2,3-Dihydro-dithiin and -dithiepine-1,1,4,4-tetroxides: Small molecule non-peptide antagonists of the human galanin hGAL-1 receptor

Article abstract of DOI:10.1016/S0968-0896(00)00062-6

The neuropeptide galanin modulates several physiological functions such as cognition, learning, feeding behavior, and depression, probably via the galanin 1 receptor (GAL-R1). Using an HTS assay based on ¹²⁵I-human galanin binding to the human

Full text of DOI:10.1016/S0968-0896(00)00062-6

Bioorganic & Medicinal Chemistry 8 (2000) 1383±1391
2,3-Dihydro-Dithiin and -Dithiepine-1,1,4,4-tetroxides:
Small Molecule Non-Peptide Antagonists of the Human
Galanin hGAL-1 Receptor
Malcolm K. Scott,* Tina M. Ross, Daniel H. S. Lee, Hoau-Yan Wang,
Richard P. Shank, Kenneth D. Wild, Coralie B. Davis, Je€rey J. Crooke,
Alexander C. Potocki and Allen B. Reitz
Drug Discovery Division, The R. W. Johnson Pharmaceutical Institute, Welsh and McKean Roads, Spring House, PA 19447, USA
Received 18 November 1999; accepted 11 February 2000
AbstractÐThe neuropeptide galanin modulates several physiological functions such as cognition, learning, feeding behavior, and
depression, probably via the galanin 1 receptor (GAL-R1). Using an HTS assay based on 125I-human galanin binding to the human
galanin-1 receptor (hGAL-R1), we discovered a series of 1,4-dithiin and dithiipine-1,1,4,4-tetroxides that exhibited binding anity
IC50's to hGAL-R1 ranging from 190 to 2700 nM. Two of the dithiepin analogues, 7 and 23, behaved pharmacologically as hGAL-
R1 antagonists in secondary assays involving adenylate cyclase activity and GTP binding to G-proteins. Analogues 7 and 23 were
also active in functional assays involving galanin, reversing the inhibitory e€ect of galanin on acetylcholine (ACh) release in rat
brain hippocampal slices and electrically-stimulated guinea pig ileum twitch. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
to carry out memory tasks in the Morris water maze,10
the sunburst radial maze,11 and delayed non-matching
to sample tests.12 In several studies, increases in feeding
were demonstrated in rats and ground squirrels using
centrally administered galanin.13 Galanin may also be
involved in emotional state via inhibition of dopami-
nergic cells in the ventral tegmentum as described in
recent work using an animal model of depression.14 The
antinociceptive e€ects of galanin have been demon-
strated in rats via intrathecal administration15 and by
the ®nding that the non-speci®c galanin receptor mod-
ulators M-35 and galantide reverse spinal analgesia
induced by morphine.16
The neuropeptide galanin, ®rst isolated in 1983 from
porcine intestine,1 is a ubiquitous 2±9±30 amino acid
peptide that is found in the mammalian central (CNS)
and peripheral (PNS) nervous systems.2 4 Human gala-
nin (1) contains 30 amino acids,5 whereas galanin found
in other species contains 29 amino acids with a C-term-
inal amide.6 In the CNS, galanin is distributed in axons
and neurons located in the thalamus, hypothalamus,
cortex, amygdala, hippocampus and spinal cord,7,8 and
in the PNS it is found in pancreatic, gastrointestinal,
bladder, and genital tissue.9
The diverse physiological e€ects of galanin in mammals
can be attributed to speci®c interaction with a series of
at least three G-protein coupled seven-transmembrane
galanin receptors (GPCRs) including GAL-R1,17 21
GAL-R2,22 27 and GAL-R3.26,28 Several recent reviews
give extensive coverage to the identi®cation, molecular
biology, and function of these sub-types.29 31 GAL-R1
is primarily localized in the brain,21 whereas GAL-R2 is
largely distributed in the periphery,24,27 and GAL-R3
mRNA is found in some peripheral tissues, with low
levels in the brain.28,32 In the human brain, GAL-R2 and
GAL-R3 are more widely distributed than GAL-R1,
which is localized in greatest abundance in the cerebral
GWTLNNSAGYLLGPHAVGNHRSFSDKNGLTS
1
Galanin has signi®cant central e€ects on cognition,
learning, emotional state, feeding behavior, and the
modulation of pain in animals. When administered
intraventricularly to rats, galanin potently impairs cog-
nitive performance as measured by the decreased ability
*Corresponding author. Tel.: +1-215-628-5608; fax: +1-215-628-
4985; e-mail: scottm@prius.jnj.com
0968-0896/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(00)00062-6

1384
M. K. Scott et al. / Bioorg. Med. Chem. 8 (2000) 1383±1391
cortex, amygdala, and substantia nigra.26,33 GAL-R1
and GAL-R3 negatively regulate adenylate cyclase,34 37
whereas GAL-R2 is only slightly inhibits the production
of cAMP, and is also coupled to phospholipase C.26,37
Table 1. Binding anities of 2-substituted-1,1,4,4-dithiintetroxides
Because of the high level of hGAL-R1 in human brain,
this receptor is an appropriate target for the discovery
of CNS drugs for treatment of various disorders.
Peptide hGAL-R1 ligands such as galantide (M-15),38
M-35,39 M-40,40 and galanin peptides attached to non-
peptide units41 have been evaluated. Recently, a non-
peptide hGAL-R1 antagonist (2) was reported with a
1.7 mM hGAL-R1 IC50.42
Compound
Z
hGALR1
IC50 (nM)
3
4
5
6
Phenyl
4-Bromophenyl
PhenylCH2(NH2)CH
2700
1000
8000
1000
PhenylCH2(4-n-BuOphenylCH2OCONH)CH
We employed an hGAL-R1 high throughput screen
(HTS) of our corporate compound collection which
resulted in a 0.14% hit rate of >50% inhibition at
25 mM. Con®rmation, IC50 determinations, and second-
ary screening of the hits identi®ed a variety of active
chemical series. From this group, we now disclose the
activity of the dithiin-1,1,4,4-tetroxide (3) chemical ser-
ies as hGAL-R1 antagonists (Tables 1±3).
with either propyl- or phenethylamine to give amide 50.
Oxidation of 50 with 30% hydrogen peroxide/acetic
acid gave 16 or 18. The reduced analogue of 7, com-
pound 52, was obtained as shown in Scheme 4. Reduc-
tion of 44 (R=4-Me, n=1) with triethysilyl hydride
gave dithiipan 51, which was oxidized with m-chloro-
perbenzoic acid (mCPBA), a€ording 52.
We describe here the structure activity relationships
(SAR) based on 3, ultimately leading to dithiipin-
1,1,4,4-tetroxide 7, the ®rst non-peptidic sub-micro-
molar hGAL-R1 antagonist (190 nM IC50).
Biological Evaluations
The binding anities of compounds 3±42 at the hGAL-
R1 were evaluated using membranes from human
Bowes melanoma cells bound to wheat germ agglutin-
conjugated scintillation proximity assay (SPA) beads
and are reported as IC50 values in Tables 1±3. Determi-
nation of agonist/antagonist activity of 3±42 was done
by measuring cAMP accumulation and GTP-gS bind-
ing. The e€ects of two compounds, 7 and 23, were
examined in two functional assays: galanin-induced
inhibition of ACh release from rat brain hippocampal
slices and the galanin-induced inhibition of ACh-stimu-
lated guinea pig ileum twitch. The results are shown in
Figures 2 and 3. The propensity of 32 and 38 to bind
irreversibly to proteins was determined (Table 4) and
cytotoxicity measurements were carried out on 3, 7, 10,
17,23, 30, 33 and 35 (Table 5).
Synthetic Chemistry
The reported synthesis of 3 and modi®cations therein
were employed for the preparation of 3±42 (Scheme
1).43 Treatment of an aryl ketone or an a-halo-aryl
ketone with either ethane- or propane-dithiol and
ꢀ
BF3 OEt2 in CH2Cl2 at ambient temperature gave the
corresponding dithiolanes 43.44 Using either bromine45,46
or NBS,47 these dithiolanes were converted to dithiins
or dithiipins 44 via dithiolane ring opening followed by
reclosure.45 Oxidation of 44 with H2O2/HOAc a€orded
compounds 3, 4, 7±15, 17 and 19±42. The phenethyl
analogues of 3, compounds 5 and 6, were obtained by
the route depicted in Scheme 2. Chloroketone 45 was
converted to the respective intermediate dithiolane 46
.
using BF3 OEt2 and ethanedithiol, followed by re¯uxing
pyridine. Oxidation of 46 provided 47 which was
converted to 5 using tris-tri¯uoroacetate boron.48 Com-
pound 6 was obtained by reaction of 5 with 4-butoxy-
benzoyl chloride. Benzamide analogues 16 and 18 were
prepared as shown in Scheme 3. Treatment of ester 48,
prepared as described above, with aqueous potassium
hydroxide in DMF a€orded acid 49, which was coupled
Results and Discussion
Modi®cation of 3 to give 4-bromo derivative 4 resulted
in an increase in micromolar binding anity at the
hGAL-R1, as did conversion to phenethyl analogue 6,
whereas the less complex phenethyl derivative 5 had
weaker binding anity (Table 1). Replacement of the

M. K. Scott et al. / Bioorg. Med. Chem. 8 (2000) 1383±1391
1385
Table 2. Binding anities of dithiipin-2-phenyl-1,1,4,4-dithiipintetr-
oxides at hGAL-R1
Table 3. Binding anities of 2-(naphthy)l- and 2-(thienyl)-1,1,4,4
dithiipintetroxides at hGALR1
Compound
X
hGAL-R1 IC50 (nM)
Compound
R
hGAL-R1 IC50 (nM)
40
41
42
1-Naphthyl
2-Thienyl
3-Thienyl
371
1300
1300
7
8
9
4-Methyl
4-Bromo
4-Fluoro
4-Phenoxy
4-Ethyl
2-Fluoro
190
260
410
410
430
430
440
490
570
577
580
586
590
600
600
640
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
stitution on the phenyl ring, phenoxycarbonyl analogue
13 and tri¯uoromethyl analogue 20 show signi®cant
binding anity for the hGAL-R1 with IC50's of 440 nM
and 600 nM respectively. The 3-iodo, bromo, methoxy,
and chloro analogues, 24±27, are less active with bind-
ing anities ranging from 850 to 1 mM while the 3-
¯uoro and 3-tri¯uoromethoxy analogues, 33 and 39,
have micromolar activity. Compound 40 (Table 3), in
which the phenyl ring of 7 was replaced with a 1-naph-
thyl group, showed good binding anity equal to the
more active 4-phenyl analogues. The 2- and 3-thienyl
derivatives, 41 and 42, were much less active than either
7 or 40, with binding anities in the micromolar range.
Finally, the saturated analogue of 7, dithiipan 52, had
no binding anity for hGAL-R1.
3-Phenyloxycarbonyl
4-Tri¯uoromethoxy
2-Bromo
4-Propionamido
2-Tri¯uoromethyl
4-Phenethylamido
4-Ethoxy
3-Tri¯uoromethyl
4-Phenyl
2-Chloro
4-Iodo
3-Iodo
3-Bromo
3-Methoxy
3-Chloro
710
850
890
910
1000
1000
1000
1100
1100
1300
1300
1580
1600
1900
2100
2700
2700
2-Methoxy
4-Ethyloxycarbonyl
4-Methoxy
4-Cyclohexyl
3-Methyl
Using methods previously described to measure cAMP
accumulation49 and GTP-gS binding,49 it was determined
that compounds 3±42 are full antagonists at hGAL-R1
(data not shown). Since 7 had the best binding anity
of the analogues of 3 at the hGAL-R1, it was evaluated
in our general receptor screening program. Compound 7
appears to be relatively speci®c for the hGAL-R1, with
no binding observed at 30 mM to hGAL-R2 and weak
or no anity for hGAL-R2, NPY1, NPY2, NPY5,
MC4R, d-opioid receptor and other GPCRs.
3-Fluoro
4-Phenylsulfonyl
2-Methyl
4-Chloro
4-Tri¯uoromethyl
H
3-Tri¯uoromethoxy
dithiintetroxide ring of 3 with a dithiipintetroxide ring
(viz. 38) had no e€ect on binding anity, whereas
corresponding 4-methylphenyl derivative 7 had an
increased binding anity by an order of magnitude
(Table 2). Interestingly, the 2- and 3-methyl analogues,
35 and 32, of 7 and the 4-tri¯uoromethyl analogue 37
were much less active, displaying binding anities in the
micromolar range. Binding anity to the hGAL-R1 can
sustain some structural variation in the 4-phenyl posi-
tion as demonstrated by the 4-bromo analogue 8 which
was slightly less active than 7 while other 4-substituted
analogues, 9±11 and 14, had signi®cant anity in the
400 nM range. Somewhat less active 4-substituted ana-
logues were amides 16 and 18, ethoxy, phenyl and iodo
analogues, 19, 21 and 23, which ranged from 577 nM to
710 nM. Binding anity decreases further for the 4-
ethoxycarbonyl, methoxy, cyclohexyl, phenylsulfonyl,
and chloro analogues, 29, 30, 31,34, and 36, which have
micromolar activity. Halogen substitution or halogen-
bearing groups in the 2-phenyl position are bene®cial as
shown by the binding anities of 2-¯uoro,-bromo,-tri-
¯uoromethyl, and -chloro analogues, 12, 15, 17 and 22,
which range from 430 nM to 640 nM whereas the 2-
methoxy analogue 28 is much less active. For 3-sub-
Since galanin inhibits acetylcholine (ACh) release in
mammalian brain,49 51 presumably via a pertussis
toxin-sensitive GI protein,49 and impairs performance in
rodent learning in a cholinergically related manner,52 we
studied the e€ects of 7 and 23 on the release of ACh
from rat cortical brain slices and cortical synaptosomes
in the presence of galanin.49 For comparative purposes,
38
we employed the galanin antagonist galantide in the
cortical synaptosome study. As shown in Figures 1 and
2, 7 and 23 blocked the inhibitory e€ect of galanin on
ACh release in a potent, dose-dependent manner in
both experiments. At 10 mM, 7 almost totally blocked
the e€ects of galanin in rat cortical brain slices. Fur-
thermore, in another assay involving the e€ect of gala-
nin on ACh modulation, the galanin-induced inhibition
of electrically-induced contraction of guinea pig ileum,
both 7 and 23 signi®cantly blocked galanin inhibition as
depicted in Figure 3.
The dithiin and dithiepine-1,1,4,4-tetroxides described
above contain a reactive segment, namely a 1,4-a,b-
unsaturated sulfone, that is capable of undergoing

1386
M. K. Scott et al. / Bioorg. Med. Chem. 8 (2000) 1383±1391
.
Scheme 1. Reagents and conditions: (a) BF3 OEt2, HSCH2(CH2)nCH2SH; (b) Br2 or NBS; (c) H2O2, HOAc.
.
Scheme 2. Reagents and conditions: (a) BF3 OEt2, HSCH2CH2SH; (b) pyridine; (c) H2O2, HOAc; (d) TFA, B(CF3CO2)3; 4-BuOPhCOCl, TEA.
Scheme 3. Reagents and conditions: (a) 3 N KOH, DMF; (b) N-methylmorpholine; (c) i-BuOCOCl; (d) RNH2; (e) H2O2, HOAc.

M. K. Scott et al. / Bioorg. Med. Chem. 8 (2000) 1383±1391
1387
Scheme 4. Reagents and conditions: (a) Et3SiH, CH2Cl2; (b) mCPBA, CH2Cl2..
Figure 3. E€ect of 7 and 23 on the galanin-induced inhibition of
guinea pig ileum twitch.
Figure 1. E€ect of 7 and 23 on the galanin (GAL)-induced inhibition
of ACH release from rat cortical brain slices.
were preincubated with BSA prior to the hGAL-R1
binding experiments. The results, which are reported in
Table 4 as the ratio of the IC50 without incubation with
BSA to the IC50 following preincubation with BSA,
indicate that the potencies of 3, 7, 10, 23, 30, 32, 33, 35
and 38 were only slightly altered, showing that pre-
incubation of the compounds with proteins does not
a€ect their activity at the hGAL-R1. In the second
study, Bowes melanoma membranes containing the
hGAL-R1 were incubated with 32 and 38, washed with
a low salt bu€er, and then treated again with 32 and 38.
Both compounds generated similar dose±response curves
compared with untreated Bowes melanoma membranes,
suggesting that these compounds do not bind irrever-
sibly to proteins or the hGAL-R1. In addition, the e€ect
of several of these analogues was evaluated on cell via-
bility. When 7, 10, 23, 30, 33 and 35 were incubated
with Bowes melanoma cells, there was no e€ect on cell
viability and the cells multiplied and maintained in cul-
ture, whereas 17 and 28 caused a decrease (Table 5).
Figure 2. E€ect of 7, 23 and galantide (GAT) on the galanin(GAL)-
induced inhibition of ACH release from rat cortical synaptosomes.
Michael-type addition by a nucleophile. The propensity
of these compounds as Michael acceptors was con-
®rmed by the disappearance of both 3 and 7 within 1 or
5 h respectively after treatment with propylamine and
diethylamine (data not shown).
Conclusions
The hGAL-R1 antagonistic activity of the title com-
pounds is an important ®nding in the discovery process
for CNS therapeutic agents based on hGAL-R1,
demonstrating that non-peptidic small molecules can
bind to the receptor with sub-micromolar anity and
elicit responses in functional assays antagonizing the
e€ect of galanin. While these compounds are precluded
as hGAL-R1 antagonist drug candidates due to their
chemically reactive nature and solubility, they have
Because of the chemically reactive nature of these com-
pounds, two studies were performed using selected ana-
logues to determine if they bound irreversibly to
proteins and, more speci®cally, to the hGAL-R1. In the
®rst study, compounds 3, 7, 10, 23, 30, 32, 33, 35 and 38

1388
M. K. Scott et al. / Bioorg. Med. Chem. 8 (2000) 1383±1391
Table 4. Binding of dithiipine analogues to BSA
the organic layer, drying, and ®ltering a€orded a crude
orange solid. This material was passed through ¯ash
silica using CH2Cl2 as eluant to give 44 (R=H, n=0) as
a white crystalline solid. A slurry of 44 (R=H, n=0,
1.0, 5.15 mmol) and HOAc (3.0 mL) was added in por-
tions to a re¯uxing solution of 30% H2O2 (2.97 mL) and
HOAc (2.97 mL) and the resulting light yellow solution
was re¯uxed for 15 min. On cooling, a crystalline solid
formed which was ®ltered and washed with water,
a€ording 3 as a light yellow crystalline solid (0.80 g,
61%). 1H NMR (300 MHz, CDCl3) d 7.42±7.66 (m, 5H,
aromatic), 6.67 (s, 1H, vinyl), 4.00 (s, 4H, SO2CH2). MS
(EI): m/e=258.
Compound
IC50 (nM)/IC50 (nM) BSA treated
3
7
0.70
0.60
2.02
0.95
1.31
0.88
1.19
0.86
0.83
10
23
30
32
33
35
38
Table 5. E€ect on cell viability of dithiin and dithiipine 1,1,4,4-tetr-
oxides
Compound 4 was prepared in a similar fashion to give a
white powder (0.037, 3%). 1H NMR (DMSO-d6) d
32;7.81 (s, 1H), 7.8 (d, 2H), 7.55 (d, 2H), 4.3 (q, 4H).
MS (ES+)=335.19, 313.13, 292.14, 277.17, 275.16,
255.2, 253.2, 242.24, 232.18, 191.16, 169.19.
Compound
% Cell viability of Bowes
melanoma cells at 18 h (5 mM)
3
7
100
95
10
17
23
28
30
33
35
95
30
100
60
100
100
90
2,3-Dihydro-2-(1-(amino)-phenethyl)-1,4-dithiin-1,1,4,4-
tetroxide (5). A solution of alpha-chloro (N-carboxy-
benzyl) phenylalanine (45, 0.7 g, 2.1 mmol) in tetra-
hydrofuran (10 mL) and ethanedithiol (0.18 mL,
2.1 mmol) was treated dropwise with boron tri¯uoro-
etherate (0.26 mL, 2.1 mmol) at room temperature. The
resulting reaction was stirred overnight at room tem-
perature, quenched carefully with 10% NH4OH, and
extracted with CH2Cl2. The organic layer was separated,
dried with MgSO4, ®ltered and evaporated. The residual
oil was taken up in pyridine (5 mL) and heated to 100 ꢁC
for 2 h. Evaporation of the solvent yielded dithiane 46 as
an oil which was treated with a solution of 30% H2O2
and acetic acid (5 mL) and stirred at ambient tempera-
ture for 72 hours. Filtration of the reaction mixture
a€orded 47 (0.175 g, 19%) as a crystalline solid.
served a purpose in providing SAR information for
other related chemical series.
Experimental
Chemistry
1
General remarks. The H NMR spectra were obtained
on a 300 MHz Bruker AC 300 NMR spectrometer with
Me4Si as an internal standard. The mass spectrum of
each compound was generated using Hewlett±Packard
1050 Series electrospray or chemical ionization mass
spectrometers. The spectral data for each compound
supported the assigned structure. Chromatographic
separations were performed using either a Chromato-
tron Model 7924T radial chromatography system or a
Biotage Flash 40 ¯ash chromatography system. Most
reagents and solvents were purchased and used without
further puri®cation.
This material (0.14 g, 0.32 mmol) was dissolved in tri-
¯uoroacetic acid (3 mL), treated with tris-tri¯uoro-
acetate boron48 (0.56 g, 1.6 mmol) and stirred for one
hour at room temperature. Methanol was added to the
reaction mixture followed by evaporation in vacuo (3Â).
The residue was taken up in 10% MeOH/dichloro-
methane solution and washed with sodium carbonate
solution. The organic layer was separated, dried with
MgSO4, ®ltered and evaporated to a residue which was
puri®ed via silica prep plate (10% MeOH/dichloro-
methane) to yield 5 (0.017 g, 18%). 1H NMR (CDCl3) d
7.4 (m, 5H), 6.95 (s, 1H), 4.35 (dd, J=1.5 Hz, 0.5 Hz,
1H), 3.85 (m, 4H), 3.4 (dd, J=1.5 Hz, 0.5 Hz, 1H), 2.6
(q, 1H), 1.4 (bs, 2H). MS (ES+)=302.1 (H+), 209.1.
2,3-Dihydro-1,4-dithiin-1,1,4,4-tetroxide (3). A mixture
of a-chloroacetophenone (2.50 g, 16.0 mmol) and eth-
anedithiol (1.69 g, 17.9 mmol) at ambient temperature
was treated dropwise with boron tri¯uoroetherate
(1.13 g, 8.0 mmol). A mild exotherm and gas evolution
ensued. After 5 min, Et2O (10 mL) was added followed
by water with thorough mixing. The organic layer was
separated and washed three times with 15% NH4OH
solution, dried over anhyd. Na2SO4, ®ltered, and eva-
porated to a yellow oil. A solution of this material and
pyridine (20 mL) was re¯uxed for 0.5 h. The reaction
mixture was cooled, diluted with water (20 mL), and
mixed thoroughly with 50:50 EtO2:hexane. The organic
layer was separated, washed with 1 N HCl solution until
acidic, water and 15% NH4OH solution. Separation of
2,3-Dihydro-2-(1-(4-(butoxyphenylcarbonyl)amino)-phen-
ethyl)-1,4-dithiin-1,1,4,4-tetroxide (6). Triethylamine
(89 mL, 0.64 mmol) was added to a solution of 5 (0.11 g,
0.32 mmol) in dioxane (1 mL) and cooled to 0 ꢁC fol-
lowed by addition of 4-butoxybenzoyl chloride (61 mL,
0.32 mmol). The reaction was stirred for 48 h at ambient
temperature, diluted with dichloromethane, and washed
with sodium bicarbonate solution. The organic layer
was separated, dried with MgSO4, ®ltered, and solvent
evaporated in vacuo to a crude oil. Puri®cation by silica
radial chromatography (10% MeOH/dichloromethane)

M. K. Scott et al. / Bioorg. Med. Chem. 8 (2000) 1383±1391
1389
1
a€orded 6 (0.038, 25%) as a crystalline solid. H NMR
(DMSO-d6) d 8.6 (d, 1H), 7.75 (d, 2H), 7.4 (s, 1H), 7.25
(d, 2H), 7.15 (m, 3H), 6.8 (d, 2H), 5.3 (bt, 1H), 4.2 (m,
4H), 4.0 (t, 2H), 3.8 (bs, 1H), 3.3 (m, 1H), 3.0 (dd, J=6
Hz, 4.5 Hz, 1.7 (m, 2H), 1.4 (m, 2H), 0.8 (t, 3H). MS
(ES+)=478.1 (H+), 191.1, 169.1.
with 1 N HCl and the organic layer was separated, dried
over anhyd. MgSO4, ®ltered and evaporated to yield acid
49 (R=4-CO2H, n=1, 0.44 g, 35%). N-Methylmor-
pholine (90 mL, 0.8 mmol) was added to a solution of 49
(0.1 g, 0.4 mmol) in chloroform (20 mL) cooled to 0 ꢁC.
After stirring for 15 min, isobutylchloroformate (60 mL,
0.4 mmol) was added and stirred for an additional
20 min followed by addition of propylamine (24 mL,
0.4 mmol). The reaction was stirred for 2 h, diluted with
saturated bicarbonate solution, and extracted with di-
chloromethane. The organic layer was separated, dried
with MgSO4, ®ltered and evaporated to yield an oil.
Puri®cation by silica radial chromatography (2%
MeOH/dichloromethane) yielded 50 (0.07 g, 60%).
2,3-Dihydro-2-(4-methylphenyl)-1,4-dithiepine-1,1,4,4-
tetroxide (7). A solution of 40-methylacetophenone
(10 g, 74.5 mmol) and CH2Cl2 (200 mL) was treated
dropwise with boron tri¯uoroetherate (12.5 mL,
104 mmol) at room temperature. To the resulting solu-
tion was added 1,3-propanedithiol (7.45 mL, 74.5 mmol)
and the reaction mixture was stirred for 4 h at room
temperature. The reaction was quenched carefully with
saturated NaHCO3 solution and extracted with di-
chloromethane. The organic layer was separated and
evaporated in vacuo. The residue was dissolved in ethyl
acetate and washed with 1 N NaOH, water, and satu-
rated NaCl solution. The organic layer was separated,
dried with MgSO4, ®ltered and solvent evaporated in
vacuo to yield a crude oil. Puri®cation of the crude oil
via ¯ash chromatography (50% CH2Cl2/hexane) a€orded
15.2 g of dithiolane 43 (R=4-Me, n=1) as a crystalline
solid. A solution of this material (2.00 g, 8.9 mmol), N-
bromosuccinimide (1.59 g, 8.9 mmol), and CHCl3
(150 mL) was stirred at room temperature for 20 min
followed by the addition of 1,8-dimethylamino naph-
thalene (1.91 g, 8.9 mmol). The reaction was stirred
for 10 min, ®ltered, and the organic solvent was eva-
porated in vacuo. The residue was dissolved in 1:1
CH2Cl2:hexane and ®ltered to yield a crude oil. Pur-
i®cation of the crude oil by ¯ash chromatography (30%
dichloromethane/hexane) yielded 0.41 g of dithiin 44
(R=4-Me, n=1) as an oil.
A solution of 50 (0.07 g, 0.24 mmol) in dichloromethane
(1 mL) was added to a solution of re¯uxing 30% H2O2
(1 mL) and acetic acid (1 mL) and re¯uxed for 5 min
followed by cooling in the freezer for 18 h. Filtration
a€orded 16 (0.032 g, 37%) as a crystalline solid. 1H
NMR (DMSO-d6) d 7.9 (d, 2H), 7.58 (s, 1H), 7.52 (d,
1H), 3.95 (t, 2H), 3.8 (t, 2H), 3.2 (q, 2H), 2.35 (m, 2H),
1.5 (m, 2H), 1.9 (t, 3H). MS (CI)=358 (H+), 299, 274,
262, 246, 230, 206, 189.
2,3-Dihydro-2-(4(phenethylaminocarbonyl)-phenyl)-1,4-
dithiepine-1,1,4,4-tetroxide (18). Using the procedure
described for 16, 18 (0.011 g, 19%) was obtained as a
1
crystalline solid. H NMR (DMSO-d6) d 8.7 (t, 1H),
7.78 (d, 2H), 7.55 (d, 2H), 7.52 (s, 1H), 7.3 (m, 5H), 4.0
(t, 2H), 3.75 (t, 2H), 3.52 (q, 2H), 2.75 (t, 2H), 2.4 (m,
2H). MS (CI)=388 (H+), 356, 324, 308, 292, 251.
2,3-Dihydro-2-(2-methylphenyl)-1,4-dithiepine-1,1,4,4-
tetroxide (32). Using the procedure described for the
preparation of 7, a solution of dithiolane 43 (R=2-Me,
n=1, 1.49 g, 6.7 mmol) and CHCl3 (136 mL), cooled in
an ice bath, was treated dropwise with a solution of Br2
(0.22 g, 7.6 mmol) and CHCl3 (34 mL). After the addi-
tion was complete, the ice bath was removed and the
reaction was stirred at room temperature for 15 min.
Excess 5% NaHCO3 was added with thorough mixing.
The organic layer was separated, dried over anhyd.
Na2SO4, ®ltered, and evaporated to an oily residue. The
addition of Et2O caused a white solid to form, which
was ®ltered. The ®ltrate was evaporated and the residue
was passed through ¯ash silica using 75:25 hexane:
CH2Cl2 to give 0.276 g (19%) of 44 (R=2-Me, n=1) as
an oil. A solution of this material (0.110 g, 0.495 mmol)
and HOAc (0.20 mL) was added to a re¯uxing solution
of 30% H2O2 (0.297 mL) and HOAc (0.297 mL). The
solution was re¯uxed 15 min, cooled in ice, and a white
crystalline solid formed that was ®ltered, washed with
water and Et2O, and dried to a€ord 32, 70 mg (50%).
1H NMR (300 MHz, DMSO-d6) d 7.22±7.40 (m, 5 H,
aromatic, vinyl), 3.90±4.00 (t, 2H, SO2CH2), 3.74±3.84
(t, 2H, SO2CH2), 2.26±2.39 (m, 2H, CH2), 2.31 (s, 3 H,
CH3). MS (EI): m/e=286.
This material (0.41 g, 1.83 mmol) was dissolved in di-
chloromethane (10 mL) and slowly added to a mixture
of 30% H2O2 (2 mL)/acetic acid (2 mL) at 100 ꢁC and
stirred at 100 ꢁC for 5 min. The reaction was cooled
and stored in the freezer overnight. Filtration a€orded
7 (0.680 g, 11%) as a crystalline solid. This material
(0.100 g) was puri®ed using a Model 215 Gilson
Automated Preparative HPLC (reverse phase column
100Â20 mm, stationary phase: C-18 on 5.4 mm beads
with 80 A pore size, mobile phase: acetonitrile/water con-
taining 0.1% TFA, 30±90% acetonitrile gradient) to give
7 as a white solid; mp (sinter 182 ꢁC) 184.5±186.5 ꢁC; 1H
NMR (300 MHz, DMSO-d6) d 7.32±7.40 (m, 3H, aro-
matic, vinyl), 7.23±7.32 (m, 2H, aromatic), 3.90±3.98 (t,
2H, SO2CH2), 3.78±3.88 (t, 2H, SO2CH2), 2.5 (m, 2H,
CH2), 2.331 (s, 3H, CH3); MS (CI): m/z 304 (M+NH+4 ).
Anal. calcd for C12H14O4S2: C 50.33, H 4.93, S 22.39;
obsd: C 50.31, H 4.92, S 22.10. Using the procedure
described above, compounds 8±15, 16, 19±26 and 28±42
were prepared.
2,3-Dihydro-2-(4(propylaminocarbonyl)-phenyl)-1,4-dithi-
epine-1,1,4,4-tetroxide (16). A stirred mixture of 48
(1.40 g, 5.0 mmol), DMF (100 mL), and 3 N aqueous
KOH was heated at 100 ꢁC for 0.5 h. The reaction was
cooled to room temperature followed by the addition of
CH2Cl2 (100 mL) and ice. The mixture was acidi®ed
Prepared similarly was compound 27.
(2-Methylphenyl)-1,4-dithiepan-1,1,4,4-tetroxide (43). A
mixture of 44 (R=4-Me, n=1, 0.165 g, 0.86 mmol),

1390
M. K. Scott et al. / Bioorg. Med. Chem. 8 (2000) 1383±1391
triethylsilyl hydride (0.164 mL, 1.03 mmol) and CH2Cl2
(10 mL) was stirred at room temperature for 15 min.
Tri¯uoroacetic acid was added (0.132 mL, 1.72 mmol) to
the reaction and the mixture was stirred overnight at
room temperature. The reaction mixture was mixed
thoroughly with 5% aqueous NaHCO3 solution, the
organic layer was separated, dried with MgSO4, ®ltered
and the solvent evaporated to yield 51 (0.11 g, 67%).
single pulses, 0.25 ms duration) at the minimum voltage
required to produce a maximal tissue response (40±
60 V). Agonists (such as galanin) were evaluated for
their ability to inhibit the electrically-evoked twitch.
Putative antagonists were evaluated for their ability to
shift the concentration±e€ect curve of an agonist to the
right. IC50 and maximum inhibition values were deter-
mined using Prism (Graphpad Software Inc., San
Diego, CA).
A solution of 51 (0.11 g, 0.5 mmol) and CH2Cl2 (5 mL)
was added to a mixture of mCPBA (0.635 g, 2.21 mmol)
and dichloromethane (20 mL) and stirred overnight at
room temperature. The reaction mixture was ®ltered,
diluted with ethyl acetate and washed with 1 N NaOH.
The organic layer was separated, dried with MgSO4,
®ltered and evaporated to yield 52 as a white powder
(0.020 g, 7%). H NMR (DMSO-d6) d 7.25 (q, 4H), 4.9
(d, 1H), 4.45 (m, 1H), 3.7 (m, 3H), 3.55 (m, 2H), 2.3 (s,
3H), 2.25 (t, 2H). MS (CI)=306 (NH4+), 240, 191.
Protein binding assay (BSA). In a separate experiment,
100 mM of compounds 3, 7, 10, 23, 30, 32, 33, 35 and 38
were ®rst incubated with 1% bovine serum albumin for
16 h at room temperature. Their respective anities to
hGAL-R1 were then measured in the binding assay as
described above.
1
Protein binding assay (Bowes melanoma cell mem-
branes). Bowes melanoma membranes were incubated
with compounds 32, 38 (10 uM), or vehicle for 1 h at
25 ꢁC and then washed extensively with 50 mM HEPES,
pH 7.4, containing 2 mM Ca, 2 mM Mg, 0.1% BSA and
protease inhibitors. These two membrane preparations
were then used in a receptor binding study as described
above. The dose±response curves generated for galanin,
32, or 38 using these membrane preparations were
almost identical to those obtained in the normal binding
assay.
Biology
Binding assays. Membranes prepared from the human
Bowes melanoma cell line that expresses hGAL-R1
(50 mg) were coupled to wheat germ agglutinin coated
scintillation proximity assay (SPA, 250 mg) beads in a
bu€er containing 50 mM HEPES, pH 7.4, 2 mM Ca++
,
2 mM Mg++, 0.1% BSA and a protease inhibitors
cocktail. 125I-Human galanin (40 pM) was then added in
the presence or absence of unknowns. Non-speci®c
binding was de®ned by 10 mM human galanin. IC50
values were determined by the Graphpad software
Prism (Graphpad Software Inc., San Diego, CA).
Cytotoxicity measurements. Human Bowes melanoma
cells were maintained in DMEM/F12 medium supple-
mented with 10% FCS and antibiotics in a 96-well cul-
ture plate. Test compounds 7, 10, 17, 23, 28, 30, 33 and
35 (1±100 mM) were added separately to each well and
incubated for 24 h at 37 ꢁC and 5% CO2. Cell viability
was determined using the MTT assay using the Cyto
Tox 96 Non-radioactive cytotoxicity assay kit (Pro-
mega) according to the vendor's protocol.
The hGAL-R2 binding study of 7 was performed using
conditions identical with the hGAL-R1 study except that
recombinant hGAL-R2 expressed on HEK293 cells was
used as the receptor source.
Secondary assays. Determinations of 35S-gGTP binding
to G-proteins and cAMP measurements were carried
out as described previously.49
Acknowledgements
The authors gratefully acknowledge helpful discussions
and scienti®c input of Drs. Barry Dubinsky and Carlos
Plata-Salaman of The R. W. Johnson Pharmaceutical
Research Institute and Dr. Jacqueline N. Crawley of
The National Institute of Mental Health.
Galanin-induced inhibition of acetylcholine release. Mea-
surements of the e€ect of 7 and 23 on galanin-induced
inhibition of ACh release from brain slices and synap-
tosomes were carried out as described previously.49
Electrical-stimulated guinea pig ileum twitch. Male,
Hartley guinea pigs (300±900 g) were euthanized by CO2
and the terminal portion of the ileum (discarding the 10
cm nearest the caecum) was quickly removed and placed
into oxygenated (95:5 O2:CO2) Kreb's bicarbonate buf-
fer (NaCl, 119 mM; KCl, 4.7 mM; CaCl2, 2.5 mM;
KH2PO4, 1.2 mM; MgSO4, 1.2 mM; NaHCO3,
25.0 mM; dextrose, 11.1 mM) at 37 ꢁC. The ileum was
cleaned of intestinal contents, blood vessels, and con-
nective tissue, cut into several pieces (approximately
2 cm each), attached to gold chains with stainless steel
clips, and then suspended between vertical platinum
electrodes in 20 ml organ baths at a tension of 1.0 g,
again bathed in oxygenated Kreb's bicarbonate bu€er
at 37 ꢁC. Tissues were stimulated electrically (0.1 Hz,
References and Notes
1. Tatemoto, K.; Rokaeus, A.; Jornwall, H.; McDonald, T. J.;
Mutt, V. FEBS Lett. 1983, 164, 124.
2. Bartfai, T.; Hokfelt, T.; Langel, U. Crit. Rev. Neurobiol.
1993, 7, 229.
3. Crawley, J. N. Regulatory Neuropeptides 1995, 59, 1.
4. Kask, K.; Berthold, M.; Bartfai, T. Life Sci. 1997, 60, 1523.
5. Bersani, M.; Johnsen, A. H.; Hojrup, P.; Dunning, B. E.;
Andreasen, J. J.; Holst, J. J. FEBS Lett. 1991, 283, 189.
6. Bartfai, T.; Langel, U. Eur. J. Med. Chem. 1995, 30, 163s.
7. Melander, T.; Hokfelt, T.; Rokaeus, A. J. Comp. Neurol.
1986, 248, 475.
8. Sko®tsch, G.; Jacobowitz, D. M. Peptides 1985, 6, 509.
9. Rokaeus, A. Trends Neurosci. 1987, 10, 158.

M. K. Scott et al. / Bioorg. Med. Chem. 8 (2000) 1383±1391
1391
10. Sundstrom, E.; Archer, T.; Melander, T.; Hokfelt, T.
Neurosci. Lett. 1988, 88, 331.
11. Malin, D. H.; Novy, B. J.; Lett-Brown, A.; Plotner, R. E.;
May, B. T.; Radulescu, S. J.; Crothers, M. K.; Osgood, L. D.;
Lake, J. R. Life Sci. 1992, 50, 939.
potential uses as therapeutic targets. In Annual Reports in
Medicinal Chemistry; Bristol, J. A., Ed.; Academic Press: New
York; 1998 Vol. 33, pp 41±50.
30. Wang, S.; Parker, E. M. Exp. Opin. Ther. Patents 1998, 8,
1225.
12. Robinson, J. K.; Crawley, J. N. Behav. Neurosci. 1993,
107, 458.
31. Wang, S.; Gustafson, E. L. Drug News Perspect. 1998, 8,
458.
13. Boswell, T.; Richardson, R. D.; Schwartz, M. W.; D'Al-
lessio, D. A.; Woods, S. C.; Sipols, A. J.; Baskin, D. G.;
Kenagy, G. J. Brain Res. 1993, 32, 379.
14. Weiss, J. M.; Bonsall, R. W.; Demetrikopoulos, M. K.;
Emery, M. S.; West, C. H. K. Ann. N. Y. Acad. Sci. 1998, 863,
364.
32. Bard, J. A.; Borowsky, B.; Smith, K. E.; Branchek, T. A;
Gerald, C. P.; Jones, K. A. World Patent WO 9815570, 1998.
33. Sullivan, K. A.; Shiao, L. L.; Cascieri, M. A. Biochem.
Biophys. Res. Commun. 1997, 233, 823.
34. Parker, E. M.; Izzarelli, D. G.; Nowak, H. P. Mol. Brain
Res. 1995, 34, 179.
15. Cridland, R. A.; Henry, J. L. Neuropeptides 1988, 11, 23.
16. Reimann, W.; Englberger, W.; Friderichs, E.; Selve, N.;
Wi€ert, B. Naunyn Schmied. Arch. Pharmacol. 1994, 350, 380.
17. Amirano€, B.; Servin, A. L.; Rouyer-Fessard, C.; Couvi-
neau, A.; Tatemoto, K.; Laburthe, M. Galanin Endocrinology
1987, 121, 284.
18. Servin, A. L.; Amirano€, B.; Rouyer-Fessard, C.; Tate-
moto, K.; Laburthe, M. Biochem. Biophys. Res. Commun.
1987, 144, 298.
19. Wynick, D.; Smith, D. M.; Ghatei, M.; Akinsanya, K.;
Bhogal, R.; Purkiss, P.; By®eld, P.; Yanaihara, N.; Bloom, S.
R. Proc. Natl. Acad. Sci. USA 1993, 90, 4231.
20. Fisone, G.; Langel, U.; Carlquist, M.; Bergman, T.; Con-
solo, S.; Hokfelt, T.; Unden, A.; Andell, S.; Bartfai, T. Eur J.
Biochem. 1989, 181, 269.
35. Wang, S.; He, C.; Maguire, M. FEBS. Lett. 1997, 411,
225±230.
36. Heuillet, E.; Bouaiche, Z.; Menager, J.; Dugay, P.; Munoz,
N.; Dubois, H.; Amirano€, B.; Crespo, A.; Lavayre, J.; Blan-
chard, J.-C.; Doble, A. Eur. J. Pharmacol. 1994, 269, 139.
37. Wang, S.; Hashemi, T.; Fried, S.; Clemmons, A. L.;
Hawes, B. E. Biochemistry 1998, 37, 6711.
38. Bartfai, T.; Bedecs, K.; Land, T.; Langel, U.; Bertorelli,
R.; Girotti, P.; Consolo, S.; Xu, X. J.; Wiesenfeld-Hallin, Z.;
Nilsson, S. Proc. Natl. Acad. Sci. USA 1991, 88, 10961.
39. Wiesenfeld-Hallin, Z.; Xu, X. J.; Langel, U.; Bedecs, K.;
Hoekfelt, T.; Bartfai, T. Proc. Natl. Acad. Sci. USA 1992, 89,
3334.
40. Langel, U.; Land, T.; Bartfai, T. Int. J. Pept. Protein Res.
1992, 39, 516.
21. Habert-Ortoli, E.; Amirano€, B.; Loquet, I.; Laburthe,
M.; Mayaux, J. F. Proc. Natl. Acad. Sci. USA 1994, 91, 9780.
22. Howard, A. D.; Tan, C.; Shiao, L. L. FEBS Lett. 1997,
405, 285.
41. Pooga, M.; Jureus, A.; Rezaei, K; Hasanvan, H.; Saar, K.;
Kask, K.; Kjellen, P.; Land, T.; Halonen, J.; Maeorg, U.; Uri,
A.; Solyom, S.; Bertfai, T.; Langel, U. J. Pept. Res. 1998, 51,
65.
23. Wang, S.; Hashemi, T.; He, C.; Strader, C.; Bayne, M.
Mol. Pharmacol. 1997, 52, 337.
24. Fathi, Z.; Cunningham, A. M.; Iben, L. G.; Battaglino, P.
B.; Ward, S. A.; Nichol, K. A.; Pine, K. A.; Wang, J.; Gold-
stein, M. E.; Iismaa, T. P.; Zimanyi, I. A. Brain Res. Mol.
Brain Res. 1997, 51, 49.
42. Chu, M.; Mierzwa, R.; Truumees, I.; King, A.; Sapidou,
E.; Barrabee, E.; Terracciano, T.; Patel, M. G.; Gullo, V. P.;
Burrier, R.; Das, P. R.; Mittleman, S.; Puar, M. S. Tetra-
hedron Lett. 1978, 387, 6111.
43. Brewer, A. D.; Davis, R. A. US Patent 4,004,018, 1977.
44. Afonso, C. A. M.; Barros, M. T.; Godinho, L. S.; May-
cock, C. D. Synthesis 1991, 575.
45. Caputo, R.; Ferreri, C.; Palumbo, G.; Capozzi, G. Tetra-
hedron 1986, 2369.
46. Caputo, R.; Ferreri, C.; Palumbo, G. Synthesis 1991, 223.
47. Jeko, J.; Timar, T.; Jaszberenyi, J. C. J. Org. Chem. 1991,
56, 223.
25. Smith, K. E.; Forray, C.; Walker, M. W. J. Biol. Chem.
1997, 272, 24612.
26. Kolakowski, L. F., Jr.; O'Neill, G. P.; Howard, A. D.;
Broussard, S. R.; Sullivan, K. A.; Feighner, S. D.; Sawzdargo,
M.; Nguyen, T.; Kargman, S.; Shiao, L.; Hreniuk, D. L.; Tan,
C. P.; Evans, J.; Abramovitz, M.; Chateauneuf, A.; Cou-
lombe, N.; Ng, G.; Johnson, M. P.; Tharian, A.; Khoshbouei,
H.; George, S. R.; Smith, R. G.; O'Dowd, B. F. J. Neurochem.
1998, 71, 2239.
48. Pless, J. Angew. Chem. Inter. Ed. 1973, 12, 147.
49. Wang, H.-Y.; Wild, K. D.; Shank, R. P.; Lee, D. H. S.
Neuropeptides 1999, 33, 197.
27. Bloomquist, B. T.; Beauchamp, M. R.; Zhelnin, L.;
Brown, S.; Gore-Willse, A. R.; Gregor, P.; Corn®eld, L. J.
Biochem. Biophys. Res. Commun. 1998, 243, 474.
28. Wang, S.; He, C.; Hashemi, T.; Bayne, M. J. Biol. Chem.
1997, 272, 31949.
50. Fisone, G.; Wu, C. F.; Consolo, S.; Nordstrom, O.;
Brynne, N.; Bartfai, T.; Melander, T.; Hokfelt, T. Proc. Natl.
Acad. Sci. USA 1987, 84, 7339±7343.
51. Ogren, S.-O.; Pramanik, A. Neurosci. Lett. 1991, 128, 253.
52. Mastropaolo, J.; Nadi, N. S.; Ostrowski, N. L.; Crawley, J.
N. Proc. Natl. Acad. Sci. USA 1988, 85, 9841.
29. Fathi, Z. Galanin receptors: recent developments and