------------------~~~------------------

Dose-Response Study of N,N-Dimethyltryptaminein Humans1. Neuroendocrine, Autonomic, and Cardiovascular Effects

Rick]. Strassman, MD, Clifford R. Qualls, PhD

Background: To begin applying basic neuropharmaco-logical hypotheses of hallucinogenic drug actions to hu-mans, we generated dose-response data for intravenouslyadministered dimethyltryptamine fumarate's (DMT) neu-roendocrine, cardiovascular, autonomic, and subjectiveeffects in a group of experienced hallucinogen users.

Methods: Dimethyltryptamine, an endogenous mam-malian hallucinogen and drug of abuse, was adminis-tered intravenously at 0.05,0.1,0.2, and O.4mglkg to 11experienced hallucinogen users, in a double-blind, sa-line placebo-controlled, randomized design. Treat-ments were separated by at least 1 week.

Results: Peak DMT blood levels and subjective effectswere seen within 2 minutes after drug administration,and were negligible at 30 minutes. Dimethyltryptaminedose dependently elevated blood pressure, heart rate, pu-pil diameter, and rectal temperature, in addition to el-evating blood concentrations of f)-endorphin, cortico-

tropin, cortisol, and prolactin. Growth hormone bloodlevels rose equally in response to all doses of DMT, andmelatonin levels were unaffected. Threshold doses for sig-nificant effects relative to placebo were also hallucino-genic (0.2 mglkg and higher). Subjects with five or moreexposures to 3,4-methylenedioxymethamphetamine dem-onstrated less robust pupil diameter effects than thosewith two or fewer exposures.

Conclusions: Dimethyltryptamine can be adminis-tered safely to experienced hallucinogen users anddose-response data generated for several measures hy-pothesized under serotonergic modulatory control.Additional studies characterizing the specific mecha-nisms mediating DMT's biological effects may proveuseful in psychopharmacological investigations ofdrug-induced and endogenous alterations in brainfunction.

(Arch Gen Psychiatry. 1994;51:85-97)

HALLUClNOGENIC drugs re-liably induce a uniqueconstellation of subjec-tive effects in humans.They elicit perceptual il-

lusions and hallucinations, primarily vi-sual but often auditory. Mood effects rangefrom euphoria to panic to bland indiffer-ence. Somatic effects include dissocia-tion, relaxation, or tension. 1"Classic" hal-lucinogens include phenethylamines (eg,mescaline), indolealkylamines (eg, psilo-cybin and dimethyltryptamine [DMT]),and the lysergamides (eg, lysergic acid di-ethylamide [LSDj).2

These unique compoundswere the fo-cus ofintensive clinical research in the 1950sand 1960s. The discovery of LSD may ar-guably have been as important to the de-velopment of a "biological" psychiatry as

From the Departments ofPsychiatry (Dr Strassman) andMedicine (Dr Qualls), Schoolof Medicine, and theDepartment of Mathematicsand Statistics (Dr Qualls),University of New Mexico,Albuquerque.

was the contemporaneous discovery ofchlorpromazine. Hallucinogens' use inclinical research was associated with ac-ceptable risks when subjects, both psy-chiatric patients and normal controls, werecarefully screened, prepared, and fol-lowed Up.3 Clinical research with hallu-cinogens, however, became increasinglydifficult with their widespread abuse byyoung adults. Highly publicized "adversereactions" to hallucinogens, primarilyLSD,4 resulted in their placement into therestrictive schedule I of the Controlled

ARCH GEN PSYCHIATRYNOL 51, FEB 199485

SUBJECTS AND METHODS

The protocol was approved by the Scientific Advisory Com-mittee of the General Clinical Research Center (GCRC) andthe Human Research Review Committee of the Universityof ew Mexico School of Medicine, Albuquerque, the ewMexico State Pharmacy Board, the US Drug EnforcementAdministration, and the US Food and Drug Administra-non." Witnessed written informed consent was obtainedfrom all subjects. Confidentiality and anonymity were main-tained throughout the study.

SUBJECTS

We recruited only experienced hallucinogen users. All sub-jects but one were recruited by word of mouth; this lastsubject responded to an announcement sent to a graduatestudent training office at a local university.

Subjects were initially interviewed byone of us (R.].S.)to screen for medical andlor psychiatric conditions and as-sess level of experience with, and ability to manage dysphoricreactions to, hallucinogenic drugs. Subjects were droppedfrom further consideration if they were taking any medica-tions long-term, had a history of psychosis not secondaryto drugs or fever, or had a chronic medical disorder.

Prospective subjects then received a Structured Clini-cal Interview for DSM-IlI-R (SCID-R), Outpatient Version ,2+

administered by a trained psychiatric research nurse. First-degree relatives' psychiatric histories were obtained duringthe SCID- R. Diagnoses were confirmed by discussion withtwo research psychiatrists. We then performed a medical his-tory, physical examination, and laboratory screening tests(including a complete blood count with differential cell count,24-item chemistry panel, thyroid functions [including thy-rotropin] , urinalysis, and electrocardiogram).

The demographic characteristics of our subjects are givenin the Table. Subject 11 was dropped afterreceiving the non-blind (0.04- and-0.4-mglkg) doses and the 0.05-, 0.1-, and0.2-mglkg doses because he had a major depressive episodeprecipitated by multiple personal stressors that appeared un-related to his participation in the study. He responded to de-sipramine hydrochloride. The average (±SEM) age (exclud-ingsubject 11) was 41.5 ± l.5years. The group was high func-tioning, with only onesubjectnot being aprofessional or studentin a professional training program. There was a high incidenceof prior major depression in the group. Only one subject wassuffering any current Axis I disorder; this was an adjustmentdisorder resulting from an impending divorce. There also wasa high incidence of first-degree relatives suffering from de-pression andlor alcohol abuse. The number of exposures tohallucinogens ranged from six to "hundreds." Two subjectsdescribed a history of cocaine dependence/abuse, currentlyin remission; the remainder had little or no exposure to co-caine or amphetamines. Not all subjects had used marijuanaequally, although all had some experience with it.

METHYLENEDIOXYMETHAMPHETAMINE USE

Of interest was the breakdown of subjects by amount ofmethylenedioxymethamphetamine (MDMA, "Ecstasy") use.All but two had used MDMA at least once. Six of the 11 sub-jects had used MDMA five or more times; five had used ittwice or less. Methylenedioxymethamphetamine is a mam-malian serotonergic neurotoxin" with equivocal long-termeffects on human neuroendocrine andlor behavioral func-non." Thus, we were interested in assessing differential ef-fects of the serotonergic agonist, DMT, in these two groupsof users. For the sake of approximating equal cell sizes, wedescribed the six former subjects as having had high expo-sure and the latter five as having had low or no exposure.

DRUG

Dimethyltryptamine fumarate was prepared by David E. Ni-chols, PhD, Purdue University, West Lafayette, Ind. Pu-rity, determined by gas chromatography-mass spectrom-etry and high-performance liquid chromatography, wasgreater than 99.9%. It was prepared for clinical adminis-tration by the Inpatient Pharmacy of the University of NewMexico Hospital in vials containing 40 mglmL. Sterility andpyrogenicity testing revealed no abnormalities in a repre-sentative sample of vials. The dimethyltryptamine fuma-rate solution was drawn into a sterile tuberculin syringe atthe appropriate dose and diluted with sterile saline to 1.0mL before administration.

PROCEDURES

Nonblind Drug Administration

First, all subjects received nonblind administrations of 0.04and 0.4 mglkg of intravenous (IV) dimethyltryptamine fu-marate in the GCRC, usually on consecutive days. Previ-ous work " had demonstrated no tolerance to the cardio-vascular and subjective effects of a full 1M dose of DMTgiven twice daily for 5 consecutive days. Thus, we did notbelieve tolerance to DMT's effects would occur at this in-terval. We decided to administer the drug IV rather than1M as had been reported previously." This was because ofour initial dose-finding studies using a subject with expe-rience smoking DMT free base, the usual form and routeof administration by recreational users." The onset and in-tensity of effect of 1.0 mglkg of 1M dimethyltryptamine fu-marate was described by this subject as significantly lessrapidly developing and hallucinogenic than his previousexperience with the smoked drug. Further dose-finding stud-ies with him and another experienced DMT smoker estab-lished that 0.4 mglkg of IV dimethyltryptamine fumarateproduced a "rush" and hallucinogenic effect comparablewith those seen with a "full dose" of smoked free base.

The nonblind days served several purposes. They al-lowed us to assess the cardiovascular and subjective

ARCH GEN PSYCHIATRYNOL 51, FEB 199486

effects of a subclinical and fully hallucinogenic dose of di-methyltryptamine in a novel and potentially anxiety-provoking setting. These days also provided the researchteam and subjects an opportunity to become acquaintedand familiar with each other in the research center settingwithout the extensive blood drawing and use of the rectalthermistor that could exaggerate any potentially paranoidreactions. Subjects also were able to "calibrate" them-selves to what to expect regarding minimal and maximaleffects of the drug, one of the purposes of the double-blindstudy being whether they could distinguish among incre-mental doses of DMT. Finally, these nonblind days pro-vided subjects with the opportunity to drop out of the studybefore we had collected extensive data on them.

Subjects were admitted to the GCRC by9 AM after hav-ing fasted since midnight. An IV line was inserted into aforearm vein and kept patent with heparinized saline. Sub-jects were allowed to relax for 30 minutes before the drugwas given. Dimethyltryptamine was infused over 30 sec-onds and flushed with 5 mL of sterile saline over the next15 seconds. Blood pressure and heart rate (HR) were takenwith an automatic cuff several times before and frequentlyduring the 30 minutes following drug administration. Theresearch team (a registered nurse and psychiatrist) sat qui-etly on either side of the subject, attentive to verbal andnonverbal cues, but did not offer any direction or adviceunless absolutely necessary during the first several min-utes, the time of the peak effect of the drug. Sublingual ni-troglycerin, and IV diphenhydramine and diazepam, wereavailable at the bedside for sustained hypertension, aller-gic reactions, or excessive anxiety, respectively. As drug ef-fects resolved, discussion focused on the nature of thesubject's experiences. After subjects ate a snack or meal,they were discharged home.

No subject withdrew voluntarily after the nonblind days.One subject was withdrawn because ofa marked diastolic bloodpressure response (to >105 rnm Hg) to the 0.04-mg/kg dose.

Double-blind, Placebo-Controlled, RandomizedDrug Administration

A computer-generated randomization sequence was codedand kept by the Inpatient Pharmacy of the University of NewMexico Hospital. Subjects were admitted by 9 AM to a dimlylit «100 lux) room on the GCRC. Two IV lines were started:a small-gauge metal needle in one arm for drug administra-tion, and a large-gauge, three-holed plastic angiocatheterforblood withdrawal. A reusable rectal thermistor probe (YS-400, Yellow Springs Instruments, Yellow Springs, Ohio), con-figured to a physiological monitor (Vita-Log PMS-8, CPS Inc,Palo Alto, Calif), was inserted. The monitor sampled rectaltemperature every minu te, and the thermistor probe has anaccuracy of ±0.03°C. Blood pressure and HR (vital signs),and pupil diameter (measured by placing a card with blackcircles differing by I-mrn increments juxtaposed 10 to 15cm from the subject's eyes), were taken at 30, 15, and

2 minutes before drug administration. Blood samples for cor-ticotropin, l3-endorphin, prolactin, GH, cortisol, and mela-tonin levels were drawn 30 and 2 minutes before drug ad-ministration. Drug and saline flush were administered over45 seconds, and time 0 was when the saline flush was com-pleted. Doses were 0.05, 0.1, 0.2, and 0.4 mg/kg of dimeth-y~tryptamine fumarate and sterile saline placebo. Bloodsampleswere drawn, and pupil diameter and vital signs measured 2,5, 10, 15,30, and 60 minutes after drug administration. Studysessions occurred at an interval of at least 1 week and tookplacefromJanuary to September 1991. The one woman sub-ject was studied during the early follicular phase of her cycle.A negative serum pregnancy test drawn the night before herstudy days was always available before drug administration.

ASSAYS

Dimethyltryptamine (free base) blood levels were assayedusing the gas chromatography-mass spectrometry techniqueof Walker et al." Limit of detectability was 1 ngimL. Cor-ticotropin and l3-endorphin levels were assayed by irnmu-noradiometricassays (Nichols Diagnostic Laboratories, SanJuan Capistrano, Calif), and lower limits of detectabilitywere0.2 and 2.2 pmol/L, respectively. Limits of detectability forcortisol, GH, and prolactin radioimmunoassays (all from Di-agnostic Products, Los Angeles, Calif) were 6 nmollL, 1 fLgIL,and 1 fLgIL,respectively. Growth hormone and prolactin val-ues below 1 fLgILwere entered as 0.9 fLgIL.The melatoninradioimmunoassay/" limit of detectability was 5 pmol/L. Allassays were performed in duplicate.

STATISTICS

Variable values are givenasmean±SEM. All analyses wereper-formed usingPCSAS Version 6.03 (Cary,NC). Several derivedvariables were computed for analyses. Baseline values for bloodpressure, HR, and pupil diameter were derived by averagingthe three values obtained before drug administration (ie, - 30,-15, and - 2 minutes). Baseline temperature was computedby averaging values for 5 minu tes immediately preceding drugadministration. Hormone baseline values were computed byavcragmgresultsfrom the - 30- and - 2-minutesamples. Maxi-mum rise above baseline is indicated .6.Max;.6.AUCrepresentsthe area under the curve above baseline, calculated by the trap-ezoidal rule.

Our primary statistical tool was the univariate analysis ofvariance (ANOV A) with repeated measures, with dose as therepeated factor. Pairwise comparisons, when the overall FvaluegeneratedaP<.05, were performed using the "contrast" state-ment in the general linear models procedure in SAS.To assessapossiblemodulatoryrole ofmethylenedioxymethamphetamineexposure on dimethyltryptamine's effects on a particularvari-able, we performed a two-way univariate repeated-measuresANOVA using MDMA history as the independent variable.

Continued on next page

ARCH GEN PSYCHlATRYNOL 51, FEB 199487

MISSING DATA

We did not analyze blood hormone level results forthe one femalesubject in the group because her veinscollapsed at some point for almost everysession, andher blood data were incomplete. However,we did ana-lyze her autonomic and cardiovascular responses.

Pupil diameter was difficult to measure consis-tently, as subjects varied widely in their ability and/ordesire to open their eyes during acute drug intoxi-cation. Thus, these measurements were obtained atthe subjects' discretion. Additionally, several sub-jects had technically poor rectal temperature trac-ings because ofdifficultieswith probe placement, andthese data could not be used. Two subjects had un-satisfactory temperature data for the 0.05- and 0.2-mglkg sessions; one had no O.l-mglkg session data;and one had no temperature data from his placebosession. Finally, one subject refused the rectal therm-istor probe, and no temperature data were collectedfrom him. As SAS cannot perform repeated-measures ANOVAon subjects with missing data, weused a less robust one-way ANOVAfor the tempera-ture and pupil diameter data, with dose as the inde-pendent variable. Fisher's Least Significant Differ-ence Test for pairwise comparisons was performed ifthe overall F value generated a significance ofP< .05.To assess the effectofDMT exposure on MDMA'sef-fects on pupil diameter and rectal temperature, weperformed a two-way ANOVAwith dose and MDMAhistory as independent factors.

Substances Act. Human studies with these drugs effec-tively ended in the mid-1970s. Despite legal restrictionsand severe penalties for manufacture, possession, and dis-tribution of hallucinogens, their use among college stu-dents has held relatively steady over the past 20 years."

Basic research into the mechanisms of action of hal-lucinogens continued, however, and contributed to re-cent advances in serotonin (5-HT) receptor pharmacol-ogy.6 Recent behavioral, in vivo, and ligand binding datasuggest agonist or partial agonist effects at the 5-HT 2,

5-HT lA, and 5-HT ic receptors.":" Dopaminergic!' and nor-adrenergic'? effects have been described but not studiedrecently. ~

Hallucinogenic drugs have diffuse and relatively well-documented biological effects, presumably related to theirserotoninergic properties. Growth hormone (GH), pro-lactin, l3-endorphin, corticotropin, and cortisol levels areall affected by serotonergic stimuli. 13 Consistent with thepresumed serotonergic properties of the classic halluci-nogens, levels of several of these hormones increase withtheir administration. For example, human prolactin bloodlevels rise in response to dimethyltrypramine." and ratplasma corticoid levels are stimulated by LSD.15 Cardio-

vascular variables, core temperature, and pupillary di-ameter also are affected by serotonergic hallucinogens.Examples of these effects include DMT's hypertensive andmydriatic.!" LSD's pyretogenic and mydriaric.F'!"and psilocybin'S hypertensive and tachycardtc '?

properties.Dimethyltryptamine is a short-acting hallucino-

gen, first discovered in hallucinogenic Amazonian snuffs. 20

It was later synthesized and determined to be active byparenteral (intramuscular [1M]) administration only."Its discovery in human body fluids prompted a flurry ofinvestigations into its possible biosynthesis in man andits role as a putative endogenous psychotogen." Diffi-culties distinguishing peripheral levels ofDMT betweennormal controls and patients with endogenous psycho-ses led to a loss of interest in the function of this hallu-cinogenic tryptamine."

It has been impossible to test in humans hypoth-eses of hallucinogen mechanisms of action developed frombasic research. Systematic psychopharmacological in-vestigations are the necessary interface bridging recentneuropharmacological findings to human effects. Dose-response investigations applying current psychiatric re-search methods to assess hallucinogens' effects can gen-erate data providing the bases for more experimentalstudies, such as selective blockade with receptor sub-type antagonists.

Several properties recommend DMT as an appro-priate drug with which to renew such investigations. Itsshort duration minimizes prolonged dysphoric reac-tions in the potentially stressful environment of a mod-em clinical research center. The lack of a widespread "folk-lore" concerning its effects provides less bias andexpectations from subjects. Finally, DMT's role in hu-man brain function has yet to be explicated in normaland pathological states. That is, understanding effects andmechanisms of action ofDMT may shed light on the eti-ology and treatment of endogenous hallucinatory con-ditions.

We have performed a dose-response study of the neu-roendocrine, cardiovascular, autonomic, and subjectiveeffects of dimethyltryptamine fumarate using a double-blind, placebo-controlled, randomized design. This ar-ticle focuses on biological data generated by this study ..

RI:5ULTS

The major findings of this study are the following: (1)Intravenous dimethyltryptamine fumarate was fully hal-lucinogenic in doses of 0.2 and 0.4 mglkg. Effects werefelt nearly instantaneously, peaked within 2 minutes af-ter injection, and resolved within 20 to 30 minutes. Thetime course of DMT blood levels matched the march ofsubjective effects. (2) Multiple biological variables rosedose dependently in response to DMT. The time courseof elevations in blood levels of pro-opiomelanocortin

ARCH GEN PSYCHlATRYNOL 51, FEB 199488

* NOS indicates not otherwise specified. Subject 11 suffered a major depression during the study and was withdrawn before completing it. His data areincluded for illustrative purposes.

(POMC) peptides (corticotropin and ~-endorphin), HR,blood pressure, and pupil diameter paralleled subjectiveeffects and DMT blood concentrations. Elevations of pro-lactin and cortisol blood levels lagged 5 to 10 minutesbehind these changes. Core temperature and GH effectsdid not begin until the first set of perturbations had al-most resolved (15 to 30 minutes) and may not have peakedby 60 minutes after DMT administration.

DMT BLOOD LEVELS

There was no measurable DMT in blood samples beforedrug administration or after saline placebo administra-

tion for any subject except for subject 4. He had mea-surable DMT levels at one point before active drug ad-ministration and twice after placebo injection. His lowlevels in these circumstances are consistent with endog-enous levels in a study of normal volunteers using a simi-lar assay.'? However, he regularly took "Chinese herbs"and "homeopathic tinctures" as tonics, unlike any of ourother subjects.

Plasma levels ofDMT peaked at the first blood draw,2 minutes after the injection was completed, closely cor-responding to peak psychological effects. Generally, a dou-bling of dose resulted in a doubling of LlMaxvalues. Thesedata are displayed in Figure 1. As previously de-

ARCH GEN PSYCHIATRYNOL 51, FEB 199489

90 ,80

70

E 60g,g

50E~£ 40>,0=~is 30

20

10

2 5 10 15 30 60Time, min

Figure 1. Mean dimethyltryptamine (free base) values in 10 subjects afterfour doses of intravenous dimethyltryptamine fumarate (O.05mglkg[squares), 0.1 mglkg [triangles}, 0,2 mglkg [diamonds}, and OA mglkg[closed circles}) and saline placebo (open circles).

scribed," DMT levels varied widely among subjects, peakvalues ranging from 32 to 204 ng/mL after the 0.4-mglkg injection.

BASELINE EFFECTS

There were no differences among baseline values for anybiological variable among the five treatment conditions.

NEUROENDOCRINE EFFECTSPOMC PEPTIDES

The aAUC and aMax for both corticotropin and!3-endorphin rose dose dependently in response to di-methyltryptamine fumarate, with the 0.4-mglkg dose'seffects being statistically greater than all other doses forboth of these variables. The effects ofplacebo and the 0.05-and O.l-mglkg doses could not be separated statistically.Of interest was the stimulation of corticotropin and !3-en-dorphin levels by saline placebo. This has the appear-ance of a conditioned response to placebo injection inall these subjects who had previously received the some-what anxiety-provoking O.4-mglkg dosejionblind. Thesedata are displayed graphically in Figure 2 (corticotro-pin) and Figure 3 (!3-endorphin).

Prolactin

The aMax and aAUC for prolactin demonstrated a sig-nificant dose-dependent stimulatory effect of dimethyl-tryptamine fumarate. The O.4-mglkg dose raised both of

these variables Significantly more than placebo and the0.05- and O.l-mglkg doses. The effectsof0.4 and 0.2 mglkgon aMax prolactin could not be distinguished from eachother, although the O.4-mglkg dose's effects on aAUCwere Significantly greater than those of 0.2 mglkg. Theeffects of 0.2,0.1, and 0.05 mg/kg of dimethyltryp-tamine fumarate and of placebo on aAUC were not sta-tistically separable. These data are graphically presentedin Figure 4. Note that prolactin effects were slightly de-layed relative to the POMC peptides.

Cortisol

The aAU C and aMax for cortisol both demonstrated dose-dependent stimulatory effects of dimethyltryptamine fu-marate. The O.4-mglkg dose raised these two variablesmore than all other doses. The effects of 0.4 mg/kg onaAUC for cortisol could not be statistically distin-guished from the effects of 0.2 mglkg, however. Neithercould the 0.4-mglkg dose's effects on aMax for cortisolbe separated statistically from 0.2 and 0.05 mglkg. Ef-fects of the 0.05-, 0.1-, and 0.2-mglkg doses on aAUCand aMax were not statistically different, although theireffectswere numerically greater than those of saline. Thesedata are graphically presented in Figure 5.

Growth Hormone

There was a trend for DMT to produce a rise in aAUCfor GH. The aMax for GH rose in response to DMT ad-ministration. Although all doses of DMT raised GH lev-

50

40

sE 30Q.

C'0.

28~ 20u

10

2 5 10 15 30Time, min

60

Figure 2. Mean blood corticotropin values in 10 subjects after four dosesof intravenous dimethyltryptamine fumarate (0.05 mglkg [squares), 0.1mglkg [triangles}, 0.2 mglkg [diamondS}, and OA mglkg [closed circles})and saline placebo (open circles).

ARCH GEN PSYCHIATRYNOL 51, FEE 199490

els relative to saline placebo, effects of different doses couldnot be distinguished from each other. These data aregraphically displayed in Figure 6. Note the delay in GHeffects relative to the other neuroendocrine results.

Melatonin

We only assayed six subjects' blood samples for melatoninfrom their 0 .4-mglkg sessions. Utilizing one-sample t tests,there was no effect of 0.4 mglkg of dimethyltryptamine fu-marate on either ~Max or ~AUC for melatonin (data notgiven), so the remaining samples were not assayed.

AUTONOMIC EFFECTS

Rectal Temperature

The ~Max and ~AUC for temperature demonstrated aSignificant effect of dimethyltryptamine fumarate. ForMUC, the O.4-mglkg dose's effects were greater than thoseof 0.1 and 0.05 mg/kg and saline, but could not be dis-tinguished from those of 0.2 mglkg. The effects of 0.2,0.1, and 0.05 mglkg of drug and of saline could not bestatistically separated. For ~Max in temperature, the 0.4-and 0.2-mglkg doses' effects were greater than those pro-duced by 0.1 or 0.05 mg/kg or placebo. The ~Max ef-fects of 0.4 and 0.2 mglkg could not be statistically sepa-rated from each other; neither could the effects of the threelatter treatments be separated. These data are graphi-cally displayed in Figure 7, where the delayed effect rela-tive to other variables can be seen.

50

30

Time, min

602 5 10 15

Figure 3. Mean blood {3-endorphin values in 10 SUbjects after four dosesof intravenous dimethyltryptamine fumarate (0,05 mglkg [squares], 0,1mglkg [triangles], 0,2 mglkg [dismonos}, and 0,4 mglkg [closed circles])and saline placebo (open circles),

7

~"'-cO

~ 5'00:

3~--~.-~-------r--------------~2 5 10 15 30

Time, min60

Figure 4. Mean blood prolactin values in 10 subjects after four doses ofintravenous dimethyltryptamine fumarate (0,05 mglkg [squares], 0,1mglkg [triangles], 0,2 mglkg [diamondS], and 0.4 mglkg [closed circles])and saline placebo (open circles),

Pupil Diameter

As discussed in the "Procedures" section, our data for pu-pil diameter contained the most missing points and shouldbe interpreted cautiously, Missing values made a calcu-lation of LlAUC pupil diameter impossible for the moredisruptive 0,2- and 0.4- mglkg doses. Furthermore, thesedata were analyzed using A OVA without repeated mea-sures, a less robust analytic tooL

The LlMax pupil diameter demonstrated significanteffects of DMT. The 0.4-mg/kg dose increased pupil di-ameter more than all other treatments, Effects of pla-cebo and 0,05, 0,1, and 0,2 mglkg could not be distin-guished from each other. These data are displayedgraphically in Figure 8.

CARDIOV ASCULAR EFFECTS

Dimethyltryptamine dose dependently elevated 6.Max andLlAUC values for both HR and mean arteriai blood pres-sure (MAP). For ~Max HR, effects of 0.4 and 0.2 mglkgwere Significantly greater than those of 0.1 and 0.05 mglkgand saline; these latter three treatments' effects were notdistinguishable from each other. Neither were the ef-fects of 0.4 and 0.2 mglkg separable from each other. Ef-fects on 6.AUC HR were less clearly separable by dosebut followed the same pattern.

The O.4-mglkg dimethyltryptamine fumarate doseraised LlAUC MAP significantly more than 0.1 and 0,05mglkg and saline. These latter three treatments' effects couldnot be distinguished from each other. Neither were the

ARCH GEN PSYCHIATRYNOL 51, FEB 199491

600

" .....,/ -, -,,-,,-, -,,-, ,-, -,,-,,,,,,,•

500"""0§"0Vl.~u

400

300

30Time, min

2 5 10 15 60

Figure 5. Mean blood cortisol values in 10 subjects after four doses ofintravenous dimethyltryptamine fumarate (0.05 mglkg [squares]. 0.1mglkg [triangles], 0.2 mglkg [diamonds), and 0.4 mglkg [closed circles})and saline placebo (open circles).

effects of 0.2 andO.4 mglkgstatistically separable. TheO.4-and 0.2- mglkg doses caused significantly greater elevationsof~MaxMAP thanO.l andO.05 mglkg and saline; the twoformer doses' effects could not be distinguished from eachother. These data are graphically displayed in Figure 9(HR) and Figure 10 (MAP).

MDMA EFFECTS

For all of the biological variables under investigation, only~Max pupil diameter values were smaller across all dosesofDMT in the high-exposure compared with the low- orno-exposure MDMAgroup (high exposure, O.6::!::0.2mm;low or no exposure, l.S::!::0.2 mm; F=l1.3, P=.002).

SUBJECTIVE EFFECTS

A detailed description of subjective effects, and the devel-opment of and preliminary data generated by our new rat-ingscale, the Hallucinogen Rating Scale (HRS) ,are presentedin a companion article." Briefly, subjects felt the onset ofeffects before the 45-second infusion was completed. Peakhallucinogenic effects (with 0.2 and 0.4 rhglkg) occurred

.usually by the first blood draw and vital signs check, at 2minutes after injection, a useful temporal reference pointfor subjects. Subjects remained moderately intoxicated foranother 10 to 20 minutes. They felt relatively normal at 20to 30 minutes after injection. Several described a relaxed,"at ease" feeling for an additional 30 minutes.

All subjects but one experienced visual hallucinatoryphenomena at 0.2 and 0.4 mglkg of dimethyltryptamine fu-

marate. Brief, poorly formed auditory hallucinations wereexperienced by most subjects. Somatic effects consisted ofan intense rush, felt throughout the body or specifically inthe head, and bodily dissociation, usually associated witha transient anxious or fearful emotional reaction. Cogni-tive effects were not particularly profound; rather, an "ob-serving ego" was maintained, alert and attentive to the per-ceptualandsomatic effects of the drug. TheO.2-mglkgdose'sperceptual perturbations, rush, and disorganizing effectswere less intense than for 0.4 mglkg, and 0.2 mglkgwas thedose preferred by many subjects. Thea. I-mg/kg dose of di-methyltryptamine fumarate produced hallucinatory phe-nomena in only one subject (the woman in the study). So-matic effects predominated, and many subj ects disliked thisdose themost. The O.05-mglkg dose was mistaken for pla-cebo by several subjects. Ifeffects were noted, they were usu-ally somatic, with a warm, relaxed, and "floating" bodilysensation.

COMMENT

We safely administered graded doses of IV DMT, an en-dogenous hallucinogen and schedule Idrug of abuse, toa group of experienced hallucinogen users. Dose-dependent responses were observed for several neuro-endocrine, cardiovascular, and autonomic variables in arandomized, double-blind, saline placebo-controlled de-sign. Peak psychological effects paralleled the time courseof DMT blood levels, which appeared, peaked, and fellrapidly. Blood levels of the POMe peptides, corticotro-pin and l3-endorphin, closely followed the course of DMT

4

I.Bt'

/",'/

./;

2 5 10 15 30Time, min

60

Figure 6. Mean blood growth hormone values in 10 subjects after fourdoses of intravenous dimethyltryptamine fumarate (0.05 mglkg [squares),0.1 mglkg [triangles}, 0.2 mglkg [diamonds}, and 0.4 mglkg [closedcircles}) and saline placebo (open circles).

ARCH GEN PSYCHlATRYNOL 51, FEB 199492

levels and hallucinogenic effects, as did pupil diameterand cardiovascular responses. Cortisol and prolactin lev-els rose in a dose-dependent fashion and followed the firstgroup of variables' course by 5 to 10 minutes. Core tem-perature and GH stimulation were most delayed, al-though GH effects were not separable by dose. Stimula-tion of daytime melatonin levels was not seen.

The threshold for significant biological effects of di-methyltryptamine relative to placebo was also that for itshallucinogenic properties: 0.2 mglkg and higher. The 0.4-mglkg dose of IV dimethyltryptamine fumarate pro-duced blood levels and psychological effects compa-rable with those seen with administration of more thantwice higher doses ofIM dimethyltryptamine hydrochlo-ride or creatine sulfate.'! However, the IV route gener-ated a more rapid onset (nearly instantaneous) and off-set for these effects than the 1M route.

Dimethyltryptamine, found in Amazonian halluci-nogenic plant mixtures and human body fluids, meetselectrophysiological." pharmacological." and behav-ioral-" criteria for hallucinogenicity in lower animals. Ithad been used safely in humans in this country and Eu-rope.14.16.31Its short duration of action was believed ca-pable of providing a discrete, more manageable halluci-nogenic state in a clinical research setting than might the8- to 12-hour effects of lysergic acid diethylamide.

Hallucinogenic drugs are serotonergic agonists or par-tial agonists, in addition to having adrenergic and dopa-minergic properties. The 5-HT receptor subtypes of inter-

0.14

0.13

0.12

0.110.10

0.09~ 0.08~~

0.07a. 0.06E'"t- 0.05<l

0.04

0.030.020.Q1

0.00-0.01-0.02

10 20 40 50 6030Time, min

Figure 7. Mean rectal temperature values relative to baseline temperaturein 10 subjects atter tour doses of intravenous dimethyltryptamine fumarate(0.05 mg/kg (squares], 0.1 mg/kg [triangles]. 0.2 mg/kg [diamonds]. and0.4 mg/kg [closed circles]) and saline placebo (open circles). Baselinetemperature values were calculated by averaging the values at the 5minutes immediately preceding dimethyltryptamine administration.

7

E 6E~.

*E'"C"§-c,

4

3 ~~r-.-~-------'----------------.2 5 10 15 30

Time. min60

Figure 8. Mean pupil diameter values in 11 subjects after tour doses atintravenous dimethyltryptamine fumarate (0.05 mg/kg [squares]. 0.1mg/kg [triangles]. 0.2 mg/kg [diamonds). and OA mg/kg [closed circles])and saline placebo (open circles).

est include the 5-HT 21lcand5-HT lA.36•37We therefore choseour measured variables because they are believed regula tedor modulated to a Significant degree by serotonergic neu-rotransmission. These variables included multiple neuro-endocrine markers, blood pressure, HR, rectal temperature,and pupil diameter. As DMT has nearly equal affinities forboth the 5-HT 2 and 5-HT lAsubtypes," many of the effectsto be discussed may devolve from activation of either; how-ever, functional interactions in humans also have been de-scribed" and may be important.

Levels of the POMC peptide hormones, corticotro-pin and l3-endorphin, both rose dose dependently in oursubjects. They rose quite quickly, peak values being seenat5 minutes after injection. With doses of 0.1 mglkg andhigher, ~AUC and ~Max values for these two hormoneswere nearly identical, confirming their corelease." Theserises in j3-endorphin and corticotropin extend hu-man40.41data demonstrating serotonergic stimulatory ef-fects. Either 5-HTlA42 or 5-HT2

43 subtypes may be in-volved.

Prolactin levels rose in our subjects in a dose-dependent manner, confirming and extending the dataof Meltzer et al.!" wherein cyproheptadine inconsis-tently blocked DMT's stimulatory effect. Multiple ex-perimental procedures'v" have suggested a stimulatoryeffect of serotonergic neurotransmission on prolactin lev-els. Subtype delineation is not clearly understood, but both5-HTlA48and 5-HT2

49 receptors appear involved.Cortisol levels rose dose dependently in our sub-

jects, consistent with the robust elevation of corticotro-pin described above, and with earlier human data." Se-

ARCH GEN PSYCHIATRYNOL 51. FEB 199493

100 •"

", ,, ,~\

90,Ii ,:1\ ~:1 \

,,~ ~ 1 ,

, ,.0 /4\ ,~ II tI

,a: 80 ,~ II

,'"

,:r: ,

70

2 5 10 15 30 60

Time, min

Figure 9. Mean heart rate values (in beats per minute [bpm]) in 11subjects after four doses of intravenous dimethyltryptamine fumarate (0.05mglkg [squares), 0.1 mglkg [triangles}, 0.2 mglkg [diamonds}, and 0.4mglkg [closed circles]) and saline placebo (open circles).

rotonin receptor activation stimulates peripheral cortisollevels, most likely via corticotropin release," with rel-evant subtypes having been discussed previously. The cor-tisol effects in our subjects did not provide as useful adifferentiation among doses as the POMC hormones.

Hallucinogen-induced hyperthermia in humans?'may occur by means of central'? and/or peripheral'" 5-HT2

activation. Dimethyltryptamine's hyperthermic effects areprobably mediated by the 5-HT Z subtype, as 5-HT lA ago-nists are hypothermic in man.? The delayed time courseof the temperature rise seen with DMT is of interest andmay reflect more complex homeostatic mechanisms con-trolling this variable than those with a more rapid timecourse.

Previous human studies have described variable ef-fects of longer-acting hallucinogenic drugs on cardio-vascular variables, including hypotension and hy-pertension, and tachycardia and br adycard ia.?"Dimethyltryptamine's excitatory cardiovascular prop-erties confirm and extend previous human data" andsuggest 5-HT2 agonism. The 5-HTz agonists usuallyelicit a rise in blood pressure, without reliable in-creases in HR.56,57 In contrast, 5-HT1A agonists gener-ally produce a drop in blood pressure, again withoutconsistent HR effects.58

Dimethyltryptamine resembles other classic hallu-cinogens in its mydriatic properties'"; this pupillary di-latation may be 5-HT2 mediated.F As the pupillary di-ameter data were both the least reliably measured (ie, wedid not use automated equipment, which also could havemeasured more dynamic features of pupillary function)

and contained the most missing values, they must be con-sidered preliminary.

We found no dose dependency of DMTs stimula-tion of blood GH levels. Meltzer et all4 demonstrated acyproheptadine-sensitive stimulatory effect of one DMTdose on peripheral GH levels in humans. Perhaps if wehad continued drawing samples for 2 hours, as did Melt-zer et al, we might have found better separation of ef-fects by dose .

Subjective effects of Iv dimethyltryptamine were con-sistent with previous reports of parenterally adminis-tered DMTI6,31,61 and were marked by perceptual, disso-ciative, affective, and cognitive alterations typical of theclassic hallucinogens. Onset from the IV route was nearlyinstantaneous, rather than taking several minutes to de-velop, and effects were shorter than previously pub-lished 1M data. However, this time course is quite simi-lar to reports of DMT smoked free base." The threesubjects in the group with experience smoking DMT freebase all volunteered that the IV route was somewhat morerapid and intense in onset and effects than smoking thecompound. In this regard, DMT is actively transportedacross the blood-brain barrier in the rat62 A more de-tailed account of these effects and their quantification bya new rating scale are described in a companion article.V

Dimethyltryptamine, as a pu tative directly acting 5-HTagonist in humans, can be most appropriately comparedwith o-chloro-z-fl-piperazinyl) pyrazine (MK-212) andm-chlorophenylpiperazine (m-CPP), although onlym-CPPhas been given intravenously to humans. Neither drug pro-duces the typical hallucinogenic syndrome in normal sub-

110

105

100

"":r:EE 95c.:«:;:

90

85

80 -L-,-,-,__,- ,- -,30 602 5 10 15

Time, min

Figure 10. Mean arterial blood pressure (MAP) values in 11 subjects afterfour doses of intravenous dimethyltryptamine fumarate (0.05 mglkg[squares), 0.1 mglkg [triangles}, 0.2 mglkg [diamonds}, and 0.4 mglkg[closed circles]) and saline placebo (open circles).

ARCH GEN PSYCHIATRYNOL 51, FEE 199494

j ects; however, behavioral effects may be more disruptivein psychiatric samples. 63.64 Both m-CPP and MK-212 havecomplex effects at multiple 5-HT receptor subtypes, in ad-dition to effects on other neurotransmitters.P

Intravenous m-CPP's biological effects are similar tothose ofDMT. Cortisol and GH stimulation is comparablein magnitude to DMT's, but the prolactin response to DMTis smaller." Similarly, GH responses are delayed relativeto those of prolactin and cortisol. Hyperthermic effects ofIV m-CPP also are relatively delayed and are greater thanthose seen with DMT. 67However, temperature data werecollected longer for IV m-CPP than for DMT, and we mayhave missed a rise of similar magnitude. Behaviorally, IVm-CPP's peak effects occur somewhat later than those ofDMT,67 perhaps related to DMT's previously mentionedactive transport into the brain.

OrallyadministeredMK-212elevatesprolactinandcor-tisol (butnot GH) blood levels, 64blood pressure, and bodyternperature/" AsMK-212 has never been given intravenouslyto humans, a comparison of the temporal sequence of itseffects with IV DMT or m-CPP is not possible.

The only biological effect demonstrated in subjects witha greater cumulative exposure to the serotonergic neuro-toxin MDMA relative to the group with less exposure wasa decreased pupillary response to all doses ofDMT. This isopposite to what might be expected based on augmentedbehavioral effects of hallucinogens in amine-depleted'" ordenervated'? animals. We are unable to explain this find-ing, except to note that our pupil diameter data were theleast complete and obj ectively measured, and were analyzedby ANOVA without repeated measures.

We erroneously hypothesized that daytime melatoninlevels migh tbe eleva ted secondary to DMT administra tion,based on the presence of 5-HT 2 recepjors on pineal tissue"and stimulation by DMT of a requisite enzyme for melato-nin synthesis. 72 However, the lack of a stimulatory effect ofDMT on melatonin is relevant regarding the specificity ofour neuroendocrine results. Previous work'? showed thatmelatonin levelsdoubled after a "stressful" long-distance race,although !3-endorphin levels rose to only half those notedin the present DMT study. These two sets of data make itunlikely that a "stress" effect (ie, purely sympathetic acti-vation) drove our neuroendocrine variables; tlso, either mela- (tonin levels would have risen or !3-endorphin effects wouldhave been much greater.

CAVEATS

Our sample of subjects had a heavy loading of personal(50%) and family (67%) histories of affective disorderand/or alcohol abuse, both of which may be associatedwith abnormalities in serotonergic responsivity. 74.75Thus,the data generated from this study may not be general-izable to strictly "normal" subjects. In addition, nine of11 subjects had been divorced, there being a total of 17divorces in this group. Although not specifically as-

sessed, this high rate of divorce may imply a certain de-gree of impulsivity within our subjects, another condi-tion in which serotonergic abnormalities have beendescribed."

One subject's blood pressure response to a screen-ing, nonblind dose of 0.04 mglkg caused him to be with-drawn from further participation. After completing thisstudy, while screening additional subjects for subse-quent protocols, another subject with a history of bor-derline essential hypertension (currently asymptom-atic) also had a robust diastolic response to 0.05 mglkgof dimethyltryptamine fumarate and was withdrawn fromfurther sessions. Thus, we recommend that subjects witha history of hypertension, even though currently asymp-tomatic, not be considered for IV DMT studies. In addi-tion, all prospective subjects must first be screened withthe low, minimally psychoactive 0.05-mglkg dose; dias-tolic blood pressure responses to more than 100 mm Hgcontraindicate exposure to higher doses of DMT.

CONCLUSION

Human data are necessary to begin bridging basic neu-ropharmacological and clinical psychopharmacologicalconcepts of hallucinogenic drugs' effects and mecha-nisms of action." Using careful screening and prepara-tion of subjects, attentive yet unobtrusive supervision ofthe acute drug intoxication, and follow-up, DMTwas safelyadministered to a group of experienced hallucinogen us-ers and normative dose-response data generated. The mostsalient biological variables, based on our data, seem tobe corticotropin, prolactin, core temperature, blood pres-sure, HR, and pupil diameter. Pupil diameter studies, ifperformed, would benefit from the use of automated equip-ment.

We believe the importance 0f these data derive from atleast two perspectives. The first is that of understanding ef-fects and mechanisms of action of these intriguing drugs.Many ofDMT's biological effects in psychiatrically healthyhallucinogen users are now much better characterized. Theproperties ofother classicandnonclassic hallucinogens (suchas 1-(2,5-dimethoxy-4-methylphenyl)-2-aminopropane(DaM) now can be assessed for their degree(s) of similar-ity. In addition, although serotonergic mechanisms are be-lieved primary in mediating hallucinogens' effects, these datacan help validate this hypothesis byperforrning human stud-ies using selective 5-HT receptor subtype antagonists. Ex-periments also can be designed to assess the relative con-tributions of noradrenergic and dopaminergic neurotrans-mission.

Second, DMT might be considered an additional se-rotonergic probe of neuroendocrine and behavioral vari-ables in research attempting to characterize neurotrans-mission in normal volunteers and psychiatrically illsubjects with presumed serotonergic abnormalities. Al-though the hallucinogenic effects of DMT are not sepa-

ARCH GEN PSYCHIATRYNOL 51, FEB 199495

rable from its threshold effects (relative to placebo) onthe multiple biological factors presented herein, these ef-fects, if judiciously managed, should not present insur-mountable difficulties. mCPP, for example, increases sev-eral patient populations' symptoms but remains a valuabletool for human serotonergic studies. However, it mustbe emphasized that until DMT's effects are mechanisti-cally characterized in normal subjects, there would notbe sufficient data to conclude that altered responses toDMT necessarily implicated altered 5-HT function.

Accepted for publication November 4, 1992.This investigation was supported by grant R03-

DA06524 from the National Institute on Drug Abuse, Rock-ville, Md; the Scottish Rite Foundation for SchizophreniaResearch, NMJ, Lexington, Mass; grant RR00997-13 fromthe National Institutes of Health General Clinical Re-search Center, Rockville; the Scott Rogers Fund of the Uni-versity of New Mexico, Albuquerque; and University of NewMexico Department of Psychiatry research funds.

We thank David E. Nichols, PhD, for synthesis of thedimethyltryptamine fumarate used in this study; CynthiaGeist, RN, and the nursing staff of the GCRC, 5E, Uni-versity of New Mexico Hospital, for nursing support; KatyZownir-Brazis, RN, for performing the SCID-R inter-views; E. Jonathan Lisansky, MD, for assistance in diag-nosis; Joy McLeod, Alberta Bland, and Susan Lee for labo-ratory assistance; David S. Schade, MD, for developmentof the dimethyltryptamine assay; and Richard Dorin, MD,and Russel Reiter, PhD, for assistance in performing as-says.

Reprint requests to Department of Psychiatry, Uni-versity of New Mexico, 2400 Tucker Ave NE, Albuquer-que, NM 87131-5326 (Dr Strasstrlan).

REFERENCES

1. Freedman OX. On the use and abuse of LSD. Arch Gen Psychiatry. 1968;18:330-347.

2. Nichols 0, Oberlender R, McKenna D. Stereochemical aspects of hallucino-genesis. In: Watson R, ed. Biochemistry and Physiology of Substance Abuse,Ifl. Boca Raton, Fla: CRC Press Inc; 1991:1-39.

3. Cohen S. Lysergic acid diethylamide: side effects and complications. J NervMent Dis. 1960;130:30-40.

4. Strassman R. Adverse reactions to psychedelic drugs: a review of the litera!ture. J Nerv Ment Dis. 1984;172:577-595.

5. Pope H Jr, lonescu-Pioggia M, Aizley H, Varma D. Drug use and life style amongcollege undergraduates in 1989: a comparison with 1969 and 1978. Am J Psy-chiatry. 1990;147:998-1001.

6. Peroutka S, Snyder S. Multiple serotonin receptors: differential binding of [3H15-hydroxytryptamine, [3Hllysergic acid diethylamide and [3Hlspiroperidol. MolPharmacal. 1979;16:687-690.

7. Glennon R, Titeler M, McKenney J. Evidencefor 5-HT, involvement in the mecha-nism of action of hallucinogenic agents. Life Sci. 1985;35:2505-2511.

8. Spencer 0 Jr, Glaser T, Traber J. Serotonin receptor subtype mediation of theinteroceptive discriminative stimuli induced by 5-methoxy-N,N-dimethyltryptamine. Psychopharmacology. 1987;93:158-166.

9. Yagaloff K, Hartig P. "'I-lysergic acid diethylamide binds to a novel seroton-ergic site on rat choroid plexus epithelial cells. J Neurasci. 1985;5:3178-3183.

10. Buckholtz N, Zhou 0, Freedman OX, Potter W. Lysergic acid diethylamide (LSD)

administration selectively downregulates serotonin, receptors in rat brain. Neu-rapsychopharmacology. 1990;3:137-148.

11. Ahn H, Makman M. Interaction of LSD and other hallucinogens with dopamine-sensitive adenylate cyclase in primate brain: regional differences. Brain Res.1979;162:77-88.

12. Horita A, Hamilton A. Lysergic acid diethylamide: dissociation of its behavioraland hyperthermic effects by DL-alpha-methyl-p-tyrosine. Science. 1969;164:78-79.

13. Van de Kar L. Neuroendocrine pharmacology of serotonergic (5-HT) neurons.Ann Rev Pharmacal Toxicol. 1991;31:289-320.

14. Meltzer H, Wiita B, Tricou B, Simonovic M, Fang V. Effects of serotonin pre-cursors and serotonin agonists on plasma hormone levels. In: Ho B, SchoolarJ, Usdin E, eds. Serotonin in Biological Psychiatryc New York, NY: Raven Press;1980:117-139.

15. Halaris A, Freedman OX, Fang V. Plasma corticoids and brain tryptophan afteracute and tolerance dosage of LSD. Life Sci. 1976;17:1467-1472.

16. Sai-Halasz A, Brunecker G, Szara S. Dimethyltryptamine: ein neues Psychoti-cum. Psychiatr Neural. 1958;135:285-301.

17. Horita A, Dille J. The pyretogenic effect of lysergic acid diethylamide. Science.1954;120:1110-1111.

18. Greiner T, Burch N, Edelberg R. Psychopathology and psychophysiology ofminimal LSD-25 dosage: a preliminary dosage-response spectrum. Arch Neu-ral Psychiatry. 1958;79:208-210.

19. Isbell H. Comparison of the reactions induced by psilocybin and LSD-25 inman. Psychopharmacologia. 1959;1 :29-38.

20. Fish M, Johnson N, Horning E. Piptadenia alkaloids: indole bases of P. per-egrina (L.) Benth. and related species JAm Chem Soc. 1955;77:5892-5895,

21. Barker S, Monti J, Christian S. N,N-dimethyltryptamine: an endogenous hal-lucinogen. Int Rev Neurabiol. 1981;22:83-110.

22. Gillin J, Kaplan J, Stillman R, Wyatt R. The psychedelic model of schizophre-nia: the case of N,N-dimethyltryptamine. Am J Psychiatry. 1976;133:203-208.

23. Strassman R. Human hallucinogenic drug research in the United States: a present-day case history and review of the process. J Psychoactive Drug. 1991 ;23:29-38.

24. Spitzer R, Williams J, Gibbon M. Structured Clinical Interview of DSM-III-R:Outpatient Version. New York, NY: Biometric Research Dept, New York StatePsychiatric Institute; 1987.

25. McKenna 0, Peroutka S. Neurochemistry and neurotoxicity of 3,4-methylenedioxymethamphetamine (MDMA, 'Ecstasy'). J Neurochem. 1990;54:14-22.

26. Price L, Ricaurte G, Krystal J, Heninger G. Neuroendocrine and mood re-sponses to intravenous L-tryptophan in 3,4-methylenedioxymethamphetamine(MDMA) users. Arch Gen Psychiatry. 1989;46:20-22.

27. Stafford P. Psychedelics Encyclopedia. Rev ed. Boston, Mass: Houghton Mif-flin Co; 1982:308-331.

28. Walker R, Mandel L, Kleinman J, Gillin J, Wyatt R, VandenHeuvel W. Improvedselective ion monitoring mass-spectrometric assay for the determination ofN,N-dimethyltryptamine in human blood utilizing capillary column gas chro-matography. J Chramatogr. 1979;162:539-546.

29. Webley G, Mehl H, Willey K. Validation of a sensitive direct assay for melatoninfor investigation of circadian rhythms in different species. J Endocrinol. 1985;106:387-394.

30. Angrist B, Gershon S, Sathananthan, Walker R, Lopez-Ramos B, Mandel L,VandenHeuvel W. Dimethyltryptamine levels in blood of schizophrenic patientsand control subjects. Psychopharmacology. 1976;47:29-32.

31. Kaplan J, Mandel L, Stillman R, Walker R, VandenHeuvel W, Gillin J, Wyatt R.Blood and urine levels of N,N-dimethyltryptamine following administration ofpsychoactive doses to human subjects. Psychopharmacologia. 1974;38:239-245.

32. Strassman RJ, Qualls CR, Uhlenhuth EH, Kellner R. Dose-response study ofN,N-dimethyltryptamine in humans, II: subjective effects and preliminary re-sults of a new rating scale. Arch Gen Psychiatry. 1994;51 :98-1 08.

33. Aghajanian G, Foote W, Sheard M. Action of psychotogenic drugs on singlemidbrain raphe neurons. J Pharmacal Exp Ther. 1970;171:178-187.

34. Deliganis A, Pierce P, Peroutka S. Differential interactions of dimethyltrypt-amine (DMT) with 5-HT'A and 5-HT, receptors. Biochem Pharmacal. 1991;41:1739-1744.

35. Jenner P, Marsden C, Thanki C. Behavioral changes induced by N,N-dimethyltryptamine in rodents. Br J Pharmacal. 1980;69:69-80.

36. Pierce P, Peroutka S. Hallucinogenic drug interactions with neurotransmitter

ARCH GEN PSYCHIATRYNOL 51, FEB 199496

receptor binding sites in human cortex. Psychopharmacology. 1989;97:118-122.

37. Sanders-Bush E, Breeding M. Choroid plexus epithelial cells in primary culture:a model of 5HT,c receptor activation by hallucinogenic drugs. Psychopharma-cology. 1991;105:340-346.

38. Charig E, Anderson I, Robinson J, Nutt 0, Cowen P. l-tryptophan and prolactinrelease: evidence for interaction between 5-HT, and 5-HT2 receptors. Hum Psy-chopharmacol. 1986;1 :93-97.

39. Vale W, Spiess J, Rivier C, Rivier J. Characterization of a 41-residue ovinehypothalamic peptide that stimulates secretion of corticotropin and s-enoor-phin. Science. 1981 ;213:1394-1396.

40. Lewis 0, Sherman B. Serotonergic regulation of adrenocorticotropin secretionin man. J Clin Endocrinol Metab. 1984;58:458-462.

41. Petraglia F, Facchinetti F, Martignoni E, Nappi G, Volpe A, Genazzani A. Sero-tonergic agonists increase plasma levels of [3-endorphin and [3-lipotropin inhumans. J Clin Endocrinol Metab. 1984;59:1138-1142.

42. Anderson I, Cowen P, Grahame-Smith D. The effects of gepirone on neuro-endocrine function and temperature in humans. Psychopharmacology. 1990;100:498-503.

43. Calogero A, Bernardini R, Margioris A, Bagdy G, Gallucci W, Munson P, Tam-arkin L, Tomai T, Brady L, Gold P, Chrousos G. Effects of serotonergic agonistsand antagonists on corticotropin-releasing hormone secretion by explanted rathypothalami. Peptides. 1989;10:189-200.

44. Kato Y, Nakai Y, Imura H, Chichara K, Ohgo S. Effect of 5-hydroxytryptophan(5-HTP) on plasma prolactin levels in man. J Clin Endocrinol Metab. 1974;38:695-697.

45. Asnis G. Eisenberg J, van Praag H, Lemus C, Friedman J, Miller A. The neu-roendocrine response to fenfluramine in depressives and normal controls. BioiPsychiatry. 1988;24:117-120.

46. Demisch L, Neubauer M. Stimulation of human prolactin secretion by mesca-line. Psychopharmacology. 1979;64:361-363.

47. Aloi J, Insel T, Mueller E, Murphy D. Neuroendocrine and behavioral effects ofmCPP in rhesus monkeys. Life Sci. 1984;34:1325-1331.

48. Smith C, Ware C, Cowen P. Pindolol decreases prolactin and growth hormoneresponses to intravenous L-tryptophan. Psychopharmacology. 1991;103:140-142.

49. Van de Kar L, Lorens S, Urban J, Bethea C. Effect of selective serotonin (5-HT)agonists and 5-HT2 antagonists on prolactin secretion. Neuropharmacology.1989;28:299-305.

50. Calogero A, Bagdy G, Szemeredi K, Tartaglia M, Gold P, Chrousos G. Mecha-nisms of serotonin receptor agonist-induced activation of the hypothalamic-pituitary-adrenal axis in the rat. Endocrinology. 1990;126:1888-1894.

51. Snyder SH, Faillace L, Hollister L. 2,5-Dimethoxy-4-methyl-amphetamine (STP):a new hallucinogenic drug. Science. 1967;158:669-670.

52. Won S, Lin M. 5-Hydroxytryptamine receptors in the hypothalamus mediatethermoregulatory responses in rabbits. Naunyn Schmiedebergs Arch Pharma-cal. 1988;338:256-261.

53. Cohen M, Fuller T, Wiley S. Evidence for 5-HT, receptors mediating contractionin vascular smooth muscle. J Pharmacal Exp Thet. 1981 ;21:421-425.

54. Denber H, Merlis S. Studies on mescaline, VI: therapeutic aspects of the mescaline-chlorpromazine combination. J Nerv Ment Dis. 1955;122:463-469.

55. Turner W, Merlis S. Effect of some indolealkylamines on man. Arch NeuralPsychiatry. 1959;81:121-129.

56. Dabire H, Chaouche- Teyara K, Cherqui C, Fournier B, Laubie M, Schmitt H.Characterization of 001, a putative 5-HT, receptor agonist in the rat. Eur J •Pharmacal. 1989;168:369-374.

57. Rittenhouse P, Bakkum E, Van de Kar L. Evidence that the serotonin agonist,001, increases renin secretion and blood pressure through both central andperipheral5-HT, receptors. J Pharmacal Exp Ther. 1991;259:58-65.

58. Dreteler G, Wouters W, Saxena P. Systemic and regional hemodynamic effectsof the putative 5-HTlA receptor agonist flesinoxan in the cat. J Cardiovasc Phar-macal.1989;14:770-776.

59. Klee G, Bertino J, Weintraub W, Callaway E III. The influence of varying dosageon the effects of lysergic acid diethylamide (LSD-25) in humans. J Nerv MentDis. 1961;132:404-409.

60. Millson 0, Haworth S, Rushton A, Wilkinson 0, Hobson S, Harry J. The effectsof a 5-HT2 antagonist (ICI 169,369) on changes in waking EEG, pupillary re-sponses and state of arousal in human volunteers. Br J Clin' Pharmacol. 1991;32:447-454.

61. Rosenberg 0, Isbell H, Miner E. Comparison of placebo, N-dimethyltryptamine,and 6-hydroxy-N-dimethyltryptamine in man. Psychopharmacology. 1963;4:39-42.

62. Yanai K, Ido T, Ishiwata K, Hatazawa J, Takahashi T, Iwata R, Matsuzawa T. Invivo kinetics and displacement study of a carbon-11-labeled halluclnouen, N,N-["C]dimethyltryptamine. Eur J Nucl Med. 1986;12:141-146.

63. Seibyl J, Krystal J, Price L, Woods W, Heninger G, Charney D. 5-HT functionin the biochemical and behavioral responses to mCPP in healthy subjects andschizophrenics. Soc Neurosci Abstr. 1989;15:1236.

64. Lowy M, Meltzer H. Stimulation of serum cortisol and prolactin secretion inhumans by MK-212, a centrally active serotonin agonist. Bioi Psychiatry. 1988;23:818-828.

65. Murphy 0, Lesch K, Aulakh C, Pigott T. Serotonin-selective arylpiperazines withneuroendocrine, behavioral, temperature and cardiovascular effects in hu-mans. Pharmacal Rev. 1991 ;43:527-552.

66. Charney 0, Goodman W, Price L, Woods W, Rasmussen S, Heninger G. Se-rotonin function in obsessive-compulsive disorder. Arch Gen Psychiatry. 1988;45:177-185.

67. Murphy 0, Mueller E, Hill J, Tolliver T, Jacobsen F. Comparative anxiogenic,neuroendocrine and other physiologic effects of m-chlorophenylpiperazine givenintravenously and orally to healthy volunteers. Psychopharmacology. 1989;98:275-283.

68. Lee H, Bastani B, Friedman L, Ramirez L, Meltzer H. Effect of the serotoninagonist, MK-212, on body temperature in schizophrenia. Bioi Psychiatry ..1992;31 :460-470.

69. Appel JB, Freedman OX. Chemically-induced alterations in the behavioral ef-fects of LSD-25. Biochem Pharmacal. 1964;13:861-869.

70. Nisbet AF, Marsden CA. Increased behavioural response to 5-methoxy-N,N-dimethyltryptamine but not to RU-24969 after intraventricular 5,7-dihydroxytryptamine administration. Eur J Pharmacal. 1984;104:177-180.

71. Ebadi M, Govitrapong P. Neural pathways and neurotransmitters affecting mel-atonin synthesis. J Neural Transm. 1986;21 (suppl):125-158.

72. Hartley R, Smith J. The activation of pineal hydroxy-indole-O-methyl-transferase by psychotomimetic drugs. J Pharm Pharmacol. 1973;25:751-752.

73. Strassman R, Appenzeller 0, Lewy A, Qualls C, Peake G. Increase in plasmamelatonin, [3-endorphin, and cortisol after a 28.5-mile mountain race: rela-tionship to performance and lack of effect of naltrexone. J Clin Endocrinol Metab.1989;69:540-545.

74. Price L, Charney 0, Delgado P, Heninger G. Serotonin function and depression:neuroendocrine and mood responses to intravenous L-tryptophan in de-pressed patients and healthy comparison subjects. Am J Psychiatry. 1991;148:1518-1525.

75. Lee M, Meltzer H. Neuroendocrine responses to serotonergic agents in alco-holics. Bioi Psychiatry. 1991;30:1017-1030.

76. Fishbein 0, Lozovsky 0, Jaffe J. Impulsivity, aggression, and neuroendocrineresponses to serotonergic stimulation in substance abusers. Bioi Psychiatry.1989;25:1049-1066.

77. Freedman OX. Hallucinogenic drug research: if so, so what? Pharmacal Bio-chem Behav. 1986;24:407-415.

ARCH GEN PSYCHlATRYNOL 51, FEB 199497