Neurotransmitter basis of the behavioral effects of hallucinogens

Animal models of serotonergic psychedelics

The serotonin 5-HT(2A) receptor is the major target of psychedelic drugs such as lysergic acid diethylamide (LSD), mescaline, and psilocybin. Serotonergic psychedelics induce profound effects on cognition, emotion, and sensory processing that often seem uniquely human. This raises questions about the validity of animal models of psychedelic drug action. Nonetheless, recent findings suggest behavioral abnormalities elicited by psychedelics in rodents that predict such effects in humans. Here we review the behavioral effects induced by psychedelic drugs in rodent models, discuss the translational potential of these findings, and define areas where further research is needed to better understand the molecular mechanisms and neuronal circuits underlying their neuropsychological effects.

Disruption of FR-40 by 5-HT agonists. I. Effects of chronic imipramine or trazodone

Doses of LSD, quipazine, 8-OHDPAT and TFMPP were established that prominently disrupted FR-40 operant response pattern in two groups of rats. Subsequently, one group received daily intraperitoneal (IP) injections of imipramine, 2.5 mg/kg, for 4 weeks, then 10 mg/kg for 2 additional weeks. The second group received 5 mg/kg/day, IP, of trazodone for the first 4 weeks, then 20 mg/kg/day for the next two weeks. For these periods and the 3 weeks after discontinuing the chronic drug treatments (washout), test doses of the 4 agonists were evaluated twice weekly in random order for their effects to decrease FR-40 reinforcements and increase pauses. No consistent, systematic changes in sensitivity to the agonist effects on FR-40 behavior were observed during chronic drug treatments, although significant effects were at times observed. However, during the washout period in the imipramine group, both LSD and 8-OHDPAT effects on reinforcements were reversed to baseline levels. The effect of 5-OHDPAT on pauses during washout in this group was also attenuated. During washout in the trazodone group, the 8-OHDPAT-induced pausing and loss of reinforcements was reduced so as to be not significantly different from baseline values. Previous studies have demonstrated antagonism of LSD- and quipazine-induced disruption of FR-40 by pretreating with the 5-HT2-selective antagonist pirenperone (28). Since chronic antidepressants down-regulate brain 5-HT2 binding sites, the effects of LSD and quipazine were expected to be attenuated, which was not the case.(ABSTRACT TRUNCATED AT 250 WORDS)

Disruption of FR-40 by 5-HT agonists. II. Effects of chronic phenelzine or isocarboxazid

Effects of chronic treatment with the monoamine oxidase inhibitors phenelzine and isocarboxazid on disruption of FR-40 operant responses by 5-HT agonists have been studied. Three groups of rats that were trained in the FR-40 operant schedule showed marked disruption by 0.1 mg/kg IP lysergic acid diethylamide (LSD), 2 mg/kg IP quipazine (Q), 0.05 mg/kg SC 8-hydroxy-2-(di-n-propylamino)tetralin (8-OHDPAT), and 1 mg/kg SC (m-trifluoromethyl-phenyl)piperazine (TFMPP), administered twice weekly in random order. Subsequently, one group received daily IP injection of phenelzine (5 and 10 mg/kg), the second group received 5 mg/kg IP of isocarboxazid, and the third group received vehicle (0.5% methyl cellulose) for 24 days (Period 1 and Period 2). For these periods and 12 days after discontinuing the MAOI treatments (Washout Period), test doses of 5-HT agonists were evaluated for their effects to decrease reinforcements (R) and increase pauses (P). No change in sensitivity to the LSD, Q and TFMPP effects on FR-40 behavior was observed in the vehicle-treated group. However, an attenuated effect of 8-OHDPAT was found in this group. In phenelzine- and isocarboxazid-treated rats the disruption of FR-40 responses by LSD and 8-OHDPAT were significantly reduced during Period 1, Period 2 and Washout Period. A significantly less effect on disruption in FR-40 responses by quipazine and TFMPP during Period 2 and the Washout Period was also seen. Since MAO inhibitors appear to down-regulate both 5-HT1 and 5-HT2 binding sites in brain, the attenuated effects of the 5-HT agonists were anticipated.(ABSTRACT TRUNCATED AT 250 WORDS)

Variability in the effects of 4-bromo-2,5-dimethoxyamphetamine (DOB) on operant behavior of squirrel monkeys

Effects of the hallucinogenic drug (+/-)-4-bromo-2,5-dimethoxyamphetamine HCl (DOB, 0.003-0.3 mg/kg) were studied in squirrel monkeys. Only decreases in responding were seen in monkeys studied under 5-min fixed-interval schedules of food presentation. These decreases were blocked by pretreatment with the 5-HT2 antagonist ketanserin (0.1-1.0 mg/kg) and by the non-selective 5-HT antagonists methysergide (0.3 mg/kg) or mianserin (0.1-1.0 mg/kg). Similar decreases in responding and antagonism by 5-HT antagonists were seen at slightly higher doses of DOM HCl (methyl rather than bromo at the 4 position). In contrast to effects under the food schedule, DOB initially produced marked increases in responding of three monkeys studied under schedules of shock avoidance. However, a complex pattern of changes in the effects of DOB emerged when the same doses were given on subsequent occasions. In one monkey, there were graded increases in responding to a peak of just over 200% of control at 0.17 mg/kg when DOB was given in a roughly ascending dose series. However, no increases in responding were observed at any dose when DOB was given on many subsequent occasions (some very widely spaced). A second monkey showed similar increases initially, but responding was suppressed by a formerly rate-increasing dose of DOB (0.1 mg/kg) on two subsequent test days. Later, this dose again produced increases in responding of about the same magnitude as seen initially, but these increases eventually diminished and were no longer observed. In the third monkey, increases in responding after the initial ascending dose series diminished in an irregular manner over the course of successive redeterminations.

Studies on the specificity of neurochemical and behavioral effects of LSD-25

Brain perfusion experiments of conscious rats engaged in operant behavior and administered fluoxetine or LSD, with or without prior injection of 5-HTP, indicate there is probably more than one functional pool of 5-HT in the CNS. Furthermore, the fact that prior loading with the precursor is necessary before unmasking an effect of LSD suggests the LSD-sensitive pool is newly synthesized and represents only a small fraction of total CNS serotonin. Separating the effects of LSD's behavioral action into pausing (disruption) and depressed responding rate, with or without pausing, enabled us to demonstrate blockade of the disruption by methysergide without blockade of the decreased responding rate. Mianserin blocks both effects of LSD's action. We suggest that behavioral effects of low doses of LSD are due to sympathetic arousal and may offer a model for agitated depression and/or anxiety and that drugs of the mianserin-type may prove useful for treating some forms of anxiety, as well as depression.

Comparative effects of LSD and lisuride: clues to specific hallucinogenic drug actions

This review compares the effects of LSD and its nonhallucinogenic congener lisuride hydrogen maleate (LHM) on various biochemical, behavioral and electrophysiological indices of neuronal function. The underlying rationale is that any differences between the effects of LSD and LHM might be relevant to neuronal actions which are unique and specific to hallucinogenic drugs and thereby provide clues to the neurobiological substrates of hallucinogenesis. In biochemical studies, LHM appears to be very similar to LSD with respect to its actions on monoaminergic (5-HT, DA, NE) systems. The major difference between the two ergots appears quantitative in nature since LHM is more potent than LSD, especially on DA neurochemistry. Needed at the present time are additional comparative studies of LSD and LHM with respect to other biochemical measures, for example on the release of 5-HT and DA and comparisons at more molecular levels such as subcellular compartmentation. Also necessary are more intensive regional analyses on specific subpopulations of 5-HT and DA systems (mesolimbic, mesostriatal and mesocortical). Behavioral studies are relatively uniform in their characterization of the greater DA-ergic activity of LHM as compared to LSD. In particular, the drug discrimination (DD) procedure has indicated a more specific interaction of LSD with 5-HT neuronal systems as compared to LHM and has successfully differentiated the relative roles of 5-HT and DA systems in the behavioral effects of LSD and LHM. Electrophysiological studies have been consistent with both biochemical and behavioral findings with respect to the much greater effect of LHM on DA receptors. In fact, the effects of LSD on DA-containing neurons are both weak and heterogeneous, again indicating a need for more detailed analyses of specific DA projection systems. The greater potency of LHM than LSD on 5-HT containing dorsal raphe neurons has lessened the attractiveness of the once popular theory that hallucinogenic efficacy is related to diminution of impulse flow in 5-HT systems but has also spawned greater interest in the possible role of postsynaptic 5-HT receptors in hallucinogenic drug action. Thus far, the most interesting finding is the ability of LSD and other hallucinogens, but not LHM, to potentiate an excitomodulatory effect of 5-HT in the facial motor nucleus. If such a phenomenon occurs more generally in the CNS, the importance of this finding will be greatly enhanced. Preliminary data is presented which suggests that LSD may also induce such an effect in a limbic forebrain structure, the nucleus accumbens.(ABSTRACT TRUNCATED AT 400 WORDS)

Antagonism of the LSD cue by putative serotonin antagonists: relationship to inhibition of in vivo [3H]spiroperidol binding

In two groups of rats trained to discriminate 0.08 or 0.16 mg/kg of lysergic acid diethylamide (LSD) from saline, pirenperone and ketanserin completely blocked the stimulus effect of LSD. Pizotifen (BC-105) blocked the LSD cue when the training dose was 0.08 mg/kg, but had variable effects in the 0.16 mg/kg of LSD-trained group. The antagonism of the 0.08 mg/kg cue occurred at doses of the antagonists which blocked [3H]spiroperidol labeled 5-HT2 receptors in the frontal cortex in vivo; binding in the striatum was unaffected by the LSD antagonists. However, in doses which produce the LSD cue, neither LSD nor the 5-HT agonist, 5-methoxy-N,N-dimethyltryptamine, which substitutes for LSD, inhibited the binding in either the cortex or the striatum. The results are discussed in relation to the possible neuropharmacological basis for the LSD cue.

The 5HT2 antagonist pirenperone reverses disruption of FR-40 by hallucinogenic drugs

Indolealkylamine and phenethylamine hallucinogens disrupted responding maintained under a fixed-ratio 40 (FR-40) schedule of reinforcement. LSD, DMT, mescaline and DOM produced dose-dependent decreases in number of reinforcers and increases in 10-sec periods of non-responding (pause-intervals). The 5HT agonist quipazine, as well as the LSD congener lisuride, altered response patterns in a similar manner. The effects of these drugs were examined after pretreatment with pirenperone, an antagonist with specificity toward the 5HT2 receptor with reference to the 5HT1 receptor. The dose-response curves for the phenethylamine hallucinogens were shifted significantly to the right and to a greater degree than were those for the indolealkylamine hallucinogens. Pirenperone also antagonized the effects of quipazine to a degree similar to that observed with the phenethylamine-type hallucinogens. Pirenperone did not significantly shift the dose-response pattern to lisuride. These data suggest that the behavioral disruption induced by these hallucinogens and quipazine relates at least in part to an effect on 5HT2 receptors, while the effects of lisuride do not involve a significant interaction at the 5HT2 receptor.

Comparison of anorexia and motor disruption by cyclazocine and quipazine

The mixed narcotic agonist-antagonist cyclazocine and the 5-HT agonist quipazine disrupt food-rewarded fixed ratio-40 (FR-40) operant behavior in rats as a dose-dependent decrease in the number of reinforcers obtained and a reciprocal increase in the number of 10-second intervals between responding ("pausing"). This disruption has been shown to result in part from interaction with 5-HT neuronal systems, and may be a consequence of: (1) disruption of cognitive processes, (2) motivational impairment, or (3) motor deficits. To identify which of these components is (are) involved in the disruption of operant responding, female Sprague-Dawley rats were tested for food consumption, spontaneous locomotor activity, or rotarod performance following intraperitoneal injection of cyclazocine, quipazine, or both. Cyclazocine decreased food consumption at doses larger than those required to disrupt operant behavior, while quipazine decreased consumption at doses disruptive to operant responding. Little effect was exerted by either drug on spontaneous locomotor activity, while rotarod performance was disrupted only by very large doses of either drug relative to effects of FR-40 behavior. These data indicate that neither drug appears to disrupt operant behavior by causing gross motor deficits. Thus, cyclazocine may disrupt operant responding by impairing cognition, while quipazine may act through food satiation mechanisms.

Behavioral effects of intracerebroventricular administration of LSD, DOM, mescaline or lisuride

The effects on a fixed ratio-40 (FR-40) operant behavior of intracerebroventricular (ICV) administration of the hallucinogens lysergic acid diethylamide (LSD), 2,5-dimethoxy-4-methylamphetamine (DOM), mescaline or the non-hallucinogenic LSD-analogue lisuride were compared with intraperitoneal (IP) administration. Infusion of LSD (8.5 to 34 micrograms) into the left lateral ventricle produced a dose-dependent decrease in reinforcers and an increase in 10-sec periods of non-responding (pause intervals). The time-course of LSD showed a shorter latency to onset after ICV than IP administration. The ED50 for doses increasing pause intervals by ICV administration was 15 micrograms. This disruption was greater than that produced by IP administration of equivalent doses of LSD (IP ED50: 19 micrograms). DOM (40 to 120 micrograms) infused into the lateral ventricle also produced a dose-dependent disruption of FR-40 behavior. ICV DOM also showed a rapid onset to peak effects, but a slower offset than LSD, and was 3 times more potent than systemic administration (ED50s: 58 micrograms ICV vs. 153 micrograms IP). Mescaline was much more potent in disrupting FR-40 behavior by the ICV route than by IP administration. The ICV ED50 for doses of mescaline increasing pause intervals was 74 micrograms, in contrast to an ED50 following systemic administration of 2251 micrograms, demonstrating a 30-fold difference in potency. Lisuride administered via the ICV route was no more potent than by IP administration with ED50s of 4 micrograms ICV and 4 micrograms IP. Lower doses of lisuride administered by both routes had a similar effect over time on pause intervals.(ABSTRACT TRUNCATED AT 250 WORDS)

Opposite effects of N,N-dimethyltryptamine (DMT) and 5-methoxy-n,n-dimethyltryptamine (5-MeODMT) on acoustic startle: spinal vs brain sites of action

The present studies examined the role of the spinal cord and the brain in mediating the effects of the hallucinogens N,N-dimethyltryptamine (DMT) and 5-methoxy-N,N-dimethyltryptamine (5-MeODMT) on the acoustic startle response in the rat. Systemic administration of these agents, which distributes to both the brain and spinal cord, produced opposite effects, as DMT depressed and 5-MeODMT increased acoustic startle. However, when administered directly into the lateral ventricle in the forebrain (intraventricular administration) 5-hydroxytryptamine (5-HT), DMT and 5-MeODMT all depressed acoustic startle, DMT and 5-MeODMT being about equipotent in this regard. In contrast, when administered directly into the spinal cord subarachnoid space (intrathecal administration), 5-HT and 5-MeODMT increased startle, whereas DMT was without effect. In another series of studies, the effects of systemically-administered DMT and 5-MeODMT on the "startle" elicited by electrical stimulation of the nucleus reticularis pontis caudalis (RPC) were determined. Since the RPC is the last nucleus of the primary startle circuit before the spinal cord, agents which act downstream from the RPC (i.e., in the lower brainstem and spinal cord) would be expected to alter RPC-elicited "startle," while agents which act upstream from the RPC would be without effect. Given systemically, 5-MeODMT markedly increased RPC-elicited "startle" while DMT was without effect. These data indicate that DMT and 5-MeODMT are equipotent in depressing startle through actions in the brain. In contrast, the difference in the effects of DMT and 5-MeODMT on acoustic startle is related to the spinal excitatory effects of 5-MeODMT which DMT does not possess. From the present results it is suggested that the relative potencies of DMT and 5-MeODMT in other behavioral measures may relate to the role of brain (equipotent) or spinal (5-MeODMT more potent than DMT) sites of action for the various behaviors.