, on the class of NMDA receptor antagonist used. The present study tested the ability of various site-selective NMDA receptor antagonists to modify cocaine-conditioned motor activity. Two procedures were used for independently assessing drug effects on spontaneous activity and expression of cocaine-conditioned behavior. In the conditioning experiments, rats were administered i.p. injections of cocaine (30 mg/kg) or saline paired with distinctive environments. Spontaneous horizontal activity was dose-dependently enhanced by dizocilpine (0.03-0.3 mg/kg) and memantine (1-30 mg/kg), but not by D-CPPene (3-(2-carboxypiperazin-4-yl)-1-propenyl-1-phosphonic acid; SDZ EAA 494; 1-10 mg/kg), ACEA-1021 (5-nitro-6,7-dichloro-1,4-dihydro-2, 3-quinoxalinedione; 3-56 mg/kg), or eliprodil (3-30 mg/kg). Higher doses of memantine, D-CPPene (1-10 mg/kg), eliprodil (3-30 mg/kg), or ACEA-1021 reduced vertical activity. Following five cocaine-environment pairings, rats displayed significant increases in motor activity when exposed to the cocaine-paired environment. The following antagonists were administered prior to the conditioning test: dizocilpine (MK-801; 0.03-0.1 mg/kg), memantine (1-10 mg/kg), D-CPPene (0.3-3 mg/kg), ACEA-1021 (3-10 mg/kg), and eliprodil (1-10 mg/kg). Of these, memantine, ACEA-1021 and, to the lesser degree, eliprodil attenuated expression of cocaine-conditioned motor activity at doses that did not significantly affect spontaneous motor activity. These results show that cocaine-conditioned behaviors can be selectively modulated by some, but not all, NMDA receptor antagonists.     

Analgesic effects of morphine and loperamide in the rat formalin test: interactions with NMDA receptor antagonists

To reveal peripheral components of opiate analgesia, effects of loperamide, opioid agonist which does not penetrate the blood-brain barrier, were examined in formalin and acute thermal pain tests in comparison with morphine. Formalin administration induces pain behaviour such licking/biting of injected paw expressed as two phases. The first phase is caused by C-fibre activation due to peripheral stimulation, the second phase attributed to ongoing input from peripheral site, leading to spinal hyperexcitability, which is dependent on N-methyl-D-aspartate (NMDA) receptor activation. Loperamide (3-10 mg/kg) and morphine (6 mg/kg) reduced formalin-induced nociceptive behaviours and these effects were reversed by naloxone methiodide (0.03-10 mg/kg), opioid receptor antagonist which poorly penetrates the blood-brain barrier. Loperamide action was enhanced only by centrally active NMDA receptor antagonists memantine (3 mg/kg) and CGP 37849 (3 mg/kg), but not by NMDA/glycineB receptor antagonists showing weak or no central nervous system (CNS) activity. Present results suggest that central NMDA receptor blockade may be necessary to enhance analgesia induced through peripheral opioid mechanisms in formalin-evoked nociception.     

Behavioral effects of NMDA receptor agonists and antagonists in combination with nitric oxide-related compounds

Responding of rats was maintained under a 120-response fixed ratio (FR) schedule of food delivery, and animals received individual and combined injections of N-methyl-D-aspartic acid (NMDA), phencyclidine hydrochloride, (+)-MK-801 hydrogen maleate (MK-801), (+/-)-2-amino-5-phosphonopentanoic acid (AP5), 7-chlorokynurenic acid (7CK), ifenprodil tartrate, N(G)-nitro-L-arginine methyl ester hydorchloride (L-NAME), 7-nitroindazole, aminoguanidine hemisulfate, L-arginine, molsidomine, sodium nitroprusside, and 8-(diethylamino)octyl 3,4,5-trimethoxybenzoate hydrochloride (TMB-8). Behavioral suppression after NMDA was completely and dose-dependently reversed by MK-801, phencyclidine, AP5, and aminoguanidine; partially and dose-dependently attenuated by molsidomine, ifenprodil, and 7CK; and not attenuated at all by L-NAME, 7-nitroindazole, or TMB-8. These findings suggested that behavioral suppression after NMDA was associated with nitric oxide from the inducible synthase. In a second series of experiments, comparable behavioral suppression by 0.1 mg/kg MK-801, but not 3 mg/kg phencyclidine, was attenuated by nitroprusside, molsidomine, and L-arginine, suggesting that suppressions from MK-801 and phencyclidine were mediated by different final common pathways, and that behavioral suppression from MK-801, but not phencyclidine, may be associated with Ca(2+)-dependent nitric oxide.     

Repeated spatial acquisition: effects of NMDA antagonists and morphine

Effects of morphine and 2 N-methyl-D-aspartate (NMDA) receptor antagonists, phencyclidine and LY235959, were studied using a within-subject, repeated-acquisition/performance procedure adapted to the Morris Swim Task. In the performance component, subjects swam to a hidden platform that was always in the same location in the pool. In the acquisition component, the platform was moved to a different place for each session. Baseline training produced rapid and direct swims to the platform in the performance component and steep within-session learning curves in the acquisition component. All 3 compounds increased swim distances, escape latencies, and slowed swim speed in a dose-dependent manner, but only morphine consistently produced selective impairments on acquisition. NMDA antagonists generally affected acquisition only at doses that also disrupted performance, although phencyclidine produced selective effects in some animals. These outcomes were different than those from studies of response chains in primates, suggesting that task and species variables may be important determinants of drug effects on acquisition.     

Antinociceptive activity of CP-101,606, an NMDA receptor NR2B subunit antagonist

1. The analgesic activity of CP-101,606, an NR2B subunit-selective N-methyl-D-aspartate (NMDA) receptor antagonist, was examined in carrageenan-induced hyperalgesia, capsaicin- and 4β-phorbol-12-myristate-13-acetate (PMA)-induced nociceptive tests in the rat.
2. CP-101,606 30 mg kg⁻¹, s.c., at 0.5 and 2.5 h after carrageenan challenge suppressed mechanical hyperalgesia without any apparant alternations in motor coordination or behaviour in the rat.
3. CP-101,606 also inhibited capsaicin- and PMA-induced nociceptive responses (licking behaviour) with ED₅₀ values of 7.5 and 5.7 mg kg⁻¹, s.c., respectively.
4. These results suggest that inhibition of the NR2B subunit of the NMDA receptor is effective in vivo at modulating nociception and hyperalgesia responses without causing the behavioural side effects often observed with currently available NMDA receptor antagonists.

     

Effect of low-affinity NMDA receptor antagonists on electrical activity in mouse cortical slices

The objective of this study was to investigate the effects of three low-affinity NMDA receptor antagonists, MRZ 2/279 (1-amino-1,3,3,5,5-pentamethyl-cyclohexane HCl), AR-R 15896AR ([+]-alpha-phenyl-2-pyridine-ethanamine diHCl) and dextromethorphan on epileptiform activity in vitro. Epileptiform discharges were elicited in DBA/2 mouse cortical slices by perfusion with Mg(2+)-free artificial cerebrospinal fluid. MRZ 2/279, AR-R 15896AR and dextromethorphan all reversibly decreased the frequency of the discharges in a concentration-dependent manner. The IC(50)'s for MRZ 2/279, AR-R 15896AR and dextromethorphan were 5.2, 10.8 and 55.9 microM, respectively. These low-affinity NMDA receptor antagonists may be proved to be clinically effective with fewer adverse effects than drugs with high-affinity for the NMDA receptor-operated channel.     

Antinociceptive effects of chronic administration of uncompetitive NMDA receptor antagonists in a rat model of diabetic neuropathic pain

Diabetic neuropathic pain remains an unmet clinical problem and is poorly relieved by conventional analgesics. N-methyl-d-aspartate (NMDA) receptors play an important role in central sensitization in neuropathic pain. Although NMDA antagonists are highly effective in reducing neuropathic pain, these agents cause severe side effects at therapeutic doses, which limit their clinical uses. Neramexane and memantine are uncompetitive NMDA antagonists with minimal side effects at therapeutic doses. Here we determined the antinociceptive effect of chronic administration of neramexane and compared its effect with that of memantine and gabapentin in a rat model of diabetic neuropathic pain. Mechanical hyperalgesia was measured with a noxious pressure stimulus, and tactile allodynia was assessed with von Frey filaments in diabetic rats induced by streptozotocin. Compared with vehicle-treated rats, treatment with neramexane (12.3, 24.6, and 49.2 mg/kg/day) for 2 weeks via an osmotic minipump produced dose-dependent and sustained effects on mechanical hyperalgesia and allodynia. Administration of memantine (20 mg/kg/day) or gabapentin (50 mg/kg/day) for 2 weeks also produced significant and persistent antinociceptive effects on mechanical hyperalgesia and allodynia. The magnitude of the antinociceptive effect produced by the intermediate and high doses of neramexane was comparable to that of gabapentin and memantine. The plasma level achieved by neramexane at 12.3, 24.6, and 49.2 mg/kg/day was 0.26 ± 0.04, 0.50 ± 0.05, and 1.21 ± 0.16 μM, respectively. These data suggest that neramexane at therapeutically relevant doses attenuates diabetic neuropathic pain. Our study provides valuable information about the therapeutic potential of chronic administration of neramexane and memantine for painful diabetic neuropathy.     

Pretreatment with morphine potentiates naloxone-conditioned place aversion in mice: effects of NMDA receptor antagonists

Acute pretreatment with opioid receptor agonists potentiates behavioral effects of opioid antagonists. This phenomenon was suggested to serve as an acute model of opioid dependence. Since antagonists acting at N-methyl-D-aspartate (NMDA) receptors were repeatedly shown to attenuate development, maintenance, and expression of opioid dependence, the present study evaluated the effects of competitive NMDA receptor antagonist, D-CPPene (SDZ EAA 494; 3-(2-carboxypiperazin-4-yl)-1-propenyl-1-phosphonic acid), and low-affinity channel blocker, 1-amino-3,5-dimethyl adamantane hydrochloride (memantine), on establishment of naloxone-conditioned place aversion in mice that were pre-exposed to morphine. Morphine (20 mg/kg) pretreatment significantly potentiated the ability of naloxone (0.01-0.3 mg/kg) to produce place aversion. The place aversion produced by naloxone (0.1 mg/kg) was attenuated by D-CPPene (1 and 3 mg/kg but not 0.1 or 0.3 mg/kg) when it was administered 3.5 h after morphine (0.5 h prior to conditioning trial with naloxone) but not 0.5 h prior to morphine. Memantine (1-10 mg/kg) had no effect under any treatment condition (0.5 h prior to morphine, simultaneously with morphine, 2 or 3.5 h after morphine). Thus, the ability of NMDA receptor antagonist to affect development and/or expression of morphine dependence may not be a good predictor of their effects on establishment of morphine-potentiated naloxone-conditioned place aversion.     

Competitive Gly/NMDA receptor antagonists

Glutamic acid (Glu) is the major excitatory neurotransmitter in the mammalian central nervous system (CNS) where it is involved in the physiological regulation of different processes. It has been well established that excessive endogenous Glu is associated with many acute and chronic neurodegenerative disorders such as cerebral ischemia, epilepsy, amiotrophic lateral sclerosis (ALS), Parkinson's and Alzheimer's diseases. In addition to the classical competitive glutamate receptor (GluR) antagonists, much effort has been directed toward the development of many different non-competitive antagonists of these receptors and, among them, compounds blocking the glycine site on the NMDA receptor complex (Gly/NMDA) have been widely investigated. Many Gly/NMDA receptor antagonists showed to be potential therapeutic agents in many neurological diseases such as stroke, epilepsy and neuropathic pain. Some of them, endowed also with favourable physicochemical properties and low secondary undesiderable effects, reached clinical trials.     

The behavioral pharmacology of NMDA receptor antagonists

There is considerable interest in the development of NMDA antagonists as potential therapeutic agents in the treatment of convulsant, neurodegenerative and anxiety disorders. Because the clinical use of phencyclidine (PCP) has been precluded by its psychotomimetic effects and abuse potential, there has been concern that other NMDA antagonists including those acting competitively might produce similar untoward effects. However, the studies in animals, reviewed here by Joyce Willetts, Robert Balster and David Leander, suggest that while there are certain similarities in the behavioral effects of PCP-like and competitive antagonists, there are also differences. These differences have implications for the development of NMDA antagonists with less likelihood for producing PCP-like side-effects.     

Genetic differences in the effects of competitive and non-competitive NMDA receptor antagonists on locomotor activity in mice

The effects of non-competitive (MK-801, phencyclidine, and ketamine) and competitive (CGP 39551, CGS 19755, and NPC 12626) N-methyl-d-aspartate (NMDA) receptor antagonists on locomotor activity in inbred CBA and C57, and in outbred NMRI mice were examined. Administration of the non-competitive NMDA antagonists produced a dose-dependent increase in well-coordinated locomotor activity at lower doses, followed by a bizarre behavioral syndrome (head weaving, body rolling, rotations, ataxia) after higher doses. The pharmacological profile of the competitive antagonists CGP 39551, CGS 19755, and NPC 12626 was more complex. CGP 39551 dose-dependently inhibited locomotor activity, whereas CGS 19755 and NPC 12626 displayed a biphasic action, that is low doses inhibited locomotor activity, whereas higher doses produced mild behavioral stimulation. The behavioral effects of NMDA antagonists appear to be genetically determined, since CBA animals were most sensitive to both noncompetitive and competitive antagonists, followed by NMRI and C57 animals. The differential effects of NMDA antagonists in various strains of mice suggest that the observed behavioral differences may be due to genetic differences in the NMDA/glutamate receptor channel complex.     

Clinically available NMDA receptor antagonists memantine and dextromethorphan reverse existing tolerance to the antinociceptive effects of morphine in mice

The tail-flick test was used to investigate the effects of chronic administration of the N-methyl-D-aspartate (NMDA) receptor antagonists, dextromethorphan, memantine and MRZ 2/579, on the development and reversal of morphine tolerance in mice in three separate experiments. Experiment 1 investigated the effects of NMDA receptor antagonists on the development of tolerance. Morphine (10 mg/kg for 6 days, twice daily) produced a 5.9-fold rightward shift of the cumulative dose-response curves. Co-administration of dextromethorphan, memantine or MRZ 2/579 between tests 1 and 2 dose-dependently (5-10 mg/kg) inhibited the development of morphine tolerance. In experiment 2, in which the effects on the reversal were investigated, morphine-tolerant mice were treated b.i.d. for an additional 6 days (between tests 2 and 3) with vehicle+vehicle, NMDA receptor antagonist+vehicle, vehicle+morphine or NMDA receptor antagonist+morphine. Morphine-tolerant mice treated with vehicle+vehicle remained morphine tolerant, whereas this residual morphine tolerance was inhibited by administration of all three NMDA antagonists (each 10 mg/kg). Morphine-tolerant mice receiving vehicle+morphine injections demonstrated an unchanged degree of antinociceptive tolerance. In these mice, the co-administration of memantine and MRZ 2/579, but not dextromethorphan, resulted in the reversal of morphine tolerance. In experiment 3, memantine and MRZ 2/579 (10 mg/kg) inhibited the acute antinociceptive effect of morphine, but dextromethorphan did not. These data indicate that low-affinity, clinically available and/or therapeutically promising NMDA receptor antagonists may be used to inhibit ongoing morphine tolerance.     

Antinociceptive activity of NK1 receptor antagonists: non-specific effects of racemic RP67580.

1. Release of substance P in the dorsal horn is considered a primary event in the perception of pain. The profile of racemic RP67580, a non-peptide antagonist at the NK1 (substance P) receptor, was examined in a range of antinociception tests on rodents. 2. At doses up to 30 mg kg-1, s.c. racemic RP67580 exhibited antinociceptive activity in writhing and formalin paw tests in mice and gerbils. Acetic acid induced writhing and the licking response to formalin were reduced to 40-50% of the level observed in vehicle-treated animals (P < 0.05). However, this agent was not active in mouse tail flick, rat paw pressure or rat and guinea-pig formalin paw tests. 3. Like racemic RP67580, the calcium channel blockers nifedipine (30 mg kg-1, i.p.) and verapamil (10 or 20 mg kg-1, s.c.) inhibited the response to formalin by approximately 60% in gerbils (P < 0.05 compared with vehicle-treated animals). 4. Evidence for calcium channel antagonist activity of RP67580 was obtained in vitro. Racemic RP67580 inhibited calcium entry into depolarized strips of guinea-pig ileum longitudinal muscle myenteric plexus (apparent KB = 587 +/- 115 nM), inhibited [3H]-diltiazem binding to rabbit skeletal membranes (IC50 = 298 nM) and depressed high threshold calcium currents in neurones cultured from rat cortex (10% inhibition at 10 microM). 5. These findings indicate that the acute antinociceptive effects of RP67580 may not be attributable to a specific interaction with NK1 receptors and may be mediated via calcium channel blockade.     

Anticonvulsant activity of competitive antagonists of NMDA receptor in genetically epilepsy-prone rats.

The potency of some competitive antagonists of N-methyl-D-aspartate (NMDA) to antagonize audiogenic seizures was evaluated in genetically epilepsy-prone rats following intraperitoneal or oral administration. The anticonvulsant effects were evaluated on seizures evoked by auditory stimulation (109 dB, 12-16 kHz) in animals placed singly under a perspex dome. Sixty and hundred-twenty minutes after intraperitoneal or 120 min after oral administration all compounds showed antiseizure activity with ED50 against clonus ranging from 11.6 to 384 mumol/kg after intraperitoneal and higher doses after oral administration. DL-(E)-2-Amino-4-methyl-5-phosphono-3-pentenoic acid (CGP 37849) and its carboxyethylester (CGP 39551) and 3-((+/-)-2-carboxypiperazin-4-yl)-1-propenyl-1-phosphonic acid (CPPene), were the most potent compounds when administered intraperitoneally and orally. 3-((+/-)-2-Carboxypiperazin-4-yl)-propyl-1-phosphonic acid ((-)-CPP) and (+)-CPP showed minor potency as anticonvulsants and 2-amino-7-phosphonoheptanoic acid (2AP7) was the least potent. Following oral treatment, the duration of action of CPPene was about 8 h, while CGP 39551 still protected against audiogenic seizures after 16 h. All compounds were anticonvulsant at doses below those at which overt behavioural side-effects were apparent. CPPene and (+/-)-CPP showed the best therapeutic index among the NMDA receptor antagonists studied. Since some of these compounds showed anticonvulsant properties after oral administration, potential use of these or other selective NMDA antagonists for antiepileptic therapy in man is suggested.     

Antinociceptive activity of clonidine and its potentiation of morphine analgesia.

The activity of clonidine and its interaction with morphine was assessed in the mouse tail flick assay. In this assay, clonidine was found to be 10 times more potent than morphine. Clonidine potentiated morphine antinociceptive activity approximately five-fold and morphine potentiated clonidine activity four-fold. Clonidine's agonstic activity was not reversed by naloxone hydrochloride (10 mg/kg) while the potentiating effect of clonidine by morphine was. Tolerance to the antinociceptive effect of morphine was observed in morphine pellet-implanted mice but no cross tolerance was observed for clonidine. These data indicate that clonidine-induced analgesia is not a result of an interaction at morphine receptors; but rather, common pathway(s) are present which appear to complement the agonistic interaction of each.     

Antinociceptive activity of propionyl esters of morphine: a reevaluation

The antinociceptive activity of the propionyl homologues of 3-O- and 6-O-acetyl- and 3,6-O-diacetylmorphine was re-investigated using materials of unequivocally established structure. Testing was in male Wistar rats at 60 min following subcutaneous administration by the rat tail-flick method. Results indicate that the antinociceptive activity of 3-O-propionylmorphine was similar to that of 3-O-acetylmorphine. 6-O-Propionylmorphine and 3,6-O-dipropionylmorphine had similar antinociceptive activity and, like 6-O-acetylmorphine, 6-O-propionylmorphine may be the pharmacologically active principle responsible for the antinociceptive activity of its disubstituted homologue.     

Mouse locomotor activity: Effect of morphine,narcotic antagonists,and the interaction of morphine and narcotic antagonists

Morphine-induced locomotor activity in mice has been investigated but there appear to be only a small number of reports on the effects of narcotic antagonists on this increased activity. Also, it is well known that some narcotic antagonists can act as both morphine antagonists and analgesics in nociceptive assays, but there have been few demonstrations of this stimulant and antagonistic action with locomotor activity studies. In this study, six compounds (cyclazocine levallorphan, diprenorphine (M-5050), nalorphine, naloxone, and naltrexone) have been investigated in regard to their activity as antagonists of morphine-induced locomotor activity and in regard to their ability to stimulate locomotor activity themselves. All six compounds antagonized the effect of morphine, but only cyclazocine and levallorphan produced any significant stimulation of locomotor activity by themselves at the doses tested. This study indicates that changes in mouse locomotor activity can be used successfully to monitor the interaction between morphine and narcotic antagonists and that locomotor activity studies can also be used to study the stimulant (agonist) properties of narcotic antagonists.Supported in part by NIMH Contract No. HSM-42-72-167.The research was carried out with animals maintained in animal care facilities fully accredited by the American Association for Accreditation of Laboratory Animal Care.