Neurophysiological effects of hallucinogens on serotonergic neuronal systems

Neuroscience & BiobehtavioralReviews, Vol. 6, pp. 509--514, 1982. Printed in the U.S.A.

Neurophysiological Effects of Hallucinogens on Serotonergic Neuronal Systems R O B E R T B. M c C A L L

The Upjohn Company, Kalamazoo, M I 49001

McCALL, R. B. Neurophysiological effects of hallucinogens on serotonergic neuronal systems. NEUROSCI. BIOBEHAV. REV. 6(4) 509-514, 1982.--Low intravenous doses of the hallucinogen d-lysergic acid diethylamide (LSD) markedly suppress the discharge of serotonin (5-HT)-containing neurons in the dorsal raphe nucleus of the rat. Microiontophoretically applied LSD also inhibits the firing of 5-HT neurons, indicating that the inhibitory effect is mediated directing on 5-HT neurons. Forebrain neurons receiving a major serotonergic input are relatively insensitive to LSD. Other indole hallucinogens (i.e., psilocin, dimethyltryptamine, and 5-methoxydimethyltryptamine) also preferentially inhibit raphe firing as compared to postsynaptic forebrain neurons. These observations led to the hypothesis that hallucinogens produce their psychoactive effects by acting preferentially upon 5-HT autoreceptors in the dorsal raphe allowing postsynaptic neurons to escape from the tonic inhibitory action of 5-HT neurons. However, problems exist with the concept that hallucinogens produce their psychoactive effects by disinhibiting postsynaptic neurons. First, the time course of the behavioral and neuronal effects of LSD do not correlate. Second, 5-HT neurons do not become tolerant to the inhibitory actions of LSD. Third, the hallucinogen mescaline fails to directly inhibit 5-HT neurons. Finally, the nonhallucinogen lisuride markedly suppresses the discharges of 5-HT neurons. These observations suggest that postsynaptic actions of hallucinogens may be of prime importance in producing their psychedelic effects. Evidence is presented to suggest that the hallucinogens may act postsynaptically to sensitize both serotonergic and noradrenergic receptors. It is suggested that a mechanism of receptor sensitization, in distinction to disinhibition, might account for the altered perceptual reactivity produced by these drugs. Serotonin

Hallucinogens

d-Lysergic acid diethylamide

BASED on the structural similarity between the hallucinogen d-lysergic acid diethylamide (LSD) and serotonin (5HT), and on experiments demonstrating that LSD acts as a potent 5-HT antagonist in the periphery, Gaddum [20] and Wooley and Shaw [47] independently proposed that L S D ' s potent psychoactive effects might be mediated by an action on central 5-HT receptors. However, it was not until Dahls t r r m and Fuxe [16] localized 5-HT-containing perikarya to the brain stem raphe nuclei and identified the areas of projection of these neurons using fluorescence histochemical techniques that the effects of LSD and other hallucinogens on serotonergic pathways could be studied [16,19]. Since that time attempts to elucidate the mechanism of action of hallucinogenic drugs have focused increasingly upon their effects at the level of the single neuron. The major emphasis of this paper will concern single cell recording studies dealing with the actions of hallucinogenic drugs on histochemically-identified serotonergic neurons and on nuclei which receive a major serotonergic input. Wherever possible an effort will be made to establish correlations between these single unit studies and relevant biochemical and behavioral data. In addition, recent studies involving the effects of hallucinogens on norepinephrine (NE)-containing neurons in the locus coeruleus will be discussed. Based on the observations that LSD reduced the turnover

C o p y r i g h t © 1982 A N K H O

Mescaline

of brain 5-HT [38], and that electrical stimulation of the midbrain raphe nuclei increased 5-HT turnover [7] in the brain, it was suggested that LSD might depress the firing of 5-HT neurons. To test this hypothesis, Aghajanian and co-workers [3,4] used microelectrode recording techniques to examine the effects of intravenous (IV) LSD on unit activity of 5-HT containing neurons in the dorsal raphe nucleus of the rat. As predicted from the biochemical studies, small doses of LSD (10-50/zg/kg) produced a complete but reversible inhibition of 5-HT neuron firing. In contrast, the discharge rate of neurons in a number of other brain areas were either unaffected or accelerated by similar doses of LSD. Similarly the indoleamine hallucinogens psilosin and N ,Ndimethyltryptamine (DMT) inhibited raphe firing [4,5]. In contrast, brom-LSD (BOL), a nonpsychotomimetic analog of LSD, failed to reduce 5-HT unit activity at doses of 25-50 /zg/kg. More recently, Mosko and Jacobs [32] demonstrated that 5-methoxy-N,N-dimethyltryptamine (5-MeODMT), an indoleamine with weak hallucinogenic properties, also depressed raphe neuron discharges. These early studies suggested that psychoactive indoleamines inhibited 5-HT cell firing in the dorsal raphe, while nonhallucinogenic indoles had little effect on serotonergic neurons. However, these data did not indicate whether systemically administered LSD depressed

I n t e r n a t i o n a l Inc.n0149-7634/82/040509-06503.00/0

510 serotonergic neurons through a direct or indirect mechanism. To answer this question Aghajanian and co-workers [1, 5, 17] determined the effects of microiontophoretically applied LSD on 5-HT cell firing. They found that very small ejecting currents (i.e., 10 nA) of dilute solutions of LSD profoundly suppressed the firing of serotonergic neurons. Direct application of 5-HT also inhibited raphe firing; however, LSD was much more potent than 5-HT in its inhibitory effects. Iontophoretic application of other hallucinogens such as psilocin, DMT, and 5-MeODMT also dramatically inhibited the discharge of 5-HT cells [17]. In contrast, BOL had no effect on the firing rate of raphe neurons, nor did it block the inhibitory action of 5-HT [1]. These data indicate that the inhibitory effect of hallucinogenic indoles and 5-HT is mediated directly on 5-HT containing neurons. More recent investigations indicate the presence of a recurrent inhibitory system in the dorsal raphe which is mediated through 5-HT axon collaterals [9]. Thus, the inhibitory effects of LSD and 5-HT are likely mediated by serotonin receptors (i.e., presynaptic receptors or autoreceptors) located on the dendrites and perikarya of 5-HT neurons [9]. If the psychoactive properties of the hallucinogens result from their ability to inhibit raphe firing, then why aren't 5-HT and its precursors, which also depress 5-HT firing, hallucinogenic? A possible explanation for this observation is provided by the work of Haigler and Aghajanian [23]. They found that neurons receiving a major serotonergic input (e.g., ventral lateral geniculate, the basolateral and cortical amygdala and the optic tectum) are relatively insensitive to LSD at iontophoretic ejection currents that are highly effective in the raphe (Fig. 1). In contrast, iontophoretic 5-HT is approximately equipotent in depressing activity of 5-HT neurons and neurons in postsynaptic areas which receive a major serotonergic input (Fig. 1). The preferential action of LSD on raphe cells suggests that LSD could release postsynaptic neurons from a tonic inhibitory 5-HT influence. As predicted, low intravenous doses of LSD accelerates the discharge rate of neurons in postsynaptic areas which receive an inhibitory 5-HT input (Fig. 1) [21, 24, 33]. More recent investigations indicate that psilocin, DMT, and 5-MeODMT also preferentially inhibited raphe firing as compared to the discharge of neurons in postsynaptic forebrain areas [5,17]. In contrast, BOL inhibited neurons in the raphe and in postsynaptic areas to approximately the same degree [4]. In these studies, the degree of each drug's preferential effect on the presynaptic 5-HT receptor correlated well with their estimated hallucinogenic potency. These studies led to the hypothesis that indole hallucinogens produced their psychoactive effects by acting preferentially upon 5-HT autoreceptors in the dorsal raphe allowing postsynaptic neurons to escape from the tonic inhibitory action of 5-HT neurons. Since the visual and limbic systems are densely innervated by 5-HT axons, the hallucinogeninduced disinhibition in these areas could account for two of the major aspects of hallucinogenic action: visual hallucinations and alterations of affect. Recently, Jacobs and co-workers have conducted an elegant series of experiments in order to examine the behavioral effects of hallucinogens in conscious, free-moving cats, while simultaneously recording 5-HT unit activity in the dorsal raphe. Although hallucinogens produce a number of behavioral effects in the cat, the most characteristic appear to be limb flicking and abortive grooming [25]. Trulson and Jacobs [39,40] found that the behavioral effects produced by low doses of LSD and 5-MeODMT were associated with a

McCALL 5HT 20

LSD 20

Raphe 5HT/LSD~

~o,o2o4o

=

5HT

2_0 /

LSD-IV

N

5Hr

20

Ventral Lateral Oeniculote

-

tn LSD\

5HT

20 ~o2o4o

[LSD-IV] |SHT |

~2_o

25

Amygdala !

I

5 MIN

FIG. 1. Comparison of the response of a neuron in the dorsal raphe nucleus with that of neurons in the ventral lateral geniculate and the basolateral nucleus of the amygdala to microiontophoretic administration of 5-HT and LSD. LSD preferentially inhibited 5-HT neurons. 5-HT inhibited 5-HT neurons and postsynaptic neurons to a similar degree. In this and the following figure, the horizontal bar indicates the iontophoretic ejection period; the number over the bar represents the ejecting current. The ordinate represents the integrated firing rate in spikes per second. Reprinted by permission of The Williams and Wilkins Co., Baltimore, MD. Copyright 1974.

decrease in the activity of 5-HT neurons. 5-MeODMT t10250/xg/kg, IM) depressed 5-HT cell firing and increased behavioral effects in a dose-dependent fashion. Importantly, the onset, peak, and discontinuation of the behavioral effects of 5-MeODMT were temporally correlated with changes in 5-HT neuronal activity. The direct correlation between the behavioral and neuronal effects of 5-MeODMT provide strong support for the "disinhibition" hypothesis of hallucinogenic action described above. Like 5-MeODMT, LSD inhibited 5-HT neuronal activity and produced characteristic limb flicks in the free-moving cat [40]. Although the peak behavioral and unit changes produced by LSD (10 and 50 /zg/kg, IP) were temporally correlated, the behavioral effects outlasted the neuronal effects by several hours. Furthermore, LSD (50 /zg/kg, IP) readministered 24 hours later failed to produce any behavioral effects (i.e., tolerance), but the depression in 5-HT cell activity was at least as large as that observed on the first day. The latter observation supports observations in the anesthetized rat which indicate that the responsiveness of 5-HT neurons in the dorsal raphe to LSD was not altered by chronic administration of LSD (I00-1000/zg/kg/day for 4 to 7 days) [42]. In contrast, one of the most impressive behavioral effects of LSD in humans is the dramatic tolerance which develops to its repeated administration [11,45]. The dissociations between the effects of LSD on behavior and raphe unit activity suggest that the constellation of behavioral effects produced by LSD and related hallucinogens are not due entirely to their preferential inhibitory action on raphe neurons. However, the observations described above

HALLUCINOGENS AND 5-HT NEURONS do not preclude the possibility that while the initiation of the behavioral effects of LSD may depend on the depression of 5-HT neuron firing, the drug might also produce changes in postsynaptic neurons which could outlast the raphe effect and modify responses to subsequent injections. In this regard, Trulson and Jacobs [41] have shown that chronic administration of LSD alters the sensitivity of postsynaptic 5-HT receptors to LSD and 5-HT. One precautionary note must be added regarding the experiments described above. Lisuride, a nonhallucinogenic analog of LSD, produces behavioral effects in the cat which are identical to those caused by hallucinogens, including limb flick and abortive grooming ]28]. The fact that lisuride is a false positive in the cat model raises the possibility that the behavior produced by hallucinogens in the cat is not a good indicator of hallucinogenic activity in man and, therefore, may not be an appropriate index with which to study the mechanism of action of these drugs. Further experiments are required to help answer this question. The phenethylamine hallucinogen mescaline produces behavioral effects which are very similar to those produced by LSD [46]. In addition, cross-tolerance occurs to the behavioral effects of mescaline and LSD [11]. A metabolic action does not account for the cross-tolerance since chronic treatment with mescaline does not reduce brain levels of LSD 1451. Finally, several investigators have suggested that mescaline's aromatic ring may assume a conformation resembling the pyrrole portion of the indole nucleus in LSD I9]. Taken together, these data suggest that mescaline and LSD act on the same receptor site to produce their psychoactive effects. If true, then the "disinhibition" hypothesis of hallucinogenic action demands that mescaline have a depressant action on 5-HT neuron firing. Aghajanian and coworkers first determined the effects of mescaline on raphe unit activity [8]. They found that intravenous administration of mescaline (2-4 mg/kg) inhibits a subpopulation of 5-HT neurons located within the ventral portion of the dorsal raphe nucleus and adjacent areas of the median raphe. These same neurons were depressed by LSD. These data suggested that a subpopulation of raphe neurons might be involved in mediating the psychoactive actions of mescaline and LSD. However, studies in which mescaline was microiontophoretically applied onto 5-HT neurons did not bear out this expectation. In agreement with previous work, Haigler and Agh~janian 1221 found that intravenous mescaline (2-8 mg/kg) inhibited a subpopulation of 5-HT neurons in the dorsal raphe. However, iontophoretic application of mescaline using low ejection currents failed to inhibit raphe firing. Some neurons were inhibited by high ejecting currents of mescaline, but this was thought to represent a local anesthetic action since the depression was associated with a decrease in action potential size. Furthermore, this depression w~s not correlated with the inhibition produced by intravenous mescaline. For example, some neurons were unaffected by the iontophoretic application of mescaline but were completely inhibited by intravenous mescaline. The failure to obtain a response to microiontophoretically applied mescaline indicates that the inhibitory effects of systemically administered mescaline on 5-HT neurons are mediated via an indirect mechanism. Biochemically, LSD and mescaline also differ in that LSD but not mescaline depresses brain serotonin metabolism [18]. Thus, it appears that LSD and mescaline differ in their mode of action upon raphe cells. If mescaline and LSD act at a common site to produce their psychoactive actions, then these data would argue against a

,

511

"disinhibition" mechanism for the action of hallucinogenic drugs. Another difficulty with the 5-HT "disinhibition" hypothesis emerges from the work of Rogawski and Aghajanian [37] involving the actions of lisuride on raphe neurons. Lisuride is a nonhallucinogenic ergoline derivative which is structurally related to LSD. Low doses of lisuride have been found to reduce 5-HT and dopamine turnover, while higher doses increase NE turnover I36]. These observations suggest that lisuride is a potent 5-HT and dopamine agonist and a weak NE receptor antagonist. Rogawski and Aghajanian found that extremely small intravenous doses of lisuride ( 1-5 p.g/kg) produced a rapid, dose-dependent suppression of the spontaneous activity of 5-HT neurons. The inhibition was relatively long-lasting but reversible; partial recovery of 15-50% of predrug baseline neuronal activity was attained within 20 minutes. On the basis of the molar intravenous dose required to achieve a complete inhibition of spontaneous raphe activity, lisuride was approximately 5-10 times as potent as LSD. Similar observations have recently been made by Laurent and Pieri [26] regarding the relative potencies of lisuride and LSD on inhibiting 5-HT cell firing. When applied by microiontophoresis (6--20 nA) lisuride markedly depressed the firing rate of all 5-HT neurons tested. When compared with LSD, the time of onset and the magnitude of the inhibition produced by iontophoretic lisuride were similar, but the duration was about twice as great as that produced by LSD. These results suggest that lisuride has a direct agonist action at the 5-HT autoreceptor. Since the "disinhibition" theory of hallucinogenic action requires that psychedelic agents exert their effect via a direct suppression of 5-HT neurons, then these findings present a serious challenge to the "disinhibition" hypothesis. However, recent studies indicate that some nonserotonergic agents act on 5-HT neurons to inhibit their discharge (i.e., a2-adrenergic receptor antagonists) [12]. Thus, the possibility remains that lisuride could act via a nonserotonergic mechanism to suppress 5-HT cell firing and, therefore, would not express hallucinogenic activity. Further experiments designed to determine pharmacological differences in the actions of LSD and lisuride will be critical in understanding the mechanism of action of the hallucinogens. As detailed above, certain problems exist with the theory that hallucinogens act preferentially on 5-HT autoreceptors to suppress raphe firing and thus produce their psychedelichallucinatory state by disinhibiting postsynaptic neurons in visual and limbic areas where 5-HT is inhibitory. First, the time course of the behavioral and neuronal effects of LSD in the free-moving cat do not correlate [40]. Second, although tolerance develops to the behavior effects of LSD {40], no such tolerance develops to the inhibitory effect of LSD on 5-HT neurons [40,42]. Third, the hallucinogen mescaline does not directly suppress the firing of 5-HT neurons in the dorsal raphe 122]. Finally, lisuride is more potent than LSD in inhibiting serotonergic neurons, but does not produce hallucinations [37]. Taken together, these observations suggest that postsynaptic action(s) of the hallucinogens may be of prime importance in producing their psychedelic effects. In this regard, the recent work of McCall and Aghajanian appears to be particularly intriguing [29--31]. These investigators studied the effects of 5-HT, NE, and several hallucinogens on facial motoneurons. Like all motor nuclei, the facial motor nucleus receives a dense 5-HT input [29,35]. In addition, the facial nucleus offers a special advantage in that interneurons are not present in this nucleus [29]. Therefore,

512

McCALL TABLE 1 I N F L U E N C E O F I N T R A V E N O U S H A L L U C I N O G E N S AND N O N H A L L U C I N O G E N S ON T H E F A C I L I T A T I N G E F F E C T O F 5-HT AND N E ON F A C I A L M O T O N E U R O N S

Drug LSD Mescaline Psilocin DMT$ Lisuride Methysergide

Number of Animals*

Dose

% Control 5-HT

% Control NE

8 6 5 5 5 5

10/~g/kg 1 mg/kg 1 mg/kg 1 mg/kg 40/zg/kg 1 mg/kg

1026 _+ 93t 763 +_ 120t 497 _+ 46? -102 -+ 4 0

976 - 1091 913 _+ 168T 491 _ 72? 469 _+ 48t 102 + 2 98 _+ 6

*One neuron was tested per animal. t'p<0.0t, Student's t-test. SAnimals were pretreated with metergoline (I00/zg/kg, IV).

any interpretation of the effects of iontophoretically applied substances on facial motoneurons cannot be confused by indirect effects on local interneurons. McCall and Aghajanian [29] found that microiontophoretic application of 10-200 nA pulses of 5-HT lasting 1 to 10 minutes failed to excite the normally quiescent motoneurons. However, small amounts of 5-HT (5-10 nA) dramatically facilitated the subthreshold and threshold excitatory effects of iontophoreticaily applied glutamate and afferent nerve stimulation. Typically, the current of glutamate required to produce an activation of facial motoneurons was reduced by at least 50% in the presence of 5-HT. Like 5-HT, NE also markedly facilitated excitatory inputs to facial motoneurons. Peripheral 5-HT antagonists, which fail to block the inhibitory effects of 5-HT in the raphe and forebrain areas [9], antagonized the facilitating effects of 5-HT but not NE [31]. This observation indicates that the receptor mediating the facilitating effects of 5-HT in the facial nucleus is distinct from those found in the dorsal raphe and postsynaptic forebrain areas. Recent anatomical [6] and intracellular recording studies [43] indicate that the facilitating effect of 5-HT is mediated by receptors located on the dendrites and somas o f the motoneurons and is associated with a slow depolarization and slight increase in membrane resistance. Thus, in the facial nucleus 5-HT functions in a manner that is not analogous to direct excitation, but rather acts as a gain setter to enhance the effects of excitatory afferent inputs. White and Neuman have recently confirmed that 5-HT and N E facilitate excitatory inputs to spinal motoneurons [44]. More recently McCall and Aghajanian studied the effects of various hallucinogens on facial motoneurons [30]. Intravenous administration of LSD (5--10 /zg/kg) or mescaline (0.5-1.0 mg/kg) had no effect by themselves on the glutamate-induced excitation of facial motoneurons (Fig. 2). In contrast, the facilitation of facial neuron excitation by iontophoretically applied 5-HT and N E was enhanced 810-fold by these hallucinogens (Fig. 2, Table 1). In a similar fashion microiontophoretically applied LSD and mescaline failed to effect the excitation of motoneurons produced by glutamate, but dramatically enhanced the facilitating effects of 5-HT and NE. The most remarkable feature of this effect was that even a brief exposure to the hallucinogens sensitized responses to 5-HT and N E for many hours. The similarity between the effects of LSD and mescaline contrasts to

G15 A

5HT 25

NE 15

LSD 25

15

25

15

i

4401 O-J

G17 LSD 5

U) B o 3601 "'

5HT 30

30

30

30

30

30

G 24 LSD 20

C

5HT 20

20

20

20

20

3301 0,...J__ k-----t

2 MIN FIG. 2. LSD potentiates the facilitating effect of iontophoretically applied 5-HT and NE on glutamate (G)-induced facial motoneuron excitation. (A) Intravenous LSD (10/zg/kg) immediately enhances the response to 5-HT and NE, an effect lasting at least 4 hours (slashed baseline). (B) Iontophoretically applied LSD (5 nA) also enhances the effect of 5-HT. (C) Higher ejecting currents of LSD (20 nA) temporarily depress the response to 5-HT. Reprinted by permission of Pergamon Press Ltd., New York, NY. Copyright 1980.

the actions of these agents in the dorsal raphe nucleus (see above). In addition, psilocin and DMT acted in a manner similar to LSD and mescaline (Table 1). Thus, it appears that the ability to potentiate the facilitating effect of 5-HT and N E may be common to all of the psychedelic hallucinogens. Importantly, the nonhallucinogens lisuride and methysergide failed to potentiate the facilitating effects of 5-HT and NE.

H A L L U C I N O G E N S AND 5-HT N E U R O N S

~.

The hallucinogens appear to potentiate the effects of monoamines on facial motoneurons by increasing the sensitivity of 5-HT and NE receptors. A precedent for such a sensitizing action can be found in the observation that low doses of LSD and mescaline facilitate uterine contractions produced by 5-HT 115 I. The mechanism by which hallucinogens sensitize 5-HT and N E receptors is not known. However, the fact that LSD and 5-HT appear to bind to different 5-HT receptor sites [13] suggests the possibility that druginduced changes in receptor sensitivity could occur, in part, through interactions between 5-HT and LSD binding sites. In any case, a sensitization of facilitatory 5-HT and NE receptors by LSD and mescaline likely accounts for the enhancement of spinal reflexes produced by these hallucinogens [ 10,27]. If the sensitizing effects of hallucinogenic drugs observed in the facial nucleus occurs in sensory pathways of the central nervous system, then a mechanism of receptor sensitization, in distinction to disinhibition, might account for the altered perceptual reactivity produced by these drugs. In this regard, the time course of the LSD enhanced effects of 5-HT and NE parallels the prolonged behavioral effects of this drug in the free-moving cat (see above, [40]). Further support for a receptor sensitizing effect of hallucinogens similar to that observed in the facial nucleus stems from the fact that the indoleamine and phenethylamine hallucinogens acted in a similar fashion, while lisuride has no effect in the facial nucleus. These observations raise the interesting possibility that relatively long-term changes in the sensitivity of noradrenergic and/or serotonergic receptors may be involved in the psychedelic actions of hallucinogens. Recent work by Aghajanian suggests that the sensitizing effects of psychedelic drugs may be more widespread than previously suspected I2]. He found that intravenously administered mescaline (0.25-2.0 mg/kg) and LSD (5-10/zg/kg) produced a prolonged suppression of firing of NE-containing neurons located in the locus eoeruleus (LC). Despite the reduction in baseline firing rate, both mescaline and LSD enhanced the reactivity of LC neurons to peripheral stimula-

5t3

tion elicited by air puff, skin pinch or stimulation of the sciatic nerve. These effects were not shared by other drugs which also depress the spontaneous firing of NE-containing neurons (e.g., L-amphetamine, desipramine, and clonidine). In addition, the excitation of LC neurons by microiontophorcticaUy applied acetycholinc, glutamate o r substance P was not enhanced by systemic doses of mescaline and LSD. Thus, the effects of the hallucinogens appear to be mediated through an effect on LC afferents rather than through a direct increase in the excitability of LC neurons. In addition, iontophoretic application of mescaline or LSD failed to enhance the reactivity of NE neurons to peripheral stimulation. The possibility exists that, like facial motoneurons, the enhanced reactivity of LC neurons induced by mescaline and LSD represents a sensitization to excitatory afferent inputs. Interestingly, many of the sensory relay nuclei which project to the LC [14] also receive a major serotonergic and/or noradrenergic input. It is interesting to speculate that the enhanced reactivity to peripheral stimuli elicited by hallucinogens results from a sensitization of 5-HT and/or NE receptors located on sensory relay neurons. On the basis of the investigations described above, it is clear that much more work is required before we will gain an understanding of the mechanisms involved in hallucinogenic drug action. The constellation of behaviors produced by the hallucinogens is likely to result from a variety of actions. Thus, it is possible that both presynaptic actions (i.e., 5-HT disinhibition) and postsynaptic effects (i.e., altered synaptic plasticity in serotonergic or noradrenergic projection areas) of hallucinogens are necessary to express their psychoactive properties. In addition, the effects of hallucinogens on other putative neurotransmitters, including peptidergic systems, must not be overlooked. In any case, a unified theory on the mechanism of action of the hallucinogens must take into account the similarity in the actions of LSD and mescaline, the differences in the actions of LSD and lisuride, and the rapid tolerance that develops to LSD.

REFERENCES 1. Aghajanian, G. K. LSD and 2-Bromo-LSD: Comparison of effects on serotonergic neurones and on neurones in two serotonergic projections areas, the ventral lateral geniculate and amygdala. Neuropharmacology 15: 521-528, 1976. 2. Aghajanian, G. K. Mescaline and LSD facilitate the activation of locus coeruleus neurons by peripheral stimulation. Brain Res. 186: 492-498, 1980. 3. Aghajanian, G. K., W. E. Foote and M. H. Sheard. Lysergic acid diethylamide: Sensitive neuronal units in the midbrain raphe. Science 161: 706--708, 1968. 4. Aghajanian, G. K., W. E. Foote and M. H. Sheard. Action of psychotogenic drugs on midbrain raphe neurons. J. Pharmac. exp. Ther. 171: 178-187, 1970. 5. Aghajanian, G. K. and H. J. Haigler. Hallucinogenic indoleamines: Preferential action upon presynaptic serotonin receptors. Psychopharmac. Cmmmm. 1: 61%629, 1975. 6. Aghajanian, G. K. and R. B. McCall. Serotonergic synaptic input to facial motoneurons: Localization by electronmicroscopic autoradiography. Neuroscience 5: 2155-2162, 1980.

7. Aghajanian, G. K., J. A. Rosecrans and M. H. Sheard. Serotonin: Release in the forebrain by stimulation of midbrain raphe. Science 156: 402-403, 1967. 8. Aghajanian, G. K., M. H. Sheard and W. E. Foote. LSD and mescaline: Comparison of effects on single units in the midbrain raphe. In: Psychotomimetic Drugs, edited by D. H. Efron. New York: Raven Press, 1970, pp. 165-176.

9. Aghajanian, G. K. and R. Y. Wang. Physiology and pharmacology of central serotonergic neurons. In: Psychopharmacology: A Generation of Progress, edited by M. A. Lipton, A. DiMascio and K. F. Killam. New York: Raven Press, 1978, pp. 171-183. 10. Andrn, N. E., H. Corrodi, K. Fuxe and H. T. HOkfe. Evidence for a central 5-hydroxytryptamine receptor stimulation by lysergic acid diethylamide. Br. J. Pharmac. Chemother. 34: 1-7, 1968. 11. Appel, J. B. and D. X. Freedman. Tolerance and crosstolerance among psychotomimetic drugs. Psychopharmacologia 13: 267-274, 1968. 12. Baraban, J. M., R. Y. Wang and G. K. Aghajanian. Reserpine suppression of dorsal raphe neuronal firing: Mediation by adrenergie system. Eur. J. Plmrmac. 52: 27-36, 1978. 13. Bennett, J. P. and S. H. Snyder. Serotonin and lysergic acid diethylamide binding in rat brain membranes: Relationship Io postsynaptic serotonin receptors. Molec. Pharmac. 12: 373389, 1976. 14. Cedarbaum, J. M. and G. K. Aghajanian. Afferent projections to the rat locus coeruleus as determined by a retrograde tracing technique, d. contp. Neurol. 178: 1-15, 1978. 15. Costa, E. Effects of hallucinogenic and tranquilizing drugs on serotonin evoked'uterine contractions. Proc. Soc. exp. Biol. Med. 91: 3%41, 1956.

514 16. Dahlstr6m, A. and K. Fuxe. Evidence for the existence of monoamine-containing neurons in the central nervous system. I. Demonstration of monoamines in the cell bodies ofbrainslem neurons. Acta physiol, stand. Suppl. 232: 1-55, 1964. 17. De Montigny, C. and G. K. Aghajanian. Preferential action of 5-methoxytryptamine and 5-methoxydimethyltryptamine on presynaptic serotonin receptors: A comparative iontophoretic study with LSD and serotonin. Neuropharmacology 16: 811818, 1977. 18. Freedman, D. X., R. Gottlieb and R. A. Lovell. Psychotomimetic drugs and brain 5-hydroxytryptamine metabolism. Biochem. Pharmac. 19: 1181-1188, 1970. 19. Fuxe, K. Evidence for the existence of monoamine-containing neurons in the central nervous system. IV. The distribution of monoamine terminals in the central nervous system. Acta physiol. Scand. Suppl. 247: 37-85, 1965. 20. Gaddum, J. H. Antagonism between lysergic acid diethylamide and 5-hydroxytryptamine. J. Physiol. 121: 15, 1953. 21. Guilbaud, G., J. M. Besson, J. L. Oliveras and J. C. Liebeskind. Suppression by LSD of the inhibitory effect exerted by dorsal raphe stimulation on certain spinal cord interneurons in the cat. Brain Res. 61" 417-422, 1973. 22. Haigler, H. J. and G. K. Aghajanian. Mescaline and LSD: Direct and indirect effects on serotonin-containing neurons in brain. Eur. J. Pharmac. 21: 53-60, 1973. 23. Haigler, H. J. and G. K. Aghajanian. Lysergic acid diethylamide and serotonin: A comparison of effects on serotonergic neurons and neurons receiving a serotonergic input. J. Pharmac. exp. Ther. 188: 688-699, 1974. 24. Horn, G. and J. M. McKay. Effects of lysergic acid diethylamide on the spontaneous activity and visual receptive fields of cells in the lateral geniculate nucleus of the cat. Expl Brain Res. 17: 271-284, 1973. 25. Jacobs, B. L., M. E. Trulson and W. C. Stern. An animal behavioral model for studying the actions of LSD and related hallucinogens. Science 194: 741-743, 1976. 26. Laurent, J. P. and L. Pieri. Lisuride and LSD decrease spontaneous firing of neurons in the raphe. Experientia 34: 926, 1978. 27. Maj, J., W. Palider and A. Rawlow. The influence of mescaline on the flexor reflex of the hindlimb of the spinal rat. J. Pharm. Pharmac. 29: 177-178, 1977. 28. Marini, J. L., B. L. Jacobs, M. H. Sheard and M. E. Trulson. Activity of a nonhallucinogenic ergoline derivative, lisuride, in an animal behavior model for hallucinogens. Psychopharmacology 73: 328-331, 1981. 29. McCall, R. B. and G. K. Aghajanian. Serotonergic facilitation of facial motoneuron excitation. Brain Res. 169:11-27, 1979. 30. McCall, R. B. and G. K. Aghajanian. Hallucinogens potentiate responses to serotonin and norepinephrine in the facial motor nucleus. Life Sci. 26: 114%1156, 1980.

McCALL 31. McCall, R. B. and G. K. Aghajanian. Pharmacological characterization of serotonin receptors in the facial motor nucleus: A microiontophoretic study. Ear..1. Pharmac. 65: 175-183, 1980. 32. Mosko, S. S. and B. L. Jacobs. Electrophysiological evidence against negative neuronal feedback from the forebrain controlling midbrain raphe unit activity. Brain Res. ! 19: 291-303, 1977. 33. Mouriz-Garcia, A., R. Schmidt and A. Arlazoroff. Effects of LSD on the spontaneous and evoked activity of retinal and geniculate ganglion cells. Psychopharmacologia 15: 382-391, 1969. 34. Nichols, D. E., W. R. Pfister, G. K. W. Yim and R. J. Cosgrove. A new view of the structural relationship between LSD and mescaline. Brabl Res. Bull. 2: 16%171, 1977. 35. Palkovits, M., M. Brownstein and J. M. Saavcdra. Scrotonin content of the brain stem nuclei in the rat. Brain Res. 80: 237249, 1974. 36. Pieri, L., H. H. Keller, W. Burkhard and M. Daprada. Effects of lisuride and LSD on cerebral monoamine systems and hallucinosis. Nature 272: 278-280, 1978. 37. Rogawski, M. A. and G. K. Aghajanian. Response of central monoaminergic neurons to lisuride: Comparison with LSD. Lift, Sci. 24: 128%1298, 1979. 38. Rosecrans, J. A., R. A. Lovell and D. X. Freedman. Effects of lysergic acid diethylamide on the metabolism of brain 5-hydroxytryptamine. Biochem. Pharmac. 16:2011-2021, 1967. 39. Trulson, M. E. and B. L. Jacobs. Effects of 5methoxy-N,N-dimethyltryptamine on behavior and raphe unit activity in freely moving cats. Eur. J. Pharmac. 54: 43-50, 1979. 40. Trulson, M. E. and B. L. Jacobs. Dissociations between the effects of LSD on behavior and raphe unit activity in freely moving cats. Science 205: 515-518, 1979. 41. Trulson, M. E. and B. L. Jacobs. Alterations of serotonin and LSD receptor binding following repeated administration of LSD. Lift, Sci. 24: 2053-2062, 1979. 42. Trulson, M. E., C. A. Ross and B. L. Jacobs. Lack of tolerance to the depression of raphe unit activity by lysergic acid diethylamide. Neuropharmacology 16: 771-774, 1977. 43. VanderMaelen, C. P. and G. K. Aghajanian. Intracellular studies showing modulation of facial mononeurone excitability by serotonin. Nature 287: 346--347, 1980. 44. White, S. R. and R. S. Neuman. Facilitation of spinal motoneuron excitability by 5-hydroxytryptamine and norepinephrine. Brain Res. 188: 11%127, 1980. 45. Winters, J. C. Tolerance to a behavioral effect of lysergic acid diethylamide and cross-tolerance to mescaline in the rat: Absence of a metabolic component. J. Pharmac. exp. Ther. 178: 625-630, 1971. 46. Wolbach, A. B., E. J. Miner and H. Isbell. Comparison of psilocin and psilocybin, mescaline, and LSD-25. Psychopharmacologia 3: 21%223, 1962. 47. Wooley, D. W. and E. Shaw. A biochemical and pharmacological suggestion about certain mental disorders. Proc. m~tn. A cad. Sci. U.S.A. 40: 228--231, 1954.