Design, synthesis, and subtype selectivity of 3,6-disubstituted β-carbolines at Bz/GABA(A)ergic receptors. SAR and studies directed toward agents for treatment of alcohol abuse

Article abstract of DOI:10.1016/j.bmc.2010.08.049

A series of 3,6-disubstituted β-carbolines was synthesized and evaluated for their in vitro affinities at α_xβ ₃γ₂ GABA_A/benzodiazepine receptor subtypes by radioligand binding assays in search of α₁ s

Full text of DOI:10.1016/j.bmc.2010.08.049

Bioorganic & Medicinal Chemistry 18 (2010) 7548–7564
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Design, synthesis, and subtype selectivity of 3,6-disubstituted b-carbolines
at Bz/GABA(A)ergic receptors. SAR and studies directed toward agents
for treatment of alcohol abuse
Wenyuan Yin a, Samarpan Majumder b, Terry Clayton a, Steven Petrou c, Michael L. VanLinn a,
Ojas A. Namjoshi a, Chunrong Ma a, Brett A. Cromer c,d, Bryan L. Roth b, Donna M. Platt e,
James M. Cook a,
⇑
a Department of Chemistry and Biochemistry, University of Wisconsin-Milwaukee, Milwaukee, WI 53201-0413, United States
b Department of Pharmacology and Division of Medicinal Chemistry, University of North Carolina-Chapel Hill Medical School and School of Pharmacy and NIMH Psychoactive
Drug Screening Program, 120 Mason Farm Road, 4072 Genetics Medicine Bldg, Chapel Hill, NC 27599-7365, United States
c Howard Florey Institute, University of Melbourne, Parkville, VIC 3010, Australia
d Department of Pharmacology, University of Melbourne, Parkville, VIC 3010, Australia
e Harvard Medical School, New England Primate Research Center, One Pine Hill Drive, PO Box 9102, Southborough, MA 01772-9102, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of 3,6-disubstituted b-carbolines was synthesized and evaluated for their in vitro affinities at
Received 28 June 2010
Revised 25 August 2010
Accepted 26 August 2010
Available online 29 September 2010
a
xb3c2 GABAA/benzodiazepine receptor subtypes by radioligand binding assays in search of a1 subtype
selective ligands to treat alcohol abuse. Analogues of b-carboline-3-carboxylate-t-butyl ester (bCCt, 1)
were synthesized via a CDI-mediated process and the related 6-substituted b-carboline-3-carboxylates
6 including WYS8 (7) were synthesized via a Sonogashira or Stille coupling processes from 6-iodo-bCCt
(5). The bivalent ligands of bCCt (32 and 33) were also designed and prepared via a palladium-catalyzed
homocoupling process to expand the structure–activity relationships (SAR) to larger ligands. Based on the
pharmacophore/receptor model, a preliminary SAR study on 34 analogues illustrated that large substit-
uents at position-6 of the b-carbolines were well tolerated. As expected, these groups are proposed to
project into the extracellular domain (LDi region) of GABAA/Bz receptors (see 32 and 33). Moreover, sub-
stituents located at position-3 of the b-carboline nucleus exhibited a conserved stereo interaction in lipo-
philic pocket L1, while N(2) presumably underwent a hydrogen bonding interaction with H1. Three novel
Keywords:
Beta-carboline
Benzodiazepine
Alcohol antagonist
GABA receptor
b-carboline ligands (bCCt, 3PBC and WYS8), which preferentially bound to a1 BzR subtypes permitted a
comparison of the pharmacological efficacies with a range of classical BzR antagonists (flumazenil,
ZK93426) from several different structural groups and indicated these b-carbolines were ‘near GABA neu-
tral antagonists’. Based on the SAR, the most potent (in vitro) a1 selective ligand was the 6-substituted
acetylenyl bCCt (WYS8, 7). Earlier both bCCt and 3PBC had been shown to reduce alcohol self-adminis-
tration in alcohol preferring (P) and high alcohol drinking (HAD) rats but had little or no effect on sucrose
self-administration.1–3 Moreover, these two b-carbolines were orally active, and in addition, were anxio-
lytic in P rats but were only weakly anxiolytic in rodents. These data prompted the synthesis of the b-
carbolines presented here.
Published by Elsevier Ltd.
1. Introduction
1.1. Alcoholism
ism,5–8 alcohol-dependent individuals represent a heterogeneous
group,9–11 and it is unlikely that a single pharmacological treat-
ment will be effective for all alcoholics. Hence, a better under-
standing of the neuromechanisms which regulate alcohol seeking
Alcohol addiction and dependence remain significant public
health concerns, impacting physical and mental well-being, family
structure and occupational stability.4 While advances have been
made in the development of novel therapies to treat alcohol-
behaviors and the design of clinically safe and effective drugs that
reduce alcohol addiction and dependence remain a high prior-
ity.7,12 While the precise neuromechanisms regulating alcohol-
seeking behaviors remain unknown, there is now compelling
evidence that the GABAA receptors within the striatopallidal and
extended amygdala system are involved in the ‘acute’ reinforcing
actions of alcohol.13–18 Among the potential GABAA receptor
isoforms within the VP regulating alcohol-seeking behaviors, GABA
⇑
Corresponding author. Tel.: +1 414 229 5856; fax: +1 414 229 5530.
E-mail address: capncook@uwm.edu (J.M. Cook).
0968-0896/$ - see front matter Published by Elsevier Ltd.
doi:10.1016/j.bmc.2010.08.049

W. Yin et al. / Bioorg. Med. Chem. 18 (2010) 7548–7564
7549
receptors containing the
a
1 receptor subtype (GABA
a
1) appear
the mammalian nervous system.54–59 All these polypeptides pos-
sess an approximate molecular mass of ꢀ50 kD and are structur-
ally related.
preeminent. Thus, Criswell et al. observed that acute alcohol
administration selectively enhanced the effects of iontophoreti-
cally applied GABA in the ventral pallidum (VP).19,20 However, no
effects were seen in the septum, ventral tegmental area (VTA),
To evaluate the role of the
a1 receptor in regulating alcohol
reinforcement, 3-propoxy-b-carboline hydrochloride (3-PBC), a
and CA1 hippocampus. These data suggest the
a
1 Bz/Gaba(A)ergic
mixed benzodiazepine (BDZ) agonist–antagonist with binding
receptor plays an important role in alcohol-motivated behaviors.
Research on the neuroanatomical basis of alcohol reward has
shown that the NACC, VTA, VP, central amygdala (CeA), and hippo-
campus are all involved in GABAergic regulation of ethanol (EtOH)
reinforcement.2,21–24 Other investigators have identified a dense
reciprocal projection from the VP to the NACC,25–27 and many of
these have been found to be GABAergic neurons.28–30 The NACC
is now well established as a substrate that regulates the reinforcing
properties of abused drugs.13 Finally, immunohistochemical31,32
and in situ hybridization studies33–35 have demonstrated that the
VP contains one of the highest concentrations of mRNA encoding
selectivity at the
the prototypical BDZ agonist zolpidem, 3-PBC exhibited a slightly
higher binding selectivity for the
1 receptor.60,61 Preliminary
a
1 receptor was developed.2 Compared with
a
behavioral studies in several species (e.g., rats, mice, and primates)
show that 3-PBC is a BDZ antagonist, exhibiting competitive bind-
ing-site interactions with BDZ agonists at low to moderate doses
(2.5–15 mg/kg).60,62,63 At higher doses (15–60 mg/kg), 3-PBC pro-
duces anxiolytic effects in the plus maze that are comparable with
those of chlordiazepoxide in alcohol preferring (P) rats.62–65 Thus
given the proposed subunit composition of the GABA receptors
within the CeA,31,32,34,66 pharmacological compounds capable of
the
a
1 subunit in the CNS. These findings, together with pharma-
exploiting the a1, a2, and a3 subunit-containing GABAA receptors
cological studies suggesting the VP plays a role in reward-medi-
represent optimal tools to evaluate the role of GABAA receptors in
alcohol reinforcement and better understand neurobehavior and
ethanol responding.
ated behaviors of psychostimulants and opiates,28,36–39 suggest a
possible role of the VP-
alcohol. Findings of previous studies concluded that inhibition of
VP- 1 receptors by the 1-preferring antagonist 3-PBC produced
marked reductions on alcohol-maintained responding.9,15,40,41
The 1-mediated suppression at the VP level by 3-PBC showed a
high degree of neuroanatomical specificity. Specifically, the 1-
a1 receptors in the euphoric properties of
a
a
1.2. a1 Subtype selective ligands
a
The b-carboline-3-carboxylate-t-butyl ester (bCCt) is a mixed
a
benzodiazepine agonist–antagonist ligand with binding selectivity
mediated suppression was not observed with the more dorsal
placements in the NACC or caudate putamen. The failure of 3-
PBC to alter alcohol self-administration in the NACC/striatum is
in agreement with previous research which has consistently re-
at
for both
a
1 receptors;2,62,67 bCCt also exhibits some affinity (albeit lower)
a2 and a3 receptors. Behavioral studies in several species
(e.g., rats, mice, primates) show that bCCt is a BDZ antagonist
exhibiting competitive binding-site interactions with BDZ agonists
over a broad range of doses.60,62,68–72 Other studies show that bCCt
produces anxiolytic effects in rodents62 and potentiates the anti-
ported a lack of expression of the
a1 transcript in the NACC and
caudate putamen.31–35
An understanding of the neuromechanisms that regulate alco-
hol drinking is key in the development of drugs to treat alcohol
addiction and dependence in humans.2 In recent years, much evi-
dence has accumulated in favor of the GABA system;22,23,42,43 how-
ever much remains unknown about the role of specific GABAA
receptor subtypes in regulating ethanol reinforcement. This is
due to both a lack of high affinity and selective ligands capable
of discriminating among the GABAA receptor subunits and the het-
erogeneity of various subunits within the known alcohol reward
circuitry.31,34 Of the potential GABAA receptors involved in the
conflict response induced by a
1 subtype agonists in primates.70
Thus, bCCt displays a weak agonist or antagonist profile depending
on the behavioral task, species, and dose employed.
In studies involving the
served to selectively reduce alcohol-motivated behaviors in a vari-
ety of experiments.2,73 However, unlike the
5 selective inverse
a1 subtype, bCCt, and 3-PBC were ob-
a
agonist RY-23, both the b-carboline antagonists bCCt and 3-PBC
displayed mixed weak agonist–antagonist profiles in vivo in alco-
hol P and HAD rats. Therefore, in addition to being able to study
the molecular basis of alcohol reinforcement, a1 Bz b-carboline li-
reinforcing properties of alcohol, evidence suggests the
within the VP may play an important role in regulating alcohol-
seeking behaviors, as mentioned above. The VP contains one of
a
1 subtype
gands which display mixed agonist–antagonist pharmacology in
alcohol P and HAD rats may be capable of reducing alcohol intake
while eliminating or greatly reducing the anxiety associated with
habitual alcohol, abstinence or detoxification. Thus, these types
of ligands may be ideal clinical agents for the treatment of alcohol
dependent individuals.2,73
Consequently, several series of structurally different com-
pounds have been synthesized which possess some a1 subtype
selectivity.67,74–77 The discovery of high affinity, saturable, and ste-
reospecific ligands for the BzR has been coupled with the demon-
stration that b-carbolines exhibited an affinity for the BzR.78–85
Some of these agents act on the BzR to induce effects that are
functionally opposite (inverse agonists/antagonists) to those of
classical BDZs. Consequently, the affinities of a wide variety of
b-carbolines have been reported on synaptosomal membranes
from this laboratory,60,72,78,80,81,86–90 and the laboratories of
others,91–96 and this prompted the study of the binding affinities
of a series of b-carbolines67 at five recombinant GABAA/BzR sub-
the highest distributions of
a1 subunits in the mesolimbic sys-
tem.32–35 Finally, acute ethanol administration has been reported
to selectively enhance the effects of iontophoretically applied
GABA in the VP. These effects correlate highly with [3H] zolpidem
binding (an a
1 subtype selective agonist).19,20
The GABAA receptor is the major inhibitory neurotransmitter
receptor of the central nervous system (CNS) and the site of action
of a variety of pharmacologically and clinically important drugs,
such as benzodiazepines, barbiturates, neuroactive steroids, anes-
thetics, and convulsants.44 It is now clear that these receptors reg-
ulate the excitability of the brain, anxiety, muscle tone, circadian
rhythms, sleep, vigilance, memory, and learning.44 There are sev-
eral disease states thought to be associated with the improper
functioning of this system, including anxiety, epilepsy,45 insom-
nia,46 depression and bipolar disorder,47,48 schizophrenia,49 as well
as mild cognitive impairment and Alzheimer’s disease.50 A role of
GABAA receptors in drug and alcohol abuse has also been re-
ported.51–53 Briefly, GABAA receptors are composed of five subunits
that form a central chloride channel and can belong to different
types (a1b3c2, a2b3c2, a3b3c2, a5b3c2 and a6b3c2) expressed from
recombinant human cell lines.91,97 In general, this series of b-carb-
oline ligands exhibited some selectivity at a1 receptor subtypes
including bCCt (1) and 3-PBC (2).2,3 These two ligands displayed
a 20-fold and 10-fold selectivity, respectively, for the a1 subtype
over the a2 and a3 receptors, as well as over 150-fold selectivity
subunit classes. A total of 19 subunits (6
a, 3b, 3c, 1d, 1e, 1p, 1h,
3q) of the GABAA receptor have been cloned and sequenced from

7550
W. Yin et al. / Bioorg. Med. Chem. 18 (2010) 7548–7564
N
N
O
N
OCH2CH2CH3
O
O
N
N
N
N
H
H
1 βCCt
2 3PBC
3 zolpidem
α1
α2
15
α3
α5
α6
α1
α2
α3
α5
α6
α1
α2
α3
α5
α6
0.72
5.3
26.7
110.8
52.3
156
>10,000 >10,000
18.9
>5,000
68.8
591 >1000
383
TMS
O
O
O
N
N
I
O
N
N
N
N
N
H
N
H
CF3
5 6- iodo-βCCt
6 6-trimethylsilanylethynyl-βCCt
4 CL 218872
α2 α3
α1
α2
α3
α5
α6
α1
α2
α3
α5
α6
α1
α5
α6
57 1964 1161 561 >10,000
14.4 44.9
65.3
123
>4000
6.8
30
36
108 1000
O
O
H
O
O
O
O
N
N
H
HN
NH
7 WYS8 (6-ethynyl-βCCt)
31 βCCt bivalent ligand
α1
α2
α3
α5
α6
α1
30
α2
α3
α5
α6
0.972 111
102
208 1980
124
100
>300 >4000
Figure 1. In vitro binding affinities of a series of a1 selective ligands (Ki in nM).
for the
a
1 site over the
a
5 subtype.2,3 bCCt (1) was more selective at
synthesis of bCCt-related bivalent ligands,103–106 it was found that
a series of 3,6-disubstituted b-carbolines (see Fig. 1), including 6-
iodo-bCCt (5) and 6-trimethylsilanyl-ethynyl-bCCt (6) (Fig. 1) pos-
sessed a1 subtype selectivity.103
the a1 subtype in vitro than the classical a1 selective agonists zol-
pidem (3) and CL 218872 (4) (Fig. 1).60,98,99 A number of in vitro
and in vivo studies employing a1 (e.g., zolpidem, CL 218872,68
bCCt, and 3-PBC3) selective ligands suggest the
a
1-containing
The rigidly linked linear bivalent ligands of bCCt at position ‘6’
GABAA/Bz receptors of the ventral pallidum (VP) play an important
role in regulating alcohol’s neurobehavioral effects; particularly
alcohol’s reinforcing properties as mentioned above.2,3,19,20,35
did bind to BzR receptors with some a1 subtype selectivity and
may provide the desired a1 selectivity through specific occupation
of the LDi region of the pharmacophore/receptor model.103
Although the two 3,6-disubstituted-b-carbolines 5 and 6 are less
potent than bCCt (1), the potent binding affinities observed for 5
and 6 at the a1 subtype has stimulated the synthesis of the bCCt
analogs: 3-substituted-b-carbolines as well as 3,6-disubstituted-
b-carbolines.
2. Structure–activity relationships
A predictive 3-D QSAR pharmacophore/receptor model for in-
verse agonist/antagonist b-carbolines was initially developed via
Comparative Molecular Field Analysis (CoMFA) and later re-
fined.100,101 Affinities of ligands from 15 different structural classes
have been evaluated.61 Based on this CoMFA study of a series of b-
carbolines, Huang et al. reported that b-carbolines bind to all diaz-
epam sensitive (DS) sites of the BzR with some selectivity at the
On the other hand, these studies also indicated that the selectiv-
ity of GABAA/BzR site ligands could be described in relation to
binding and pharmacological efficacy in vitro. This efficacy was
based on the capacity of a ligand to modulate GABAergic func-
tion.54 BzR ligands act to modulate chloride flux over a continuum
ranging from positive to negative modulation, with neutral antag-
onists acting theoretically, at a point on the continuum, with zero
intrinsic efficacy (e.g., they bind to the receptor but exhibit no
activity).74 Consequently, the pharmacological profiles of bCCt
a
1-containing receptor isoform and this was confirmed by the
in vitro binding affinity of these ligands.102 A lipophilic region
(LDi) of the pharmacophore receptor model appears to be larger
in the a1
, a2 and a3-containing receptor isoforms and important
for a1 subtype selectivity.61 More recently, during the design and
and 3-PBC at recombinant a1b3c2, a2b3c2, a3b3c2, a4b3c2, a5b3c2

W. Yin et al. / Bioorg. Med. Chem. 18 (2010) 7548–7564
7551
and
a
6b3c2 receptor subtypes expressed in Xenopus oocytes were
Hydrolysis of ester 8 to provide the acid 9, was then followed by
esterification in t-butanol with CDI to provide BCCt (1). The synthe-
sis of 3PBCꢁHCl (2) was more difficult to scale up due to the com-
plex last step (14?2). It began with bCCE (8) from Scheme 1,
which was heated with hydrazine to furnish hydrazide (13) in
82% yield. The hydrazide (13) was stirred with nitrous acid to pro-
vide an azide, which was unstable, and was converted into 3-ami-
no-b-carboline (14) when stirred with acetic acid (Scheme 1) via a
Curtius rearrangement. The last step, originally developed on a
100 mg scale, has now been scaled up to 4 g levels to furnish
3PBC hydrochloride salt in reasonable yield for studies in primates.
In Scheme 2, the b-carboline alkyl esters 16–20 as well as chiral
bCCt analogs 21–24, and 25, 26 were prepared via the CDI-medi-
ated process described above (see Scheme 1).113 Briefly, when
b-carboline-3-carboxylic acid 9 was treated with 1,1-carbonydiim-
idazole (CDI) in dry DMF, the imidazole derivative 27 which
resulted was subsequently transformed into the desired esters by
treating it with the corresponding alcohols (individually) in the
investigated.2,3,64
The results of this study illustrated that bCCt was a near ‘neu-
tral’ antagonist (i.e., little or no efficacy) at all receptor subtypes.
In fact, the level of intrinsic efficacy of bCCt in oocytes was less
at some receptor subtypes than the classical non-selective antago-
nist flumazenil (Ro 15-1788, for which intrinsic efficacy at all BZ-
sensitive GABAA subtypes was relatively low, but not zero). To
date, no compound has been characterized that exhibits zero effi-
cacy at all BzR subtypes, raising the possibility that a compound
labeled as an ‘antagonist’ may indeed exhibit functional activity gi-
ven the right circumstances. For example, more recently, the effi-
cacies of both bCCt and 3-PBC in the selective reduction of alcohol
responding and production of anxiolytic effects were demon-
strated in P and HAD rats following oral administration.107 When
compared with naltrexone treatment, these reductions in alcohol
responding were more selective and longer in duration.107 In sum-
mary, the antagonist bCCt exhibited either a neutral or low-effi-
cacy agonist response at GABA receptors in oocytes. Although
there has been some debate in the literature at present as to
whether a ligand’s binding or efficacy selectivity was ‘the more
salient factor’ in determining a ligand’s capacity to function as
an alcohol antagonist,3,67,107 the knowledge of the efficacy of an
individual putative anti-alcohol reward ligand across all GABAA
receptors is indeed critical to the knowledge of their mode of ac-
tion in the CNS.
presence of DBU in a one-pot sequence. The key potential a1 chiral
selective analogs CMD-30 R/S isomers (21 and 22) can be synthe-
sized by the CDI method in 90% yield (individually) on 10 g scale.
The required starting chiral alcohols were obtained by asymmetric
reduction of the corresponding trifluoromethyl ketones with
(+)-DIP-chloride,114,115 or the (ꢂ)-DIP-enantiomer. The 6-substi-
tuted-iodo-b-carboline-3-carboxylates 28 were then prepared as
intermediates to generate different functionality at position-6
through a palladium-mediated cross coupling process. For exam-
ple, as illustrated in Scheme 2, bCCt (1) was treated with
I2/CF3COOAg in chloroform to provide 6-iodo-bCCt 5 (see also
28a) in 80% yield and the 6-substituted targets 29–31 were ob-
tained in 65–83% yields via a Stille coupling process employing
commercially available substituted tributyl-stannanes. The substi-
tution by halogen occurred at position-6 as indicated by analysis of
5 (or 28a) by NMR spectroscopy especially with One Dimensional
Nuclear Overhauser Effect (NOE) experiments.
Depicted in Scheme 3 are the synthetic routes for the bCCt-re-
lated bivalent ligands 32 and 33.103 In order to efficiently effect a
palladium-mediated Sonogashira process at position-6 of b-carbo-
lines (a reactive electron-rich indole heterocycle), protection/
deprotection of the indole Na–H group in b-carbolines 34 and 35
was necessary. The Boc protected 6-ethynyl-b-carboline-3-carbox-
ylic acid t-butyl ester (34) was prepared directly from 6-substi-
tuted acetylenyl bCCt (7), which was initially termed WYS8. The
common intermediate, iodo-bCCt 5 (see also 28a) was then con-
verted into the 6-substituted trimethylsilylacetylenyl bCCt (6) via
a Sonogashira coupling process.116,117 At this point, TBAF was em-
ployed to remove the trimethylsilyl group to provide the 6-substi-
tuted acetylenyl bCCt analog WYS8 (7) as well.
Based on the limited availability of data on the series of a1
‘binding’ and ‘efficacy’ selective b-carbolines (bCCt, 3PBC) as anti-
alcohol agents,108 the present study was designed to expand the
SAR and search for better a1 subtype selective ligands. These com-
pounds may be promising modulators of alcohol-related co-mor-
bid behaviors in alcohol dependence via the GABAA/BzR system.
Although recent evidence suggests a salient role for GABAergic
mechanisms in the regulation of excessive alcohol drinking and
the negative affective states associated with abstinence, decreased
GABAergic tone stemming from chronic alcohol use and with-
drawal may serve to generate anxiety.109 Thus, compounds that
enhance GABAergic tone may be effective and safe treatments for
both excessive alcohol drinking and the negative affective states
associated with abstinence and may represent novel pharmaco-
therapies to treat alcoholism.
In this regard, the chemistry and pharmacological evaluation of
a series of structurally modified analogues of bCCt (1) as selective
and potent a1 subtype-preferring ligands are described. The syn-
thesis of the a1 selective compound 7 (WYS8) and the structure–
activity relationships (SAR) of 3,6-disubstituted b-carbolines are
also presented. The established pharmacophore/receptor mod-
el61,110 of BDZ binding sites was employed to design ligands with
respect to the LDi region at position-6, as well as characterize the
The ester 7, was then protected with a Boc group at the N(1)
position to afford 34 under standard conditions. A Sonogashira pro-
cess was then employed to couple 34 with Boc protected iodo-bCCt
(35) to provide the rigid two-carbon linked bivalent ligand 32 of
bCCt. The Boc protecting group was removed thermally by heating
in cumene at high dilution and bivalent ligand 32 was obtained.
The bisacetylenic bridged ligand 33 was synthesized from the
Boc protected 6-ethynyl-bCCt 34 via a homocoupling process,118
followed by the removal of the Boc group under thermal conditions
in cumene at high dilution.
binding pocket L1 at position-3. Protein-ligand docking of the
a1
subtype GABAA receptor protein and WYS8 illustrated the agree-
ment between the pharmacophore/receptor model and BzR site
prediction based on homology modeling.110–113
3. Chemistry
The synthesis of the ligands under study is outlined in Schemes
1–3. The important precursor b-carboline-3-carboxylate ethyl es-
ter (bCCE, 8) and its corresponding acid (9) were the intermediates
required for large-scale synthesis of bCCt (1), as well as an inter-
mediate required for the synthesis of the new b-carbolines. As
4. Results and discussion
outlined in Scheme 1,
D,L
-tryptophan 10 was converted into tetra-
4.1. bCCt bivalent ligands
hydro-b-carboline (11) via a Pictet–Spengler reaction on kilogram
scale. Fischer esterification of 11, followed by oxidation with acti-
vated MnO2 provided the intermediate BCCE (8) on 200 g scale.
The in vitro biological protocols employed in the present study
follow the published procedure119,120 and are detailed in Section 6.

7552
W. Yin et al. / Bioorg. Med. Chem. 18 (2010) 7548–7564
O
CO2
COOH
C2H5OH/2eq H2SO4
reflux, 83%
OC2H5
HCHO / NaOH
37 ο C, 95%
NH
NH
H3N
N
N
H
N
H
H
12, 200g/12L
10, 1kg/22L
11, 500g/12L
MnO2/benzene
reflux, 80%
O
O
O
O
OC2H5
CDI/DMF;
OH
NaOH / H2O
reflux, 90%
N
N
N
DBU, 50eq t-BuOH
85 o C, 75%
N
H
N
N
H
H
8, βCCE 50g/1L
1, βCCt 5g
9, 30g/2L
NH2NH2, EtOH
reflux, 16 hr,
82%
O
CH3CH2CH2OH
isoamylnitrite
KSCN/CuSCN;
O
NaNO2, HCl
o C;
NH2
2HOAc
NHNH2
0
N
N
2HCl
N
HOAc (aq)
76 %
ether, HCl(g)
58%
N
N
H
N
H
H
13
14
2, 3PBC
Scheme 1. Large scale synthesis of bCCt and synthesis of 3PBC.
Although the
a
1b3c2 BzR/GABAergic subtype is very similar in
knowledge these are the first two bivalent ligands in the b-carb-
oline series, which bind to BzR. Further pharmacological evalua-
tion in vivo of the bCCt bivalent ligand with the longer rigid
linker should shed light on the above hypothesis and this would
also provide some tools to determine the size and exact location
of the LDi region.
structure to the a2 and a3 subtypes, there are slight differ-
ences.61,121 One major difference is in region LDi, which appears lar-
ger in the a1 subtype than in either the
a2, or a3 or a5 subtypes.
This is located near position-6 of bCCt (1) (Fig. 3).60,110,122
The western region of the pharmacophore/receptor model tol-
erates positive lipophilic interactions in this area that corresponds
to region LDi (a region in the pharmacophore adjacent to the
extracellular domain of the receptor) of the unified pharmaco-
phore/receptor model. In this region, bulky substituents are toler-
ated and occupation of this area with substituents appears to
enhance affinity at a1 subtypes. This knowledge provided an
opportunity to introduce a linker between two pharmacophoric
b-carboline-3-carboxylate residues in order to design selective
and rigid bivalent ligands. As described in the Introduction, initial
4.2. WYS8
A series of 6-substituted-b-carboline-3-carboxylates have been
synthesized and bound in vitro to the a1b3c2 BzR subtype prefer-
entially as compared to other subtypes (see Tables 1–3).61,102 These
ligands have also been modeled in the GABAA/BzR pharmacophore
model, and the 6-substitutents align well in the LDi region.123 Occu-
pation of this region should lead to enhanced selectivity of a ligand
efforts to find a novel series of
a
1-preferring ligands focused on
at the
carbolines, 6-trimethylsilanylethnyl-bCCt 6 has been recently syn-
thesized and found in vitro to prefer the 1 subtype. However, the
most selective ligand for the 1 subtype was WYS8 (7). This 1 sub-
type selective ligand was 100-fold more selective over the other
subtypes. This was the most a1 subtype selective ligand reported,
to date, to these authors’ knowledge. This 6-substituted acetylenyl
bCCt 7 was 214-fold more selective for a1 isoforms over a5 iso-
forms. Studies of SAR in Table 1 confirmed the occupation of region
a1-containing isoform. Among the new 3,6-disubstituted-b-
design and synthesis of bCCt bivalent ligands. Although the a1
subtype selectivity was not amplified with the particular acetyle-
nyl linked bivalent ligand 32, the ligand does bind preferentially
at a1 subtypes (Table 1). It was proposed the two-carbon linker
was not long enough and that crowding between the second bCCt
unit and the receptor protein decreased the binding affinity at the
a1 subtype, thereby negating some of the potential selectivity.
However, these rigidly linked linear bivalent ligands 32 and 33
fit the GABAA/BzR pharmacophore/receptor model very well
(Fig. 4).110 The unit at C-6, presumably, protrudes into the extra-
cellular domain of the BzR, as previously expected,102,103 and
bound to BzR with some a1 subtype selectivity.102,103 To our
a
a
a
LDi of the receptor pharmacophore model did enhance a1 selectiv-
ity in comparison to the affinity of the non-selective ligand diaze-
pam or the a5 selective ligand, RY080. As illustrated in the two
dimensional Figure 5, full occupation of the LDi lipophilic region

W. Yin et al. / Bioorg. Med. Chem. 18 (2010) 7548–7564
7553
O
O
O
OH
ROH, DBU
OR1
CDI / DMF
rt, 2 h
N
N
N
N
N
N
H
N
H
N
H
9
27
CF3
CF3
24
, R1 =
21, R1 =
15, R1 = -CH2CF3
16
, R1 = -CH(CF3)2
17, R1 = -CH2CCl3
18
F
F
, R1 = -CH(CH3)CCl3
22
, R1 =
25, R1 =
26, R1 =
19, R1 = -CH(CH2CH3)CH3
F
20
, R1 = -CH(CH2CH3)CF3
23, R1 =
O
O
I
R2
OR1
OR1
R2SnBu3
CF3COOAg/I2
CHCl3, reflux, 80%
N
N
N
N
Pd(PPh3)4/DMF
rt, 65% - 83%
H
H
29a
, R2 =
5(28a)
, R1 = -C(CH3)3,
28b, R1 = -CH2CF3,
28c
O
S
, R1 = -CH2CH3,
28d, R1 = -CH(CF3)2
30a, R2 =
31a
, R2 =
Scheme 2. CDI-mediated esterification of 3-substituted b-carbolines followed by the conversion into 3,6-disubstituted b-carbolines.
by b-carbolines may account for the potency/selectivity of this
class of ligands at the a1 subunit. Analysis of the in vitro binding
data for this series of bulky 6-substituted b-carbolines (Table 1)
has shown some selectivity for the a1 receptor subtype. In addi-
tion, it is important to note that binding affinity in this series of li-
gands of greater than 400 nM usually results in zero efficacy at the
subtype at pharmacologically relevant concentrations.
Homology models employed here of the GABAA receptor were
as described previously111,124 except that a number of alternative
models were considered for loop C, which was two residues short-
er than the template and hence built from a loop database. The fi-
nal model was selected based on assessment of model quality125
and consistency with published mutational data,126–130 particu-
larly with the T207 side-chain appropriately positioned facing
the benzodiazepine-binding pocket. Positioning of WYS8 in the
BzR was executed using a genetic algorithm (FlexiDockÒ). Flexible
docking provides a means of docking ligands into protein active
sites.
ent which interacted at L1. The L1 pocket tolerates linear groups up
to four carbons in length. Comparison of the in vitro receptor bind-
ing affinity of the ligands depicted in Table 2b indicated the ethers
36 and 37 bind potently to a1 subtypes while 34 does not; binding
affinity is lost, illustrating that the substituent at the 3-position is
to large to allow the ligand to bind. Likewise, the 3-benzyloxy b-
carboline 35 is also too bulky to fit the L1 pocket despite its lipo-
philic nature. The second factor was the ability of the substituent
at position-3 to release electron density to the pyridine ring. This
enhanced the basicity of the nitrogen atom at N(2) which resulted
in a greater ligand-receptor interaction at H1. Analysis of the bind-
ing affinities of the novel trifluoroalkyl esters of b-carboline-3-car-
boxylic acid further supported this hypothesis (Table 2a). The
trifluoroalkyl esters exhibited reduced binding affinity at all recep-
tor subtypes when compared to their corresponding alkyl esters
(20 vs 19, 15 vs bCCE). Since the trifluoromethyl was a strong elec-
tron-withdrawing group, when compared to the corresponding
alkoxycarbonyl moiety, the 3-trifluoroalkoxycarbonyl substituent
would decrease electron density to the pyridine (N2) ring reducing
the basicity of the nitrogen atom. This would result in a weaker li-
gand-receptor interaction at H1. In addition, the trifluoroalkyl
group was less lipophilic than the corresponding alkyl moiety,
which may result in a weaker interaction at L1. Ramachandran
and Hanzawa have reported that trifluoromethyl groups are nearly
as large as isopropyl or t-butyl functions.114,133 It was possible, the
trifluoromethyl substituted ligands are simply too large to exert
high affinity binding; however, bCCT (1), WY-B-24 (25) and CM-
4.3. 3-Substituted b-carbolines
It was initially believed by Braestrup, Loew, and others that an
ester moiety at position-3 of b-carbolines was required for a ligand
to exhibit high affinity binding at Bz binding sites.84,95,131,132 How-
ever, high affinity binding of b-carbolines including the antagonist
3-propoxy b-carboline (3-PbC, Table 2b) demonstrated this was
not the case.81,88,89 Examination of data from additional stud-
ies78,88,89 have suggested that at least two factors affected high
affinity binding at BzR with respect to 3-alkoxy substituted b-carb-
olines,60,79,102,123 one of which was the lipophilicity of the substitu-
A-77 (26) all bound with good potency to a1 BzR subtypes (see Ta-
bles 1 and 2), and these ester functions occupy a large molecular
volume.

7554
W. Yin et al. / Bioorg. Med. Chem. 18 (2010) 7548–7564
Si
O
H
O
O
I
TMS
H
TBAF/H2O
90%
O
O
O
N
N
N
N
H
PdCl2(PPh3)2/CuI
N
H
N
H
THF/Et3N
92%
6
7 (WYS8)
5(28a)
1.2 eq (Boc)2O
CH2Cl2, DMAP
90%
1.2 eq (Boc)2O
CH2Cl2, DMAP
90%
1.2 eq (Boc)2O
CH2Cl2, DMAP
90%
O
Si
H
O
O
I
TMS
H
TBAF/H2O
90%
O
O
O
N
N
N
N
Boc
N
PdCl2(PPh3)2/CuI
N
Boc
THF/Et3N
94%
Boc
35
36
34
1) PdCl2(PPh3)2, CuI,
BrCH2CO2Et,
THF/(iPr)2NH, 60%
2) cumene, reflux
55%
1) 34, PdCl2(PPh3)2, CuI,
THF/Et3N,
2) cumene, reflux
80%
O
N
O
O
O
O
O
O
O
N
N
N
N
H
N
H
N
N
H
H
32
33
Scheme 3. Preparation of 6-substituted acetylenyl bCCt (7, WYS8) and related bivalent ligands.
remained more potent in vitro at all 5 BzR subtypes. Because the
receptor subtype selectivity remained about the same for the (R)
and (S) isomers, this indicated the stereoenvironment in lipophilic
pocket L1 was highly conserved across the entire series of BzR sub-
types in agreement with earlier work on the binding affinities of
the enantiomers of the framework-constrained 4,5-substituted
pyrroloimidazobenzodiazepines.134 It is possible that lipophilic
pocket L1 is simply a large area in the pharmacophore/receptor
model with only small steric differences between receptor sub-
types. More work will be required to determine if this is the case.
A similar result was observed in the case of (R) and (S) isomers of
CMD-30. The (R)-enantiomer CMD-30 R (22) bound slightly tighter
to the receptor subtype than the (S) isomer (21) with almost 70-
fold more selectivity for the a1 subtype over the a5 isoform. In
addition, previously it was reported that a hydrogen bond between
the N(9) H atom of a b-carboline and the secondary site A2 in the
receptor pharmacophore was required for potent inverse agonist
activity in vivo.78,79 Therefore, a series of ligands with the Boc pro-
tection at position-9 such as 40 and 41 were evaluated and were
not a1 subtype selective ligands. In fact, they did not bind to BzR
at all in agreement with previous work.123
Figure 2. Efficacy of bCCt in modulating GABA at recombinant GABA1–5 receptors2
in Xenopus oocytes: comparison with other BzR antagonists. Modulation of GABAA
a1b3
c2, a2b3c2, a3b3c2, a4b3c2, and a5b3c2 receptor subunit combinations
expressed in Xenopus oocytes by bCCt (open bars), flumazenil (shaded bars), and ZK
93426 (black bars). A saturating concentration (1–10
voltage-clamped oocytes along with an EC50 of GABA.
lM) was coapplied over
4.4. Chiral 3-substituted b-carbolines
4.5. Efficacy of a1-preferring ligands in oocytes at GABAA
receptor channels
Examination of the binding data for the enantiomeric pair of b-
carboline sec-butyl esters 23 and 24 (Table 3) indicated that the
(R)-enantiomer 24 bound tighter to the receptor subtypes than
the (S) isomer 23. Although both enantiomers exhibited approxi-
The physiological efficacy of bCCt, as compared to other Bz
antagonists, was investigated across all diazepam sensitive (DS)
receptor subunits at recombinant a1, a2, a3, and a5 receptor sub-
mately a fourfold selectivity for the a1b3c2 subtype, the (R) isomer
units in the Xenopus oocytes assay and is depicted in Figure 2 by

W. Yin et al. / Bioorg. Med. Chem. 18 (2010) 7548–7564
7555
Figure 3. Orthogonal views of CoMFA contour maps for the affinity of 6-benzyl-substituted b-carbolines at the
a1b3c
2 BzR. Orthogonal view of CoMFA contour maps for the
a1b3c
2 receptor subtype with 6-benzyl-substituted b-carbolines modeled by Huang.60 Green contours represent areas of positive steric interaction at a contribution level of
85%, which would result in reduced binding affinity. Blue contours represent areas of positive charge interaction at a level of 85%, which would increase the affinity of a
ligand.
Table 1
Affinities (Ki = nM) of 3,6-disubstituted b-carbolines at
a
xb3c2 (x = 1–3, 5, 6) receptor subtypes
R6
R3
N
N
H
Ligands
R6
R3
a1
a2
a3
a4
a5
a6
1(bCCt)
8(bCCE)
2(3-PBC)
H
H
H
CO2tBu
CO2Et
OnPr
0.72
1.2
5.3
15
4.9
52.3
18.9
5.7
68.8
1000
ND
1000
111
26.8
591
>5000
2700
>1000
TMS
TMS
TMS
H
6(WYB14)
CO2tBu
CO2CH2CF3
CO2Et
6.8
30
36
2000
200
108
1444
47
1000
>3000
2000
1980
6b(WY-B-25)
6c(WY-B-99-l)
7(WYS8)
17
59
88
4.4
4.5
111
5.58
102
2000
2000
CO2tBu
0.972
208
H
I
I
I
I
7b(WY-B-26-2)
28a(iodo-bCCt)
28b(WY-B-20)
28c(iodo-bCCE)
28d(WY-B-08)
CO2CH2CF3
CO2tBu
CO2CH2CF3
CO2Et
4.5
14.4
12
4.8
78
44.6
44.9
39
31
301
42.7
123
47
34
131
2000
>4000
2000
1000
3000
124
65.3
122
286
681
2000
>4000
3000
1000
3000
CO2CH(CF3)2
29a(WYS13)
29b(WYB27-l)
30a(WYS12)
30b(WYB27-2)
CO2tBu
2.4
26
13
27.5
117
314
72
NA
163
127
2861
449
5000
2000
5000
2000
O
O
S
S
CO2CH2CF3
CO2tBu
143
166
13
3000
NA
37
CO2CH2CF3
9.2
2000
31a(WYS15)
CO2tBu
3.63
25
2.02
137
44.3
125
NA
76.5
299
5000
2000
31b(WYB29-2)
CO2CH2CF3
2000
CMA57
CM-A-82a
CM-A-87
F
COC3H7
CO2tBu
CO2tBu
3.7
2.78
1.62
27
8.93
4.54
40
24.5
14.7
NA
1000
1000
254
7.49
4.61
>2500
1000
1000
C(CH3)3
F
Bcct
Bcct
Bcct
32(WYS2)
33(WYS6)
30
124
100
>300
5000
>300
5000
>4000
5000
Bcct
120
1059
3942
The affinity of compounds at GABAA/BzR recombinant subtypes was measured by competition for [3H]flunitrazepam binding to HEK cell membranes expressing human
119
receptors of composition
a1b3c2
,
a2b3c2
,
a
3b3c2
,
a4b3c2
,
a5b3c
2, and
a6b3c2
.
Data represent the average of at least three determinations with a SEM of 5%. The most
selective ligands are in bold type.
Harvey et al.2,64 In comparison to other BzR antagonists such as flu-
mazenil and ZK 93426, as mentioned, bCCt exhibited either a neu-
tral or low-efficacy agonist response at GABA a1 (96 7%), a2
(99 10%), a3 (108 6%), and a4 (107 5%) receptors. However, a
low-efficacy partial inverse agonist response was observed at the
a5 receptor (88 7% of the GABA response). Flumazenil exhibited
an efficacy profile that was qualitatively similar to bCCt at the a1
(99 5%),
a
3 (118 7%), and
a5 (96 6%) subtypes. At the
a2 recep-
tor, flumazenil produced
a
low-efficacy agonist response
(115 4%), while bCCt was GABA neutral (99 10%). Flumazenil
also produced a qualitatively similar response to bCCt at the a4
receptor, albeit the magnitude of GABA potentiation by flumazenil

7556
W. Yin et al. / Bioorg. Med. Chem. 18 (2010) 7548–7564
Figure 4. Alignment of bivalent ligands 32 and 33 within the included volume of the
a1b3c
2 subtype.
Table 2a
Affinities (Ki = nM) of 3-substituted b-carbolines at
a
xb3c2 (x = 1–3, 5, 6) receptor subtypes
R6
R3
N
N
H
Ligands
R6
R3
CO2Et
CO2CH2CF3
CO2CH(CF3)2
a1
a2
a3
a4
a5
a6
BCCE
15
H
H
H
H
H
H
H
1.2
3.0
3.99
10
2000
18
4.9
24.5
8
33
2000
60
5.7
41.7
32
1000
>500
1000
1000
3000
NA
26.8
125.7
461
189
2000
216
2700
>2000
2000
2000
5000
16(WYB09-1)
17(WYB23-1)
18(WYB17)
19(CMA64)
20(CMA69)
CO2CH2CCl3
43
CO2CH(CH3)CCl3
CO2CH(CH3)C2H5
CO2CH(CF3)C2H5
2000
116
1000
>2000
>2000
1000
1000
NA
1000
F
F
O
O
25(WY-B-24)
26(CM-A-77)
22.0
33.5
177
44.8
3000
1000
422
3000
3000
O
F
N
N
N
H
O
1000
1000
1000
N
H
The affinity of compounds at GABAA/BzR recombinant subtypes was measured by competition for [3H]flunitrazepam binding to HEK cell membranes expressing human
119
receptors of composition
a1b3c2
,
a
2b3c2
,
a
3b3c2
,
a4b3c2
,
a5b3c
2, and
a
6b3c2
.
Data represent the average of at least three determinations with a SEM of 5%. The most
selective ligands are in bold type.
far exceeded that of bCCt (132 6% vs 108 6%, respectively).
However, it is important to note, with regard to a4 6b3c2 sub-
types, the agonist effect was observed at 10 M, far above that re-
quired for agonist efficacy at the DS subtypes. In contrast, ZK 93426
produced a clear agonist profile, potentiating GABAergic activity by
5. Conclusion
/a
l
Ethanol allosterically modulates the GABA receptor complex to
open the chloride channel and hyperpolarize cells. At the pharma-
cological level, the effects of ethanol can be antagonized with
GABA antagonists.109 Unfortunately, the paucity of high affinity
subtype selective ligands capable of discriminating among the var-
ious GABAA receptor subtypes has, thus far, precluded study of the
precise role GABAA subunits play in mediating EtOH-maintained
responding.
137 8ꢂ148 11% across the a1
–a4 subtypes, but was GABA neu-
tral at the 5 receptor (96 6%). These findings suggested that bCCt
a
had no appreciable intrinsic efficacy. The rationale for referring to
this agent as a ‘mixed agonist–antagonist’ was based on the fact
that, despite the ability to potentiate GABA at certain receptor sub-
types, it was ‘GABA neutral’ at select doses. In addition, at select
doses, bCCt and 3-PbC were capable of competitive antagonism
of classical benzodiazepine agonists,60,64,72 therefore, the develop-
ment of subtype-selective antagonists for GABAA receptors, such as
A series of b-carboline ligands described here has exhibited
some selectivity at the a1 receptor subtype which included b-carb-
oline-3-carboxylate-t-butyl ester (bCCt, 1) and 3-propoxy-b-carbo-
line hydrochloride (3-PBC, 2). These ligands displayed a 20-fold
and 10-fold selectivity, respectively, for the a1 subtype over the
a2 and a3 receptors, as well as over 150-fold selectivity for
the a1 site over the a5 subtype.2,3 bCCt (1) was more selective at
the a1 subtype in vitro than the classical a1 selective agonists zol-
pidem (3) and CL 218872 (4).60,98,99 bCCt and 3-PBC are capable of
competitive antagonism of classical benzodiazepine agonists,3,60,72
therefore, the development of subtype-selective antagonists for
bCCt, which targeted the GABAA a1 receptor as a weak agonist–like
antagonist,74 should facilitate efforts to understand the anti-alco-
hol action of b-carbolines in nonhuman primates and humans as
well.
In the NIMH supported PDSP screen neither bCCT, 3PBC, nor
WYS8 exhibited significant interactions at other receptors (see
http://pdsp.med.unc.edu for details).

W. Yin et al. / Bioorg. Med. Chem. 18 (2010) 7548–7564
7557
Table 2b
did enhance a1 selectivity in comparison to the affinity of the
non-selective ligands or the a5 selective ligands. Full occupation
of the LDi lipophilic region by b-carbolines may account for the po-
Affinities (Ki = nM) of ether-substituted b-carbolines at axb3c2 (x = 1–3, 5, 6) receptor
subtypes
Ligands
a1
a2
a3
a5
a6
tency/selectivity of this class of ligands at the a1 subunit. Based on
the SAR, the most potent a1 selective ligand was 6-substituted
acetylenyl bCCt (WYS8, 7). It was suggested the attenuation of
EtOH-motivated responding effected by WYS8 (7) was mediated
via the a1 selective antagonism of the GABAA/BzR receptor.2 In re-
gard to ester functions at C(3), although both (R) and (S) enantio-
O
N
350.2 3000
3000
3000 10,000
N
H
34
mers exhibited approximately
a fourfold selectivity for the
a
1b3c2 subtype, the (R) isomer remained more potent in vitro at
O
O
all five BzR sites. Two factors affected high affinity binding at BzR
with respect to b-carbolines, one of which was the lipophilicity
of the substituent which interacted at L1. The second factor was
the ability of the substituent at position-3 to release electron den-
sity to the pyridine ring.
830
3000
3000
10,000 10,000
N
N
N
H
35
The most selective ligand for a1 subtypes, to date, to these
authors’ knowledge was WYS8 (7). This 6-substituted acetylenyl
N
H
36.9
24.9
245
194
245
1000
1000
bCCt (7) was 214-fold more selective for a1 isoforms over a5 iso-
forms. WYS8 can clearly be differentiated from non-selective BDZs
by its selective binding affinity at the a1 receptor subunit and re-
duced capacity to potentiate GABA in Xenopus oocytes.
36
O
O
Innate elevations of the a1 and a2 subunits of the HAD rat may
contribute to the capacity of novel b-carboline ligands to function
as both anxiolytic agents and alcohol antagonists in this genetic rat
line.107 These differences may explain the capacity of these novel
b-carboline ligands to block alcohol drinking and still exhibit anx-
iolytic actions in the P and HAD alcoholic rats. WYS8 may be a suit-
able ligand to evaluate as a preclinical agent to reduce alcohol
dependence. Its reduced efficacy at the a1–2 subunits in potentiat-
ing GABA may render it a safe BDZ receptor ligand devoid of syn-
ergistic interactions with alcohol.
N
N
123.6 139.2
1000 10,000
N
H
37
N
H
818
859
10,000 10,000
38
O
O
6. Experimental section
N
N
N
H
5.3
52.3
25.1
68.8
28.2
591
826
1000
1000
6.1. Biological and pharmacological testing
2
Methods for in vitro receptor binding and efficacy in oocytes
follow previous work.129,130 Competition binding assays were per-
formed in a total volume of 0.5 mL at 4 °C for 1 h using [3H] fluni-
N
H
6.43
trazepam as the radiolabel. For these binding assays, 20–50 lg of
membrane protein harvested with hypotonic buffer (50 mM
Tris–acetate pH 7.4 at 4 °C) was incubated with the radiolabel as
previously described.119 Nonspecific binding was defined as radio-
39
The affinity of compounds at GABAA/BzR recombinant subtypes was measured by
competition for [3H]flunitrazepam binding to HEK cell membranes expressing
human receptors of composition
6b3c2
119 Data represent the average of at least three determinations with a SEM of
5%.
activity bound in the presence of 100 lM diazepam and repre-
a1b3c2, a2b3c2, a3b3c2, a4b3c2, a5b3c2, and
sented less than 20% of total binding. Membranes were harvested
with a Brandel cell harvester followed by three ice-cold washes
onto polyethyleneimine-pretreated (0.3%) Whatman GF/C filters.
Filters were dried overnight and then soaked in Ecoscint A liquid
scintillation cocktail (National Diagnostics; Atlanta, GA). Bound
radioactivity was quantified by liquid scintillation counting. Mem-
brane protein concentrations were determined using an assay kit
from Bio-Rad (Hercules, CA) with bovine serum albumin as the
standard.
a
.
GABAA receptors which targets the GABAA a1 receptor as a weak
agonist–like antagonist,74 should facilitate efforts to understand
the anti-alcohol actions of b-carbolines in nonhuman and human
primates alike. Compared with Naltrexone, the reductions in alco-
hol responding were more selective and longer in duration.107 The
The results are summarized in Tables 1–3.
goal of the present study was to identify novel a1 GABAA subtype-
The electrophysiological analyses of all selective compounds
were performed with whole cell variation of the patch-clamp-tech-
nique, in HEK cells employing GABA concentrations around the
subtype-specific EC203,120 to depict the quantitative efficacy differ-
ence (i.e., GABA modulation) and qualitative subunit modulation
(i.e., subunit type) of these ligands relative to diazepam.
Melting points were taken on a Thomas–Hoover melting point
apparatus or an Electrothermal Model IA8100 digital melting
point apparatus and are reported uncorrected. Proton NMR spec-
tra were recorded on a Bruker 250- or 300-MHz multiple-probe
preferring ligands that may serve as prototypes for further evalua-
tion of clinical efficacy. These types of compounds may provide
treatments for excessive alcohol drinking and the negative affec-
tive states associated with abstinence. Ligands that enhance GAB-
Aergic tone may be effective and safe treatments for both excessive
alcohol drinking and the negative affective states associated with
abstinence. This may represent novel, new pharmacotherapies to
treat alcoholism.
Studies of the structure–activity relationships confirmed that
occupation of region LDi of the receptor pharmacophore model

7558
W. Yin et al. / Bioorg. Med. Chem. 18 (2010) 7548–7564
Table 3a
Affinities (Ki = nM) of chiral 3-substituted b-carbolines at
axb3c2 (x = 1–3, 5, 6) receptor subtypes
Ligands
a1
a2
a3
a4
a5
a6
O
H
O
CF3
21
22
23
24
90
931
172
>3000
1847
>2000
N
N
N
N
N
H
CMD-30(S)
O
O
O
H
CF3
O
27.0
17
343.3
59
453
88
>3000
NA
1847
1444
69
>2000
>3000
>2000
N
H
CMD-30(R)
O
H
N
H
O
H
7.7
32.5
43
NA
N
H
The affinity of compounds at GABAA/BzR recombinant subtypes was measured by competition for [3H]flunitrazepam binding to HEK cell membranes expressing human
119
receptors of composition
a1b3c2
,
a2b3c2
,
a
3b3c2
,
a4b3c2
,
a5b3c
2, and
a6b3c2
.
Data represent the average of at least three determinations with a SEM of 5%.
Table 3b
Affinities (Ki = nM) of Boc protected 3-substituted b-carbolines at axb3c2 (x = 1–3, 5, 6) receptor subtypes
Ligands
a1
a2
a3
a4
a5
a6
O
O
40
450
5000
ND
ND
5000
5000
N
N
Boc
WYS20
O
I
O
41
ND
ND
ND
ND
1847
ND
N
N
Boc
WYS21
ND = not determined yet (see previous tables for details of receptor binding). Data represent the average of at least three determinations with a SEM of 5%.
instrument. Infrared spectra were recorded on a Nicolet DX FTIR
BX V5.07 spectrometer or a Mattson Polaris IR-10400 instrument.
Low-resolution mass spectral data (EI/CI) were obtained on a
Hewlett-Packard 5985B GC-mass spectrometer, while high reso-
lution mass spectral data were taken on a VG autospectrometer
(Double Focusing High Resolution GC/Mass Spectrometer, UK).
Microanalyses were performed on a CE Elantech EA1110 elemen-
tal analyzer. Analytical TLC plates employed were E. Merck Brink-
man UV active silica gel (Kieselgel 60 F254) on plastic, and silica
gel 60b for flash chromatography was purchased from E. M. Lab-
oratories. All chemicals were purchased from Aldrich Chemical
Co. unless otherwise stated. All solvents were dried according to
the published procedures.
pound 11: mp 295 °C (lit. mp 293 °C)135 (lit. mp 286 °C);136 (IR
(KBr) 3600–2300, 1630 cmꢂ1; MS (CI, CH4), m/z (relative intensity)
217 (M++1, 50), 216 (62), 169 (59), 144 (100). This material was
employed directly in the next step.
6.1.2. Ethyl 1,2,3,4-tetrahydro-9H-pyrido[3,4-b]indole-3-
carboxylate (12)
The 1,2,3,4-tetra-hydro-9H-pyrido[3,4-b]indole-3-carboxylic acid
11 (500 g, 2.3 mol) was dissolved in anhydrous ethanol (9 L) in a
12 L (three neck) flask, and concd sulfuric acid (98%, 245 mL,
4.6 mol) was carefully added to the solution until most of the solid
dissolved. The reaction mixture was heated to reflux under nitro-
gen until the starting material was no longer detected by TLC on
silica gel (48 h), and the solution became homogeneous. The reac-
tion solution was cooled and the solvent removed under reduced
pressure. The residue was dissolved in H2O (6.4 L) and the pH of
the solution adjusted to 8 with cold aq NH4OH (concd) after which
a precipitate formed. This mixture was then extracted with CHCl3
(6 ꢃ 2.5 L). The combined organic layers were dried (Na2SO4) and
the solvent was removed under reduced pressure to yield a light
tan solid which was dried in a vacuum oven at 100 °C to provide
12 (465 g, 83%). Compound 11: mp 150 °C (lit. mp 149–
150 °C);90,137 1H (300 MHz, CDCl3) d 1.33 (t, J = 7.32 Hz, 3H), 2.45
(br s, 1H), 2.88 (dd, J = 9.70 Hz, J = 15.37 Hz, 1H), 3.13 (dd,
J = 4.76 Hz, J = 15.5 Hz, 1H), 3.77 (dd, J = 4.76 Hz, J = 9.70 Hz, 1H),
6.1.1. 1,2,3,4-Tetrahydro-9H-pyrido[3,4-b]indole-3-carboxylic
acid (11)
D,L-Tryptophan (1000 g, 4.9 mol) was added to a solution of aq
sodium hydroxide (12 L, 0.4 N) after which the mixture was stirred
until it dissolved. Formaldehyde (560 mL of a 37% aq solution,
6.9 mol) was added and the solution was allowed to stir for three
days at 37 °C. Glacial acetic acid (400 mL) was added which re-
sulted in the precipitation of a solid as a fine suspension. The mix-
ture was allowed to stir for two days, after which additional solid
formed. The solid was filtered from the medium, washed with
water (4 ꢃ 1000 mL), and dried to give 11 (953 g, 90.0%). Com-

W. Yin et al. / Bioorg. Med. Chem. 18 (2010) 7548–7564
7559
ica gel/ethyl acetate) indicated the absence of starting material.
The hot solution was filtered through a bed of Celite to remove
the MnO2 and the filter cake was washed with hot benzene. The
benzene layers were allowed to cool. A precipitate formed and
was collected by vacuum filtration, which provided (100–120 g,
50–60%) of pure b-carboline-3-carboxylic acid ethyl ester 8 (bCCE).
The benzene which remained in the filtrate was removed under re-
duced pressure to provide 25–35 g of additional bCCE, but as crude
material. The crude material could be purified by recrystalization
from ethanol. Compound 8: mp 225–227 °C (lit. 224–
229 °C);67,137,138 1H (300 MHz, DMSO-d6) d 1.36 (t, J = 6.95 Hz,
3H), 4.37 (q, J = 6.95 Hz, 2H), 7.37–7.24 (m, 1H), 7.68–7.57 (m,
2H), 8.38 (d, J = 7.87 Hz, 1H), 8.90 (s, 1H), 8.97 (s, 1H), 10.7 (br,
1H); MS (CI, CH4) m/e 241 (M++1, 47), 195 (22), 168 (100), 140
(9). This material was employed directly in the next step.
S2
S3
Cl
L2
L3
L1
H
S1
H2
N
O
N
R
O
O
LDi
N
N
H
H1
6.1.4. b-Carboline-3-carboxylic acid (9)
b-Carboline-3-carboxylic acid ethyl ester 8 (30.0 g, 0.126 mol)
was suspended in aq NaOH (10%, 1.5 L) and heated to reflux until
all the material had gone into solution (1 h). The heating was con-
tinued for an additional 3 h. The reaction mixture was cooled to rt
and acidified by addition of ice-cold aq concd HCl to pH 4. The pre-
cipitate which resulted was stirred overnight. The solid was col-
lected by vacuum filtration and washed with H2O (2 ꢃ 150 mL).
The product was dried at 80 °C under vacuum for 24 h to provide
9 (26.1 g, 99%). Compound 9: mp 220–221 °C (lit. mp 220 °C);67
A2
Figure 5a. Overlap of diazepam and bCCt in the pharmacophore/receptor model.
The structure of WYS8 and diazepam in a simple representation of the pharma-
cophore model. WYS8 (7) (blue line) and diazepam (black line) fitted to the
inclusive pharmacophore model for the BzR. Sites H1 and H2 represent hydrogen
bond donor sites on the receptor protein complex, while A2 represents a hydrogen
bond acceptor site necessary for potent inverse activity in vivo. L1, L2, L3, and LDi are
four lipophilic regions in the binding pharmacophore. Descriptors S1, S2, and S3 are
regions of negative steric repulsion.
IR (KBr) 3260, 2970, 1710 cmꢂ1
;
1H NMR (CDCl3) d 7.31 (t,
J = 7.32 Hz, 1H), 7.69–7.57 (m, 2H), 8.38 (d, J = 7.87 Hz, 1H), 8.90
(s, 1H), 8.96 (s, 1H), 12.10 (s, 1H); MS (CI, CH4), m/e (M++1, 269).
Anal. Calcd for C16H16N2O2 (0.55 H2O): C, 69.07; H, 6.19; N,
10.07. Found: C, 68.81; H, 5.77; N, 10.00.
6.1.5. b-Carboline-3-carboxylic acid t-butyl ester (1)
4.07 (br s, 2H), 4.26 (q, J = 7.14 Hz, 2H), 7.18–7.08 (m, 2H), 7.29 (d,
J = 7.87 Hz, 1H), 7.48 (d, J = 7.68 Hz, 1H), 8.17 (br s, 1H); MS (CI
CH4) m/e 144 (97.3), 245 (M+1, 87), 244 (100), 183 (6), 171 (33),
144 (83). This material was employed directly in the next step.
To a solution of carbonyl diimidazole (28.2 g, 0.177 mol) in
anhydrous DMF (1.2 L) was added dry b-carboline-3-carboxylic
acid 9 (25 g, 0.118 mol) under argon. The reaction mixture was ini-
tially a pale yellow-colored suspension, but after stirring for
30 min, a purple or red-colored solution resulted. The reaction
mixture was stirred for an additional 2 h at rt and carbon dioxide
was released during the reaction. Analysis by TLC (silica gel) indi-
cated the absence of starting material on the baseline. To this reac-
tion mixture was added dry DBU (18 g, 0.118 mol) and dry freshly
distilled t-butyl alcohol (437 g/560 mL, 50 equiv). The mixture
was heated at 85 °C for 18 h until analysis by TLC indicated the
disappearance of the imidazole intermediate. The solvent was then
6.1.3. Ethyl 9H-pyrido[3,4-b]indole-3-carboxylate (8)
Into a round-bottom flask (12 L) equipped with a reflux con-
denser and an overhead stir was added 1,2,3,4-tetrahydro-b-carb-
oline-3-carboxylic acid ethyl ester 12 (200 g, 0.86 mol) and dry
benzene (8 L). The solution was allowed to heat to reflux at which
time activated MnO2 (200 g) was added to the flask. Additional
quantities of activated MnO2 were added until analysis by TLC (sil-
Figure 5b. WYS8 docked in the BzR site of the a1 subtype GABAA receptor. The a1 and c2 subunits are rendered in yellow and blue, respectively.

7560
W. Yin et al. / Bioorg. Med. Chem. 18 (2010) 7548–7564
removed under reduced pressure. The residue was partitioned be-
tween CH2Cl2 (1.2 L) and H2O (800 mL). The organic layer was sep-
arated and the H2O layer was extracted with CH2Cl2 (2 ꢃ 500 mL).
The combined organic layer was washed with an aq solution of 10%
K2CO3, water, brine and dried (Na2SO4). The solvent was removed
under reduced pressure and the residue was purified by flash chro-
matography (silica gel, EtOAc/hexane = 1:1) to provide bCCt (20 g,
65%) as a white solid. bCCt can be recrystallized from EtOAc to pro-
vide white crystals 1: mp 301–303 °C (lit. mp 298–300);80 IR (KBr)
DMF (0.5 mL) was slowly syringed into the reaction mixture at
ꢂ6 °C. The mixture was stirred at 0 °C for 8 h until analysis by
TLC (silica gel) indicated the disappearance of the imidazole inter-
mediate. The reaction mixture was then poured into ice water
(30 mL) and extracted with CH2Cl2 (3 ꢃ 40 mL). The combined or-
ganic layers were washed with H2O (5 ꢃ 40 mL), brine and dried
(Na2SO4). The solvent was removed under reduced pressure and
the residue was purified by flash chromatography (silica gel,
EtOAc/hexanes = 2:1) to provide 21 (0.113 g, 78%) as a white solid.
3500–3400, 3200–3000, 1610, 1560, 1370, 1340 cmꢂ1
;
1H
Compound 21: mp 239–241 °C; ½a D27
= ꢂ9.62 (c 0.81, in CHCl3); IR
ꢄ
(300 MHz, DMSO-d6) d 1.74 (s, 9H), 7.37 (t, J = 7.48 Hz, 1H), 7.63
(t, J = 7.66 Hz, 1H), 7.73 (d, J = 8.52 Hz, 1H), 8.24 (d, J = 7.89 Hz,
1H), 8.85 (s, 1H), 9.12 (s, 1H), 8.97 (s, 1H), 10.19 (br, 1H); The spec-
tral data for 1 were identical to those reported in the literature.88
(NaCl) 3266, 1725, 1502 cmꢂ1 1H NMR (300 MHz, CDCl3) d 1.63
;
(d, J = 6.6 Hz, 3H), 5.75 (m, 1H), 7.40 (t, J = 7.5 Hz, 1H), 7.64 (t,
J = 7.6 Hz, 1H), 7.72 (d, J = 8.1 Hz, 1H), 8.25 (d, J = 7.9 Hz, 1H),
8.91 (s, 1H), 9.18 (s, 1H), 10.04 (br s, 1H); MS (EI) m/e (relative
intensity) 308 (M+, 17), 168 (100), 140 (21). Exact mass Calcd for
6.1.6. 2,2,2-Trifluoroethyl b-carboline-3-carboxylate 15
C15H11F3N2O2: 308.0773. Found: 308.0773. Anal. Calcd for C15
Compound 15 was prepared from b-carboline-3-carboxylic acid
9 and 2,2,2-trifluoroethyl alcohol following the procedure em-
ployed for the preparation of 1. Compound 15: mp 264–266 °C;
H11F3N2O2: C, 58.45; H, 3.60; N, 9.09. Found: C, 58.15; H, 3.63;
N, 8.88.
IR (NaCl) 3275, 1735 cmꢂ1
;
1H NMR (300 MHz, CDCl3) 4.87 (m,
6.1.11. (S)-1,1,1-Trifluorobutan-2-ol
2H), 7.42 (m, 1H), 7.65 (m, 2H), 8.24 (d, J = 7.9 Hz, 1H), 8.93 (s,
1H), 9.09 (s, 1H), 9.10 (br s, 1H); 13C NMR (75.5 MHz, CDCl3) d
60.4, 60.8, 112.8, 118.9, 120.7, 121.2, 122.7, 127.8, 129.2, 134.5,
135.2, 138.1, 141.3, 164.5; MS (EI) m/e (relative intensity) 294
(M+, 30), 195 (7), 168 (100), 167 (42). Anal. Calcd for C14H9F3N2O2:
C, 57.15; H, 3.08; N, 9.52. Found: C, 57.22; H, 3.14; N, 9.23.
(S)-1,1,1-Trifluorobutan-2-ol was prepared following the liter-
ature procedure.114 To an oven-dried, 25 mL round-bottom flask
was transferred (ꢂ)-DIP-Chloride (10.68 g, 33 mmol) in a glove
box. Then 1,1,1-trifluorobutan-2-one (4.03 g, 32 mmol) was added
at rt under argon. The reaction mixture was stirred at rt for 10 h.
Ethyl ether was added and the reaction solution was cooled to
0 °C followed by addition of acetaldehyde (1.6 g, 1.1 equiv). The
reaction mixture was allowed to warm to rt and stirring was con-
6.1.7. 9H-b-Carboline-3-carboxylic acid 2,2,2-trifluoro-1-
trifluoromethyl-ethyl ester 16
tinued for 24 h while the second equivalent of
a-pinene was lib-
Compound 16 was prepared following the procedure employed
for the preparation of 1. Compound 16: 1H NMR (300 MHz, DMSO-
d6) d 7.05–7.21 (m, 1H), 7.33–7.44 (t, 1H), 7.61–7.72 (m, 2H), 8.48
(d, J = 7.89 Hz, 1H), 9.09 (s, 2H), 12.31 (s, 1H); 13C NMR (75.5 MHz,
DMSO-d6) d 65.9, 66.8, 112.9, 120.1, 120.9, 122.8, 127.8, 129.4,
133.3, 134.9, 138.4, 141.3, 162.9. This material was pure by TLC
(silica gel).
erated. An aq solution of sodium hydroxide (2.5 M, 30 mL) was
added and the solution which resulted was extracted with ether
(3 ꢃ 30 mL). The ether layer was dried (Na2SO4) and fractionally
distilled through a Pyrex distilling column packed with glass
beads. The desired alcohol was collected along with EtOH (20%).
(S)-1,1,1-Trifluorobutan-2-ol: 1H NMR (300 MHz, CDCl3) 1.04 (dt,
J = 0.6 Hz and J = 7.4 Hz, 3H), 1.58 (m, 1H), 1.72 (m, 1H), 3.82
(m, 1H). This material was used in a later step without further
purification.
6.1.8. 9H-b-Carboline-3-carboxylic acid 2,2,2-trichloro-ethyl
ester 17
Compound 17 was prepared following the procedure employed
for the preparation of 1. Compound 17: 1H NMR (300 MHz, CDCl3)
d 5.17 (s, 2H), 7.38–7.44 (m, 1H), 7.63–7.74 (m, 2H), 8.27 (d,
J = 7.95 Hz, 1H), 8.95 (s, 1H), 9.21 (s, 1H), 9.66 (s, 1H). This material
was pure by TLC (silica gel).
6.1.12. (R)-1,1,1-Trifluoroisopropyl b-carboline-3-carboxylate
22
Compound 22 was prepared from the acid 9 and (R)-1,1,1-triflu-
oropropan-2-ol following the procedure employed for preparation
of (S)-1,1,1-trifluoro-propan-2-ol 21. Compound 22: mp 239–
241 °C; ½a 2D7
ꢄ
= 8.73 (c 0.88, in CHCl3); The spectral data for 22 were
identical to those for 21; however, the optical rotation was in the
6.1.9. 9H-b-Carboline-3-carboxylic acid 2,2,2-trifluoro-1-
methyl-ethyl ester 18
opposite direction.
Compound 18 was prepared following the procedure employed
for the preparation of 1. Compound 18: mp 247–249 °C; IR (NaCl)
6.1.13. (S)-b-Carboline-3-carboxylic acid sec-butyl ester (23)
To a solution of carbonyl diimidazole (1.53 g, 9.4 mmol) in
anhydrous DMF (50 mL) was added b-carboline-3-carboxylic acid
9 (1.0 g, 4.7 mmol). The reaction mixture was stirred for 2 h at rt
and carbon dioxide was released during the reaction. Analysis by
TLC (silica gel) indicated the absence of starting material on the
baseline. To this reaction mixture was added dry DBU (0.72 g,
4.7 mmol) and dry (S)-butyl alcohol (1.13 g, 15.2 mmol). The mix-
ture which resulted was heated at 55 °C for 8 h until analysis by
TLC (silica gel) indicated the disappearance of the imidazole inter-
mediate. The solvent was then removed under reduced pressure.
The residue was partitioned between CH2Cl2 (100 mL) and H2O
(100 mL). The organic layer was separated and the H2O layer was
extracted with CH2Cl2 (2 ꢃ 80 mL). The combined organic layer
was washed with H2O (3 ꢃ 100 mL), brine and dried (Na2SO4).
The solvent was removed under reduced pressure and the residue
was purified by flash chromatography (silica gel, EtOAc/hex-
ane = 2:1) to provide 23 (0.96 g, 76%) as a white solid. Compound
2359, 1729, 1345, 1251, 1092, 729, 450 cmꢂ1 1H NMR (300 MHz,
;
CDCl3) d 1.82 (d, J = 6.1 Hz, 3H), 5.96 (q, 1H), 7.42 (t, 1H), 7.67
(m, 2H), 8.26 (d, J = 7.6 Hz, 1H), 8.8 (s, 1H), 8.9 (s, 1H), 9.5 (br, s,
1H); 13C NMR (75.5 MHz, CDCl3) d 16.5, 79.1, 112.3, 118.5, 121.2,
121.5, 121.9, 128.8, 129.2, 133.9, 137.4. This material was pure
by TLC (silica gel).
6.1.10. (S)-1,1,1-Trifluoroisopropyl b-carboline-3-carboxylate
(21)
To a solution of carbonyl diimidazole (0.168 g, 1.03 mmol) in
anhydrous DMF (5 mL), b-carboline-3-carboxylic acid 9 (0.10 g,
0.47 mmol) was added. The reaction mixture which resulted was
stirred for 2 h at rt until analysis by TLC (silica gel) indicated the
absence of starting material on the baseline. The solution which re-
sulted was then cooled to ꢂ6 °C and this was followed by addition
of (S)-1,1,1-trifluoropropan-2-ol (2.3 equiv) which was contami-
nated with some EtOH. The dry DBU (100 mg, 0.68 mmol) in dry

W. Yin et al. / Bioorg. Med. Chem. 18 (2010) 7548–7564
7561
23: mp 212–213 °C; [
a
]
;
25 = 35.6 (CHCl3, c 1.43); IR (KBr) 3222,
1H NMR (300 MHz, CDCl3) d 0.99 (t,
6.1.17. 6-Ethynl-9H-b-carboline-3-carboxylic acid t-butyl ester
(7)-WYS8
1706, 1622, 1494 cmꢂ1
J = 7.5 Hz, 3H), 1.40 (d, J = 6.3 Hz, 3H), 1.79 (m, 2H), 5.27 (m, 1H),
7.34 (t, J = 7.9 Hz, 1H), 7.57 (t, J = 7.8 Hz, 1H), 7.79 (d, J = 8.3 Hz,
1H), 8.21 (d, J = 7.9 Hz, 1H), 8.44 (s, 1H), 9.10 (s, 1H), 11.47 (br s,
1H); MS (CI, CH4) m/e (relative intensity) 269 (M++1, 100), 241
(15), 213 (41), 195 (14). Anal. Calcd for C16H16N2O2: C, 71.62; H,
6.01; N, 10.44. Found: C, 71.33; H, 6.09; N, 10.26.
To a solution of 6-trimethyl-silanyl-b-carboline-3-carboxylic
acid t-butyl ester 6 (850 mg, 2.32 mmol) in THF (10 mL) was added
1.2 equiv of TBAF (2.8 mL of 1 M TBAF/THF solution) at 0 °C and
then the solution was allowed to warm to rt. After consumption
of the starting material as indicated by TLC, H2O (10 mL) was
added and the mixture was extracted with CH2Cl2 (3 ꢃ 20 mL).
The combined organic layer was concentrated under reduced pres-
sure and the residue was chromatographed on a short silica gel col-
umn (EtOAc/hexane = 4:1) to give 7 (620 mg, 92%) as a white solid.
Compound 7: 1H NMR (300 MHz, CDCl3) d 1.75 (s, 9H), 3.54 (s, 1H),
7.62–7.85 (m, 2H), 8.41 (s, 1H), 8.83 (s, 1H), 9.33 (s, 1H), 11.1 (s,
1H); EIMS 292 (M+, 25), 236 (12), 192 (100), 164 (30). This material
was pure by TLC (silica gel) and used directly in a later step.
6.1.14. (R)-b-Carboline-3-carboxylic acid sec-butyl ester 24
Compound 24 was prepared in 75% yield following the proce-
dure for preparation of 1. Compound 24: [
1.25). The spectral data for 22 were identical to those for 23, except
a
]
25 = ꢂ35.2 (CHCl3, c
the optical rotation was opposite in direction.
6.1.15. 6-Iodo-9H-b-carboline-3-carboxylic acid t-butyl ester (5)
Into a round-bottom flask (250 mL) was added CF3CO2Ag
(1.03 g, 4.67 mmol), b-carboline-3-carboxylic acid t-butyl ester 9
(1.02 g, 3.83 mmol), and CHCl3 (100 mL). This was followed by
addition of iodine (1.15 g, 4.66 mmol). The reaction mixture was
allowed to stir at rt for 6 h after which another portion of CF3CO2Ag
(500 mg, 2.26 mmol) was added and stirring continued for another
10 h at reflux. Analysis by TLC (silica gel) indicated that most of the
bCCt had disappeared. The reaction mixture was filtered through a
bed of Celite to remove the solid salts and the filter cake was
washed with EtOH (3 ꢃ 50 mL). The solvent was removed under
reduced pressure and the residue was purified by flash chromatog-
raphy (silica gel, EtOAc/hexane = 4:1) to provide 5 (908 mg, 65%) as
a white solid. Compound 5: mp 347–348 °C (dec); IR (NaCl) 3223,
6.1.18. 6-Thiophen-2-yl-9H-b-carboline-3-carboxylic acid t-
butyl ester (30a). Representative procedure for preparation of
6-subsitituted b-carbolines
A solution of 6-iodo-b-carboline-3-carbo-xylic acid t-butyl ester
5 (265 mg, 0.67 mmol) in dry toluene (15 mL) was degassed under
vacuum and purged with dry N2 through the solution three times.
The mixture was then heated to 140 °C under nitrogen after which
tetrakis(triphenylphosphine) palladium (0) (77 mg, 0.067 mmol,
10 mol %) and 2-(tributylstannyl) thiophene (718 mg, 2.01 mmol)
were added in one portion. The mixture was heated to reflux under
nitrogen. After 12 h, the mixture was allowed to cool to rt and the
precipitate which resulted was removed by vacuum filtration. The
filtrate was concentrated under reduced pressure and the residue
was treated with a saturated aq solution of NaHCO3 (30 mL) and
extracted with CH2Cl2 (3 ꢃ 25 mL). The combined extracts were
washed with brine and dried (Na2SO4). The solvent was removed
under reduced pressure and the residue was purified by flash chro-
matography (silica gel, EtOAc/hexane = 5:1) to provide a white so-
lid 30a (195 mg, 83%). Compound 30a: 1H NMR (300 MHz, CDCl3) d
2.07 (s, 9H), 7.15 (t, 1H), 7.31 (d, J = 3 Hz, 1H), 7.40 (d, J = 3 Hz, 1H),
7.89 (q, 2H), 8.44 (s, 1H), 8.89 (s, 1H), 9.32 (s, 1H), 11.2 (s, 1H). This
material was pure by TLC (silica gel).
1710, 1485, 1323, 1245, 1161, 1104, 1020 cmꢂ1
;
1H NMR
(300 MHz, CDCl3) d 1.68 (s, 9H), 7.63 (d, J = 8.6 Hz, 1H), 7.85 (dd,
J = 1.5 Hz and J = 8.6 Hz, 1H), 8.12 (s, 1H), 8.53 (d, J = 1.4 Hz, 1H),
9.31 (s, 1H), 11.47 (br s, 1H); 13C NMR (75.7 MHz, CDCl3) d 28.3,
80.8, 83.6, 115.1, 118.1, 123.9, 126.7, 131.2, 134.1, 136.8, 137.5,
138.7, 140.4, 164.9; MS (EI800) m/e (relative intensity) 395
(M++1, 18), 367 (20), 339 (100), 268 (33). Anal. Calcd for
C16H15IN2O2: C, 48.75; H, 3.84; N, 7.11. Found: C, 49.01; H, 3.91;
N, 6.95.
6.1.16. 6-Trimethylsilanylethynyl-9H-b-carboline-3-carboxylic
acid t-butyl ester (6)
6.1.19. 6-Furan-2-yl-9H-b-carboline-3-carboxylic acid t-butyl
ester 29a
Into a round-bottom flask (50 mL) which contained a solution of
degassed THF/Et3N (10 mL/2 mL), was added 6-iodo-9H-b-carbo-
line-3-carboxylic acid t-butyl ester 5 (400 mg, 1.02 mmol), bis[tri-
phenylphosphine] palladium dichloride (35 mg, 5 mol %) and
copper(I) iodide (7 mg, 5 mol %). Note: Practically, on a small scale,
CuI could be used up to 10–15 mol % because of its lower molecu-
lar weight compared to the palladium catalyst; on a bigger scale,
both Pd(PPh3)2Cl2 and CuI can be used as low as 0.5–1 mol %. The
reaction mixture was then degassed two times with an oil pump
at ꢂ78 °C, and then the trimethylsilyl acetylene (300 mg,
3.06 mmol) was added into the mixture and it was degassed one
more time at ꢂ78 °C. The reaction mixture was gradually allowed
to warm to rt and stirred at rt for an additional 0.5–1 h until all the
starting material had disappeared (TLC analysis indicated that the
original red spot changed color to purple, since both s.m. and prod-
uct had very similar Rf values). The solvent was removed under re-
duced pressure at this point and the residue was purified by flash
Compound 29a was prepared following the procedure for prep-
aration of 30a. Compound 29a: 1H NMR (300 MHz, CDCl3) d 1.63 (s,
9H), 6.63 (m, 1H), 6.97 (d, J = 6 Hz, 1H), 7.70 (d, J = 9 Hz, 1H), 7.77
(s, 1H), 7.96 (d, J = 9 Hz, 1H), 8.73 (s, 1H), 8.90 (s, 1H), 8.96 (s, 1H),
12.1 (s, 1H); 13C NMR (75.5 MHz, CDCl3) d 28.4, 80.7, 102.3, 106.3,
112.4, 113.2, 117.4, 117.8, 127.1, 127.8, 133.6, 134.1, 138, 140.6,
142.6, 154.1, 165.0, 166.0. This material was pure by TLC (silica
gel).
6.1.20. 6-Furan-2-yl-9H-b-carboline-3-carboxylic acid 2,2,2-
trifluoro-ethyl ester 29b
6-Furan-2-yl-9H-b-carboline-3-carboxylic acid 2,2,2-trifluoro-
ethyl ester 29b was prepared following the procedure for prepara-
tion of 30a. Compound 29b: 1H NMR (250 MHz, CDCl3) d 4.82–4.92
(m, 2H), 6.54 (s, 1H), 6.75 (s 1H), 7.38 (s, 1H), 7.65 (d, J = 10 Hz, 1H),
7.96 (d, J = 10 Hz, 1H), 8.5 (s, 1H), 8.93 (s, 1H), 9.13 (s, 1H), 9.59 (s,
1H); 13C NMR (75.5 MHz, CDCl3) d 60.9, 105.3, 112.5, 112.8, 113.3,
117.6, 119.2, 122.3, 123.8, 125.3, 127.9, 134.7, 135.3, 138.6, 140.6,
142.7, 154.1, 164.4. This material was pure by TLC (silica gel).
chromatography (silica gel, EtOAc/hexane = 4:1) to provide
6
(340 mg, 92%) as a white solid. 1H NMR (300 MHz, CDCl3) d 1.71
(s, 9H), 1.75 (s, 9H), 7.67 (m, 2H), 8.26 (d, J = 7.6 Hz, 1H), 8.8 (s,
1H), 8.9 (s, 1H), 9.5 (br, 1H); 13C NMR (75.5 MHz, CDCl3) d 0.021,
28.3, 81.9, 92.6, 105.7, 113.2, 114.9, 117.7, 121.3, 125.5, 128.3,
132.4, 133.8, 138.2, 138.4, 141.5, 165.4; EIMS 364 (M+, 38), 293
(12), 264 (100), 249 (48), 124 (60). This material was employed di-
rectly in the next step.
6.1.21. 6-Iodo-b-carboline-3,9-dicarboxylic acid di-tert-butyl
ester (35)
To a solution of 6-iodo-b-carboline-3-carboxylic acid t-butyl es-
ter 5 (2 g, 5.06 mmol) in anhydrous CH2Cl2 (30 mL), (Boc)2O
(1.32 g, 6.07 mmol) and DMAP(123 mg, 1.01 mmol) were added.

7562
W. Yin et al. / Bioorg. Med. Chem. 18 (2010) 7548–7564
The reaction mixture was allowed to stir at rt for half an hour until
analysis by TLC (silica gel) indicated that the starting material had
been converted into the Boc protected indole 35. The solvent was
removed under reduced pressure and the residue was purified by
flash chromatography (silica gel, EtOAc/hexane = 5:95) to provide
35 (2.3 g, 92%) as a white solid. Compound 35: mp 317–320 °C;
IR (NaCl) 2975, 2917, 1728, 1457, 1343, 1238, 1154, 1119, 1031,
6-ethynyl-b-carboline-3,9-dicarboxylic acid di-tert-butyl ester 34
(1.6 g, 4.1 mmol) and 6-iodo-b-carboline-3,9-dicarboxylic acid di-
tert-butyl ester 35 (2.1 g, 4.25 mmol) in THF/TEA (30 mL; 4:1).
The reaction mixture which resulted was degassed, and back-filled
three times with argon. The reaction mixture was then allowed to
stir at rt for about 1 h until analysis by TLC (silica gel) indicated the
starting materials were absent. The solution was concentrated un-
der reduced pressure and the residue was chromatographed on a
silica gel column with CH2Cl2 as the eluent to give 1,2-bis(b-carb-
oline-3,9-dicarboxylic acid di-tert-butyl ester) ethyne (2.97 g, 95%)
as a white solid: mp 305–307 °C; IR (NaCl) 2972, 2929, 1737, 1559,
1466, 1338, 1156, 1102, 823, 624 cmꢂ1; 1H NMR (300 MHz, CDCl3)
d 1.72 (s, 9H), 1.79 (s, 9H), 7.86 (d, J = 8.97 Hz, 1H), 8.32 (s, 1H),
8.47 (d, J = 8.79 Hz, 1H), 8.68 (s, 1H), 9.63 (s, 1H); 13C NMR
(75.5 MHz, CDCl3) d 28.2, 82.1, 85.8, 89.0, 116.0, 116.7, 118.6,
123.6, 124.3, 131.5, 133.3, 136.1, 138.1, 139.1, 143.0, 149.7,
164.2; MS (FAB) 759 ((M+H)+, 13). Anal. Calcd for C44H46N4O8
(H2O): C, 68.03; H, 6.22; N, 7.21. Found: C, 68.10; H, 6.22; N,
7.21.
811 cmꢂ1 1H NMR (300 MHz, CDCl3) d 1.71 (s, 9H), 1.77 (s, 9H),
;
7.91 (d, J = 8.972 Hz, 1H), 8.22 (d, J = 8.972 Hz, 1H), 8.42 (s, 1H),
8.61 (s, 1H), 9.61 (s, 1H); 13C NMR (75.5 MHz, CDCl3) d 28.1, 28.2,
82.1, 85.8, 87.2, 115.9, 118.5, 125.7, 129.9, 130.6, 135.7, 138.9,
143.0, 149.6, 164.1; MS (EI) m/e (relative intensity) 494 (M+, 42),
438 (71), 338 (100), 294 (70), 268 (8), 212 (8), 168 (35). Anal. Calcd
for C21H23IN2O4 (0.1 C6H14): C, 51.60; H, 4.85; N, 5.57. Found: C,
51.84; H, 4.88; N: 5.45.
6.1.22. 6-Trimethylsilanylethynyl-b-carboline-3,9-dicarboxylic
acid di-tert-butyl ester (36)
Dichlorobis(triphenylphosphine)palladium(II) (140 mg; 2 mol %),
and CuI (40 mg; 2 mol %) were added to a solution of 6-iodo-
b-carboline-3,9-dicarboxylic acid di-tert-butyl ester 35 (4.9 g;
10 mmol) in anhydrous THF (30 mL) and triethylamine (10 mL).
The mixture was degassed, and back-filled three times with argon.
Then trimethylsilyl acetylene (1.08 g; 11 mmol) was added with
stirring under argon. After the mixture was allowed to stir for
1 h, the solvents were removed under vacuum and the residue
was chromatographed on a short column (silica gel, hexane/
CH2Cl2 = 7:3) to give 36 (4.36 g, 94%) as a white solid. Compound
36: mp 334–336 °C (dec); IR (NaCl) 2974, 2137, 1732, 1559,
1476, 1469, 1368, 1343, 1309, 1247, 1156, 1109, 872, 842,
The 1,2-bis(b-carboline-3,9-dicarboxylic acid di-tert-butyl es-
ter) ethyne 32 (800 mg, 1.05 mmol) was added to a distilled solu-
tion of cumene (40 mL), which had been degassed. The reaction
vessel was evacuated and refilled with nitrogen three times. The
temperature was then brought to reflux for about 30 min until a
yellow precipitate had formed. The mixture which resulted was fil-
tered and washed with hexane to give pure dimer 32 (545 mg,
93%). Compound 32: mp >350 °C (dec); IR (KBr) 3227, 1716,
1327, 1162, 738, 450 cmꢂ1 1H NMR (300 MHz, DMSO) d 1.62 (s,
;
9H), 7.70–7.80 (m, 2H), 8.7 (s, 1H), 8.94 (s, 1H), 8.99 (s, 1H),
12.25 (s, 1H); 13C NMR (300 MHz, CDCl3) d 28.3, 80.8, 113.3,
114.6, 117.9, 125.8, 127.5, 131.9, 134.2, 138.0, 140.8, 164.9; MS
(FAB) 559 ((M+H)+, 41). This material was pure by TLC (silica gel).
760 cmꢂ1 1H NMR (300 MHz, CDCl3) d 0.31 (s, 9H), 1.71 (s, 9H),
;
1.78 (s, 9H), 7.77 (d, J = 8.97 Hz, 2H), 8.25 (s, 1H), 8.41 (d,
J = 8.61 Hz, 2H), 8.65 (s, 1H), 9.62 (s, 1H); 13C NMR (75 MHz, CDCl3)
d 0.11, 28.1, 28.2, 82.1, 85.8, 94.4, 104.4, 116, 116.5, 118.7, 123.5,
124.7, 131.6, 133.7, 136.2, 138.1, 139.2, 142.9, 149.7, 164.1; MS
(EI) m/e (relative intensity) 465 (M+, 30), 409 (100), 365 (21), 308
(80), 262 (25), 249 (40). This material was pure by TLC (silica
gel) and used in the next step.
6.1.25. 1,4-Bis(9H-b-carboline-3-carboxylic acid tert-butyl ester)
buta-1,3-diyne (33)
In
a round-bottom flask (200 mL), PdCl2(PPh3)2 (58 mg,
2 mol %), CuI (16 mg, 2 mol %), and N,N-diisopropyl-ethylamine
(534 mg, 4.92 mmol) were added and the mixture stirred under ar-
gon. The flask was evacuated (degassed) and refilled with argon.
The THF (40 mL) and 6-ethynyl-b-carboline-3,9-dicarboxylic acid
di-tert-butyl ester 34 (1.6 g, 4.1 mmol) were then added (under ar-
gon) to the above mixture. To this flask methyl bromoacetate
(410 mg, 2.5 mmol) was added, and the reaction mixture was stir-
red at rt for 6–8 h. The progress of this reaction was monitored by
TLC on silica gel. After the reaction was complete, 8–10 g of silica
gel was added and the solvent was removed under vacuum. The so-
lid residue (a plug) was then placed on a column and subjected to
column chromatography (CH2Cl2) to give 1,4-bis(b-carboline-3,9-
dicarboxylic acid di-tert-butyl ester) buta-1,3-diyne (2.08 g, 65%):
1H NMR (300 MHz, CDCl3) d 1.74 (s, 9H), 1.81 (s, 9H), 7.82 (d,
J = 8.79 Hz, 1H), 8.34 (s, 1H), 8.48 (d, J = 8.79 Hz, 1H), 8.69 (s, 1H),
9.65 (s, 1H); 13C NMR (75.5 MHz, CDCl3) d 28.2, 74.0, 81.2, 82.2,
86.0, 116.1, 116.9, 117.2, 123.7, 125.5, 131.3, 134.0, 136.2, 138.1,
139.6, 143.2, 149.6, 164.1; MS (FAB) 783 ((M+H)+, 100).
The 1,4-bis-(b-carboline-3,9-dicarboxylic acid di-tert-butyl es-
ter) buta-1,3-diyne (411 mg, 0.5 mmol) was added to a distilled
solution of cumene (20 mL), which had been degassed. The reac-
tion mixture was evacuated and refilled with nitrogen three times.
The temperature was then brought to reflux for about 30 min until
a yellow precipitate had formed. The mixture was filtered and the
solids washed with hexane to give pure dimer 33. Compound 33:
mp >350 °C (dec) IR(KBr) 3424, 1708, 1627, 1466, 1369, 1302,
1251, 1154, 1107, 1025, 846, 645 cmꢂ1; 1H NMR (300 MHz, DMSO)
d 1.63 (s, 9H), 7.60–7.69 (m, 2H), 8.40 (s, 1H), 8.96 (s, 1H), 9.04 (s,
1H), 12.5 (s, 1H); 13C NMR (75.5 MHz, CDCl3) d 28.3, 78.8, 82.5,
112.9, 118.1, 121.0, 122.1, 129.5, 131.5, 134.8, 141.1, 163.0,
6.1.23. 6-Ethynyl-b-carboline-3,9-dicarboxylic acid di-tert-butyl
ester 34
To
a solution of 6-trimethylsilanylethynyl-b-carboline-3,9-
dicarboxylic acid di-tert-butyl ester 36 (2.14 g, 4.6 mmol) in THF
(20 mL), 1.2 equiv of TBAF (5.52 mL of 1 M TBAF/THF solution) at
0 °C was added and then the solution was allowed to warm to rt.
After consumption of the starting material as indicated by TLC (sil-
ica gel), H2O (10 mL) was added and the mixture was extracted
with CH2Cl2 (20 mL, 3ꢃ). The combined organic layer was concen-
trated under reduced pressure and the residue was chromato-
graphed on a short column (silica gel, hexane/CH2Cl2 = 4:1) to
give 34 (1.66 g, 92%) as a white solid. Compound 34: mp 226–
229 °C; IR (NaCl) 3303, 2978, 2346, 2232, 1734, 1622, 1560,
1463 cmꢂ1 1H NMR (300 MHz, CDCl3) d 1.54 (s, 9H), 1.72 (s, 9H),
;
1.79 (s, 9H), 3.17 (s, 1H), 7.79 (d, J = 8.76 Hz, 2H), 8.27 (s, 1H),
8.44 (d, J = 8.94 Hz, 2H), 8.67 (s, 1H), 9.64 (s, 1H); 13C NMR
(75.5 MHz, CDCl3) d 27.3, 28.1, 28.2, 67.8, 82.1, 83.0, 85.1, 85.5,
116.0, 116.7, 123.5, 124.9, 133.7, 136.2, 138.1, 139.4, 143.0,
164.2; MS (EI) m/e (relative intensity) 393 (M+, 32), 338 (13), 321
(26), 293 (100), 167 (28), 139 (24). Anal. Calcd for C23H24N2O4
(0.05 CH2Cl2): C, 69.78; H, 6.12; N, 7.06. Found: C, 69.70; H, 6.10;
N, 6.81.
6.1.24. 1,2-Bis(9H-b-carboline-3-carboxylic acid tert-butyl ester)
ethyne (32)
Dichlorobis (tri-phenylphosphine)palladium(II) (60 mg, 2 mol %)
and copper iodide (16 mg, 2 mol %) were added to a mixture of

W. Yin et al. / Bioorg. Med. Chem. 18 (2010) 7548–7564
7563
131.9, 134.2, 138.0, 140.8, 164.9; MS (FAB) 583 (M+, 100). This
material was pure by TLC (silica gel).
The synthesis of the ligands in Table 2b had been reported pre-
viously in Ref. 89.
41. Lumeng, L.; Murphy, J. M.; McBride, W. J.; Li, T.-K. Genetic Influences on Alcohol
Preference in Animals; Oxford University Press: New York, 1995. pp 165–201.
42. June, H. L.; Eggers, M. W.; Warren-Reese, C.; DeLong, J.; Ricks-Cord, A.; Durr, L.
F.; Cason, C. R. Eur. J. Pharmacol. 1998, 350, 151.
43. June, H. L.; Torres, L.; Cason, C. R.; Hwang, B. H.; Braun, M. R.; Murphy, J. M.
Brain Res. 1998, 784, 256.
44. Sieghart, W.; Ernst, M. Curr. Med. Chem.: Cent. Nerv. Syst. Agents 2005, 5, 217.
45. Bowser, D. N.; Wagner, D. A.; Czajkowski, C.; Cromer, B. A.; Parker, M. W.;
Wallace, R. H.; Harkin, L. A.; Mulley, J. C.; Marini, C.; Berkovic, S. F.; Williams,
D. A.; Jones, M. V.; Petrou, S. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 15170.
46. Bateson, A. N. Sleep Med. 2004, 5, S9–S15.
47. Otani, K.; Ujike, H.; Tanaka, Y.; Morita, Y.; Katsu, T.; Nomura, A.; Uchida, N.;
Hamamura, T.; Fujiwara, Y.; Kuroda, S. Neurosci. Lett. 2005, 381, 108.
48. Dean, B.; Scarr, E.; McLeod, M. Brain Res. Mol. Brain Res. 2005, 138, 145.
49. Guidotti, A.; Auta, J.; Davis, J. M.; Dong, E. B.; Grayson, D. R.; Veldic, M.; Zhang,
X. Q.; Costa, E. Psychopharmacology 2005, 180, 191.
Acknowledgments
This work was supported in part by NIMH 046851 (J.M.C.) as
well as NIAAA AA016179 (DMP) and NCRR RR00168 (DMP). We
acknowledge support of this work by the Research Growth Initia-
tive of the University of Wisconsin-Milwaukee and the Lynde
and Harry Bradley Foundation.
50. Maubach, K. Drug Targets CNS Neurol. Disord. 2003, 2, 233.
51. Barrett, A. C.; Negus, S. S.; Mello, N. K.; Caine, S. B. J. Pharmacol. Exp. Ther. 2005,
315, 858.
References and notes
52. Anthenelli, R.; Schuckit, M. A. Inter. J. Addict. 1990/1991, 25, 81.
53. Schuckit, M. A. Alcohol Clin. Exp. Res. 1988, 12, 465.
54. Barnard, E.; Skolnick, P.; Olsen, R.; Möhler, H.; Sieghart, W.; Biggio, G.;
Braestrup, C.; Bateson, A.; Langer, S. J. Pharmacol. Exp. Ther. (Pharmacol. Rev.)
1998, 50, 291.
1. Cook, J. M.; Van Linn, M. L.; Yin, W., 2009, 92 pp. CODEN: PIXXD2. (UWM
Research Foundation, Inc., USA). WO 2009-US45014 WO 2009143445 A1 CAN
151:571093 AN 2009:1468879.
2. Harvey, S. C.; Foster, K. L.; McKay, P. F.; Carroll, M. R.; Seyoum, R.; Woods, J. E.;
Grey, C.; Jones, C. M.; McCane, S.; Cummings, R.; Mason, D.; Ma, C. R.; Cook, J.
M.; June, H. L. J. Neurosci. 2002, 22, 3765.
55. Simon, J.; Wakimoto, H.; Fujita, N.; Lalande, M.; Barnard, E. A. J. Biol. Chem.
2004, 279, 41422.
3. June, H. L.; Foster, K. L.; McKay, P. F.; Seyoum, R.; Woods, J. E.; Harvey, S. C.;
Eiler, W. J. A.; Grey, C.; Carroll, M. R.; McCane, S.; Jones, C. M.; Yin, W. Y.;
Mason, D.; Cummings, R.; Garcia, M.; Ma, C. R.; Sarma, P.; Cook, J. M.; Skolnick,
P. Neuropsychopharmacology 2003, 28, 2124.
4. Kessler, R. C.; Frank, R. G. Psychiatr. Med. 1997, 27, 861.
5. O’Malley, S. S.; Jaffe, A. J.; Chang, G.; Schottenfeld, R. S.; Meyer, R. E.;
Rounsaville, B. Arch. Gen. Psychiatry 1992, 49, 881.
56. Sieghart, W.; Sperk, G. Curr. Top. Med. Chem. 2002, 2, 795.
57. Mossier, B.; Togel, M.; Fuchs, K.; Sieghart, W. J. Biol. Chem. 1994, 269,
25777.
58. Togel, M.; Mossier, B.; Fuchs, K.; Sieghart, W. J. Biol. Chem. 1994, 269, 12993.
59. Pirker, S.; Schwarzer, C.; Wieselthaler, A.; Sieghart, W.; Sperk, G. Neuroscience
2000, 101, 815.
60. (a) Cox, E. D.; Diaz-Arauzo, H.; Huang, Q.; Reddy, M. S.; Harris, B.; Mckernan, R.
M.; Skolnick, P.; Cook, J. M. J. Med. Chem. 1998, 41, 2537; (b) Huang, Q. Ph.D.
Thesis, University of Wisconsin-Milwaukee, 1998.
6. Volpicelli, J. R.; Alterman, A. I.; Hayashida, M.; O’Brien, C. P. Arch. Gen.
Psychiatry 1992, 49, 876.
7. Kranzler, H. R. Alcohol 2000, 35, 537.
8. Spanagel, R.; Zieglgansberger, W. Trends Pharmacol. Sci. 1997, 18.
9. Cloninger, C. R. Science 1987, 236, 410.
61. Huang, H.; He, X.; Ma, C. R.; Liu, R. Y.; Yu, S.; Dayer, C. A.; Wenger, G. R.;
McKernan, R.; Cook, J. M. J. Med. Chem. 2000, 43, 71.
62. Carroll, M.; Woods, J. E., II; Seyoum, R. A.; June, H. L. Alcohol Clin. Exp. Res.
2001, 25.
10. Li, T.-K. J. Stud. Alcohol 2000, 61, 5.
11. Li, T.-K.; Crabb, D. W.; Lumeng, L. Neuropharmacology of Ethanol; Birkhauser:
Boston, 1991. pp 107–124.
12. Johnson, B. A.; Ait-Daoud, N. Psychopharmacology 2000, 149, 327.
13. Koob, G. F.; Roberts, A. J.; Schulteis, G. et al. Alcohol Clin. Exp. Res. 1998, 22, 3.
14. June, H. L.; Cason, C. R.; Cheatham, G.; Liu, R. Y.; Gan, T.; Cook, J. M. Brain Res.
1998, 794, 103.
15. McBride, W. J.; Li, T.-K. Crit. Rev. Neurobiol. 1998, 12, 339.
16. Allain, H.; Belliard, S.; Decertaines, J.; Bentueferrer, D.; Bureau, M.; Lacroix, P.
Dementia 1993, 4, 347.
63. Rowlett, J. K. Unpublished results, 2010.
64. June, H. L.; Cook, J. M.; Ma, C. U.S. Patent Appl. Publ., 2003, 60 pp. CODEN:
USXXCO. (USA). US 2002-329100 US 20030176456 AN 2003:737374 CAN
139:241570.
65. Foster, K. L.; McKay, P. F.; Seyoum, R.; Milbourne, D.; Yin, W.; Sarma, P. V. V. S.;
Cook, J. M.; June, H. L. Neuropsychopharmacology 2004, 29, 269.
66. Araki, T.; Tohyama, M. Mol. Brain Res. 1992, 12, 293.
67. Cox, E. D.; Hagen, T. J.; Mckernan, R. M.; Cook, J. M. Med. Chem. Res. 1995, 5,
710.
17. Heimer, L.; Alheid, G. F. The Basal Forebrain: Anatomy and Function; Plenum
Press: New York, 1991. pp 1–42.
68. Griebel, G.; Perrault, G.; Letang, V.; Grainger, P.; Avenet, P.; Schoemaker, H.
Psychopharmacology 1999, 146, 205.
18. Heimer, L.; Zahm, D. S.; Churchill, P.; Kalivas, W. Neuroscience 1991, 41, 89.
19. Criswell, H. E.; Simson, P. E.; Duncan, G. E.; Mc Cown, T. J.; Herbert, J. S.;
Morrow, L. J. Pharmacol. Exp. Ther. 1993, 267, 522.
69. Carroll, M. E.; Carmona, G.; May, S. J. Exp. Anal. Behav. 1991, 18, 361.
70. Paronis, C. A.; Cox, E. D.; Cook, J. M.; Bergman, J. Psychopharmacology 2001,
156, 461.
20. Criswell, H. E.; Simson, P. E.; Knapp, D. J.; Devaud, L. L.; Mc Cown, T. J.; Duncan,
G. E. J. Pharmacol. Exp. Ther. 1995, 273, 525.
71. Rowlett, J. K.; Tornatzky, W.; Cook, J. M.; Ma, C. R.; Miczek, K. A. J. Pharmacol.
Exp. Ther. 2001, 297, 247.
21. June, H. L.; Zuccarelli, D.; Torres, L.; Craig, K. S.; DeLong, J.; Allen, A.; Braun, M.
R.; Cason, C. R.; Murphy, J. M. J. Pharmacol. Exp. Ther. 1998, 284, 1006.
22. June, H. L.; Harvey, S. C.; Foster, K. L.; McKay, P. F.; Cummings, R.; Garcia, M.;
Mason, D.; Grey, C.; McCane, S.; Williams, L. S.; Johnson, T. B.; He, X. H.; Rock,
S.; Cook, J. M. J. Neurosci. 2001, 21, 2166.
23. Hyytia, P.; Koob, G. F. Eur. J. Pharmacol. 1995, 283, 151.
24. Nowak, K. L.; McBride, W. J.; Lumeng, L.; Li, T.-K.; Murphy, J. M.
Psychopharmacology 1998, 139, 108.
25. Nauta, H. J.; Smith, G. P.; Faull, R. L. M.; Domesick, V. B. Neuroscience 1978, 3.
26. Zahm, D. S.; Heimer, L. J. Comp. Neurol. 1988, 272, 516.
27. Groenwegen, H. J.; Berende, H. W. Neuroscience 1993, 57, 113.
28. Churchill, L.; Kalivas, P. W. J. Comp. Neurol. 1994, 345, 579.
29. Kuo, H.; Chang, H. T. J. Comp. Neurol. 1992, 321, 626.
30. Morgenson, G. J.; Nielson, M. Brain Res. Bull. 1983, 11, 309.
31. Fritschy, J. M.; Mohler, H. J. Comp. Neurol. 1995, 359, 154.
32. Turner, J. D.; Bodewitz, G.; Thompson, C. L.; Stephenson, F. A. Anxiolytic Beta
Carbolines: From Molecular Biology to the Clinic; Springer: New York, 1993. pp
29–49.
72. Shannon, H. E.; Guzman, F.; Cook, J. M. Life Sci. 1984, 35, 2227.
73. Rowlett, J. K.; Spealman, R. D.; Lelas, S.; Cook, J. M.; Yin, W. Y.
Psychopharmacology 2003, 165, 209.
74. Rowlett, J. K.; Cook, J. M.; Duke, A. N.; Platt, D. M. CNS Spectr. 2005, 10, 40.
75. Liu, R.; Zhang, P.; Mckernan, R. M.; Wafford, K. A.; Cook, J. M. Med. Chem. Res.
1995, 5, 700.
76. Liu, R.; Hu, R. J.; Zhang, P.; Skolnick, P.; Cook, J. M. J. Med. Chem. 1996, 39, 1928.
77. Li, X. Y.; Cao, H.; Zhang, C. C.; Furtmueller, R.; Fuchs, K.; Huck, S.; Sieghart, W.;
Deschamps, J.; Cook, J. M. J. Med. Chem. 2003, 46, 5567.
78. Allen, M. S.; Tan, Y. C.; Trudell, M. L.; Narayanan, K.; Schindler, L.; Martin, M. J.;
Schultz, C. A.; Hagen, T. J.; Koehler, K. F.; Codding, P.; Skolnick, P.; Cook, J. J.
Med. Chem. 1990, 33, 2343.
79. Allen, M. S.; Laloggia, A. J.; Dorn, L. J.; Martin, M. J.; Constantino, G.; Hagen, T.
J.; Koehler, K. F.; Skolnick, P.; Cook, J. J. Med. Chem. 1992, 35, 4001.
80. Cain, M.; Weber, R. W.; Guzman, F.; Cook, J. M.; Barker, S. A.; Rice, K. C.;
Crawley, J. N.; Paul, S. M.; Skolnick, P. J. Med. Chem. 1982, 25, 1081.
81. Hagen, T. J.; Skolnick, P.; Cook, J. M. J. Med. Chem. 1987, 30, 750.
82. Cook, J. M.; Diaz-Arauzo, H.; Allen, M. S. NIDA Res. Monogr. 1990, 133.
83. Schweri, M.; Cain, M.; Cook, J. M.; Paul, S.; Skolnick, P. Pharmacol. Biochem.
Behav. 1982, 17, 457.
33. Churchill, L.; Bourdelais, A.; Austin, M. C.; Lolait, S. J.; Mahan, L. C.; O’Carroll, A.
M.; Kalivas, P. W. Synapse 1991, 8, 75.
34. Wisden, W.; Laurie, D. J.; Monyer, H.; Seeburg, P. H. J. Neurosci. 1992, 12,
1040.
84. Fryer, R. I.; Cook, C.; Gilman, N. W.; Walser, A. Life Sci. 1986, 39, 1947.
85. Trullas, R.; Ginter, H.; Jackson, B.; Skolnick, P.; Allen, M. S.; Hagen, T. J.; Cook, J.
M. Life Sci. 1988, 43, 1193.
86. Ninan, P. T.; Insel, T. M.; Cohen, R. M.; Cook, J. M.; Skolnick, P.; Paul, S. M.
Science 1982, 218, 1332.
87. Mendelson, W. B.; Cain, M.; Cook, J. M.; Paul, S. M.; Skolnick, P. Science 1983,
219, 414.
88. Hagen, T. J.; Guzman, F.; Schultz, C.; Cook, J. M.; Shannon, H. E. Heterocycles
1986, 24, 2845.
35. Duncan, G. E.; Breese, G. R.; Criswell, H. E.; McCown, T. J.; Herbert, J. S.;
Devaud, L. L.; Morrow, A. L. Neuroscience 1995, 64, 1113.
36. Napier, T. C.; Chrobak, J. J. Neuroreport 1992, 3, 609.
37. Hubner, C. B.; Koob, G. F. Eur. J. Pharmacol. 1995, 283, 151.
38. Gong, W.; Justice, J. B. Prog. Neuropsychohpharmacol. Biol. Psychiatry 1997, 21,
839.
39. Gong, W.; Neill, D. Brain Res. 1996, 707, 64.
40. Cicero, T. J.. In Biochemistry and Pharmacology of Ethanol; Plenum Press: New
York, 1979; Vol. 2. pp 533–560.
89. Allen, M. S.; Hagen, T. J.; Trudell, M. L.; Codding, P. W.; Skolnick, P.; Cook, J. M.
J. Med. Chem. 1988, 31, 1854.

7564
W. Yin et al. / Bioorg. Med. Chem. 18 (2010) 7548–7564
90. Diaz-Arauzo, H.; Evoniuk, G. E.; Skolnick, P.; Cook, J. M. J. Med. Chem. 1991, 34,
1754.
91. Wafford, K. A.; Bain, C. J.; Whiting, P. J.; Kemp, J. A. Mol. Pharmacol. 1993, 44, 437.
92. Braestrup, C.; Nielsen, M. Nature 1981, 294, 472.
93. Venault, P.; Chapouthier, G.; de-Carvalho, L. P.; Simiand, J.; Morre, M.; Dodd, R.
H.; Rossier, J. Nature 1986, 321, 864.
94. Lippke, K. P.; Schunack, W. G.; Wenning, W.; Müller, W. E. J. Med. Chem. 1983,
26, 499.
95. Corda, M. G.; Blaker, W. D.; Mendelson, W. B.; Guidotti, A.; Costa, E. Proc. Natl.
Acad. Sci. U.S.A. 1983, 80, 2072.
96. Havoundjian, H.; Reed, G. F.; Paul, S. M.; Skolnick, P. J. Clin. Invest. 1987, 79,
473.
111. Cromer, B.; Morton, C.; Parker, M. W. Trends Biochem. Sci. 2002, 27, 280.
112. Trudell, J. R. Biochim. Biophys. Acta 2002, 1565, 91.
113. Ma, C. Ph.D. Thesis, University of Wisconsin-Milwaukee, 2000.
114. Ramachandran, P. V.; Teodorovic, A. V.; Brown, H. C. Tetrahedron 1993, 49,
1725.
115. Ramachandran, P. V.; Gong, B.; Brown, H. C. J. Org. Chem. 1995, 60, 41.
116. Sonogashira, K.; Tohda, Y.; Hagihara, N. Tetrahedron Lett. 1975, 50, 4467.
117. Sonogashira, K. Organometal. Chem. 2002, 653, 46.
118. Lei, A.; Srivastava, M.; Zhang, X. J. Org. Chem. 2002, 67, 1969.
119. Choudhary, M. S.; Craigo, S.; Roth, B. L. Mol. Pharmacol. 1992, 42, 627.
120. Lüddens, H.; Korpi, E. R.; Seeburg, P. H. Neuropharmacology 1995, 34, 245.
Review.
97. Hadingham, K. L.; Wingrove, P.; Le-Bourdelles, B.; Palmer, K. J.; Ragan, C. I.;
Whiting, P. J. Mol. Pharmacol. 1993, 43, 970.
121. Huang, Q.; Zhang, W. J.; Liu, R. Y.; McKernan, R. M.; Cook, J. M. Med. Chem. Res.
1996, 6, 384.
98. Sanger, D. J.; Benavides, J.; Perrault, G.; Morel, E.; Cohen, C.; Joly, D.; Zivkovic,
B. Neurosci. Biobehav. Rev. 1994, 18, 355.
122. Huang, Q.; Liu, R. Y.; Zhang, P. W.; He, X. H.; McKernan, R.; Gan, T.; Bennett, D.
W.; Cook, J. M. J. Med. Chem. 1998, 41, 4130.
99. McKernan, R. M.; Rosahl, T. W.; Reynolds, D. S.; Sur, C.; Wafford, K. A.; Atack, J.
R.; Farrar, J.; Cook, G.; Ferris, P.; Garrett, L.; Bristow, L.; Marshall, G.; Macaulay,
A.; Brown, N.; Howell, O.; Moore, K. W.; Carling, R. W.; Street, L. J.; Castro, J. L.;
Ragan, C. I.; Dawson, G. R.; Whiting, P. J. Nat. Neurosci. 2000, 3, 587.
100. Braestrup, C.; Nielsen, M.; Olsen, R. Proc. Natl. Acad. Sci. U.S.A. 1980, 77, 2288.
101. Nielsen, M.; Braestrup, C. Nature 1980, 286, 606.
102. Huang, Q.; Cox, E. D.; Gan, T.; Ma, C. R.; Bennett, D. W.; Mckernan, R. M.; Cook,
J. M. Drug. Des. Discov. 1999, 16, 55.
103. Yin, W.; Sarma, P. V. V. S.; Ma, J.; Han, D.; Chen, J. L.; Cook, J. M. Tetrahedron
Lett. 2005, 46, 6363.
104. Portoghese, P. S. J. Med. Chem. 2001, 44, 2259.
105. Portoghese, P. S.; Lin, C. E.; Farouzgrant, F.; Takemori, A. E. J. Med. Chem. 1994,
37, 1495.
106. Halazy, S. Exp. Opin. Ther. Patents 1999, 9, 431.
123. He, X. H.; Huang, Q.; Ma, C. R.; Yu, S.; McKernan, R.; Cook, J. Drug Des. Discov.
2000, 17, 131.
124. O’Mara, M.; Cromer, B.; Parker, M.; Chung, S. H. Biophys. J. 2005, 88, 3286.
125. Arnold, K.; Bordoli, L.; Kopp, J.; Schwede, T. Bioinformatics 2006, 22, 195.
126. Buhr, A.; Baur, R.; Sigel, E. J. Biol. Chem. 1997, 272, 11799.
127. Buhr, A.; Schaerer, M. T.; Baur, R.; Sigel, E. Mol. Pharmacol. 1997, 52, 676.
128. Buhr, A.; Sigel, E. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 8824.
129. Mihic, S. J.; Whiting, P.; Klein, R. L.; Wafford, K.; Harris, R. A. J. Biol. Chem.
1994, 269, 32768.
130. Sigel, E.; Schaerer, M. T.; Buhr, A.; Baur, R. Mol. Pharmacol. 1998, 54, 1097.
131. Braestrup, C.; Honore, T.; Nielsen, M. C.; Peterson, E. N.; Jensen, L. H. Biochem.
Pharmacol. 1984, 33, 859.
132. Lawson, J.; Uyeno, E. T.; Nienow, J.; Loew, G. H.; Toll, L. Life Sci. 1984, 34,
2007.
107. June, H. L., Sr.; Foster, K. L.; Eiler, W. J. A. I.; Goergen, J.; Cook, J. B.; Johnson, N.;
Mensah-Zoe, B.; Simmons, J. O.; June, H. L., Jr.; Yin, W.; Cook, J. M.; Homanics,
G. E. Neuropsychopharmacology 2007, 32, 137.
108. Bell, R. L.; Rodd, Z. A.; Lumeng, L.; Murphy, J. M.; McBride, W. J. Addict. Biol.
2006, 11, 270.
109. Koob, G. F. Biochem. Pharmacol. 2004, 68, 1515. review and references cited
therein.
110. Clayton, T.; Chen, J. L.; Ernst, M.; Richter, L.; Cromer, B. A.; Morton, H. N.;
Kaczorowski, C. C.; Helmstetter, F. J.; Furtmuller, R.; Ecker, G.; Parker, M. W.;
Sieghart, W.; Cook, J. M. Curr. Med. Chem. 2007, 14, 2755.
133. Hanzawa, Y.; Kawagoe, K.; Ito, M.; Kobayashi, Y. Chem. Pharm. Bull. 1987, 35,
1633.
134. Liu, R.; Zhang, P.; Gan, T.; Mckernan, R. M.; Cook, J. M. Med. Chem. Res. 1997, 7,
25.
135. Saiga, Y.; Iijima, I.; Ishida, A.; Miyagishima, T.; Shigezane, K.; Oh-Ishi, T.;
Matsumoto, M.; Matsuoka, Y. Chem. Pharm. Bull. 1987, 35, 3262.
136. Eftink, M. R.; Jia, J.; Hu, D.; Ghiron, C. A. J. Phys. Chem. 1995, 99, 5713.
137. Plate, R.; Nivard, R. J. F.; Ottenheijm, H. C. J.; Kardos, J.; Simonyi, M.
Heterocycles 1986, 24, 3105.
138. Moody, C. J.; Ward, J. G. J. Chem. Soc. Perkin Trans. 1 1984, 2895.