Dihydrobenzofuran analogues of hallucinogens. 4. Mescaline derivatives

Article abstract of DOI:10.1021/jm970219x

Dihydrobenzofuran and tetrahydrobenzodifuran functionalities were employed as conformationally restricted bioisosteres of the aromatic methoxy groups in the prototypical hallucinogen, mescaline (1). Thus, 4-(2- aminoethyl)-6,7-dimethoxy-2,3-dihydrobenzofu

Full text of DOI:10.1021/jm970219x

J . Med. Chem. 1997, 40, 2997-3008
2997
Articles
Dih yd r oben zofu r a n An a logu es of Ha llu cin ogen s. 4.1 Mesca lin e Der iva tives2
Aaron P. Monte,† Steve R. Waldman, Danuta Marona-Lewicka, David B. Wainscott,‡ David L. Nelson,‡
Elaine Sanders-Bush,§ and David E. Nichols*
Department of Medicinal Chemistry and Molecular Pharmacology, School of Pharmacy and Pharmacal Sciences,
Purdue University, West Lafayette, Indiana 47907, Lilly Research Laboratories, Eli Lilly and Company,
Indianapolis, Indiana 46285, and Department of Pharmacology, School of Medicine, Vanderbilt University Medical Center,
Nashville, Tennessee 37232
Received March 31, 1997X
Dihydrobenzofuran and tetrahydrobenzodifuran functionalities were employed as conforma-
tionally restricted bioisosteres of the aromatic methoxy groups in the prototypical hallucinogen,
mescaline (1). Thus, 4-(2-aminoethyl)-6,7-dimethoxy-2,3-dihydrobenzofuran hydrochloride (8)
and 1-(8-methoxy-2,3,5,6-tetrahydrobenzo[1,2-b:5,4-b′]difuran-4-yl)-2-aminoethane hydrochlo-
ride (9) were prepared and evaluated along with 1 for activity in the two-lever drug
discrimination (DD) paradigm in rats trained to discriminate saline from LSD tartrate (0.08
mg/kg). Also, 1, 8, and 9 were assayed for their ability to displace [3H]ketanserin from rat
cortical homogenate 5-HT2A receptors and [3H]8-OH-DPAT from rat hippocampal homogenate
5-HT1A receptors. In addition, these compounds were evaluated for their ability to compete
for agonist and antagonist binding to cells expressing cloned human 5-HT2A, 5-HT2B, and 5-HT2C
receptors. Finally, agonist efficacy was assessed by measurement of phosphoinositide hydrolysis
in NIH 3T3 cells expressing the rat 5-HT2A or 5-HT2C receptors. Although 1 fully substituted
for LSD in the DD assays (ED50 ) 33.5 µmol/kg), neither 8 nor 9 substituted for LSD, with
just 50% of the rats administered 8 selecting the drug lever, and only 29% of the rats
administered 9 selecting the drug lever. All of the test compounds had micromolar affinity for
the 5-HT1A and 5-HT2A receptors in rat brain homogenate. Curiously, the rank order of affinities
of the compounds at 5-HT2A sites was opposite their order of potency in the behavioral assay.
An evaluation for ability to stimulate phosphoinositide turnover as a measure of functional
efficacy revealed that all the compounds were of approximately equal efficacy to serotonin in
5-HT2C receptors. At 5-HT2A receptors, however, 8 and 9 were significantly less efficacious,
eliciting only 61 and 45%, respectively, of the maximal response. These results are consistent
with the proposed mechanism of action for phenethylamine hallucinogens, that such compounds
must be full agonists at the 5-HT2A receptor subtype. In contrast to the 2,5-dimethoxy-
substituted phenethylamines, where rigidification of the methoxy groups had no deleterious
effect on activity, the loss of activity in the 3,4,5-trioxygenated mescaline analogues may suggest
that the 3 and 5 methoxy groups must remain conformationally mobile to enable receptor
activation.
In tr od u ction
structure to hallucinogen-like, or psychedelic,6 ac-
tivity.7-12 In addition, the psychic-behavioral effects
produced by mescaline were frequently used as a
benchmark against which the effects of other hal-
lucinogens were compared, and the potencies of novel
compounds prepared in SAR investigations were for
many years expressed in terms of “mescaline units”.13,14
Peyote (Lophophora williamsii), a member of the
Cactaceae family, represents one of the earliest known
hallucinogenic plants, the use of which was exclusively
limited to the New World.3 The main constituent of
peyote, mescaline (3,4,5-trimethoxy-â-phenethylamine,
1), was first identified by Heffter in 1896 and was
recognized as the active hallucinatory principle on the
basis of rudimentary pharmacological testing and self
experiments.4 Following the first chemical synthesis of
mescaline in 1919,5 this alkaloid served as the proto-
typical structure in more than 75 years of structure-
activity relationship (SAR) studies linking molecular
* Address correspondence to: Dr. David E. Nichols, Dept. of
Medicinal Chemistry and Molecular Pharmacology, School of Pharmacy-
-RHPH, Purdue University, West Lafayette, IN 47907. Phone: (765)
494-1461 FAX: (765) 494-1414. email: drdave@pharmacy.purdue.edu.
† Current address: Department of Chemistry, Cowley Hall, Uni-
versity of Wisconsin-La Crosse, La Crosse, WI 54601.
‡ Lilly Research Laboratories.
The successive SAR studies of 1-like hallucinogens
have ultimately led to agents of general structure 2 (R
) H, CH3; X ) alkyl, halo, alkoxy, alkylthio, etc.) which
currently are among some of the most potent hal-
lucinogens known.8,12,15,16 As such, compounds 2 have
§ Vanderbilt University Medical Center.
X Abstract published in Advance ACS Abstracts, August 15, 1997.
S0022-2623(97)00219-7 CCC: $14.00 © 1997 American Chemical Society

2998 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 19
Monte et al.
Sch em e 1a
a
Reagents: (a) (COCl)2, PhH; CH2N2, Et2O; (b) AcOH; (c) Pd/C, H2, EtOH; (d) BCl3, CH2Cl2; (e) H2CO, (CH3)2NH (aqueous), heat;
CH3I, iPrOH; (f) KCN, H2O, heat; (g) CH3I, K2CO3, Me2CO, reflux; (h) Pd/C, H2, EtOH, HCl.
typically been shown to have low nanomolar affinities
for serotonin 5-HT2A and 5-HT2C receptors,17-25 the
primary sites believed to be involved in the mediation
of the unique behavioral effects of hallucinogenic
drugs.26-29 Given the high potency and site-selectivity
of 2 derivatives, they have served as convenient proto-
types for further SAR studies using rigid analogs to
probe the topography of the serotonin 5-HT2 agonist
binding sites.30,31 Recently, compounds 3,32 4,33,34 5,35
and 6 and 736 have all been synthesized and evaluated
pharmacologically in our laboratory for hallucinogen-
like behavioral activity and serotonin receptor affinities,
to study the possible “active” conformations of the
aromatic methoxy groups of 2 at the serotonin agonist
binding site. We have reported that the dihydrobenzo-
furan structure serves as an effective, conformationally-
restricted bioisostere of the aromatic methoxy group,
and that the orientation of alkoxy groups exemplified
by compound 7 most likely represents the active con-
formation of these groups during serotonin receptor
activation.36
and our established synthetic methodology to the mes-
caline prototype 1, a structure containing three aromatic
methoxy groups. Because 1 has been shown to have
action at both serotonin and dopamine (DA) recep-
tors,37,38 and because hallucinogenic activity has gener-
ally been correlated with the overall hydrophobicity of
the drug molecule,39-41 it was hypothesized that 8 and
9 could ultimately have potential as probes of 5-HT, DA,
or other central neurotransmitter receptors. As mol-
ecules with increased hydrophobicity relative to 1, they
might also be useful in evaluating the role of this
physicochemical property in the production of hal-
lucinogenic effects. Thus, we now report on the syn-
thesis and preliminary pharmacological evaluation of
rigid analogs of mescaline in which one (8) or two (9) of
the three aromatic methoxy groups of 1 are tethered
into rotationally constrained dihydrofuran rings.
Ch em istr y
The dihydrobenzofuranyl mescaline analog 8 was
synthesized successfully as shown in Scheme 1. Con-
struction of the furan ring was accomplished using the
method of J ung and Abrecht.42 Thus, 2,3,4-trimethoxy-
benzoic acid was converted to its corresponding diazo-
methylketone which was cyclized to the benzofuranone
10 upon treatment with acetic acid. Catalytic hydro-
genation of the ketone then afforded the parent dihy-
drobenzofuran 11 in excellent yield. Initial unsuccessful
attempts to elaborate the side chain of 11 involved the
standard sequential procedure of electrophilic formyla-
tion, condensation of the aldehyde with nitromethane,
and subsequent reduction of the nitrostyrene with
As an extension of our previous efforts with dihy-
drobenzofuranyl analogs of hallucinogenic phenethyl-
amines, we sought to apply a similar design rationale

Dihydrobenzofuran Analogues of Hallucinogens
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 19 2999
Sch em e 2a
a
Reagents: (a) H2CO, HCl (g), concd HCl, dioxane; (b) NaCN, DMSO; (c) concd HCl, heat; (d) BH3, THF; H2O; (e) BBr3 (excess),
CH2Cl2; CH3OH; (f) K2CO3, Me2CO, reflux; (g) CH3I, K2CO3, Me2CO, reflux; (h) Cl2CHOCH3, SnCl4, CH2Cl2; H2O, ice; (i) CH3NO2, NH4OAc;
(j) LiAlH4, THF; H2O.
LiAlH4 as described previously.43 Because the 5-posi-
tion of 11 was more susceptible to electrophilic attack,
this position was first blocked by aromatic bromination
which occurred in a facile and near-quantitative man-
ner. (The bromine atom would then be removed during
the subsequent LAH reduction step). Attempts to
formylate the 5-bromo compound at the 4-position,
however, gave complex mixtures of products that were
extremely difficult to purify, and this method was
abandoned. A possible explanation for the difficulties
with this formylation may lie in the fact that some
demethylation of the 7-methoxy group may have oc-
curred in the presence of the Lewis acid catalyst. The
steric crowding of this group causes it to lie out of plane,
thus decreasing the overlap between the 7-methoxy lone
pair electrons and the aromatic π system. Conse-
quently, the 7-methoxy group is more basic than a
planar aromatic ether and would be expected to be more
labile to cleavage by Lewis acids such as the tin(IV)
chloride used in this formylation step.
Fortunately, an alternative method for constructing
phenethylamines existed in the literature that was well
suited for mescaline-like compounds.44,45 This method-
ology effects direct aromatic substitution at the 4-posi-
tion as shown in Scheme 1. Thus, the 7-methoxy group
of 11 was regioselectively demethylated using boron
trichloride in dichloromethane.46 The ring was func-
tionalized at the 4-position of 12 under mild Mannich
conditions, and the resulting crude Mannich base was
quaternized to the methiodide salt 13 by addition of
iodomethane to the reaction mixture.47 Displacement
of trimethylamine from 13 with cyanide ion produced
the phenylacetonitrile 14 in good yield. This phenol was
then remethylated with iodomethane to give 15, and
catalytic hydrogenation of the nitrile in the presence of
HCl afforded the desired target compound 8 in good
overall yield.
synthesized successfully as shown in Scheme 2. Chlo-
romethylation of 1,2,3-trimethoxybenzene according to
the method of Dallacker et al.48 proceeded in fair yield
to give the bis-chloromethyl compound 16. Both chlo-
rine atoms of 16 were displaced with cyanide using KCN
in DMSO to afford the bis-nitrile 17.49 After acidic
hydrolysis to the corresponding bis-acetic acid 18,
reduction using the selective borane-THF method of
Yoon et al.50 gave diol 19 in reasonable yield considering
the difficulty of workup for this water-soluble interme-
diate.
The two dihydrofuran rings were formed simulta-
neously employing a methodology developed recently in
our laboratory.34,36 Thus, essentially conducting three
steps in one pot, 19 was first tri-demethylated and
dibrominated with excess BBr3, quenching with metha-
nol to avoid aqueous workup of the highly water soluble
and unstable pyrogallol intermediate. Removal of metha-
nol under vacuum and immediate reflux with potassium
carbonate in acetone led to ring closure to give the
cyclized, phenolic intermediate. Once the complete
disappearance of the dibromo compound was noted by
TLC, iodomethane was added to the reaction mixture
to O-methylate the remaining phenolic oxygen, giving
20 in excellent overall yield from 19. Final elaboration
of the sidechain then proceeded according to standard
methods.43 Friedel-Crafts formylation of 20 using tin-
(IV) chloride catalyst51,52 gave aldehyde 21 in modest
yield. The low yield was likely due to the partial
cleavage of the out-of-plane 8-methoxy group, as dis-
cussed above for the synthesis of the monofuran deriva-
tive 8. Nitrostyrene 22 was formed by condensation of
21 with nitromethane using ammonium acetate as the
catalyst, and the final target compound 9 was obtained
in good yield by LAH reduction of 22.
The tetrahydrobenzodifuran mescaline analog 9 was

3000 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 19
Monte et al.
Ta ble 1. Results of the Drug Discrimination Studies in
LSD-Trained Rats and Radioligand Competition Studies at
[3H]Ketanserin-Labeled Rat 5-HT2A and
[3H]8-OH-DPAT-Labeled Rat 5-HT1A Receptors
DD studies
Ki (nanomolar)
ED50
drug (µmol/kg)
5-HT2A
sites
5-HT1A
sites
95% CI
na
1
8
9
33.5
(20.9-53.5) 8-12 5500 ( 600 6900 ( 800
50% @ 57.8 NSb
29% @ 55.2 NSb
8-12 2500 ( 300 6200 ( 400
8-12 NTc NTc
6-15 2100 ( 200 6900 ( 800
23 50% @ 16.2 NSb
a
Number of animals tested at each dose. b NS ) no substitution.
c Not tested in this preparation.
P h a r m a cology
Compounds 1, 8, and 9 were initially evaluated in the
two-lever drug discrimination assay in a group of rats
trained to discriminate the effects of ip injections of
saline from those of LSD tartrate (0.08 mg/kg), accord-
ing to methods described previously (Table 1).53 For the
compounds not fully substituting for LSD in this assay,
the percentage of rats selecting the drug lever at the
given dosage levels is reported in the table. Addition-
ally, 1, 8, and 9 were tested for their ability to compete
for radioligand binding to 5-HT1A and 5-HT2A receptor
sites (Table 1).53 Briefly, the ability of test compounds
to displace 0.75 nM [3H]8-OH-DPAT from rat hipp-
campal homogenate and 0.75 nM [3H]ketanserin from
rat frontal cortex homogenate was measured. In addi-
tion, the compounds were evaluated in cells expressing
cloned human 5-HT2A, 5-HT2B, and 5-HT2C receptors
that were labeled either with [3H]antagonists or radio-
labeled agonists (Table 2). Finally, as a measure of
efficacy, compounds were assessed for their ability to
stimulate phosphoinositide hydrolysis in cell lines ex-
pressing the rat 5-HT2A and 5-HT2C receptors (Figures
1 and 2).
F igu r e 1. Effects of mescaline 1 (0), 8 (1), 9 (b), and 23 (O)
on phosphoinositide hydrolysis in NIH 3T3 fibroblasts ex-
pressing the rat 5-HT2A receptor. The curves are representa-
tive of experiments carried out on the same day. The mean
Emax values (3-5 determinations) for 1, 23, 8, and 9 were 92,
96, 61, and 45%, respectively, of the maximum 5-HT response.
The mean EC50 values ( SEM were 4 ( 1, 4 ( 2, 3 ( 1, and
2 ( 1 µM, respectively.
Resu lts a n d Discu ssion
The results of the drug discrimination (DD) studies
in LSD-trained rats and the results of the radioligand
competition experiments at [3H]ketanserin-labeled rat
5-HT2A and [3H]8-OH-DPAT-labeled rat 5-HT1A recep-
tors are presented in Table 1. In the behavioral assays,
mescaline 1 was the most potent of the test compounds,
substituting for LSD with an ED50 of 33.5 µmol/kg.
Neither of the new rigid analogs fully substituted for
LSD. For the dihydrobenzofuran 8, only 50% of the rats
selected the drug lever when administered 57.8 µmol/
kg of test drug. When a similar dosage of the tetrahy-
drobenzodifuran 9 was administered, only 29% of the
rats selected the drug lever.
F igu r e 2. Effects of mescaline 1 (0), 8 (1), 9 (b), and 23 (O)
on phosphoinositide hydrolysis in NIH 3T3 fibroblasts ex-
pressing the rat 5-HT2C receptor. The curves are representa-
tive of experiments carried out on the same day. The mean
Emax values (three determinations) of 1, 23, 8, and 9 were 97,
95, 82, and 70% respectively, of the maximum 5-HT response.
The mean EC50 values ( SEM were 24 ( 1, 2 ( 1, 2 ( 1, and
1 ( 0.2 µM, respectively.
molecular hydrophobicities anticipated. Barfknecht et
al.39 previously correlated hallucinogenic activity with
octanol-water partition coefficients for a series of
substituted phenylalkylamines and found that the
optimum hydrophobicity was centered at log P ) 3.15.
Since mescaline has a log P of only 1.18, it was felt that
increased hydrophobicities of 8 and 9 (however, the
increase resulting from the added methylene groups is
It was initially envisioned that if 8 and 9 retained
the necessary pharmacophoric elements, they might
have somewhat increased in vivo hallucinogen-like
activity compared to 1 due to the slightly greater
Ta ble 2. Results of the Radioligand Competition Studies of 8 and 9 at Cloned Human 5-HT2A, 5-HT2B, and 5-HT2C Receptors (Ki
values ( SEM in nanomolar)
agonist radioligands
125I]DOI
antagonist radioligands
[3H]rauwolscine
5-HT2B
[
[3H]serotonin
5-HT2B
[3H]ketanserin
5-HT2A
[3H]mesulergine
5-HT2C
compound
5-HT2A
5-HT2C
1
8
9
551.0 ( 35.3
368.4 ( 17.9
335.3 ( 35.7
215.6 ( 25.0
302.8 ( 2.9
118.3 ( 16.0
61.5 ( 5.3
795.1 ( 65.5
523.0 ( 53.0
301.5 ( 70.0
555.1 ( 41.0
NT
NT
NT
6405 ( 834
4443 ( 319
NT
311 ( 38
205 ( 7.6
NT
1105 ( 57
654 ( 52
NT
23
177.2 ( 18.0

Dihydrobenzofuran Analogues of Hallucinogens
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 19 3001
somewhat offset by the more effective hydrogen-bonding
of the constrained heterocyclic oxygens) would enhance
their in vivo activity. Nevertheless, in the behavioral
assays, the rank order of potency for this series de-
creased with increasing molecular hydrophobicity. In
view of the somewhat greater brain concentrations of 8
and 9 that should be achieved, relative to 1, the ability
to activate some essential biological target appeared to
have been lost by compounds 8 and 9. These results
are even more surprising when it is recalled that similar
transformations leading to molecules such as 7 afforded
extremely potent compounds.36 Clearly, there seem to
be some fundamental differences between the molecular
mechanisms of action of the 3,4,5- and 2,4,5-substituted
phenethylamine hallucinogens.
At this point, it was decided to assay a related
compound, 3,5-dimethoxy-4-ethoxyphenethylamine (es-
caline; 23), reported to be significantly more potent than
mescaline in humans.10 To our surprise, 23 also failed
to produce generalization in LSD-trained rats, but did
produce LSD-appropriate responding in 50% of the rats
at a dose that was approximately one-half the ED50 of
mescaline.
the three 5-HT2 receptor subtypes with those of 1, 8,
and 9 revealed no obvious basis for the decreased in vivo
potencies of 8 and 9.
This prompted us to examine the relative efficacies
of the compounds in activating phosphoinositide turn-
over in cells expressing the 5-HT2A receptor, anticipating
that perhaps compounds 8 and 9 might be partial
agonists or antagonists. These compounds were com-
pared to mescaline 1 and 23 (Figure 1). All four
phenethylamine derivatives stimulated [3H]-IP3 hy-
drolysis in a transfected cell line expressing the 5-HT2A
receptor. Consistent with the drug discrimination data,
compounds 8 and 9 were partial agonists, eliciting a
maximum response less than that obtained with sero-
tonin. Also in agreement with the drug discrimination
data, compound 9 was a weaker agonist than was 8.
Curiously, however, in these studies, escaline was a full
agonist, in contrast to its low efficacy in the drug
discrimination paradigm, with an Emax virtually identi-
cal to that of mescaline. This latter result contrasts
with the partial substitution for 23 in the drug dis-
crimination assay.
Similar phosphoinositide turnover experiments were
then carried out in cells expressing the rat 5-HT2C
receptor (Figure 2) to determine whether agonist prop-
erties at this receptor might provide more insight into
the behavioral effects of the four drugs. The profile was
very similar to that found for the 5-HT2A receptor.
While mescaline was a full agonist at the 5-HT2C
receptor, compounds 8 and 9 had somewhat reduced
efficacy. Escaline 23, once again, appeared to be equally
efficacious to mescaline 1. At this receptor, however, 1
had a much larger EC50 than did the other three
compounds. We believe this latter result has bearing
on the drug discrimination results, especially for 23, as
discussed later.
In the in vitro assays, none of the test compounds had
significant affinity for 5-HT1A receptors, results that are
consistent with what is known of the general SAR of
hallucinogenic phenylalkylamines. Curiously, however,
the rank order of affinities of the test compounds for
5-HT2A receptors was opposite the potencies obtained
in the behavioral assays. Thus, while mescaline 1 had
the lowest affinity for 5-HT2A sites, the rigid analog 8
had more than twice the potency at these sites. These
results are unusual because 5-HT2A receptor affinity of
similar phenethylamines typically parallels the potency
in behavioral assays for hallucinogenic activity. Indeed,
the 5-HT2 receptor subtype is commonly accepted as the
primary site of action through which the unique behav-
ioral effects of this class of drugs are mediated.26-29
Although mescaline represents one of the least potent
hallucinogens, one would still expect its in vitro and in
vivo pharmacological profiles to be in agreement with
the known phenylalkylamine SAR. Furthermore, the
related 3,5-dimethoxy-4-ethoxyphenethylamine (23) had
a Ki of 2.1 µM at these 5-HT2A sites, virtually identical
to 8, yet is reported to be 6-8 times more potent than
mescaline 1 as an hallucinogen in humans.10
The data from the binding studies in cloned human
receptors are no more revealing (Table 2). At [3H]-
antagonist-labeled 5-HT2A, 5-HT2B, or 5-HT2C receptors,
increased rigidification in going from 8 to 9 led to
increased affinity. Hallucinogens are generally consid-
ered to act as agonists, so it was even more surprising
when data from the competition studies employing the
agonist ligands [125I]DOI and [3H]serotonin were ob-
tained also showing increased affinity as the molecules
become more rigid. When one considers that the
4-ethoxy homologue 23 is considerably more potent than
1 as an hallucinogen,10 a comparison of its affinities for
As one final assay to compare these compounds and
attempt to identify additional features in their in vitro
pharmacology that might serve to explain the behavioral
results, we evaluated the ability of all four compounds
to inhibit the accumulation of [3H]DA or [3H]5-HT in
rat whole brain synaptosomes. The IC50 values of all
compounds were >10 µM, with no obvious trends to
parallel the observed behavioral potencies (unpublished
results).
We believe that the loss of LSD-like activity in the
drug discrimination assay for compounds 8 and 9 may
lie in their reduced 5-HT2A agonist efficacies, compared
to 1. This point is particularly emphasized by the
lowered degree of substitution of 9 relative to 8, and
the parallel reduction of agonist efficacy. These results
support the current hypothesis that hallucinogens are
agonists at 5-HT2A receptors.
Somewhat more problematic, however, is the inter-
pretation of the behavioral data for compound 23. To
our knowledge, examples of false negatives among
hallucinogen analogues in this assay have been un-
known up to the present time. The affinity of the
4-ethoxy homologue 23 for the 5-HT2A receptor, as well
as its ability to stimulate phosphoinositide turnover in
the cloned receptor, are consistent with its activity as a
mescaline-like agent in man. Although the radioligand
competition experiments indicate only about a two-fold
increase of affinity over 1, the enhanced hydrophobicity

3002 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 19
Monte et al.
of 23 could be part of the explanation for its 6-8 fold
higher clinical potency.
robust interoceptive cue than do 5-HT2A agonists, at
least in rats.
In summary, we have extended our previously de-
scribed rigid analog approach, using tetrahydrobenzodi-
furan functionalities as rotationally restricted bioiso-
steres of the aromatic methoxy groups of mescaline.
Although the rigid analogs synthesized here had in-
creased overall hydrophobicity and enhanced affinity for
serotonin 5-HT2 receptors, they had decreased efficacy
and were not as potent as 1 in a behavioral assay for
hallucinogen-like activity. These results may point to
different (i.e. out-of-plane) methoxy group conformations
for 3,4,5-trimethoxyphenylalkylamines related to mes-
caline as compared with the 2,5-dimethoxyphenylalkyl-
amines. The novel rigid analogs 8 and 9 may serve as
useful ligands for probing possible alternate mecha-
nisms of action for mescaline and similar structures, and
will be valuable tools in future pharmacological inves-
tigations of the SAR of hallucinogenic agents.
Escaline 23 also has full efficacy at the 5-HT2C
receptor and furthermore has significantly greater
potency than mescaline at that site, at least in the
cloned rat receptor. That result suggests that 5-HT2C
agonism should be a relatively more prominent feature
of the pharmacology of 23 compared with 1. No studies
have been carried out to compare the relative strengths
of the interoceptive cues produced in rats by 5-HT2A and
5-HT2C receptor agonists. It seems possible, therefore,
that the interoceptive cue induced by the increased
intensity of the 5-HT2C agonist action of 23 may be
masking its 5-HT2A stimulus properties. It might be
noted in this context that the EC50 values for compounds
8 and 9 (Figure 2) are also significantly lower than for
1. Thus, while we believe that lowered 5-HT2A efficacy
of 8 and 9 is sufficient to explain their behavioral effects,
stimulus masking by agonism at the 5-HT2C receptor
could also be invoked as a partial explanation. Of
course, both mechanisms may be important.
Exp er im en ta l Section
Ch em istr y. Melting points were determined using a
Thomas-Hoover apparatus and are uncorrected. 1H-NMR
spectra were recorded using either a 500 MHz Varian VXR-
500S or 300 MHz Bruker ARX-300 NMR spectrometer. Chemi-
cal shifts are reported in δ values (ppm) relative to tetrameth-
ylsilane (TMS) as an internal reference (0.03% v/v). Abbre-
viations used in NMR analyses are as follows: s ) singlet, d
) doublet, t ) triplet, dd ) doublet of doublets, dt ) doublet
of triplets, td ) triplet of doublets, q ) quartet, p ) pentet, m
) multiplet, b ) broad, Ar ) aromatic. Chemical ionization
mass spectra (CIMS) using methane as the carrier gas were
obtained with a Finnigan 4000 spectrometer. IR measure-
ments were taken with a Perkin-Elmer 1600 Series FTIR
spectrophotometer. Elemental analyses were performed by the
Purdue University Microanalysis Laboratory and were within
(0.4% of the calculated values unless otherwise noted. Thin-
layer chromatography (TLC) was typically performed using
Baker-flex silica gel IB2-F, plastic-backed plates with fluores-
cent indicator (2.5 × 7.5 cm, J .T. Baker), eluting with CH2-
Cl2, and visualizing with UV light at 254 nm and/or I2 vapor
unless otherwise noted. Plates used for radial centrifugal
chromatography (“Chromatotron”, Harrison Research, Palo
Alto, CA) were prepared from Silica Gel 60 PF2-54 containing
gypsum. Most reactions were carried out under an inert
atmosphere of dry nitrogen.
6,7-Dim et h oxyb en zofu r a n -3-on e (10). Oxalyl chloride
(7.2 mL, 81.0 mmol) in 27 mL of dry benzene was added
dropwise to a solution of 12.05 g (56.8 mmol) of 2,3,4-
trimethoxybenzoic acid (Aldrich, Milwaukee, WI) in 150 mL
of benzene, with two drops of DMF added as a catalyst. The
mixture was stirred for 3.5 h, and the volatiles were removed
under reduced pressure to give a quantitative yield of crude
benzoyl chloride as an oil that spontaneously crystallized to a
white solid.
Following literature methods,55,56 an ethereal solution of
diazomethane was cautiously prepared from Diazald (N-
methyl-N-nitroso-p-toluenesulfonamide) in a well-ventilated
fume hood. A solution of 17 g of KOH in 23 mL of water, 66
mL of carbitol (2-(2-ethoxyethoxy)ethanol), and 20 mL of ether
was stirred in a Diazald apparatus (Aldrich, Milwaukee, WI)
and warmed to 60 °C over an oil bath. Diazald (48.42 g, 226.1
mmol) in 150 mL of anhydrous ether was added slowly to the
warm solution, and the clear, yellow distillate that formed was
collected and kept cold in an ice-salt bath. The crude benzoyl
chloride prepared above was dissolved in 50 mL of anhydrous
ether and added dropwise to the cold diazomethane solution.
After 10 min, the ice-salt bath was removed, and the mixture
was stirred for an additional 2 h. The solvent and excess
diazomethane were removed by stirring under reduced pres-
sure (water aspirator), to afford the intermediate diazometh-
ylketone as a viscous yellow-orange oil. To the oil was added
It is possible that other pharmacological factors might
be involved in the different behavioral effects of these
compounds. For example, it has been reported previ-
ously that the behavioral effects of 1 are mediated
through both 5-HT and dopamine (DA) receptors.37
Although we have not examined dopamine receptor
affinities for any of these molecules, their ability to
inhibit accumulation of dopamine into rat whole brain
synaptosomes did not differ significantly (unpublished
study). It is still possible that other unidentified
variables that we did not consider may be important.
The efficacy data at the 5-HT2A receptor, however, seem
adequate to explain the decreased LSD-like effects of
compounds 8 and 9.
The decreased efficacy of compounds 8 and 9 at the
5-HT2A receptor relative to 1 or 23 is most interesting.
When one incorporates the 2- and 5-methoxy groups of
2,4,5-substituted hallucinogenic amphetamines into di-
hydrofuran rings, activity is completely retained.36 In
the present instance, however, the same type of in-
creased rigidification leads to decreased agonist efficacy,
while at the same time increasing affinity. If the
previously proposed receptor model is correct,36 involv-
ing hydrogen bonding sites for the methoxy oxygens, the
present results suggest that the 3- and 5-methoxy
groups in mescaline-related compounds may not lie in
conformations that place them in the plane of the
aromatic ring. Implicit in this observation is the
possibility that these methoxy groups may reorient
themselves upon receptor binding, to achieve a comple-
mentary fit and induce an activated conformational
state of the receptor. This would necessarily involve
out-of-plane conformations for the methoxy groups.
Therefore, future molecular modeling studies involving
comparisons of 3,4,5- and 2,4,5-trisubstituted pheneth-
ylamines should take this possibility into account.
Further pharmacological evaluation of 1, 8, 9, and 23
appears warranted to elucidate fully the mechanism of
action of these mescaline analogues, and to understand
how the mechanism(s) of action for mescaline might
differ from the action of hallucinogenic 2,5-dimethoxy-
phenylalkylamines. These results also suggest the
possibility that 5-HT2C agonists may produce a more

Dihydrobenzofuran Analogues of Hallucinogens
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 19 3003
80 mL of glacial acetic acid, giving an immediate, strong
evolution of gas and heat. The mixture was allowed to stir in
a room temperature water bath for 1.5 h. Solvent removal
under reduced pressure gave a transparent red-brown oil that
crystallized spontaneously. Recrystallization from methyl
ethyl ketone gave 8.30 g (77% from 2,3,4-trimethoxybenzoic
acid) of pure product 10 as fluffy, white needles: mp 122-
124 °C; 1H NMR (CDCl3) δ 3.95 (s, 3, CH3O), 4.0 (s, 3, CH3O),
4.7 (s, 2, CH2), 6.7 (d, 1, ArH), 7.4 (d, 1, ArH); CIMS m/z 195
(M + 1). Anal. (C10H10O4) C, H.
6,7-Dim eth oxy-2,3-d ih yd r oben zofu r a n (11). The ben-
zofuranone 10 (7.42 g, 38.2 mmol) was added to a suspension
of 2.5 g of 10% Pd-C in 300 mL of absolute ethanol in a Parr
flask, and the mixture was shaken under 70 psi of H2 for 80
h. The reaction mixture was filtered through Celite, and the
solvent was removed on the rotary evaporator to give a pale
yellow oil. Kugelrohr distillation gave 5.95 g (86%) of es-
sentially pure 11 as a colorless oil. An analytical sample was
further purified by radial centrifugal chromatography (1 mm
silica plate, CH2Cl2): bp 70 °C at 0.02 mm Hg; 1H NMR
(CDCl3) δ 3.2 (t, 2, ArCH2CH2O), 3.8 (s, 3, CH3O), 3.9 (s, 3,
CH3O), 4.6 (t, 2, ArCH2CH2O), 6.4 (d, 1, ArH), 6.8 (d, 1, ArH);
CIMS m/z 181 (M + 1). Anal. (C10H12O3) C, H.
7-Hydr oxy-6-m eth oxy-2,3-dih ydr oben zofu r an (12). The
dimethoxy benzofuran 11 (5.72 g, 30.4 mmol) was dissolved
in 84 mL of methylene chloride and stirred over a solid CO2-
acetone bath. Boron trichloride (33 mmol as a 1 N CH2Cl2
solution) was introduced to the mixture. After 0.5 h of stirring,
the cooling bath was allowed to melt and warm slowly to room
temperature. Stirring was continued for 4.5 h, and the
reaction was cooled on ice and quenched by the dropwise
addition of 138 mL of water. The organic phase separated,
and the aqueous phase was extracted with methylene chloride
(5 × 20 mL). The combined organic extracts were washed with
brine (2 × 50 mL) and concentrated under reduced pressure
to give a yellow oil. Water (400 mL) was poured over the oil,
and 15 mL of 5 N NaOH was added. The basic solution was
stirred under nitrogen for 2 h and was then acidified with
concentrated HCl. The aqueous mixture was extracted with
5 × 50 mL of CH2Cl2, and the extracts were combined. The
organic phases were washed with brine, dried (MgSO4), and
evaporated to yield a red-orange oil. Kugelrohr distillation
gave a colorless oil that spontaneously crystallized to a white
solid upon cooling. Recrystallization of this solid from ethyl
acetate-hexane afforded 2.98 g (59%) of pure 12 as dense,
colorless, needles: mp 71-72 °C; 1H NMR (CDCl3) δ 3.2 (t, 2,
ArCH2CH2O), 3.9 (s, 3, CH3O), 4.6 (t, 2, ArCH2CH2O), 5.3 (s,
1, ArOH), 6.4 (d, 1, ArH), 6.7 (d, 1, ArH); CIMS m/z 167 (M +
1). Anal. (C9H10O3) C, H.
4-(Cyan om eth yl)-7-h ydr oxy-6-m eth oxy-2,3-dih ydr oben -
zofu r a n (14). To a stirred solution of 12 (574 mg, 3.45 mmol)
in 5 mL of methanol was added dimethylamine (0.47 mL of a
40% aqueous solution, 3.7 mmol), followed by formaldehyde
(0.28 mL of a 37% aqueous solution, 3.7 mmol). The mixture
was stirred at room temperature for 30 min and then outfitted
with a condenser and brought to gentle reflux over a steam
bath. After 4.5 h, TLC analysis indicated the presence of
unreacted starting material, so more dimethylamine (0.10 mL
of 40% solution, 0.8 mmol) and formaldehyde (0.06 mL of 37%
solution, 0.8 mmol) were added to drive the reaction to
completion. After three more hours of reflux, the reaction
mixture was cooled, and all volatiles were removed under
reduced pressure, leaving the crude benzylamine as a white
solid. The benzylamine was dried under high vacuum and was
taken up, with warming, in 10 mL of 2-propanol. Methyl
iodide (0.4 mL, 6.4 mmol) was added, and the mixture was
stirred at room temperature. After 2 min, a flocculent white
solid began to form. Stirring was continued for 5 h, and the
precipitate was collected by vacuum filtration and dried under
high vacuum to give 1.11 g (88%) of the quaternary methiodide
13 as a fine white powder that was sufficiently pure to carry
on to the next step. Attempts to purify further the methiodide
salt by recrystallization resulted in rapid decomposition: mp
159-162 °C (begins to decompose at 97 °C and does not melt
cleanly); CIMS m/z 224 [(M + 1) - CH3I], 179 [(M + 1) -
N(CH3)3+I-].
The crude methiodide salt 13 (5.74 g, 15.7 mmol) was
immediately taken up in 63 mL of water containing 7.41 g
(113.8 mmol) of potassium cyanide, and the mixture was
heated to reflux. All solids dissolved on heating, but as the
reaction proceeded a small amount of precipitate became
visible. After 5 h, the reaction was cooled and cautiously
acidified with 6 N HCl (warning! HCN produced), giving a
substantial quantity of solid. The acidified mixture was stirred
for 2 h in the fume hood, and the solid material was carefully
collected by suction filtration. Extensive air drying on the
filter paper afforded 2.54 g (79%) of the title compound 14 as
an off-white solid: mp 100-104 °C; 1H NMR (CDCl3) δ 3.2 (t,
2, ArCH2CH2O) 3.6 (s, 2, ArCH2CN), 3.9 (s, 3, CH3O), 4.7 (t,
2, ArCH2CH2O), 5.3 (s, 1, ArOH), 6.4 (s, 1, ArH); CIMS m/z
206 (M + 1), 1.79. Anal. (C11H11NO3) C, H, N.
4-(Cya n om et h yl)-6,7-d im et h oxy-2,3-d ih yd r ob en zofu -
r a n (15). A mixture of 4-(cyanomethyl)-7-hydroxy-6-methoxy-
2,3-dihydrobenzofuran (14, 400 mg, 1.95 mmol), powdered
anhydrous potassium carbonate (539 mg, 3.90 mmol), and
methyl iodide (0.24 mL, 3.9 mmol) in 4.6 mL of acetone was
heated at reflux and stirred under N2 for 20 h. The reaction
mixture was cooled, filtered through Celite, and evaporated
to give an oily yellow solid. This was taken up in ethyl acetate,
washed with 10% sodium bicarbonate (2 × 25 mL) and brine
(25 mL), and dried (MgSO4). Filtration, followed by solvent
removal under reduced pressure afforded a dark yellow oil that
crystallized spontaneously. The oil was purified on the Chro-
matotron (4 mm silica plate, CH2Cl2) to give 345 mg (81%) of
15 as a yellow solid: mp 87-89 °C; 1H NMR (CDCl3) δ 3.2 (t,
2, ArCH2CH2O), 3.6 (s, 2, ArCH2CN), 3.8 (s, 3, CH3O), 3.9 (s,
3, CH3O), 4.7 (t, 2, ArCH2CH2O), 6.4 (s, 1, ArH); CIMS m/z
220 (M + 1), 193. Anal. (C12H13NO3) C, H, N.
4-(2-Am in oet h yl)-6,7-d im et h oxy-2,3-d ih yd r ob en zofu -
r a n Hyd r och lor id e (8). In a Parr hydrogenation flask, 1.85
g (8.44 mmol) of 6,7-dimethoxy-4-(cyanomethyl)-2,3-dihy-
drobenzofuran 15 and 1.0 g of 10% Pd-C were combined in
125 mL of ethanol containing 2 mL of concentrated HCl. The
mixture was shaken under 60 psi of hydrogen for 24 h, during
which time the theoretical amount of hydrogen was taken up.
Filtration through Celite to remove the catalyst, followed by
evaporation of solvent in vacuo gave the crude product as a
white solid. Recrystallization from ethanol-ethyl acetate gave
1.93 g (88%) of pure 8‚HCl as glistening white crystals: mp
206-208 °C; 1H NMR (HCl salt in DMSO) δ 2.8 (t, 2, NH3-
CH2CH2Ar), 3.0 (t, 2, NH3CH2CH2Ar), 3.1 (t, 2, ArCH2CH2O),
3.69 (s, 3, CH3O), 3.74 (s, 3, CH3O), 4.6 (t, 2, ArCH2CH2O),
6.4 (s, 1, ArH), 8.1 (bs, 3, NH3); CIMS m/z 224 (M + 1). Anal.
(C12H17NO3‚HCl) C, H, N.
1,5-Bis(ch lor om eth yl)-2,3,4-tr im eth oxyben zen e (16).48
A mixture of 49.0 g (0.29 mol) of 1,2,3-trimethoxybenzene
(Aldrich, Milwaukee, WI), 200 mL of 1,4-dioxane, and 130 mL
of concentrated HCl was stirred vigorously, and hydrogen
chloride gas was continuously bubbled through the clear
solution. Formaldehyde (90 mL of a 37% aqueous solution)
was added in three 30 mL portions at 30 min intervals. The
formalin addition caused the reaction to become exothermic
with the formation of a white precipitate. The mixture was
stirred for 3 h, and the HCl gas flow was stopped. After
stirring for another 1 h, the mixture (now returned to room
temperature) was cooled further on an ice bath, and 150 mL
of concentrated HCl was added. The mixture was extracted
with 5 × 100 mL of ether, and the extracts were combined
and washed with 2 × 100 mL of 10% K2CO3, 2 × 100 mL of
water, and brine. The organic phase was dried over MgSO4
and filtered through Celite, and the solvent was removed
under reduced pressure to afford 65 g (85%) of crude product
as an orange oil. The oil was purified by Kugelrohr distillation
(bp 120 °C at 0.4 mmHg) to give 31.87 g (41%) of pure 16 as
a colorless oil that crystallized slowly on standing: mp 43 °C
(lit.48 mp 43 °C); 1H NMR (CDCl3) δ 3.89 (s, 3, ArOCH3), 3.99
(s, 6, ArOCH3), 4.58 (s, 4, ArCH2Cl), 7.12 (s, 1, ArH), CIMS
m/z 265 (M + 1), 229, 195.
1,5-Bis(cyan om eth yl)-2,3,4-tr im eth oxyben zen e (17). Fol-
lowing the method of Friedman and Schecter,49 20 g (0.075
mol) of the bis-chloro compound 16 was added portionwise to
a stirred mixture of 8.5 g (0.17 mol) of sodium cyanide in 65

3004 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 19
Monte et al.
mL of dry DMSO warmed over an 80 °C oil bath. After 20
min, the reaction appeared complete (TLC) and was poured
onto 400 mL of ice. The dark mixture was extracted with 4 ×
100 mL of ether, and the extracts were combined and washed
with 5 × 100 mL of H2O and 2 × 100 mL of brine. After drying
(MgSO4) and filtration through Celite, the solvent was removed
on the rotary evaporator, and the oil was stirred under high
vacuum to give 16.2 g (87%) of 17 as a clear, brown oil that
was sufficiently pure to carry on to the next step. An
analytical sample was prepared by Kugelrohr distillation to
give a colorless oil: bp 150 °C at 0.15 mm Hg; 1H NMR (CDCl3)
δ 3.63 (s, 4, ArCH2CN), 3.85 (s, 3, ArOCH3), 3.95 (s, 6,
ArOCH3), 7.02 (s, 1, ArH); CIMS m/z 247 (M + 1), 220. Anal.
(C13H14N2O3) C, H, N.
afford 2.8 g (57% from diol 19) of pure 20 as a waxy, white
solid: mp 63-64 °C; 1H NMR (CDCl3) δ 3.12 (td, 4, ArOCH2-
CH2, J ) 8.7 Hz, 0.9 Hz), 3.96 (s, 3, ArOCH3), 4.59 (t, 4,
ArOCH2CH2, J ) 8.7 Hz), 6.67 (s, 1, ArH); CIMS m/z 193 (M
+ 1). Anal. (C11H12O3) C, H.
4-F or m yl-8-m eth oxy-2,3,5,6-tetr a h yd r oben zo[1,2-b:5,4-
b′]d ifu r a n (21). A solution of 2.8 g (0.015 mol) of 20 was
dissolved in 50 mL of dry CH2Cl2 and cooled over an ice bath.
Tin(IV) chloride (3.4 mL, 0.029 mol) was added to the solution
via syringe, and the mixture was stirred for 5 min. Two, 1.3
mL portions (0.029 mol) of R,R-dichloromethyl methyl ether
were added at 30 min intervals, the ice bath was removed,
and the mixture was allowed to stir at room temperature for
4 h. The reaction was quenched by the addition of 50 mL of
ice cold 3 N HCl and diluted with 100 mL of ether, and the
layers were separated. The organic phase was washed with
5 × 50 mL of 6 N HCl, 2 × 50 mL of 5% NaHCO3, and brine,
dried (MgSO4), and filtered through Celite and a thin pad of
silica gel. Removal of solvent under reduced pressure gave a
yellow oil that crystallized spontaneously. The solid was
recrystallized from ethyl acetate-hexane to give 1.3 g (40%)
of 21 as pale yellow needles: mp 125-126 °C; 1H NMR (CDCl3)
δ 3.44 (t, 4, ArOCH2CH2, J ) 8.7 Hz), 4.10 (s, 3, ArOCH3),
4.68 (t, 4, ArOCH2CH2, J ) 8.7 Hz), 10.08 (s, 1, ArCHO); CIMS
m/z 221 (M + 1). Anal. (C12H12O4) C, H.
8-Meth oxy-4-(2-n itr oeth en yl)-2,3,5,6-tetr a h yd r oben zo-
[1,2-b:5,4-b′]d ifu r a n (22). The aldehyde 21 (0.75 g, 3.41
mmol) was stirred under nitrogen with 10 mL of nitromethane
and 0.26 g (3.41 mmol) of ammonium acetate over a 100 °C
oil bath. After 5 h, the volatiles were removed on the rotary
evaporator. The residue was taken up in 10% CH2Cl2 in ether,
and the organic phase was washed with 2 × 50 mL of water,
4 × 50 mL of 3 N HCl, and 50 mL of brine. After drying
(MgSO4), filtration through Celite, and removal of solvent
under reduced pressure, the crude yellow-orange product was
purified on the Chromatotron (4 mm silica plate, CH2Cl2). The
pure solid was crystallized from methanol to give 0.49 g (55%)
of 22 as bright yellow needles: mp 157-158 °C; 1H NMR
(CDCl3) δ 3.29 (t, 4, ArOCH2CH2, J ) 8.7 Hz), 4.03 (s, 3,
ArOCH3), 4.70 (t, 4, ArOCH2CH2, J ) 8.7 Hz), 7.35 (d, 1,
ArCHdCH, J ) 13.5 Hz), 7.96 (d, 1, ArCHdCH, J ) 13.9 Hz);
CIMS m/z 264 (M + 1). Anal. (C13H13NO5) C, H, N.
1-(8-Meth oxy-2,3,5,6-tetr a h yd r oben zo[1,2-b:5,4-b′]d ifu -
r a n -4-yl)-2-a m in oeth a n e Hyd r och lor id e (9). A solution of
the nitroolefin 22 (0.53 g, 2.01 mmol) in 50 mL of dry THF
was added dropwise to a stirred suspension of 0.25 g (6.04
mmol) of LiAlH4 in 50 mL of dry THF. The mixture was
heated at reflux over an oil bath for 3 h and quenched by first
cooling in an ice bath and then cautiously adding 5 mL of 5 N
KOH with vigorous stirring. The mixture was filtered through
Celite, washing the filter cake with ether, and the volatiles
were removed on the rotary evaporator. The aqueous residue
was taken up in 75 mL of 3 N HCl and washed with 3 × 50
mL of ether. The aqueous phase was cooled over an ice bath
and made strongly basic with the addition of 5 N KOH. The
free amine was then extracted into 5 × 40 mL of CH2Cl2, and
the extracts were combined and washed with 2 × 100 mL of
brine. After drying (MgSO4), filtration through Celite, and
removal of the solvent in vacuo, 0.35 g (72%) of the free amine
9 was obtained as a yellow oil. The hydrochloride salt was
formed by the addition of 1 N HCl in anhydrous ethanol, and
recrystallization from ethanol-ethyl acetate gave 222 mg
(40%) of 9‚HCl as a white, crystalline solid: mp 282-283 °C;
1H NMR (free base in CDCl3) δ 1.35 (bs, 2, NH2), 2.58 (t, 2,
ArCH2CH2, J ) 7.3 Hz), 2.88 (t, 2, ArCH2CH2, J ) 7.3 Hz),
3.10 (t, 4, ArOCH2CH2, J ) 8.7 Hz), 3.93 (s, 3, ArOCH3), 4.61
(t, 4, ArOCH2CH2, J ) 8.7 Hz); CIMS m/z 236 (M + 1), 219;
Anal. (C13H17NO3‚HCl) C, H, N.
1,5-Bis(ca r boxym eth yl)-2,3,4-Tr im eth oxyben zen e (18).
The bis-nitrile 17 (11 g, 0.045 mol) was hydrolyzed by reflux
for 4 h with 50 mL of concentrated HCl. Some acid precipi-
tated upon cooling to room temperature, and the mixture was
thoroughly cooled in an ice-water bath to effect the precipita-
tion of more solid. The water-soluble precipitate was collected
on a Bu¨chner funnel and rinsed carefully with ice-cold water.
Air-drying the solid overnight, and then under high vacuum,
1
gave 9.38 g (74%) of 18 as a white solid: mp 155-157 °C; H
NMR (DMSO-d6) δ 3.45 (s, 4, ArCH2COOH), 3.76 (s, 6,
ArOCH3), 3.79 (s, 3, ArOCH3), 6.78 (s, 1, ArH), 12.23 (s, 2,
COOH); CIMS m/z 285 (M + 1), 267, 239, 225. Anal.
(C13H16O7) C, H.
1,5-Bis(2-h yd r oxyeth yl)-2,3,4-tr im eth oxyben zen e (19).
The diacid 18 (7.7 g, 0.027 mol) was suspended in 250 mL of
dry THF and stirred vigorously while cooling on an ice bath.
To the mixture was cautiously added 73 mL of BH3/THF (0.073
mol as a 1 N solution) via syringe. The mixture was stirred
for 24 h while the ice bath was allowed to melt and warm to
room temperature. After carefully quenching the reaction with
40 mL of H2O, the THF was removed on the rotary evaporator.
The aqueous residue was saturated with K2CO3 and extracted
with 5 × 50 mL of 10% THF in diethyl ether. The extracts
were combined and washed with brine, dried (MgSO4), and
filtered through Celite. Removal of solvent under reduced
pressure gave 5.7 g (82%) of a thick yellow oil that was purified
by Kugelrohr distillation (115-135 °C at 0.1 mmHg) to give a
waxy solid. The solid was recrystallized from ethyl acetate-
petroleum ether to afford 3.7 g (53%) of pure diol 19 as a white
1
solid: mp 52 °C; H NMR (CDCl3) δ 2.38 (bs, 2, OH), 2.80 (t,
4, ArCH2CH2OH, J ) 6.4 Hz), 3.76 (t, 4, ArCH2CH2OH, J )
6.4 Hz), 3.83 (s, 6, ArOCH3), 3.88 (s, 3, ArOCH3), 3.74 (s, 1,
ArH); CIMS m/z 257 (M + 1), 239, 225. Anal. (C13H20O5) C,
H.
8-Met h oxy-2,3,5,6-t et r a h yd r ob en zo[1,2-b:5,4-b′]d ifu -
r a n (20). A solution of 6.6 g (0.026 mol) of the diol 19 in 150
mL of CH2Cl2 was cooled to -78 °C by stirring over a solid
CO2-acetone bath. Boron tribromide (17 mL, 0.18 mol) was
added to the mixture via syringe, and the solution was stirred
for 17 h while allowing the bath to warm to room temperature
and evaporate. The reaction was quenched by cooling again
to -78 °C and adding 100 mL of methanol. After warming to
room temperature, the volatiles were removed on the rotary
evaporator, and another 100 mL of methanol was added and
evaporated. The intermediate dibromide (20) was further
dried by stirring under high vacuum to give an unstable brown
oil that was taken directly to the next step.
Under an atmosphere of N2, acetone (250 mL) and K2CO3
(35 g, 0.26 mol) were added to the oil, and the mixture was
heated at reflux over an oil bath for 5 h. Iodomethane (1.62
mL, 0.026 mol) was then added to the solution, and reflux was
continued for another 6 h. The reaction was cooled to room
temperature and filtered through Celite, and the acetone was
removed on the rotary evaporator. The dark residue was
partitioned between ether and water, and the layers were
separated. The ether phase was washed with 3 × 50 mL of 5
N NaOH, 3 × 50 mL of 3 N HCl, and brine, dried over MgSO4,
and filtered through Celite. Solvent removal under reduced
pressure gave a yellow oil that spontaneously crystallized after
standing at room temperature. The crude product was purified
on the Chromatotron (two 4 mm silica plates, CH2Cl2) to give
a solid that was recrystallized from ethyl acetate-hexane to
P h a r m a cology Meth od s. Dr u g Discr im in a tion Stu d -
ies. The procedures for the drug discrimination assays were
essentially as described in previous reports.33,36 Twenty male
Sprague-Dawley rats (Harlan Laboratories, Indianapolis, IN)
weighing 200-220 g at the beginning of the drug discrimina-
tion study were used as subjects trained to discriminate LSD
tartrate from saline. None of the rats had previously received
drugs or behavioral training. Water was freely available in

Dihydrobenzofuran Analogues of Hallucinogens
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 19 3005
was used to determine
method of Litchfield and Wilcoxon59
the individual home cages and a rationed amount of supple-
mental feed (Purina Lab Blox) was made available after
experimental sessions to maintain approximately 80% of free-
feeding weight. Lights were on from 0700 to 1900. The
laboratory and animal facility temperature was 22-24 °C, and
the relative humidity was 40-50%. Experiments were per-
formed between 0830 and 1700 each day, Monday-Friday.
Six standard operant chambers (Model E10-10RF, Coul-
bourn Instruments, Lehigh Valley, PA) consisted of modular
test cages enclosed within sound-attenuated cubicles with fans
for ventilation and background white noise. A white house
light was centered near the top of the front panel of the cage,
which was also equipped with two response levers, separated
by a food hopper (combination dipper pellet trough, Model E14-
06, module size 1/2), all positioned 2.5 cm above the floor. Solid
state logic in an adjacent room, interfaced through a Med
Associates interface to a 486-based microcomputer, controlled
reinforcement and data acquisition with a locally-written
program.
the ED50 (log-probit analysis as the dose producing 50% drug-
lever responding) and 95% confidence interval (95% CI). This
method also allowed for tests of parallelism between dose-
response curves of the drug and the training drug. If 50% or
more of the animals tested were disrupted at a dose where
the nondisrupted rats gave 80% SDL, no ED50 was calculated.
P h a r m a cology Meth od s. Ra d ioliga n d Com p etition
Assa ys in Ra t Br a in Hom ogen a te. Male Sprague-Dawley
rats (Harlan Laboratories, Indianapolis, IN) weighing 175-
199 g were used. The animals were kept in groups of five rats
per cage, at the same conditions described above, but with free
access to food and water.
[3H]Ketanserin and [3H]-8-OH-DPAT were purchased from
New England Nuclear (Boston, MA) at specific activities of 61
and 135.5-216 Ci/mmol, respectively. (+)-LSD tartrate was
obtained from the National Institute on Drug Abuse. Cin-
anserin was a gift from the SQUIBB Institute for Medical
Research, and 5-HT was purchased from Sigma (St. Louis,
MO).
A fixed ratio (FR) 50 schedule of food reinforcement (Bioserv
45 mg dustless pellets) in a two-lever paradigm was used. The
drug discrimination procedure details have been described
elsewhere.57,58 After habituation to the experimental condi-
tions (one week after isolation in the individual home-cages
and at the beginning of the food deprivation), the rat’s initial
shaping was started. During the first 2-3 sessions, rats were
trained only to associate a characteristic noise (click) after
lever pressing with a delivered food pellet (without drug
injections). Initially, rats were shaped to lever press on an
FR1 schedule so that one food pellet was dispensed for each
press. Half of the rats were trained on drug-L (left), saline-R
(right) and the other half on drug-R, saline-L to avoid
positional preference. Training sessions lasted 15 min and
were conducted at the same time each day. Levers were
cleaned between animals with 10% ethanol solution to avoid
olfactory cues. Only one appropriate lever was present during
the first 10 sessions of initial learning (after beginning to
administer saline or training drug, i.p. 30 min before sessions).
Afterwards, both levers were present during all following
phases of training, but reinforcements were delivered only
after responses on the appropriate lever. Presses on the
incorrect lever had no programmed consequences. As respond-
ing rates stabilized (during the next 15 sessions), the schedule
of reinforcement was gradually increased to a FR50. Once at
the FR50, training continued until an accuracy of at least 85%
(number of correct presses × 100/number of total presses) was
attained for eight of ten consecutive sessions (approximately
40-60 sessions). Once criterion performance was attained,
test sessions were interspersed between training sessions,
either one or two times per week. At least one drug and one
saline session separated each test session. Rats were required
to maintain the 85% correct responding criterion on training
days in order to be tested. In addition, test data were
discarded when the accuracy criterion of 85% was not achieved
on the two training sessions following a test session. Test
drugs were administered ip 30 min prior to the sessions; test
sessions were run under conditions of extinction, with rats
removed from the operant chamber when fifty presses were
emitted on one lever. If fifty presses on one lever were not
completed within 5 min the session was ended and scored as
a disruption. Treatments were randomized at the beginning
of the study.
The procedure of J ohnson et al.60 was employed. Briefly,
the frontal cortex or hippocampal brain regions from 20-40
rats were pooled and homogenized (Brinkman Polytron, setting
6 for 2 × 20 s) in 4 or 8 volumes of 0.32 M sucrose for frontal
cortex or hippocampus, respectively. The homogenates were
centrifuged at 36000g for 10 min, and the resulting pellets
were resuspended in the same volume of sucrose. Separate
aliquots of tissue suspension were then frozen at -70 °C until
assay.
For each separate experiment, a tissue aliquot was thawed
slowly and diluted 1:25 with 50 mM Tris HCl (pH ) 7.4). The
homogenate was then incubated at 37 °C for 10 min and
centrifuged twice at 36500g for 20 min with an intermittent
wash. The resulting pellet was resuspended in 50 mM Tris
HCl with 0.5 mM Na2EDTA, 0.1% Na ascorbate, and 10 mM
pargyline HCl (pH 7.4). In experiments with [3H]ketanserin,
either 10 mM MgCl2 or 5.7 mM CaCl2 were included, respec-
tively. A second preincubation for 10 min at 37 °C was
conducted, and the tissues were then cooled in an ice bath.
All experiments were performed with triplicate determina-
tions using the appropriate buffer to which 200-400 µg of
protein was added, giving a final volume of 1 mL. The tubes
were allowed to equilibrate for 15 minutes at 37 °C before
filtering through Whatman GF/C filters using a cell harvester
(Brandel, Gaithersburg, MD) followed by two 5 mL washes
using ice-cold Tris buffer. Specific binding was defined as that
displaceable with 10 µM cinanserin in the [3H]ketanserin
binding study and with 10 µM 5-HT in the [3H]8-OH-DPAT
binding study. Filters were air-dried, placed into scintillation
vials with 10 mL of Ecolite scintillation cocktail, allowed to
sit overnight before counting at an efficiency of 37% for tritium,
and directly counted in a γ counter for [125I]-ligand at an
efficiency of 79.4%.
Ra d ioliga n d Com p etition Exp er im en ts Usin g Clon ed
Hu m a n Recep tor s. All chemicals were obtained from the
sources previously described.61 [3H]-5-HT was purchased from
DuPont-NEN (Wilmington, DE) or Amersham Corporation
(Arlington Heights, IL) at 22.8-26.7 or 81-91 Ci/mmol,
respectively. [125I]DOI (2200 Ci/mmole), [3H]rauwolscine (70-
90 Ci/mmol), and [3H]ketanserin (60-78.7 Ci/mmol) were
purchased from Dupont-NEN (Wilmington, DE).
Mem br a n e P r ep a r a tion fr om Tr a n sfor m ed Cell Lin es.
Membranes were prepared essentially as previously de-
scribed61 using AV12 cell lines (Syrian hamster fibroblast,
ATCC no. CRL 9595) stably transformed with the human
5-HT2A, 5-HT2B, or 5-HT2C receptors. In brief, cells expressing
the receptor of interest were grown in suspension and har-
vested by centrifugation. The cell pellets were then resus-
pended in a minimal volume of a hypotonic buffer, 50 mM Tris-
HCl, pH 7.4, and frozen at -70 °C until needed. On the day
of the assay, the membrane suspension was thawed and
diluted to 35 mL per 0.5 × 109 cells with 50 mM Tris-HCl, pH
7.4. The combination of hypotonic buffer and vortexing was
sufficient to lyse the cells for the membrane preparation. After
vortexing, the preparation was centrifuged at 39000g for 10
min at 4 °C, and the resulting membrane pellet was resus-
The training drug was (+)-lysergic acid diethylamide tar-
trate (LSD; 0.08 mg/kg, 186 nmol/kg, NIDA). All drugs were
dissolved in 0.9% saline and were injected intraperitoneally
in a volume of 1 ml/kg, 30 min before the sessions.
Data from the drug discrimination study were scored in a
quantal fashion, with the lever on which the rat first emitted
50 presses in a test session scored as the “selected” lever. The
percentage of rats selecting the drug lever (%SDL) for each
dose of test compound was determined. The degree of sub-
stitution was determined by the maximum %SDL for all doses
of the test drug. “No substitution” is defined as 59% SDL or
less, and “partial” substitution is 60-79% SDL. If the drug
was one which completely substituted for the training drug
(at least one dose resulted in a %SDL ) 80% or higher) the

3006 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 19
Monte et al.
pended and incubated at 37 °C for 10 minutes and then
centrifuged at 39000g for 10 min at 4 °C. This pellet was
resuspended and centrifuged one more time, and the final
membrane pellet was resuspended (using a Tissumizer, setting
65 for 15 s) in Tris-HCl, pH 7.4, for cells expressing the human
5-HT2B receptor, in Tris-HCl, pH 7.4, containing MgCl2 and
EDTA for [125I]DOI binding to 5-HT2A or 5-HT2C receptors, or
in Tris-HCl, pH 7.6, for [3H]ketanserin and [3H]mesulergine
binding to 5-HT2A and 5-HT2C receptors, respectively.
[3H]Mesu ler gin e Bin d in g to th e Hu m a n 5-HT2C Recep -
tor . This assay was adapted from that described by Pazos et
al.65 Membranes were prepared as described above. Final
concentrations for the 0.8 mL-assays were 0.74-0.82 nM [3H]-
mesulergine, varying concentration of competing compound,
and 50 nM Tris-HCl, final pH 7.6. Nonspecific binding was
determined using 3 mM mianserin. Assay tubes were incu-
bated for 30 min at 37 °C, after which the samples were
filtered, washed, and radioactivity determined as for the [3H]-
ketanserin binding assay described above.
5-HT2B [3H]-5-HT Bin d in g Stu d ies. Human 5-HT2B re-
ceptor binding assays using [3H]-5-HT were performed as
previously described.61 The assay was automated using a
Biomek 1000 (Beckman Instruments, Fullerton, CA). [3H]-5-
HT in Tris-HCl containing CaCl2, pargyline, and L-ascorbic
acid, adjusted to pH 7.4, was added to drug dilutions, spanning
six log units, in water. Then 200 µL of membrane resuspen-
sion (approximately 100-150 µg protein) was added with
mixing and incubated for 15 min at 37 °C. The total incubation
volume was 800 µL, and all incubations were performed in
triplicate. The final concentration of CaCl2, pargyline, Tris,
and L-ascorbic acid was 3 mM, 10 µM, 50 mM, and 0.1%,
respectively. The assay was terminated by vacuum filtration
through Whatman GF/B filters which had been presoaked with
0.5% polyethylenimine (w/v) and precooled with 4 mL ice-cold
wash buffer (50 mM Tris-HCl, pH 7.4), using a Brandel cell
harvester (Model MB-48R; Brandel, Gaithersburg, MD). The
filters were then washed rapidly four times with 1 mL of ice-
cold wash buffer. The amount of [3H]-5-HT trapped on the
filters was determined by liquid scintillation spectrometry
(Ready Protein, LS 6000IC, Beckman Instruments, Fullerton,
CA). The final [3H]-5-HT concentration for competition studies
was approximately 2 nM (range ) 1.7-2.5 nM). The actual
free radioligand concentration was determined by sampling
the supernatant of identical tubes where bound ligand was
removed by centrifugation. Nonspecific binding was defined
with 10 µM 5-HT or 10 µM 1-naphthylpiperazine (1-NP). The
amount of protein was determined by the method of Brad-
ford,62 with bovine serum albumin as the standard.
Sta tistica l An a lysis for Ra d ioliga n d Bin d in g. Nonlin-
ear regression analysis for the competition curves was per-
formed as previously described61 to determine IC50 values.
These were converted to Ki values by the method of Cheng
and Prusoff.66
Mea su r em en t of P h osp h oin ositid e Tu r n over in NIH
3T3 Cells. 5-HT2A and 5-HT2C receptor cDNAs were the gift
of David J ulius (UCSF, San Francisco, CA). Clonal cell lines
were generated in NIH 3T3 fibroblasts (American Type
Culture Collection, Rockville, MD) by calcium phosphate
precipitation and screening by radioligand binding. Cell lines
expressing 5-HT2A receptors (∼5000 fmol/mg protein) and
5-HT2C receptors (∼3000 fmol/mg protein) are referred to as
3T3-2A and 3T3-2C, respectively. [3H]Myo-Inositol (20-25 Ci/
mmol) was purchased from Dupont/NEN Corporation (Boston,
MA). Penicillin, streptomycin, and Dulbecco’s modified eagle
medium (DMEM) were obtained from GIBCO/BRL (Grand
Island, NY). Bovine serum was purchased from Hyclone
(Logan, UT) and tissue culture plates, from Falcon (Lincoln
Park, NJ ).
Fibroblasts were grown in DMEM supplemented with
penicillin (5 U/ml), streptomycin (5 mg/ml), and calf serum
(9%). For the PI hydrolysis assay, cells were plated into 24-
well culture plates and allowed to adhere for 30 h in DMEM
supplemented with 9% calf serum. Cells were subsequently
washed once with 0.5 mL serum-free DMEM and then
incubated for 18 h in 0.5 mL serum-free, inositol-free DMEM
supplemented with 5 mCi/mL [3H]myo-inositol. Labeled cells
were subsequently used for determination of [3H]inositol
phosphate (IP) formation as described originally by Berridge
et al.67 and modified by Grotoweil and Sanders-Bush.68 Dose-
response curves were fit using Prism, a commercially available
computer graphics/analysis program (GraphPad, San Diego,
CA).
5-HT2A,2C [
125I]DOI Bin d in g Stu d ies. Human 5-HT2A or
5-HT2C binding studies were performed essentially as described
for [3H]-5-HT binding to the 5-HT2B receptor with the following
exceptions. The assay buffer contained, in final concentration,
10 µM pargyline, 9.75 mM MgCl2, 0.5 mM (ethylenedinitrilo)-
tetraacetic acid, disodium salt (EDTA), 0.1% sodium ascorbate,
and 50 mM Tris-HCl, pH 7.4. Incubations were performed at
37 °C for 30 min with approximately 40 and 30 µg protein for
the 5-HT2A and 5-HT2C receptors, respectively, and then
filtered and washed as described above. The amount of [125I]-
DOI trapped on the filters was determined using a γ counter.
Nonspecific binding was determined with 10 µM mianserin
for 5-HT2C and 1 µM ketanserin for 5-HT2A receptors. The final
concentration of [125I]DOI was approximately 0.07-0.15 nM.
[3H]Keta n ser in Bin d in g to th e Hu m a n 5-HT2A Recep -
tor . Membranes were prepared as described above, and the
assay conditions were essentially as previously described.63
Assays consisted of 0.8 mL total volume containing 50 mM
Tris-HCl, 100 nM prazosin (to block potential binding of [3H]-
ketanserin to R1-adrenergic receptors), 0.4-0.5 nM [3H]-
ketanserin, and varying concentrations of the competing
compound of interest (final pH 7.6). Mianserin, 3 µM, was
used to define the level of nonspecific binding. Tubes were
incubated at 37 °C for 15 min and then rapidly filtered and
washed as described above. The amount of [3H]ketanserin
trapped on the filters was determined by liquid scintillation
spectrometry.
Ack n ow led gm en t. This work was supported by
NIH grants DA02189 (D.E.N.) and DA05181 (E.S.-B.)
from the National Institute on Drug Abuse. A. Monte
was supported by a National Institutes of Health-
National Institute of General Medical Sciences Predoc-
toral Training Grant.
Refer en ces
(1) For parts 1-3 in this series, see refs 32, 33, and 36, respectively.
(2) Abstracted in part from the Ph.D. thesis of A. Monte, Purdue
University, 1995.
(3) Kapadia, G. J .; Fayez, M. B. E. Peyote Constituents: Chemistry,
Biogenesis, and Biological Effects. J . Pharm. Sci. 1970, 59,
1699-1727.
(4) Heffter, A. Ueber Cacteenalkaloide. Chem. Ber. 1896, 29, 216-
227.
(5) Spa¨th, E. Uber Die Anhalonium-Alkaloidie. 1. Anhalin and
Mezcalin. Monatsh. Chem. 1919, 40, 129-154.
(6) J affe, J . H. Drug Addiction and Drug Abuse. In The Pharma-
cological Basis of Therapeutics, 8th ed.; Gilman, A. G., Rall, T.
W., Nies, A. S., Taylor, P., Eds.; Pergamon Press: New York,
1990; pp 553-557.
(7) Braun, U.; Braun, G.; J acob, P.; Nichols, D. E.; Shulgin, A. T.
Mescaline Analogs: Substitutions at the 4-Position. In NIDA
Research Monograph 22, Quantitative Structure-Activity Rela-
tionships of Analgesic, Narcotic Antagonists, and Hallucinogens,
Barnett, G., Trsic, M., Willette, R., Eds.; U.S. Government
Printing Office: Washington, D.C., 1978; pp 27-32.
(8) Nichols, D. E.; Glennon, R. A. Medicinal Chemistry and Structure-
Activity Relationships of Hallucinogens. In Hallucinogens:
Neurochemical, Behavioral, and Clinical Perspectives; J acobs,
B. L., Ed.; Raven Press: New York, 1984; pp 95-142.
(9) Nichols, D. E. Studies of the Relationship Between Molecular
Structure and Hallucinogenic Activity. Pharmacol. Biochem.
Behav. 1986, 24, 335-340.
[3H]Ra u w olscin e Bin d in g to th e Hu m a n 5-HT2B Re-
cep tor . This assay is based on
a previously described
procedure.64 Membrane preparation and the filtration binding
assay were essentially as described above. Conditions specific
to this assay were as follows (all concentrations given as final
concentrations): 2 nM [3H]rauwolscine, 500 nM efaroxan (to
mask rauwolscine binding to R2-adrenergic receptors), and 50
mM Tris-HCl, pH 7.4. Tubes were incubated at 37 °C for 20
min and then rapidly filtered as described above. Nonspecific
binding was defined in the presence of 10 µM 1-naphthylpip-
erazine.

Dihydrobenzofuran Analogues of Hallucinogens
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 19 3007
(10) Shulgin, A. T.; Shulgin, A. Pihkal: A Chemical love Story;
Transform Press: Berkeley, 1991; 978 pp.
(11) Nichols, D. E.; Oberlender, R.; McKenna, D. J . Stereochemical
Aspects of Hallucinogenesis. In Biochemistry and Physiology
of Substance Abuse; Watson, R., Ed.; CRC Press: Boca Raton,
1991; pp 1-39.
(32) Nichols, D. E.; Hoffman, A. J .; Oberlender, R. A.; Riggs, R. M.
Synthesis and Evaluation of 2,3-Dihydrobenzofuran Analogues
of the Hallucinogen 1-(2,5-Dimethoxy-4-methylphenyl)-2-ami-
nopropane: Drug Discrimination Studies in Rats. J . Med.
Chem. 1986, 29, 302-304.
(33) Nichols, D. E.; Snyder, S. E.; Oberlender, R.; J ohnson, M. P.;
Huang, X. 2,3-Dihydrobenzofuran Analogues of Hallucinogenic
Phenethylamines. J . Med. Chem. 1991, 34, 276-281.
(34) Waldman, S. R.; Monte, A. P.; Bracey, A.; Nichols, D. E. One-
Pot Claisen Rearrangement/O-Methylation/Alkene Isomerization
in the Synthesis of Ortho-Methoxylated Phenylisopropylamines.
Tetrahedron Lett. 1996, 37, 7889-7892.
(35) Monte, A. P.; Marona-Lewicka, D.; Cozzi, N. V.; Nelson, D. L.;
Nichols, D. E. Conformationally Restricted Tetrahydro-1-ben-
zoxepin Analogs of Hallucinogenic Phenethylamines. Med.
Chem. Res. 1995, 5, 651-663.
(36) Monte, A. P.; Marona-Lewicka, D.; Wainscott, D. B.; Nelson, D.
L.; Nichols, D. E. Dihydrobenzofuran Analogues of Hallucino-
gens. 3. Models of 4-Substituted 2,5-Dimethoxyphenylalkyl-
amine Derivatives with Rigidified Methoxy Groups. J . Med.
Chem. 1996, 39, 2953-2961.
(37) Trulson, M. E.; Crisp, T.; Henderson, L. J . Mescaline Elicits
Behavioral Effects in Cats by an Action at Both Serotonin and
Dopamine Receptors. Eur. J . Pharmacol. 1983, 96, 151-154,
and references therein.
(38) Peroutka, S. J .; Hamik, A.; Harrington, M. A.; Hoffman, A. J .;
Mathis, C. A.; Pierce, P. A.; Wang, S. S.-H. (R)-(-)-[77Br]4-Bromo-
2,5-dimethoxyamphetamine Labels a Novel 5-Hydroxytryptamine
Binding Site in Brain Membranes. Mol. Pharmacol. 1988, 34,
537-542.
(12) Nichols, D. E. Medicinal Chemistry and Structure-Activity
Relationships. In Amphetamine and Its Analogs; Cho, A. K.,
Segal, D. S., Eds.; Academic Press, Inc.: San Diego, 1994; pp
3-41.
(13) Shulgin, A. T.; Sargent, T.; Naranjo, C. Structure-Activity
Relationships of One-Ring Psychotomimetics. Nature 1969, 221,
537-541.
(14) Shulgin, A. T.; Dyer, D. C. Psychotomimetic Phenylisopropyl-
amines. 5. 4-Alkyl-2,5-dimethoxyphenylisopropylamines. J .
Med. Chem. 1975, 18, 1201-1204.
(15) Glennon, R. A. Hallucinogenic Agents as Discriminative Stimuli:
A Correlation with Serotonin Receptor Affinities. Psychophar-
macology 1980, 68, 155-158.
(16) Glennon, R. A.; Young, R.; Benington, F.; Morin, R. D. Behavioral
and Serotonin Receptor Properties of 4-Substituted Derivatives
of the Hallucinogen 1-(2,5-Dimethoxyphenyl)-2-aminopropane.
J . Med. Chem. 1982, 25, 1163-1168.
(17) Rasmussen, K.; Glennon, R. A.; Aghajanian, G. K. Phenethyl-
amine Hallucinogens in the Locus Coeruleus: Potency of Action
Correlates with Rank Order of 5-HT2 Binding Affinity. Eur. J .
Pharmacol. 1986, 132, 79-82.
(18) Heller, W. A.; Baraban, J . M. Potent Agonist Activity of DOB at
5-HT2 Receptors in Guinea Pig Trachea. Eur. J . Pharmacol.
1987, 138, 115-117.
(19) Lyon, R. A.; Davis, K. H.; Titeler, M. [3H]DOB (4-Bromo-2,5-
Dimethoxyphenylisopropylamine) Labels a Guanyl Nucleotide-
Sensitive State of Cortical 5-HT2 Receptors. Mol. Pharmacol.
1987, 31, 194-199.
(39) Barfknecht, C. F.; Nichols, D. E.; Dunn, I., W. J . Correlation of
Psychotomimetic Activity of Phenethylamines and Amphet-
amines with 1-Octanol-Water Partition Coefficients. J . Med.
Chem. 1975, 18, 208-210.
(20) Lyon, R. A.; Titeler, M.; Seggel, M. R.; Glennon, R. A. In-
dolealkylamine Analogs Share 5-HT2 Binding Characteristics
with Phenylalkylamine Hallucinogens. Eur. J . Pharmacol.
1988, 145, 291-297.
(21) Sanders-Bush, E.; Burris, K. D.; Knoth, K. Lysergic Acid
Diethylamide and 2,5-Dimethoxy-4-methylamphetamine are
Partial Agonists at Serotonin Receptors Linked to Phospho-
inositide Hydrolysis. J . Pharmacol. Exp. Ther. 1988, 246, 924-
928.
(22) Titeler, M.; Lyon, R. A.; Glennon, R. A. Radioligand Binding
Evidence Implicates the Brain 5-HT2 Receptor as a Site of Action
for LSD and Phenylisopropylamine Hallucinogens. Psychophar-
macology 1988, 94, 213-216.
(23) Pierce, P. A.; Peroutka, S. J . Hallucinogenic Drug Interactions
with Neurotransmitter Receptor Binding Sites in Human Cortex.
Psychopharmacology 1989, 97, 118-122.
(24) Seggel, M. R.; Yousif, M. Y.; Lyon, R. A.; Titeler, M.; Roth, B.
L.; Suba, E. A.; Glennon, R. A. A Structure-Affinity Study of
the Binding of 4-Substituted Analogs of 1-(2,5-Dimethoxyphen-
yl)-2-aminopropane at 5-HT2 Serotonin Receptors. J . Med.
Chem. 1990, 33, 1032-1036.
(25) Glennon, R. A.; Raghupathi, R.; Bartyzel, P.; Teitler, M.;
Leonhardt, S. Binding of Phenylalkylamine Derivatives at
5-HT1C and 5-HT2 Serotonin Receptors: Evidence for a Lack of
Selectivity. J . Med. Chem. 1992, 35, 734-740.
(40) Nichols, D. E.; Shulgin, A. T.; Dyer, D. C. Directional Lipophilic
Character in
a Series of Psychotomimetic Phenethylamine
Derivatives. Life Sci. 1977, 21, 569-576.
(41) Gupta, S. P.; Singh, P.; Bindal, M. C. QSAR Studies on
Hallucinogens. Chem. Rev. 1983, 83, 633-649.
(42) J ung, M. E.; Abrecht, S. Improved Synthesis of 3-Substituted
7-Methoxybenzofurans. Useful Intermediates for Preparation
of Morphine Analogues. J . Org. Chem. 1988, 53, 423-425.
(43) Monte, A. P.; Marona-Lewicka, D.; Cozzi, N. V.; Nichols, D. E.
Synthesis and Pharmacological Examination of Benzofuran,
Indan, and Tetralin Analogues of 3,4-(Methylenedioxy)amphet-
amine. J . Med. Chem. 1993, 36, 3700-3706.
(44) Short, J . H.; Dunnigan, D. A.; Ours, C. W. Synthesis of
Phenethylamines from Phenylacetonitriles Obtained by Alky-
lation of Cyanide Ion with Mannich Bases from Phenols and
Other Benzylamines. Tetrahedron 1973, 29, 1931-1939.
(45) Aboul-Enein, M. N.; Eid, A. I. A Novel Route for the Synthesis
of Mescaline. Acta Pharm. Suec. 1979, 16, 267-270.
(46) Carvalho, C. F.; Sargent, M. V. Boron Trichloride as a Selective
Demethylating Agent for Hindered Ethers. J . Chem. Soc.,
Chem. Commun. 1984, 227-229.
(47) Caldwell, W. T.; Thompson, T. R. Nuclear Methylation of
Phenols. A New Synthesis of Intermediates in the Preparation
of Antisterility Factors. J . Am. Chem. Soc. 1939, 61, 765-767.
(48) Dallacker, F.; Meunier, E.; Limpens, J .; Lipp, M. Preparations
of Derivatives of Trimethoxybenzene. Monatsch. 1960, 91,
1077-1088.
(49) Friedman, L.; Schecter, H. Preparation of Nitriles from Halides
and Sodium Cyanide. An Advantageous Nucleophilic Displace-
ment in Dimethyl Sulfoxide. J . Org. Chem. 1960, 26, 877-879.
(50) Yoon, N. M.; Pak, C. S.; Brown, H. C.; Krishnamurthy, S.; Stocky,
T. P. Selective Reductions. XIX. The Rapid Reaction of Car-
boxylic Acids with Borane-Tetrahydrofuran. A Remarkably
Convenient Procedure for the Selective Conversion of Carboxylic
Acids to the Corresponding Alcohols in the Presence of Other
Functional Groups. J . Org. Chem. 1973, 38, 2786-2792.
(51) Nichols, D. E.; Barfknecht, C. F.; Rusterholz, D. B.; Benington,
F.; Morin, R. D. Asymmetric Synthesis of Psychotomimetic
Phenylisopropylamines. J . Med. Chem. 1973, 16, 480-483.
(52) Lewin, A. H.; Parker, S. R.; Fleming, N. B.; Carroll, F. I.
Formylation of Arenes by R,R-Dichloromethyl Methyl Ether. An
Improved Experimental Procedure. Org. Prep. Proc. Int. 1978,
10, 201-204.
(26) Branchek, T.; Adham, N.; Macchi, M.; Kao, H.-T.; Hartig, P. R.
[3H]DOB (4-Bromo-2,5-Dimethoxyphenylisopropylamine) and
[3H]Ketanserin Label Two Affinity States of the Cloned Human
5-Hydroxytryptamine2 Receptor. Mol. Pharmacol. 1990, 38,
604-609.
(27) Titeler, M.; Leonhardt, S.; Weisberg, E. L.; Hoffman, B. J . 4-[125I]-
Iodo-(2,5-dimethoxy)phenylisopropylamine and [3H]Ketanserin
Labelling of 5-Hydroxytryptamine2 (5-HT2) Receptors in Mam-
malian Cells Transfected with a Rat 5-HT2 cDNA: Evidence for
Multiple States and Not Multiple 5-HT2 Receptor Subtypes.
Mol. Pharmacol. 1990, 38, 594-598.
(28) Pierce, P. A.; Kim, J . Y.; Peroutka, S. J . Molecular Structural
Basis of Ligand Selectivity for 5-HT2 Versus 5-HT1C Cortical
Receptors. Naunyn-Schmiedeberg’s Arch. Pharmacol. 1992, 346,
4-11.
(29) Sanders-Bush, E. Neurochemical Evidence That Hallucinogenic
Drugs Are 5-HT1C Receptor Agonists: What Next? In NIDA
Research Monograph 146, Hallucinogens: An Update; Lin, G. C.,
Glennon, R. A., Eds.; NIH Publication No. 94-3872, NIDA:
Rockville, 1994; pp 203-213.
(30) Nichols, D. E.; Barfknecht, C. F.; Long, J . P.; Stanbridge, R. T.;
Howell, H. G.; Partyka, R. A.; Dyer, D. C. Potential Psychoto-
mimetics. 2. Rigid Analogs of 2,5-Dimethoxy-4-methylphenyl-
isopropylamine (DOM, STP). J . Med. Chem. 1974, 17, 161-166.
(31) Nichols, D. E.; Weintraub, H. J . R.; Pfister, W. R.; Yim, G. K.
W. The Use of Rigid Analogues to Probe Hallucinogen Receptors.
In NIDA Research Monograph 22, Quantitative Structure-
Activity Relationships of Analgesics, Narcotic Antagonists, and
Hallucinogens; Barnett, G., Trsic, M., Willette, R., Eds.; U.S.
Government Printing Office: Washington, D.C., 1978; pp 70-
83.
(53) Monte, A. P.; Marona-Lewicka, D.; Kanthasamy, A.; Sanders-
Bush, E.; Nichols, D. E. Stereoselective LSD-like Activity in a
Series of d-Lysergic Acid Amides of (R)- and (S)-2-Aminoalkanes.
J . Med. Chem. 1995, 38, 958-966.
(54) Nichols, D. E.; Hoffman, A. J .; Oberlender, R. A.; J acob, P., III;
Shulgin, A. T. Derivatives of 1-(1,3-Benzodioxol-5-yl)-2-butan-
amine: Representatives of a Novel Therapeutic Class. J . Med.
Chem. 1986, 29, 2009-2015.
(55) De Boer, T. J .; Backer, H. J . Diazomethane. Organic Syntheses;
Wiley: New York, 1963; Collect. Vol. IV, pp 250-253.
(56) Black, T. H. The Preparation and Reactions of Diazomethane.
Aldrichimica Acta 1983, 16, 3-10.

3008 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 19
Monte et al.
(57) Oberlender, R.; Nichols, D. E. Drug Discrimination Studies With
MDMA and Amphetamine. Psychopharmacology 1988, 95, 71-
76.
(58) Oberlender, R.; Nichols, D. E. (+)-N-Methyl-1-(1,3-Benzodioxol-
5-yl)-2-Butanamine as a Discriminative Stimulus in Studies of
3,4-Methylenedioxymethamphetamine-Like Behavioral Activity.
J . Pharmacol. Exp. Ther. 1990, 255, 1098-1106.
(59) Litchfield, J . T.; Wilcoxon, F. A Simplified Method of Evaluating
Dose-Effect Experiments. J . Pharmacol. Exp. Ther. 1949, 96,
99-113.
(64) Nelson, D. L.; Baez, M.; Kursar, J . D.; Sasso, D. A.; Wainscott,
D. B. [3H]Rauwolscine as an Antagonist Radioligand for the
Cloned Human Serotonin2B (5-HT2B) Receptor. Soc. Neurosci.
Abstr. 1995, 21, Part 2: 1124.
(65) Pazos, A.; Hoyer, D.; Palacios, J . M. Mesulergine, a Selective
Serotonin-2 Ligand in the Rat Cortex, Does not Label these
Receptors in Porcine and Human Cortex: Evidence for Species
Differences in Brain Serotonin-2 Receptors. Eur. J . Pharmacol.
1985, 106, 531-538.
(60) J ohnson, M. P.; Mathis, C. A.; Shulgin, A. T.; Hoffman, A. J .;
Nichols, D. E. [125I]-2-(2,5-Dimethoxy-4-Iodophenyl) Aminoeth-
ane ([125I]-2C-I) as a Label for the 5-HT2 Receptor in Rat Frontal
Cortex. Pharmacol. Biochem. Behav. 1990, 35, 211-217.
(61) Wainscott, D. B.; Cohen, M. L.; Schenk, K. W.; Audia, J . E.;
Nissen, J . S.; Baez, M.; Kursar, J . D.; Lucaites, V. L.; Nelson,
D. L. Pharmacological Characteristics of the Newly Cloned Rat
5-HT2F Receptor. Mol. Pharmacol. 1993, 4, 419-426.
(62) Bradford, M. M. A Rapid and Sensitive Method for the Quan-
titation of Microgram Quantities of Protein Utilizing the Prin-
ciple of Protein-Dye Binding. Anal. Biochem. 1976, 72, 248-
254.
(66) Cheng, Y.; Prusoff, W. H. Relationship Between the Inhibition
Constant (Ki) and the Concentration of an Inhibitor that Causes
50% Inhibition (I50) of an Enzymatic Reaction. Biochem. Phar-
macol. 1973, 22, 3099-3108.
(67) Berridge, M. J .; Downes, C. P.; Hanley, M. R. Lithium amplifies
agonist-dependent phosphatidylinositol responses in brain and
salivary glands. Biochem. J . 1982, 206, 587-595.
(68) Grotowiel, M. S.; Sanders-Bush, E. Regulation of serotonin2A
receptors in heterologous expression systems. J . Neurochem.
1994, 63, 1255-1260.
(63) Nelson, D. L.; Lucaites, V. L.; Audia, J . E.; Nissen, J . S.;
Wainscott, D. B. Species Differences in the Pharmacology of the
5-Hydroxytryptamine2 Receptor: Structurally Specific Differ-
entiation by Ergolines and Tryptamines. J . Pharmacol. Exp.
Ther. 1993, 265, 1272-1279.
J M970219X