-
Sigma Xi, The Scientific Research Society
Mechanisms of Action of LSD: The study of serotonin-containing
neurons in the brain mayprovide the key to understanding
drug-induced hallucinations and their relationship todreams and
psychosisAuthor(s): Barry L. Jacobs and Michael E. TrulsonSource:
American Scientist, Vol. 67, No. 4 (July-August 1979), pp.
396-404Published by: Sigma Xi, The Scientific Research
SocietyStable URL: http://www.jstor.org/stable/27849328 .Accessed:
29/01/2014 06:32
Your use of the JSTOR archive indicates your acceptance of the
Terms & Conditions of Use, available at
.http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars,
researchers, and students discover, use, and build upon a wide
range ofcontent in a trusted digital archive. We use information
technology and tools to increase productivity and facilitate new
formsof scholarship. For more information about JSTOR, please
contact [email protected].
.
Sigma Xi, The Scientific Research Society is collaborating with
JSTOR to digitize, preserve and extend accessto American
Scientist.
http://www.jstor.org
This content downloaded from 134.117.10.200 on Wed, 29 Jan 2014
06:32:04 AMAll use subject to JSTOR Terms and Conditions
-
Mechanisms of Action of LSD
The study of serotonin-containing neurons in the brain may
provide the key to understanding drug induced hallucinations and
their relationship to dreams and psychosis
Barry L. Jacobs Michael E. Trulson
"Last Friday, April 16,1943,1 was forced to stop my work in the
laboratory in the middle of the afternoon and to go home, as I was
seized by a peculiar restlessness associated with a sensation of
mild dizzi ness. Having reached home, I lay down and sank in a kind
of drunkenness which was not unpleasant and which was char
acterized by extreme activity of imagi nation. As I lay in a dazed
condition with my eyes closed (I experienced daylight as
disagreeably bright) there surged upon me an uninterrupted stream
of fantastic
images of extraordinary plasticity and vividness and accompanied
by an intense, kaleidoscope-like play of colors. This condition
gradually passed off after about two hours."
This description, from the journal of Albert Hofmann (1), a
Swiss chemist working in Basel during World War II, heralds the
dawn of the era of ex
perimental study of hallucinogenic drugs, for Hofmann's
experience re sulted from his accidental ingestion of an unknown
quantity of d-lysergic
Barry Jacobs, Associate Professor of Psychol ogy at Princeton
University and a member of the interdepartmental graduate program
in
neuroscience, received his doctorate at UCLA and was a
postdoctoral fellow in the psychia try department at Stanford. In
1977-78 he was a visiting scientist at the Salk Institute. His
research interests include the neural bases of complex mammalian
behavior, animal models
of human psychopathology, psychoactive drugs, and sleep. Michael
Trulson, who re ceived his doctorate in biopsychology from the
University of Iowa in 1974, has been at Princeton since then as
a lecturer and director
of the neurochemistry laboratory in the psy chology department.
He is currently an Alfred P. Sloan fellow. His research interests
are
primarily in the area of brain neurochemistry and behavior.
Address for Professor Jacobs:
Program in Neuroscience, Department of Psychology, Princeton
University, Princeton, NJ 08544.
acid diethylamide (LSD). Three days later, in an attempt to
verify that the episode was indeed attributable to the ingestion of
LSD, Hofmann took what he thought would be a small quantity of the
drug, 250 fig. As it turned out, this was approximately five times
the dose necessary to pro duce intense hallucinations in an av
erage adult male. His personal ac count of the drug's effects was
re
markably accurate, since virtually everything he described has
been confirmed by subsequent laboratory studies with large numbers
of subjects under controlled conditions.
The effects of LSD can be divided into three general categories
(2): so
matic symptoms?dizziness, weak ness, tremors, nausea, creeping
or
tingling sensations on the skin, and blurred vision; perceptual
symp toms?altered shapes and colors, vi sual hallucinations,
synesthesia (a
mixing of senses, such as the trans formation of sounds into
changes in visual perception), and a distorted time sense;
affective and cognitive symptoms?large and rapid mood changes,
difficulty in thinking, de personalization, and dreamlike feel ing.
The small amount of LSD needed to produce these profound psycho
logical effects makes it the most po tent psychoactive drug known,
on a
microgram-for-microgram basis.
During the past fifteen years, re search in the field of
neuroscience has
greatly elucidated the neurobiological mechanisms that mediate
LSD's psychological effects. The primary purpose of this article is
to describe the research that has led to our
present understanding of these mechanisms. Much of the
discussion will be centered on serotonin, a brain
neurotransmitter that appears to play a key role in the action
of LSD. More generally, we will consider brain se rotonin in the
broad context of its role in mammalian behavior.
The discussion of the mechanism of action of LSD also
illustrates the revolution that has occurred during the past two
decades in our under standing of the action of a variety of
psychoactive drugs, such as cocaine, amphetamine, and
chlorpromazine (the most frequently prescribed an tipsychotic).
Although these drugs act through a number of different
mechanisms, they share the common property of altering chemical
neuro transmission in the brain. Therefore, our story logically
begins with a brief review of chemical neurotransmis sion.
The human brain is comprised of tens of billions of nerve cells,
or neurons. Neurons bring information to the brain concerning the
current status of, or the occurrence of any change in, the body's
internal milieu and the external world. Neurons also call into
action many of the glands and mus cles of the body. By acting upon
each other, in ways we are just beginning to understand, neurons
provide the tremendous integrative capacity of the brain that
underlies human
memory, thought, and emotion.
The action of neuron upon neuron is carried out by means of
small amounts of chemicals released from the terminal endings of
the neurons into the minute gaps, or synapses, between them (Fig.
1). These mole cules then cross the synapse and im
pinge upon the dendrites or the cell body of the receiving
neuron and produce either excitation or inhibi
396 American Scientist, Volume 67
This content downloaded from 134.117.10.200 on Wed, 29 Jan 2014
06:32:04 AMAll use subject to JSTOR Terms and Conditions
-
Figure 1. One neuron makes contact with an other not by actually
touching but by com
municating across minute spaces, called sy napses, by means of
chemical neurotransmit ters. The classical synapse is between the
axon terminal of one neuron and the dendrites or cell body of
another neuron. Fig. 2 shows details of synaptic
neurotransmission.
tion. If the summated effect of the hundreds, or even thousands,
of in puts that a neuron receives favors excitation and reaches a
threshold level, the cell will "fire" and transmit an electrical
impulse down its axon.
When the impulse reaches the ter minal portion of the axon and
causes the release of some small proportion of the chemical
neurotransmitter stored there, the entire cycle begins again.
LSD and serotonin: Historical overview For many years scientists
had known of a blood-borne chemical that pro duced vasoconstriction
(a serum factor that affected blood vessel tonus, hence the name
serotonin) and of a substance present in the gut that caused
intestinal motility. In the mid-twentieth century, serotonin, the
single compound producing both these effects, was isolated and syn
thesized, and its molecular structure elucidated as
5-hydroxytryptamine. Soon thereafter, in 1953-54, serotonin
was found to be present in the mam malian central nervous system
(CNS) in significant quantities and to be concentrated in varying
amounts in different regions of the brain. This led to the proposal
that serotonin was a CNS neurotransmitter. In the twenty-five years
since that time, an impressive body of evidence has been
accumulated to indicate that seroto nin does indeed act as a
central neu rotransmitter.
Structurally, the serotonin molecule is similar to a portion of
the larger LSD molecule in that they both con tain an indole
nucleus. Based on this structural similarity, Gaddum, working in
England, and Wooley, in the United States, independently proposed
that LSD might act to block the synaptic action of serotonin in the
brain (3, 4). They buttressed their argument by demonstrating that
LSD exerted a powerful blocking effect on serotonin's action in
peripheral tissue studied in vitro. This hypothesis at tracted a
great deal of attention, be cause, in addition to being considered
a hallucinogenic drug, LSD was thought by many to be a prototypic
psychotomimetic drug?i.e. one whose effects mimicked psychosis.
Thus, the argument went, a key ele ment in the etiology of
psychosis might be either decreased amounts of brain serotonin or
the synthesis of an endogenous compound that anta gonized
serotonin's action in the brain.
This ambitious hypothesis was soon shaken by the report that
brom-LSD, which is LSD with a single bromine atom attached,
although as effective as LSD in blocking serotonin's action on
peripheral tissue, was devoid of potent psychic effects in humans
(5). It was therefore unlikely that the ability of LSD to block the
action of serotonin could account for its hal lucinogenic action,
since this ability
was shared by a close analog of LSD that was nonhallucinogenic.
It is worth noting, however, that both sides of this argument were
based on studies of LSD's action on peripheral tissue. Because of
the general inac cessibility of brain tissue, there had been very
little direct test of the central effects of LSD. The remain der of
this paper will be concerned with examining these latter studies,
but because so many of them involve brain serotonin, we must first
provide
postsynaptic neuron
Figure 2. An axon terminal of a serotonin containing neuron
(presynaptic neuron) makes synaptic contact with one of its target
neurons
(postsynaptic neuron). L-tryptophan, the amino acid precursor of
serotonin, is brought to the neuron by the blood. Serotonin is syn
thesized from tryptophan inside serotonergic nerve terminals and is
stored in packets called vesicles. When an action potential invades
the axon terminal, the vesicles release their con tents into the
synaptic gap and bombard the postsynaptic neuron to produce either
excita tion or inhibition. Serotonin is inactivated by being taken
back up into the terminal, where it is catabolized by monoamine
oxidase (MAO) to form 5-hydroxyindoleacetic acid (5 HIAA).
serotonin
Figure 3. The indole nucleus structure of the serotonin molecule
is similar to that of several hallucinogenic drugs, including LSD
(as shown in Fig. 6).
some background information about this neurotransmitter.
Basics of brain serotonin Serotonin is a small and relatively
simple molecule that is synthesized from the essential amino acid
tryp tophan. After a meal, tryptophan is transported, by the blood,
from the gut to serotonin-containing neurons, where serotonin
synthesis takes place (Figs. 2 and 3). The newly formed serotonin
is stored in the axon ter
1979 July-August 397
This content downloaded from 134.117.10.200 on Wed, 29 Jan 2014
06:32:04 AMAll use subject to JSTOR Terms and Conditions
-
minai in packets called vesicles. When an action potential
invades the axon
terminal, the serotonin molecules are released into the synaptic
gap and bombard the serotonin receptors on the postsynaptic, or
target, neuron. Inactivation of serotonin's synaptic action is
accomplished primarily through a process by which the mol ecules
are taken back into the axon terminal. Serotonin is destroyed by
the enzyme monoamine oxidase, and the resulting major catabolite of
se rotonin, 5-hydroxyindoleacetic acid (5-HIAA), can be measured in
brain tissue, urine, and cerebrospinal fluid.
In the mid-1960s, a seminal series of studies employing the
newly devel
oped technique of fluorescence his
tochemistry detailed the localization of the cell bodies and
axon terminals of the CNS neurons that utilize sero tonin as their
transmitter (6, 7).
When brain tissue is treated with
para-formaldehyde gas and then bombarded with ultraviolet light,
serotonin-containing neurons give off a yellow fluorescence (Fig.
4). This
technique revealed that CNS seroto nin was localized almost
exclusively within neurons of the raphe nuclei (small clusters of
cells on the midline of the brainstem) and their axon ter minals.
The total number of these neurons, at least in the rat brain, where
they have been studied most
intensively, is no more than 10,000-20,000. Thus, they represent
a very small proportion of all CNS neurons.
The serotonin-containing neurons (or raphe neurons) have widely
ramifying axons that are often sent out over
great distances. For example, those in the posterior portion of
the brainstem send their axons down the length of the spinal cord,
and those in the dor sal and median raphe nuclei of the anterior
brainstem send their axons into various portions of the forebrain.
In the forebrain, raphe neurons most
heavily innervate portions of the vi sual system and portions of
the limbic system, a group of structures known to be important in
emotional experi ence and expression (Fig. 5). Al
though most of these anatomical de tails have been worked out in
the rat,% they have relevance to humans be cause the pattern of
distribution of serotonin in the central nervous sys tem is fairly
constant across a variety of mammalian species.
Figure 4. Serotonin-containing neurons are
made visible through the use of fluorescence
histochemistry. In this photomicrograph of the dorsal raphe
nucleus in the midbrain of the rat, each oval yellow spot is the
cell body of a sero
tonin-containing neuron; the dark areas at the bottom on both
sides are large fiber tracts
(medial longitudinal fasciculi). The photomi crograph covers
approximately 1 mm of brain tissue from top to bottom.
(Photomicrograph courtesy of George Aghajanian, Yale Univer
sity.)
LSD and serotonin: Modern era The first report that LSD had a
sig nificant effect on brain neuro transmission was published in
1961 by Daniel Freedman (8). He found that a single injection of
LSD in creased the level of brain serotonin in the rat by 24
percent, while brom LSD, in even higher doses, failed to affect
brain serotonin. Thus, although LSD and brom-LSD had similar ef
fects on serotonin in the periphery, their central effects were
significantly different. In an important extension of this study,
Freedman and his col leagues reported that in addition to increases
in brain serotonin, LSD produced significant decreases in the brain
level of the major metabolite of serotonin, 5-HIAA (9). These
findings led to the hypothesis that LSD might inactivate or depress
the activity of serotonin neurons. The increase in brain serotonin
was attributed to its accumulation within neurons that were no
longer releasing it due to their inactivity.
Fortunately, because of the pioneer ing anatomical studies
employing fluorescence histochemistry, this hypothesis could be
directly exam ined by recording the electrical ac
tivity of serotonergic (i.e. serotonin containing) cell bodies
localized in
tight clusters in the brain stem of the rat. In 1968, George
Aghajanian in serted microelectrodes into the region of the dorsal
raphe nucleus of anes thetized rats and, after obtaining a stable
baseline sample of the cell's activity, he intravenously adminis
tered a low dose of LSD (10). Nor
mally, the discharge of these neurons in anesthetized rats is
slow (1-2 spikes/second) and regular. Following the injection of
LSD, as hypothesized, these serotonergic cells displayed an
abrupt and complete cessation of ac
tivity. By contrast, the activity of
neighboring nonserotonergic neurons was either unaffected or
slightly in creased by LSD. In a subsequent study, Aghajanian
reported that brom-LSD, even in much higher doses, had a much
smaller effect on
serotonergic neurons than LSD (11). Finally, through the use of
microion
tophoresis, which permits the direct application of small
amounts of sub stances to single neurons, Aghajanian demonstrated
that LSD depressed the activity of serotonergic neurons
through an effect directly on the cell body and had little
direct effect on
any of the other CNS neurons that were studied (12).
LSD's specific depression of raphe neuronal activity, in
conjunction with the uniformly inhibitory synaptic action of
serotonin in the forebrain, produces a disinhibition of raphe
target, or postsynaptic, neurons. Be cause the densest aggregations
of these target neurons are in areas of the brain that mediate
processing of visual or emotive information, we have an obvious
mechanism for ex
plaining the major affective, percep tual, and cognitive effects
of LSD. Thus, it is hypothesized that LSD acts to depress the
activity of seroto nin-containing neurons, which, through
disinhibition, cause a release of activity of neurons in the visual
system, the limbic system, and many other brain areas (13). This
model does not preclude the possibility that LSD may also exert a
direct action on other brain neurons, including sero tonin target
cells.
The argument that a significant pro
398 American Scientist, Volume 67
This content downloaded from 134.117.10.200 on Wed, 29 Jan 2014
06:32:04 AMAll use subject to JSTOR Terms and Conditions
-
portion of LSD's psychic effects can be attributed to its action
on sero
tonergic neurons is buttressed by the fact that other
hallucinogens which produce similar psychic effects, such as
psilocin, N,N dimethyltryptamine (DMT), and 5-methoxy DMT (5
MeODMT), also have an indole nu cleus structure (Fig. 6) and
depress raphe unit activity (11, 14). On the other hand,
psychoactive drugs that elicit different psychic effects, such
as
A-tetrahydrocannabinol (THC?the active component of marijuana)
or
amphetamine, neither have the indole nucleus structure nor
suppress the activity of serotonin-containing neurons. The few
human experiments that bear on this issue also provide indirect
support for the serotonin
hypothesis. For example, drugs that decrease brain serotonin
levels pot entiate the effects of LSD in humans, and drugs that
increase brain sero tonin levels decrease the effects.
It was against this backdrop of re search that we began our own
studies of LSD and, more generally, the brain serotonin system. We
felt that a number of important issues remained to be resolved.
Since we are ulti
mately interested in behavior, would the effects of
hallucinogenic drugs on
serotonergic neurons be the same in freely moving animals as
they were in anesthetized or immobilized animals? On a
microgram-for-microgram basis, why is LSD so much more potent than
other hallucinogens? What role, if any, do other neurotransmitter
systems play in the action of halluci nogenic drugs? What mediates
the rapid and dramatic diminution in LSD's psychological effects
that fol lows its repeated administration? Is the decrease in
activity of brain se rotonin neurons that is produced by LSD
approximated by any normal physiological condition?
An animal model for LSD Resolution of many of these issues
necessitated examining the behav ioral effects of LSD. Since the
use of human subjects in such experiments is precluded for ethical
reasons, we turned to animal experiments. How ever, using animals
in studies dealing with variables like hallucinations, which are
based exclusively on self report, raises the unanswerable question
of how to discern what a nonverbal organism is feeling, thinking,
or perceiving. Another tack,
hippocampus
PP- P --
...MFB
amygdala
Figure 5. The axons of serotonin-containing neurons often
project over great distances. Clusters of cell bodies (dots) of
serotonin
containing neurons, both the dorsal and me dian raphe nuclei as
well as the group of cells labeled B-9, are shown in cross section
through the midbrain of the rat. Axons from these cell bodies
ascend into the forebrain to make con
tact with structures such as the lateral genic ulate nucleus
(LGNV, a portion of the visual
system) and the hippocampus and amygdala (portions of the limbic
system). The medial forebrain bundle (MFB) is one of the major
pathways connecting the midbrain raphe neurons with their forebrain
target cells.
which obviates the need to impute an
underlying similarity of state, in volves using some aspect of
animal behavior (or physiology) as a model of the human variable.
Such models are founded on the assumption that their validity can
be established through demonstrating that changes in the animal
directly parallel changes in humans. The changes need not be
homologous, or even analogous, but merely must covary
systematically.
In order to qualify as an animal model for the actions of
hallucinogenic drugs in humans, a particular behavior
would have to change specifically in response to this class of
drug and no other, vary in frequency or magnitude in a
dose-dependent manner, be elicited by drug doses within the human
range, and closely parallel the
major parameters of the drug's effects in humans (e.g.
development of tol erance).
We chose the cat as our experimental animal because of its vast
behavioral repertoire and the ease with which any behavioral
changes could be ob served. Although previous studies had examined
the effects of LSD and related hallucinogens on behavior in a
variety of species, few of them had explored a complete dose range
or
closely examined the behavioral changes. Most important, none of
them had reported an effect that was demonstrated to be specific to
LSD. Accordingly, in our initial study, we administered LSD to cats
and at tempted to compile a complete and detailed record of all
major behavioral changes, with special attention to
Figure 6. The hallucinogenic drugs LSD, psi- have an indole
nucleus structure similar to that
locin, and DMT (N,N-dimethyltryptamine) of the serotonin
molecule (Fig. 3).
1979 July-August 399
This content downloaded from 134.117.10.200 on Wed, 29 Jan 2014
06:32:04 AMAll use subject to JSTOR Terms and Conditions
-
behaviors that might be evoked spe cifically by LSD.
We discovered that, in addition to increased frequency of head
and body shakes, grooming, investigatory be havior, and
hallucinatory-like re
sponses, two other behaviors, not
previously reported, were observed with high probability under
LSD (15, 16). Limb flicks and abortive grooming increased in
frequency in direct relation to the dose of LSD
(beginning from a baseline of essen
tially zero in saline-treated animals and progressively
increasing in ani mals treated with 2.5, 10, 25, and 50
Mg/kg of body weight). Limb flicking is a species-specific
behavior seen in normal cats almost exclusively in re
sponse to the presence of a foreign substance, such as water, on
the paw. The paw is lifted and rapidly and
repetitively shaken or snapped out ward from the body (Fig. 7).
In abor tive grooming, the cat orients to the body surface as if to
groom but does not perform the consummatory grooming response
(bite, lick, or
scratch) or performs it in midair. Apparently, as the cat begins
to groom, it becomes distracted and never finishes the grooming
sequence.
Whether the limb flicks, abortive grooming, head and body
shakes, and the like actually represent halluci nations is
irrelevant, since the model simply employs these measures as
parallels to hallucinations in hu mans.
The specificity of these behavioral changes is indicated by the
fact that they are never seen in response to single injections of
many other classes of psychoactive drugs, such as THC, amphetamine,
caffeine, atropine (an anticholinergic), and chlorphenira mine (an
antihistaminic). Nor were they seen in response to brom-LSD, the
nonhallucinogenic relative of LSD, or tryptamine, a nonhallucino
genic indole nucleus compound. Most important, however, these
behaviors
were elicited by hallucinogens that are structurally related to
LSD (DMT, 5-MeODMT, and psilocin) and known to depress the activity
of serotonin-containing neurons (17). They were also elicited by a
drug that blocks serotonin's action on its target neurons and by a
drug that decreases brain levels of serotonin by inhibiting its
synthesis. Furthermore, when LSD is administered to cats previ
ously treated with a serotonin-syn
thesis inhibitor, the behavioral effects of the two drugs are
synergistic (18). Therefore, besides helping to estab lish the
validity of the model, the data from these drug studies also
buttress the serotonin hypothesis of halluci nogenic drug
action.
In an attempt to extend the useful ness of this model to known
actions of hallucinogens in humans, we ob served something
surprising. If an adult human is given an effective dose of LSD (1
Mg/kg) on day 1 and then again on days 2,3, and 4, there will be a
marked decrease in the effective ness of the drug in producing
psychic change by the second day, and an al most complete loss of
effectiveness by day 4 (19, 20). This effect is called tolerance.
Paralleling these studies, we gave LSD to cats in an intermedi
Figure 7. When cats are given LSD, or other related
hallucinogenic drugs, they display several behaviors that are seen
exclusively in
response to these drugs. One of these is the limb flick, which
is seen in normal cats only in
response to the presence of a foreign substance on the paw. The
paw is raised from the ground and then rapidly shaken or flicked
away from the body.
ate (10 Mg/kg) or a high (50 Mg/kg) dose on day 1 and then
administered an additional 50 Mg/kg dose on day 2 (21). Much to our
surprise, the single 10 Mg/kg pretreatment produced a nearly
complete blockade of the be havioral effects of the 50 Mg/kg dose,
and the single 50 Mg/kg pretreatment made the second 50 Mg/kg dose
as
behaviorally ineffective as an injec tion of saline. In
subsequent behav ioral studies we found that this tol erance to
LSD, which was complete 24 hours after the initial injection,
actually had begun to develop within two hours after the injection,
at a time when the drug itself was still exerting its primary
effect! We shall return to this finding later, when we discuss
research on the mechanism mediating tolerance.
One of the most intriguing aspects of these studies was the fact
that the
magnitude of the peak behavioral effect of LSD was significantly
greater than that produced by psilo cin, DMT, 5-MeODMT, serotonin
receptor blockade, or serotonin de pletion. For example, a 50
fig/kg dose of LSD produced an average of ap proximately 40 limb
flicks per hour, whereas the other drugs, regardless of dose,
typically produced 5-10 limb flicks per hour. Since all of these
drugs, including LSD, were known to block serotonin
neurotransmission, we reasoned that the magnitude of LSD's
behavioral effect must be at tributable to some additional action.
Therefore, we began to explore the possibility that, in addition to
sero tonin, other neurotransmitters might be involved.
Studies in several other laboratories had indicated that LSD
also acted to mimic the action of the neuro transmitter dopamine
(22, 23). We confirmed this directly with electro physiological
studies that examined the effect of LSD on the activity of
dopaminergic neurons in the rat brain (24). Furthermore, using a
simple behavioral model in the rat, we found that another very
potent hallucino genic drug, DOM (2,5-dimethoxy-4
methylamphetamine), also had a
significant dopaminergic effect, whereas the indole nucleus
halluci nogens psilocin, DMT, and 5 MeODMT were virtually devoid of
dopaminergic action (25). Since a previous study had reported that
DOM also depressed the activity of serotonin-containing neurons,
we
hypothesized that DOM might be as behaviorally effective as LSD
in our cat model. This was confirmed with behavioral studies, in
which we found that DOM produced approximately 40 limb flicks per
hour (17).
Thus, the most potent hallucinogenic drugs may be those that
both inacti vate brain serotonin and mimic brain dopamine.
Serotonin inactivation may be necessary and sufficient for
hallucinogenesis (psilocin, DMT, and 5-MeODMT have serotonergic,
but no dopaminergic, action), while the dopaminergic action may
modulate the amplitude of the effect. This is supported by clinical
evidence. When patients who are having "bad trips" on LSD are given
antipsychotic drugs, which are potent dopamine-receptor blockers
(e.g. chlorpromazine), they
400 American Scientist, Volume 67
This content downloaded from 134.117.10.200 on Wed, 29 Jan 2014
06:32:04 AMAll use subject to JSTOR Terms and Conditions
-
typically report a diminution in the intensity of the experience
but a continuation of the hallucinatory activity. The dopaminergic
action of hallucinogenic drugs might also be relevant to the fact
that they are fre quently considered to mimic psy chosis (26),
since the preponderance of current biochemical evidence bearing on
schizophrenia indicates an overactivity in the brain dopamine
system. Thus, consideration of a
dopaminergic action of hallucinogenic drugs, in addition to
their action on brain serotonin, seems to explain what previously
appeared to be somewhat anomalous and disparate findings.
Serotonin neurons and behavior The next step in our research was
a crucial one for us. We were interested in recording the activity
of seroto nin-containing neurons in freely moving cats so that
behavior could be studied concomitantly. Over the past ten years,
in conjunction with Dennis
McGinty and Ronald Harper at UCLA, we have developed and re
fined a technique first used by James Olds. Basically, it involves
recording single neuron activity by means of bundles of insulated
microwires that can be advanced through the brain in small steps by
an attached mechani cal microdrive (Fig. 8). These elec trodes
differ from classical metal
microelectrodes in that they are flexible and have much larger
diam eters at the tips (32 vs. 0.1 to 1.0 ^ m). This method allows
us to maintain recordings from single cells even fol lowing rather
violent movements on the part of the cat (Fig. 9), and therefore
allows us to study the ac tivity of the same neuron over long
periods of time (often several days).
Prior to examining the effects of hal lucinogenic drugs on both
behavior and the activity of serotonin-con taining neurons in the
cat, we felt that it was important to provide a general context for
these data by first char acterizing the spontaneous activity of
these neurons across the sleep wakefulness-arousal continuum (27).
During a quiet waking state, sero
tonergic neurons discharge with the slow, regular pattern that
character izes the activity of serotonergic neu rons in
anesthetized rats. However, this activity can be modulated in both
directions, depending on the state of
MN
t ........
Xgg
vil*
Figure 8. A special apparatus was devised to record the
electrical activity of single neurons in awake, freely moving
animals. A cat is placed in a large, soundproofed box that is
electrically shielded. The animal's behavior can be con
tinuously monitored and recorded by videotape through a one-way
mirror in one wall of the chamber. Various gross electrodes, such
as
those for recording the EEG, as well as the microelectrodes, are
attached to a standard connector and the entire assembly is secured
to the skull with an acrylic. During an experi
ment, the cat is connected to various amplifiers and a polygraph
machine by means of a flexible cable attached to the connector on
the animal's head.
* * * * * * * * * *.*.* * *
*T *
Figure 9. An oscilloscope records the electrical
activity of a single serotonin-containing neuron in an awake,
freely moving cat, such as the one
shown in Fig. 8. In this 20-second sample of
activity, each vertical line is a neuronal dis
charge.
the cat. The activity increases during periods when the cat
becomes active, and briefly increases still further in
response to an arousing or alerting stimulus (e.g. a click or
flash). On the other hand, the activity of these cells decreases as
the cat becomes quies cent and drowsy (Fig. 10). Their ac
tivity decreases still further when the cat enters the first
phase of sleep, and finally ceases during the next stage of
sleep (termed REM sleep because of the appearance of rapid eye
move
ments). These data, in conjunction with other evidence, recently
led us to propose that the oft-noted phe nomenological similarity
of dreams (which occur most vividly in REM sleep) and drug-induced
hallucina tions might be mediated, in part, by a common
neurochemical event?
inactivation of central serotonergic neurotransmission (28).
With the basic characterization of the activity of these neurons
completed, we turned to directly examining the behavioral effects
of hallucinogenic drugs, while simultaneously recording the
activity of serotonin-containing neurons. We will first describe
our results with 5-MeODMT (29), be cause they were the most
straight forward. This drug produced dose dependent decreases in
the activity of serotonergic neurons and dose-de
pendent increases in specific behav iors (the limb flick
response is the
most reliable and easiest to quantify). Furthermore, the onset,
offset, and peak of the behavioral effects of 5 MeODMT were
temporally corre
1979 July-August 401
This content downloaded from 134.117.10.200 on Wed, 29 Jan 2014
06:32:04 AMAll use subject to JSTOR Terms and Conditions
-
f:;|j||||i?^|]:;:
llJpllllll
ISHHlill
lliiilllll
im 30 min
45 m?n
6-hrs
Figure 10. The activity of serotonin-containing neurons varies
dramatically as a function of the state of the animal. This figure
illustrates the
activity of the same serotonin-containing neuron across the
complete sleep-waking arousal continuum. Each strip shows 60 sec
onds of data, and each vertical line represents a cell discharge,
or spike. The highest rate of
discharge is seen during active waking; the ac
tivity slows somewhat during quiet waking and as the animal
becomes drowsy. A dramatic
slowing occurs as the animal enters sleep (slow-wave sleep-1
through slow-wave sleep-3), until the cell finally becomes silent
during the REM phase of sleep, when vivid dreaming typically takes
place.
Figure 11. An injection of LSD, in a dose of 50
Hg/kg of body weight, rapidly and dramatically affects the
activity of serotonin-containing neurons. Within 15 minutes the
cell's activity has significantly slowed, and it reaches its
nadir
approximately 45-60 minutes after the injec tion. This cell's
activity was maximally de creased by about 75% from the pre-drug
base line. The cellular activity returns to normal in 4-6
hours.
lated with the onset, offset, and peak of the changes in neural
activity.
By directly correlating behavioral changes with changes in the
activity of serotonin-containing neurons, this
study provided perhaps the strongest direct support for the
serotonin hy pothesis of hallucinogenic drug ac tion. One of the
most interesting as
pects was the finding that significant behavioral changes were
often asso ciated with small decreases in raphe unit activity (e.g.
15-20 percent). This indicated that rather subtle varia tions in
the outputs of these neurons
might have profound behavioral ef fects.
When we used this approach to ex amine the behavioral and
electro
physiological effects of LSD, the re sults were in general very
similar to those seen with 5-MeODMT, but with two important
differences. First, a high dose of LSD (50 Mg/kg) pro duced a
depression of serotonergic neuronal activity that lasted for ap
proximately 4 hours (Fig. 11), while the behavioral effects lasted
for at least 6-8 hours. Second, when the 50 jug/kg dose was
re-administered the next day, it produced little or no be havioral
effect, but the neuronal
change was as large as that on the previous day (30, 31).
These two somewhat anomalous findings?i.e. behavior outlasting
neuronal change and neuronal change without behavior?are probably
in terrelated. It appears that even while
LSD is exerting its primary depres sant effect on serotonin
neurons, it is also producing a change in some set of
postsynaptic neurons that will outlast the primary effect and
continue to mediate the behavioral change. This is supported by the
experiment de scribed above, in which we saw toler ance develop to
a single dose of LSD while this dose was still exerting its primary
behavioral effect.
General support of these notions about changes in postsynaptic
re ceptors came from experiments using gross behavioral measures in
rats, in which we observed that repeated ad ministration of LSD
markedly re duces the sensitivity of serotonergic target neurons to
LSD (32). We have also found that repeated adminis
402 American Scientist, Volume 67
This content downloaded from 134.117.10.200 on Wed, 29 Jan 2014
06:32:04 AMAll use subject to JSTOR Terms and Conditions
-
tration of LSD decreases the number of postsynaptic binding
sites for both serotonin and LSD and may affect the affinity of
serotonin for its re ceptor sites (33). Repeated adminis tration of
LSD might therefore result in a decreased capacity of LSD and/or
serotonin to stimulate neurons post synaptic to serotonergic
neurons. Thus, tolerance to LSD is not med iated by a change in
responsivity or
sensitivity on the part of serotonergic neurons, but an
important change seems to occur at the next neuron in the series.
What is not clear at present, however, is the exact nature of the
change in the postsynaptic neurons.
Why is the relationship between be havior and neuronal change
so
straightforward with 5-MeODMT and apparently so complex with
LSD? We do not yet have the answer to this at the cellular level,
but we do know that no other hallucinogenic drug approaches LSD in
its capacity to produce rapidly developing, long lasting, and
dramatic tolerance. This remains one of the more intriguing
unanswered questions regarding LSD's mechanism of action.
General implications Since animals, including humans, do not
commonly ingest plants contain ing hallucinogenic compounds, the
system of serotonin neurons must have evolved to subserve some adap
tive function other than mediating hallucinations. A large
behavioral literature indicates that serotonin
may play a general inhibitory role with respect to a variety of
sensory motor processes. Blockade of seroto
nin neurotransmission, whether by destruction of serotonergic
neurons, inhibition of its synthesis, or blockade of its receptors,
consistently produces an animal that is hypersensitive to virtually
all environmental stimuli and hyperactive in virtually all sit
uations. Through a general inhibitory function, serotonin neurons
may serve to modulate an organism's be havior and maintain it
within nar
rowly specified limits.
It seems reasonable that some basic role would be served by
neurons whose cell bodies are localized in the lower, more
primitive, portions of the brain and whose axons reach widely and
diffusely throughout the central nervous system. Consistent with
this
concept is the fact that these neurons appear to be among the
first to dif ferentiate during the development of the mammalian
CNS. This is also supported by the fact that serotonin neurons
discharge, in a wide variety of situations, with an almost
clocklike regularity?e.g. in anesthetized rats
(34), in awake, freely moving cats (27), and even when examined
in 400-yum thick slabs of rat brain tissue main tained in vitro
(35).
The slowness and regularity of the
activity of these neurons probably denotes a tonic neuronal
function, as
opposed to a rapid and variable ac
tivity, which would carry more in formation and subserve a more
dy namic function. A group of farsighted neuroanatomists, who were
among the first to study the raphe, specu lated that it is
"probably a primitive part of the brainstem which shows relatively
little differentiation during the phylogenetic ascent of the verte
brates. Correspondingly, one would be inclined to ascribe to it
relatively simple, but fundamental and impor tant tasks in the
function of the brain" (36).
Our studies of serotonin neurons indicate that as overall level
of motor
activity or arousal increases, so does the activity of these
cells. Recipro cally, as the animal becomes quies cent and drowsy,
the activity of these cells declines, possibly because this
inhibitory control is no longer neces sitated. As the animal enters
sleep, the cells fire still more slowly, and during REM sleep, a
state in which tonic muscle activity is abolished, the cells stop
firing. If we consider the action of hallucinogenic drugs in this
context, we see a fully awake animal with a brain serotonin system
func tioning as though the animal were
asleep. This may provide an impor tant insight into
understanding hal lucinations and perhaps, more gen erally, other
altered states of con sciousness. In a given behavioral sit uation,
an altered state of conscious ness may occur when a key brain
mechanism, such as the serotonin
system, functions in a manner that is
appropriate to a different behavioral situation.
We have tried to explain the behav ioral effects of LSD in terms
of
changes in the activity of single brain cells. As with any
complex behavioral process and any centrally acting drug,
a complete explanation will, of ne
cessity, involve a good deal more than the activity of one set
of neurons. There are also many unanswered questions about the
actions of LSD, such as the precise mechanism underlying tolerance.
An important sign of health and vitality in any sci entific field
is the ability to undergo change and revision. Neuroscience is at
present one of the more vigorous fields of scientific
investigation, and we therefore have no doubt that the story we
have told will undergo sig nificant modification and exten
sion.
References 1. A. Hofmann. 1968. Psychotomimetic
agents. In Drugs Affecting the Central Nervous System, ed. A.
Burger, pp. 184 85. Marcel Dekker.
2. L. E. Hollister. 1968. Chemical Psychoses, p. 34. Charles C.
Thomas.
3. J. H. Gaddum and K. A. Hameed. 1954.
Drugs which antagonize 5-hydroxytryp tamine. Brit. J. Pharm.
9:240-48.
4. D. W. Wooley and E. Shaw. 1954. A bio chemical and
pharmacological suggestion about certain mental disorders. PNAS
40:228-31.
5. A. Cerletti and E. Rothlin. 1955. Role of
5-hydroxytryptamine in mental disease and its antagonism to
lysergic acid deriv atives. Nature 176:785-86.
6. A. Dahlstrom and K. Fuxe. 1964. Evidence for the existence of
monoamine-containing neurons in the central nervous system. I.
Demonstration of monoamines in the cell bodies of brainstem
neurons. Acta Physiol. Scand. 62:Suppl. 232,1-55.
7. K. Fuxe. 1965. Evidence for the existence of monoamine
neurons in the central ner vous system. IV. The distribution of
monoamine terminals in the central ner vous system. Acta Physiol.
Scand. 64:
Suppl. 247, 41-85.
8. D. X. Freedman. 1961. Effects of LSD-25 on brain serotonin.
J. Pharm. Exp. Ther. 134:160-66.
9. J. A. Rosecrans, R. A. Lovell, and D. X. Freedman. 1967.
Effects of lysergic acid
diethylamide on the metabolism of brain
5-hydroxytryptamine. Biochem. Pharmac. 16:2011-21.
10. G. K. Aghajanian, W. E. Foote, and M. H. Sheard. 1968.
Lysergic acid diethylamide: Sensitive neuronal units in the
midbrain
raphe. Science 161:706-08.
11. G. K. Aghajanian, W. E. Foote, and M. H. Sheard. 1970.
Action of psychotogenic drugs on midbrain raphe neurons. J. Pharm.
Exp. Ther. 171:178-87.
12. H. J. Haigler and G. K. Aghajanian. 1974.
Lysergic acid diethylamide and serotonin: A comparison of
effects on serotonergic neurons and neurons receiving a seroton
ergic input. J. Pharm. Exp. Ther. 188: 688-99.
13. G. K. Aghajanian, H. J. Haigler, and J. L.
Bennett. 1975. Amine receptors in the CNS III.
5-Hydroxytryptamine in brain. In
1979 July-August 403
This content downloaded from 134.117.10.200 on Wed, 29 Jan 2014
06:32:04 AMAll use subject to JSTOR Terms and Conditions
-
Handbook of Psychopharmacology, Vol.
6, ed. L. L. Iversen, S. D. Iversen, and S. H.
Snyder. Plenum Press.
14. S. S. Mosko and B. L. Jacobs. 1977. Elec
trophysiological evidence against negative neuronal feedback
from the forebrain
controlling mid-brain raphe unit activity. Brain Res.
119:291-303.
15. B. L. Jacobs, M. E. Trulson, and W. C. Stern. 1976. An
animal behavior model for
studying the actions of LSD and related
hallucinogens. Science 194:741-43.
16. B. L. Jacobs, M. E. Trulson, and W. C. Stern. 1977.
Behavioral effects of LSD in the cat: Proposal of an animal
behavior
model for studying the actions of halluci
nogenic drugs. Brain Res. 132:301-14.
17. B. L. Jacobs, M. E. Trulson, A. D. Stark, and G. R.
Christoph. 1977. Comparative effects of hallucinogenic drugs on
behavior of the cat. Commun. Psychopharm. 1: 243-54.
18. M. E. Trulson and B. L. Jacobs. 1976. LSD acts
synergistically with serotonin deple tion: Evidence from behavioral
studies in cats. Pharm. Biochem. Behau. 4:231-34.
19. L. S. Cholden, A. Kurland, and C. Savage. 1955. Clinical
reactions and tolerance to LSD in chronic schizophrenia. J.
Nerv.
Ment. Dis. 122:211-21.
20. H. Isbell, R. E. Belleville, H. F. Fraser, A.
Wikler, and C. R. Logan. 1956. Studies on
lysergic acid diethylamide (LSD-25). I. Effects in former
morphine addicts and
development of tolerance during chronic intoxication. Arch.
Neurol. Psychiat. 76: 468-78.
21. M. E. Trulson and B. L. Jacobs. 1977. Usefulness of an
animal behavioral model in studying the duration of action of LSD
and the onset and duration of tolerance to LSD in the cat. Brain
Res. 132:315-26.
22. K. Von Hungen, S. Roberts, and D. F. Hill. 1974. LSD as an
agonist and antagonist at central dopamine receptors. Nature 252:
588-89.
23. L. Pieri, M. Pieri, and W. Haefely. 1974. LSD as an agonist
of dopamine receptors in the striatum. Nature 252:586-88.
24. G. R. Christoph, D. M. Kuhn, and B. L. Jacobs. 1977.
Electrophysiological evi dence for a dopaminergic action of
LSD:
Depression of unit activity in the sub stantia nigra of the rat.
Life Sei. 21: 1585-96.
25. M. E. Trulson, A. D. Stark, and B. L. Ja cobs. 1977.
Comparative effects of hallu
cinogenic drugs on rotational behavior in rats with unilateral
6-hydroxydopamine lesions. Eur. J. Pharm. 44:113-19.
26. M. B. Bowers. 1972. Acute psychosis in duced by
psychotomimetic drug abuse. Arch. Gen. Psychiat. 27:437-40.
27. M. E. Trulson and B. L. Jacobs. 1979.
Raphe unit activity in freely moving cats: Correlation with
level of behavioral arousal. Brain Res. 163:135-50. This con firmed
and extended the results of a pre vious study by D. J. McGinty and
R. M.
Harper, 1976, Dorsal raphe neurons: De pression of firing during
sleep in cats, Brain Res. 101:569-75.
28. B. L. Jacobs. 1978. Dreams and halluci nations: A common
neurochemical mech
anism mediating their phenomenological similarity. Neurosci.
& Biobehav. Rev. 2: 59-69.
29. M. E. Trulson and B. L. Jacobs. 1979. Ef fects of
5-methoxy-N,N-dimethyltrypta
mine on behavior and raphe unit activity in freely-moving cats.
Eur. J. Pharm. 54: 43-50.
30. M. E. Trulson and B. L. Jacobs. In press. Dissociations
between the effects of LSD on behavior and raphe unit activity
in
freely moving cats. Science.
31. M. E. Trulson, C. A. Ross, and B. L. Jacobs. 1977. Lack of
tolerance to the depression of raphe unit activity by lysergic acid
di
ethylamide. Neuropharm. 16:771-74.
32. M. E. Trulson, C. A. Ross, and B. L. Jacobs. 1976.
Behavioral evidence for the stimu lation of CNS serotonin receptors
by high doses of LSD. Psychopharm. Comm. 2: 149-64.
33. M. E. Trulson and B. L. Jacobs. In press. Alterations of
serotonin and LSD receptor binding following repeated
administration of LSD. Life Sei.
34. S. S. Mosko and B. L. Jacobs. 1974. Mid brain raphe neurons:
Spontaneous activity and response to light. Physiol. Behau.
13:589-93.
35. S. S. Mosko and B. L. Jacobs. 1976. Re
cording of dorsal raphe unit activity in vitro. Neurosci. Lett.
2:195-200.
36. E. Taber, A. Brodai, and P. Walberg. 1961. The raphe nuclei
of the brainstem in the cat. I. Normal topography and cyt?archi
tecture and general discussion. J. Comp.
Neurol. 116:161-88.
"Whatever happened to elegant solutions?"
404 American Scientist, Volume 67
This content downloaded from 134.117.10.200 on Wed, 29 Jan 2014
06:32:04 AMAll use subject to JSTOR Terms and Conditions
Article Contentsp. 396p. 397p. 398p. 399p. 400p. 401p. 402p.
403p. 404
Issue Table of ContentsAmerican Scientist, Vol. 67, No. 4
(July-August 1979), pp. 388-504Front MatterSigma Xi News [pp.
388-389, 503-504]Letters to the Editors [pp. 390,
392-395]Mechanisms of Action of LSD: The study of
serotonin-containing neurons in the brain may provide the key to
understanding drug-induced hallucinations and their relationship to
dreams and psychosis [pp. 396-404]Climate and the Ocean:
Measurements of changes in sea-surface temperature should permit us
to forecast certain climatic changes several months ahead [pp.
405-416]Self-Awareness in Primates: The sense of identity
distinguishes man from most but perhaps not all other forms of life
[pp. 417-421]The Therapy of Diabetes: Insulin therapy does not
always prevent the serious complications of diabetes; current
research is directed at more closely reproducing the functioning of
a healthy pancreas [pp. 422-431]The Feeding Mechanisms of Baleen
Whales: Since Robert Sibbald first described baleen whales in 1692,
we have come to distinguish three typesthe right whales, grazers on
copepods; the finner whales, engulfers of krill and fish; and the
gray whale, a forager of the sea bottom [pp. 432-440]Ancient
Hydraulic Techniques in the Chiapas Highlands: Strategies used by
the Maya in southeastern Mexico for efficient management of soil
and water resources provide evidence of cultural change and
population growth [pp. 441-449]Radiocarbon Dating with
Accelerators: Direct detection of 14C promises to revolutionize
radiocarbon dating [pp. 450-457]Nitrogen Fixation: Basic to
Applied: The increasing cost of nitrogen fertilizerwhich is
essential for high-yielding cerealshas triggered research into the
mechanism of natural nitrogen fixation [pp. 458-466]The Scientists'
BookshelfReview: untitled [pp. 467-467]Review: untitled [pp.
467-468]Review: untitled [pp. 468-469]Review: untitled [pp.
469-469]Review: untitled [pp. 469-470]Review: untitled [pp.
470-470]Physical SciencesReview: untitled [pp. 470-470]Review:
untitled [pp. 470, 472]Review: untitled [pp. 472-472]Review:
untitled [pp. 472-472]Review: untitled [pp. 472-472]Review:
untitled [pp. 472-472]Review: untitled [pp. 472-473]Review:
untitled [pp. 473-473]Review: untitled [pp. 473-473]Review:
untitled [pp. 473-473]
Earth SciencesReview: untitled [pp. 473-474]Review: untitled
[pp. 474-474]Review: untitled [pp. 474-474]Review: untitled [pp.
474-474]Review: untitled [pp. 474-475]Review: untitled [pp.
475-475]Review: untitled [pp. 475-476]Review: untitled [pp.
476-476]
Life SciencesReview: untitled [pp. 476-476]Review: untitled [pp.
476-477]Review: untitled [pp. 477-477]Review: untitled [pp.
477-477]Review: untitled [pp. 477-478]Review: untitled [pp.
478-478]Review: untitled [pp. 478-478]Review: untitled [pp.
478-478]Review: untitled [pp. 478-479]Review: untitled [pp.
479-479]Review: untitled [pp. 479-479]Review: untitled [pp.
479-479]Review: untitled [pp. 479-480]Review: untitled [pp.
480-480]Review: untitled [pp. 480-480]Review: untitled [pp.
480-480]Review: untitled [pp. 480-480]Review: untitled [pp.
480-481]Review: untitled [pp. 481-481]Review: untitled [pp.
481-481]Review: untitled [pp. 481-482]Review: untitled [pp.
482-482]Review: untitled [pp. 482-482]Review: untitled [pp.
482-483]Review: untitled [pp. 483-483]Review: untitled [pp.
483-483]Review: untitled [pp. 483-483]Erratum: Handbook of
Physiology, Section 9: Reactions to Environmental Agents [pp.
483-483]Review: untitled [pp. 483-483]Review: untitled [pp.
483-483]Review: untitled [pp. 483-484]Review: untitled [pp.
484-484]Review: untitled [pp. 484-484]Review: untitled [pp.
484-484]Review: untitled [pp. 484-484]Review: untitled [pp.
485-485]
Behavioral SciencesReview: untitled [pp. 485-485]Review:
untitled [pp. 485-485]Review: untitled [pp. 485-486]Review:
untitled [pp. 486-486]Review: untitled [pp. 486-486]Review:
untitled [pp. 486-487]Review: untitled [pp. 487-487]Review:
untitled [pp. 487-487]Review: untitled [pp. 487-488]Review:
untitled [pp. 488-488]Review: untitled [pp. 488-488]Review:
untitled [pp. 488-489]
Mathematics and Computer ScienceReview: untitled [pp.
489-489]Review: untitled [pp. 489-489]Review: untitled [pp.
489-489]Review: untitled [pp. 489-490]Review: untitled [pp.
490-490]Review: untitled [pp. 490-490]Review: untitled [pp.
490-490]
Engineering and Applied SciencesReview: untitled [pp.
490-490]Review: untitled [pp. 490-491]Review: untitled [pp.
491-491]Review: untitled [pp. 491-491]Review: untitled [pp.
491-491]Review: untitled [pp. 491-492]Review: untitled [pp.
492-492]Review: untitled [pp. 492-492]
History and Philosophy of ScienceReview: untitled [pp.
493-493]Review: untitled [pp. 493-493]Review: untitled [pp.
493-493]Review: untitled [pp. 493-494]
Books Received for Review [pp. 494-502]Back Matter